Oncogenes, Fusion Genes and Tumor Suppressor Genes

BCR-ABL disrupts PTEN nuclear-cytoplasmic shuttling through phosphorylation-dependent activation of HAUSP

Abstract

Chronic myeloid leukemia (CML) is a myeloproliferative disorder characterized by the t(9;22) translocation coding for the chimeric protein p210 BCR-ABL. The tumor suppressor phosphatase and tensin homolog (PTEN) has recently been shown to have a critical role in the pathogenesis of CML. Nuclear localization and proper nuclear-cytoplasmic shuttling are crucial for PTEN’s tumor suppressive function. In this study, we show that BCR-ABL enhances HAUSP-induced de-ubiquitination of PTEN in turn favoring its nuclear exclusion. We further demonstrate that BCR-ABL physically interacts with and phosphorylates HAUSP on tyrosine residues to trigger its activity. Importantly, we also find that PTEN delocalization induced by BCR-ABL does not occur in the leukemic stem cell compartment due to high levels of PML, a potent inhibitor of HAUSP activity toward PTEN. We therefore identify a new proto-oncogenic mechanism whereby BCR-ABL antagonizes the nuclear function of the PTEN tumor suppressor, with important therapeutic implications for the eradication of CML minimal residual disease.

Introduction

The tumor suppressor gene phosphatase and tensin homolog (PTEN) has recently been shown to have a critical role in the pathogenesis of BCR-ABL-mediated leukemogenesis and myeloproliferative disorders.1, 2, 3 PTEN is one of the most frequently mutated/deleted or silenced tumor suppressors in human cancer.4 PTEN antagonizes the PI3K-AKT pathway in the cytosol, while suppressing cell proliferation in the nucleus through the enhancement of APC-Cdh1 function.5 Loss of PTEN function therefore leads to the aberrant activation of the PI3K–AKT pathway, cell growth, survival and proliferation. The involvement of PTEN in tumorigenesis was originally attributed to its genetic inactivation in several cancer types. Recently, it has been shown that PTEN is functionally haploinsufficient and its tumor suppressive function could be impaired after subtle expression downregulation, even in the presence of a wild-type gene copy.6 Furthermore, nuclear-cytoplasmic shuttling of PTEN is crucial for its proper tumor suppressive function. PTEN is normally located both in the nucleus and in the cytosol. The ability of PTEN to localize in the nucleus is essential for both the maintenance of genomic integrity and growth suppression, while PTEN nuclear localization is frequently lost in many cancer specimens.5, 6, 7 PTEN nuclear-cytoplasmic shuttling is regulated by mono-ubiquitination.8 HAUSP/USP7 de-ubiquitinates PTEN, resulting in its cytosolic accumulation.9 HAUSP itself is negatively regulated by the promyelocytic leukemia gene (PML), which in turn is critical for the control of hematopoietic stem cell maintenance and chronic myeloid leukemia (CML) pathogenesis.10, 11 In this study, we demonstrate that nuclear-cytosolic shuttling of PTEN is impaired in CML through an aberrant PML/HAUSP cross-talk.

Materials and methods

Cells

NIH3T3, HEK293 and K562 cells were obtained from ATCC (Manassas, VA, USA); 32D and WEHI-3B were from DSMZ (Braunschweig, Germany). HEK293 cells were used as a standard cell line for biochemical experiments (in particular in ubiquitination assays) due to the highest efficacy of co-transfection of multiple plasmids. To better characterize PTEN compartmentalization, immunofluorescence was performed in NIH3T3 cells, which have an optimal nuclear-cytoplasmic ratio for visualizing cellular localization. BCR-ABL 32D and K562 were selected for use as conventional CML cellular models. Parental 32D cell line was cultured in the presence of Wehi-3B conditioned medium as a source of IL-3.

Plasmids, antibodies and reagents

p210 BCR-ABL (gift from Daley GQ) was cloned into pcDNA-3.1 (Invitrogen, Grand Island, NY, USA, #V790-20) and pBABE-puromycin vectors (gift from Dr Bob Weinberg; Addgene plasmid #1764); BCR-ABL Kinase Dead (K1172R) was generated by site direct mutagenesis; Myc-tagged pRK5 plasmid expressing PTEN was reported elsewhere;9 Myc-tag HAUSP-WT and catalytically inactive HAUSP-CS were reported elsewhere.9, 10, 11, 12 GFP-PTEN constructs were described elsewhere.13 HAUSP mutants were generated by site-directed mutagenesis, as described below (Quikchange II-XL; Stratagene, Santa Clara, CA, USA, #200523). HA-Ub plasmid was described elsewhere.9, 10, 11, 12, 13 Antibodies were BCR (Cell Signaling, Danvers, MA, USA and Milan, Italy, #3902), ABL (Cell Signaling, # 2862), Myc-tag (Cell Signaling, #2276), PTEN (Cascade Bioscience, Winchester, MA, USA, #6H2.1), PML (Santa Cruz, Dallas, TX, USA and Heidelberg, Germany #H-238), β-Actin (Sigma, Santa Cruz, MO, USA, #A5316), HA (Covance, Dedham, MA, USA, #16B12), Phosphotyrosine (Santa Cruz, #7020), Histone H3 (Cell Signaling, #9715), GFP (Cell Signaling, #2555; Santa Cruz, #390394), Tubulin (Cell Signaling, #2146), HAUSP (Santa Cruz, #sc30164) and Vinculin (Abcam, Cambridge, UK, #ab137331). Secondary antibodies were goat anti-mouse IgG-HRP (Santa Cruz, #2005) and goat anti-rabbit IgG-HRP (Santa Cruz, #2004). A concentration of 5 μM imatinib was used for 6 h or the indicated time.

