Transcriptional Control and Signal Transduction

A pathway from leukemogenic oncogenes and stem cell chemokines to RNA processing via THOC5

Abstract

THOC5 is a member of the THO complex that is involved in processing and transport of mRNA. We have shown previously that hematopoietic stem cells have an absolute requirement for THOC5 for survival and that THOC5 is phosphorylated on tyrosine 225 as a consequence of leukemogenic protein tyrosine kinase (PTK) action. We have investigated pathways for THOC5 phosphorylation to develop an understanding of THO complex modulation by tyrosine kinase (TK) oncogenes in leukemias. We demonstrate that THOC5 phosphorylation is mediated by Src PTK and CD45 protein tyrosine phosphatase action and that this event is sensitive to oxidative status. We show that THOC5 phosphorylation is elevated in stem cells from patients with chronic myeloid leukemia (CML) and that this phosphorylation is sensitive to the frontline drugs used in CML treatment. Further we show that THOC5 Y225 phosphorylation governs mRNA binding. In addition, CXCL12 is shown to induce THOC5 Y225 phosphorylation, and site-directed mutagenesis demonstrates that this modulates motile response. In conclusion, we delineate a signaling pathway stimulated by leukemogenic PTKs, chemokines and oxidative stress that can affect THO complex mediation of gene expression describing mechanisms for post-transcriptional regulation of protein levels.

Introduction

THOC5/Fms-interacting protein is a member of the THO complex, which in turn is a part of the TREX (transcription/export) complex. TREX has been conserved from yeast to man and has been shown to be required for coupled transcription elongation and nuclear export of mRNAs.1 In Drosophila melanogaster, decreases in the THO complex reduce overall protein synthesis by approximately half, although there is an increase in specific proteins, including those involved in DNA repair.2 In yeast, a similar effect is observed, suggesting a relationship between the THO complex and response to adverse changes in the cellular environment.3 The THO complex is associated with the messenger ribonucleoparticle, the nascent RNA and the (nuclear) exported RNA. Mutations in the THO complex can lead to decreased polyadenylation of RNAs and their enhanced rate of degradation. Heat-shock protein 70 is one gene product that can be regulated by THO complex; specifically, THOC5 binds heat-shock protein 70 messenger ribonucleoparticle and associates with the Tap-p15 bulk mRNA nuclear export receptor to facilitate export.4, 5 THOC5 has been implicated in the regulation of differentiation and development via modulation of transcription factor C/EBPα,6 a gene often mutated in human leukemias. This protein regulates adipocyte and neutrophil/macrophage development.6 Indeed, THOC5 was originally named Fms-interacting protein and identified as a substrate and binding partner of the macrophage-colony-stimulating factor receptor.7 THOC5 is a 75 kDa phosphoprotein found both in the nucleus and cytoplasm and is phosphorylated on tyrosine via activation of the macrophage-colony-stimulating factor receptor TK. It is also tyrosine phosphorylated on Y225 as a consequence of leukemogenic protein TK (PTK) action.8

THOC5 knockouts are embryonically lethal in the mouse, and we have shown that after induction of THOC5 knockdown, mice die from rapid hematopoietic failure due to primitive hematopoietic cell depletion.9 This collapse is far swifter than observed with ATM-null mice, where self-renewal is compromised.10 Thus in hematopoietic terms, we have a protein implicated in stem cell survival, which is a target for downstream action of receptor PTKs and leukemogenic PTKs and which modulates hematopoietic transcription factor action.

How these effects are achieved is not clear. Subcellular localization of THOC5 has been shown to be in a dynamic equilibrium with the protein shuttling between the cytosol and nucleus.11 This raises the possibility that THOC5 protein could be critical in conveying survival/development signals, for example, from the macrophage-colony-stimulating factor receptor to the spliceosome/mRNA export complex. We have investigated pathways for THOC5 phosphorylation to determine its biological role. We report that THOC5 phosphorylation is sensitive to cellular oxidative status, an event mediated by Src PTK and CD45 protein tyrosine phosphatase and the stem cell chemokine CXCL12. The resultant phosphorylation governs mRNA binding and may modulate motility. We have elucidated a receptor-mediated activation of CD45/Src kinases/THOC5 phosphorylation that is affected by leukemogenic oncogenes and found to occur in leukemia stem cells.

Materials and methods

Cell lines

Ba/F3 cells were transfected with either MSCV retroviral vector or MSCV containing BCR/ABL, NPM/ALK, THOC5 and phosphomutants (Y to F) of NPM/ALK and THOC5 as described previously.12 K562 cells were maintained in RPMI (Gibco, Paisley, UK) with 10% (v/v) horse serum (Biowest, Nuaillé, France).

Patient samples

Human samples were leukapheresis products taken at the time of diagnosis from patients with chronic phase chronic myeloid leukemia (CML), with informed consent in accordance with the Declaration of Helsinki and approval of the National Health Service Greater Glasgow Institutional Review Board. The CD34+ population was enriched using CliniMACS (Miltenyi Biotec, Bisley, UK) according to standard protocols, before storage as aliquots at −150 °C. CML CD34+ samples were cultured in serum-free media consisting of Iscove’s modified Dulbecco’s medium (Sigma-Aldrich Company Ltd, Gillingham, UK) supplemented with serum substitute (bovine serum albumin/insulin/transferrin, BIT9500; StemCell Technologies, Grenoble, France), 2 mM L-glutamine, 105 U/ml penicillin, 100 mg/ml streptomycin, 0.1 mM 2-mercaptoethanol and 0.8 μg/ml low-density lipoprotein (Sigma-Aldrich). Serum-free medium was supplemented with a growth factor cocktail of 0.20 ng/ml recombinant human stem cell factor, 1 ng/ml recombinant human interleukin-6, 0.20 ng/ml recombinant human granulocyte colony-stimulating factor, 0.05 ng/ml leukemia inhibitory factor, 0.2 ng/ml MIPα (macrophage inflammatory protein) (StemCell Technologies). Non-CML samples were CD34+-enriched leukapheresis products maintained and used as described for CML CD34+ samples.

