Myeloma

The feed-forward loop between YB-1 and MYC is essential for multiple myeloma cell survival

Article metrics

Abstract

Y-box binding protein 1 (YB-1) functions as a translational regulator and has been suggested to elevate MYC mRNA translation via an internal ribosome entry segment (IRES) point mutation in multiple myeloma (MM). We show that YB-1-mediated translation of MYC mRNA occurs independently of the reported IRES mutation, as 87 MM patients (n=88) and all tested human MM cell lines (HMCLs) were negative for the mutation. We show for the first time that positive MYC staining predicts YB-1 co-expression in malignant plasma cells and YB-1/MYC co-expression increases from 30% in medullary to 70% in extramedullary MM. YB-1 knockdown in HMCLs reduced both MYC protein levels and MYC mRNA in the polysomal fraction, providing a mechanism by which YB-1 controls MYC translation. MYC transcription of YB-1 is demonstrated in HMCLs as MYC knockdown resulted in reduced YB-1 protein and mRNA levels. Furthermore, MYC activation in non-malignant mouse embryonic fibroblasts (MEFs) increased YB-1 mRNA, clearly indicating that MYC drives YB-1 transcription. Importantly, perturbation of the MYC/YB-1 oncogenic circuit leads to apoptosis in HMCLs. Here, we demonstrate that these two proteins co-regulate each other via combined transcriptional/translational activity establishing their pivotal role in MM cell survival. We therefore suggest that targeting the YB-1/mRNA interaction provides a new strategy for MM drug development.

Introduction

Multiple myeloma (MM) is a malignant haematopoietic disorder of terminally differentiated B cells that are mainly localized in the bone marrow, where they lead to bone destruction and impaired haematopoiesis. Despite the development of novel drugs (that is, immune-modulatory drugs and proteasome inhibitors), MM remains an incurable disease.1 The final stages of MM are characterized by rapid disease progression and the development of drug resistance. Therefore, novel therapies based on a better understanding of the molecular pathogenesis of MM are urgently needed.

Y-box binding protein 1 (YB-1) belongs to the family of DNA/RNA binding proteins with a highly conserved cold-shock domain.2, 3, 4 Poor patient prognosis in breast carcinoma and non-small cell lung carcinoma correlates with nuclear localization of YB-1 protein, recently identified as phosphorylated YB-1 (pYB-1). Furthermore, pYB-1 activates genes involved in breast cancer cell growth as well as in survival.5, 6, 7, 8 We have shown that cytoplasmic YB-1 is overexpressed in MM, and correlates with disease progression, cell survival and drug resistance.9 However, the mechanism by which cytoplasmic YB-1 contributes to the malignant phenotype in MM is largely unknown.

Mouse models of plasma cell (PC) dysfunction can provide in-vivo evidence of YB-1 expression and function in normal and malignant PCs. In the iMycΔEμ transgenic mice, pristane induces inflammation granulomas in the peritoneum and PC tumour (PCT) formation in all animals between 2 and 6 months.10, 11 Like iMycΔEμ, many murine PCT models depend on MYC protein expression.12 The IL-6-dependent mouse model of PCT used here, H2-Ld-hIL6, is MYC independent, here the B-cell growth and survival factor IL-6 are constitutively overexpressed and PC neoplasms occur spontaneously.10, 11, 13

MYC protein expression has been detected in MM samples and correlates with disease progression.14, 15 Indirect perturbation of MYC transcription via JQ1, a BET bromodomain protein inhibitor, is associated with cellular senescence in MM.16 YB-1 was identified as one of the three internal ribosome entry segment (IRES) trans-acting factors capable of enhancing MYC family mRNA translation, particularly in the presence of an MYC IRES point mutation in human MM cell lines (HMCLs).17, 18 The binding of YB-1 via its conserved cold-shock domain to specific mRNAs in the cytoplasm is thought to regulate their stability and accessibility for the translational machinery.17, 19, 20, 21, 22, 23 The mechanism by which YB-1 together with MYC regulates malignant PC survival is unknown. Furthermore, no data exist on their co-expression in malignant PCs.

In the current study, we show that both MM primary tissue and HMCLs lack the IRES point mutation. We describe the co-expression of MYC and YB-1 in malignant PCs and show that both proteins reciprocally regulate each other via a translational/transcriptional circuit, and show the mechanism by which YB-1 affects MYC translation. This YB-1/MYC oncogenic circuit is the first description of its kind in tumour cells and it is crucial for the survival of malignant PCs.

Materials and methods

Mice

We analysed paraffin-embedded tissue from both iMycΔEμ (n=15) and H2-Ld-HuIL-6 mice (n=12). Mice were maintained on a 12-h light cycle according to guidelines similar to Institutional Animal Care and Use Committee. For mouse embryonic fibroblast (MEF) retrieval, mice were euthanized by cervical dislocation after isofluran anaesthesia 13.5 days post coitus.

Cell culture

All products were from Life Technologies, Darmstadt, Germany unless otherwise stated. Homozygous R26MER/MER and littermate wild-type control MEFs were isolated and maintained in DMEM supplemented with Penicillin/Streptomycin and 10% fetal bovine serum (FBS).24 For MYC activation, cells were placed in low serum (0.2% FBS) overnight, stimulated with 100 nM 4-Hydroxytamoxifen (4-OHT) (Sigma-Aldrich, Taufkirchen, Germany) for the indicated durations.

HMCLs AMO-1 and MM.1S were grown in RPMI-1640 medium supplemented with 2 mM L-glutamine, 1 mM sodium pyruvate, 100 μg/ml gentamicin, 20% FBS for AMO-1 and 10% FBS for MM.1S. The human P493-6 was cultivated in RPMI-1640 and 10% FBS, 0.1 μg/ml doxycycline (Sigma) was used to repress MYC expression.

