Coordinated effects of microRNA-494 induce G₂/M arrest in human cholangiocarcinoma

Abstract

MicroRNA (miRs) have emerged as salient regulators in cancer homeostasis and, recently, as putative therapeutics. Cholangiocarcinomas (CCA) are aggressive cancers with survival usually measured in months. mRNA arrays followed by pathway analysis revealed that miR-494 is a major modulator of the cell cycle progression from gap 2 (G₂) to mitosis (M). We performed fluorescence activated cell sorting (FACS) as well as differential interference contrast (DIC) microscopy, and confirmed that miR-494 induces a significant arrest in G₂/M in CCA cells. Furthermore, we verified that miR-494 modulates the protein level of six genes involved in the G₂/M transition: Polo-like Kinase 1 (PLK1), pituitary tumor-transforming gene 1 (PTTG1), Cyclin B1 (CCNB1), cell-division cycle 2 (CDC2), cell-division cycle 20 (CDC20) and topoisomerase II α (TOP2A). Next, we identified direct binding of miR-494 to the open reading frame (ORF) and downregulation of PTTG1 and TOP2A. In summary, our findings suggest that miR-494 has a global regulatory role in cell cycle progression, exerted by concerted effects on multiple proteins involved in gap 1 (G₁) to synthesis (S), as described previously, as well as G₂ to M progression. Therefore, it appears that the simultaneous effects of a single miR species on multiple targets along the same canonical pathway is advantageous for the usage of miRs as therapeutics. In addition, our data suggest that miRs act within a narrow range. miR expression above the upper threshold does not appear to induce further effects, which is reassuring in terms of off-target effects of miR surrounding noncancerous tissue.

Introduction

Cholangiocarcinomas (CCA) are cancers thought to arise from epithelial cells lining the biliary tree.¹ Their prognosis is dismal, in part due to the late diagnosis.² Surgery remains the only curative option, but it is offered only to a minority of patients, who are diagnosed relatively early in the course of the disease.³ An improved understanding of the pathogenesis of CCA as well as novel diagnostic and therapeutic approaches are direly needed for this cancer.

Several previous studies identified miRs that are dysregulated in CCA as well as distinct roles played by miRs in CCA genesis or progression.^4-8 Thus, miRs rise as salient regulators of human CCA and may offer both novel insights into pathogenesis and hope for the developing of new therapeutics.

Recent landmark papers shed light on miR function in terms of binding specificity to their targets and in terms of the magnitude of their effect on targets. In contrast to previous views, it is now generally accepted that the majority of miR function is exerted through decreasing the amount of target mRNA rather than through inhibition of protein translation alone.⁹ In addition, there is growing evidence that miRs exert their function by coordinated inhibition of multiple targets within same pathway.¹⁰ The miR effect on any given target is usually modest, but the cumulative effect on multiple targets becomes phenotypically significant.^11,12

In a previous study, we found that miR-494 is downregulated in human CCAs.¹³ To obtain a comprehensive and unbiased view regarding the effects of miR-494 in cancer cells, we performed mRNA arrays on cells overexpressing miR-494 and on negative controls, respectively. By employing pathway analysis and then confirming the results with western blotting, we found that miR-494 exerts moderate effects on multiple molecules along the canonical G₁-S transition pathway, as described earlier,¹³ as well as on multiple molecules involved in the progression through G₂ and M phases of cell cycle. In the present study, we focused on the relation between miR494 and G₂-M cell cycle progression.

