Regulation of D-cyclin translation inhibition in myeloma cells treated with mTOR inhibitors: Rationale for combined treatment with ERK inhibitors and rapamycin

Myeloma cell lines and transfections

Human U266 multiple myeloma isogenic cell lines were transfected as previously described (6). Briefly, the constitutively activated myristoylated-AKT (AKT) cDNA expression vector was purchased from Upstate (Charlottesville, VA). Stable transfection of U266 cells with activated AKT or empty vector (EV) control was accomplished by electroporation (230 V for 25 msec). Stable transfections were selected in neomycin (350 mg/ml) and successful transfection was determined by Western blotting with antibodies specific for total and phosphorylated AKT (Ser473). Transient transfections of cells with constitutively activated cDNA MEK1 expression vector (Upstate) or empty vector control was performed using the AMAXA Nucleofection System (AMAXA Inc, Gaithersberg, MD). The transfection efficiency was typically >80% as determined by tranfection of cells with a green fluorescent protein plasmid vector. Successful transfection was determined by Western blotting with antibodies specific for phosphorylated ERK, a downstream substrate for MEK1.

Adenovirus expressing wild-type PTEN has been described previously (11). The genome of this vector has deletions of the El and E3 regions and protein IX gene and expresses its transgene under control of the human cytomegalovirus immediate-early promoter/enhancer. As corresponding control, we used the same virus containing enhanced green fluorescent protein (EGFP) as its transgene. OPM-2 cells were transduced with adenovirus for 2 hours with an MOI of 100 (1). Virus containing solutions were removed after 2 hours and the cells were incubated in the presence or absence of rapamycin for an additional 48 hours.

Reagents

All antibodies were purchased from Cell Signaling Technology (Beverly, MA), except the cyclin D1 antibody (BD Biosciences, Palo Alto, CA). The cyclin D1 antibody was specific for D1 and did not cross-react with D2 or D3. The ERK kinase assay kit was purchased from Cell Signaling Technology. Rapamycin was purchased from Calbiochem (San Diego, CA). The CCI-779 was provided by Wyeth-Ayerst (Pearl River, NY). The MEK inhibitor, U0126, and the p38 inhibitor, SB202190, were purchased from Promega (San Louis Obispo, CA). For testing cell survival, viability was determined by MTT assays as previously described (12).

Animals

Four-six week old male NOD/SCID mice were obtained from Jackson Laboratories (Bar Harbor, ME). The mice were maintained 4/cage in pathogen-free conditions. All animal studies were conducted in accordance with protocols approved by the Animal Research Committee of the West Los Angeles Veterans Administration Medical Center.

Xenograft Model

We used the murine myeloma xenograft model of Leblanc et al (13) with minor modifications as preciously described (6). The cell lines were mixed with matrigel and were then injected subcutaneously (200 μl/mouse containing 3 × 10⁷ U266 cells). Mice were randomized into drug treated or control groups (8–14 mice/group) when the tumor volume reached approximately 200–400 mm3. CCI-779 was injected IP daily X5, followed by 2 days of no drug and then 5 additional injections as previously described (6). Mice were routinely euthanized when tumors reached >2000 mm3 in volume.

Immunohistochemistry

At day +13 mice were euthanized with CO2 and the tumor mass was excised. The tumor was bisected using a razor blade: one half of the tumor was immediately placed in 10% buffered formaldehyde overnight, and the other half was frozen for protein extraction. Formaldehyde fixed tumors were embedded in paraffin and cut into 5 μm-thick serial sections using standard histological procedures. Immunohistochemical staining with anti-human Ki-67 antibody was conducted using standardized automated methods (14). Sections were counterstained with hematoxylin/eosin. The proliferation index was determined by assaying the area of Ki-67 staining using the MetaMorph software from 10 randomly selected fields at 20× magnification as previously described (6). Tumor nodules were primarily composed of tumor cells with approximately 2% of the area made up of vasculature (as determined by staining with anti-CD34 antibody) as reported in a previous study (15). In Ki-67 stained sections, stroma and vasculature (as determined by microscopy) were excluded from the analysis.

Western Blot Analysis

Protein was extracted in ice-cold lysis buffer and 25 μg of protein from each sample were boiled and separated by 10% SDS-PAGE. After blocking in 10% non-fat dried milk, the membranes were washed and then incubated with primary antibodies for 1 hr followed by incubation with 1 μg/ml horseradish peroxidase-labeled secondary antibodies.