Immunohistochemistry

Slides with 4 μm thick serial sections were deparaffinized, subject to heat induced epitope retrival then blocked appropriately before the primary antibody: polyclonal human N-terminal-specific YB-1 (ABIN155053, Antibodies-Online, Aachen, Germany), and biotin-conjugated rat anti-mouse CD138 (Becton Dickinson, Heidelberg, Germany) both at 1:100, MYC (N-262) (Santa Cruz Biotechnology, Heidelberg, Germany) and at 1:50 and incubated overnight at 4 °C.25 Human tissue was stained with anti-YB-1 at 1:1000 (clone EP2708Y, Epitomics, Burlingame, CA, USA) and visualized with the CSAII kit (DAKO) as per instructions. Fluorescent secondary antibodies were Cy2, Cy3 and Cy5 (Jackson Immuno Research Laboratories, West Grove, PA, USA), and biotinylated secondary antibodies followed by streptavidin conjugated to either HRP or AP were visualized with DAB substrate and/or AP substrate kit (all from Vector Laboratories, Burlingame, CA, USA). Fluorescent-labelled slides were counterstained using DAPI (Sigma). Images were captured using a Nikon Eclipse TE 2000-U with either a DS-Qi1Mc (5.03) or Dxm1200 digital camera, with NIS-Elements BR version 3.0 software (all Nikon, Düsseldorf, Germany).

Western blot

Cells were lysed in radio-immunoprecipitation assay buffer (50 mM Tris (pH 8.0), 150 mM NaCl, 1% NP-40, 0.5% deoxycholate, 0.1% sodium dodecyl sulphate (SDS)) supplemented with protease inhibitor cocktail (Roche, Grenzach-Wyhlen, Germany) on ice for 15 min. Protein concentration was quantified using Bradford assay reagent (Bio-Rad, München, Germany). Protein lysate (20–30 μg protein/lane) was separated on SDS-polyacrylamide gels and transferred onto Immobilon-P PVDF membrane (Millipore, Schwalbach, Germany). Membranes were blocked with 5% BSA in TBS-T (20 mM Tris (pH 7.4), 0.5 M NaCl, 0.1% Tween-20) and incubated for 1 h in antibody solution (5% BSA-TBS-T). Primary antibodies used were anti-MYC rabbit polyclonal (1:500, clone # N-262; Santa Cruz Biotechnology), anti-YB-1 rabbit monoclonal (1:10.000, clone # EP2708Y; Epitomics), anti-TCTP rabbit polyclonal (1:10 000, ab37506; Abcam, Cambridge, UK). Secondary antibodies used were donkey anti-rabbit IgG-HRP and sheep anti-mouse IgG-HRP. Validation of equal protein loading via HRP-labelled β-actin antibody (1:10 000, clone # l-19; Santa Cruz Biotechnology).

FACS and caspase activity

For apoptosis detection, cells were incubated in binding buffer (1 × PBS, 5 mM EDTA, 0.5% BSA) containing 2.5 μg/ml propidium iodide, washed with PBS and evaluated by flow cytometry (CyFlow SL, Partec, Münster, Germany). Analyses using a dot plot with forward scatter and sideward scatter, followed by FL-1 vs FL-3 detected propidium iodide-positive staining. Apoptotic cells show a decrease in forward scatter and an increase in sideward scatter. The apoptotic cells were identical with the propidium iodide-positive cells. Discrimination between viable and apoptotic cells was done according to the forward scatter/sideward scatter plot.

Caspase activity was measured with the cell permeable CaspACE FITC-VAD-FMK In Situ Marker (Promega, Mannheim, Germany). The marker was incubated with 100 000 cells at a final concentration of 10 μM for 20 min before FACS analysis.

Affymetrix microarrays and data analysis

Total RNA from PCT was isolated using the RNeasy Plus Mini Kit (Qiagen, Hilden, Germany). RNA quality and mRNA enrichment in the immunoprecipitation (IP) samples were assessed using the Agilent 2100 bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). In all, 150 ng of RNA was used for cDNA synthesis followed by in-vitro transcription, labelling and fragmentation according to the GeneChip 3’IVT Express Kit (Affymetrix, Santa Clara, CA, USA). The cRNAs were subjected to either genome U133 Plus 2.0 or 430 2.0 arrays and hybridized overnight at 45 °C. Washing and scanning were performed using the GeneChip Fluidics station 450 and the GeneChip scanner (Affymetrix). PCT input RNA, IP and HMCL IP samples were processed simultaneously. Data were MAS5 normalized and ranked downwards by signal intensity. For Venn diagram, the first 1000 entries of PCT IP were pre-selected, doubles excluded and PCT input RNA signals used as control to assess transcripts enriched in IP only. Of this list, the first 500 candidates were compared with the first 1000 entries, pre-selected for unique entries of AMO-1 and MM.1S IP data using Microsoft Access. The described candidate lists of AMO-1, MM.1S and PCT IP as well as the created overlaps were annotated to pathways (C2) using the GSEA molecular signature database.26

Density gradient centrifugation

In all, 1.5 × 107 cells were washed with ice-cold PBS containing 100 μg/ml cycloheximide (CHX), pelleted (3 min, 2000, r.p.m., 4 °C), and lysed in 400 μl buffer containing 100 mM KCl, 20mM Tris (pH 7.5), 5 mM MgCl2, 100 μg/ml CHX, 0.5% (w/v) NP-40, a 1:1000 mix of protease inhibitors and 1 mM DTT. After incubation on ice for 10 min, cell debris was pelleted by centrifugation (10 000 r.p.m., 10 min, 4 °C). The supernatant was then separated by centrifugation in a 5–45% sucrose gradient (SW41Ti, 34 500 r.p.m., 2 h, 4 °C). Gradient fractions were separated by SDS-PAGE and analysed by western blot or pooled for RNA isolation (fractions 1–10, 11–23) and subsequent real-time PCR analysis.

Statistical analysis

A two-tailed Student’s t-test was applied to perform statistical analysis. Results were considered significant at P<0.05. Calculations were performed with Prism GraphPad 4.0c (GraphPad Software, La Jolla, CA, USA). All data are presented ±s.d.

Extended Material and methods description can be found within Supplementary Information.