Results

We previously identified miR-494 to be downregulated in human CCA (by comparing the expression of miR-494 in 43 CCA vs. 30 normal specimens) and reported that its reinforcement results in cancer growth inhibition, in part through reinforcing the G₁ to S transition checkpoint.¹³ We transfected two CCA cell lines with miR-494 and a non-specific mimic (NSM), respectively, and performed mRNA arrays to identify in an unbiased fashion the genes whose expression is downregulated by miR-494. Interestingly, aside from “cell cycle G₁/S checkpoint regulation,” ingenuity pathway analysis (IPA) also indicated that miR-494 appears to coordinately affect several genes involved in the “mitotic roles of Polo-like kinase” and “cell cycle G₂/M checkpoint regulation” canonical pathways (Table 1). Of note, IPA ranks top canonical pathways based on the number of genes in a specific pathway that display decreased expression following treatment with miR-494; therefore, it is likely that miR-494 exerts its functional effects through these top canonical pathway. We queried IPA in terms of gene networks that are modulated by miR-494. As Figure 1 shows, miR-494 appears to modulate the largest number of targets in the metaphase-to-anaphase transition stage within the canonical pathway “mitotic roles of Polo-like kinase.” Figure 2 shows the reported putative miR-494 targets within the canonical pathway “Cell cycle: G₂/M checkpoint regulation.”

Table 1. Top canonical pathways. cDNA microarrays and gene pathway analysis reveal miR-494 regulated canonical pathways. Ingenuity pathway analysis summary reports that the top two canonical pathways modulated by miR-494 are “Mitotic roles of Polo-like kinase” and “Cell cycle G₂/M checkpoint regulation.”.

Name	p-value	Ratio
Mitotic Roles of Polo-Like Kinase	8.28E-06	6/58(0.103)
Cell Cycle: G2/M DNA Damage Checkpoint Regulation	2.00E-05	5/44(0.114)
Cell Cycle: G1/S Checkpoint Regulation	9.89E-05	5/58/(0.086)
Citrate Cycle	1.06E-04	4/57(0.07)
Small Cell Lung Cancer Signaling	3.62E-04	5/88(0.057)

Figure 1. Genes with decreased expression upon miR-494 stimulation are involved in the “mitotic roles of Polo-like kinase” canonical pathway. miR-494 downregulates multiple aspects within the mitotic roles of Polo-like kinase: centrosome separation and maturation, mitotic entry and metaphase to anaphase transition and mitotic exit. Most of the reported putative targets are involved in metaphase-to-anaphase Transition (CDC20, PRC1, CCNB1, CDC2 and PTTG1). The figures display molecules involved in the respective pathways. The molecules that are downregulated by miR-494 and therefore represent putative targets are depicted in gray color.

Figure 2. Genes with decreased expression upon miR-494 stimulation are involved in the G₂/M checkpoint regulation. MiR-494 downregulates genes involved in the regulation of transition from G₂ to M, including TOP2A, CDC2 and CCNB1. The figures display molecules involved in the respective pathways. The molecules that are downregulated by miR-494 and therefore represent putative targets are depicted in gray.

To obtain experimental validation of the effects of miR-494 on cell cycle progression from S to the end of M, we utilized FACS analyses. Cell ploidy, however, is unchanged from the end of S to the beginning of cytokinesis; therefore, we sought to identify additional markers that can differentiate between successive stages in this progression. Since most miR-494 putative targets involved in G₂/M arrest were found in the metaphase-to-anaphase transition stage, we sought a cellular marker that displays changing expression at this point of the cell cycle. Cyclin B1 was previously reported to display increasing levels starting late in the synthesis (S) phase of the cell cycle and to drop precipitously, due to degradation, at the transition between metaphase and anaphase.¹⁴ Therefore, double staining with propidium iodide (PI) and CCNB1 allows for quantifying cell distribution in two compartments: (1) from the end of S to the metaphasic plate (4n cells as determined by PI staining and rising CCNB1) and (2) from the metaphasic plate to cytokinesis (4n cells as determined by PI staining and negative expression of CCNB1), as previously described.¹⁴ For this experiment, we have employed HuCCT1 cells that were infected with MSCV-IRES-enhanced-GFP carrying miR-494 (MIEG3–494V) or with a negative control (MIEG3-EmptyVirus, or MIEG3-EV), as previously described.¹³ HuCCT1-MIEG3–494V cells stably express miR-494 at approximately 2.5-fold more than HuCCT1-MIEG3-EV. Of note, the level of miR-494 in HuCCT1-MIEG3–494V is approximately the same as in normal, non-cancerous tissue.¹³ To stimulate cells to arrest in mitosis, a low dose of vinblastine [100 nM (nM) for 2 h] was used. We found that cells expressing miR-494 at higher levels (HuCCT1-MIEG3–494V) displayed a higher percent of cells in G₂/M, consistent with a reinforced checkpoint (Fig. 3). Thus, we determined that cells expressing higher levels of miR-494 have reinforced cell cycle checkpoints in the progression from S to the end of M. To identify the precise checkpoint that is reinforced by miR-494, we determined the ratio of the number of cells in G₂-metaphase (4n and CCNB1-positive) and metaphase-cytokinesis (4n and CCNB1-negative), respectively, to the number of cells in G₂/M (4n). Of the total number of cells in G₂/M, the G₂-metaphase compartment contained significantly more cells in HuCCT1-MIEG3–494V than in HuCCT1-MIEG3-EV (74.91% ± 2.8 vs. 64.38% ± 3.9, p-value < 0.0001, unpaired Student t-test, results of three independent experiments, each with four replicates per condition, for a total of 12 replicates per condition). A representative analysis for one of the replicates is shown in Figure 3. These data confirmed our hypothesis that miR-494 induces cancer cell arrest in the G₂-metaphase compartment.