Cell Cycle Analysis

Cell cycle analysis of hypotonic propidium iodide (PI)-stained cells was determined by fluorescence-activated cell sorting (FACS) with a Becton-Dickinson FACScaliber. Histograms generated by FACS were analyzed by ModFit Cell Cycle Analysis Software (Verity, Topsham, ME) to determine the percentage of cells in each phase.

Polysome Analysis

Extraction and display of polysomes was performed as previously described (9). Briefly, U266 isogenic cells (4×10⁶ cells) were lysed in ice-cold lysis buffer supplemented with 100 μg/ml cycloheximide at 4°C. Following removal of nuclei and mitochondria, supernatants were layered onto 15–50% sucrose gradients and spun at 38,000 RPM for 2 h at 4°C in a SW-40 rotor (Beckman Instruments). The gradients were then fractionated into eleven 1 ml fractions using an ISCO Density Gradient Fractionator at a flow rate of 3 ml/min. The polysome profile of the gradient was determined by UV absorbance at 260 nm. The RNA from individual fractions was extracted using phenol/chloroform and precipitated in EtOH. The RNA was then processed for Northern blots. Radioactive probes were generated using the Ambion MaxiScript kit (Ambion, Austin TX).

Dicistronic reporter assay

The parental dicistronic reporter construct utilized in these studies was pRF (a kind gift of A. Willis, University of Leicester, UK) (16). The pRCD1F construct, which contains the cyclin D1 5′-UTR (GenBankTM accession number NM053056) and the pRp27F construct, which contains the 365-nucleotide IRES sequence (17) of p27Kip1 subcloned into the intracistronic region of pRF as previously described (18). Cells were transfected with 25 μg plasmid DNA by electroporation. The cells were then incubated with or without drugs for 18 hours, washed twice in PBS and lysed in passive lysis buffer (Promega). The firefly and Renilla luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega). Transfection efficiency was measured by β-galactosidase activity using a β-galactosidase enzyme assay system (Promega).

Statistics

Student t-test was used to determine significance of differences between groups.

AKT regulates the anti-proliferative response to mTOR inhibitors in vitro and in vivo

In previous studies, U266 multiple myeloma cells, stably expressing an activated AKT gene, were significantly more sensitive than empty vector control U266 cells to the in vivo anti-tumor effects of CCI-779 in a xenograft model (6, 15). In the current study, we used the same isogenic pair of myeloma cell lines to investigate the mechanisms of AKT’s regulatory effects on rapamycin sensitivity. The expression of myristoylated AKT, constitutively phosphorylated on both threonine 308 and serine 473 residues is demonstrated in Figure 1A. In contrast, empty vector cells (U266EV) express an AKT molecule that is minimally phosphorylated on threonine 308 and without phosphorylation of serine 473. Cells were treated with IGF-1 (400 ng/ml) as a positive control for AKT activation and, as shown, the empty vector control cells were capable of AKT phosphorylation on both residues when stimulated by IGF-1. As shown in Figure 1B, AKT clearly regulated the in vitro anti-proliferative effect of rapamycin as measured by its effects on cell number (top panel) and the MTT assay (bottom panel). For the AKT-transfected cells in the MTT assay (bottom panel, open circles, U266AKT), the ED50 was approximately 1nM, while the ED50 was >100 nM in the EV-control transfected cells (closed circles, U266EV). As has been reported in previous studies, flow cytometric analysis (not shown) confirmed that rapamycin did not induce significant apoptosis in either isogenic cell line (1).

As an in vivo correlate of the above in vitro MTT proliferation data, we also analyzed the Ki-67 proliferation marker in U266 tumors harvested from CCI-779 treated mice. The cell lines were injected sub-cutaneously into NOD/SCID mice and, when the tumors reached 200–400 mm3 of volume, some mice were treated with 10 days of increasing doses of CCI-779 by IP injections (6). Tumors were then excised and immuno-stained for Ki-67 expression. As shown in Figure 1C, CCI-779 had a much greater inhibitory effect on Ki-67 staining in AKT-transfected tumors (dark bars) (ED50 = 10mg/kg) compared to EV controls (open bars) (ED50 >20 mg/kg). At all concentrations of CCI-779, there was no significant effect on Ki-67 expression in the EV-transfected tumor cells. Thus, AKT regulates the cell cycle inhibitory effect of mTOR inhibitors in this myeloma model in vivo as well as in vitro.

