UBC Faculty Research and Publications

Cold shock Y-box protein-1 proteolysis autoregulates its transcriptional activities van Roeyen, Claudia R; Scurt, Florian G; Brandt, Sabine; Kuhl, Vanessa A; Martinkus, Sandra; Djudjaj, Sonja; Raffetseder, Ute; Royer, Hans-Dieter; Stefanidis, Ioannis; Dunn, Sandra E; Dooley, Steven; Weng, Honglei; Fischer, Thomas; Lindquist, Jonathan A; Mertens, Peter R Aug 27, 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12964_2013_Article_350.pdf [ 1.7MB ]
JSON: 52383-1.0223555.json
JSON-LD: 52383-1.0223555-ld.json
RDF/XML (Pretty): 52383-1.0223555-rdf.xml
RDF/JSON: 52383-1.0223555-rdf.json
Turtle: 52383-1.0223555-turtle.txt
N-Triples: 52383-1.0223555-rdf-ntriples.txt
Original Record: 52383-1.0223555-source.json
Full Text

Full Text

RESEARCH Open AccessCold shock Y-box protein-1 proteolysisautoregulates its transcriptional activitiesClaudia RC van Roeyen1†, Florian G Scurt2†, Sabine Brandt2, Vanessa A Kuhl1, Sandra Martinkus1, Sonja Djudjaj2,Ute Raffetseder1, Hans-Dieter Royer3, Ioannis Stefanidis4, Sandra E Dunn5, Steven Dooley6, Honglei Weng6,Thomas Fischer7, Jonathan A Lindquist2 and Peter R Mertens2*AbstractBackground: The Y-box protein-1 (YB-1) fulfills pleiotropic functions relating to gene transcription, mRNAprocessing, and translation. It remains elusive how YB-1 shuttling into the nuclear and cytoplasmic compartments isregulated and whether limited proteolysis by the 20S proteasome releases fragments with distinct function(s) andsubcellular distribution(s).Results: To address these questions, mapping of domains responsible for subcellular targeting was performed.Three nuclear localization signals (NLS) were identified. NLS-1 (aa 149-156) and NLS-2 (aa 185-194) correspond toresidues with unknown function(s), whereas NLS-3 (aa 276-292) matches with a designated multimerization domain.Nuclear export signal(s) were not identified. Endoproteolytic processing by the 20S proteasome before glycine 220releases a carboxy-terminal fragment (CTF), which localized to the nucleus, indicating that NLS-3 is operative.Genotoxic stress induced proteolytic cleavage and nuclear translocation of the CTF. Co-expression of the CTF andfull-length YB-1 resulted in an abrogated transcriptional activation of the MMP-2 promoter, indicating anautoregulatory inhibitory loop, whereas it fulfilled similar trans-repressive effects on the collagen type I promoter.Conclusion: Compartmentalization of YB-1 protein derivatives is controlled by distinct NLS, one of which targets aproteolytic cleavage product to the nucleus. We propose a model for an autoregulatory negative feedback loopthat halts unlimited transcriptional activation.Keywords: Cold shock protein, DbpB, YBX1, Nuclear localization signal, Post-translational modification, RNA/DNAbinding proteinBackgroundCold shock proteins (CSP) are amongst the most con-served proteins in evolution, sharing a cold shock domain(CSD) from pro- to eukaryotes [1]. Numerous functionshave been unravelled for members of this protein family.In bacteria CSPs are co-ordinately up-regulated followinga decrease in temperature to rescue bacterial growth [2].In eukaryotic cells CSPs are involved in the transcriptionalregulation of genes related to cell proliferation (e.g. DNApolymerase-α [3], cyclins A and B1 [4], FAS receptor [5]).Further target genes coordinate matrix synthesis anddegradation [6], inflammatory responses (e.g. IL-2 [7],GM-CSF [8]), and antigen presentation (major humanleukocyte antigen [9], ABC transporters [10]).Y-box protein (YB)-1 is the prototypic member of thecold shock protein family in humans. YB-1 acts in a cell-context specific manner on gene transcription, e.g. ofmatrix-metalloproteinase (MMP)-2 [11] and granulocytemacrophage-colony stimulating factor (GM-CSF) genes[8]. Furthermore, YB-1 has been isolated as a major com-ponent of messenger ribonucleoprotein particles (mRNPs)that guide mRNA storage, for instance of GM-CSF [12]and renin [13], and is involved in translation processes[14-16]. The specific association of YB-1 with mRNAevidenced its regulatory role in mRNA processing inconcert with splicing factors, such as SRp30c [17].* Correspondence: peter.mertens@med.ovgu.de†Equal contributors2Department of Nephrology and Hypertension, Diabetes and Endocrinology,Otto-von-Guericke University Magdeburg, Leipziger Str 44, 39120Magdeburg, GermanyFull list of author information is available at the end of the article© 2013 van Roeyen et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of theCreative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use,distribution, and reproduction in any medium, provided the original work is properly cited.van Roeyen et al. Cell Communication and Signaling 2013, 11:63http://www.biosignaling.com/content/11/1/63The plethora of functions fulfilled by YB-1 necessitatessubcellular protein shuttling with high stringency. Specificprotein domains, denoted nuclear export signals (NES)and nuclear localization signals (NLS), may coordinatesuch multifunctional shuttling and tasking [18]. Coordi-nated YB-1 protein shuttling has been characterized within vitro cell models. Cell stress exerted by hyperthermia[19], cytotoxic agents [20], and ultraviolet irradiation [20]alters a predominant cytoplasmic YB-1 localization in un-stressed cells to a nuclear preponderance. Cytokines IL-2[21] and IFN-γ [6,22] are also reported to induce a similarnuclear shuttling. In vivo data have been mostly collectedwith cancer tissue. YB-1 has been detected in the cytoplas-mic and/or nuclear compartments [23,24]. Nuclear YB-1localization has been described as a negative prognosticmarker for cancers of the breast [25], prostate [26], syno-via [23,26], and lung [24]. Discussions have focused on theunderlying mechanism(s) for poor cancer prognosis, e.g.the chemotherapy insensitivity of cells with high levelsof nuclear YB-1 expression may be due to upregulated ex-pression of multidrug resistance-1 (MDR-1 [10]) and theABC transporter MRP2 [27].Given the aforementioned tightly regulated subcellulardistribution of YB-1 in inflammatory diseases and can-cer, the present study was initiated to elaborate theunderlying mechanisms that orchestrate YB-1 proteinshuttling. Firstly, differences in subcellular targeting offluorescently-tagged YB-1 domains was assessed in vitrousing laser scanning microscopy [4]. Additionally, nu-clear localization signals (NLS) that target domains ofthe protein, e.g. following endoproteolysis, to the nuclearcompartment were fine-mapped. The functional rele-vance of a predefined carboxy-terminal fragment (CTF),that readily shuttles to the nuclear compartment, wasunraveled, indicating an auto-inhibitory regulatory loopin gene transcription.ResultsSubcellular localization of YB-1 deletion constructsOur starting hypothesis was that YB-1 protein fragmentsmay be directed to different cellular compartments. Ana-lyses of the subcellular distribution for full-length andtruncated YB-1-GFP fusion proteins has been describedin HeLa cells [4]. We first confirmed these results inour model system. Fusion proteins encompassing eitherthe full-length YB-1 or various deletions, possessing aC-terminal green fluorescent protein tag, were introducedinto rat mesangial cells (RMC; Figure 1A). Some constructsencode for proteins with truncations of the C-terminaldomain (denoted basic/acidic (B/A) or charged zipperdomain); depicted in Figure 1A. To preserve comparabilitywith previous results, we chose to introduce the same ex-pression constructs used by Jurchott et al. [4]. Of note, theprotein fragments span aa 1–317 of the YB-1 protein(accession number J03827)[9] and not of YBX1 (accessionnumber NM_004559), that has a disparate length of 324amino acids due to a later annotation of the database.Cells were grown without cell cycle synchronization inmedium containing 10% FCS, as performed by Jurchottet al. [4]. For all immunofluorescence analyses, at least100 transfected RMCs were assessed for their subcellularfluorescence distribution. Cells were grouped into fivecategories: (i.) exclusive nuclear (N), (ii.) exclusive cyto-plasmic (C), (iii.) a pattern of uneven, predominantlynuclear (NC) or (iv.) predominantly cytoplasmic (CN) or(v.) even nuclear/cytoplasmic (NC) distribution. Laserscanning microscopy detected the full-length YB-1protein fused to eGFP exclusively in the cytoplasm(Figure 1B). Transfection of the control eGFP plasmidresulted in staining of both the nuclear and cytoplasmiccompartments (Figure 1B, second panel). The N-terminal YB-1 domains (aa 21–147) strictly localized tothe cytoplasmic