Transfection and retroviral transduction

Expression of BCR-ABL, PTEN and HAUSP was achieved via transfection with Lipofectamine 2000 (Invitrogen) or X-treme GeneHP (Roche, Basel, Switzerland) or transduction with retrovirus. Retrovirus was produced by transfection of pBABE-BCR-ABL in ecotropic phoenix packaging cells (ATCC). Forty-eight hours post transfection, viral media was collected and filtered through a 0.44-μM filter before application to the indicated cells in the presence of 4 μg/ml polybrene. Infection was repeated after 24 h. Forty-eight hours post infection, cells were selected in 5 μg/ml puromycin and stable populations of cells were then used for downstream experiments. Stable K562 cell lines expressing GFP-PTEN constructs and HAUSP constructs were selected using G418 after transfection.

Immunofluorescence and flow cytometry

Immunofluorescence was performed by fixing cells with 4% PFA, permeabilizing cells with 0.1% Triton and blocking with bovine serum albumin. Slides were then incubated overnight with the indicated primary antibodies. Secondary antibodies were Alexafluor-488 (Invitrogen, #A-11078) and Alexafluor-543 (Invitrogen, #A-11030). Cells were mounted on a slide with Vectashield Mounting Media with DAPI (Vector Laboratories, Burlingame, CA, USA, #H-1200). At least 100 cells per experiment were assessed. Statistical analysis was performed on data from three independent experiments. Quantification of apoptosis by flow cytometry was performed as described elsewhere.10

Immunoprecipitation and western immunoblotting

Proteins were extracted with a buffer containing 150 mM NaCl, 1 mM EDTA, 50 mM HEPES (pH 7.5), 1% Triton X-100 and 10% glycerol. For immunoprecipitations, lysates were precleared with normal mouse IgG antibody (Santa Cruz, #2025) or normal rabbit IgG (Santa Cruz, #2027) and Protein G-PLUS-Agarose (Santa Cruz, #sc-2002).

Kinase assay

An in vitro kinase assay with purified proteins was performed as previously described.14 Purified full-length ABL (Invitrogen, #P3049) and full-length His-GST-HAUSP (Invitrogen, #11681-H20BL-50) were resuspended in kinase buffer (50 mM HEPES, 150 mM NaCl, 1 mM MgCl2, 1 mM MnCl2, 10 mM NaF, 1 mM NaVO3, 5% glycerol, 1% NP-40, 1 mM DTT, 1 mM PMSF) and 10 μM ATP (Sigma). After 30 min at 37 °C, proteins were boiled and separated on an 8% SDS–PAGE, transferred onto nitrocellulose, and incubated with anti-phosphotyrosine and anti-HAUSP antibodies.

Immunohistochemistry

Immunohistochemistry experiments were performed on formalin-fixed, paraffin-embedded tissues using an anti-PTEN antibody (Cell Signaling, #9188) according to the manufacturer’s protocols.

Mass spectrometry

After anti-Myc immunoprecipitation, proteins were resolved in an 8% polyacrylamide gel and stained with Coomassie Brilliant Blue. The bands of interest were excised and in-gel tryptic digestion was performed. Briefly, bands were excised and destained with acetone, 100 mM NH4HCO3, dried under vacuum, reduced in DTT and NH4HCO3 and treated with 55 mM iodoacetamide in 100 mM NH4HCO3. Gel pieces were dried again, rehydrated in 20 μl of modified trypsin (Roche) solution (0.01 mg/ml) and digested overnight at room temperature. Peptides were then extracted by using TFA 0.2%. Aliquots of 2 μl of each sample were mixed with an equal volume of α-cyano-4-hydroxycinnamic acid matrix (saturated in 50% acetonitrile) (Sigma-Aldrich, St Louis, MO, USA) and the mixture was dropped onto the MSP 96 target ground steel plate of an MALDI-TOF mass spectrometer (Bruker Daltonics) in reflector mode. A mixture of peptides (peptide calibration standard, Bruker Daltonics (Milan, Italy)) was used for internal mass calibration. A peak list was created for each sample and Mascot Server (version 3.3, Matrix Sciences, London, UK) was used to compare our spectra with Swiss-Prot database. The parameters used were trypsin as an enzyme, Homo sapiens as species, ±100 p.p.m. as a mass tolerance and one as missed cleavage sites.

