Biochimica et Biophysica Acta 1843 (2014) 2886–2899

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r .com/ locate /bbamcr

PTPN13 regulates cellular signalling and β-catenin function duringmegakaryocytic differentiation

José L. Sardina a,b, Guillermo López-Ruano a,c, Rodrigo Prieto-Bermejo a,c, Beatriz Sánchez-Sánchez a,c,Alejandro Pérez-Fernández a,c, Luis Ignacio Sánchez-Abarca d, José Antonio Pérez-Simón d, Luis Quintales b,Jesús Sánchez-Yagüe a, Marcial Llanillo a,c, Francisco Antequera b, Angel Hernández-Hernández a,c,⁎a Department of Biochemistry and Molecular Biology, University of Salamanca, Salamanca, Spainb IBFG, Instituto de Biología Funcional y Genómica, CSIC, Salamanca 37007, Spainc IBSAL (Instituto de Investigación Biomédica de Salamanca), Salamanca 37007, Spaind Department of Hematology, Hospital Universitario Virgen del Rocío/IBIS/CSIC/University of Seville, Spain

⁎ Corresponding author at: Departamento de BioquímicEdificio Departamental, Plaza Doctores de la Reina s/n, 370923 294 781; fax: +34 923 294 579.

E-mail address: angelhh@usal.es (A. Hernández-Herná

http://dx.doi.org/10.1016/j.bbamcr.2014.08.0140167-4889/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 5 May 2014Received in revised form 8 August 2014Accepted 26 August 2014Available online 3 September 2014

Keywords:MegakaryopoiesisPTPN13β-CateninHematopoiesisWnt signalling

PTPN13 is a high-molecular weight intracellular phosphatase with several isoforms that exhibits a highlymodular structure. Although in recent years different roles have been described for PTPN13, we are still farfrom understanding its function in cell biology. Here we show that PTPN13 expression is activated duringmegakaryocytic differentiation at the protein and mRNA level. Our results show that the upregulation ofPTPN13 inhibits megakaryocytic differentiation, while PTPN13 silencing triggers differentiation. The ability ofPTPN13 to alter megakaryocytic differentiation can be explained by its capacity to regulate ERK and STATsignalling. Interestingly, the silencing of β-catenin produced the same effect as PTPN13 downregulation. Wedemonstrate that both proteins coimmunoprecipitate and colocalise. Moreover, we provide evidence showingthat PTPN13 can regulate β-catenin phosphorylation, stability and transcriptional activity. Therefore, the abilityof PTPN13 to control megakaryocytic differentiation must be intimately linked to the regulation of β-cateninfunction. Moreover, our results show for the first time that PTPN13 is stabilised upon Wnt signalling, whichmakes PTPN13 an important player in canonical Wnt signalling. Our results show that PTPN13 behaves asan important regulator of megakaryocytic differentiation in cell lines and also in murine haematopoieticprogenitors. This importance can be explained by the ability of PTPN13 to regulate cellular signalling, andespecially through the regulation of β-catenin stability and function. Our results hold true for differentmegakaryocytic cell lines and also for haematopoietic progenitors, suggesting that these two proteins may playa relevant role during in vivo megakaryopoiesis.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

PTPN13 (a.k.a. PTP-Bas, PTPL1, hPTP1E, FAP-1, and PTP-BL in mice,hereafter referred to as mPTPN13) is a high-molecular weight intracel-lular phosphatase with several isoforms that shows a highly modularstructure. Besides the catalytic PTP domain at the C-terminus, it has aKINDdomain, a FERMdomain and 5 PDZ domains. This structure allowsPTPN13 to maintain numerous interactions with a variety of ligands(reviewed in [1]), which, besides its catalytic ability, suggests an impor-tant role as a scaffolding protein. PTPN13 is broadly distributed indifferent tissues [1] and its expression is regulated throughout mousedevelopment. Its mRNA is ubiquitously expressed in early embryos,

a y BiologíaMolecular, Lab. 122,07 Salamanca, Spain. Tel.: +34

ndez).

but at later stages its distribution seems to be restricted to epithelialor neuronal tissues [2], suggesting an important developmental role.In zebrafish, the knockdown of PTPN13 causes severe defects in cellmigration during gastrulation [3]. Deletion of the mPTPN13 catalyticdomain hampers motor nerve repair in mice [4], while mPTPN13-deficient mice show an enhanced differentiation of T helper cells [5].Although it has been suggested that PTPN13 would be involved inapoptosis, cytokinesis and cell cycle progression (reviewed in [1,6]),its role in cell biology remains a puzzle. Moreover, several studieshave demonstrated the involvement of PTPN13 in cancer. Its expressionis altered in different types of cancer, and it seems that, depending onthe context, this protein could play a role as a tumour suppressor ortumour promoter (reviewed in [7]).

The canonical Wnt pathway involves the stabilization of β-catenin,an intriguing protein because of its dual independent functions as acell adhesion molecule and as a transcription cofactor [8]. The phos-phorylation of β-catenin at serine and threonine residues by GSK3 and

2887J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

CK1 targets the protein for proteasome degradation [9]. However, β-catenin can also be phosphorylated at tyrosine residues; this has beenlinked to the regulation of its interaction with cadherin [10], its nuclearlocalization [11] and its transcriptional activity [12]. This would implythat β-catenin function could be regulated through dephosphorylationby protein tyrosine phosphatases (PTPs). In fact, it has been suggestedthat β-catenin could be regulated by different PTPs [10], includingPTPN13 [13,14].

In the present work we show that PTPN13 and β-catenin are impor-tant regulators ofmegakaryocytic differentiation in bothmegakaryocyt-ic cell lines and primary mouse haematopoietic progenitor cells. Theexpression of PTPN13 is activated during megakaryocytic differentia-tion and its upregulation or downregulation alters this process; thiscan be explained in terms of the ability of PTPN13 to affect the signallingpathways that are important for differentiation. Moreover, we haveobserved that PTPN13 coimmunoprecipitates and colocalise with β-catenin in megakaryocytic cells, and that its expression is activatedupon Wnt3a stimulation. Interestingly, the downregulation of β-cateninproduced a phenotype that was very similar to that produced by thedownregulation of PTPN13. We show that in this scenario β-cateninstability and function are regulated by PTPN13, suggesting that theimportance of PTPN13 for megakaryocytic differentiation is intimatelylinked to the regulation of β-catenin.

2. Materials and methods

2.1. Reagents

Cell culture media and supplements were from LONZA (Barcelona,Spain). Thrombopoietin (TPO) and IL 1β were from Cellgenix GmbH(Freiburg, Germany). Phorbol 12-myristate 13-acetate (PMA), dimethylsulfoxide (DMSO), TRI reagent, anti-β tubulin and MTT were fromSigma-Aldrich Spain. Antibodies against PTPN13, p-ERK, ERK 1, Stat3,Stat5, GATA-1, β-Catenin, β-Catenin and p-β-Catenin (S33, Y86 orY654) were from Santa Cruz Biotechnology, CA, USA. Antibodies againstp-Stat3 (Y705) and p-Stat5 (Y694)were fromCell signalling Technology,Danvers MA, USA. Human CD41 and CD61 antibodies and the apoptosisdetection kit were from Inmunostep, Salamanca, Spain. Antibodies todetect murines CD41 and CD61 were from eBioscience, Barcelona,Spain. Donkey anti-rabbit Alexa 488 and goat anti-mouse Alexa 568 andProlong Gold Antifade Reagent were from life Technologies, Madrid,Spain. Polyethylenimine, Linear (MW 25,000) was from Polysciences,Inc. (Eppelheim, Germany). G418 sulphate was from Calbiochem,Darmstadt, Germany. RNAspin mini kit was from GE Healthcare LifeSciences Europe (Barcelona, Spain). RevertAid M-MuLV retrotransciptaseand ribonuclease inhibitor were from Fermentas (Barcelona, Spain).

2.2. Cell lines

The human erythroleukemic HEL cells were grown in RPMI mediumsupplemented with 10% FBS, 100 U/ml penicillin, 100 U/ml streptomy-cin and 2 mM L-glutamine. The murine cell line L8057 was grown in50% RPMI/50% IMDM, 15% FBS, 100 U/ml penicillin, 100 U/ml strepto-mycin and 2 mM L-glutamine. Megakaryocytic differentiation wasinduced by 20 nM PMA (K562 and HEL cells) or 100 nM PMA (L8057cells) treatment. pHM6-PTPL1 and pHM6-PTPL1-C/S plasmids [15]were used to generate stable cell lines over-expressing wild-typePTPN13 or a catalytically inactive mutant. Cells were transfected byelectroporation and selected with G418 as reported previously [16].

2.3. Lentiviral production for RNA interference

Two different sequences were designed to generate small hairpinRNAs (shRNA) in order to interferewith gene expression. The sequencesof shRNA1 and shRNA2 for each target were as follows: PTPN13: 5′-CCATGAAGATTCTGATAAA-3′ and 5′-TGGACGAGTTCTAGAATTA-3′; β-

Cat: 5′-CCATGGAACCAGACAGAAA-3′ and 5′-CCACTAATGTCCAGCGTTT-3′; mPTPN13: 5′-GACCGAATTCGAGAGAGAT-3′ and 5′-CAAAGACGATTCCACTTAC-3′; mβ-Cat: 5′-AGGACAAGCCACAGGATTA-3′ and 5′-GCACCATGCAGAATACAAA-3′. An oligonucleotide against firefly luciferasewas used as a control [17]. Homology to unrelated genes was discardedby Basic local alignment search tool analysis. Target sequences werecloned into pLVTHM, as described previously [16]. For lentivirus produc-tion, the packaging vectors pMD2G and psPAX2 together with pLVTHMwere transfected into 293T with 0.05 mg/ml polyethylenimine in150 mM NaCl. Viral supernatants were harvested at 48 h and 72 hafter transfection and concentrated by ultracentrifugation at 76,000 ×gfor 2 h at 16 °C. Precipitated viruseswere resuspended in a small volumeand the viral titre was measured by the infection of 293T cells.