compartment, whereas the C-terminaldomain aa 146–317 exhibited a nuclear localization(Figure 1B). Truncations within the YB-1 C-terminaldomain, encoded by constructs P146-225 and P224-317,revealed that these are targeted to the nucleus, similarlyto the fusion protein encompassing aa 172–225. Theprotein fragment P146-172 fused to eGFP was localizedin both compartments. Of note, YB-1-eGFP fusion pro-tein encoded by a longer construct covering aa 21–262was exclusively detected in the cytoplasm, indicatingthat the nuclear localization signal(s) residing within aa172–225 are not operative in a more extensive proteincontext that includes the N-terminal domains (Figure 1B).With the exception of the construct encoding for aa21–262 all YB-1 deletion constructs behaved similarly,indicating that for the tested model systems there are nomajor differences with regard to YB-1 protein targeting.Fine-mapping of nuclear localization signalsTo narrow down the nuclear localization signals withinthe YB-1 protein, a computer-based search for knownNLS was performed using the NUCDISC program(http://psort.nibb.ac.jp; [28]). The search revealed fourhits, all residing within the C-terminal basic/acidicdomain, that are (i.) aa 149–156, (ii.) 185–194, (iii.) 243–249 and (iv.) 276–292. These motifs were tested inisolation by fusing them to a DsRed fluorescent tag atthe N-terminus. The subcellular localization was deter-mined following expression of the respective fluorescentproteins in RMCs (Figure 2A, B). To readily visualizethe cellular compartments a plasmid encoding for cyanfluorescent protein (CFP) was co-introduced. CFP is pre-dominantly detected within the nucleus at 552 - 627 nm.CFP was chosen as the DsRed tag fluorescence spectrumoverlaps with that of propidium iodine, thus precludingthis method for nuclear counterstaining.van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 2 of 16http://www.biosignaling.com/content/11/1/63The residues encompassing aa 149–156, aa 185–194and aa 276–292 conferred an exclusive nuclear fluores-cence pattern, whereas the aa 243–249 motif did not(Figure 2A, B). Since the motif at aa 149–156, denotedNLS-1, did not localize in the nucleus in the longer pro-tein fragment encoded by P146-172 (Figure 1A, B), wetherefore focused our attention on the motifs at residuesaa 185–194 (NLS-2) and 276–292 (NLS-3) and evalu-ated their minimal composition for nuclear shuttling.Mutational analyses of the nuclear localization signalsNLS-2 and −3The NLS-2 at residue aa 185–194 has been described byBader and Vogt in chicken YB-1 [29]. As a general rule,NLS are comprised of at least seven residues with a highcontent of basic amino acids [18]. Therefore we generatedfive different constructs by introducing mutations tonarrow down the minimal requirement(s) for nuclearlocalization and specifically address the question whetherFigure 1 Localization of YB-1 and deletion constructs of YB-1. A. Schematic overview on GFP-tagged YB-1 (deletion) constructs. B. Localization ofWT YB-1-GFP and YB-1 deletion constructs in proliferating RMC. Cells were transfected with expression vectors encoding for WT YB-1-GFP, YB-1deletion constructs, or GFP. The subcellular localization was determined by confocal laser scanning microscopy, cell architecture was visualized byPI-staining. In the right column a merged overlay of PI staining and GFP fluorescence is shown. The percentages of cells with GFP-staining only in thenucleus (N), predominantly in the nucleus and weak in the cytoplasm (NC), in both compartments equally (NC), predominantly in the cytoplasm (CN),and only in the cytoplasm (C) are provided. 100 transfected cells were assessed for each plasmid.van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 3 of 16http://www.biosignaling.com/content/11/1/63the tyrosine residue at aa 188 is required for functionality(Figure 2C). With constructs PNLS 185 I and PNLS 185 II,an even cellular distribution of fusion proteins was ob-served (Figure 2D), indicating that the centrally located ar-ginines alone are not sufficient for targeting. Exchange ofeither an arginine within the N-terminal portion (PNLS185 III) or a central arginine (PNLS 185 IV) with the neu-tral amino acid glycine had a minimal effect on nuclearlocalization, when compared to the wild-type NLS-2motif. Replacing the tyrosine with phenylalanine at aa 188(PNLS 185 V) also did not alter the nuclear localization, in-dicating that the minimal functional requirements ofNLS-2 are independent of tyrosine 188 phosphorylation.Inspection of the domains at aa 276–292 (NLS-3) re-vealed a bipartite composition of this motif. Both “arms” ofthe motif, PPQRRYRR and RRRRP, exhibit characteristicsFigure 2 Mapping the putative nuclear localization signals (NLS) of YB-1. A. Diagram of DsRed-tagged abbreviated residues correspondingto putative NLS. The diagram summarizes the four putative NLS and their relative localization within the full-length YB-1 protein. B. Subcellularlocalization of putative NLS in RMC. Cells were transfected with expression vectors encoding for DsRed-tagged potential NLS of YB-1 and CFP,respectively. The subcellular localization was determined using confocal laser scanning microscopy. The cellular morphology was visualized byco-transfection with CFP, that enriches within the nucleus. The histograms provide quantification of the staining pattern. 100 transfected cellswere assessed for each construct. C. Mutational analyses of the NLS-2 (PNLS185) motif. Diagram on DsRed-tagged mutated PNLS185 expressionvectors generated to express the wild-type or mutated sequences (N: nuclear localization). D. Localization of the mutated PNLS185 expressionproteins in RMC. Cells were transfected with DsRed-tagged expression vectors encoding for mutated PNLS185 and CFP, respectively. Thehistograms provide quantification of the staining pattern. 100 transfected cells were assessed for each construct.van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 4 of 16http://www.biosignaling.com/content/11/1/63of nuclear localization signals (Figure 3A). These motifsare separated by an interspersed linker comprised offour amino acids. Testing of the partite motifs in isola-tion, encoded by plasmids PNLS276 I and PNLS276 II,resulted in even cellular distribution of fluorescent pro-tein. Deletion of the “linker” motif (NFNY; encoded byplasmid PNLS276 III) or extension of the “linker” byintroduction of two additional glycine residues (encodedby plasmid PNLS276 IV) did not impair functionality of nu-clear targeting. Furthermore, we tested whether substitu-tion of either tyrosine residue affects nuclear localization.Tyrosines were mutated to phenylalanine in two separateconstructs (PNLS276 V and VI), nevertheless, NLS-3remained operative (Figure 3A, B).Figure 3 Mutational analyses of NLS-3 (PNLS276) and immunoblotting of fractionated cell lysates. A. Diagram of DsRed-tagged mutatedPNLS276 expression vectors generated to express wild-type or mutated amino acid sequences (N: nuclear localization). B. RMC were transfectedwith EGF-tagged expression vectors encoding for the mutated PNLS276 sequence. The subcellular localization was assessed by laser scanningmicroscopy. The histograms provide quantification of the staining pattern. 100 transfected cells were assessed for each construct. C. Amino acidsequences of the peptides used for immunization and affinity purification. D. Nuclear YB-1 protein is phosphorylated at NLS-3 in non-differentiated monocytic, but not in differentiated THP-1 cells. Nuclear and cytoplasmic extracts were isolated from non-differentiated anddifferentiated THP-1 cells. Equal protein concentrations were loaded from each fraction. Immunoblotting was performed with the polyclonalantibodies generated against the unphosphorylated NLS-3 or tyrosine 281 phosphorylated sequence (P-NLS-3). Band intensities were quantifiedand normalized for nuclear CREB or cytoplasmic vinculin content. The cytoplasmic fraction has a 10× larger volume than the nuclear fraction.van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 5 of 16http://www.biosignaling.com/content/11/1/63Phosphorylation at tyrosine 281 of NLS-3 correlates withnuclear YB-1 shuttlingThe results obtained with the point-mutated NLS-3 motif(tyrosine exchanges) indicated that phenylalanine residueshad no effect on the NLS-3 functionality in non-stimulated mesangial cells. Given that such results do notexclude phosphorylation of the tyrosine residues withsubsequent alteration of functionality, we extended ouranalyses. As a model system we chose monocytic THP-1cells, which express high