Preparation of cellular nuclei and western blotting

Nuclear proteins were enriched using a nuclear isolation kit from Active motif (la Hulpe, Belgium) with some modifications as described previously.13 Western blotting was performed using standard protocols. Antibodies used were actin (Sigma-Aldrich; A5060), α-tubulin (Santa Cruz, Heidelberg, Germany; sc-5286), phospho-CRK-L (Cell Signalling Technology, Hitchin, UK; no. 3181), GAPDH (Cell Signalling Technology; no. 5174), heat-shock protein 70 (Cell Signalling Technology; no. 4875), lamin A/C (Cell Signalling Technology; no. 2032), myc (Cell Signalling Technology; no. 2276) and Src and phosphotyrosine 416 Src (Cell Signalling Technology; nos. 2109 and 2101). Antibodies to THOC5 (F6D/11) were produced by the Leukemia and Lymphoma Research-funded Immunodiagnostics Unit (Oxford, UK) and the anti-phospho-Y225 THOC5 by Eurogentec (Seraing, Belgium).

Inhibitor treatments

Cells were treated with ATM inhibitor KU-55933 (ref. 14) (Tocris Bioscience, Bristol, UK) at 10 μM for 4 h, MEK1/2 inhibitor U0126 (ref. 15) (Cell Signalling Technology) at 10 μM for 1 h, Src inhibitor SU6656 (ref. 16) at 20 μM for 1 h and phosphatidylinositol-3 kinase inhibitor LY294002 (ref. 17) at 10 μM for 4 h (Calbiochem, Nottingham, UK). Etoposide VP-16 (PCH Pharmachemie, Haarlem, The Netherlands) was used at 20 μM for 20 min, unless stated otherwise, hydrogen peroxide (H2O2) (Sigma-Aldrich) at 50 μM for 20 min and 4-hydroxynonenal (HNE) at 10 μM for 20 min (Cambridge Bioscience, Cambridge, UK). Imatinib mesylate (Selleck Chemicals, Newmarket, UK) was used at 5 μM for 4 h and at 5 μM overnight on human primary cells. Dasatinib18 (Bristol-Myers Squibb, Uxbridge, UK) was used at 50 nM on K562 and 150 nM overnight for human primary cells. Nilotinib19 (Novartis, Camberley, UK) was used at 5 μM overnight for human primary cells.

Flow cytometry

For cell surface staining, CD34+ cells were incubated with the appropriate antibodies for 20 min in the dark in 2% fetal calf serum/phosphate-buffered saline. Antibodies used were anti-CD34-APC, CD38-PerCP, CD90-PE (BD Biosciences, Oxford Science Park, Oxford, UK). For intracellular staining, CD34+ and K562 cells were re-suspended in ‘Fix and Perm’ following the manufacturer’s instructions (Merck Chemicals Ltd, Nottingham, UK). Primary antibody was added at room temperature for 1 h at 1/100 dilution and anti-rabbit FITC (Sigma-Aldrich) secondary for 30 min, and also at 1/100 dilution. Samples were analyzed by flow cytometry using the FACSCanto Flow Cytometer (Becton Dickinson, Oxford, UK). Data analysis was performed with FACSDiva (Becton Dickinson) and FlowJo (Tree Star, Olten, Switzerland) software.20

Immunocytochemistry

K562 cells and CML CD34+ cells were left untreated or treated with the indicated drugs for 24 h. Cells were harvested and spotted onto glass microscope slides coated with poly-L-lysine and were subsequently fixed with 3.7% (w/v) formaldehyde for 15 min and permeablized using a 0.25% (w/v) Triton-phosphate-buffered saline solution for 20 min. Cells were blocked with 5% (w/v) bovine serum albumin-phosphate-buffered saline and stained with the indicated primary antibodies and appropriate secondary antibodies. Cells were concurrently stained with 4', 6-diamidino-2-phenylindole (Vectashield, Vector Laboratories Ltd, Peterborough, UK). Cells were imaged using a Zeiss Imager M1 AX10 fluorescence microscope (Carl Zeiss, Birmingham, UK) and subjected to deconvolution (AxioVision Software; Carl Zeiss) for image manipulation. Fluorescent signal was measured in three dimension by Image Processing and Analysis in Java (Image J) program.

Transient transfection

Transient transfection was achieved using the Amaxa Cell Line Nucleofector II device (Lonza, Verviers, Belgium) following the manufacturer’s protocols optimized for Ba/F3 cells (Kit V and Program X-001, Lonza). Constructs used contained wild-type Src or viral (activated) Src.21 Transfection efficiency was measured with the vector pmaxGFP. Cells were lysed 24 h post-electroporation and the phosphorylation of THOC5 assessed by western blot analysis.

Chemotaxis and cell motility assays

Chemotaxis assays were performed as described previously22 using a Boyden chamber assay (Fisher Scientific, Loughborough, UK). The number of cells migrating through a 5 μm filter in response to 200 ng/ml CXCL12 was assessed over 6 h.