We considered the possibility that an AKT-dependent regulation of sensitivity to mTOR inhibitors might be due to differences in the ability of rapamycin or CCI-779 to inhibit the mTOR pathway. To examine this possibility, we assayed the phosphorylation of the mTOR substrate, P70 S6 Kinase (P70), after exposure to increasing concentrations of rapamycin. Figure 1D demonstrates representative results from one of three independent experiments where phosphorylation of P70 was analyzed. Interestingly, although expressing little activated AKT (6), empty vector-transfected (U266EV) cells constitutively express significant amounts of basal phosphorylated P70 (Fig 1D). This may be due to the IL-6 autocrine nature of these cells (19) that results in several activated signal cascades, some of which, may induce down stream phosphorylating effects on P70, independent of AKT (20). The levels of P70 phosphorylation were only slightly more upregulated in the AKT-transfected U266AKT cells (Fig 1D). Nevertheless, in the presence of rapamycin, a comparable degree of inhibited P70 phosphorylation occurs between the two U266 cell lines. As shown in Figure 1D, rapamycin induced approximately 50% inhibition at 0.1 nM rapamycin and >95% inhibition at 1 nM in both cell lines. Thus, these results rule out the possibility that differences in the ability of rapamycin to curtail mTOR activity can explain the AKT-dependent differential sensitivity of these multiple myeloma cell lines to growth inhibition.

AKT activity regulates effects on cyclin D1 expression and Rb phosphorylation

In previous studies in other tumor models (8–10), downregulation of D-cyclin expression induced by mTOR inhibitors played an important role in the observed cell cycle arrest. Thus, we evaluated the effects on D-cyclins in U266 cells. U266 cells predominantly express cyclin D1 (secondary to a t(11;14) translocation of cyclin D1) to drive G1-S transit (21, 22). As shown in Figure 2A (one of three separate comparable experiments), rapamycin was considerably more potent at inhibiting expression of cyclin D1 in U266AKT cells, decreasing expression by almost 50% at the lowest concentration tested (0.1 nM) and by >75% at 1 nM (bar graph below immunoblots). In contrast, rapamycin had little consistent effect on cyclin D1 expression in control U266EV cells (left panel, Fig 2A). Basal levels of cyclin D1 (in the absence of rapamycin) were comparable between the two cell lines, indicating that the sensitization to mTOR inhibitors caused by heightened AKT activity is not due to a higher baseline level of an mTOR-cyclin D1 pathway that is more important to growth of “high-AKT” myeloma cells.

To test if a differential effect on cyclin D1 expression was also present in the in vivo xenograft model, protein was extracted from tumors excised after the last CCI-779 treatment, followed by immunoblot assay for cyclin D1 and actin expression. As shown in Figure 2B, a similar AKT dependent effect on cyclin D1 expression was identified following in vivo treatment. CCI-779 effectively inhibited cyclin D1 expression in AKT-transfected tumors when used at 10 or 20 mg/kg mouse but similar doses had little effect on expression in EV-transfected tumors. Data shown are for pooled tumor lysate of 4 mice/group. The ratio of cyclin D1/actin expression measured by densitometry is shown below the representative western blot.

Cyclins interact with CDKs to phosphorylate Retinoblastoma protein (Rb). Since D-cyclins are the major determinants of CDK4/6 holoenzyme activity, decreased expression of cyclin D1 should prevent Rb phosphorylation in an AKT-dependent fashion. To test this notion, we also immunoblotted Rb, assaying its phosphorylation state in isogenic U266 cells treated with rapamycin in vitro (Fig 2C). Although some U266 clones have loss of Rb expression (23), our U266 lines as well as others (24) do express Rb. As shown, phosphorylation of Rb residues (807/811 and 780) were inhibited in AKT-transfected U266AKT cells compared to empty vector controls (U266EV). This was best seen when testing phosphorylation at 807/811 residues where 0.1 nM rapamycin inhibited phosphorylation >50% and 10 nM completely ablated phosphorylation in AKT-transfected cells. In contrast, rapamycin had little effect on Rb phosphorylation at the 807/811 residues in the U266EV cells. In experiments not shown, the ability of rapamycin to increase expression of the p27 CDK inhibitor was not AKT-dependent with approximately a doubling of p27 protein levels in both cell lines at 0.1–10 nM concentrations. Thus, AKT activity determined the ability of mTOR inhibitors to down-regulate cyclin D1 expression and Rb phosphorylation while having no effect on upregulation of p27.