Results

Lack of point mutation in MYC IRES for MM

It was reported that both YB-1 and the PTB-1 (polypyrimidine tract-binding protein-1) enhance MYC mRNA translation in a cap-independent manner in vitro.17 These authors described a point mutation in the MYC IRES sequence, located in exon 1 of the MYC gene (C to T transition at position 5429, RefSeqGene NG_007161.1), which enhances YB-1/PTB-1 binding to the IRES, subsequently leading to increased MYC protein synthesis.18 This mutation was found in 42% (n=19) of MM and 20% (n=10) of MGUS (Monoclonal gammopathy of undetermined significance) cases analysed.18, 27 We performed bidirectional sequencing of this MYC gene segment of CD138-positive cells from 88 primary patient samples, encompassing different genetic MM subgroups (IgH translocation, hyperdiploid cases, 13q and 17p deletions). In 87 cases, we identified the wild-type MYC sequence only. One single patient sample had the C to T transition, but the transition was also present in the germline control. Our findings (C/C=0.989 and C/T=0.011) are equivalent to the reported single-nucleotide polymorphism for this region (rs4645949) with the genotype frequency C/C=0.988 and C/T=0.012 in the NIH PDR90 population set.

Additionally, we sequenced the reported region in both HMCLs (AMO-1, MM.1S) and HEK293T cells and could not detect the C to T point mutation despite high MYC and YB-1 protein levels (Figure 4a; Supplementary Figure 3). This verifies that MYC mRNA is translated independently of the reported IRES mutation.

Immunoreactivity of MYC in malignant PCs predicts YB-1 expression

Animal models showing PC dysfunction provide an in-vivo system to uncover important biological facets of malignant PCs like MM. PCTs occur spontaneously in IL-6-ms mice, but are induced with pristane injections in iMycΔEμ mice. In the absence of pristane injections, iMycΔEμ mice develop other B-lineage neoplasms by 12 months of age. Both the IL-6-ms and the iMycΔEμ transgenic mice have been backcrossed into BALB/c allowing for the use of BALB/c as control. The distribution and frequency of PCs as identified by CD138 staining in gut was similar in untreated iMycΔEμ mice (data not shown) and wild-type BALB/c mice (Figure 1aii), suggesting a normal PC distribution and frequency of untreated mice. However, in IL-6-ms mice, we found PC expansions in all mice to varying degrees. These PC abnormalities have been placed in several different subcategories.13 For simplicity, we have condensed these categories to CD138-positive cell clusters.

Figure 1
figure1

YB-1/CD138 expression in BALB/c, iMycΔEμ and IL-6-ms PCs. (a) All mice used in this study were backcrossed into BALB/c. Therefore, we stained BALB/c intestine for YB-1 and CD138 to show the normal expression of the two proteins. Expression of YB-1 (ai) in the intestinal villi, as well as CD138 staining of PCs (aii) is shown. In the overlay shown CD138-positive cells are negative for YB-1 (aiii). (b) Staining in the iMycΔEμ mice. PC expansions as illustrated by CD138 staining (bii) were also positive for YB-1 (bi). (c) YB-1 (ci) and CD138 (cii, overlay ciii) colocalization were found in 50% of the IL-6-ms population studied. In the other half of the studied animals, we found PC expansions that were negative for YB-1 (diiii). In all cases, the scale bar represents 20 μm.

YB-1 has been detected in MM, though never documented in CD138-positive cells in animal models.9 We first analysed the normal distribution of YB-1-positive cells in areas where PC expansions are known to occur, namely lymph nodes, spleen and intestine (n=4). In BALB/c control mice and untreated PCT-free iMycΔEμ, we found weakly positive YB-1 cells in spleen using avidin/biotin amplification (data not shown) and background staining in intestinal villi (n=4) (Figure 1ai). Although there are no published reports for YB-1 staining in adult mice, our results are in agreement with the published data on human tissue.28

CD138-positive cell clusters in both pristane-injected iMycΔEμ mice and IL-6-ms were unambiguous (Figures 1b and d). In all injected iMycΔEμ mice studied, we found large clusters of PCs in the peritoneum. Strikingly, 100% CD138-positive clusters in iMycΔEμ mice showed an intense YB-1 expression (Figure 1b, n=12). Importantly, this YB-1/CD138 co-expression is absent in normal PCs (Figure 1aiii). In the IL-6-ms model, CD138-positive cell clusters were either in the lymph nodes, peritoneum or spleen. Interestingly, although all mice had CD138-positive cell aggregates, only 50% of mice studied (n=10) showed a strong correlation between YB-1 and CD138 expression (Figures 1c and d). As illustrated in Figure 1d, three of the five animals with large clusters of CD138-positive cells were YB-1 negative. Therefore, the appearance of CD138-positive cell clusters could not be used to predict the presence of YB-1 in IL-6-ms. These data represent the novel finding that YB-1 is differentially expressed in neoplastic CD138-positive B-cells in the two models.

We then asked if the genotype of the two mouse models was responsible for the difference in the YB-1 staining pattern. The iMycΔEμ mouse model was driven by MYC expression and was 100% YB-1/CD138 positive. The IL-6-ms mouse model depends on the constitutive overexpression of IL-6 and showed only 50% YB-1/CD138 correlation. We therefore examined if MYC expression was related to YB-1 expression. Importantly, normal PCs were negative for both MYC (Figure 2b) and YB-1 (Figure 1aiii). Using serial sections to study MYC/YB-1 co-expression, we show that MYC overexpression in iMycΔEμ mice is limited to CD138-positive cell expansions (Figure 2aii) and thus correlates with YB-1 staining (Figure 2ai). In the IL-6-ms animals, 100% of YB-1/CD138-positive cells were also positive for MYC, revealed using the sensitive avidin/biotin amplification and DAB substrate (Figure 2c). IL-6-ms animals with no YB-1 expression had no MYC expression in CD138-positive cells. The novel finding that MYC/CD138-positive malignant PCs co-express YB-1 independent of the genetic background supports our notion of an important link between these two proteins in mouse primary tissue.