levels of YB-1 and differentiateinto adhering, amoeboid shaped macrophages followingprolonged incubation with phorbol-12-myristate (PMA;72 hours [30]). In accordance with previous work, amarked down-regulation of YB-1 expression was observedfollowing PMA stimulation. It is known that PMA incuba-tion affects intracellular signalling cascades and proteinphosphorylation [30]. Western blot analyses were per-formed with two different polyclonal affinity-purified anti-bodies, prepared with unphosphorylated (NLS-3) andphosphorylated (P-NLS-3) peptides (Figure 3C). Phospho-specific antibody was affinity-purified following capture ofthe non-phospho-specific polyclonal immunoglobulins(flow-through following unphosphorylated peptide col-umn incubation). This approach yielded highly specificimmunoglobulin preparations suitable to determine thephosphorylation status, as confirmed in control blots withisolated peptides (not shown).Fractionation of THP-1 cell lysates was performed andthe phosphorylation status at NLS-3 assessed. Immuno-blotting for cytoplasmic and nuclear marker proteins,vinculin and CREB, respectively, indicated successful frac-tionation. Undifferentiated THP-1 cells expressed YB-1protein abundantly, with ~70% of nonphosphorylated YB-1 protein localizing to the nucleus (Figure 3D). Incubationof THP-1 cells with PMA resulted in decreased YB-1 pro-tein expression (−80%) [30]. The YB-1 localization in dif-ferentiated THP-1 cells remained predominant nuclear(Figure 3D, left panel).The same cell lysates were subjected to analyses with thephospho-specific antibody (Figure 3D, right panel). As a re-sult, phosphorylation is only detected in the nuclear frac-tion of YB-1 in undifferentiated THP-1 cells. CytoplasmicYB-1 was not detected, suggesting that cytoplasmic YB-1is non-phosphorylated at NLS-3. In PMA-differentiatedTHP-1 cells, phosphorylation at tyrosine 281 was no longerdetected. These results suggest that phosphorylation oftyrosine 281 in NLS-3 takes place and appears to correlatewith nuclear protein shuttling (see also Additional file 1:Figure S1 and Additional file 2: Figure S2).Two putative cytoplasmic retention (CRS)/nuclear exportsignals (NES) exist within the YB-1 N-terminal domainsYB-1 protein residues aa 21–147 and aa 21–262 werefused to eGFP at the C-terminus. The resultant hybridproteins yielded exclusive cytoplasmic fluorescence pat-terns, indicating the presence of operative CRS or NESwithin the domains spanning aa 21–147 (Figure 1B).Next, four deletion constructs were designed thatencoded for partially overlapping domains within theN-terminus: aa 1–57, aa 52–101, aa 52–146 and aa 69–146. The results indicate that only aa 52-101/eGFPshowed an exclusive cytoplasmic localization (summa-rized in Figure 4A, B).A computer-aided search for nuclear export signals(NetNES 1.1 Server, http://www.cbs.dtu.dk; [31]) identifiedone potential motif at aa 100–109 (LRSVGDGETV). Simi-lar to classical NES, this motif contains three of the fourproperly spaced hydrophobic residues and is rich in theamino acids serine, glutamate, and aspartate. This motifwas tested in isolation, fused to eGFP at the C-terminus,yielding an even cellular distribution (Figure 4B). Of note,fluorescent protein encoded by P52–101 was visualized ina punctuated pattern and pronounced in vicinity to the nu-clear membranes. An interaction of YB-1 with actin myofi-brils has been demonstrated by co-immunoprecipitationstudies [32]. By double immunofluorescence staining wewere able to demonstrate co-localization of actin fiberswith ectopically expressed fluorescent P52-101 protein(Figure 4C).The carboxy-terminal YB-1 cleavage fragment is targetedto the nucleusSorokin et al. [33] described that YB-1 is endopro-teolytically processed by the 20S proteasome betweenamino acids 119/220. The released N-terminal proteinfragment was found to be transcriptionally active in thenucleus after thrombin stimulation of endothelial cells[34], whereas the carboxy-terminal fragment (CTF) hasnot been evaluated further. NLS-3 is situated within the105 aa encompassing the CTF , suggesting that the CTFmay also be targeted to the nucleus. The subcellulartargeting of the CTF was analyzed by means of a constructencoding the CTF (aa 220–324) fused to eGFP at theC-terminus. This fluorescent protein was exclusively local-ized in the nucleus, with a pronounced speckled pattern(Figure 4D). We next asked whether over-expression ofCTF affects the subcellular localization of full-length YB-1protein, given that it may act as a decoy protein for thedimerization motif. The cellular content of untagged CTFwas increased by means of expression vector pSG5(CTF).The co-introduced fluorescent full-length YB-1 proteinwas detected by laser scanning microscopy. In a reciprocalapproach, full-length YB-1 protein levels were increasedusing the expression vector pSG5(YB-1) and the CTFfluorescent protein detected. Under both conditions thedescribed compartmentalization of the proteins remainedunaltered (Figure 4E).van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 6 of 16http://www.biosignaling.com/content/11/1/63YB-1 proteolysis and subcellular targeting of domainsfollowing genotoxic stressThe nuclear shuttling of YB-1 has been described underconditions of cellular stress. To address the question,whether proteolytic processing of YB-1 and nuclearshuttling of the resulting protein fragments takes placecytotoxic stress experiments were designed. Doxorubi-cin, a common drug used to treat cancers of the bladder,breast, lung, or ovary, was added to the culture mediumof MCF-7 breast cancer and rat mesangial cells in theabsence and presence of the proteasome inhibitor MG-132. The subcellular distribution of endogenous YB-1protein was assessed using affinity-purified polyclonalantibodies targeting epitopes within the protein N- andC-terminus (Figure 5A). Immunofluorescence micros-copy revealed that antibody preparations were specificfor YB-1, (Additional file 3: Figure S3). In unchallengedcells, the C-terminal antibody detected YB-1 predomin-antly within the cytoplasm. A concentration-dependentshift to the nucleus was visualized following doxorubicinincubation (DOXO, 0.6 and 1.2 μg/ml; Figure 5B).In order to assess whether cleavage of YB-1 is aFigure 4 Mapping of the potential nuclear export signals (NES) of YB-1 and subcellular localization of the carboxy-terminal YB-1protein fragment (CTF). A. Diagram on putative NES within the YB-1 protein. The composition of generated NES-constructs, denoted P1–57,P52–101, P52–146, and P69–146 with GFP-tags is provided. B. Localization of the tested constructs, denoted P69–146, P52–146, and P100–109,with GFP-tags in RMCs. The histograms provide quantification of the staining pattern. 100 transfected cells were assessed for each construct.C. Co-localization of P52–101 and α-smooth muscle actin in RMC. Cells were transfected with P52–101 (GFP-tagged) and after fixation, immuno-stained for α-smooth muscle actin (TRITC labelling). In the right column (GFP + TRITC) an overlay of TRITC and GFP staining is presented. D. Ratmesangial cells were transfected with an expression vector encoding for GFP-tagged CTF of YB-1 (pCTF). The subcellular localization was assessedusing confocal laser scanning microscopy. The histogram provides quantification of the staining pattern. 