THOC5–mRNA complex isolation

Ba/F3 cells expressing NTAP-THOC5 and NTAP-THOC5 Y225F in MSCV were produced as described previously.23 The THOC5–mRNA complexes were isolated and reverse transcription-polymerase chain reaction performed as described previously.4 Briefly, cells were lysed in lysis buffer (10 mM Tris, 150 mM NaCl, 1 mM phenylmethylsulfonyl fluoride, 0.4% NP-40) by freeze thawing and complexes isolated by incubation with streptavidin sepharose (GE Healthcare, Amersham, UK) in lysis buffer with 10 mM mercaptoethanol and 0.5 mM EDTA. Reverse transcription was carried out using oligo(dT) primers and the Omniscript reverse transcriptase kit from Qiagen (Crawley, UK). Primer pairs for Gigyf2 were 5′-IndexTermCGCCGACTGGAAGAGAACC-3′ and 5′-IndexTermTTGCTGTGTTAGACTGCTGAC-3′ (235 bp).

Results

Structure function analysis on NPM/ALK reveals the key residues for activation of tyrosine phosphorylation

We have recently demonstrated that the THOC5 protein is a downstream target of the leukemogenic PTKs BCR/ABL, FIP1L/PDGFRα, TEL/PDGFR and NPM/ALK.8 To discover the potential pathways lying downstream from the oncogenic PTKs, NPM/ALK mutants were expressed in Ba/F3 cells, and the effect of site-directed mutagenesis on the ability to promote phosphorylation on tyrosine 225 of THOC5 assessed. Mutation of NPM/ALK to create a kinase-dead version showed that the kinase activity of NPM/ALK is required for the observed increase in THOC5 phosphorylation (Figure 1a).

Figure 1
figure1

THOC5 Y225 phosphorylation. (a) Cells expressing different tyrosine mutants of NPM/ALK were lysed and their ability to phosphorylate THOC5 on Y225 assessed by western blot analysis with a phospho-Y225 (pY225) phospho-specific antibody. Cells were treated with 50 μM H2O2 for the times shown (b) before being lysed and the levels of Y225-phosphorylated THOC5 assessed by western blot analysis. (c) Following isolation, 10 μg of nuclear and 30 μg of cytoplasmic lysates from Ba/F3 cells were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and the distribution of THOC5 and pY225 THOC5 assessed by western blot analysis. Tubulin and histone H3 are used as loading controls and markers of fractionation quality. (d) Control MSCV cells and myc-tagged THOC5 Y225F mutant-transfected cells were subject to nuclear and cytoplasmic fractionation. Cellular distribution of endogenous and Y225F mutant THOC5 was then assessed by western blot analysis with anti-THOC5 and anti-myc antibodies on 10 μg of nuclear and 30 μg of cytoplasmic lysates. Tubulin and lamin are used as loading controls and markers of fractionation quality.

Initial analysis of point-mutated NPM/ALK identified Y461F, Y567F and Y644F triple mutant as deficient in the ability to promote Y225 THOC5 phosphorylation (Figure 1a). Single mutation analysis then identified Y567F, but not Y461F, Y644F NPM/ALK as deficient in this respect (Figure 1a). The Y567 phosphoresidue of NPM/ALK is known to bind ShcA.24 All three isoforms of the ShcA family are phosphorylated by PTKs, including p66 Shc, which can enhance reactive oxygen production by mitochondria.

Based on these data and the known ability of leukemogenic PTKs to stimulate reactive oxygen production, we assessed the ability of H2O2 to promote THOC5 Y225 phosphorylation. Addition of H2O2 to Ba/F3 cells led to a transient increase in THOC5 Y225 phosphorylation (Figure 1b).

THOC5 is part of the THO complex that is critical for the processing and transport of RNA. Under normal conditions, THOC5 has a nuclear:cytoplasmic location and has been shown to shuttle between these compartments. We have previously shown THOC5 not only to be phosphorylated as a consequence of macrophage-colony-stimulating factor stimulation but also indeed to interact with its receptor c-fms.11 We therefore hypothesized that the phosphorylation of THOC5 at Y225 was a cytosolic event. Western blot analysis of nuclear and cytosolic fractions showed phospho-Y225 to reside predominantly in the cytoplasm (Figure 1c). We also generated an myc-tagged phosphomutant Y225F THOC5 to determine if phosphorylation influenced subcellular localization. This was not the case as the proportion of Y225F THOC5 in the nucleus was the same as native THOC5 (Figure 1d).

As cellular stress can be caused by chemotherapeutic agents that can promote oxidative stress,25 we also assessed the effects of etoposide on THOC5 phosphorylation. We have previously published that the appropriate concentration of etoposide to induce double-strand breaks in DNA (as a measure of biological response) in Ba/F3 cells is 20 μM.26 Figure 2a demonstrates that treatment of cells with etoposide (20 μM) led to a transient increase in Y225 phosphorylation. THOC5 has been shown to be a target for the ATM kinase in respect of serine phosphorylation. ATM is a DNA damage-associated protein kinase as is the PTK c-ABL. To investigate the potential kinases that may be involved in the phosphorylation of THOC5, we undertook a study using protein kinase inhibitors implicated in downstream signaling from DNA damage. Figure 2b illustrates that while the inhibition of abl had no effect on Y225 phosphorylation, ATM inhibition increased THOC5 phosphorylation. Although one cannot discard the possibility, the differences are due to off-target effects. ATM inhibition has been shown to increase reactive oxygen species,10 whereas Abl deficiency results in antioxidant gene overexpression,27 which leads to opposite response to oxidative stress as compared with ATM.28 This could account for the different consequences of their inhibition on THOC5 phosphorylation. With respect to kinases downstream of stress-induced signaling (etoposide or H2O2), Src inhibitor completely abrogated THOC5 Y225 phosphorylation. Unexpectedly, MEK inhibition led to an increase in Y225 phosphorylation. Further studies treating the cells with the MEK and Src inhibitors, while simultaneously subjecting the cells to DNA damage by etoposide treatment, showed that Src activity is required for the etoposide-induced increase in THOC5 phosphorylation (Figure 2c). The MEK inhibitor again resulted in an increase in phosphorylation. As already stated, we have previously shown that leukemogenic oncogenes lead to an increase in Y225 phosphorylation. As these oncogenes are known to activate Src, we asked the question whether inhibition of Src in oncogene-transfected cells would lead to a decrease in Y225 phosphorylation. This was indeed the case indicating that Y225 phosphorylation is downstream of Src following DNA damage and oncogene action (Figure 2d). Furthermore, the induction of THOC5 Y225 phosphorylation following H2O2 treatment was prevented when Src was inhibited (Figure 2e). It has been reported that oxidative stress is linked to Src activation via lipid peroxidation.29 As HNE is produced during oxidative stress, we asked whether HNE could lead to THOC5 Y225 phosphorylation and whether this was via Src activation. Treatment of cells with HNE led to an increase in phospho-Y225 (Figure 2f), which was inhibited if Src inhibitors were present (Figure 2g). Transient transfection of Ba/F3 cells with both active and inactive Src mutants further demonstrated that THOC5 Y225 is phosphorylated as a consequence of Src activity (Figure 2h). Transfection of only the active Src led to an increase in Y225 phosphorylation.