2.4. Cell line transduction with lentiviruses

HEL and L8057 cell lines were transduced with lentiviruses in low-serum medium (1% FBS) by spinoculation (centrifugation at 1800 ×g,90 min at 32 °C), and viruses were removed 24 h later. The efficiencyof transduction was monitored by GFP expression. A multiplicity ofinfection (MOI) between 50 and 100 was used in order to achievecomplete infection of all target cells.

2.5. Megakaryocytic differentiation analyses

Megakaryocytic differentiationwas studied via analysis ofmegakaryo-cytic markers (CD41 and CD61) by flow cytometry and May–Grünwald–Giemsa staining, as described elsewhere [16]. Acetylcholinesterase activ-itywas used as amarker ofmegakaryocytic differentiation inmurine cells.Activity was measured in permeabilized L8057 cells with 0.2% Triton X-100, as described previously [18]. HEL cells treated with 20 nM PMAduring 7 days were fixed with 2.5% paraformaldehyde (15 min, 4 °C),permeabilized with 70% ethanol (30 min, 4 °C), and stained withpropidium iodide (PI) (50 μg/ml) in the presence of RNAse (50 μg/ml)(1 h, 37 °C). Cells were acquired with a Becton-Dickinson, FACSCaliburcytometer using CellQuest software (BD Biosciences), to assess DNAcontent.

2.6. Immunoblotting and immunoprecipitation

Cells resuspended in MLB lysis buffer (25 mM HEPES, pH 7.5,150 mM NaCl, 1% Igepal, 10% glycerol, 10 mM MgCl2, 1 mM EDTA,25 mM NaF, 1 mM Na2VO4, plus proteinase inhibitors) were incubatedon ice for 20 min, and non-soluble material was eliminated by centrifu-gation. Protein concentrations were determined using the Bradfordassay. The samples were then subjected to sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and the proteins weretransferred onto polyvinylidene fluoride (PVDF) membranes, asdescribed previously [16]. Quantification of bands was performedby densitometry as previously described [19], and it is shown belowthe blots. PTPN13 or β-catenin was immunoprecipitated with specificantibodies and the proteins present in the immunoprecipitate wereanalysed by immunoblotting.

2.7. Quantitative RT-PCR

RNAwas extractedwith TRI reagent (SIGMA). 200U of RevertAidM-MuLV retrotransciptase was used to generate cDNA. The RT reactionwas performed at 37 °C for 1 h in the presence of 20 U of ribonucleaseinhibitor. The reactions were run for 40 cycles at the appropriateannealing temperature for each pair of oligos on a StepOne Real-TimePCR System (Applied Biosystems). Analysis of the PCR data wasperformed with the comparative CT method (ΔΔCT) [20], usingactin as endogenous control. The oligo sequences were as follows:Human actin: Forward: 5′-CACCACACCTTCTACAATGA-3′, Reverse:5′-ACATGATCTGGGTCATCTTC-3′; PTPN13: Forward: 5′-AGTAAGCC

2888 J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

TAGCTGATCCTG-3′, Reverse: 5′-TGGATCTTTTCAACATCTGA-3′; β-Cat: Forward: 5′-TTGGTTCACCAGTGGATTCT-3′, Reverse: 5′-AATTTGAAGGCAGTCTGTCG-3′.

2.8. Cell viability and proliferation analyses

Cell viability was determined by staining with annexin V-PE/7-AAD,and proliferation was analysed by the MTT assay as described [16].

2.9. Immunocytochemistry

PTPN13, β-catenin and CD61 expressions were analysed by immu-nocytochemistry as described previously [21]. Cells were fixed with2.5% paraformaldehyde (30 min, RT), permeabilized with 0.1% TritonX-100 (10 min, RT), and blocked with 1% BSA (10 min, RT). Then, cellswere incubated with primary antibodies, rabbit anti-PTPN13 (1:50),mouse anti-β-catenin (1:50), or mouse anti-CD61 (1:50) (2 h, RT).Donkey anti-rabbit Alexa 488 (1:4000) and goat anti-mouse Alexa568 (1:4000) were used as secondary antibodies (1 h, RT). Cells weremounted with Prolong Gold Antifade reagent (Invitrogen). Imageswere taken with a Leica TCS SP5 Confocal Laser Scanning Microscope,using LAS AF software (Leica Application Suite Advanced Fluorescence).

2.10. Transcriptional analyses

A segment of 918 bp corresponding to the PTPN13/JNK3 (MAPK10)bidirectional promoter was amplified by PCR and cloned into the pGL3luciferase reporter plasmid. The forward and reverse oligo sequenceswere as follows: 5′ GGGGTCGTCGGCTGCATTTTCAAC 3′ and 5′ GGAGCGAAAGCAGAGAGAGGAGGAA 3′. This plasmid was transfected byelectroporation in HEL cells. Promoter activity was analysed bymeasur-ing the luciferase activity of transfected cells. pCMV-βGal was used as atransfection control.

The luciferase TOP-FLASH reporter was used to analyse β-catenintranscriptional activity [12]. A stable β-catenin mutant was generatedby site-directed mutagenesis of S33Y and cloned into the pWPI vector.The reporter plasmid was cotransfected into 293T cells in the presenceor absence of pWPI-β-catenin S33Y. pHM6-PTPL1 and pHM6-PTPL1-C/S plasmids [15] were used to over-express wild-type PTPN13 or thecatalytically inactive mutant respectively. To interfere with PTPN13expression, PTPN13-shRNA2 cloned into pLVTHMwas used, as describedabove. Luciferase activitywas normalisedwith respect toβ-galactosidasetransfection control as done elsewhere [22].

2.11. Bone marrow cell purification and lentiviral transduction

All animal protocols were approved by the University of SalamancaAnimal Care and Use Committee and were approved by the BioethicsCommittee of the University of Salamanca. C57BL/6 mice were pur-chased from the University of Salamanca Animal Facility Unit. Lineagemarker-depleted (Lin−) cells purified from complete bone marrowfrom mice by magnetic sorting, using the mouse Lineage cell depletionkit (Miltenyi Biotec), according to the manufacturer's instructions. Lin−

cell lentiviral transduction was conducted by spinoculation as reportedabove, with MOIs between 200 and 300 per cell, in IMDM, 20% FBS,100 U/ml penicillin, 100 U/ml streptomycin and 2 mM L-glutamine,supplemented with 10 ng/ml mSCF, 10 ng/ml mIL3, 10 ng/ml mTPOand 10 ng/ml mFlt3-L [23]. 24 h later, after removal of the viruses, cellswere subjected to treatment with 100 ng/ml TPO and 10 ng/ml IL1βover 6 or 12 days to induce ex vivo megakaryocytic differentiation.Lentiviral transduced cells were identified byGFP expression, anddiffer-entiation analysis was performed in these cells, analysing the levelsof megakaryocytic markers (CD41, CD61 and CD42a), ploidy, and thecolony forming ability. Colony forming assays were performed asdescribed before [19], in the presence of 100 ng/ml TPO and 10 ng/mlIL1β. The number of megakaryocytic colonies was counted on day 14.

2.12. Statistical analyses

Data are expressed as means ± SD. Statistical analyses wereperformed using Student's t-test or analysis of variance (ANOVA)followed by Scheffé test. Differenceswere considered statistically signif-icant when p b 0.05.

3. Results

3.1. The upregulation of PTPN13 hinders megakaryocytic differentiation

PTPN13 is highly expressed in human platelets (our unpublisheddata), which suggests a role for this protein in platelet physiology orduring megakaryocytic differentiation. Human HEL cells are often usedas a model to study megakaryocytic differentiation after stimulationwith low doses of phorbol esters [16]. We found that PTPN13 proteinexpression increased during megakaryocytic differentiation in HELcells (Fig. 1A). To determine whether these results were due to proteinstabilization or to regulation at transcription level, we analysed PTPN13mRNA levels and the activity of the ptpn13 gene promoter. Our resultsshowed that themRNA levels increased rapidly duringHEL cell differen-tiation (Supplementary Fig. 1A) and, in agreement with this, we alsodetected an increase in promoter activity in HEL cells upon PMA stimu-lation (Fig. 1B). Together, these data suggest that ptpn13 gene transcrip-tion is activated during megakaryocytic differentiation.

To test the involvement of PTPN13 inmegakaryocytic differentiationwe generated stable cell lines that overexpressed the protein (Fig. 1C).PTPN13 overexpression significantly decreased the levels of the CD41and CD61 megakaryocytic markers, while the expression of CD45haematopoietic marker remained unchanged (Fig. 1D and E). Neithercell viability (Supplementary Fig. 1B) nor the proliferative capacity(Supplementary Fig. 1C) was modified by PTPN13 overexpression.Morphological analysis revealed that, after PMA stimulation, cells withPTPN13 upregulation showed smaller size (Fig. 1F), and reducedexpression of CD61 at the cell surface (Fig. 1G), reflecting a less differen-tiated phenotype. The frequency of polyploid cells was also reduced byPTPN13 overexpression (Fig. 1H). Moreover, these effects did not seemto be completely dependent on the catalytic activity of the protein, sincethe overexpression of a catalytically inactive mutant produced anoutcome similar to that of the wild-type protein.

3.2. Dowregulation of PTPN13 triggers megakaryocytic differentiation

Next, we reduced PTPN13 levels in HEL cells with two differentshRNA sequences (Fig. 2A and B). The levels of megakaryocytic markerswere significantly higher when the PTPN13 level was downregulated,while the expression of CD45 haematopoietic marker remainedunchanged (Fig. 2C). At morphological level, in comparison withcontrol cells the downregulation of PTPN13 levels generated largercells (Fig. 2D), and immunostaining revealed a clear increase in CD61expression at the cell surface (Fig. 2E), in keeping with the flow cytom-etry analysis (Fig. 2C). The frequency of polyploid cells after PMA stim-ulation was also increased by PTPN13 silencing (Fig. 2F). PTPN13silencing also impaired cell viability (Supplementary Fig. 1D) and thecellular proliferation rate (Supplementary Fig. 1E).

