Haploinsufficiency of the ESCRT Component HD-PTP Predisposes to Cancer

Report

Haploinsufficiency of the ESCRT Component HDPTP Predisposes to Cancer Graphical Abstract

Authors Sanaz Manteghi, Marie-Claude Gingras, Dmitri Kharitidi, ..., Michael Witcher, Jerry Pelletier, Arnim Pause

Correspondence [email protected]

In Brief Manteghi et al. demonstrate that HDPTP/PTPN23 is a haploinsufficient tumor suppressor located on 3p21.3. They show that Ptpn23+/ mice are predisposed to spontaneous tumor formation and Mycinduced lymphomagenesis via integrindependent survival and invasion. Importantly, they reveal that PTPN23 haploinsufficiency is common in human cancers and correlates with poor survival.

Highlights d

d

d

d

Ptpn23+/ loss predisposes mice to spontaneous tumorigenesis Ptpn23+/ loss accelerates Myc-induced lymphomagenesis Ptpn23+/ promotes integrin-dependent B cell lymphoma survival and invasion PTPN23 heterozygous alterations are frequent in human cancers

Manteghi et al., 2016, Cell Reports 15, 1893–1900 May 31, 2016 ª 2016 The Author(s). http://dx.doi.org/10.1016/j.celrep.2016.04.076

Cell Reports

Report Haploinsufficiency of the ESCRT Component HD-PTP Predisposes to Cancer Sanaz Manteghi,1,2 Marie-Claude Gingras,1,2 Dmitri Kharitidi,1,2 Luc Galarneau,1,2 Maud Marques,4 Ming Yan,1,2 Regina Cencic,1,2 Francis Robert,1,2 Marile`ne Paquet,3 Michael Witcher,4 Jerry Pelletier,1,2 and Arnim Pause1,2,* 1Department

of Biochemistry, McGill University, Montre´al, QC H3G 1Y6, Canada Cancer Research Center, McGill University, Montre´al, QC H3A 1A3, Canada 3De ´ partement de Pathologie et de Microbiologie, Faculte´ de Me´decine Ve´te´rinaire, Universite´ de Montre´al, Saint-Hyacinthe, QC J2S 2M2, Canada 4Departments of Oncology and Experimental Medicine, Lady Davis Institute, Jewish General Hospital, McGill University, Montre ´ al, QC H3T 1E2, Canada *Correspondence: [email protected] http://dx.doi.org/10.1016/j.celrep.2016.04.076 2Goodman

SUMMARY

Endosomal sorting complexes required for transport (ESCRT) drive cell surface receptor degradation resulting in attenuation of oncogenic signaling and pointing to a tumor suppressor function. Here, we show that loss of function of an ESCRT protein (HD-PTP encoded by the PTPN23 gene, located on the tumor suppressor gene cluster 3p21.3) drives tumorigenesis in vivo. Indeed, Ptpn23+/ loss predisposes mice to sporadic lung adenoma, B cell lymphoma, and promotes Myc-driven lymphoma onset, dissemination, and aggressiveness. Ptpn23+/ derived tumors exhibit an unaltered remaining allele and maintain 50% of HD-PTP expression. Consistent with the role of HD-PTP in attenuation of integrin recycling, cell migration, and invasion, hemizygous Ptpn23+/ loss increases integrin b1-dependent B cell lymphoma survival and dissemination. Finally, we reveal frequent PTPN23 deletion and downregulation in human tumors that correlates with poor survival. Altogether, we establish HD-PTP/PTPN23 as a prominent haploinsufficient tumor suppressor gene preventing tumor progression through control of integrin trafficking.

receptors and attenuate downstream signaling (Raiborg and Stenmark, 2009). Therefore, ESCRT proteins were proposed to act as tumor suppressor genes (TSGs) (Mattissek and Teis, 2014). Clinical evidence also supports this notion. For instance, downregulation of the ESCRT member HCRP1 correlates with reduced therapeutic efficacy of cetuximab in ovarian cancer (Wittinger et al., 2011). However, due to the lack of viable homozygous knockout mouse models, clear demonstration that ESCRTs act as tumor suppressors has not been made yet. The histidine domain containing protein tyrosine phosphatase (HD-PTP) is a ubiquitously expressed pseudophosphatase encoded by the PTPN23 gene (Gingras et al., 2009a, 2009b). HD-PTP is also a key ESCRT-associated protein required for receptor degradation that was reported to inhibit migration and invasion by preventing integrin, Src, and focal adhesion kinase (FAK) signaling (Castiglioni et al., 2007; Kharitidi et al., 2015; Lin et al., 2011; Mariotti et al., 2009). We hypothesized that PTPN23 might be a haploinsufficient TSG since it maps to 3p21.3, a prevalent TSG cluster frequently hemizygously deleted in many human tumors, but rarely mutated on the remaining allele (Angeloni 2007; Ji et al., 2005). In accord, PTPN23 TSG function was suggested since HD-PTP overexpression delays xenograft growth of testicular germ cell tumors (Tanaka et al., 2013). However, the HD-PTP/PTPN23 TSG function and mechanism has not been yet demonstrated in vivo by loss-of-function studies. Therefore, we decided to directly investigate the tumor predisposition of Ptpn23 hemizygous knockout mice (Gingras et al., 2009a).