Figure 2
figure2

Mechanism of THOC5 phosphorylation. (a) Ba/F3 cells were treated with 20 μM etoposide for the times indicated before being lysed and the levels of THOC5 phosphorylation assessed by western blot analysis. (b) Western blot analysis of THOC5 phosphorylation in Ba/F3 cells following treatment with 5 μM imatinib (ABLi), 10 μM KU-55933 (ATMi), 10 μM LY294002 (phosphatidylinositol-3 kinase inhibitor PI3Ki)) for 4 h or 10 μM U0126 (MEK1/2i) and 20 μM SU6656 (SFKi) for 1 h. (c) Cells were treated with 20 μM etoposide with or without a 1 h preincubation with 20 μM SU6656 (SFKi) or 10 μM U0126 (MEK1/2i). Levels of phospho-Y225 THOC5 were then assessed by western blot analysis. (d) Cells expressing the leukemogenic oncogenes BCR/ABL, NPM/ALK and Tel/PDGFRβ were treated with 20 μM SU6656 (SFKi) for 1 h and the level of phospho-Y225 THOC5 assessed by western blot analysis. Cells were treated with 50 μM H2O2 (e) or 10 μM HNE (g) for 20 min with or without a 1 h preincubation with 20 μM SU6656 (SFKi) or with 10 μM HNE alone for the times indicated (f) and the level of phospho-Y225 THOC5 assessed by western blot analysis. (h) Cells were transfected with pcDNA constructs containing either GFP, activated viral or wild-type Src and the affect on Y225 THOC5 phosphorylation assessed by western blot.

THOC5 phosphorylation is elevated in primary CML stem cells

Having established the fact that THOC5 Y225 was a target for numerous leukemogenic PTKs, we were keen to demonstrate that THOC5 phosphorylation was present in primary cells. To confirm we were seeing human leukemia-associated effects, we firstly confirmed that BCR/ABL led to an increase in THOC5 phosphorylation in a CML patient-derived cell line. K562 cells were treated with the BCR/ABL inhibitor dasatinib and then subjected to immunofluorescence microscopy using the phospho-Y225 THOC5 antibody (Figure 3a): the loss of phospho-Y225 observed following dasatinib treatment was further confirmed by western blot (Figure 3b) and phospho-CRKL assay demonstrated successful inhibition of BCR/ABL kinase activity. Flow cytometry measurements also demonstrated the sensitivity of phospho-Y225 levels to dasatinib treatment (Figure 3c). These observations were then repeated in primary cells.

Figure 3
figure3

THOC5 Y225 phosphorylation is sensitive to dasatinib treatment in human CML cells. K562 cells were treated with 50 nM dasatinib for 24 h before THOC5 phospho-Y225 levels were assessed by immunofluorescence (a), western blot analysis (b) or flow cytometry (c). Western blot analysis of CRKL and phospho-CRKL were used as controls for the efficacy of dasatinib treatment and GAPDH as a loading control. Flow cytometry data are displayed as mean fluorescence intensity±s.e.m.

First, we demonstrated by immunofluorescence (Figure 4a) and western blot (Figure 4b) that phospho-Y225 levels were sensitive to dasatinib treatment in the bulk CD34+ cell population from patients with CML. We then compared phospho-Y225 levels from CML and healthy individuals by flow cytometry not only in the bulk CD34+ cell population but also in the more primitive CD34+CD38− cells (Figure 4c). All three cell populations showed higher levels of expression of phospho-Y225 THOC5 in CML than normal controls.

Figure 4
figure4

THOC5 Y225 phosphorylation levels are higher in primary CML patients when compared with normal controls and is sensitive to BCR/ABL kinase inhibitors. CD34+ cells from patients with CML were treated with 50 nM dasatinib for 24 h before THOC5 phospho-Y225 levels were assessed by immunofluorescence (a) and western blot analysis (b). Heat-shock protein 70 (HSP90) was used as a loading control in the western blot analysis. (c) Flow cytometric comparison of THOC5 phospho-Y225 expression in CML (n=4) and normal (n=3) CD34+, CD34+CD38+ and CD34+CD38− cells. Results are displayed as mean fluorescence intensity±s.e.m. Student’s t-test indicates that the CD34+ and the CD34+CD38− samples are significantly different between normal and CML cells (P=0.019 and 0.018, respectively). (d) THOC5 phospho-Y225 levels were assessed in the populations shown by flow cytometry following 24 h treatment with 5 μM imatinib, 150 nM dasatinib or 5 μM nilotinib. Results are displayed as mean fluorescence intensity (n=4)±s.e.m.