ThedownregulationofmPTPN13 inmouse L8057 cells (Supplementa-ry Fig. 2A) led to higher acetylcholinesterase (AchE) activity (Supplemen-tary Fig. 2B) and to a significant increase in the levels of megakaryocyticmarkers (Supplementary Fig. 2C). In conclusion, PTPN13 or mPTPN13 si-lencing leads cells to exhibit a more differentiated phenotype.

3.3. Changes in the level of PTPN13 alter cellular signalling

Megakaryocytic differentiation is highly dependent on sustainedERK activation [16,24], and also on the activation of STAT transcriptionfactors [25]. It has been described that PTPN13 can affect ERK signalling

WT

20 m

Emptyvector

20 m

C/S

20 m

A B

Luci

fera

seac

tivity

Rel

ativ

eun

its

02468

1012

(-)

PM

A

n=7

p <

0.00

1

n=7

PTPN13

0 2 4 8 16 24 32 48 (h)

-Tubulin

1 2 10 16 21 28 32 42 PTPN13-Tubulin

Empt

yve

ctor

WT

C/S

PTPN13

-Tubulin

C

1 1.6 1.8 PTPN13/ -Tubulin

F

D

0

20

40

60

80

100

1 2 30

20

40

60

80

100

CD41

Mea

n Fl

uore

scen

ce(%

)

CD61

Em

pty

vect

or

WT

C/S

WT

C/S

n=4 n=4

p<0.

001

p<0.

001

n=4 n=4

p<0.

001

p<0.

001

n=4

Em

pty

vect

or

GWTC/S

Emptyvector

H2N

4N

8N=9.9%

16N=1.2%

>4N=11.1%

2N

4N

8N=6.4%

>4N=6.6%

2N

8N=5.9%

>4N=6.1%

Eve

nts

IPEmpty Vector C/S WT0

2468

101214

8N >4N

EmptyVector

C/S

WT

n=6

n=6n=6

p<0.

05

n=5

p<0.

01

n=6

p<0.

01

n=5

p<0.

01

Cel

ls(%

)

n=4

0

50

100

150

CD45

WT

C/S

Em

pty

vect

or

n=6n=6

n=6

E

0

50

100

150

200

CD45

WT

C/S

Em

pty

vect

or

n=6 n=4

n=5

0

20

40

60

80

100

120

0

20

40

60

80

100

120

CD41 CD61

WT

C/S

Em

pty

vect

or

WT

C/S

Em

pty

vect

or

n=7n=4

n=5

n=7

n=4n=5p<

0.00

1

p<0.

001

p<0.

001

p<0.

05

Mea

n Fl

uore

scen

ce(%

)

β

β

β

β/

μ μ μ

Fig. 1. Ectopic expression of PTPN13 hinders megakaryocytic differentiation in HEL cells. A) PTPN13 protein expression is activated during megakaryocytic differentiation of HEL cells. Arepresentativewestern blot of three different experiments is shown. B) The activity of the ptpn13 promoter increases significantly inHEL cells subjected tomegakaryocytic differentiation.C) Stable cell lines overexpressingwild-type PTPN13 (WT) or a catalytically inactivemutant (C/S)were generated in HEL cells. A representative western blot of four different experimentsis shown. The levels of CD41, CD61 and CD45 were analysed in the absence (D) or presence (E) of PMA treatment for 48 h. Cells overexpressing PTPN13 show a significant reduction inmegakaryocyticmarker levels (CD41 and CD61). F) Cellmorphologywas assessed byMay–Grünwald–Giemsa staining after treatmentwith 20nMPMA for 48h.G) Expression of theCD61megakaryocyticmarkerwas analysed by immunofluorescence (in red) in cells overexpressing PTPN13and control cells, after treatmentwith 20nMPMA for 48h.Nucleiwere identifiedbyDAPI staining (in blue). CD61 immunostainingwas reduced in associationwith PTPN13 overexpression (empty vector: 12.7± 4.6;Wt: 4.9± 0.9; C/S: 5.3± 1.7. Means± SDof 8 differentimages from 2 different experiments). H) The DNA content was analysed in cells overexpressing PTPN13 and in control cells, after 7 days treatment with 20 nM PMA. The percentage ofhigh-ploidy cells is shown (left panel). A histogram representing DNA content is shown from a representative experiment (right panel). The frequency of high-ploidy cells was decreasedby PTPN13 overexpression. The data show means ± SD. Statistical significance is indicated in the figure.

2889J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

in other systems [26,27], and that STATs can act as mPTPN13 substratesin a haematopoietic context [5]. To better understand how PTPN13might regulate megakaryocytic differentiation, we analysed the level

of activation of ERK, STAT3 and STAT5 in stable cell lines in whichPTPN13 had been overexpressed or downregulated. The experimentsshowed that the activation levels of the three signalling proteins

n=3

p<0.

001

n=3

p<0.

001

B

PTP

N13

RN

Ai 1

PTP

N13

RN

Ai 2

Rel

ativ

eex

pres

sion

Luc

RN

Ai

1.21

0.80.60.40.2

0

A

PTPN13

-Tubulin

Luc

RN

Ai

PTPN

13 R

NA

i1

PTPN

13 R

NA

i2

1 0.2 0.03 PTPN13/ -Tubulin

D Luc RNAi

20 m

PTPN13 RNAi 2

20 m

PTPN13 RNAi 1

20 m

E Luc RNAi PTPN13 RNAi1 PTPN13 RNAi2

n=6

n=6

p<0.

001

n=5

p<0.

001

Cel

ls(%

)

n=6

F

05

1015202530

8N >4N

Luc RNAi

PTPN13RNAi 1

PTPN13RNAi 2

2N

4N

8N=6.9%16N=0.9%

>4N=7.8%Eve

nts

IPLuc RNAi PTPN13 RNAi 1 PTPN13 RNAi 2

2N

4N

8N=12.1%

16N=1.7%

>4N=14%

2N

4N

8N=13.3%

16N=1.9%

>4N=18.3%

n=6

n=5

p<0.

001

p<0.

001

C

n=4

n=8

p < 0.001CD41

0

50

100

150

200

250

n=4

n=8p < 0.001

Mea

n Fl

uore

scen

ce(%

)

PMA - - - - + + + +

Luc

RN

AiPT

PN13

RN

Ai1

PTPN

13 R

NAi

2P

TPN

13 R

NA

i1+2

Luc

RN

AiPT

PN13

RN

Ai1

PTPN

13 R

NAi

2PT

PN13

RN

Ai1+

2

CD61p< 0.001

0306090

120150180

p< 0.001

n=8n=8

PTPN

13 R

NAi

2

Luc

RN

AiPT

PN13

RN

Ai1

PTPN

13 R

NAi

1+2

Luc

RN

AiP

TPN

13 R

NAi

1PT

PN13

RN

Ai2

PTPN

13 R

NAi

1+2

Mea

n Fl

uore

scen

ce(%

)

n=4n=4

PMA - - - - + + + +

n=8

Mea

n Fl

uore

scen

ce(%

)

0

50

100

150

200

CD45

n=4n=8

n=4

Luc

RN

AiPT

PN13

RN

Ai1

PTP

N13

RN

Ai2

PTP

N13

RN

Ai1+

2

Luc

RN

AiPT

PN13

RN

Ai1

PTP

N13

RN

Ai2

PTPN

13 R

NAi

1+2

N.S. N.S.

PMA - - - - + + + +

μ μ μ

β

β

Fig. 2. PTPN13 silencing triggersmegakaryocytic differentiation in HEL cells. Stable cell lines inwhich PTPN13 expression had been reduced by RNAiwere generated by lentiviral infection. Twodifferent shRNA sequences against PTPN13were used. A sequence against firefly luciferasewas used as a control. A) PTPN13 protein levels were reduced by the two different shRNA sequencesused. A representativewestern blot of four different experiments is shown. B) RT-qPCR analysis revealed thatmRNA levels were also significantly reduced in the same cell lines. C) The levels ofCD41, CD61 and CD45markerswere analysed in the presence or absence of PMA treatment. PTPN13 silencing induced a significant increase inmegakaryocyticmarker levels (CD41 and CD61).D) Cell stainingwithMay–Grünwald–Giemsa showed that PTPN13 silencing induced amore differentiated phenotype. E) Expression of the CD61megakaryocytic marker was analysed by im-munofluorescence (in red) in cells treatedwith 20 nMPMA for 48 h. Nuclei were identified by DAPI staining (in blue). PTPN13 silencing increased CD61 immunostaining (Luc RNAi: 5.3± 1.2;PTPN13 RNAi 1: 13.5 ± 3.9; PTPN13 RNAi 2: 10.2± 2.2. Means± SD of 8 different images from 2 different experiments). F) The DNA content was analysed in PTPN13-silenced and in controlcells, after 7 days of 20 nM PMA treatment. The percentage of high-ploidy cells is shown (left panel). A histogram representing DNA content is shown from a representative experiment (rightpanel). The frequency of high-ploidy cells was increased by PTPN13 downregulation. The data showmeans ± SD. Statistical significance is indicated in the figure.

2890 J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

A B

D

C0 10 30 60 0 10 30 60 (min)

ERK

pERK

PTPN13 RNAiLuc RNAi

1 7 2 1.3 2 10 6 3 pERK/ERK

tSTAT5

pSTAT5

0 30 60 120 0 30 60 120 (min)

PTPN13 RNAiLuc RNAi

1 7.5 4.6 2 1 7 7.3 5.6 pSTAT5/STAT5

STAT3

pSTAT3

1 2.5 1.3 pSTAT3/STAT3

tSTAT5

pSTAT5

1 2.3 1.8 pSTAT5/STAT5

Luc

RN

Ai

PTPN

13 R

NA

i1PT

PN13

RN

Ai2

ERK

pERK

1 2.3 1.8 pERK/ERK

1 0.93 0.95 -Tubulin

Luc

RN

Ai

STAT5

GAPDH

PTPN

13 R

NA

i2

PTPN

13 R

NA

i1

1 31 43 GATA-1/GAPDH

1 0.95 1.3 STAT5/GAPDH

GATA-1

STAT3

pSTAT3

tSTAT5

pSTAT5

ERK

WT

C/S

pERK

1 0.6 0.7 pERK/ERK

1 0.7 0.6 pSTAT3/STAT3

1 0.5 0.8 pSTAT5/STAT5

1 0.96 0.9 -Tubulin

-Tubulin

Empt

yve

ctor

WT

C/S

STAT5

GAPDH

GATA-1

1 0.86 0.25 GATA-1/GAPDH

1 1 1 STAT5/GAPDH

Empt

yve

ctor

β

ββ

Fig. 3.Modification of PTPN13 levels alters cellular signalling. The levels of activation of signalling pathways important for megakaryocytic differentiation were analysed in stable HEL celllines in which PTPN13 was overexpressed or downregulated. A) Overexpression of wild-type PTPN13 (Wt) or a catalytically inactive mutant (C/S) significantly reduced the level ofactivation of ERK, STAT3 and STAT5. B) By contrast, PTPN13 downregulation induced the activation of ERK, STAT3 and STAT5. C) ERK and STAT5 activation upon PMA stimulation. Theactivation of both signalling pathways was more marked in cells in which PTPN13 had been silenced. D) The levels of GATA-1, the master transcription factor for megakaryocyticdifferentiation, were downregulated by PTPN13 overexpression (left panel) and upregulated by PTPN13 silencing (right panel). All panels show representative western blots of at leastthree different experiments.