AKT regulates effects on cyclin D translation

In other models, inhibition of mTOR causes an immediate downstream dephosphorylation of p70S6 kinase and 4E-BP1, which results in depressed translation of cell cycle proteins such as D-type cyclins (5). Thus, we tested for possible AKT-dependent effects on D-cyclin translation using the ribosomal loading assay described by Zong et al (25). This assay is based on the fact that actively translated mRNA species are associated with multiple ribosomes (polysomes), while mRNAs sequestered with single ribosomes (monosomes) are inactively or poorly translated (25). Following separation of polysomes from monosomes, a Northern blot analysis performed on their associated RNA can then estimate translational state, defined as the ratio of polysome-associated RNA to monosome-associated RNA signal intensity. The two U266 isogenic cell lines were thus treated with or without 10 nM rapamycin for 48 hours and the cellular extracts prepared for polysome analysis as previously described (9). The polysomes and monosomes were separated by sucrose gradient centrifugation and fractionated into eleven 1-ml fractions. The UV absorbance (254 nm) of each fraction was measured to generate a polysome profile of the sucrose gradient that was used to differentiate between the polysome (fractions 1–3) and monosome (fractions 4–11) fractions.

The Northern blot analysis of cyclin D1 or actin mRNA on total mRNA isolated from the corresponding fractions is shown in Figure 3A. The relative amounts of cyclin D1 mRNA in polysome versus monosome fractions were determined by densitometric analysis of equally exposed autoradiographs and the percent polysome RNA is shown to the right of each Northern blot. As can be seen, rapamycin (RAPA) at 10 nM was very effective in inhibiting translational efficiency of cyclin D1 in U266AKT cells (right panel), decreasing the percent of cyclin D1 mRNA associated with the polysomal fractions from 55% to 36%. As expected, rapamycin also effectively inhibited translational efficiency of the control actin RNA in these cells (decrease from 60% to 26%, lower right panel). In contrast, the empty vector-transfected U266EV cell line (left panel), expressing quiescent AKT, was completely resistant to rapamycin in terms of inhibition of cyclin D1 translation, with an actual increase in polysomal-associated RNA from 45% to 61% after rapamycin exposure. However, the drug was still able to significantly inhibit actin translation in this latter cell line (from 72% to 50% as shown). Thus, mTOR effects on translational efficiency of cyclin D1 are AKT-dependent, while those on actin translation are not. Rapamycin prevented translation in AKT-over-active cells but translation is maintained in “low-AKT” cells. This experiment was repeated with similar results. The results of translational efficiency are summarized in Figure 3B with the data representing the mean ± SD of the experiments.

AKT-dependent effects on cyclin D1 IRES activity

When mTOR inhibitors prevent cap-dependent translation, an alternative mechanism of translation is cap-independent where the 40S ribosomal subunit binds a transcript via an IRES in the RNA’s 5′-UTR. As AKT activity could regulate the ability of MM cells to maintain cyclin D1 translation and expression in the face of mTOR inhibition, we theorized that the kinase achieved this by its effects on IRES activity. The cyclin D1 transcript has a previously described IRES sequence in its 5′-UTR (18). Thus, to test our hypothesis, we utilized the dicistronic reporter vector in which the cyclin D1 5′-UTR, containing its IRES, was subcloned into the intracistronic space between the Renilla and firefly luciferase open reading frames (ORF) in the parental pRF vector (Fig 3C). This generated the pRCD1F vector whose firefly luciferase translation would be driven by the 5′-UTR. As an additional control, the 5′-UTR of p27Kip1CDK inhibitor was also subcloned into the intracistronc space to generate pRp27F (Fig 3C). Prior work confirmed the presence of an IRES sequence in the p27 5′-UTR (17). Following transfection of isogenic U266 MM cell lines with the dicistronic vectors, the reporter constructs were tested for their ability to direct 5′-UTR-mediated cap-independent translation (firefly luciferase). Results were normalized for transfection efficiency by co-transfection with a β-galactosidase construct. Transfected cells were treated with or without rapamycin for 18 hours.