Mutagenesis

HAUSP mutants were generated using the Quick-change II-XL (Stratagene, #200523) site-directed mutagenesis kit. Primers were Y243F forward primer: 5′-IndexTermGAAAGGCTGTGTTCATGATGCCAAC-3′, reverse primer: 5′-IndexTermGTTGGCATCATGAACACAGCCTTTC-3′; Y213F forward primer: 5′-IndexTermAAGCACACAGGCTTCGTCGGCTTAAAGAAT-3′, reverse primer: 5′-IndexTermATTCTTTAAGCCGACGAAGCCTGTGTGCTT-3′; Y367F forward primer: 5′-IndexTermTTGTGGATTTTGTGGCAGTAGA-3′, reverse primer 5′-IndexTermTCTACTGCCACAAAATCCACAA-3′; Y564F forward primer: 5′-IndexTermGAAGCCCATCTCTTTATGCAAGTGCAGATA-3′, reverse primer: 5′-IndexTermTATCTGCACTTGCATAAAGAGATGGGCTTC-3′; Y411F forward primer: 5′-IndexTermACTGATGAGATTTATGTTTGACCCTCAGAC-3′, reverse primer 5′-IndexTermGTCTGAGGGTCAAACATAAATCTCATCAGT-3′; Y878F forward primer 5′-IndexTermGAAACTTTACTTTCAGCAGCTTAA-3′, reverse primer 5′-IndexTermTTAAGCTGCTGAAAGTAAAGTTTC-3′; and Y947F forward primer 5′-IndexTermAATTGTAAGCTTCAAAATCATTGGTG-3′, reverse primer 5′-IndexTermCACCAATGATTTTGAAGCTTACAATT-3′.

Ubiquitination assay

Ubiquitination assays were performed as previously reported.9 Briefly, HEK293 cells were transfected with HA-Ub plasmid. Cells were then lysed by incubation with two volumes of 2% SDS at 95 °C for 10 min. Then eight volumes of 1% Triton X-100 were added. Sonicated samples were incubated with anti-Myc antibody for immunoprecipitations. Anti-HA antibody was used to detect ubiquitin adducts by western blotting.

Primary CML samples

Primary CML bone marrow samples were collected at the San Luigi Hospital (Orbassano, Italy) with informed consent at the time of diagnosis from five untreated patients and three normal individuals as described elsewhere.10 In particular, bone marrow aspirate, performed for routine diagnostic procedures, was treated accordingly to the Miltenyi Biotec protocol for CD34 purification (Miltenyi Biotec, Bergisch Gladbach, Germany, #130-094-531). Lineage-positive cells were magnetically labeled with a cocktail of biotin-conjugated antibodies and anti-Biotin microbeads. Lineage-negative fractions were then labeled with anti-CD34 and anti-CD38 antibodies and FACS sorted to collect the indicated cellular sub-populations.

Ethics statement

This project was reviewed and approved by the San Luigi Hospital Institutional Ethical Committee (Code #10/2013).

Results

While investigating the expression of PTEN in NIH3T3 cells overexpressing BCR-ABL, we discovered that these cells surprisingly displayed nuclear exclusion of PTEN (Figure 1a). Nuclear exclusion of PTEN was almost completely reversed by the BCR-ABL inhibitor imatinib (Figure 1b). A catalytically inactive ‘kinase-dead’ (KD) BCR-ABL mutant was incapable of modulating PTEN localization, suggesting that BCR-ABL-mediated PTEN localization is dependent on ABL kinase activity (Figure 1b). Similarly, stable BCR-ABL expression in 32D cells triggered nuclear exclusion of PTEN, which was reverted by imatinib treatment and not observed in cells expressing BCR-ABL-KD (Figures 1c and d).