2891J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

decreased when PTPN13 was overexpressed (Fig. 3A) and increasedwhen the phosphatase was downregulated (Fig. 3B). Moreover, theability of PTPN13 to affect cell signalling did not seem to depend on itscatalytic ability, since overexpression of the PTPN13 catalytic inactivemutant produced an effect similar to that caused by the wild-typeprotein (Fig. 3A). Under PMA stimulation, the activation of ERK andSTAT5was alwaysmoremarked in cells inwhichPTPN13wasdownreg-ulated than in control cells (Fig. 3C).

We also looked at the levels of GATA-1, the master transcriptionfactor for erythrocytic and megakaryocytic differentiation. GATA-1levels were reduced in cells overexpressing PTPN13, especially whenthe wild-type version of the protein was upregulated (Fig. 3D, leftpanel). Moreover, GATA-1 levels were elevated in cells knocked-downfor PTPN13 (Fig. 3D, right panel). This effect seemed to be specific forGATA-1, since the expression of STAT5 transcription factor was notaffected by the modification of PTPN13 levels.

2892 J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

3.4. PTPN13 and β-catenin coimmunoprecipitate and colocalise

The physical interaction betweenmPTPN13 and the tumour suppres-sor APC has been described previously [13], and mPTPN13 seems to beable to dephosphorylate β-catenin in pancreatic cells, thus antagonizingWnt signalling [14]. Bearing in mind these previous results, we wereprompted to analyse the possible relationship between PTPN13 and β-catenin during megakaryocytic differentiation. To accomplish this, westudied whether both proteins formed part of the same protein complexin coimmunoprecipitation assays. We immunoprecipitated PTPN13 andβ-catenin from HEL cell extracts independently, and regardless of theantibody used for the immunoprecipitation the other proteinwas alwayspresent in the precipitate (Fig. 4A), showing that these two proteinsformed part of the same complex in haematopoietic cells.

Immunocytochemistry analyses revealed that both proteins weredistributed between the cytoplasm and the nucleus. Moreover, theimages showed a strong colocalisation signal for both proteins (Fig. 4B).

3.5. The dowregulation ofβ-catenin triggersmegakaryocytic differentiation

To test the possible role of β-catenin in megakaryocytic differentia-tion we generated two stable HEL cell lines in which its expressionwas downregulated with two different shRNAs (Fig. 5A and B). Thelevels of megakaryocytic markers were significantly higher when β-catenin expression was downregulated, while the expression of theCD45 marker was not affected (Fig. 5C). The interference of β-cateninproduced large cells,with amorphology reminiscent of that ofmegakar-yocytes (Fig. 5D), with a higher expression of CD61 at the cell surface asshown by immunostaining (Fig. 5E), in keepingwith the flow cytometryanalysis. Moreover, GATA-1 protein was upregulated upon β-cateninsilencing (Fig. 5F). The frequency of polyploid cells was also augmentedby β-catenin silencing (Fig. 5G). As was the case of PTPN13, β-cateninsilencing seemed to impair cell viability (Supplementary Fig. 3A), andthe proliferation rate tended to be lower, although without statisticalsignificance (Supplementary Fig. 3B). The silencing of mβ-catenin alsoled to a more differentiated phenotype in the mouse L8057 cell line(Supplementary Fig. 4). Interestingly, these features are very reminis-cent of the downregulation of PTPN13 or mPTPN13, suggesting that

B

DAPI

A

WB: -PTPN13

WB: - -Cat

IPα

-PTP

N13

IP α

- β-C

at

(-)

Cel

llys

ate

75 m

75 m

75 m

β

β

α

α μ

μ

μ

Fig. 4. PTPN13 and β-catenin belong to the same protein complex and colocalise. A) HEL cellpresence of both proteins in the immunoprecipitates was analysed by immunoblotting. The eby immunostaining with specific antibodies in HEL cells treated with 20 nM PMA at different t

the involvement of this phosphatase in megakaryopoiesis may berelated to the regulation of β-catenin function.

3.6. PTPN13 regulates β-catenin phosphorylation and stability

Since one of the main ways of regulating β-catenin function isthrough its stability, we next analysed whether PTPN13 was regulatingβ-catenin turnover. Our results revealed that β-catenin levels werehigher in PTPN13 over-expressing cells (Fig. 6A, left panel), especiallywhen the wild-type PTPN13 was overexpressed, suggesting thatPTPN13 catalytic activity would be necessary to regulate β-catenin sta-bility. In contrast, when PTPN13 was downregulated, β-catenin levelswere reduced with respect to the control cells (Fig. 6A, right panel).Moreover, β-catenin decay under cycloheximide treatment was notice-ably faster in the cells in which PTPN13 had been knocked down(Fig. 6B), suggesting that PTPN13 protects β-catenin from degradation.Since β-catenin degradation is triggered by phosphorylation, we exam-ined the levels of β-catenin Ser33 phosphorylation in the presence ofproteasome inhibitors to avoid protein degradation. It turned out thatthis modification was diminished in cells over-expressing PTPN13,especially in cells over-expressing the wild-type protein, and wasaugmented by PTPN13 knockdown (Fig. 6C). These results are consis-tent with the notion that PTPN13 canmodulate β-catenin phosphoryla-tion and turnover. Although it cannot be expected that PTPN13 wouldregulate β-catenin Ser33 phosphorylation directly, PTPN13 catalyticactivity did seem to be necessary for this purpose, since the effectsdetected were stronger in cells over-expressing the wild-type versionof PTPN13 than in those expressing a catalytically inactive mutant(Fig. 6C).

It has been shown previously that mβ-catenin can be dephosphory-lated in vitro bymPTPN13 [13] and hencewewere prompted to analysewhether β-catenin tyrosine phosphorylation could be modified byPTPN13 in vivo. Our results showed that β-catenin phosphorylation attyrosines 86 and 654 was significantly increased in cells in whichPTPN13 had been downregulated (Fig. 6D), suggesting that in thehaematopoietic context β-catenin could be an in vivo substrate forPTPN13 in human cells.

-Catenin PTPN13 Merge

0H

24H

75 m75 m75 m

75 m75 m75 m

75 m75 m75 m

8H

μ μ μ

μμμ

μ μ μ

extracts were immunoprecipitated with PTPN13- and β-catenin-specific antibodies. Thexperiment was repeated four times. B) PTPN13 and β-catenin localisation was analysedimes. Nuclei were identified by DAPI staining.

B

E

D -Cat RNAi2Luc RNAi

20 m 20 m

-Cat RNAi1

20 m

-Cat RNAi1 -Cat RNAi2Luc RNAi

A

Luc

β-C

atRN

Ai

RN

Ai1

RN

Ai2

β-Cat

β

β β

-Tubulin

1 0.3 0.5 β β

ββ

β β

β β

-Cat/ -Tubulin

G

2N

4N

8N=6.9%16N=0.9%

>4N=7.8%Eve

nts

IP

2N

4N

16N=1%

>4N=10%

2N

4N

8N=12.7%

16N=1.9%

>4N=14.8%

8N=9%

0

5

10

15

20

25

30

35

40

8N >4N

Luc RNAi

β-Cat RNAi 1

β-Cat RNAi 2n=6

n=4

p<0.

05

n=6

p<0.

05

Cel

ls(%

)

n=6

n=4

p<0.

05

n=6

p<0.

05

-Cat RNAi1 -Cat RNAi2Luc RNAi

FLu

cR

NA

i

-Cat

RN

Ai1

GATA-1

-Cat

RN

Ai2

STAT5

GAPDH

1 5.5 4.1 GATA-1/GAPDH

1 1.2 0.7 STAT5/GAPDH

Rel

ativ

eex

pres

sion

n=3

p<0.

0001

n=3

p<0.

0001

LucRNAi

β-CatRNAi 1

β -CatRNAi 2

1.21

0.80.60.40.2

0

0

100

200

300

400

500

1 2

CD41

n=4

n=8

p < 0.001

n=4

n=8

p < 0.001

Mea

n Fl

uore

scen

ce(%

)

C

Luc

RN

Ai

β-C

atR

NA

i1β

-Cat

RN

Ai2

Luc

RN

Ai

β-C

atR

NA

i1β

-Cat

RN

Ai2

PMA - - - - + + + +

β -C

atR

NA

i1+2

0

50

100

150

200

250

300

1 2

CD61

n=4

n=8

p < 0.001

n=4

n=8

Luc

RN

Ai

β-C

atR

NA

i1β

-Cat

RN

Ai2

Luc

RN

Ai

β-C

atR

NA

i1β

-Cat

RN

Ai2

PMA - - - - + + + +

β -C

atR

NA

i1+2

β -C

atR

NA

i1+2

p < 0.01

0

20

40

60

80

100

120

140CD45

n=4 n=8

N.S.

n=4 n=8

N.S.