Figure 3. MiR-494 induces arrest in G₂-metaphase. The upper two panels display flow cytometry data for HuCCT1-EV (cells expressing baseline miR-494 levels), and the lower two panels display flow cytometry results for HuCCT1–494V (cells expressing 2.5 fold higher levels of miR-494). The left-most panels display cell distribution following staining with PI (X axis) and Cyclin B1 (Y axis). Cyclin B1 increases steadily from G₀/G₁ to metaphase, then its levels drop precipitously. The cell population that is tetraploid and has high CCNB1 is in G₂-metaphasem and the cell population that is tetraploid but expressed low CCNB1 is in metaphase-cytokinesis. The right-most panels display only tetraploid cells after gating in the left-most panels. The figure shows that while 68.8% of cells are in G₂-metaphase in HuCCT1-EV, approximately 78.14% of cells are in the same phase in HuCCT1–494V (p-value < 0.001, unpaired Student t-test).

To obtain independent evidence of miR-494-induced cell cycle arrest in G₂/M, as well as to gain further mechanistic insights, we performed live cell differential interference contrast (DIC) microscopy. HuCCT1-MIEG3–494V and HuCCT1-MIEG3-EV cells were plated the day before the experiment, and the media was changed with media containing 100 nM vinblastine immediately prior to recording time-lapse images. A total number of 10 time-lapse movies were recorded for HuCCT1-EV and HuCCT1–494V, respectively. A total of 22 cell divisions were observed for HuCCT1–494V and 19 cell divisions for HuCCT1-EV (Table S1). All movies were carefully reviewed, and the time between the disappearance of nucleoli (Ns) until the metaphasic plate was first seen was recorded (MP, Fig. 4; and Movies S1 and S2, demonstrating G₂-metaphase duration in HuCCT1-EV and HuCCT1–494V, respectively). This period of time corresponds, roughly, to G₂-metaphase. The time from the last image of nucleoli to appearance of metaphasic plate was 19.7 ± 8.3 min for HuCCT1-MIEG3-EV and prolonged to 39.8 ± 13.0 (mean ± SD) minutes for HUCCT1-MIEG3–494V (p < 0.0001, Student unpaired t-test). These data suggest that cells expressing miR-494 at levels similar to normal tissue (HuCCT1-MIEG3–494V) spend approximately twice as long in G₂/M compared with cancer cells expressing lower levels of miR-494. These data are consistent with the results obtained by FACS and double staining with PI and CCNB1 and suggest a reinforcement of checkpoints within G₂-metaphase induced by miR-494.