100 transfected cells were counted.E. Localization of WT YB-1-GFP and CTF in co-transfected RMC. Cells were transfected with an expression vector encoding for WT YB-1-GFPand pSG-5(CTF), encoding for the untagged CTF. Furthermore, RMC were transfected with pCTF (GFP-tagged CTF) and the expression vectorpSG5(YB-1) encoding for untagged YB-1. The subcellular localization was determined using confocal laser scanning microscopy. The histogramsprovide quantification of the staining pattern for 100 transfected cells.van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 7 of 16http://www.biosignaling.com/content/11/1/63Figure 5 (See legend on next page.)van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 8 of 16http://www.biosignaling.com/content/11/1/63prerequisite for nuclear translocation, proteasome inhibi-tor MG-132 (10 μM) was added to the cell culturemedium and the cells challenged with doxorubicin. Incells preincubated with MG-132, most YB-1 proteinremained cytoplasmic with only a minor fraction shuttlingto the nucleus (Figure 5B). From this observation it maybe concluded that proteasomal cleavage appears to be aprerequisite for nuclear translocation; although additional“activation” mechanisms may be operative. Next, theantibody specific for an epitope within the proteinN-terminus was utilized. In unchallenged cells, theN-terminal antibody detected YB-1 predominantly withinthe cytoplasm, similar to the results obtained with theC-terminal antibody. However, following genotoxic stress(Figure 5C), some YB-1 protein was still detected peri-nuclear. Preincubation with MG-132 yielded a predo-minantly cytoplasmic fluorescence pattern. The resultssuggest that cell stress-dependent protein cleavage isfollowed by nuclear shuttling of the protein C-terminaldomain. To confirm that such a cleavage event occurs,cell fractionation was performed with rat mesangial cellsexposed to increasing concentrations of doxorubicin (0.6,1.2, and 2.4 μg/ml), followed by immunoblotting with theaforementioned antibodies (Figure 5D and Additional file2: Figure S2). As a result, a concentration-dependent nu-clear accumulation of full-length YB-1 was detected withboth antibodies. In addition, a protein fragment with arelative molecular weight ~28 kDa was detected followingdoxorubicin exposure. This fragment is found exclusivelyin the nuclear fractions using the antibody recognizingthe C-terminal epitope and is phosphorylated at NLS-3(Additional file 2: Figure S2). Similar results were ob-tained using MCF-7 breast cancer cells (Additional file 4:Figure S4A/B). Again, a concentration-dependent shift ofYB-1 to the nuclear compartment was visualized fol-lowing doxorubicin incubation, a proteasome-dependentcleavage was evidenced by the same subcellular alter-ations as described for rat mesangial cells. Cell viabilitywas assessed by trypan blue exclusion in control experi-ments with increasing concentrations of doxorubicin(Additional file 4: Figure S4C).The CTF influences the transcriptional activity of full-length YB-1In the following, we tested whether ectopically expressedCTF interferes with gene transcription orchestrated byoverexpressed full-length YB-1, e.g. target gene MMP-2.Rat mesangial cells were transfected with the reporterconstruct pGL2MMP-2/RE-1 that harbours a YB-1 re-sponsive enhancer element derived from the rat MMP-2promoter [35,36]. The experimental set-up included co-transfections with empty expression vector pSG5, pSG5(YB-1), pSG5(CTF), or the combination of the latter. Asdescribed before, full-length YB-1 overexpression en-hanced the transcriptional activity of the MMP-2 pro-moter more than 100-fold under the chosen conditions([11], Figure 6A). A similar induction of gene transcrip-tion was observed with ectopically expressed CTF, ran-ging between 100- and 200-fold even at low doses ofplasmid DNA. When both expression plasmids, encod-ing full-length YB-1 or CTF, were co-transfected, thetranscriptional activity was markedly repressed. Quanti-fication revealed luciferase values at background levels(Figure 6A). The abrogation of full-length YB-1 tran-scriptional activity was determined with increasing con-centrations of CTF expression plasmid. Thus, the CTFmay functionally interfere with the trans-stimulatory ef-fect of full-length YB-1 on target gene expression,whereas the CTF alone is capable of trans-activatinggene transcription to a comparable extent as full-lengthYB-1 protein.A trans-repressive effect of YB-1 on gene transcription,e.g. of the col1α1 promoter, has also been described [37].We next wished to evaluate whether the CTF alone mayfulfill such a repressive effect. Rat mesangial cells weretransfected with the well characterized promoter reporterconstruct pGLα1-2.3 [37]. Ectopic overexpression of full-length YB-1 led to a suppression of transcriptional pro-moter activity by more than 90%. Ectopic overexpressionof CTF alone resulted in a similar reduction of promoteractivity by ~70% (Figure 6B). Co-expression of both, full-length YB-1 protein and CTF, did not interfere with therepressive effect.(See figure on previous page.)Figure 5 Subcellular localization of YB-1 protein fragments following genotoxic stress in the absence and presence of proteasomeinhibitor. A. Schematic of YB-1 protein with its centrally localized cold shock domain. Polyclonal peptide-derived affinity-purified antibodies weregenerated against two epitopes localized either in the N- or C-terminus. B. Distribution of endogenous YB-1 protein was assessed byimmunofluorescence microscopy in rat mesangial cells following immunodetection with the anti-YB-1 antiserum directed against the C-terminus(primary antibody). Murine anti-vinculin was used to visualize the cell structure. Fluorescently labelled secondary antibodies anti-rabbit IgG(Fab)-Cy3 and anti-mouse IgG(Fab)-FITC were used for detection. Nuclei were visualized by DAPI staining. Rat mesangial cells were incubated for 16 hwith doxorubicin at increasing concentrations (0.6 and 1.2 μg/ml) in the absence or presence of proteasome inhibitor MG-132 (10 μmol/l).Images were taken at ×63 magnification. C. Distribution of endogenous YB-1 protein was assessed by immunofluorescence microscopy in ratmesangial cells according to the protocol outlined in A with polyclonal YB-1 antiserum directed against the N-terminus (N-Term, primaryantibody). D. Immunoblotting of fractionated cell lysates from rat mesangial cells exposed to doxorubicin at increasing concentrations (0.6 , 1.2,and 2.4 μg/ml). Cytoplasmic and nuclear proteins were separated and purity ascertained by detection of vinculin and CREB.van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 9 of 16http://www.biosignaling.com/content/11/1/63DiscussionOur first quest was to determine the subcellularlocalization of full-length YB-1 protein as well as aseries of truncated protein fragments (Figure 1). Weconfirmed in resting cells that full length YB-1 is pri-marily cytoplasmic. Construct encoding aa 21 to 262yielded an equal fluorescent signal in the cytoplasm andnucleus, whereas it was exclusively cytoplasmic the pre-vious study [4]. For the other constructs, no differencesregarding subcellular targeting of YB-1 protein wasdetected in vitro.Next we mapped the functional motifs and sub-domains within the YB-1 protein that either confernuclear shuttling/retention or cytoplasmic localization/retention. Analogous to the report by Bader et al. [29]we detected functionality of the NLS at aa 185–194 (re-ferred to here as NLS-2) in rat cells. In addition, twonovel nuclear localization motifs located at aa 149–155(NLS-1) and aa 276–292 (NLS-3) were mapped. Inspec-tion of motif NLS-3 immediately suggested a bipartitecomposition. Mutational analyses confirmed that bothparts are indeed required for functionality, at the sameFigure 6 The carboxy-terminal YB-1 protein fragment (CTF) regulates transcriptional activity. A. Transcriptional activity of CTF on theMMP2 promotor construct pGL2MMP-2/RE-1. Rat mesangial cells were transfected with equal amounts of DNA (as indicated). After 48 hours celllysates were prepared and luciferase assays were performed. All experiments were performed in triplicates and confirmed in three independentexperiments. B. Transcriptional activity of the CTF on the Col1a1 promotor construct pGLα1-2.3. Rat mesangial cells were transfected with theindicated expression vectors and incubated for 48 hours. Luciferase assays were performed thereafter using cell lysates. All experiments wereperformed in triplicates and confirmed in three independent experiments.van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 10 of 16http://www.biosignaling.com/content/11/1/63time alterations of the spatial organization of the twohalves did not impair nuclear shuttling of a fluorescent fu-sion protein. Both NLS sequence motifs (NLS-2 and −3)contain tyrosine residues that are potential sites for phos-phorylation, e.g. to regulate functionality of the NLS. Todetermine whether phosphorylation take place, we gener-ated polyclonal peptide-based antibodies against the non-phosphorylated NLS-3 domain and the same antigenphosphorylated at tyrosine 281. The results indicate thatmost of the YB-1 protein phosphorylated at tyrosine 281within NLS-3 appears to be in the nucleus, whereas thecytoplasmic YB-1 protein appears unphosphorylated atthis residue (Figure 3 and Additional file 2: Figure S2).These results indicate that phosphorylation at tyrosine281 is accompanied by nearly exclusive nuclear loca-lization of full-length YB-1, suggesting that the phosphory-lated NLS-3 may act dominantly on the subcellularlocalization of Y-box protein. However, the data also showthat this phosphorylation is not essential for nuclearlocalization. In addition, the data suggest that the loca-lization of YB-1 may be highly regulated after cell stimula-tion. Stenina