Figure 1
figure1

BCR-ABL induces nuclear exclusion of PTEN. (a) Left panel: NIH3T3 cells were transduced with pBABE-BCR-ABL retrovirus and immunofluorescence on endogenous PTEN was performed to detect PTEN compartmentalization (green). DAPI was used to visualize nuclei (blue). Right panel: western immunoblot on empty vector (1) and BCR-ABL-expressing (2) NIH3T3 cells. Anti-tubulin was used as a loading control. (b) Quantification and statistical analysis of PTEN localization in NIH3T3 cells with and without BCR-ABL expression or imatinib treatment. Where indicated, cells were treated with 1 μM imatinib for 8 h. Data shown are mean+s.d. from three independent experiments. KD, ‘kinase-dead’ mutant. P-values shown were calculated by two-tailed t-test. (c) Parental 32D cell line and stable BCR-ABL expressing cells were stained with PTEN antibody (green) and DAPI (blue). Lower panel: western immunoblot on cells expressing empty vector (1), BCR-ABL (2), BCR-ABL plus treatment with imatinib (3), or KD BCR-ABL. Anti-ABL, anti-phosphotyrosine (P-TYR) or anti-tubulin (loading control) antibodies were used. (d) Quantification and statistical analysis of PTEN localization in 32D cells with and without BCR-ABL expression or imatinib treatment. When indicated, cells were treated with 1 μM imatinib for 8 h Data shown are mean+s.d. from three independent experiments. P-values shown were calculated by two-tailed t-test.

Previous studies have shown that nuclear-cytoplasmic shuttling of PTEN is regulated by mono-ubiquitination. Mono-ubiquitinated PTEN is predominantly nuclear, while de-ubiquitinated PTEN is mostly cytosolic.4 The de-ubiquinase HAUSP is responsible for de-ubiquitination of PTEN, resulting in PTEN translocation to the cytosol.9 Furthermore, PML negatively regulates HAUSP, thereby favoring PTEN nuclear retention.9 Therefore, we hypothesized that BCR-ABL may regulate PTEN localization via PML or HAUSP.

We first tested whether BCR-ABL regulates PTEN de-ubiquitination. The expression of BCR-ABL resulted in a marked reduction in mono-ubiquitinated PTEN (Figure 2a). To determine the involvement of HAUSP in BCR-ABL-induced de-ubiquitination of PTEN, we performed a PTEN ubiquitination assay in HEK293 cells in the presence of wild-type HAUSP or a catalytically inactive HAUSP mutant, HAUSP-CS.9, 10, 11, 12 BCR-ABL-induced de-ubiquitination of PTEN was rescued by HAUSP-CS (Figure 2b).

Figure 2
figure2

BCR-ABL promotes PTEN de-ubiquitination through HAUSP. (a) HEK293 cells were transfected with the indicated plasmids and ubiquitination assessed by immunoprecipitation/western blotting. Asterisks mark ubiquitinated PTEN. (b) Ubiquitination assay in the presence of wild-type HAUSP and CS-HAUSP. Asterisks mark ubiquitinated PTEN. (c) Co-immunoprecipitation of HAUSP and BCR-ABL in transiently transfected HEK293 cells. P-TYR: phosphotyrosine. (d) In vitro kinase assay with purified ABL and HAUSP proteins performed in the absence or presence of 1 μM imatinib treatment for the times indicated. P-TYR: phosphotyrosine. (e) Immunoprecipitation of HAUSP Y243F mutant and anti-phosphotyrosine immunoblot. Where indicated, cells were treated with imatinib for 6 h. (f) Coomassie staining of Myc-HAUSP immunoprecipitation. Red arrows indicate the bands at the sizes expected for BCR-ABL and HAUSP. (g) BCR and ABL peptides in the immunoprecipitate from (f) identified by mass spectrometry.

We next assessed whether BCR-ABL interacts with HAUSP. To this end, we co-transfected Myc-tagged HAUSP and BCR-ABL in HEK293 cells and performed an immunoprecipitation against Myc-HAUSP. BCR-ABL interacted with tyrosine-phosphorylated HAUSP (Figure 2c), suggesting that BCR-ABL may phosphorylate HAUSP.

To test whether HAUSP tyrosine phosphorylation by BCR-ABL is direct, we performed a kinase assay with purified ABL and HAUSP protein. ABL directly phosphorylated HAUSP on tyrosine residues, and this phosphorylation was inhibited by treatment with imatinib (Figure 2d). HAUSP protein sequence analysis revealed a putative recognition motif for phosphorylation by Abelson (I/V/L-Y-X-X-P/F) at HAUSP tyrosine Y243 (242-V-Y-M-M-P-246). The HAUSP Y243F mutant showed significantly reduced BCR-ABL-induced HAUSP phosphorylation, which in turn was completely abrogated by imatinib treatment (Figure 2e). These data strongly suggested that BCR-ABL phosphorylates HAUSP on Y243 as well as additional HAUSP tyrosine residues.