Dasatinib, in addition to inhibiting BCR/ABL, also inhibits Src. We were therefore interested whether the other two drugs commonly used in the treatment of CML, namely imatinib, and nilotinib, which have no Src inhibitory activity, also inhibited THOC5 Y225 phosphorylation. Figure 4d clearly shows that phospho-Y225 levels were reduced by all three drugs in all three cell populations. The potential role of THOC5 phosphorylation and its regulation by kinase inhibitors in primary leukemic stem cells led us to examine its function more closely.

THOC5 Y225 involvement in stem cell homing chemokine signaling pathways

There are a variety of cellular stimuli that can activate the Src kinase family. We therefore considered whether there were other hematopoietic stem cell-relevant pathways to THOC5 phosphorylation. CXCL12 is less able to promote chemotaxis in the presence of BCR/ABL.30 This suggests some overlap in CXCL12 and BCR/ABL signaling in hematopoietic cells. As BCR/ABL induces THOC5 phosphorylation8 and CXCL12 can promote activation of Src, we therefore investigated its effects on Y225 THOC5 phosphorylation. This would also demonstrate whether G-protein-coupled receptor agonists can affect THOC5 phosphorylation. We found that CXCL12 stimulation led to a transient increase in THOC5 phosphorylation (Figure 5a). Further evidence that this was a direct consequence of CXCL12-mediated Src activation was obtained by investigating THOC5 phosphorylation status in CD45-null mast cells. In CD45 wild-type cells, Src was activated as illustrated by Src autophosphorylation on Y416, and THOC5 phosphorylated. In the absence of CD45, Src was not activated and THOC5 was not phosphorylated, indicating that Src and CD45 are upstream of THOC5 phosphorylation (Figure 5b). We next investigated whether the phosphorylation of THOC5 had an effect on CXCL12-induced motility. Motility assays were performed on cells transfected with THOC5 and Y225F mutant THOC5. Figure 5c clearly shows that the presence of the Y–F mutation increases the cellular response to CXCL12. Thus, the data infer the Y225 site has a role in motile regulation via an as yet unidentified mechanism.

Figure 5
figure5

CXCL12 induced phosphorylation of THOC5. (a) Ba/F3 cells were treated with 200 ng/ml of CXCL12 for the times indicated before the cells were lysed and phospho-Y225 THOC5 levels assessed by western blot analysis. (b) Mast cells isolated from CD45-null and CD45 wild-type (wt) mice were lysed and expression of the proteins indicated assessed by western blot analysis. Actin was included as a loading control. (c) The ability of cells to migrate in response to 200 ng/ml CXCL12 was assessed in Boydon chamber assay. The number of cells migrating in the presence and absence of CXCL12 are shown. (d) THOC5–mRNA complexes were isolated from Ba/F3 NPM/ALK cells expressing NTAP wtTHOC5 or NTAP Y225F THOC5 by binding to streptavidin sepharose. Cell lysates (input) and purified samples (bound) were analyzed by western blot for THOC5 expression and by reverse transcription-polymerase chain reaction (RT-PCR) for Gigyf2 expression.

Phosphorylation of Y225 modulates mRNA binding

We have previously identified mRNAs that bind to THOC5 and rely on THOC5 for export from the nucleus.4 Here we tested whether mutation of tyrosine 225 had an effect on the ability of THOC5 to bind to just such an mRNA. THOC5 and THOC5 Y225F fused to streptavidin-inducing peptide (pNTAP) were expressed in Ba/F3 cells and the THOC5 containing messenger ribonucleoparticle enriched with streptavidin conjugated to sepharose. Western blot analysis of the cell lysates (input) and enriched samples (bound) demonstrate the isolation of THOC5–messenger ribonucleoparticle complex (Figure 5d, upper panels). Reverse transcription-polymerase chain reaction analysis (Figure 5d, lower panels) clearly showed that mutation of Y225 abolishes THOC5’s ability to bind to the mRNA for Gigyf2. This implicates the phosphorylation of tyrosine 225 of THOC5 via leukemogenic kinases in the modulation of mRNA export and hence translational control. Of note is the larger input of endogenous THOC5 in the lysate of THOC5 Y225F-expressing cells to ensure normalization of the two TAP-THOC constructs and subsequent co-enrichment of endogenous THOC5.

Discussion

We have previously demonstrated that leukemogenic PTKs promote the phosphorylation of THOC5 on tyrosine 2258 and that THOC5 is a key protein in the maintenance of hemopoietic stem cells in vivo.9 The aims of this study were to identify the mechanism and consequences of THOC5 tyrosine phosphorylation.

Inferences from structure/function analysis of NPM/ALK implicated oxidative phosphorylation in the induction of THOC5 phosphorylation. Oxidative stress did indeed lead to an increase in THOC5 Y225 phosphorylation and we were able to further demonstrate that genotoxic stress increased THOC5 phosphorylation via oxidative stress including lipid peroxidation, and also activation of Src kinase (see Figure 6). Src family kinases such as Lyn have been shown to have a role in CML progression to blast crisis as well as being a direct source of imatinib and nilotinib resistance in CML,31, 32 suggesting that THOC5 phosphorylation might be involved in the pathogenesis of leukemias. We showed that THOC5 tyrosine phosphorylation is higher in primary human CML CD34+ cells than in normal cells and is expressed in the primitive CML CD34+CD38−CD90+ leukemia stem cells. While Abl inhibition had no effect on THOC5 phosphorylation in the non-transformed Ba/F3 cells that express c-Abl, this elevated THOC5 phosphorylation in primary CML cells was sensitive to Abl inhibitors when the active PTK BCR/ABL was present. The fact that imatinib and nilotinib are as effective as dasatinib (BCR/ABL and Src inhibitor) in reducing THOC5 phosphorylation suggests that BCR/ABL kinase inhibition may be sufficient to decrease THOC5 phosphorylation. Other leukemogenic PTKs enhance production of reactive oxygen species and this effect can be reversed by TK inhibitors.33 Foxo3a can protect the CML stem cell via increased expression of antioxidative stress proteins.34, 35 Nonetheless, there is a BCR/ABL-associated oxidative stress-related protein phosphorylation event seen in the CML stem cells, THOC5 phosphorylation.