et al. [34] and Sorokin et al. [33] observeda nuclear localization of the truncated YB-1 proteincontaining N-terminal portions of the protein in stimu-lated cells. For the amino-terminal domains of YB-1, astrict cytoplasmic localization was observed when theexpressed YB-1 deletion/GFP-fusion proteins harbour aa52–101. In an attempt to define nuclear export signal-containing domains further constructs were designed thatmapped to diverse regions of the protein N-terminus (seeFigure 6). However all of these “truncated” constructs losttheir subcellular specification.In summary, we could identify 3 different NLS, but noNES within the YB-1 protein. The existence of three differ-ent NLS within a protein underscores a careful regulationof its subcellular localization. The localization of phos-phorylated NLS-3 within the nucleus of unstimulated cellsraises many questions regarding its regulation. However,further scrutiny of the other NLSs shows that they alsocontain tyrosine residues, several for which the phosphor-ylation has already been reported (PhosphoSitePlus).Thus, the regulation of tyrosine phosphorylation and itsrole in the nuclear localization of YB-1 is an area that re-quires further study. Previous reports by Bader et al. haveidentified aa 137–183 in participating as a multimerizationdomain of chicken YB-1 [29]. YB-1 multimerized in thepresence and absence of DNA and RNA templates [38].Therefore, it is possible that in the aggregates ofmultimerized YB-1, the region harboring the NLS-1 ishidden and thus no longer accessible to proteins involvedin nuclear import.Information on the subcellular localization of YB-1protein in healthy tissue or in inflammatory diseases isscarce. In healthy kidney tissue, YB-1 was almostexclusively detected in the nucleus of glomerular andtubulointerstitial resident cells [39]. Following the induc-tion of mesangioproliferative disease, a temporally andspatially coordinated up-regulation of YB-1 was detectedin the cytoplasm of mesangial cells [39]. Such a tightregulation of the subcellular protein specification is oneaspect that must be fulfilled to explain its involvementin pleiotropic functions, ranging from gene transcriptionto pre-mRNA splicing, mRNA translation, secretion, andreceptor interaction [40,41].In the studies by Sorokin et al. [33] the focus wasplaced on the N-terminal fragment. Somewhat sur-prisingly, we did not detect this fragment with ourN-terminal antibody after stress induction (Figure 5D),however we clearly detect the CTF, which may suggest adifferential processing of YB-1 that is signal-dependent.Our inability to detect the N-terminal fragment of YB-1suggests that it is either post-translational modified,thus masking the epitope, or degraded. While there areno reported phosphorylation sites within the epitope (aa10–22), there are multiple lysine residues that are sitesof ubiquitinylation (PhosphoSitePlus® [42]). Addition-ally, YB-1 associates via its N-terminus with FBX33, acomponent of an SCF E3–ubiquitin ligase; an inter-action that targets YB-1 for proteasomal degradation[43]. However, the issue of whether proteolytic process-ing of YB-1 occurs is still disputed [44,45].We extended the functional analyses by creating an ex-pression plasmid encoding for the CTF only. The effect ofectopically expressed CTF on gene transcription from en-hancer and silencer elements, respectively, of YB-1 targetgenes MMP-2 and collagen type I [11,37] were analyzed.In rat mesangial cells ectopic overexpression of the CTFresulted in increased transcriptional activity of the MMP-2 response element-1, as was shown before for full-lengthYB-1 (Figure 6). Notably, co-expression of CTF and full-length YB-1 resulted in a loss of transcriptional trans-activation of the same element. Thus, the carboxy-terminal YB-1 protein fragment aa 220–324, lacking thedescribed DNA- or RNA-binding cold shock domain [46],acts dominant-negatively on full-length YB-1-dependentgene transcription. The CTF, which is composed of basic/acidic repeats, is generally thought to mediate protein-protein interactions [1]. However, the B/A motifs from theC-terminus of Y-box proteins have also been shown tohave nucleic acid-binding activity [46,47]. Thus, a DNAbinding mechanism would be an obvious explanation,however other possibilities could be envisioned (e.g. adecoy function that influences protein-protein interactionsto promote or inhibit complex formation or YB-1oligomerization). Further investigations will be required toelucidate this mechanism of action.A hypothetical model for the functionality of the CTFwas set up, emphasizing the need for cooperative proteinvan Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 11 of 16http://www.biosignaling.com/content/11/1/63interactions to direct gene transcription and also thenovel regulatory ramifications with the CTF acting ongene transcription. In normal, unstressed cells full-length YB-1 is primarily cytoplasmatic (Figure 7). In thecase of genotoxic stress, a subset of the full-length YB-1protein is cleaved by the 20S proteasome. Both the full-length protein and the cleavage products localize to thenucleus, resulting in a loss of transcriptional activity atthe MMP-2 promoter. This effect may be dependent onthe half lives of the distinct protein fragments (Figure 7).It will be of interest to see whether other functions ofYB-1, such as mRNA binding and translation processes,are also regulated by the CTF. Of note there was atrans-repressive effect of the CTF on the collagen type Ipromoter element, that was not subjected to interferencewith wild-type protein activities. Thus it appears that theactivating and repressive activities of YB-1 on gene tran-scription may involve mechanisms that reflect a require-ment for different levels of YB-1 oligomerization and/orperhaps different binding partners.ConclusionsThe full spectrum of signals that induce either the phos-phorylation and/or proteolytic processing of YB-1 stillYB-1MMP-2RE1CSDCSDno stressexperimentalmanipulationwith increasedCTF expressionMMP-2RE1genotoxic stressYB-1MMP-2RE1CSD20S proteasomecytosolcytosolcytosolnucleusnucleusnucleusCTFNTFCSDCTFYB-1 CSDCSDCSDCSDFigure 7 Model of the functional activities for the C-terminal fragment (CTF) of YB-1. In non-stressed cells full-length YB-1 protein ispredominantly cytoplasmic and may shuttle between the nucleus and cytoplasm. In the nucleus, full-length YB-1 binds to the RE-1 elementwithin the MMP-2 promotor and trans-activates gene transcription. Following genotoxic stress, full-length YB-1 is predominantly localized in thenuclear compartment. In addition, cleavage of full-length YB-1 protein by the 20S proteasome takes place; whether this occurs in the cytoplasmor nucleus has not been investigated. The cleaved C-terminal fragment (CTF) also resides within the nuclear compartment. Co-localization offull-length YB-1 and CTF in the nuclear compartment results in loss of transcriptional trans-regulation of the MMP-2 promoter. The MMP-2promoter is transcriptionally activated by ectopically expressing CTF, that readily shuttles to the nucleus.van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 12 of 16http://www.biosignaling.com/content/11/1/63remains to be determined. It is interesting to speculateto what extent these processes may contribute to path-ology, particularly as enhanced levels of nuclear YB-1are often associated with a poor prognosis in cancer[45,48,49]. Proteasome inhibitors have been introducedinto therapeutical regimens of hematological disorderslike multipe myeloma. Of note, experimental evidenceindicates that proteasomal inhibitors like bortezomibmay exert profound antiinflammatory activities in kidneydiseases, like lupus nephritis [50] and necrotizing glom-erulonephritis [51]. Future studies will address the im-portant question of whether the prevention of YB-1cleavage via proteasomal inhibition contributes to theseanti-inflammatory activities.MethodsCell cultureRat mesangial cells were established as previously de-scribed [52,53]. Human monocytic THP-1 as well as hu-man breast adenocarcinoma MCF-7 (Michigan CancerFoundation-7) cells were obtained from the ATCC.RMCs and THP-1 cells were grown in RPMI 1640medium supplemented with 10% fetal calf serum,2 mM L-glutamin, 100 μg/ml of streptomycin and 100U/ml penicillin at 37°C in humidified 5% CO2. MCF-7cells were grown in DMEM medium supplemented asdescribed above.PlasmidsPlasmids encoding for the WT YB-1 fusion protein andthe GFP-tagged deletion constructs were obtained fromK. Jürchott (Max-Delbrück Center, Berlin, Germany).For characterization