Luc

RN

Ai

β-C

atR

NA

i1β

-Cat

RN

Ai2

β -C

atR

NA

i1+2

Luc

RN

Ai

β-C

atR

NA

i1β

-Cat

RN

Ai2

β -C

atR

NA

i1+2

PMA - - - - + + + +

β -C

atR

NA

i1+2

μ μ μ

β-C

at

Fig. 5. β-Catenin silencing triggers megakaryocytic differentiation. Stable cell lines in which β-Catenin expression had been reduced by RNAi were generated by lentiviral infection. Twodifferent shRNA sequences against β-catenin were used. A sequence against firefly luciferase was used as a control. A) β-Catenin protein levels were reduced by the two different shRNAsequences used. A representativewestern blot of three different experiments is shown. B) RT-qPCR analysis showed thatmRNA levelswere also significantly reduced in the same cell lines.C) The levels of CD41, CD61 and CD45markers were analysed in the presence or absence of PMA treatment. β-Catenin silencing induced a significant increase in megakaryocytic markerlevels (CD41 and CD61). D) Cell staining withMay–Grünwald–Giemsa revealed that β-catenin silencing induced a more differentiated phenotype. E) Expression of the CD61megakaryo-cyticmarker was analysed by immunofluorescence (in red) in cells treatedwith 20 nMPMA for 48 h. Nuclei were identified byDAPI staining (in blue).β-Catenin silencing increased CD61immunostaining (LucRNAi: 5.3±1.2;β-catenin RNAi 1: 15.6±5.3;β-cateninRNAi 2: 10.3±4.2.Means±SDof 8 different images from2different experiments). F) The levels ofGATA-1,the master transcription factor for megakaryocytic differentiation, were upregulated by β-catenin silencing. A representative western blot of four different experiments is shown. G) TheDNA content was analysed in β-catenin-silenced and in control cells, after 7 days of 20 nM PMA treatment. The percentage of high-ploidy cells is shown (left panel). A histogramrepresenting DNA content is shown from a representative experiment (right panel). The frequency of high-ploidy cells was increased by β-catenin downregulation. The data showmeans ± SD. Statistical significance is indicated in the figure.

2893J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

A

E

WT

C/S

C

Luc

RN

Ai

PTPN

13 R

NA

i1PT

PN13

RN

Ai2

β-Cat

β-Tubulin

p- -Cat (S33)

WT

C/S

Luc

RN

Ai

PTPN

13 R

NA

i1

PTPN

13 R

NA

i2

β-Cat

β

β

-Tubulin

0 0.5 1 2 0 0.5 1 2 (h)

B PTPN13 RNAiLuc RNAi

Luc RNAi

PTPN13RNAi

Time (h)

β-C

aten

inle

vel

n=5

p<0.

05

n=5n=5

n=5n=5

n=5

p<0.

05p<0.

05

0.5

0.75

0.25

0

1

0 0.5 1 1.5 2

1 2.5 5 1 0.4 0.3 β-Cat/ -Tubulin

1 0.8 0.05 1 3.2 2.2 p- -Cat(S33)/β-Cat1 1 1.2 1 1.1 0.9 -Tubulin

D

β

β

ββ

ββ

-Cat

β-Tubulin

p- -Cat (Y654)

p- -Cat (Y86)

Luc

RN

Ai

PTPN

13 R

NA

i1

PTPN

1 R

NA

i21 5.8 6.1 p- -Cat (Y86)/β-Cat1 1.8 2 p- -Cat (Y654)/β-Cat1 1.1 1.1 -Tubulin

Top-

flash

act

ivity

β-Cat S33Y: - + +

PTP

N13

RN

Ai1

n=9

n=7

n=10***

***

###

0

10

20

30

40

0

5

10

15

PTP

N13

C/S

PTP

N13

WT

β-Cat S33Y: - + + +

Top-

flash

act

ivity

n=7

n=7

n=7

n=10******

***+++

++

Empt

yve

ctor

Empt

yve

ctor

Fig. 6. PTPN13 regulates β-catenin stability and transcriptional activity. A) β-Catenin protein levels were analysed by immunoblotting in HEL cells in which PTPN13 had beenoverexpressed (first panel) or silenced (second panel). A representativewestern blot of three different experiments is shown. B) β-Catenin protein decay under cycloheximide treatmentin HEL cells in which PTPN13 was downregulated and in control cells. A representative experiment is shown. The plot shows the densitometric quantification of 5 different experiments(means± SE). C) β-Catenin phosphorylation at Ser33 was analysed by immunoblotting in HEL cells in which PTPN13was overexpressed (first panel) or silenced (second panel) treatedwith 2.5 μMofMG132 proteasome inhibitor for 14 h. A representativewestern blot of three different experiments is shown. D)β-Cateninphosphorylation at Tyr86 or Tyr654was analysedby immunoblotting in HEL cells in which PTPN13 was silenced. A representative western blot of three different experiments is shown. E) HEK293T cells transfected with a TOP-FLASHreporter plasmid and a β-galactosidase expression plasmid as a control in the presence or absence of an expression plasmid for the β-catenin stable mutant (S33Y), together with anshRNAi against PTPN13, or expression plasmids for wt PTPN13 or a catalytically inactive PTPN13 mutant (C/S). TOP-FLASH reporter activity was monitored 48 h after transfection andnormalised with respect to a β-galactosidase transfection control. The data show means ± SD. Statistical analysis was carried out using the analysis of variance (ANOVA) followed bythe Scheffé test. ***p b 0.001 when compared to cells transfected only with the TOP-FLASH reporter plasmid (−); ###p b 0.001 when compared to cells transfected with PTPN13RNAi; +++p b 0.001, ++p b 0.01, when compared to cells transfected with the β-catenin S33Y mutant only.

2894 J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

3.7. PTPN13 regulates β-catenin transcriptional activity

Besides the regulation of its stability, the regulation of β-catenintranscriptional activity is also important to control its function. There-fore, we wondered whether PTPN13 regulates the transcriptional activ-ity of β-catenin. Since our data had shown that PTPN13 regulated β-catenin stability, we decided to use a β-catenin S33Y stable mutant tofacilitate the readout from these experiments. Overexpression of β-catenin S33Y alone produced a significant increase in TOP-FLASHreporter activity, as expected. Moreover, the tuning of PTPN13 levels

was seen to affect β-catenin transcriptional activity. When PTPN13was knocked down, TOP-FLASH reporter activity was further increased(Fig. 6E, left panel), suggesting that PTPN13 would inhibit β-catenintranscriptional activity. In agreement with this, the overexpression ofPTPN13 decreased reporter activity significantly (Fig. 6E, right panel).β-Catenin phosphorylation at tyrosine 654 seems to increase thebinding of the protein to the basal RNA pol II machinery [12], and alsoseems to be important for increasing β-catenin nuclear localization[11]. These previous reports are in agreement with our findings, sincethe strong increase in transcriptional activity detected when PTPN13

PMA - - - - + + + +

A

PTPN13

GAPDH

0 1 2 4 8 (h)

-Cat

1 2.2 3.5 3 2.9 PTPN13/GAPDH1 0.9 2.6 2.5 1.7 -Cat/GAPDH

H0 2 4 8 16 (h)PTPN13

GAPDH1 1.2 2 1.1 0.9 PTPN13/GAPDH1 1.4 1.8 1.9 1.3 -Cat/GAPDH

-Cat

G

PMA - - - + + +0

50

100

150

200

Luc

RN

Ai

β-C

atR

NAi

β-C

atR

NAi

+ PT

PN13

RN

AiLu

cR

NAi

β-C

atR

NAi

β-C

atR

NAi

+ PT

PN13

RN

Ai

CD41

n=7

n=7n=7

n=7

n=7n=5++

+

++

N.S.N.S.

B

Mea

n Fl

uore

scen

ce(%

)

0

50

100

150

PMA - - - + + +

Luc

RN

Ai

β-C

atR

NAi

β-C

atR

NAi

+ P

TPN

13 R

NAi

Luc

RN

Ai

β-C

atR

NAi

β-C

atR

NAi

+ P

TPN

13 R

NAi

CD45

n=5n=7 n=7 n=7 n=7n=7

Mea

n Fl

uore

scen

ce(%

)

PMA - - - + + +

CD61

n=5

n=7

n=7n=7

0

50

100

150

200

Luc

RN

Ai

β-C

atR

NA

iβ

-Cat

RN

Ai+

PTPN

13 R

NAi

Luc

RN

Ai

β-C

atR

NAi

β-C

atR

NAi

+ P

TPN

13 R

NAi

n=7

n=7

N.S.N.S.

++++

+ +

Mea

n Fl

uore

scen

ce(%

)

F

0

50

100

150

Em

pty

Vec

tor

Em

pty

Vec

tor +

Luc

RN

Ai

WT

WT

+ β

-Cat

RN

AiE

mpt

yV

ecto

rE

mpt

yV

ecto

r + L

ucR

NAi WT

WT

+ β

-Cat

RN

Ai

CD41

n=7

n=6

n=7n=6

n=7n=7n=6n=6

**

++++

N.S.

E

Mea

n Fl

uore

scen

ce(%

)

PMA - - - - + + + +0

20406080

100120

Em

pty

Vec

tor

Em

pty

Vec

tor +

Luc

RN

Ai WT

WT

+ β

-Cat

RN

AiE

mpt

yV

ecto

rE

mpt

yV

ecto

r + L

ucR

NAi WT

WT

+ β

-Cat

RN

Ai

CD61

n=6n=6

n=6n=4

n=6n=6

n=6 n=5**

++ +

+++N.S.