Figure 4. DIC images illustrate cell transition from the end of G₂ to cytokinesis. DIC images were obtained for HuCCT1-EV and HuCCT1–494V, respectively. The following times were recorded for every cell that divided during the time-lapse experiment: the last image of nucleoli (A), the disappearance of nucleoli (B), first appearance of metaphasic plate (C), daughter chromosome separation (D) and the start of cytokinesis (E). The time between disappearance of chromosomes and metaphasic plate was recorded for every dividing cell and compared between HuCCT1-EV and HuCCT1–494V. N, nucleus; Ns, nucleoli; -Ns, lack of nucleoli; MP, metaphasic plate.

We determined that reinforcing miR-494 expression downregulates target genes within G₂/M checkpoint regulation pathway, as well as within “mitotic roles of Polo-like kinase” pathway. Next, we set to further delineate the precise effects of miR-494 on its downstream targets. Therefore, we examined the mRNA species reported to be downregulated by miR-494 within these pathways. We reasoned that this approach complements well target identification through strategies employing search engines, since miRs act predominantly through decreasing target mRNA levels, as opposed to translation inhibition.⁹ As seen in Figure 5, reinforcing the expression of miR-494 results in decreased protein levels of Polo-like kinase 1 (PLK1), pituitary tumor-transforming gene 1 (PTTG1), Cyclin B1 (CCNB1), cell-division cycle 2 (CDC2), cell-division cycle 20 (CDC20), and topoisomerase II alpha (TOP2A). We therefore concluded that treatment of cancer cells with miR-494 reinstates checkpoints within G₂/M through the coordinated downregulation of PLK1, PTTG1, CCNB1, CDC2, CDC20 and TOP2A, resulting in delayed cell cycle progression.

Figure 5. (A) Protein expression of miR-494 target genes decrease upon miR-494 stimulation. Representative western blots for CCNB1, CDC20, PLK1, PTTG1, TOP2A and CDC2 are shown. Equal protein loading was performed, as shown by β-actin. The figure also includes densitometric analyses demonstrating a significant decrease in the protein level upon miR-494 stimulation.

To study the mechanism of miR-494 directed downregulation of its target genes, we searched for conserved binding sites in the 3′UTR of these genes by employing TargetScan (www.targetscan.org). However, we found that none of these mRNA putative targets had a conserved binding site in their 3′UTR. Since all of these putative targets display lower protein levels, but more importantly, lower mRNA levels with miR-494 treatment, we reasoned that these mRNA species might have miR-494 binding sites outside the 3′UTR region. Of note, conserved miR binding sites are as widespread in the open reading frame as they are in the 3′UTR¹⁵ and are also common in the 5′UTR regions.¹⁶ Therefore, we manually searched the sequences of every target gene for putative miR-494 binding sites. We searched for the following sequences: 7mer-1A (TGTTTCA), 7mer-m8 (ATGTTTC) and 8mer (ATGTTTCA). We identified a 7mer-1A in the ORF of PTTG1 and another 7mer-1A in the ORF of TOP2A (Fig. 6A). Luciferase reporter assay experiments were performed by cloning a portion of the ORF that contained the binding site. Cells transfected with the PTTG1 ORF fragment showed an average of 20% reduction in luciferase activity upon treatment with miR-494 compared with a non-specific microRNA mimic (NSM). In addition, cells transfected with the TOP2A ORF fragment showed an average of 32% reduction in luciferase activity upon treatment with miR-494 compared with a non-specific microRNA mimic (NSM). These decreases were statistically significant, with a p-value of less than 0.001 (Student unpaired t-test). Upon miR-494 binding site mutation, the effect of miR-494 on PTTG1 as well as TOP2A was lost, as evinced by similar luciferase activity between miR-494 and NSM-treated cells (Fig. 6B).