of the NLS and NES, DNA frag-ments of the YB-1 sequence (gene AC J03827) werecloned into the vector pDsRed2-C1 or pEGFP-N1 (BDBiosciences Clontech, Heidelberg, Germany). The fusionproteins were tagged with DsRed2 at the N-terminus orEGFP at the C-terminus, respectively. Inserts were gen-erated by PCR using the full-length WT YB-1 vector astemplate and primers as described in Additional file 5:Table S1, and digested with EcoRI/BamHI or BglII/EcoRI. For the generation of the small deletion con-structs and mutational analyses annealed oligonucleo-tides (Invitrogen, Karlsruhe, Germany), as described inAdditional file 6: Tables S2 and Additional file 7: TableS3, were ligated into pDsRed2-C1 or pEGFP-N1 vectors,respectively. All nucleotide sequences were verified byautomated sequencing.Transient transfection and laser scanning microscopyRMC were grown to 60-80% confluency on coverslips in6-well culture plates and transiently transfected withFugene® 6 according to the manufacturers instructions(Roche, Basel, Switzerland). Briefly, 2 μg plasmid DNAand 6 μl Fugene solution were gently mixed in 80 μlserum-free RPMI 1640 medium and incubated for1 hour. RMC transfected with pDSRed2 derived vectors(1 μg) were cotransfected with 1 μg of pCFP vector (BDBiosciences Clontech, Heidelberg, Germany). The trans-fected cells were washed twice with PBS and then fixedwith 4% paraformaldehyde/PBS for 30 minutes. In trans-fection studies with GFP-tagged vectors, nuclear stainingwas achieved by adding propidium iodine in a con-centration of 5 μg/ml for 10 min at room temperature.Cells were mounted with immuno-mount (Shandon,Pittsburgh, USA) and analyzed by laser scanning micros-copy (Axiovert 100 M confocal laser scanning micro-scope, Carl Zeiss, Oberkochen, Germany), using a dualparameter setup and dual wave-length excitation at488 nm and 543 nm for detection of GFP/PI fluores-cence or 458 nm and 543 nm for DsRed/CFP fluores-cence. All transfection experiments were performed atleast three times.Luciferase assaysFor luciferase activity measurements, RMC were trans-fected with pSG5, pSG5(YB-1), and/or pSG5(CTF) ex-pression vectors together with the promoter constructspGL2MMP-2/RE-1 in 6 or 12 well plates and incubatedfor 48 hours. After 48 hours, cells were harvested in100 μl reporter lysis buffer and luciferase assays wereperformed with 20 μl of lysates using the Luciferase assaysystem (Promega, Madison, WI, USA). All assays wereperformed in triplicate.THP-1 differentiationTHP-1 cells were differentiated into adherent macrophage-like cells by incubation with phorbol-12-myristatefor 72 hours (PMA, 100 nM; Sigma-Aldrich, Seelze,Germany). Nuclear and cytoplasmic cell extracts were pre-pared as previously described [11]. Protein concentrationswere determined sing the Bio-Rad protein assay and bo-vine serum albumin as a standard. Proteins were subjectedto SDS-PAGE, transferred to nitrocellulose, and detectedwith suitable polyclonal antibodies to NLS-3, or P-NLS-3(Eurogentech, Köln, Germany), vinculin, or CREB. Proteinbands were visualized using ECL (Amersham Biosciences,Piscataway, NJ, USA). Band intensities were quantifiedusing the Scion Image software. The YB-1 protein contentin cells without PMA-treatment was set to one.AntibodiesTwo peptide-derived rabbit polyclonal anti-YB-1 sera,recognizing epitopes aa 21–37 and 306–321 of YB-1,were generated as previously described [39,40,54,55].Sera were used at a dilution of 1:100 for immunofluores-cence (IF) and 1:1000 for Western blotting (WB). Amonoclonal antibody against Vinculin was purchasedvan Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 13 of 16http://www.biosignaling.com/content/11/1/63from (Fitzgerald Industries, Acton, Massachusetts, USA)and used at 1:100 for IF and 1:1000 for WB. A monoclonalantibody against CREB was purchased from (CellSignaling, Danvers, MA, USA) and used at 1:1000 for WB.Horse radish peroxidise-linked anti-rabbit and -mouse anti-bodies (SouthernBiotech, Birmingham, Alabama , USA) forWB (dilutions 1:5000 to 1:10,000). Cy3-labeled anti-rabbitantibody (Sigma-Aldrich, Seelze, Germany) and FITC-labeled anti-mouse antibody (Dako Deutschland GmbH,Hamburg, Germany) were used for IF.Cell viabilityA trypan blue (Sigma-Aldrich, Seelze, Germany) exclu-sion assay was performed as described [52].Cytoplasm and nuclear fractionationRMC and MCF-7 cells were washed twice with PBS andlysed in Nuclear Extraction Buffer A (10 mM Hepes,10 mM KCl, 0.1 mM EDTA) containing complete prote-ase inhibitor cocktail (Roche, Basel, Switzerland) at 4°Cfor 15 min and centrifuged at 15,000 × g for 3 min at4°C. Supernatants containing cytoplasmic proteins werecollected. Pellets resuspended in Nuclear Extraction Buf-fer A and centrifuged at 15,000 × g for 3 min at 4°C. Su-pernatants were decanted. Steps were repeated twomore times. Pellets resuspended in Nuclear ExtractionBuffer B (20 mM Hepes, 0.4 M NaCl, 1 mM EDTA, 10%Glycerol) containing complete protease inhibitor cocktail(Roche) at 4°C for 20 min and centrifuged at 25,000 × gfor 5 min at 4°C. Supernatants containing nuclear pro-teins were collected and protein concentrations of thedifferent fractions determined by BioRad protein assay(BioRad, Munich, Germany) with BSA as a standard.Western blottingDenatured protein samples were separated by electro-phoresis in 10% SDS-PAGE, transferred onto nitrocellu-lose membranes, blocked with 5% milk in PBS-tween(PBST), washed with PBST, and incubated with primaryantibodies (described above) diluted in PBST overnightat 4°C. Primary antibodies were detected using HRP-conjugated secondary antibodies (described above) di-luted in PBST for two hours at room temperature.Immunofluorescence microscopyMCF-7 and RMC cells were grown on glass coverslips.Twenty-four hours later, the cells were treated with dif-ferent concentration of doxorubicin for fourteen hours.After incubation with doxorubicin, the cells werewashed twice with PBS to remove non-adherent cellsand fixed with 4% paraformaldehyde in PBS. Two differ-ent peptide-derived rabbit polyclonal anti-YB-1 anti-bodies and mouse anti-viniculin were used (1:100).Secondary antibodies Cy3-labeled anti-rabbit (Sigma-Aldrich, Seelze, Germany), and FITC-labeled anti-mouse (Dako Deutschland GmbH, Hamburg, Deutschland)were diluted 1:300 . Nuclei were counterstained with DAPI(Invitrogen, Karlsruhe, Germany). Cells were mountedwith fluorescence mounting medium (Dako DeutschlandGmbH, Hamburg, Germany) and analyzed using a fluores-cence microscope (DM6000 B; Leica Microsystems GmbH,Darmstadt, Germany) equipped with a CCD Camera(DFC340 FX; Leica Microsystems GmbH, Darmstadt,Germany) and a 63/1.4 objective. Separate images weretaken and later merged using ImageJ™ software.Additional filesAdditional file 1: Figure S1. Antibody specificity testing with andwithout calf intestinal alkaline phosphatase treatment. Rat mesangial cellswere left untreated (-) or treated (+) with the protein tyrosinephosphatase inhibitor pervanadate (PV) to maximize phosphorylation.Cell lysates were loaded in duplicate, separated by SDS-PAGE, andtransferred onto membranes. The membrane was cut and blocked. Inaddition, one membrane was incubated in alkaline phosphate buffer with1U calf intestinal alkaline phosphatase (CIP)/ μg protein for 60 minutes(+CIP). The membranes were the incubated with primary antibody asindicated and visualized using ECL detection. As seen, CIP treatmentcompletely ablates the signal detected using the pNLS3 antibodydemonstrating its phospho-specificity. The pan-phosphotyrosine antibody[4G10] shows that not all tyrosine phosphorylation has been removed.The YB-1 and β-tubulin signals are comparable. The position of theprotein standards and the relative molecular weight (MW) in kiloDaltons(kDa) are indicated.Additional file 2: Figure S2. Subcellular localization of YB-1 proteinfragments following genotoxic stress. Immunoblotting of fractionated celllysates from rat mesangial cells exposed to doxorubicin for 14 h atincreasing concentrations (0.6 , 1.2, and 2.4 μg/ml). Cytoplasmic andnuclear proteins were separated and purity ascertained by detection ofvinculin and CREB. Additionally, blotting with the pNLS3 antibody showsthat the phosphorylated C-terminal fragment (p28) is found exclusively inthe nuclear fraction.Additional file 3: Figure S3. Antibody specificity testing withpreincubation of immunization peptides in MCF-7 cells. Distribution ofendogenous YB-1 protein was assessed by immunofluorescencemicroscopy in MCF-7 cells with a peptide-derived affinity purifiedpolyclonal YB-1 antiserum directed against the N-terminus (primaryantibody). Upper left panel: untreated N-terminal antibody. Upper middlepanel: antibody mixed with 0.1 μg/ml of immunizing peptide (YB-1amino acids 21 to 37: SAADTKPGTTGSGAGSG). Upper right panel:antibody mixed with with 1 μg/ml of immunizing peptide. Middlepanels: Murine anti-vinculin antibody was utilized to visualize the cellstructure. Lower panels: Nuclei were visualized with DAPI. Images weretaken at ×63 magnification.Additional file 4: Figure S4. Subcellular localization of YB-1 proteinfragments following genotoxic stress in the absence and presence ofproteasomal inhibitor in MCF-7 breast cancer cells. 