Mea

n Fl

uore

scen

ce(%

)

0

50

100

150

200

PMA - - - - + + + +

Em

pty

Vec

tor

Em

pty

Vec

tor +

Luc

RN

Ai

WT

+ β

-Cat

RN

Ai

Em

pty

Vec

tor +

Luc

RN

Ai WT

WT

+ β

-Cat

RN

AiWT

Em

pty

Vec

tor

CD45

n=6n=6n=6 n=6

n=5n=5

n=6

n=5

Mea

n Fl

uore

scen

ce(%

)

2N

4N

8N=6.9%16N=0.9%

>4N=7.8%Eve

nts

IP

2N

4N

16N=1%

>4N=10%

2N

4N

8N=10.5%

16N=0.7%

>4N=11.3%

8N=9%

-Cat RNAi1 -Cat RNAi1+PTPN13 RNAi 20

5

10

15

8N >4N

Luc RNAi

β-Cat RNAi 1

β-Cat RNAi 1 + Luc RNAi 2

n=6n=6

n=6n=6

n=5n=5

+++

+++

N.S.N.S.C

Luc RNAi

Cel

ls(%

)

n=6 n=4 n=4n=5

n=4n=5

LUC

RN

Ai

β-C

at R

NA

i 1

β

/ PTP

N13

R

NA

i 2

Rel

ativ

e ex

pres

sion

PTPN13-CAT

1.21

0.80.60.40.2

0

02468

101214

n=7

n=5

n=4

n=5

N.S.

++ ++*

Cel

ls(%

)

02468

10121416

n=7

n=5

n=4

n=5

N.S.

+++++*

Cel

ls(%

)

Em

pty

Vec

tor

Em

pty

Vec

tor +

Luc

RN

Ai W

T

WT

+ β

-Cat

RN

Ai

Em

pty

Vec

tor

Em

pty

Vec

tor +

Luc

RN

Ai WT

WT

+ β

-Cat

RN

Ai

1.6

D

Rel

ativ

e ex

pres

sion

CTNNB1 PTPN13

n=5 n=5

n=5

n=2

-CAT

00.20.4

1.41.21.00.80.6

Em

pty

Vec

tor

Em

pty

Vec

tor +

Luc

RN

Ai WT

WT

+ β

-Cat

RN

Aip

< 0.

05

p <

0.05

p <

0.00

1

p <

0.00

1

p <

0.01

++

IP

Eve

nts

2N

4N

8N=9.9%

16N=1.2%

>4N=11.1%

Empty Vector Empty Vector +Luc RNAi

8N=9.9%

>4N=11%

16N=1%

2N4N

2N

8N=5.9%

>4N=6.1%

WT

4N8N=7.1%

16N=1.6%

>4N=8.7%4N

2N

WT+PTPN13 RNAi 1

β

β

β

β

β

ββ

β-C

at R

NA

i 1

Fig. 7. PTPN13 is an upstream regulator ofβ-catenin and it is stabilised byWnt-signalling. A) PTPN13was silenced in one of theβ-catenin-silenced cells.mRNA levelswere analysed byRT-qPCR.B) The levels of CD41, CD61 and CD45 markers were analysed in the presence or absence of PMA treatment. ++p b 0.01, +p b 0.05, when compared to control cells (Luc RNAi). C) The DNAcontent was analysed after 7 days of 20 nM PMA treatment. The percentage of high-ploidy cells is shown (left panel). A histogram representing DNA content is shown from a representativeexperiment (right panel). ++p b 0.01, +p b 0.05, when compared to control cells (Luc RNAi). D) Wt PTPN13-overexpressing cell line or the control cell line (empty vector) was transducedwithRNAis against PTPN13 and luciferase respectively.mRNA levelswere analysed byRT-qPCR. E) The levels of CD41, CD61 andCD45markerswere analysed in thepresence or absence of PMAtreatment. +++p b 0.001, ++p b 0.01, +p b 0.05, when compared to control cells (Empty vector or Empty vector + Luc RNAi). *p b 0.05, when compared toWT. F) The DNA content wasanalysed after 7 days of 20 nM PMA treatment. The percentage of high-ploidy cells is shown (left panel). A histogram representing DNA content is shown from a representative experiment(right panel). +++p b 0.001, ++p b 0.01, when compared to control cells (Empty vector or Empty vector + Luc RNAi). *p b 0.05, when compared to WT. PTPN13 and β-catenin levelswere analysed in HEL treated with 20 mM LiCl (G) or 100 ng/ml Wnt3a (H). A representative western blot of three different experiments is shown. The data show means ± SD. Statisticalsignificance is indicated in the figure.

2895J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

2896 J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

was down-regulated (Fig. 6E) could be directly related to an increase intyrosine 654 phosphorylation, as observed (Fig. 6D).

3.8. PTPN13 is involved in the canonical Wtn/β-catenin signalling pathway

The coincidence in phenotype when PTPN13 and β-catenin aresilenced and the fact that PTPN13 regulates different aspects of β-catenin function strongly suggested that these two proteins belongedto the samepathway, andwe surmised that PTPN13 should be upstreamfrom β-catenin.

To test this further, the expression of PTPN13was downregulated inone of the β-catenin-silenced cell lines (Fig. 7A). The double-silencedcell line also showed significantly higher levels of megakaryocyticmarkers (Fig. 7B) and a higher proportion of polyploid cells (Fig. 7C),in comparison with control cells. However, these features in thedouble-silenced cell line were not increased when compared with theparental cell line where only β-catenin was silenced. In other words,the double silencing did not produce a stronger phenotype.

Moreover, β-catenin expression was downregulated in cells overex-pressing PTPN13 (Fig. 7D). Interestingly, the decrease inmegakaryocyt-ic markers and ploidy induced by PTPN13 overexpression was partiallyrescued by β-catenin downregulation, levels very similar to those ofcontrol cells being reached (Fig. 7E and F).

These results support the notion that PTPN13 and β-catenin belongto the same pathway, PTPN13 being an upstream regulator of β-catenin. β-Catenin stabilisation is one of the main features of the so-called canonical Wnt pathway. This pathway has recently been relatedto megakaryopoiesis [28,29], and hence we thought it of interest toanalyse whether PTPN13 was involved in this signalling. Our resultsshowed that the inhibition of GSK3β by lithium chloride treatmentinduced a mild increase not only in β-catenin as expected, but also inPTPN13 (Fig. 7G). Moreover, stimulation with Wnt3a induced thestabilisation of both, β-catenin and PTPN13 (Fig. 7H), suggesting thatPTPN13 stabilisation must be part of the canonical Wnt pathway inmegakaryocytic cells. To the best of our knowledge, we show here forthe first time that PTPN13 stabilisation must be part of the canonicalWnt signalling pathway.

3.9. Downregulation of mPTPN13 or mβ-catenin triggers megakaryocyticdifferentiation of haematopoietic progenitor cells

In order to check whether our observations were meaningful in amore physiological context we decided to analyse the involvementof both proteins in the megakaryocytic differentiation of murinehaematopoietic progenitor cells.

Under ex vivo culture conditions in the presence of TPO and IL1β,murine bonemarrow Lin− cells undergomegakaryocytic differentiation,as reflected by the significant increase in the levels of megakaryocyticmarkers (CD41 and CD61) (Fig. 8A and B). After 12 days of treatmentthese cells showed a dramatic morphological change corresponding tomegakaryocytic differentiation (Fig. 8C).

To test the involvement of mPTPN13 or mβ-catenin in themegakar-yocytic differentiation of the progenitor cells, shRNAs against mβ-catenin or mPTPN13 were introduced into bone marrow Lin− cells bylentiviral infection, and an shRNA against firefly luciferase was usedas a control [16,23]. The results showed that the interference ofmPTPN13 and mβ-catenin produced a more differentiated pheno-type in murine haematopoietic progenitor cells, because these cellsshowed a significant increase in the levels of megakaryocytic markers(Fig. 8D and E), and in the frequency of high-ploidy cells in comparisonwith control cells (Fig. 8F). Moreover, the number of megakaryocyticcolonieswas increased bymβ-catenin silencing and tended to be higherwhen mPTPN13 was downregulated (Fig. 8G).

In sum, these results strongly suggest that PTPN13 and β-cateninmay be important regulators of megakaryopoiesis in vivo.

4. Discussion

PTPN13 is high-molecular weight phosphatase with a modularstructure. In spite of the broad number of interactions reported forPTPN13 [1], its molecular function remains obscure. Here we studiedthe role of PTPN13 and β-catenin in megakaryocytic differentiation. Tothe best of our knowledge, this is the first description of the importanceof PTPN13 in the homeostasis of megakaryocytic differentiation.

Our results show that the expression of PTPN13 is activated duringmegakaryocytic differentiation at the protein andmRNA levels, suggest-ing an important role for this protein during this differentiation. Inagreement with this, we found that altering PTPN13 levels dramaticallyaffected the differentiation of both, cell lines and haematopoieticprogenitors.

Experiments aimed at the overexpression or downregulation ofPTPN13 suggested that this protein would behave as a negative regula-tor of megakaryocytic differentiation, probably through the regulationof signalling pathways required for this process. It has been shownpreviously that PTPN13 downregulation leads to a hyperactivation ofERK signalling in epithelial cancer [26] and breast cancer [27] cells.Consistent with this, we detected an increase in ERK activation whenPTPN13 was downregulated and observed the opposite result whenthe phosphatase was upregulated. Moreover, we observed that PTPN13was also able to influence the activation of STAT3 and STAT5, in agree-ment with previous reports [5]. A sustained activation of ERK [24] andSTAT protein activation [30] is involved in megakaryocytic differentia-tion. Our results are consistent with the notion that PTPN13 would con-trol cellular signalling duringmegakaryocytic differentiation.We suggestthat PTPN13 upregulationmight be important for attenuating the signal-ling cascades driving such differentiation. Interestingly, PTPN13 catalyticactivity did not seem to be required to mediate this effect, since overex-pression of a catalytically inactive mutant or the wild type protein pro-duced the same effect regarding differentiation and cellular signalling,favouring a role as a scaffolding protein in this context.

Although signalling regulation by PTPN13 would explain its impor-tance for maintaining megakaryocytic differentiation, we wonderedwhether there might be additional mechanisms. We observed thatPTPN13 interacted with β-catenin in megakaryocytic cells. Interesting-ly, we noted a striking coincidence in the phenotype elicited by thesilencing of β-catenin and PTPN13. Our results show that PTPN13 regu-lates β-catenin stability. Curiously, while this workwas in preparation itwas reported that PTPN13 influences BCR-ABL-expressing cell prolifer-ation and survival through the regulation of β-catenin stability [31].Moreover, our results show that β-catenin regulation by PTPN13 isnot dependent exclusively on BCR-ABL signalling, and this may beimportant in a physiological context, such asmegakaryocytic differenti-ation for example, and not only for maintaining CML leukemic cellproliferation and survival.