Figure 3D depicts reporter expression as a ratio of firefly/Renilla luciferase activity where the activity in the pRF vector is normalized to a value of 1. As shown in the figure, the presence of the cyclin D1 5′-UTR IRES (pRCD1F) enhanced reporter expression >7 fold in empty vector-transfected cells (open bars) but only 1.8 fold in the AKT-transfected cell line (closed bars), demonstrating that the over-active AKT had a significant inhibitory effect on cyclin D1 IRES activity (p<0.05). The Renilla luciferase activities were similar for both cell lines, indicating that cap-dependent translational mechanisms were independent of AKT activity and the marked change in firefly to Renilla ratio was completely due to increased firefly luciferase expression. In contrast to the above, although the p27 5′-UTR of pRp27F vector also demonstrated significant IRES activity (approximately 6 fold over pRF), this was not AKT dependent as activity was comparable between the two isogenic MM cells lines. This indicates the specificity of the AKT regulating effect on cyclin D1 IRES function. When cyclin D1 IRES activity was measured in the cells following exposure to rapamycin, Renilla luciferase expression was decreased 30–40% in both cell lines attesting to the ability of rapamycin to inhibit mTOR and subsequent cap-dependent translation. The firefly luciferase expression was also slightly decreased in both cell lines such that the firefly-to-Renilla ratios shown were minimally altered. Importantly, the AKT-dependent differences in cyclin IRES activity was maintained between the isogenic cell lines following exposure to rapamycin. The experiment shown in Figure 3D was repeated three times with similar results.

Studies in a second myeloma cell model

To rule out that the above results in isogenic U266 cell lines were a peculiarity of that particular MM line, we also studied the OPM-2 myeloma cell line. In our previous study (6), PTEN-null OPM-2 express hyperactive AKT and OPM-2 tumor grown in NOD/SCID mice is extremely sensitive to CCI-779, with a remarkable decrease in cyclin D1 expression observed even at the lowest CCI-779 dose (0.4 mg/kg) tested. In order to test whether AKT activation was responsible for this hypersensitivity to mTOR inhibition, we transiently transfected OPM-2 cells with wild type PTEN using an adenovirus vector. At an MOI of 100:1, 90% of cells were successfully transduced. Figure 4A demonstrates that adenovirus-mediated restoration of PTEN activity in OPM-2 cells results in a marked downregulation of phosphorylated AKT (ser 473) after 24 hours compared to OPM-2 cells infected with control empty vector (EV) virus. We next tested if downregulating AKT activity in OPM-2 cells affected cyclin D1 protein levels and IRES activity. As demonstrated in figure 4B and 4C, while cyclin D1 expression in PTEN-null OPM-2 control cells (Fig 4B, left panel) is extremely sensitive to treatment with rapamycin, with inhibition of protein expression evident after treatment with 1 nM rapamycin, there is little effect on wild-type PTEN (right panel). In similar fashion, the dicistronic reporter expression (Fig 4C) demonstrates a significant increase in IRES activity in OPM-2 cells transfected with wild type PTEN to down-regulate AKT activation (p<0.05). These data collectively provide support that the AKT-dependent regulation of D-cyclin translation and rapamycin responsiveness in U266 cells is not restricted just to that cell line.

An ERK inhibitor curtails D-cyclin IRES activity and sensitizes resistant “low AKT” MM cells to rapamycin

Both ERK and p38 MAPK activity can facilitate IRES function of some transcripts, like c-myc (26, 27), and it is known that AKT can potentially down regulate ERK (28) and p38 (29) activity. Thus, we tested the notion that AKT’s regulatory role could be mediated via effects on these MAP kinases. In the U266 isogenic MM clones, AKT also regulated these MAPK pathways. As shown in Figure 5A, although the expression levels of total ERK and p38 were comparable between the two cell lines, the levels of phosphorylated ERK and phosphorylated p38 were significantly down regulated in the AKT-transfected cells with complete ablation of ERK phosphorylation and an approximate 60% decrease in p38 phosphorylation. Thus, we next investigated whether AKT regulates cyclin D1 IRES activity via these inhibitory effects on the p38 or ERK pathways.