To further define the pattern of HAUSP phosphorylation by BCR-ABL, we co-transfected BCR-ABL and Myc-HAUSP in HEK293 cells, then lysed cells and performed immunoprecipitation with an anti-Myc antibody. After separation by SDS–PAGE and Coomassie staining, we observed two bands whose molecular weight could correspond to HAUSP and BCR-ABL (Figure 2f). The bands were next excised and analyzed by mass spectrometry after in-gel tryptic digestion. The obtained peptide spectra were compared with those collected in the Mascot Database. Our results confirmed that our spectra corresponded respectively to BCR-ABL (5% of coverage of total sequence) and HAUSP (27% of coverage of total sequence). These data further supported the notion that BCR-ABL and HAUSP physically interact (Figure 2g). Furthermore, additional mass spectrometry analysis confirmed that HAUSP is phosphorylated on several other tyrosine residues that are highly conserved among species, such as Y213, Y411, Y564, Y878 and Y947 (Figures 3a and b). Therefore, we generated additional HAUSP tyrosine mutants to establish the contribution of other tyrosines to HAUSP phosphorylation by BCR-ABL. Expression of these mutants demonstrated that Y878 and Y947 significantly contribute to the tyrosine phosphorylation of HAUSP (Figures 3c and d; Supplementary Figure S1). Interestingly, these two residues are located in the C-terminal regulatory HUBL domain of HAUSP, which has recently been shown to functionally regulate the activity of HAUSP.15

Figure 3
figure3

BCR-ABL phosphorylates HAUSP on tyrosine residues. (a) Mass spectrometry analysis of HAUSP protein in the presence of BCR-ABL revealed additional phosphotyrosine residues in HAUSP. HAUSP cleavage peptides recovered from the MALDI spectrum. Peak shifts observed in HAUSP peptides correspond to 80 Da, the weight of phosphate group. M/z: mass/charge ratio; observed (obs) versus calculated (calc) masses. (b) Schematic representation of HAUSP protein with tyrosine residues highlighted that had evidence of phosphorylation in the mass spectrometry data. (c) Immunoprecipitation of HAUSP mutants in the presence of BCR-ABL with anti-Myc antibodies (HAUSP) and western immunoblotting using an anti-phosphotyrosine (pTyr) antibody. TM refers to the triple-mutant Myc-HAUSP-Y243F/Y878F/Y947F. (d) Quantification of the ratio of tyrosine-phosphorylated HAUSP to total HAUSP in immunoprecipitation of HAUSP mutants in the presence of BCR-ABL. Data shown are mean+s.d. of three independent experiments after quantification of relative levels from densitometry of western blot data. (e) Mono-ubiquitination assay, performed with HAUSP-WT and HAUSP-TM. Red asterisks indicate ubiquitinated-PTEN bands. (f) PTEN compartmentalization in BCR-ABL-expressing NIH3T3 cells transfected with the HAUSP-WT, HAUSP-TM (Y243F/Y878F/Y947F) and catalytically inactive HAUSP-CS. Data shown are mean+s.d., n>100 cell per condition. P-values shown were calculated by two-tailed t-test.

To determine whether tyrosine phosphorylation affects HAUSP activity toward PTEN, we next examined PTEN mono-ubiquitination in the presence of HAUSP-WT and Y243F/Y878F/Y947F mutant HAUSP (Triple Mutant, TM) in HEK293 cells. Indeed, HAUSP-TM was less effective at inducing PTEN de-ubiquitination than WT HAUSP (Figure 3e). These data indicate that BCR-ABL phosphorylation of HAUSP modulates HAUSP’s deubiquitinase activity toward PTEN.

To further assess the involvement of HAUSP in BCR-ABL-induced PTEN delocalization, NIH3T3 cells were transfected with HAUSP-WT, -CS and -TM. Catalytically inactive HAUSP-CS, and to a lesser extent, HAUSP-TM, antagonized PTEN delocalization induced by BCR-ABL (Figure 3f; Supplementary Figure S2).

Next, we determined whether forced expression of PTEN in different cellular compartments or the expression of HAUSP mutants would impact BCR-ABL-driven cellular proliferation. We stably expressed GFP-PTEN WT, GFP-PTEN-NES (only cytosolic) or GFP-PTEN-NLS (only nuclear) constructs13 and HAUSP-WT, HAUSP-CS or HAUSP-TM9, 10, 11, 12 in BCR-ABL-positive cell lines. Since HAUSP regulates p53 protein stability and compartmentalization, we assessed the biological effects of these constructs in the CML blast crisis cell line, K562. K562 cells are dependent on the tyrosine kinase activity of BCR-ABL, express endogenous PTEN and are characterized by p53 loss of function. Stable expression of GFP-PTEN-WT and GFP-PTEN-NLS had similar effects on the growth arrest of K562 cells, while GFP-PTEN-NES was surprisingly less effective in the inhibition of proliferation (Figure 4a). These data suggest that nuclear PTEN function contributes substantially to the overall ability of PTEN to oppose BCR-ABL-mediated proliferation. Stable expression of inactive HAUSP-CS and, to a lesser extent, HAUSP-TM was also associated with a reduction in cellular proliferation of K562 (Figure 4b). These observations suggest that in a BCR-ABL addicted cell line, nuclear PTEN and the regulatory protein HAUSP have an essential role in mediating BCR-ABL-induced proliferation.