1 2. A. Distribution ofendogenous YB-1 protein was assessed by immunofluorescencemicroscopy in MCF-7 cells following immunodetection with the anti-YB-1antiserum directed against the C-terminus (primary antibody). Murineanti-vinculin antibody was utilized to visualize cell structures.Fluorescence labelled secondary antibodies consisted of anti-rabbit IgG(Fab)-Cy3 and anti-mouse IgG(Fab)-FITC. Nuclei were visualized by DAPIstaining. MCF-7 cells were incubated for 16 h with doxorubicin atincreasing concentrations (0.6 and 1.2 μg/ml) in the absence or presenceof proteasome inhibitor MG-132 (7.5 and 10 μmol/l). Images were takenat ×63 magnification. B. Distribution of endogenous YB-1 protein wasassessed by immunofluorescence microscopy in MCF-7 cells according tothe protocol outlined in A with polyclonal YB-1 antiserum directedvan Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 14 of 16http://www.biosignaling.com/content/11/1/63against the N-terminus (N-Term, primary antibody). C. Cytotoxicity assaywith increasing concentrations of doxorubicin. Rat mesangial cells(1 × 106/well) were seeded in 24-well plates in RPMI medium (with10% FCS) followed by treatment with the indicated concentrations ofdoxorubicin for 16 h. Cell viability was then measured using Trypan bluereagent.Additional file 5: Table S1. Primers used for the cloning of deletionconstructs.Additional file 6: Table S2. Primers used for the cloning of smalldeletion constructs.Additional file 7: Table S3. Primers used for the cloning of mutationalanalyses constructs.AbbreviationsCFP: Cyan fluorescent protein; CRS: Cytoplasmic retention signal; CSD: Coldshock domain; CSP: Cold shock protein; CTF: Carboxy-terminal fragment;DMEM: Dulbecco’s Modified Eagles Medium; DOXO: Doxorubicin;eGFP: Enhanced green fluorescent protein; FITC: Fluorescein isothiocyanate;GM-CSF: Granulocyte macrophage-colony stimulating factor; IF: Immunefluorescence; IFNγ: Interferon gamma; IL-2: Interleukin 2; MCF-7: MichiganCancer Foundation-7; MDR-1: Multidrug resistance 1; MMP-2: Matrixmetalloproteinase 2; mRNPs: Messenger ribonucleoprotein particles;MRP2: Multidrug resistance protein 2; NES: Nuclear export signals;NLS: Nuclear localization signals; PBS: Phosphate buffered saline;PBST: Phosphate buffered saline tween; RMC: Rat mesangial cells;SDS-PAGE: Sodium dodecylsulfate-polyacrylamide gel electrophoresis;WB: Western blotting; YB-1: Y-box protein-1.Competing interestsThe authors declare no competing financial interests.Authors’ contributionsC.R.C.vR., F.G.S. designed and performed experiments, analyzed andinterpreted results, and wrote the manuscript; S.B., V.A.K., S.M., S.D., U.R.performed experiments and analyzed data; H-D.R. designed and contributedvital reagents; S.E.D., S.D., H.W. designed experiments, interpreted results, andwrote the manuscript; T.F. designed contributed vital reagents, analyzed andinterpreted results; J.A.L. designed and performed experiments, analyzed andinterpreted results, and wrote the manuscript; and P.R.M. directed the study,designed experiments, supervised the work, interpreted results, and editedthe manuscript. All authors read and approved the final manuscript.AcknowledgementsThis study was supported by grants from the DeutscheForschungsgemeinschaft (DFG), Sonderforschungsbereich 854 (project 01)and Me1365/7-1 to PRM and RO4036/1-1 to CRCvR.Author details1Department of Nephrology and Clinical Immunology, RWTH AachenUniversity, Aachen, Germany. 2Department of Nephrology and Hypertension,Diabetes and Endocrinology, Otto-von-Guericke University Magdeburg,Leipziger Str 44, 39120 Magdeburg, Germany. 3Breast Cancer Research,Center of Advanced European Studies and Research, Caesar, Bonn, Germany.4Department of Nephrology, University of Thessaly, Larissa, Greece.5Laboratory for Oncogenomic Research, Departments of Pediatrics andExperimental Medicine, Child and Family Research Institute, University ofBritish Columbia, Vancouver, British Columbia, Canada. 6MolecularHepatology, Department of Medicine II, University of Heidelberg, Mannheim,Germany. 7Department of Hematology and Oncology, Otto-von-GuerickeUniversity Magdeburg, Magdeburg, Germany.Received: 13 February 2013 Accepted: 12 August 2013Published: 27 August 2013References1. Wolffe AP: Structural and functional properties of the evolutionarilyancient Y-box family of nucleic acid binding proteins. Bioessays 1994,16:245–251.2. Jones PG, VanBogelen RA, Neidhardt FC: Induction of proteins in responseto low temperature in Escherichia coli. J Bacteriol 1987, 169:2092–2095.3. En-Nia A, Yilmaz E, Klinge U, Lovett DH, Stefanidis I, Mertens PR:Transcription factor YB-1 mediates DNA polymerase alpha geneexpression. J Biol Chem 2005, 280:7702–7711.4. Jurchott K, Bergmann S, Stein U, Walther W, Janz M, Manni I, Piaggio G,Fietze E, Dietel M, Royer HD: YB-1 as a cell cycle-regulated transcriptionfactor facilitating cyclin A and cyclin B1 gene expression. J Biol Chem2003, 278:27988–27996.5. Homer C, Knight DA, Hananeia L, Sheard P, Risk J, Lasham A, Royds JA,Braithwaite AW: Y-box factor YB1 controls p53 apoptotic function.Oncogene 2005, 24:8314–8325.6. Higashi K, Inagaki Y, Suzuki N, Mitsui S, Mauviel A, Kaneko H, Nakatsuka I:Y-box-binding protein YB-1 mediates transcriptional repression ofhuman alpha 2(I) collagen gene expression by interferon-gamma.J Biol Chem 2003, 278:5156–5162.7. Chen CY, Gherzi R, Andersen JS, Gaietta G, Jurchott K, Royer HD, Mann M, KarinM: Nucleolin and YB-1 are required for JNK-mediated interleukin-2 mRNAstabilization during T-cell activation. Genes Dev 2000, 14:1236–1248.8. Diamond P, Shannon MF, Vadas MA, Coles LS: Cold shock domain factorsactivate the granulocyte-macrophage colony-stimulating factorpromoter in stimulated jurkat T cells. J Biol Chem 2001, 276:7943–7951.9. Didier DK, Schiffenbauer J, Woulfe SL, Zacheis M, Schwartz BD:Characterization of the cDNA encoding a protein binding to the majorhistocompatibility complex class II Y box. Proc Natl Acad Sci U S A 1988,85:7322–7326.10. Bargou RC, Jurchott K, Wagener C, Bergmann S, Metzner S, Bommert K,Mapara MY, Winzer KJ, Dietel M, Dorken B, Royer HD: Nuclear localizationand increased levels of transcription factor YB-1 in primary humanbreast cancers are associated with intrinsic MDR1 gene expression.Nat Med 1997, 3:447–450.11. Mertens PR, Harendza S, Pollock AS, Lovett DH: Glomerular mesangial cell-specific transactivation of matrix metalloproteinase 2 transcription ismediated by YB-1. J Biol Chem 1997, 272:22905–22912.12. Capowski EE, Esnault S, Bhattacharya S, Malter JS: Y box-binding factorpromotes eosinophil survival by stabilizing granulocyte-macrophagecolony-stimulating factor mRNA. J Immunol 2001, 167:5970–5976.13. Skalweit A, Doller A, Huth A, Kahne T, Persson PB, Thiele BJ:Posttranscriptional control of renin synthesis: identification of proteinsinteracting with renin mRNA 3′-untranslated region. Circ Res 2003,92:419–427.14. Skabkin MA, Kiselyova OI, Chernov KG, Sorokin AV, Dubrovin EV, YaminskyIV, Vasiliev VD, Ovchinnikov LP: Structural organization of mRNAcomplexes with major core mRNP protein YB-1. Nucleic Acids Res 2004,32:5621–5635.15. Evdokimova V, Ruzanov P, Imataka H, Raught B, Svitkin Y, Ovchinnikov LP,Sonenberg N: The major mRNA-associated protein YB-1 is a potent 5′cap-dependent mRNA stabilizer. Embo J 2001, 20:5491–5502.16. Fraser DJ, Phillips AO, Zhang X, van Roeyen CR, Muehlenberg P, En-Nia A,Mertens PR: Y-box protein-1 controls transforming growth factor-beta1translation in proximal tubular cells. Kidney international 2008, 73:724–732.17. Raffetseder U, Frye B, Rauen T, Jurchott K, Royer HD, Jansen PL, Mertens PR:Splicing factor SRp30c interaction with Y-box protein-1 confers nuclearYB-1 shuttling and alternative splice site selection. J Biol Chem 2003,278:18241–18248.18. Cartwright P, Helin K: Nucleocytoplasmic shuttling of transcription factors.Cell Mol Life Sci 2000, 57:1193–1206.19. Stein U, Jurchott K, Walther W, Bergmann S, Schlag