The dicistronic reporter assay was again used in MM cells treated with either a p38 or MEK/ERK inhibitor. After transfection, U266EV cells were cultured for 18 hrs without or with U0126 (a MEK inhibitor) or SB202190 (a p38 inhibitor). Our results shown in Figure 5B, demonstrate that treatment with the MEK/ERK inhibitor, but not p38 inhibitor, significantly diminished cyclin D1 IRES activity (measured by firefly luciferase activity) in these cells (p<0.05). In contrast, there was no significant effect of the ERK inhibitor on Renilla luciferase expression. In comparison, U266AKT cells (right side of Fig 5B) treated with U0126 demonstrated a complete abrogation of the relatively small cyclin D1 IRES activity in these cells (from 1.6 fold activity decreased to 1 fold). The ability of the MEK inhibitor to abrogate concurrent ERK activity in the empty vector cells is shown in Figure 5C. This inhibitor completely ablated ERK in vitro kinase activity in these cells. Although the p38 inhibitor used at 25 μM, had no effect on IRES activity, it was capable of significant inhibition of p38 phosphorylation (data not shown).

Effects of combination treatment

The above data indicated that MEK/ERK activity supports cyclin D1 IRES function and suggests that the ability of AKT to inhibit ERK played a role in its regulatory effects on IRES activity and subsequent sensitivity to mTOR inhibition. This suggested that ERK activity was an important determinant of the ability of rapamycin-resistant MM cells to prevent D-cyclin down regulated expression and G1 arrest. To test this hypothesis, the isogenic cell lines were exposed to rapamycin or the U0126 MEK/ERK inhibitor and cell cycle analysis assessed (Fig 6A). In these experiments, we used the MEK/ERK inhibitor at 1 μM, a concentration which inhibited ERK activity by >80% (Fig. 5C), because there was some modest cytotoxicity when U0126 was used alone at 10 μM. As shown in Figure 6A, although the EV-transfected myeloma cells were relatively resistant to the effects of either rapamycin or the ERK inhibitor used alone, exposure to the combination of drugs significantly induced G1 arrest (p<0.01). There was no detectable apoptosis resulting from combined therapy. The successful sensitization of the “low AKT” MM cells to G1 arrest mirrored the successful down regulation of cyclin D1 protein expression when combination therapy was used (Fig 6B). In contrast to this sensitization, the ERK inhibitor did not cause an increase in G1 arrest when it was used in combination with rapamycin in the “high-AKT” MM cell line (data not shown). This result was expected as these latter cells have barely detectable ERK activity (Fig 5A). The results also argue against a non-specific G1 toxic effect of the ERK inhibitor when used in combination.

The inhibitory effect of rapamycin on p70S6 kinase activity and the subsequent interference with a p70-induced negative feedback activity on IRS-1, with resulting PI3-K/AKT activation, was previously characterized in multiple myeloma cell lines (30). It is also known that ERK activity can participate in p70S6K activation (20). Thus, one possible explanation for the sensitizing effect of the ERK inhibitor was its ability to activate AKT (mediated via p70 inhibition). Theoretically, it would convert the “low AKT’ resistant cell into a “high AKT” sensitive cell. Therefore, we tested effects of the ERK inhibitor on AKT phosphorylation in EV-transfected U266 cells and compared them to effects of rapamycin. As shown in figure 6C, the empty vector U266 line demonstrated a significant upregulation of AKT phosphorylation on both serine 473 and threonine 308 when it was exposed to either the U0126 MEK/ERK inhibitor or rapamycin. The effects were more marked for the serine 473 residue than threonine 308. The ability of each inhibitor to increase AKT phosphorylation when used alone was comparable. There was no additive enhancement of phosphorylation when both were used together. We believe that this effect on AKT can be explained by both inhibitors’ ability to down-regulate p70 with the subsequent interruption of the negative feedback loop. Similar treatment of AKT-transfected U266 cells with the inhibitors did not increase AKT phosphorylation (not shown), probably because the AKT transgene is already constitutively activated in these cells.

Next, we asked if reactivation of the MEK/ERK pathway conferred resistance to rapamycin in “high-AKT” cell lines by enhancing cyclin D1 IRES activity. In order to test this, we transiently transfected U266 isogenic cell lines with a constitutively activated MEK1 plasmid or empty vector control and either pRF or pRCD1F dual luciferase reporter vector. Successful transfection of constitutively activated MEK1 was confirmed by Western blot for upregulated ERK phosphorylation in U266AKT cells (Western blot shown in insert). As shown in Figure 6D, upregulation of ERK activity in U266EV cells increased cyclin D1 IRES activity by about 1.5 fold. However, constitutively activated MEK1 had little impact on IRES activity in the AKT-transfected cells. Thus, although ERK activity is required for IRES function (Fig 5B), it is not sufficient. As described in Discussion, there is another potential reason for why MEK/ERK activation was not capable of enhancing IRES activity in the AKT-transfected cells.