Figure 4
figure4

Effect of PTEN and HAUSP mutants on proliferation of BCR-ABL-positive K562 cells. (a) Upper panel: growth curve of stable K562 cell lines expressing GFP-PTEN-WT, GFP-PTEN-NES and GFP-PTEN-NLS. Cells were plated and then quantified by counting at each time point shown. Lower panel: western immunoblot on K562 stable cell lines to verify accurate separation of nuclear and cytoplasmic cell fractions. C, cytoplasmic; N, nuclear; Ø, empty vector. Anti-vinculin and anti-Histone H3 antibodies were used as loading controls and to verify adequate separation of cytoplasmic and nuclear fractions. (b) Upper panel: growth curve of K562 cells expressing HAUSP-WT, HAUSP-CS and HAUSP-TM. Lower panel: western immunoblot of HAUSP-expressing K562 stable cell lines to confirm expression of HAUSP.

To determine the relevance of these findings to human disease, we studied the cellular compartmentalization of PTEN in human CML samples. Primary cells collected at the moment of diagnosis, in five untreated patients, were FACS sorted to obtain LinCD34+CD38 (stem/progenitor cells, SPCs), LinCD34+CD38+ (committed progenitors) and differentiated Lin+ cells (Supplementary Figure S3). PTEN showed a diffuse cytosolic and nuclear localization in the normal bone marrow cells (Figure 5a). Only rare differentiated cells showed PTEN nuclear exclusion (data not shown). In contrast, CML samples showed a dramatic nuclear exclusion of PTEN in hematopoietic LinCD34+CD38+ progenitor cells. Surprisingly, however, CML LinCD34+CD38 SPCs retained nuclear PTEN even in the presence of BCR-ABL (Figure 5a). We further assessed the expression of PTEN in unsorted CML samples by immunohistochemistry. PTEN immunohistochemistry revealed weaker expression in the myeloid compartment of CML samples compared with normal bone marrow. Importantly, CML myeloid cells were characterized by a predominant nuclear exclusion of PTEN (Figures 5b and c). Together, these observations confirm that BCR-ABL regulates PTEN localization in CML progenitor cells.

Figure 5
figure5

PTEN compartmentalization in primary human CML samples. (a) Left panel: representative immunofluorescence images from LinCD34+CD38 SPCs, LinCD34+CD38+ progenitors, and Lin+ primary cells collected from normal bone marrow or CML samples and stained for PTEN (green). Propidium iodide (red) was used to visualize nuclei. Right panel: quantification of PTEN cellular compartmentalization. CML samples, n=5. Normal samples, n=3. Data shown are mean+s.d. of >50 cells analyzed from each sample. (b) Immunohistochemistry using an anti-PTEN antibody performed on paraffin-embedded bone marrow from two CML or normal bone marrow. Scale bar, 50 μM. Red arrows show representative progenitors cells with diffuse PTEN compartmentalization in normal bone marrow and nuclear exclusion in CML samples. (c) Zoom-in of images from (b) showing details of PTEN immunohistochemistry on cells in normal bone marrow and CML. Red arrows show representative progenitors cells with diffuse PTEN compartmentalization in normal bone marrow and nuclear exclusion in CML samples.

However, we also unexpectedly discovered that CML SPCs appear to be protected from BCR-ABL-induced PTEN delocalization. This observation may intriguingly be related to the reason why CML leukemic stem cells are resistant to tyrosine kinase inhibitors and could be of great importance from a therapeutic standpoint.16, 17, 18 We therefore investigated the possible mechanism that could explain this surprising observation. We have previously shown that PML protein levels in normal and leukemic hematopoiesis are maximally expressed in the stem cell compartment, with a subsequent reduction during differentiation.10 Similarly, the number of PML nuclear bodies decreases as differentiation proceeds. PML acts as an indirect regulator of PTEN localization though its ability to oppose HAUSP function.9 We therefore hypothesized that PML could oppose the effects of BCR-ABL on hematopoietic stem cell while a reduction in PML levels during hematopoietic stem cell differentiation could favor the change in PTEN localization triggered by the fusion protein. Indeed, we observed that PML levels directly correlate with PTEN nuclear localization (Figure 6a; Supplementary Figures S4A and B). In LinCD34+CD38 cells nuclear PTEN was associated with a higher number of PML nuclear bodies, while in the more differentiated cells, cytosolic PTEN was associated with a lower number of PML nuclear bodies (Figure 6a). In line with these data, treatment of LinCD34+CD38 cells with arsenic trioxide, which promotes PML degradation,10 triggered PTEN nuclear exclusion (Figure 6b).