Figure 6
figure6

Schematic summary of the results depicting the pathway from agonist to mRNA modulation. Agonists and oncogenes can induce changes in oxidative status within the cell and/or activate CD45 and/or Src. This leads to the phosphorylation of the THOC5 protein in a region known to bind mRNA. The mutation of Y225 to phenylalanine increases response to CXCL12, suggesting that the phosphorylation event is involved in a feedback loop to potentiate motile response. This may involve the mRNA binding function of THOC5 and translation. We propose THOC5 phosphorylation as a means of instructing the translational apparatus to change course or amplitude of activity due to altered cellular environment. This is affected by oncogenic PTKs with likely adverse consequences on a number of pathways via altered proteome.

Src activation has been shown to be regulated by CD45 tyrosine phosphatase, a key transmembrane glycoprotein involved in hematopoiesis and stem cell motility.36 CD45 is able to dephosphorylate Src on either inhibitory tyrosine 527 or on the activating tyrosine 416, thus potentially having antagonistic action on different Src family kinase members and at different stages of development.37 Using CD45-null/wild-type mast and Lin− cells, we have shown that CD45 phosphatase is required for Src activation (as determined by its autophosphorylation on tyrosine 416) and THOC5 phosphorylation on tyrosine 225. As CD45 also regulates motility, a process impaired in leukemia, we further investigated the connection between THOC5 and CD45 with stimulation of CD45 by CXCL12/CXCR4. CXCL12 is a chemokine expressed by the bone marrow stromal cells, which has a key role in the migration of hematopoietic stem/progenitor cells to bone marrow during fetal development and then their retention during adult life.38, 39 Our data showed that THOC5 tyrosine phosphorylation is mediated by the CXCL12/CXCR4 pathway in such a way as to suggest that THOC5 phosphorylation participates in a negative feedback loop. CXCL12 leads to the phosphorylation of THOC5, whereas cells expressing the Y–F mutant of tyrosine 225 display an enhanced response to CXCL12. These observations are consistent with the fact that BCR/ABL induces THOC5 phosphorylation and display a reduced ability to respond to CXCL12.23 It could be hypothesized that inhibition of CXCR4 with perixafor, for example, in the presence of CXCL12, would inhibit the phosphorylation of THOC5.

Our results are thus developing an understanding on THO complex modulation by TK oncogenes in leukemias and chemotactic factors that control stem cell retention in the marrow. This may, in part, be an explanation for the loss of stem cells so rapidly in THOC5-inducible knockout mice.9 The CXCL12/CXCR4 pathway is perturbed in CML and has recently been reported to contribute to non-pharmacological drug resistance.40 Indeed, it has been proposed that CXCR4 inhibition in conjunction with TK inhibition could over-ride drug resistance in CML and reduce residual disease.41 We are currently assessing the role of THOC5 in motility and how this influences leukemic cell behavior. Preliminary proteomic data we have generated on the THOC5 interactome suggests that in addition to members of the THO complex, THOC5 interacts with numerous proteins involved in motility (T Tamura and AD Whetton, unpublished data). Thus, it may be a multifunctional protein.

THOC5 is required for primitive cell survival9 and has been shown to modulate the specific nuclear export of mRNAs involved in cell development and hematopoiesis.4 Given the role of reactive oxygen species in stem cell biology and THOC5 phosphorylation, we asked whether tyrosine 225 phosphorylation has any part in mRNA binding and export. Using a known THOC5 binding mRNA, we were able to demonstrate that indeed THOC5 phosphorylation on tyrosine 225 affected the mRNA binding properties of THOC5. Tyrosine 225 is actually within the mRNA binding domain of THOC5 (T Tamura, unpublished observations) and as we show that phosphorylation of this residue affects mRNA binding. Given that Y225 is a downstream target of numerous leukemogenic PTKs, it can be hypothesized that this phosphorylation event modulates mRNA processing and contributes to the pathogenesis of leukemia and as such offers a novel pathway for intervention in disease treatment.

In conclusion, we propose that THOC5 phosphorylation is part of a mechanism to inform the transcription/translation machinery of a change in environment requiring modulation of translation. Leukemogenic PTKs and receptors modulate the pathway by affecting THOC5 phosphorylation via reactive oxygen species production and Src TK activation (summary depicted in Figure 6). This in turn modulates THOC5 mRNA binding and expression of transcripts for key entities in transformation, such as Myc (T Tamura, unpublished observations).

References

  1. 1

    Reed R, Cheng H . TREX, SR proteins and export of mRNA. Curr Opin Cell Biol 2005; 17: 269–273.

  2. 2

    Rehwinkel J, Herold A, Gari K, Kocher T, Rode M, Ciccarelli FL et al. Genome-wide analysis of mRNAs regulated by the THO complex in Drosophila melanogaster. Nat Struct Mol Biol 2004; 11: 558–566.

  3. 3

    Gallardo M, Luna R, Erdjument-Bromage H, Tempst P, Aguilera A . Nab2p and the Thp1p–Sac3p complex functionally interact at the interface between transcription and mRNA metabolism. J Biol Chem 2003; 278: 24225–24232.