PTPN13 can reduce S33β-catenin phosphorylation, which increasesits stability. Owing to its catalytic specificity, PTPN13would be expectedto regulate S33β-catenin phosphorylation indirectly. Notwithstanding,it is important to note that PTPN13 catalytic activity seems to berequired for the control of β-catenin stability, since overexpression ofa PTPN13 catalytically inactive mutant had a much weaker effect thanoverexpression of the wild-type version of the protein.

Different aspects of β-catenin function, such as its interaction withcadherin or its transcriptional activity, can be regulated by tyrosinephosphorylation [32]. Upon PTPN13 silencing, a strong increase in β-catenin phosphorylation occurs at tyrosines 86 and 654, together witha strong increase in β-catenin transcriptional activity, suggesting thatPTPN13 could also regulate β-catenin transcriptional activity, modulat-ing phosphorylation at certain tyrosine residues such as Tyr 654 [11,12].

Our results show that PTPN13 stabilisation is part of the canonicalWntsignalling pathway,whichmust be important to increaseβ-catenin levels.It has recently been suggested that Wnt signalling in megakaryocytes isrequired for platelet formation [28]. However, another report suggests

B

D

n=7

n=5 n=7

n=7

n=7n=6

Mea

n Fl

uore

scen

ce(%

)

0

50

100

150

200

250

300

350

Luc

RN

Ai

mP

TPN

13 R

NAi

1

mβ

-Cat

RN

Ai1

Luc

RN

Ai

mP

TPN

13 R

NAi

1

mβ

-Cat

RN

Ai1

CD41 CD61

p>0.

05

p>0.

01

p>0.

01

p>0.

05

20 m

0H 12 Days

20 m

C

n=23

E

F

A

7.7%

5 Days

0 Days

CD61

CD41

30.3%

12 Days

50.6%

CD61

n=7

n=7

n=10n=17

n=10n=13p

< 0.

01

p <

0.01

p <

0.01

p <

0.01

p < 0.01

p < 0.01

CD42d

Luc

RN

Ai

mP

TPN

13 R

NAi

1

mP

TPN

13 R

NAi

2

All

mP

TPN

13 R

NA

i1+2

mβ

-Cat

RN

Ai1

mβ

-Cat

RN

Ai2

All

mβ

-Cat

RN

Ai1+

2 0500

100015002000250030003500400045005000

Mea

n Fl

uore

scen

ce

Luc

RN

Ai

mP

TPN

13 R

NAi

1

mP

TPN

13 R

NAi

2

All

mP

TPN

13 R

NAi

1+2

mβ

-Cat

RN

Ai1

mβ

-Cat

RN

Ai2

All

mβ

-Cat

RN

Ai1+

2 0200400600800

100012001400160018002000 p

< 0.

05

p <

0.05

p <

0.05

p < 0.05p < 0.01

n=18

CD42d

n=7

n=5

n=8n=13

n=6

n=12

Mea

n Fl

uore

scen

ce(%

)

Luc

RN

Ai

mP

TPN

13 R

NAi

1

mP

TPN

13 R

NAi

2

All

mP

TPN

13 R

NA

i1+2

mβ

-Cat

RN

Ai1

mβ

-Cat

RN

Ai2

All

mβ

-Cat

RN

Ai1+

2

0

500

1000

1500

2000

2500

3000

3500

n=4

n=5n=4 n=9 n=7

n=10 n=17

p <

0.05

p <

0.01

p <

0.05

p <

0.05

p < 0.01

p < 0.01CD41

Mea

n Fl

uore

scen

ce(%

)

0

5

10

15

20

25

30

35

40

45

50

Luc

RN

Ai

mP

TPN

13 R

NAi

1

mP

TPN

13 R

NAi

2

mβ

-Cat

RN

Ai1

mβ

-Cat

RN

Ai2

Luc

RN

Ai

mP

TPN

13 R

NAi

1

mP

TPN

13 R

NAi

2

mβ

-Cat

RN

Ai1

mβ

-Cat

RN

Ai2

n=7

n=7

n=5

n=5

n=6

n=5

n=8

n=8

n=7

n=7

p <

0.05

p <

0.05

p <

0.05

p <

0.05

p <

0.05

p <

0.05

p <

0.01

p <

0.01

Cel

ls(%

)

G

0

20

40

60

80

100

120

140

160

n=3

n=3

n=4

p <

0.05

Luc

RN

Ai

mP

TPN

13 R

NAi

1+2

mβ

-Cat

RN

Ai1+

2

CFU

s/35

000

cells

0

500

1000

1500

2000

2500

3000

CD61 CD41

0H5 Days12 Days

n=2

n=3

n=4

n=2

n=3

n=4

p=0.

001

p=0.

005

p <

0.01

p=0.

001

p <

0.01

Mea

n Fl

uore

scen

ceμ μ

Fig. 8. Silencing of mPTPN13 or mβ-catenin triggers megakaryocytic differentiation of haematopoietic progenitors. Ex vivo megakaryocytic differentiation of Lin− cells purified frommouse bone marrow. Differentiation was stimulated by treatment with 100 ng/ml TPO and 10 ng/ml IL1β for 6 or 12 days, and followed by analysis of specific markers (CD41, CD61and CD42d). A) Flow cytometry diagram showing the percentage of CD41+CD61+ cells, B) and marker levels along differentiation. C) Cell staining with May–Grünwald–Giemsa atdays 0 and 12. Lin− cells were modified by lentiviral infection with shRNAs against mPTPN13 or mβ-catenin. An shRNA against firefly luciferase was used as a control. Cells carryingthe different shRNAswere identified byGFP expression. The levels of megakaryocyticmarkers at day 6 (D) and day 12 (E)were significantly higher in cells transducedwith shRNA againstmPTPN13 or mβ-catenin. F) The DNA content was analysed at day 12. The percentage of high-ploidy cells is shown. G) Megakaryocytic colony-forming assay.

2897J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

2898 J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

thatWnt signalling would inhibit the expansion of primitive megakaryo-cytes [29]. In agreement with this latter report, our results show thatPTPN13 and β-catenin levels can influence the ex vivo megakaryocyticdifferentiation of murine haematopoietic progenitors. To reconcile theseapparently discrepant results, we suggest that the levels ofWnt signallingmust be strictly regulated during megakaryopoiesis. The stabilisation ofPTPN13 and β-catenin by Wnt signalling in megakaryocyte progenitorsmust be important to control the balance between proliferation anddifferentiation at this stage, and an excessive accumulation of these twoproteins would slow differentiation down. Moreover, a progressivestabilisation of PTPN13 and β-catenin during differentiation would beconsistent with the involvement of Wnt signalling during later stages ofmegakaryocyte maturation and platelet release, as suggested elsewhere[28]. In the future it would be interesting to analyse the role of PTPN13and β-catenin in terminal megakaryocyte maturation.

Finally, we show the importance of PTPN13 for the regulation ofmegakaryocytic differentiation. In part, this can be explained throughthe regulation of ERK or STAT signalling. Moreover, we believe thatthe regulation of β-catenin stability and function by PTPN13 is essentialfor the homeostasis of megakaryopoiesis. Here we provide strongevidence of the existence of a functional relationship between PTPN13and β-catenin during megakaryocytic differentiation. Moreover, weshow that PTPN13 levels increase upon Wnt3a signalling; this mightbe important for β-catenin stabilisation, and places PTPN13 as a compo-nent of canonical Wnt signalling. Our results hold true for megakaryo-cytic cell lines and also for murine haematopoietic progenitors cells,and hence we suggest that PTPN13 and β-catenin might be importantregulators of in vivo megakaryopoiesis. In future studies we shallanalyse the role of these proteins in other haematopoietic lineages.

5. Conclusions

Our results show that PTPN13 behaves as an important regulatorof megakaryocytic differentiation in cell lines and also in murinehaematopoietic progenitors. This can be explained by the ability ofPTPN13 to regulate the signalling pathways that control this type of dif-ferentiation. Our data also show that β-catenin can also alter megakar-yocytic differentiation. We show that PTPN13 regulates β-cateninstability and function, as a part of canonical Wnt signalling, and thatthis is required for correct cell differentiation.

List of abbreviations7-AAD 7-aminoactinomycin DCML chronic myeloid leukaemiaERK extracellular signal-regulated kinasesDMSO dimethyl sulfoxideFBS foetal bovine serumFERM 4.1/ezrin/radixin/moesinHEL human erythroleukemiaIL interleukinIMDM Iscove's Modified Dulbecco's MediumKIND kinase non-catalytic C-lobe domainMOI multiplicity of infectionMTT 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyltetrazoliumbromidePDZ PSD-95/Discs large/ZO-1PMA phorbol 12-myristate 13-acetatePTP protein tyrosine phosphatasePVDF polyvinylidene fluorideRPMI Roswell Park Memorial InstituteSDS-PAGE sodium dodecyl sulphate-polyacrylamide gel electrophoresisshRNA small hairpin RNASTAT signal transducer and activator of transcriptionTGF transforming growth factorTPO thrombopoietinTREM triggering receptor expressed on myeloid cells

Author contributions

JLS and GLR performed experiments and analysed data. RPB, BSS,APF and LISA performed experiments. JAPS, JSY and MLL discussed theresults and edited the paper. LQ analysed the data. FA analysed thedata and edited the paper. AHH designed the experiments, performedexperiments and wrote the paper.

Acknowledgements

We thank our colleagues for providing us with different mate-rials: the pHM6-PTPL1 and pHM6-PTPL1-C/S plasmids were provid-ed by Dr G. Freiss (IRCM, Montpellier, France). The TOP/FLASHreporter plasmid was provided by Dr M. Duñach (UniversidadAutónoma de Barcelona, Spain). Lentiviral vectors were kindly pro-vided by Dr D. Trono (School of Life Sciences and Frontiers in GeneticsProgram, Lausanne, Switzerland). L8057 cellswere a gift fromDr P. Vyas(Wetherall Institute of Molecular Medicine, Oxford, UK). We thankA. Prieto Martín and Dr R. Bernal for their assistance with the morpho-logical analysis. This work was supported by MCINN (BFU2006-10362,BFU2009-10568, CSD2007-00015, BFU2011-28467 and BFU2011-28804), by JCyL (SA010A10-2, SA126A07 and BIO/SA59/13) and byFEDER-FIS (PS09/01075). G.L.P. and J.L.S. were recipients of a predoctoralfellowship from the Regional Government of Castilla y León-FEDER, Spain.