Figure 6
figure6

PTEN compartmentalization is regulated by PML levels and by arsenic trioxide treatment in CML. (a) Left panel: LinCD34+CD38 SPCs and LinCD34+CD38+ progenitors were sorted using FACS and stained with PTEN (green) and PML (red) antibodies. Right panel: numbers of PML nuclear bodies and statistical analysis of PTEN cellular compartmentalization. Where indicated, cells were treated with 0.15 μM arsenic trioxide (ATO) for 72 h. Data shown are mean+s.d. of n>50 cells analyzed for each condition. P-values were calculated using a two-tailed t-test. (b) Left panel: LinCD34+CD38 cells from primary CML samples were treated with ATO for 72 h then cells were fixed and stained with anti-PTEN (green) or anti-PML (red) antibodies. Right panel: PML levels detected by western immunoblotting of primary cells with and without ATO treatment. (c) FACS analysis using Annexin V and propidium iodide (PI) in primary cells, cultured as reported in (b), then treated with 5 μM imatinib and ATO alone or in combination for additional 16 h. (d) Model of the BCR-ABL/PML/HAUSP/PTEN network in CML SPCs (left panel), progenitors (middle panel) or SPCs after ATO treatment (right panel). In the stem cell pool, high levels of PML expression inhibit HAUSP de-ubiquitination of PTEN (left). In the progenitor cells, PML levels are substantially reduced leading to HAUSP activity toward PTEN (middle). Treatment of the stem cell pool with ATO (right) causes PML degradation, thereby releasing inhibition of HAUSP.

To further determine the effects of arsenic trioxide on CML SPCs, we next treated LinCD34+CD38 cells with arsenic trioxide for 72 h, as reported.10 We have previously shown that arsenic trioxide treatment promotes the cell-cycle entry of leukemic stem cells and that this treatment could be therefore used to target the quiescent stem cell pool.10 As shown in Figure 6b, PML levels were reduced, with a consequent nuclear exclusion of PTEN. Critically, arsenic trioxide treatment of CML SPCs promoted sensitivity to imatinib, as assessed by induction of apoptosis (Figure 6c).

Discussion

Our data demonstrate that BCR-ABL promotes the exclusion of PTEN from the nucleus. We observed that BCR-ABL is able to directly regulate PTEN compartmentalization though an interaction with HAUSP, which de-ubiquitinates PTEN. BCR-ABL regulates HAUSP through direct tyrosine phosphorylation of several residues. PTEN nuclear exclusion may in turn result in a proliferative advantage and aberrant expansion of hematopoietic progenitors in CML. By contrast, PTEN nuclear exclusion by BCR-ABL is countered by PML in the stem cell compartment, where it is highly expressed. By opposing the function of HAUSP, PML maintains the quiescence of leukemia stem cells. Thus, whereas PML opposes HAUSP, favoring PTEN nuclear function, BCR-ABL acts in opposition to PML function to promote PTEN nuclear exclusion (Figure 6d).

The PML/HAUSP/PTEN network may also explain why CML is not always curable by tyrosine kinase inhibitors such as imatinib. It is indeed well accepted that CML LSCs are biologically resistant to tyrosine kinase inhibitors, in a BCR-ABL-independent manner.16, 17, 18 CML is composed of oncogene-addicted progenitor cells, differentiated cells and an LSC pool whose survival and maintenance are independent of BCR-ABL activity, allowing LSCs to resist treatment with the BCR-ABL inhibitor imatinib.16, 17, 18

Thus, our data attribute a critical and druggable role to the PML/HAUSP/PTEN pathway in CML pathogenesis and suggest combination arsenic trioxide and imatinib therapy as a strategy for elimination of CML LSCs and the treatment of CML.

References

  1. 1

    Peng C, Chen Y, Yang Z, Zang H, Osterby L, Rosmarin AG et al. PTEN is a tumor suppressor in CML stem cells and Bcr-Abl induced leukemia in mice. Blood 2010; 115: 626–635.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  2. 2

    Yilmaz OH, Valdez R, Theisen BK, Guo W, Ferguson DO, Wu H et al. Pten dependence distinguishes haematopoietic stem cells from leukaemia-initiating cells. Nature 2006; 441: 475–482.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  3. 3

    Zhang J, Grindley JC, Yin T, Jayasinghe S, He XC, Ross JT et al. PTEN maintains haematopoietic stem cells and acts in lineage choice and leukaemia prevention. Nature 2006; 441: 518–522.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  4. 4

    Salmena L, Carracedo A, Pandolfi PP . Tenets of PTEN tumor suppression. Cell 2008; 133: 403–414.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  5. 5

    Song MS, Carracedo A, Salmena L, Song SJ, Egia A, Malumbres M et al. Nuclear PTEN regulates the APC-CDH1 tumor-suppressive complex in a phosphatase-independent manner. Cell 2011; 144: 187–199.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  6. 6

    Alice Berger H, Knudson AG, Pandolfi PP . A continuum model for tumour suppression. Nature 2011; 476: 163–169.