  4. 4

    Guria A, Tran DDH, Ramachandran S, Koch A, El Bounkari O, Dutta P et al. Identification of mRNAs that are spliced but not exported to the cytoplasm in the absence of THOC5 in mouse embryo fibroblasts. RNA—A Publication of the RNA Society 2011; 17: 1048–1056.

  5. 5

    Katahira J, Inoue H, Hurt E, Yoneda Y . Adaptor Aly and co-adaptor Thoc5 function in the Tap-p15-mediated nuclear export of HSP70 mRNA. EMBO J 2009; 28: 556–567.

  6. 6

    Mancini A, El Bounkari O, Norrenbrock AF, Scherr M, Schaefer D, Eder M et al. FMIP controls the adipocyte lineage commitment of C2C12 cells by downmodulation of C//EBPalpha. Oncogene 2007; 26: 1020–1027.

  7. 7

    Tamura T, Mancini A, Joos H, Koch A, Hakim C, Dumanski J et al. FMIP, a novel Fms-interacting protein, affects granulocyte/macrophage differentiation. Oncogene 1999; 18: 6488–6495.

  8. 8

    Pierce A, Carney L, Hamza HG, Griffiths JR, Zhang L, Whetton BA et al. THOC5 spliceosome protein: a target for leukaemogenic tyrosine kinases that affects inositol lipid turnover. Br J Haematol 2008; 141: 641–650.

  9. 9

    Mancini A, Niemann-Seyde SC, Pankow R, El Bounkari O, Klebba-Faerber S, Koch A et al. THOC5/FMIP, an mRNA export TREX complex protein, is essential for hematopoietic primitive cell survival in vivo. BMC Biol 2010; 8: 1.

  10. 10

    Ito K, Hirao A, Arai F, Matsuoka S, Takubo K, Hamaguchi I et al. Regulation of oxidative stress by ATM is required for self-renewal of haematopoietic stem cells. Nature 2004; 431: 997–1002.

  11. 11

    Mancini A, Koch A, Whetton AD, Tamura T . The M-CSF receptor substrate and interacting protein FMIP is governed in its subcellular localization by protein kinase C-mediated phosphorylation, and thereby potentiates M-CSF-mediated differentiation. Oncogene 2004; 23: 6581–6589.

  12. 12

    Pierce A, Whetton AD, Owen-Lynch PJ, Tavernier J, Spooncer E, Dexter TM et al. Ectopic interleukin-5 receptor expression promotes proliferation without development in a multipotent hematopoietic cell line. J Cell Sci 1998; 111: 815–823.

  13. 13

    Pierce A, Williamson A, Jaworska E, Griffiths JR, Taylor S, Walker M et al. Identification of nuclear protein targets for six leukemogenic tyrosine kinases governed by post-translational regulation. PLoS One 2012; 7: e38928.

  14. 14

    Hickson I, Yan Z, Richardson CJ, Green SJ, Martin NMB, Orr AI et al. Identification and characterization of a novel and specific inhibitor of the ataxia-telangiectasia mutated kinase ATM. Cancer Res 2004; 64: 9152–9159.

  15. 15

    Old WM, Shabb JB, Houel S, Wang H, Couts KL, Yen C-y et al. Functional proteomics identifies targets of phosphorylation by B-Raf signaling in melanoma. Mol Cell 2009; 34: 115–131.

  16. 16

    Li X, Brunton VG, Burgar HR, Wheldon LM, Heath JK . FRS2-dependent SRC activation is required for fibroblast growth factor receptor-induced phosphorylation of sprouty and suppression of ERK activity. J Cell Sci 2004; 117: 6007–6017.

  17. 17

    Ringshausen I, Schneller F, Bogner C, Hipp S, Duyster J, Peschel C et al. Constitutively activated phosphatidylinositol-3 kinase (PI-3K) is involved in the defect of apoptosis in B-CLL: association with protein kinase C8. Blood 2002; 100: 3741–3748.

  18. 18

    Copland M, Hamilton A, Eirick LJ, Baird JW, Allan EK, Jordanides N et al. Dasatinib (BMS-354825) targets an earlier progenitor population than imatinib in primary CML but does not eliminate the quiescent fraction. Blood 2006; 107: 4532–4539.

  19. 19

    Jorgensen HG, Allan EK, Jordanides NE, Mountford JC, Holyoake TL . Nilotinib exerts equipotent antiproliferative effects to imatinib and does not induce apoptosis in CD34(+) CML cells. Blood 2007; 109: 4016–4019.

  20. 20

    Hamilton A, Elrick L, Myssina S, Copland M, Jorgensen H, Melo JV et al. BCR-ABL activity and its response to drugs can be determined in CD34(+) CML stem cells by CrkL phosphorylation status using flow cytometry. Leukemia 2006; 20: 1035–1039.

  21. 21

    Woodcock SA, Rooney C, Liontos M, Connolly Y, Zoumpourlis V, Whetton AD et al. Src-induced disassembly of adherens junctions requires localized phosphorylation and degradation of the Rac activator Tiam1. Mol Cell 2009; 33: 639–653.

  22. 22

    Whetton AD, Lu YN, Pierce A, Carney L, Spooncer E . Lysophospholipids synergistically promote primitive hematopoietic cell chemotaxis via a mechanism involving Vav 1. Blood 2003; 102: 2798–2802.

  23. 23

    Pierce A, Unwin RD, Evans CA, Griffiths S, Carney L, Zhang L et al. Eight-channel iTRAQ enables comparison of the activity of six leukemogenic tyrosine kinases. Mol Cell Proteomics 2008; 7: 853–863.