Figure 2
figure2

Colocalization of YB-1 and CD138 predicts MYC expression. Studying the co-expression of YB-1 and MYC was not possible in the same section as both antibodies were made in rabbit. We used serial sections to study the relationship of the two proteins. (ai) iMycΔEμ mice co-express YB-1 in all CD138-positive cell expansions. The PCT model is driven by MYC overexpression, therefore there is a significant amount of MYC easily detected with fluorescence (aii). (b) Notably, there is no MYC expression in non-malignant PCs. (c) In the IL-6-ms model, only PC expansions expressing YB-1 (ci) co-expressed MYC (cii). Importantly, these animals were stained using DAB as there was no signal using fluorescence techniques. In this example, there is a Peyer’s patch (PP) directly adjacent to this PC expansion, in which there are two cells identified with a green arrow (expressing only CD138) and a red arrow (YB-1 positive) showing again that this co-expression was limited to CD138 PC aggregates, and not a ubiquitous feature of PCs in these animals. Example of extramedullary plasmacytomas from mesenteries with strong PC infiltration showed MYC (brown) and YB-1 (blue) co-expression (d) and CD138 staining (e). All CD138-positive cells express YB-1 and only a portion also expresses MYC. Insets are expansions of area marked. Examples of YB-1 and YB-1/MYC expression are identified. Arrows point to cells expressing YB-1 and arrowheads identify cells expressing both MYC and YB-1. In all panels, the scale bar is 20 μm.

Next, we tested whether MYC expression also predicted YB-1 expression in human tissue. In MM, YB-1 expression was present in 90% and MYC in 60% of all samples (20% positive MM cells). In all, 30% (6/20 cases) had 20% overlap in MYC and YB-1 staining suggesting co-expression (Supplementary Table 1; Supplementary Figure 1). Additionally, MYC/YB-1 double staining in extramedullary plasmacytomas (Figures 2d and e) showed that MM cells within all cases studied expressed YB-1, and 70% (7/10 cases) showed co-expression of MYC (20% MYC-positive MM cells; Supplementary Table 1). Therefore, YB-1/MYC co-expression increases with disease progression.

MYC mRNA co-immunoprecipitates with YB-1 protein in mouse primary tumours and HMCLs

The cytoplasmic YB-1 expression, as shown in Figures 1 and 2, suggested a role in mRNA translation in malignant PCs. To date, no data exist on the function of YB-1 for mRNA translation in primary mouse PCTs and MM. Therefore, we immunoprecipitated YB-1 bound mRNAs from a primary PCT of the iMycΔEμ mouse. Figure 3a shows the enrichment of RNAs in the range of 200–2000 nucleotides after YB-1 IP (blue input RNA vs red YB-1 bound). We subsequently analysed YB-1 bound and input RNA using whole genome microarrays and found among others YB-1 mRNA itself, α-tubulin, EEF2 and UBC mRNA enriched in the PCT IP sample, which is in agreement with published data from both in-vitro systems and cell lines.29, 30, 31 Importantly, MYC mRNA was found within the top 100 entries. Moreover, we observed a significant enrichment of several mRNAs associated with MYC pathways among the top 500 IP-specific entries (Supplementary Table 2).

Figure 3
figure3

IP of YB-1-associated mRNAs. (a) Bioanalyzer scan of the RNA samples before (blue) and after (red) the YB-1 IP of an iMycΔEμ PCT. The IP sample (red) shows an increase for smaller RNAs, for example, mRNAs. (b) Venn diagram showing shared mRNAs between IPs of iMycΔEμ PCT, AMO-1 and MM.1S. The pre-selected first 500 YB-1-associated transcripts from the PCT were compared with those from AMO-1 and MM.1S IP. Altogether, 138 transcripts are shared between the three YB-1 IP samples, with YB-1 mRNA itself as well as MYC and TCTP among them (Supplementary Table 3).

The relevance of the detected YB-1 bound mRNAs in PCT for human MM cells was evaluated via YB-1 IP in AMO-1 and MM.1S. Comparing the first 500 transcripts from PCT IP with those of AMO-1 and MM.1S, we found 138 shared transcripts with YB-1 and MYC among them (Figure 3b; Supplementary Table 3). The translationally controlled tumour protein 1 (TCTP) was present among the first 15 transcripts in all IP samples. We performed TCTP knockdown experiments to delineate its anti-apoptotic function in HMCLs and found that TCTP knockdown alone has no measurable influence on MM cell viability (Supplementary Figure 2). However, as TCTP is bound to and is translationally regulated by YB-1, we used it as an internal control in subsequent experiments. In summary, our YB-1 IP analysis in different models of PC neoplasm revealed a set of shared transcripts including MYC, which might contribute to the malignant phenotype of MM and PCT.

YB-1 knockdown reduces MYC protein expression

As mentioned above, HMCLs have a wild-type MYC IRES sequence and YB-1 protein binds MYC and TCTP mRNA. To further characterize their translational control by YB-1, we performed transient shRNA-mediated knockdown of YB-1 in HMCLs. Western blot analysis 72 h after shRNA transfection shows that the protein level of YB-1 is significantly reduced (Figure 4a). Furthermore, the knockdown of YB-1 is accompanied by a reduction of MYC and TCTP protein expression. We engineered AMO-1-tet and HEK293T-tet cell lines to express a doxycycline-inducible shRNA against YB-1. The AMO-1-tet cell line shows a strong decrease in YB-1 and MYC protein level at 3–4 days after doxycycline treatment (Supplementary Figure 3). For HEK293T-tet, 5 days of doxycycline treatment lead to a complete loss of the YB-1 protein as well as MYC, indicating that the translational regulation by YB-1 is independent of the selected cell line model as well as the mutational status of the MYC IRES (Supplementary Figure 3).