    Article  Google Scholar 

  7. 7

    Shen WH, Balajee AS, Wang J, Wu H, Eng C, Pandolfi PP et al. Essential role for nuclear PTEN in maintaining chromosomal integrity. Cell 2007; 128: 157–170.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  8. 8

    Salmena L, Pandolfi PP . Changing venues for tumour suppression: balancing destruction and localization by monoubiquitylation. Nat Rev Cancer 2007; 7: 409–413.

    CAS  Article  PubMed  Google Scholar 

  9. 9

    Song MS, Salmena L, Carracedo A, Egia A, Lo-Coco F, Teruya-Feldstein J et al. The deubiquitinylation and localization of PTEN are regulated by a HAUSP–PML network. Nature 2008; 455: 813–817.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  10. 10

    Ito K, Bernardi R, Morotti A, Matsuoka S, Saglio G, Ikeda Y et al. PML targeting eradicates quiescent leukaemia-initiating cells. Nature 2008; 453: 1072–1078.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  11. 11

    Ito K, Bernardi R, Pandolfi PP . A novel signaling network as a critical rheostat for the biology and maintenance of the normal stem cell and the cancer-initiating cell. Curr Opin Genet Dev 2009; 19: 51–59.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  12. 12

    Lim SK, Shin JM, Kim YS, Baek KH . Identification and characterization of murine mHAUSP encoding a deubiquitinating enzyme that regulates the status of p53 ubiquitination. Int J Oncology 2004; 24: 357–364.

    CAS  Google Scholar 

  13. 13

    Trotman LC, Wang Z, Alimonti A, Chen Z, Teruya-Feldstein J, Yang H et al. Ubiquitination regulates PTEN nuclear import and tumor suppression. Cell 2007; 128: 141–156.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  14. 14

    Ferretti R, Palumbo V, Di Savino A, Velasco S, Sbroggio M, Sportoletti P . Morgana/chp-1, a ROCK inhibitor involved in centrosome duplication and tumorigenesis. Dev Cell 2010; 18: 486–495.

    CAS  Article  PubMed  Google Scholar 

  15. 15

    Faesen AC, Dirac AM, Shanmugham A, Ovaa H, Perrakis A, Sixma TK . Mechanism of USP7/HAUSP activation by its C-terminal ubiquitin-like domain and allosteric regulation by GMP-synthetase. Mol Cell 2011; 44: 147–159.

    CAS  Article  PubMed  Google Scholar 

  16. 16

    Corbin AS, Agarwal A, Loriaux M, Cortes J, Deininger MW, Druker B . Human chronic myeloid leukemia stem cells are insensitive to imatinib despite inhibition of BCR-ABL activity. J Clin Invest 2011; 121: 396–409.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  17. 17

    Hamilton A, Helgason V, Schemionek M, Zhang B, Myssina S, Allan EK et al. Chronic myeloid leukemia stem cells are not dependent on Bcr-Abl kinase activity for their survival. Blood 2012; 119: 1501–1510.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  18. 18

    Perl A, Carroll M . BCR-ABL kinase is dead; long live the CML stem cell. J Clin Invest 2011; 121: 22–25.

    CAS  Article  PubMed  Google Scholar 

Download references

Acknowledgements

We thank all members of the Pandolfi laboratory for comments and discussion and Thomas Garvey for critical editing of the manuscript. We appreciate the help of Dr Enrico Bracco, from the Department of Clinical and Biological Sciences at the University of Turin, for spectrometry analysis. AM was supported by postdoctoral fellowship for research abroad. AHB is supported by NIH grant R01CA142787. This work was supported by NIH grants to PPP and AIRC grant to GS.

Author contributions

The experiments were conceived and designed by AM, CP, SC, AHB, GS and PPP. Mass Spectrometry analysis was performed in collaboration with BP. Experiments were performed by AM, CP, SC, UF and BP. The paper was written by AM, AHB and PPP.

Author information

Affiliations

Authors

Corresponding author

Correspondence to P P Pandolfi.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies this paper on the Leukemia website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Morotti, A., Panuzzo, C., Crivellaro, S. et al. BCR-ABL disrupts PTEN nuclear-cytoplasmic shuttling through phosphorylation-dependent activation of HAUSP. Leukemia 28, 1326–1333 (2014). https://doi.org/10.1038/leu.2013.370

Download citation

Keywords

  • chronic myeloid leukemia
  • BCR-ABL
  • HAUSP
  • PTEN
  • PML

Further reading