PM, Royer HD:Hyperthermia-induced nuclear translocation of transcription factor YB-1leads to enhanced expression of multidrug resistance-related ABCtransporters. J Biol Chem 2001, 276:28562–28569.20. Ohga T, Koike K, Ono M, Makino Y, Itagaki Y, Tanimoto M, Kuwano M,Kohno K: Role of the human Y box-binding protein YB-1 in cellularsensitivity to the DNA-damaging agents cisplatin, mitomycin C, andultraviolet light. Cancer Res 1996, 56:4224–4228.21. Tsujimura S, Saito K, Nakayamada S, Nakano K, Tsukada J, Kohno K, TanakaY: Transcriptional regulation of multidrug resistance-1 gene byinterleukin-2 in lymphocytes. Genes Cells 2004, 9:1265–1273.22. Dooley S, Said HM, Gressner AM, Floege J, En-Nia A, Mertens PR: Y-boxprotein-1 is the crucial mediator of antifibrotic interferon-gamma effects.J Biol Chem 2006, 281:1784–1795.van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 15 of 16http://www.biosignaling.com/content/11/1/6323. Oda Y, Ohishi Y, Saito T, Hinoshita E, Uchiumi T, Kinukawa N, Iwamoto Y,Kohno K, Kuwano M, Tsuneyoshi M: Nuclear expression of Y-box-bindingprotein-1 correlates with P-glycoprotein and topoisomerase II alphaexpression, and with poor prognosis in synovial sarcoma. J Pathol 2003,199:251–258.24. Gessner C, Woischwill C, Schumacher A, Liebers U, Kuhn H, Stiehl P, JurchottK, Royer HD, Witt C, Wolff G: Nuclear YB-1 expression as a negativeprognostic marker in nonsmall cell lung cancer. Eur Respir J 2004,23:14–19.25. Janz M, Harbeck N, Dettmar P, Berger U, Schmidt A, Jurchott K, Schmitt M,Royer HD: Y-box factor YB-1 predicts drug resistance and patientoutcome in breast cancer independent of clinically relevant tumorbiologic factors HER2, uPA and PAI-1. Int J Cancer 2002, 97:278–282.26. Gimenez-Bonafe P, Fedoruk MN, Whitmore TG, Akbari M, Ralph JL, EttingerS, Gleave ME, Nelson CC: YB-1 is upregulated during prostate cancertumor progression and increases P-glycoprotein activity. Prostate 2004,59:337–349.27. Geier A, Mertens PR, Gerloff T, Dietrich CG, En-Nia A, Kullak-Ublick GA,Karpen SJ, Matern S, Gartung C: Constitutive rat multidrug-resistanceprotein 2 gene transcription is down-regulated by Y-box protein 1.Biochem Biophys Res Commun 2003, 309:612–618.28. Nakai K, Horton P: PSORT: a program for detecting sorting signals inproteins and predicting their subcellular localization. Trends Biochem Sci1999, 24:34–36.29. Bader AG, Vogt PK: Inhibition of protein synthesis by Y box-bindingprotein 1 blocks oncogenic cell transformation. Mol Cell Biol 2005,25:2095–2106.30. Raffetseder U, Rauen T, Djudjaj S, Kretzler M, En-Nia A, Tacke F,Zimmermann HW, Nelson PJ, Frye BC, Floege J, et al: Differential regulationof chemokine CCL5 expression in monocytes/macrophages and renalcells by Y-box protein-1. Kidney international 2009, 75:185–196.31. la Cour T, Gupta R, Rapacki K, Skriver K, Poulsen FM, Brunak S: NESbaseversion 1.0: a database of nuclear export signals. Nucleic Acids Res 2003,31:393–396.32. Ruzanov PV, Evdokimova VM, Korneeva NL, Hershey JW, Ovchinnikov LP:Interaction of the universal mRNA-binding protein, p50, with actin: apossible link between mRNA and microfilaments. J Cell Sci 1999,112(Pt 20):3487–3496.33. Sorokin AV, Selyutina AA, Skabkin MA, Guryanov SG, Nazimov IV, Richard C,Th’ng J, Yau J, Sorensen PH, Ovchinnikov LP, Evdokimova V: Proteasome-mediated cleavage of the Y-box-binding protein 1 is linked to DNA-damage stress response. Embo J 2005, 24:3602–3612.34. Stenina OI, Shaneyfelt KM, DiCorleto PE: Thrombin induces the release ofthe Y-box protein dbpB from mRNA: a mechanism of transcriptionalactivation. Proc Natl Acad Sci U S A 2001, 98:7277–7282.35. Harendza S, Pollock AS, Mertens PR, Lovett DH: Tissue-specific enhancer-promoter interactions regulate high level constitutive expression ofmatrix metalloproteinase 2 by glomerular mesangial cells. J Biol Chem1995, 270:18786–18796.36. Mertens PR, Alfonso-Jaume MA, Steinmann K, Lovett DH: A synergisticinteraction of transcription factors AP2 and YB-1 regulates gelatinase aenhancer-dependent transcription. J Biol Chem 1998, 273:32957–32965.37. Norman JT, Lindahl GE, Shakib K, En-Nia A, Yilmaz E, Mertens PR: The Y-boxbinding protein YB-1 suppresses collagen alpha 1(I) gene transcriptionvia an evolutionarily conserved regulatory element in the proximalpromoter. J Biol Chem 2001, 276:29880–29890.38. Izumi H, Imamura T, Nagatani G, Ise T, Murakami T, Uramoto H, Torigoe T,Ishiguchi H, Yoshida Y, Nomoto M, et al: Y box-binding protein-1 bindspreferentially to single-stranded nucleic acids and exhibits 3′→5′exonuclease activity. Nucleic Acids Res 2001, 29:1200–1207.39. van Roeyen CR, Eitner F, Martinkus S, Thieltges SR, Ostendorf T, BokemeyerD, Luscher B, Luscher-Firzlaff JM, Floege J, Mertens PR: Y-box protein 1mediates PDGF-B effects in mesangioproliferative glomerular disease.J Am Soc Nephrol 2005, 16:2985–2996.40. Rauen T, Raffetseder U, Frye BC, Djudjaj S, Muhlenberg PJ, Eitner F, LendahlU, Bernhagen J, Dooley S, Mertens PR: YB-1 acts as a ligand for notch-3receptors and modulates receptor activation. J Biol Chem 2009,284:26928–26940.41. Frye BC, Halfter S, Djudjaj S, Muehlenberg P, Weber S, Raffetseder U, En-NiaA, Knott H, Baron JM, Dooley S, et al: Y-box protein-1 is actively secretedthrough a non-classical pathway and acts as an extracellular mitogen.EMBO reports 2009, 10:783–789.42. Hornbeck PV, Kornhauser JM, Tkachev S, Zhang B, Skrzypek E, Murray B,Latham V, Sullivan M: PhosphoSitePlus: a comprehensive resource forinvestigating the structure and function of experimentally determinedpost-translational modifications in man and mouse. Nucleic Acids Res2012, 40:D261–D270.43. Lutz M, Wempe F, Bahr I, Zopf D, von Melchner H: Proteasomaldegradation of the multifunctional regulator YB-1 is mediated by anF-Box protein induced during programmed cell death. FEBS letters 2006,580:3921–3930.44. Cohen SB, Ma W, Valova VA, Algie M, Harfoot R, Woolley AG, Robinson PJ,Braithwaite AW: Genotoxic stress-induced nuclear localization ofoncoprotein YB-1 in the absence of proteolytic processing. Oncogene2010, 29:403–410.45. Woolley AG, Algie M, Samuel W, Harfoot R, Wiles A, Hung NA, Tan PH, HainsP, Valova VA, Huschtscha L, et al: Prognostic association of YB-1expression in breast cancers: a matter of antibody. PLoS One 2011,6:e20603.46. Ladomery M, Sommerville J: Binding of Y-box proteins to RNA:involvement of different protein domains. Nucleic Acids Res 1994,22:5582–5589.47. Sommerville J, Ladomery M: Masking of mRNA by Y-box proteins. FASEB J1996, 10:435–443.48. Jurchott K, Kuban RJ, Krech T, Bluthgen N, Stein U, Walther W, Friese C,Kielbasa SM, Ungethum U, Lund P, et al: Identification of Y-box bindingprotein 1 as a core regulator of MEK/ERK pathway-dependent genesignatures in colorectal cancer cells. PLoS Genet 2010, 6:e1001231.49. Lasham A, Print CG, Woolley AG, Dunn SE, Braithwaite AW: YB-1:oncoprotein, prognostic marker and therapeutic target? Biochem J 2013,449:11–23.50. Neubert K, Meister S, Moser K, Weisel F, Maseda D, Amann K, Wiethe C,Winkler TH, Kalden JR, Manz RA, Voll RE: The proteasome inhibitorbortezomib depletes plasma cells and protects mice with lupus-likedisease from nephritis. Nat Med 2008, 14:748–755.51. Bontscho J, Schreiber A, Manz RA, Schneider W, Luft FC, Kettritz R:Myeloperoxidase-specific plasma cell depletion by bortezomib protectsfrom anti-neutrophil cytoplasmic autoantibodies-inducedglomerulonephritis. J Am Soc Nephrol 2011, 22:336–348.52. Mertens PR, Espenkott V, Venjakob B, Heintz B, Handt S, Sieberth HG:Pressure oscillation regulates human mesangial cell growth and collagensynthesis. Hypertension 1998, 32:945–952.53. Reisdorff J, En-Nia A, Stefanidis I, Floege J, Lovett DH, Mertens PR:Transcription factor Ets-1 regulates gelatinase a gene expression inmesangial cells. J Am Soc Nephrol 2002, 13:1568–1578.54. Mertens PR, Alfonso-Jaume MA, Steinmann K, Lovett DH: YB-1 regulation ofthe human and rat gelatinase A genes via similar enhancer elements.J Am Soc Nephrol 1999, 10:2480–2487.55. Hanssen L, Frye BC, Ostendorf T, Alidousty C, Djudjaj S, Boor P, Rauen T,Floege J, Mertens PR, Raffetseder U: Y-box binding protein-1 mediatesprofibrotic effects of calcineurin inhibitors in the kidney. J Immunol 2011,187:298–308.doi:10.1186/1478-811X-11-63Cite this article as: van Roeyen et al.: Cold shock Y-box protein-1proteolysis autoregulates its transcriptional activities. Cell Communicationand Signaling 2013 11:63.van Roeyen et al. Cell Communication and Signaling 2013, 11:63 Page 16 of 16http://www.biosignaling.com/content/11/1/63


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items