Appendix A. Supplementary data

Supplementarymaterial available online at BBAMol Cell Reswebpage.Supplementary data associated with this article can be found, in theonline version, at doi: http://dx.doi.org/10.1016/j.bbamcr.2014.08.014.

References

[1] K.S. Erdmann, Theprotein tyrosinephosphatasePTP-Basophil/Basophil-like. Interactingproteins and molecular functions, Eur. J. Biochem. 270 (2003) 4789–4798.

[2] W. Hendriks, J. Schepens, D. Bachner, J. Rijss, P. Zeeuwen, U. Zechner, H. Hameister,B. Wieringa, Molecular cloning of a mouse epithelial protein-tyrosine phosphatasewith similarities to submembranous proteins, J. Cell. Biochem. 59 (1995) 418–430.

[3] M. van Eekelen, V. Runtuwene, W. Masselink, J. Den Hertog, Pair-wise regulation ofconvergence and extension cell movements by four phosphatases via RhoA, PLoSOne 7 (2012) e35913.

[4] D.G. Wansink, W. Peters, I. Schaafsma, R.P. Sutmuller, F. Oerlemans, G.J. Adema, B.Wieringa, C.E. van der Zee, W. Hendriks, Mild impairment of motor nerve repairin mice lacking PTP-BL tyrosine phosphatase activity, Physiol. Genomics 19 (2004)50–60.

[5] M. Nakahira, T. Tanaka, B.E. Robson, J.P. Mizgerd, M.J. Grusby, Regulation of signaltransducer and activator of transcription signaling by the tyrosine phosphatasePTP-BL, Immunity 26 (2007) 163–176.

[6] G. Freiss, D. Chalbos, PTPN13/PTPL1: an important regulator of tumor aggressiveness,Anticancer Agents Med. Chem. 11 (2011) 78–88.

[7] O.D. Abaan, J.A. Toretsky, PTPL1: a large phosphatase with a split personality, CancerMetastasis Rev. 27 (2008) 205–214.

[8] H. Clevers, R. Nusse, Wnt/beta-catenin signaling and disease, Cell 149 (2012)1192–1205.

[9] H. Aberle, A. Bauer, J. Stappert, A. Kispert, R. Kemler, Beta-catenin is a target for theubiquitin–proteasome pathway, EMBO J. 16 (1997) 3797–3804.

[10] J. Lilien, J. Balsamo, The regulation of cadherin-mediated adhesion by tyrosinephosphorylation/dephosphorylation of beta-catenin, Curr. Opin. Cell Biol. 17(2005) 459–465.

[11] T. Kajiguchi, A. Katsumi, R. Tanizaki, H. Kiyoi, T. Naoe, Y654 of beta-catenin isessential for FLT3/ITD-related tyrosine phosphorylation and nuclear localization ofbeta-catenin, Eur. J. Haematol. 88 (2012) 314–320.

[12] J. Piedra, D. Martinez, J. Castano, S. Miravet, M. Dunach, A.G. de Herreros, Regulationof beta-catenin structure and activity by tyrosine phosphorylation, J. Biol. Chem. 276(2001) 20436–20443.

[13] K.S. Erdmann, J. Kuhlmann, V. Lessmann, L. Herrmann, V. Eulenburg, O. Muller, R.Heumann, The Adenomatous Polyposis Coli-protein (APC) interacts with the proteintyrosine phosphatase PTP-BL via an alternatively spliced PDZ domain, Oncogene 19(2000) 3894–3901.

[14] H.J. Welters, A. Oknianska, K.S. Erdmann, G.U. Ryffel, N.G. Morgan, The proteintyrosine phosphatase-BL,modulates pancreatic beta-cell proliferation by interactionwith the Wnt signalling pathway, J. Endocrinol. 197 (2008) 543–552.

[15] M. Dromard, G. Bompard, M. Glondu-Lassis, C. Puech, D. Chalbos, G. Freiss, Theputative tumor suppressor gene PTPN13/PTPL1 induces apoptosis through insulinreceptor substrate-1 dephosphorylation, Cancer Res. 67 (2007) 6806–6813.

2899J.L. Sardina et al. / Biochimica et Biophysica Acta 1843 (2014) 2886–2899

[16] J.L. Sardina, G. Lopez-Ruano, L.I. Sanchez-Abarca, J.A. Perez-Simon, A. Gaztelumendi,C. Trigueros, M. Llanillo, J. Sanchez-Yague, A. Hernandez-Hernandez, p22phox-dependent NADPH oxidase activity is required for megakaryocytic differentiation,Cell Death Differ. 17 (2010) 1842–1854.

[17] J.I. Diaz-Hernandez, A. Almeida, M. Delgado-Esteban, E. Fernandez, J.P. Bolanos,Knockdown of glutamate–cysteine ligase by small hairpin RNA reveals that bothcatalytic and modulatory subunits are essential for the survival of primary neurons,J. Biol. Chem. 280 (2005) 38992–39001.

[18] M.M. Rocio, J. Sanchez-Yague, A. Hernandez-Hernandez, M. Llanillo, Amphiphilicand hydrophilic forms of acetylcholinesterase from sheep platelets, Biochim.Biophys. Acta 1415 (1998) 163–173.

[19] B. Sanchez-Sanchez, S. Gutierrez-Herrero, G. Lopez-Ruano, R. Prieto-Bermejo, M.Romo-Gonzalez, M. Llanillo, A. Pandiella, C. Guerrero, J.F. Miguel, F. Sanchez-Guijo,C. Del Canizo, A. Hernandez-Hernandez, NADPH oxidases as therapeutic targets inchronic myelogenous leukemia, Clin. Cancer Res. 20 (2014) 4014–4025.

[20] T.D. Schmittgen, K.J. Livak, Analyzing real-time PCR data by the comparativeC(T) method, Nat. Protoc. 3 (2008) 1101–1108.

[21] C. Cantin, J. Holguera, L. Ferreira, E. Villar, I. Munoz-Barroso, Newcastle disease virusmay enter cells by caveolae-mediated endocytosis, J. Gen. Virol. 88 (2007) 559–569.

[22] A. Hernandez-Hernandez, A. Ferrus, Prodos is a conserved transcriptional regulatorthat interacts with dTAF(II)16 in Drosophila melanogaster, Mol. Cell. Biol. 21 (2001)614–623.

[23] J. Torres-Torronteras, A. Gomez, H. Eixarch, L. Palenzuela, G. Pizzorno, M. Hirano, A.L.Andreu, J. Barquinero, R. Marti, Hematopoietic gene therapy restores thymidinephosphorylase activity in a cell culture and a murine model of MNGIE, Gene Ther.18 (2011) 795–806.

[24] A.M. Whalen, S.C. Galasinski, P.S. Shapiro, T.S. Nahreini, N.G. Ahn, Megakaryocyticdifferentiation induced by constitutive activation of mitogen-activated proteinkinase kinase, Mol. Cell. Biol. 17 (1997) 1947–1958.

[25] K. Kaushansky, Historical review: megakaryopoiesis and thrombopoiesis, Blood 111(2008) 981–986.

[26] A.C. Hoover, G.L. Strand, P.N. Nowicki, M.E. Anderson, P.D. Vermeer, A.J.Klingelhutz, A.D. Bossler, J.V. Pottala, W.J. Hendriks, J.H. Lee, ImpairedPTPN13 phosphatase activity in spontaneous or HPV-induced squamous cellcarcinomas potentiates oncogene signaling through the MAP kinase pathway,Oncogene 28 (2009) 3960–3970.

[27] P.D. Vermeer, M. Bell, K. Lee, D.W. Vermeer, B.G. Wieking, E. Bilal, G. Bhanot, R.I.Drapkin, S. Ganesan, A.J. Klingelhutz, W.J. Hendriks, J.H. Lee, ErbB2, EphrinB1,Src kinase and PTPN13 signaling complex regulates MAP kinase signaling in humancancers, PLoS One 7 (2012) e30447.

[28] I.C. Macaulay, J.N. Thon,M.R. Tijssen, B.M. Steele, B.T. MacDonald, G.Meade, P. Burns,A. Rendon, V. Salunkhe, R.P. Murphy, C. Bennett, N.A. Watkins, X. He, D.J. Fitzgerald,J.E. Italiano Jr., P.B. Maguire, Canonical Wnt signaling in megakaryocytes regulatesproplatelet formation, Blood 121 (2013) 188–196.

[29] P. Paluru, K.M. Hudock, X. Cheng, J.A. Mills, L. Ying, A.M. Galvao, L. Lu, A.Tiyaboonchai, X. Sim, S.K. Sullivan, D.L. French, P. Gadue, The negative impact ofWnt signaling on megakaryocyte and primitive erythroid progenitors derivedfrom human embryonic stem cells, Stem Cell Res. 12 (2014) 441–451.

[30] I. Matsumura, J. Ishikawa, K. Nakajima, K. Oritani, Y. Tomiyama, J. Miyagawa,T. Kato, H. Miyazaki, Y. Matsuzawa, Y. Kanakura, Thrombopoietin-induceddifferentiation of a human megakaryoblastic leukemia cell line, CMK, involvestranscriptional activation of p21(WAF1/Cip1) by STAT5, Mol. Cell. Biol. 17(1997) 2933–2943.

[31] W. Huang, L. Bei, E.A. Eklund, Fas-associated phosphatase 1 (Fap1) influencesbetacatenin activity in myeloid progenitor cells expressing the Bcr-abl oncogene,J. Biol. Chem. 288 (2013) 12766–12776.

[32] R.L. Daugherty, C.J. Gottardi, Phospho-regulation of beta-catenin adhesion andsignaling functions, Physiology (Bethesda) 22 (2007) 303–309.