Figure 6. MiR-494 directly interacts with binding sites in the ORF of PTTG1 and TOP2A. (A) The seed of miR-494 is complementary to a 7mer-1A at position 473–479 in the ORF of PTTG1. (B) The seed of miR-494 is complementary to a 7mer-1A at position 1407–1413 in the ORF of TOP2A. (C and D) Y-axis: relative luminescence normalized to the luminescence level in NSM (non-specific mimic) treatment. X-axis: treatment conditions. NSM, non-specific mimic, 494M, miR-494 mimic; PTTG1 WT, correct orientation fragment of PTTG1 ORF containing miR-494 binding site; PTTG1 Mut, fragment of PTTG1 ORF containing a mutated miR-494 binding site; TOP2A WT, correct orientation fragment of TOP2A ORF containing miR-494 binding site; TOP2A Mut, fragment of TOP2A ORF containing a mutated miR-494 binding site. Shown is the standard error of the mean. MiR-494 induces a statistically significant decrease in luminescence (p-value < 0.001, Student t-test) of the forward PTTG1, as well as the forward TOP2A fragment vs. NSM.

The experiments reported in this manuscript as well as experiments reported previously¹³ suggest that miR-494 upregulation in cancer cells with reduced levels of miR-494 induces G₁ arrest as well as G₂/M arrest. In contrast, the impact of miR-494 depletion in normal cells expressing high levels of miR-494 is not known. To address this issue, we employed H69 normal cholangiocytes that were transfected with miR-494 hairpin inhibitor (miR-494Inh) and a non-specific inhibitor (NSI), respectively. Of note, miR-494 level in H69 cells is similar to its level in normal human cholangiocytes.¹³ Upon inhibition of miR-494, H69 cells displayed a G₁/S checkpoint release, as evinced by a decreased percentage of cells in the G₁ phase of the cell cycle (Fig. 7A). Furthermore, upon concomitant treatment with vinblastine, miR-494 inhibition induces G₂/M release, as evinced by a decrease of the percent of cells in G₂/M phase of the cell cycle (Fig. 7B). These loss-of-function studies are in accord with the gain-of-function studies and strongly suggest that miR-494 functions as a general cell cycle progression inhibitor exerted through checkpoints at the G₁ to S transition as well as within G₂/M. In addition, these studies suggest that miR-494 function is highly dependent on cellular context as well as external stimuli. Of interest, treatment of HuCCT1 cells with miR-494 inhibitor does not change the cell cycle distribution.¹³ We reason that since HuCCT1 cells have already low levels of miR-494, further inhibition of miR-494 does not have any impact onto cell cycle distribution. These findings argue that miR effects are exerted over a narrow miR level range. In our experiments, we set the level of miR-494 in HuCCT1 cells at 1, and found that levels less than 1 do not impact the cell cycle. Furthermore, nonmalignant H69 cholangiocytes have miR-494 levels of approximately 7. HuCCT1–494V cells, expressing miR-494 levels of 2.5 already show cell cycle profiles similar to normal cholangiocytes, and we suspect that upregulating miR-494 further would not have additional effects.

Figure 7. MiR-494 inhibition results in cell cycle release in normal cholangiocytes. (A) Flow cytometry of H69 non-malignant cholangiocytes treated with NSI (nonspecific inhibitor, left panel) or with 494 Inh (miR-494 Inhibitor, right panel). The percent of cells in G₀/G₁ decreased from 51.26% (NSI) to 45.94% (494Inh) upon treatment with miR-494 inhibitor (p-value < 0.001, unpaired Student t-test). (B) Flow cytometry of H69 non-malignant cells treated with vinblastine as well as NSI or 494Inh. The percent of cells in G₂/M decreased from 24.1% to 21.7% upon treatment with miR-494 inhibitor (p-value < 0.001, unpaired Student t-test).