Figure 4
figure4

YB-1 knockdown alters MYC and TCTP protein but not corresponding mRNA levels, and changes the polysome profile. (a) shRNA-mediated YB-1 knockdown leads to decreased YB-1, TCTP and MYC protein levels in AMO-1 and MM.1S cells 72 h after electroporation of pSUPER/YB-1 as indicated by western blot analysis, β-actin served as loading control. (b) Real-time PCR quantification of mRNA expression 48 h after YB-1 shRNA transfection shows reduced YB-1 mRNA levels in AMO-1 and MM.1S, whereas the expression of MYC and TCTP remains unchanged (n=4). (c) Characteristic fractionation profile of untreated AMO-1-tet cells (dashed line) compared with YB-1 knockdown induced AMO-1-tet cells after 4 days of doxycycline treatment (solid line). Cytosolic extracts were separated on 5–45% (w/v) sucrose density gradients. Twenty-one fractions were collected automatically and recorded according to their OD 254 nm. Note the change in overall polysome profile of YB-1 knockdown cells. Western blot analysis of YB-1 using two adjacent fractions following fractionation of untreated AMO-1-tet whole-cell extract (overall 23 fractions) shows YB-1 polysome and subpolysome association.

YB-1 controls the translational availability of MYC mRNA

The observed decrease of MYC and TCTP protein levels could be due to reduced mRNA transcription or translation. To confirm that knockdown of YB-1 influences the translation of the aforementioned transcripts, we performed real-time PCR quantification 48 h after transfection with YB-1 shRNA and determined MYC and TCTP mRNA expression levels compared with the control-transfected cells (Figure 4b). Indeed, the mRNA levels of TCTP and MYC are not affected by YB-1 knockdown (AMO-1 log2 2−ΔΔCT MYC=−0.18±0.19 TCTP=−0.03±0.14, MM.1S log2 2−ΔΔCT MYC=0.09±0.11 TCTP=−0.09±0.07), whereas YB-1 mRNA itself is downregulated (AMO-1 log2 2−ΔΔCT YB-1=−1.9±0.46, MM.1S log2 2−ΔΔCT YB-1=−2.7±0.38) (Figure 4b). These results demonstrate that reduced protein levels of MYC and TCTP occur independently of their mRNA abundance.

We used sucrose gradient centrifugation to further investigate the potential mechanism by which YB-1 knockdown inhibits MYC mRNA translation, and compared the polysome profile of YB-1 knockdown vs untreated control AMO-1-tet cells (Figure 4c). This profile shows a clear decline of polysomes and an increase in the subpolysomal messenger ribonucleoprotein particles after 4 days of doxycycline treatment. Western blot analysis of untreated AMO-1-tet cells showed YB-1 association within subpolysomal and polysomal fractions.32 Following gradient centrifugation, we determined by real-time PCR the MYC mRNA abundance in the subpolysomes (pooled fractions 1–10) and polysomes (pooled fractions 11–23). YB-1 knockdown led to a 16% reduction of MYC mRNA in the polysomes compared with untreated control cells, (n=2). Moreover, there was a 42% increase of MYC mRNA in the subpolysomal, translationally inactive fraction, (n=2). Hence, YB-1 binds to MYC mRNA and permits its translation.

YB-1 knockdown activates caspases to induce apoptosis

In addition to reduced MYC protein expression, YB-1 knockdown induced apoptosis in the transfected cell lines as demonstrated by the decrease in cell viability of AMO-1 (53±3%, P<0.01) and MM.1S (56±6%, P<0.001) 72 h after transfection (Figure 5a). The CaspACE FITC-VAD-FMK staining showed 44% of the AMO-1 cells (72 h) and 38% of MM.1S cells (96 h) harbour active caspases confirming the induction of apoptosis (Figure 5b). This novel finding of caspase activation describes the underlying mechanism for decreased MM cell viability in the absence of YB-1 and MYC.

Figure 5
figure5

YB-1 knockdown and decrease of MYC expression activate caspases. (a) Cell viability was assayed by FACS analysis after 72 h and shows reduced numbers of viable cells for AMO-1 (53±3%, P<0.01) and MM.1S (56±6%, P<0.001) compared with control-treated cells (n=4). (b) FACS analysis using the CaspACE in situ assay shows increased Caspase-3 activity in AMO-1 (44%) and MM.1S cells (38%) as indicated, dashed lines correspond to shRNA-treated cells, grey histograms to control cells, one representative example shown.

MYC knockdown decreases YB-1 expression and induces apoptosis

Although MYC protein has been documented in MM no data are available on the direct inhibition of MYC expression in MM and its relevance for MM survival.14, 15 Therefore, we performed direct MYC knockdown experiments in AMO-1 and MM.1S (Figure 6a). Analysis of cell viability 72 h after shRNA transfection shows that the reduced MYC levels decreased the viable cell population to 38±9% (P<0.001) for AMO-1 and 27±12% (P<0.001) for MM.1S compared with control transfections (Figure 6b). Caspase activation was detected in 61% of AMO-1 (72 h) and 46% of MM.1S (96 h) cells confirming the induction of apoptosis (Figure 6c). These data show that MYC expression is crucial for MM cell survival. Interestingly, the loss of MYC also affected YB-1 protein levels, suggesting a possible reciprocal modulation of MYC and YB-1 in HMCLs (Figure 6a).

Figure 6
figure6

MYC knockdown alters YB-1 expression and induces apoptosis in HMCLs. (a) Western blot analysis 72 h after transfection with MYC shRNA illustrates the decrease of MYC and YB-1 protein levels in HMCLs. (b) Cell survival of MYC knockdown cells compared with control-treated cells is reduced in AMO-1 (38±9%, P<0.001) and MM.1S (27±12%, P<0.001), (n=5). (c) MYC knockdown induces Caspase-3 activity in AMO-1 (61%) and MM.1S (46%) as indicated, measured by FACS using the CaspACE in situ assay, dashed lines correspond to shRNA-treated cells, grey histograms to control cells, one representative example shown. (d) Real-time PCR quantification of mRNA expression 48 h after MYC shRNA transfection shows reduced MYC as well as YB-1 mRNA levels in both HMCLs (n=5). (e) Western blot analysis of the P493-6 human B-cell line, which carries a conditional MYC allele was treated with doxycycline for 24 h and shows loss of MYC protein as well as reduced YB-1 levels. (f) Real-time PCR quantification of MEFs from MycERT2 transgenic mice. 4-OHT treatment induces MYC transcriptional activity leading to an increase of YB-1 mRNA expression after 4 h, with a significant increase to threefold after 24 h (**P<0.01) (n=2).