  24. 24

    Tartari CJ, Gunby RH, Coluccia AML, Sottocornola R, Cimbro B, Scapozza L et al. Characterization of some molecular mechanisms governing autoactivation of the catalytic domain of the anaplastic lymphoma kinase. J Biol Chem 2008; 283: 3743–3750.

  25. 25

    Siitonen T, Alaruikka P, Mantymaa P, Savolainen ER, Kavanagh TJ, Krejsa CM et al. Protection of acute myeloblastic leukemia cells against apoptotic cell death by high glutathione and gamma-glutamylcysteine synthetase levels during etoposide-induced oxidative stress. Ann Oncol 1999; 10: 1361–1367.

  26. 26

    Griaud F, Williamson AJK, Taylor S, Potier DN, Spooncer E, Pierce A et al. BCR/ABL modulates protein phosphorylation associated with the etoposide-induced DNA damage response. J Proteomics 2012; e-pub ahead of print 15 June 2012 doi:10.1016/j.jprot.2012.06.003.

  27. 27

    Cao C, Leng YM, Liu X, Yi YP, Li P, Kufe D . Catalase is regulated by ubiquitination and proteosornal degradation. Role of the c-Abl and Arg tyrosine kinases. Biochemistry 2003; 42: 10348–10353.

  28. 28

    Li BJ, Wang XY, Rasheed N, Hu YY, Boast S, Ishii T et al. Distinct roles of c-Abl and Atm in oxidative stress response are mediated by protein kinase C delta. Genes Dev 2004; 18: 1824–1837.

  29. 29

    Dittmann K, Mayer C, Kehlbach R, Rothmund M-C, Peter Rodemann H . Radiation-induced lipid peroxidation activates src kinase and triggers nuclear EGFR transport. Radiother Oncol 2009; 92: 379–382.

  30. 30

    Salgia R, Quackenbush E, Lin J, Souchkova N, Sattler M, Ewaniuk DS et al. The BCR/ABL oncogene alters the chemotactic response to stromal-derived factor-1 alpha. Blood 1999; 94: 4233–4246.

  31. 31

    Donato NJ, Wu JY, Stapley J, Gallick G, Lin H, Arlinghaus R et al. BCR-ABL independence and LYN kinase overexpression in chronic myelogenous leukemia cells selected for resistance to STI571. Blood 2003; 101: 690–698.

  32. 32

    Gioia R, Leroy C, Drullion C, Lagarde V, Etienne G, Dulucq S et al. Quantitative phosphoproteomics revealed interplay between Syk and Lyn in the resistance to nilotinib in chronic myeloid leukemia cells. Blood 2011; 118: 2211–2221.

  33. 33

    Koptyra M, Falinski R, Nowicki MO, Stoklosa T, Majsterek I, Nieborowska-Skorska M et al. BCR/ABL kinase induces self-mutagenesis via reactive oxygen species to encode imatinib resistance. Blood 2006; 108: 319–327.

  34. 34

    Naka K, Hoshii T, Muraguchi T, Tadokoro Y, Ooshio T, Kondo Y et al. TGF-beta-FOXO signalling maintains leukaemia-initiating cells in chronic myeloid leukaemia. Nature 2010; 463: 676–680.

  35. 35

    Yalcin S, Marinkovic D, Mungamuri SK, Zhang X, Tong W, Sellers R et al. ROS-mediated amplification of AKT/mTOR signalling pathway leads to myeloproliferative syndrome in Foxo3(−/−) mice. EMBO J 2010; 29: 4118–4131.

  36. 36

    Saunders AE, Johnson P . Modulation of immune cell signalling by the leukocyte common tyrosine phosphatase, CD45. Cell Signal 2010; 22: 339–348.

  37. 37

    Hermiston ML, Zikherman J, Zhu JW . CD45, CD148, and Lyp/Pep: critical phosphatases regulating Src family kinase signaling networks in immune cells. Immunol Rev 2009; 228: 288–311.

  38. 38

    Kucia M, Reca R, Miekus K, Wanzeck J, Wojakowski W, Janowska-Wieczorek A et al. Trafficking of normal stem cells and metastasis of cancer stem cells involve similar mechanisms: pivotal role of the SDF-1-CXCR4 axis. Stem Cells 2005; 23: 879–894.

  39. 39

    Whetton AD, Graham GJ . Homing and mobilization in the stem cell niche. Tr Cell Biol 1999; 9: 233–238.

  40. 40

    Tabe Y, Jin L, Iwabuchi K, Wang RY, Ichikawa N, Miida T et al. Role of stromal microenvironment in nonpharmacological resistance of CML to imatinib through Lyn/CXCR4 interactions in lipid rafts. Leukemia 2012; 26: 883–892.

  41. 41

    Weisberg E, Azab AK, Manley PW, Kung AL, Christie AL, Bronson R et al. Inhibition of CXCR4 in CML cells disrupts their interaction with the bone marrow microenvironment and sensitizes them to nilotinib. Leukemia 2012; 26: 985–990.

Download references

Acknowledgements

This work was supported by Leukemia and Lymphoma Research UK Program 08004 and Grant 08071 and a CRUK Program C11074/A11008. This study was also supported by the Glasgow and Manchester Experimental Cancer Medicine Centers (ECMC), which is funded by Cancer Research UK and by the Chief Scientist’s Office Scotland (Glasgow). FG was additionally sponsored by the British Council with an Entente Cordiale Scholarship.

Author information

Correspondence to A Pierce.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Griaud, F., Pierce, A., Gonzalez Sanchez, M. et al. A pathway from leukemogenic oncogenes and stem cell chemokines to RNA processing via THOC5. Leukemia 27, 932–940 (2013) doi:10.1038/leu.2012.283

Download citation

Keywords

  • mRNA export complex THOC
  • motility
  • CML

Further reading