Discussion

The origins of cancer chemotherapy can be traced to the discovery that nitrogen mustard acts as an effective anticancer agent.¹⁷ This discovery was followed by decades of developing empirical cytotoxic chemotherapeutics.¹⁸ Later years have brought the concept of “targeted cancer therapy,” which is based on a detailed knowledge of cancer pathways and the development of agents that can specifically target key molecules believed to be salient to cancer growth and progression.¹⁸ Another fundamental early discovery was that combination chemotherapy was more effective than single agents.¹⁹ It is therefore intuitive that therapy that combines multiple targeted cancer agents acting along the same cancer-relevant pathway would be worth exploring. Recent data suggest the miRs act predominantly by decreasing the levels of multiple target mRNAs.⁹ In addition, the effects of miRs on any given target mRNA/protein is usually modest, suggesting that miRs do not act as on-off switches on a single target, but rather as rheostats modulating multiple targets with a significant phenotypic effect.^11,12 Prior observations suggested that miRs may regulate not only a single pathway protein, but rather coordinate protein expression in a pathway-wide fashion.¹⁰ The data presented in this paper argue that miR-494, through moderate, but coordinated, inhibition of multiple mRNA species involved in the progression from G₂ to cytokinesis, induces a phenotypically significant G₂/M arrest. Interestingly, in conjunction with data presented previously,¹³ these findings suggest that miR-494 may act as a master regulator of cell cycle. Overall, we found that miR-494 regulates a total of 11 targets, of which five are involved in the transition from G₁ to S, and six are involved in the progression from G₂ to cytokinesis. Also of note, the action of miR-494 on these targets is convergent toward cell cycle arrest.

Our data suggest that miR-based therapies may offer unexpected benefits compared with efforts employing siRNA constructs targeting a single gene.²⁰ First, a coordinated action on multiple targets involved along a cancer-relevant pathway, such as progression through cell cycle, may be more difficult to evade by cancer cells. Recent experience with targeted molecules, such as PLX4032, a class I RAF selective inhibitor, demonstrated a staggering initial response rate of 80% in B-RAF-positive melanomas;²¹ however, most cancers lose response due to acquisition of novel mutations in B-RAF or activation of alternative survival pathways.^22,23 Therefore, coordinated action at multiple levels along the same canonical pathway, such as exerted by miR-494 on cell cycle progression, may offer a viable alternative to single targeted therapy. Second, since miRs are part of the cell-intrinsic physiologic mechanisms, their in vivo delivery may have less toxicity compared with other therapies. In addition, our data suggest that miR-494 acts within a narrow range, and its downregulation below or upregulation above a certain threshold does not appear to have any added and undesirable effects on cell cycle progression. Cell or tissue specificity of delivery is crucial for developing any in vivo therapies. In case of imperfect specificity, the result of upregulating miRs in normal tissues with already high levels of that particular miR may not have any effect. Our data brings further support to the parallel and convergent miR mode of action, in accord with published manuscripts.^24,25 However, further studies are needed to demonstrate or refute these hypotheses.

Perceived contradictory roles for miRs have been previously reported. For example, a recent manuscript reported that miR-494 is upregulated in retinoblastoma vs. normal retina.²⁶ Other examples include both oncogenic²⁷ as well as tumor-suppressive effects of miR-31,²⁸ miR-215^29,30 and others. PhenomiR is a miR function database that was generated by manual curation and offers facile access to manuscripts reporting miR functions in a variety of processes and organs.³¹ These previous data suggest that miR function may be highly tissue- and/or process-specific. The data presented in the current manuscript further argues that miR function may be cell context-specific. While miR-494 has globally coherent cell cycle-arrest effects, its effects are exerted predominantly on a specific phase of the cell cycle, depending on the cellular context. We showed that while H69 cells treated with a miR-494 inhibitor demonstrated a G₁ release, adding nocodazole results in a miR-494-induced predominant G₂/M release. We believe that this context-specific miR action is furthermore potentially useful in devising miR-based anticancer therapies, since cancer cells suffer significant changes in response to treatment. An miR-based therapy that continues to exert convergent anticancer effects regardless of cellular context would obviously be desirable.