MYC directly regulates YB-1 transcription in vitro

We performed real-time PCR on HMCLs with transient MYC knockdown to elucidate the transcriptional regulation of the YB-1 gene by MYC. Figure 6d shows that shRNA-mediated MYC knockdown reduces both MYC mRNA, and YB-1 mRNA in AMO-1 (log2 2−ΔΔCT MYC=−0.96±0.09, YB-1=−0.65±0.17) and MM.1S (log2 2−ΔΔCT MYC=-0.75±0.27, YB-1=−0.41±0.06) 48 h after transfection. These data suggest that YB-1 transcription is mediated by MYC. Similarly, using the P493-6 B-cell line,33 which has an MYC gene under the control of a tetracycline-responsive promoter, shows that addition of doxycycline to the cells represses MYC protein expression and consequently reduced YB-1 protein levels after 24 h (Figure 6e).

The above data from HMCLs and an independent B-cell line were generated by a reduction of MYC protein levels. It remained unclear if there was a direct link between MYC nuclear translocation and YB-1 protein expression. To answer this question, we used non-malignant MEFs from R26MER mice, which express a switchable 4-OHT-dependent variant of MYC (MycERT2).24, 34 Upon binding of 4-OHT, MycERT2 rapidly translocates to the nucleus and activates transcription. Real-time PCR analysis shows that addition of 4-OHT to MEFs homozygous for the MycERT2 allele, but not to littermate wild-type MEFs, leads to a 1.5-fold increase in YB-1 mRNA after 4 h of treatment (Figure 6f). After 24 h of treatment, the YB-1 mRNA level further increases by threefold (P<0.01), compared with the 4-OHT treated wild-type MEFs (n=2).

These data show a direct link between MYC nuclear translocation and YB-1 mRNA expression. This finding of YB-1 transcription by MYC was detected in three independent cell lines and clinical evidence of this relationship was verified by YB-1 staining in the classical MYC-driven Burkitt lymphoma (n=5; Supplementary Figure 4).

Discussion

It is known that cytoplasmic YB-1 is expressed in a subset of MM patients, its expression is associated with drug resistance and it serves as a translational regulator of several proteins. Nonetheless, how YB-1 mediates its effects is unclear.9, 35 Coincidentally, 60% of MM patients show an MYC activation signature.14, 15, 16 A functional role has been suggested for YB-1assisted MYC mRNA translation in MM involving a mutation in the MYC IRES.17 However in our sequencing data, the MM patient mutation frequency equalled that of healthy individuals, indicating that an association with MM does not exist. Perhaps, the disparity in these results lies in the controls used; our germline controls were from the same patient as were the MM cells. Furthermore, data from the NCBI dbSNP (of 168 individuals) match our results, thus separately contradicting the previously published work.

Most transgenic mouse models available for studies on late B-cell malignancies depend directly on deregulated MYC (for review, see Potter12). Using both MYC-dependent (iMycΔEμ) and MYC-independent (H2-Ld-HuIL-6) models, we found MYC/CD138 expression predicts YB-1 co-expression in malignant PCs independent of genetic background (Figures 1 and 2). YB-1/MYC co-expression in malignant PCs (mouse PCT, human extramedullary plasmacytomas and MM cells) showed the clinical relevance of this previously unidentified relationship; furthermore, the percentage of cells co-expressing both proteins increases with disease progression. Previous studies had lower YB-1-positive cases as reported here, perhaps because of the different antibodies used in each study.9 In our studies, the overwhelming majority of MM and extramedullary plasmacytoma cells expressed YB-1 (Figure 2d; Supplementary Table 1) independent of MYC expression, suggesting an important role for YB-1 expression before acquired genetic abnormalities leading to MYC overexpression.

The strong cytoplasmic YB-1 localization implicated a function in mRNA translation for primary PCs and HMCLs. We identified potential mRNA targets using YB-1 IP and microarray analysis, and found 138 mRNAs bound to YB-1 in both human and mouse models of PC dysfunction, iMycΔEμ tumour and HMCLs, including YB-1, MYC and TCTP mRNA (Figure 3; Supplementary Table 3). The presence of known targets of YB-1 assisted translation (α-tubulin, EEF2, UBC) validated our approach, but the majority of mRNAs presented here have not been identified as YB-1 targets before and hence are novel.29, 31 GSEA analysis of IP transcripts in each sample showed a significant enrichment for either directly MYC-driven pathways or those that had an indirect relationship to MYC (Supplementary Table 2). Taken together, the data on co-expression and direct binding of YB-1 to wild-type MYC mRNA imply that YB-1 is involved in MYC overexpression.

Transient shRNA-mediated YB-1 knockdown in HMCLs decreased MYC protein expression (Figure 4a), whereas MYC mRNA levels remained unchanged (Figure 4b), indicating YB-1 assisted mRNA translation. Density gradient centrifugation and subsequent real-time PCR analysis showed a shift of MYC mRNA from the polysome to the translationally inactive subpolysomal messenger ribonucleoprotein particles (Figure 4c). Therefore, MYC translation is directly dependent on YB-1 in the polysome, describing the mechanism by which YB-1 knockdown affects MYC mRNA translation.

In HMCLs, the reduction of MYC protein levels decreased both YB-1 protein and mRNA expression (Figures 6a and d), and in non-malignant MEFs from MycERT2 transgenic animals MYC activation led to YB-1 mRNA transcription (Figure 6f). These findings in independent models suggest that MYC transcription of YB-1 is not MM specific, and therefore YB-1 is likely to be expressed in any MYC-driven tumour. This suggestion is supported by the preliminary data showing YB-1 expression in Burkitt lymphoma (Supplementary Figure 4). Importantly, however, the immunohistological data showing YB-1 expression independent of MYC demonstrate that MYC is not the only transcription factor controlling YB-1 expression. For example, Twist has been identified as a YB-1 transcription factor in other cancer entities.36

Either MYC or YB-1 knockdown in HMCLs leads to a loss of both proteins, induces caspase activation and subsequently apoptosis, indicating that both YB-1 and MYC expression are crucial for MM cell survival (Figures 5 and 6a–c). JQ1, which inhibits BET bromodomain proteins, indirectly decreased transcription of MYC and subsequently induced cell-cycle arrest and senescence in MM.1S at 48 h.16 We found caspase activation at 96 h after knockdown of either YB-1 or MYC protein in MM.1S (Figures 5b and 6c). An explanation for the differences in outcome may be in the different time windows used, as 48 h is perhaps not long enough for YB-1 degradation. Importantly, our work shows that only a subset of MM cells express MYC, and therefore JQ1 will at best slow but not stop MM. This is in contrast to the finding that the majority of malignant PCs express YB-1, thus defining YB-1 as a better drug target.