INTRODUCTION RESULTS Cell surface receptors are key regulators of critical cellular processes such as survival, proliferation, migration, and invasion. Upon ligand stimulation, activated receptors are ubiquitinated, internalized, and either sorted for degradation or recycled to the cell surface. Misregulation of these processes results in cell surface receptor stabilization and signaling persistence, a well-recognized hallmark of cancer (Mellman and Yarden, 2013). The endosomal sorting complexes required for transport (ESCRT) coordinate the sorting and degradation of cell surface

Ptpn23+/ Mice Are Predisposed to B Cell Lymphoma and Lung Adenoma To investigate the in vivo functions of HD-PTP/PTPN23, C57BL/6; 129/OlaHsd F1 Ptpn23+/ mice were backcrossed into C57BL/6 or 129/OlaHsd backgrounds (Gingras et al., 2009a). As we reported, Ptpn23 / deletion was embryonic lethal (Gingras et al., 2009a), while Ptpn23+/ animals were born at the expected Mendelian frequency, developed normally, and were fertile.

Cell Reports 15, 1893–1900, May 31, 2016 ª 2016 The Author(s). 1893 This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

A Mesenteric lymphoma B

Tumor incidence P=0.005 ** Ptpn23+/+ Ptpn23+/-

Tumor incidence %

60 40

+/+

Ptpn23

+/-

Ptpn23

HD-PTP Actin

P=0.003

**

20

Lung adenoma

C C57BL/6 (Spleen)

129/OlaHsd (Lung) +/+/+ Ptpn23 Ptpn23

0

HD-PTP Lymphoma Lung adenoma C57BL/6 129/OlaHsd

E

(A) Representative images of the indicated Ptpn23+/ tumors. (B) Percent tumor incidence in the indicated strains: mesenteric lymphomas (16–18 months; Ptpn23+/ : 12/23 versus Ptpn23+/+: 3/24) and lung adenomas (15–18 months; Ptpn23+/ : 15/59 versus Ptpn23+/+: 2/50). (C) Western blot analysis of HD-PTP levels in the indicated normal tissues. Actin was used as a loading control. (D and E) Immunohistochemical analyses of the Ptpn23+/ mesenteric lymphoma (D) and lung adenoma (E) performed with the indicated antibodies. Insets indicate magnified areas and dashed lines delimit the tumor edge. (F and G) Genomic qPCR quantification of the Ptpn23 allele number in normal (N) and tumor (T) tissues isolated from lymph nodes (n = 5) (F) and lung (n = 2) (G). (H) Western blot analysis of HD-PTP levels in normal (N) and tumor (T) lymph node tissues of the indicated genotypes. Actin was used as a loading control. Data are expressed as mean ± SEM. See also Figure S1.