Recent data suggest that miRs act predominantly through decreasing mRNA levels and significantly less through decreasing protein level in the setting of unaltered mRNA level.⁹ For this reason, we elected to search for miR-494 targets through an unbiased, mRNA-based strategy, rather than through complementarity-based online search engines. Further validating our approach, and somewhat surprisingly, of the six mRNA targets validated in this manuscript, none was reported as a miR-494 putative target by TargetScan, one of the leading search engines. The explanation rests with the fact that complementarity search engines seek miR binding sites in the 3′UTR of mRNA alone. However, as previously reported, miR binding sites are as widespread in the ORF as they are in the 3′UTR and are also common in the 5′UTR regions.¹⁶ Indeed, we identified two 7mer1A sequences in the ORF of PTTG1 and of TOP2A, and the luciferase reporter assay confirmed direct interaction of miR-494 with these sites. While no 8mer, 7mer-1A, 7mer-m8 or 8mer was identified in the ORF of CDC2, CDC20, PLK1 or CCNB1, this does not exclude the presence of other miR-responsive elements within the ORF, 3′UTR or 5′UTR of these genes. Further studies are necessary to carefully identify all miR-494-responsive elements in these genes. In addition, as previously reported for miR-24, the possibility of “seedless” miR-494-responsive elements exists as well.³²

Materials and Methods

Cell lines

HuCCT1 and TFK1 human cholangiocarcinoma cell lines were maintained in Dulbelcco’s modified Eagle media (DMEM) supplemented with 10% fetal calf serum (FCS), 1000 U/mL penicillin/streptomycin (P/S) in a humidified incubator at 37°C and 5% CO₂, as previously described.³³ H69 cells, a gift from Dr. D. Jefferson (Tufts University), are normal human intrahepatic cholangiocytes transformed with SV-40. They were derived from a normal liver prior to liver transplantation,³² and they were maintained as previously described.³⁴

RNA extraction

Total RNA was isolated with TRIzol reagent (Invitrogen) following the supplier’s protocol.

Western blotting

Cells were lysed in Laemmli sample buffer (Bio-Rad) supplemented with a protease inhibitor (SIGMA). Protein concentration was measured with a BCA protein assay kit (Pierece). Cell lysates were electrophoresed on 4% to 15% polyacrylamide gels (Bio-Rad) at 140V for 1 h and electrotransferred to Hybond-P membranes (GE Healthcare) at 80V for 2 h. The membranes were blocked with TBS containing 5% skim milk and 0.1% Tween 20 for an hour and then incubated with the primary antibody. CCNB1, CDC2, PTTG1, TOP2A, CDC20 and PLK1 antibodies (Cell Signaling Technology) were used according to the manufacturer’s protocols. The membranes were incubated after washing with the secondary antibodies, horseradish peroxidase-conjugated goat anti rabbit IgG (Calbiochem) and developed using enhanced chemiluminescence-plus reagent (GE Healthcare). Densitometry was performed by using the ImageJ software (National Institutes of Health).

Luciferase reporter assay

A portion of the ORF of PTTG1 and TOP2A, respectively, containing miR-494 binding site was amplified using linker primers containing XbaI restriction sites. Next, we employed Gene Tailor Site-Directed Mutagenesis System (Invitrogen) to introduce mutations in the miR-494 binding site. The sequences of primers are as follows: TOP2A Fw;5′-GCCTCTAGACCACTGAGTGTACGCTTA-3′, TOP2A Rv;5′-GCCTCTAGAATCTTGGTCCTGATCTGTC-3′, PTTG1 Fw; 5′-GCGTCTAGA AGCTCTGTTCCTGCCTCA-3′, PTTG1 Rv; 5′-GCGTCTAGACGACAGAATGCTTGAAGG-3′. Primer sequences for mutagenesis are as follows: TOP2A Fw;5′-AGCCAAAACTTTGGCGGCCTTGGTG-3′, TOP2A Rv;5′-GCCAAAGTTTTGGCTGAATCTCCCT-3′, PTTG1 Fw;5′-GAGAGCTTGAAAAGCGCTGGGCCCC-3′, PTTG1 Rv;5′-GCTTTTCAAGCTCTCTCTCCTCGTC-3′