Our study revealed a feed-forward loop in that MYC transcribes YB-1 and YB-1 in turn regulates MYC mRNA translation independently of the reported IRES mutation, resulting in a co-expression of the two proteins in malignant PCs. We suggest that this co-regulation is likely to arise in any tumour where both MYC and YB-1 are overexpressed. Our results demonstrate that targeting of YB-1/mRNA interaction is important for the development of novel treatment strategies in MM. These findings make YB-1 a potential drug target as its perturbation results in apoptosis in MYC-dependent tumours.

References

  1. 1

    Rajkumar SV . Treatment of multiple myeloma. Nat Rev Clin Oncol 2011; 8: 479–491.

  2. 2

    MacDonald GH, Itoh-Lindstrom Y, Ting JP . The transcriptional regulatory protein, YB-1, promotes single-stranded regions in the DRA promoter. J Biol Chem 1995; 270: 3527–3533.

  3. 3

    Homer C, Knight DA, Hananeia L, Sheard P, Risk J, Lasham A et al. Y-box factor YB1 controls p53 apoptotic function. Oncogene 2005; 24: 8314–8325.

  4. 4

    Rangan VS, Oskouian B, Smith S . Identification of an inverted CCAAT box motif in the fatty-acid synthase gene as an essential element for modification of transcriptional regulation by cAMP. J Biol Chem 1996; 271: 2307–2312.

  5. 5

    Bargou RC, Jurchott K, Wagener C, Bergmann S, Metzner S, Bommert K et al. Nuclear localization and increased levels of transcription factor YB-1 in primary human breast cancers are associated with intrinsic MDR1 gene expression. Nat Med 1997; 3: 447–450.

  6. 6

    Shibahara K, Sugio K, Osaki T, Uchiumi T, Maehara Y, Kohno K et al. Nuclear expression of the Y-box binding protein, YB-1, as a novel marker of disease progression in non-small cell lung cancer. Clin Cancer Res 2001; 7: 3151–3155.

  7. 7

    To K, Fotovati A, Reipas KM, Law JH, Hu K, Wang J et al. Y-box binding protein-1 induces the expression of CD44 and CD49f leading to enhanced self-renewal, mammosphere growth, and drug resistance. Cancer Res 2010; 70: 2840–2851.

  8. 8

    Bergmann S, Royer-Pokora B, Fietze E, Jurchott K, Hildebrandt B, Trost D et al. YB-1 provokes breast cancer through the induction of chromosomal instability that emerges from mitotic failure and centrosome amplification. Cancer Res 2005; 65: 4078–4087.

  9. 9

    Chatterjee M, Rancso C, Stuhmer T, Eckstein N, Andrulis M, Gerecke C et al. The Y-box binding protein YB-1 is associated with progressive disease and mediates survival and drug resistance in multiple myeloma. Blood 2008; 111: 3714–3722.

  10. 10

    Park SS, Shaffer AL, Kim JS, duBois W, Potter M, Staudt LM et al. Insertion of Myc into Igh accelerates peritoneal plasmacytomas in mice. Cancer Res 2005; 65: 7644–7652.

  11. 11

    Suematsu S, Matsusaka T, Matsuda T, Ohno S, Miyazaki J, Yamamura K et al. Generation of plasmacytomas with the chromosomal translocation t(12;15) in interleukin 6 transgenic mice. Proc Natl Acad Sci USA 1992; 89: 232–235.

  12. 12

    Potter M . Neoplastic development in plasma cells. Immunol Rev 2003; 194: 177–195.

  13. 13

    Kovalchuk AL, Kim JS, Park SS, Coleman AE, Ward JM, Morse HC et al. IL-6 transgenic mouse model for extraosseous plasmacytoma. Proc Natl Acad Sci USA 2002; 99: 1509–1514.

  14. 14

    Chng WJ, Huang GF, Chung TH, Ng SB, Gonzalez-Paz N, Troska-Price T et al. Clinical and biological implications of MYC activation: a common difference between MGUS and newly diagnosed multiple myeloma. Leukemia 2011; 25: 1026–1035.

  15. 15

    Skopelitou A, Hadjiyannakis M, Tsenga A, Theocharis S, Alexopoulou V, Kittas C et al. Expression of C-myc p62 oncoprotein in multiple myeloma: an immunohistochemical study of 180 cases. Anticancer Res 1993; 13: 1091–1095.

  16. 16

    Delmore JE, Issa GC, Lemieux ME, Rahl PB, Shi J, Jacobs HM et al. BET bromodomain inhibition as a therapeutic strategy to target c-Myc. Cell 2011; 146: 904–917.

  17. 17

    Cobbold LC, Wilson LA, Sawicka K, King HA, Kondrashov AV, Spriggs KA et al. Upregulated c-myc expression in multiple myeloma by internal ribosome entry results from increased interactions with and expression of PTB-1 and YB-1. Oncogene 2010; 29: 2884–2891.

  18. 18

    Chappell SA, LeQuesne JP, Paulin FE, deSchoolmeester ML, Stoneley M, Soutar RL et al. A mutation in the c-myc-IRES leads to enhanced internal ribosome entry in multiple myeloma: a novel mechanism of oncogene de-regulation. Oncogene 2000; 19: 4437–4440.

  19. 19

    Landsman D . RNP-1, an RNA-binding motif is conserved in the DNA-binding cold shock domain. Nucl Acids Res 1992; 20: 2861–2864.