Amplicons were cut with XbaI and cloned into an XbaI site just downstream of the firefly luciferase structural gene in vector pGL4.13 (Promega). After sequence verification, we obtained plasmid clones containing correctly oriented inserts. 6,000 cells per well were seeded onto 96-well plates on the day prior to transfection. Cells were transfected with miR-494 mimic (Pre-miR™ miRNA Precursor Molecules for miR-494, Ambion) or the control (Pre-miR™ miRNA Precursor Negative Control #2, Ambion). Twenty-four hours later, the constructed pGL4.13 vector and an internal control pRL-CMV (Renilla luciferase) vector were transfected. Forty-eight hours after plasmid vector transfection, the luciferase reporter assay was performed using a Dual-Glo Luciferase Assay System (Promega). Luminescence intensity was measured by VICTOR2 fluorometry (Perkin Elmer), and the luminescence intensity of firefly luciferase was normalized to that of Renilla luciferase.

Differential interference contrast (DIC) microscopy

2 × 10⁴ cells per well were plated onto the LabTek chambered #1.0 Biosilicase Coverglass System (Nunc) the day before the assay. Prior to starting the imaging, the media was replaced with media containing vinblastine 100 nM. Live images of cells were observed using in microscopy chamber placed inside an incubator unit to maintain cells at 37°C, 5% CO₂ and optimal humidity. The time from last image of nucleoli to forming metaphasic plate was measured in each cell.

Cell cycle analysis by flow cytometry

Flow cytometric analysis of DNA content was performed to assess cell cycle phase distribution. In all, 2 × 10⁵ cells per sample were used for assay. Cells were washed with PBS (Cellgro) and fixed with ice-cold 70% ethanol overnight. After washing again, cells were incubated with staining buffer containing 2mg/mL Rnase (Roche) and 0.1 mg/mL propidium iodide (Roche) with 400 μL of PBS. For double staining with PI and anti-CCNB1 antibody, anti CCNB1 antibody was added and incubated for 2 h, and then AlexaFlour 633, goat anti-mouse IgG (Invitrogen) was used as a secondary antibody and incubated for 30 min. The DNA content was analyzed using FACS Calibur (BD Biosciences) and Cell Quest software (BD Biosciences) for histogram analysis. Vinblastine treatment, where applicable, was performed 2 h prior to harvesting cells, at a final concentration of 100 nM. Attached cells and cells in the medium were collected and processed as above.

Retroviral vectors, viral supernatant production and viral transduction

MSCV-based bicistronic retroviral vectors MIEG3³⁵ were used to express miR-494. The genomic DNA sequence from -80 to +80 of miR-494 was amplified using PCR primers flanked by EcoRI (5′) and XhoI (3′) and cloned into the multiple cloning site of MIEG3. The expression of miR-494 was linked with expression of enhanced green fluorescence protein (eGFP) via internal ribosome entry site 2 (IRES2). The plasmid DNA was used to generate viral supernatant from 293T Phoenix-gp cells as previously described.³⁶ Briefly, Phoenix-gp cells were grown to 70% confluence in a 10 cm tissue culture treated dish (Corning, Inc.). Eight micrograms (μg) of plasmid DNA of interest together with 10 μg MLV gag-pol plasmid and 3 μg VSVG envelope plasmid were co-transfected using Lipofectamine 2000 (Life Technologies) following the manufacturer’s protocol. The cells were then incubated at 37°C in 5% CO₂. Eight milliliters of viral supernatant was collected every 24 h and stored at -80°C until used. To stably express miR-494 via retroviral mediated gene transfer, 1x10⁵ HuCCT cells were plated in each well of a 6-well plate and grown in DMEM overnight at 37°C and 5% CO₂. Subsequently, they were incubated with 3 mL of viral supernatant containing 8 mg/mL of hexadimethrine bromide (Polybrene, Sigma-Aldrich). After 6–8 h, the viral supernatant was discarded, and fresh DMEM was added. Two days after transduction, cells were harvested and sorted for eGFP expression using a fluorescence activated cell sorter (FACSVantage SE DiVa, Becton Dickinson).

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Supplemental Materials

Supplemental materials may be found here: www.landesbioscience.com/journals/cc/article/21105