  20. 20

    Kloks CP, Spronk CA, Lasonder E, Hoffmann A, Vuister GW, Grzesiek S et al. The solution structure and DNA-binding properties of the cold-shock domain of the human Y-box protein YB-1. J Mol Biol 2002; 316: 317–326.

  21. 21

    Evdokimova V, Ruzanov P, Imataka H, Raught B, Svitkin Y, Ovchinnikov LP et al. The major mRNA-associated protein YB-1 is a potent 5' cap-dependent mRNA stabilizer. EMBO J 2001; 20: 5491–5502.

  22. 22

    Bader AG, Vogt PK . Phosphorylation by Akt disables the anti-oncogenic activity of YB-1. Oncogene 2008; 27: 1179–1182.

  23. 23

    Ashizuka M, Fukuda T, Nakamura T, Shirasuna K, Iwai K, Izumi H et al. Novel translational control through an iron-responsive element by interaction of multifunctional protein YB-1 and IRP2. Mol Cell Biol 2002; 22: 6375–6383.

  24. 24

    Murphy DJ, Junttila MR, Pouyet L, Karnezis A, Shchors K, Bui DA et al. Distinct thresholds govern Myc's biological output in vivo. Cancer Cell 2008; 14: 447–457.

  25. 25

    Kunder S, Calzada-Wack J, Holzlwimmer G, Muller J, Kloss C, Howat W et al. A comprehensive antibody panel for immunohistochemical analysis of formalin-fixed, paraffin-embedded hematopoietic neoplasms of mice: analysis of mouse specific and human antibodies cross-reactive with murine tissue. Toxicol Pathol 2007; 35: 366–375.

  26. 26

    Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA 2005; 102: 15545–15550.

  27. 27

    Paulin FE, West MJ, Sullivan NF, Whitney RL, Lyne L, Willis AE . Aberrant translational control of the c-myc gene in multiple myeloma. Oncogene 1996; 13: 505–513.

  28. 28

    Ponten F, Jirstrom K, Uhlen M . The Human Protein Atlas--a tool for pathology. J Pathol 2008; 216: 387–393.

  29. 29

    Dong J, Akcakanat A, Stivers DN, Zhang J, Kim D, Meric-Bernstam F . RNA-binding specificity of Y-box protein 1. RNA Biol 2009; 6: 59–64.

  30. 30

    Skabkina OV, Lyabin DN, Skabkin MA, Ovchinnikov LP . YB-1 autoregulates translation of its own mRNA at or prior to the step of 40S ribosomal subunit joining. Mol Cell Biol 2005; 25: 3317–3323.

  31. 31

    Tanaka T, Ohashi S, Funakoshi T, Kobayashi S . YB-1 binds to GluR2 mRNA and CaM1 mRNA in the brain and regulates their translational levels in an activity-dependent manner. Cell Mol Neurobiol 2010; 30: 1089–1100.

  32. 32

    Evdokimova VM, Kovrigina EA, Nashchekin DV, Davydova EK, Hershey JW, Ovchinnikov LP . The major core protein of messenger ribonucleoprotein particles (p50) promotes initiation of protein biosynthesis in vitro. J Biol Chem 1998; 273: 3574–3581.

  33. 33

    Schuhmacher M, Staege MS, Pajic A, Polack A, Weidle UH, Bornkamm GW et al. Control of cell growth by c-Myc in the absence of cell division. Curr Biol 1999; 9: 1255–1258.

  34. 34

    Indra AK, Warot X, Brocard J, Bornert JM, Xiao JH, Chambon P et al. Temporally-controlled site-specific mutagenesis in the basal layer of the epidermis: comparison of the recombinase activity of the tamoxifen-inducible Cre-ER(T) and Cre-ER(T2) recombinases. Nucl Acids Res 1999; 27: 4324–4327.

  35. 35

    Cobbold LC, Spriggs KA, Haines SJ, Dobbyn HC, Hayes C, de Moor CH et al. Identification of internal ribosome entry segment (IRES)-trans-acting factors for the Myc family of IRESs. Mol Cell Biol 2008; 28: 40–49.

  36. 36

    Shiota M, Izumi H, Onitsuka T, Miyamoto N, Kashiwagi E, Kidani A et al. Twist promotes tumor cell growth through YB-1 expression. Cancer Res 2008; 68: 98–105.

Download references

Acknowledgements

We would like to thank C Linden, Institute for Immunobiology, for cell sorting, M Göbel, IZKF Würzburg, for the Bioanalyzer runs; G Bornkamm for the P493-6 cell line; M Nikiforov and M Chatterjee for plasmids encoding shRNA against MYC and YB-1, respectively; and N Königl for technical assistance. This work was supported by the Deutsche Forschungsgemeinshaft (DFG, CRU 216 and SFB-TR 17), by the Interdisciplinary Centre for Clinical Research (IZKF) of Würzburg University and by the Multiple Myeloma Research Foundation (MMRF, USA).

AUTHOR CONTRIBUTIONS

KSB and KB designed, performed and analysed experiments, and contributed to writing the manuscript; RB reviewed the manuscript; ME designed, performed and analysed experiments in HMCLs, and contributed to the writing of the manuscript; EL, SW, ME and AR performed and analysed the IP experiments, EL contributed to the writing of the manuscript. MK and ME performed and analysed density gradient centrifugation. DM and ME performed and analysed the data on MEFs and carefully reviewed the manuscript; CL was responsible for the studies on the point mutation in the MYC IRES, and reviewed the manuscript; SJ provided transgenic mouse tissue samples and reviewed the manuscript. AM and AR stained and evaluated MM material, RM provided and evaluated extramedullary plasmacytomas.

Author information

Correspondence to K Bommert.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Leukemia website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Bommert, K., Effenberger, M., Leich, E. et al. The feed-forward loop between YB-1 and MYC is essential for multiple myeloma cell survival. Leukemia 27, 441–450 (2013) doi:10.1038/leu.2012.185

Download citation

Keywords

  • translational control
  • YBX1
  • oncogenic circuit

Further reading