CD3

SPC

B220

CCSP

D

Actin

Figure 1. Ptpn23+/ Mice Are Predisposed to Spontaneous Tumor Formation

confirmed a benign adenoma phenotype originated from type II pneumocytes (Figure 1E). Together, these data indicate that both tumor types were of primary F G H C57BL/6 Lymph node Lung 2.5 2.5 origin. (Lymph node) Genomic qPCR, exon sequencing, and N N 2.0 2.0 western blot analysis of Ptpn23+/ tumors + +/ +/ +/ 3 3 3 revealed that the remaining wild-type T 1.5 1.5 n2 n2 n2 tp Ptp Ptp Ptpn23 allele was unaltered in terms of P T N N sequence and expression levels (Figures HD-PTP 1.0 1.0 1F–1H and S1C). Thus, the inactivation Actin of the remaining Ptpn23 allele is not 0.5 0.5 N N T required for tumorigenesis, implying a haploinsufficient tumor suppressor 0.0 0.0 /+ /+ +/ +/ +/ +/ + + function. 3 3 3 3 2 2 23 23 n2 n2 pn pn To examine Ptpn23 TSG mechanism, pn Ptp pn Ptp Pt Pt Pt Pt we employed mouse embryonic fibroblasts (MEFs) isolated from Ptpn23+/+ and Ptpn23+/ mice. While Ptpn23+/ loss did not affect cell proliferation, it Interestingly, by 72 weeks of age, Ptpn23+/ mice developed promoted focus formation and loss of contact inhibition in spontaneous mesenteric lymphoma (C57BL/6) and lung ade- immortalized MEFs (Figures S1D–S1G). To address Ptpn23 noma (129/OlaHsd), with a 4- to 5-fold increased incidence function in oncogene-driven cell transformation, primary and compared to Ptpn23+/+ littermates (Figures 1A and 1B). Consis- immortalized MEFs were transduced with retrovirally delivered tent with hemizygous Ptpn23 deletion, a 50% reduction in H-RasV12 oncogene. The Ptpn23 status did not influence HD-PTP expression was detected (Figures 1C, S1A, and S1B), H-RasV12-induced growth arrest in primary MEFs (Figure S1H) suggesting that the remaining allele does not transcriptionally or H-RasV12-mediated soft agar growth in immortalized MEFs compensate the allele loss. Immunohistochemistry analysis of (Figure S1I). However, H-RasV12 significantly increased the mesenteric lymph node tumors using B cell (B220)- and T cell Ptpn23+/ proliferation rate of immortalized MEFs (Figure S1J). (CD3)-specific markers confirmed a lymphoma of B cell origin Altogether, our results demonstrate that Ptpn23+/ contributes (Figure 1D). Similarly, staining of lung tumors with lung cell line- to cell transformation following immortalization and/or oncogene age markers (Clara cells: CCSP and type II pneumocytes: SPC) expression. 100 μm

Ptpn23 allele number

Ptpn23 allele number

1.0 mm

1894 Cell Reports 15, 1893–1900, May 31, 2016

2.5 mm

250 μm

Figure 2. Ptpn23 Hemizygosity Accelerates Myc-Driven Lymphomagenesis

B B-cells (Eμ−myc) Ptpn23+/-

Ptpn23+/+

Pt

-

pn Pt 23 + /+ p Pt n23 + pn /Pt 23 + pn /+ 23 +/

A

Eμ-Myc Ptpn23 truncated β-geo cassette Ptpn23 wild type

HD-PTP Actin

Eµ-myc Ptpn23+/+ Eμ-myc (n=42) Ptpn23+/- Eμ-myc (n=34)

80

Median onset 87 P<0.001 43

60 40 20

Ptpn23+/+ Eμ-myc (n=34) Ptpn23+/- Eμ-myc (n=29)

100

0

80

Median survival 125 P<0.001 73

60 40 20 0

0

50 100 Tim e (days)

E 600

150

Ptpn23+/+ Eμ-myc (n=24) Ptpn23+/- Eμ-myc (n=19)

0

50

F

100 150 200 Tim e (days)

250

Ptpn23+/+ Eμ-myc (n=4) Ptpn23+/- Eμ-myc (n=4) T

3 Ptpn23 allele number

Tumor free (%)

100

D

Survival (%)

C

(A) Representative PCR-based DNA genotyping performed using primers specific for the Ptpn23 wild-type (WT) or mutated (truncated/b-geo) allele and the Em-myc transgene. (B) Representative western blot analysis of HDPTP levels in bone marrow-derived B cells. Actin is shown as loading control. (C and D) Kaplan-Meier analysis of tumor onset (C) and overall survival (D) in the indicated strains. The median onset and survival are indicated. (E) Average spleen weight of the indicated strains (70 days). (F) Genomic qPCR quantification of the Ptpn23 allele number in normal (N) and tumor (T) lymph node tissues. (G) Relative Ptpn23 mRNA levels quantified by qRT-PCR in the indicated lymphoma samples. (H) Representative western blot analysis of the HD-PTP protein levels in the indicated lymph node tumors (LNT). Tubulin is shown as a loading control. Data are expressed as mean ± SEM. See also Figure S2.

N

Ptpn23 mRNA levels

Spleen weight (mg)

125 days) (Figures 2C and 2D). Pathology analysis confirmed a lymphoblastic lym400 phoma phenotype, accompanied by 2 P=0.01 advanced splenomegaly in Ptpn23+/ T N ** Em-myc animals (Figure 2E). An intact 200 1 Ptpn23 wild-type allele was preserved and expressed in Ptpn23+/ Em-myc lymphoma when compared to Em-myc 0 0 tumors (Figures 2F–2H and S2B). AltoEμ-myc gether, our data confirmed that Ptpn23 acts as a haploinsufficient TSG in vivo in G H a Myc-driven lymphomagenesis mouse Ptpn23+/+ Eμ-myc (n=5) 1.5 Ptpn23+/- Eμ-myc (n=9) model. Eµ-myc (LNT) The p53-ARF axis is crucial for lym+/+ +/Ptpn23 Ptpn23 phoma development, and its loss pro1.0 motes Myc-induced lymphomagenesis HD-PTP (Eischen et al., 1999). While p53 exP=0.04 pression and integrity (assessed by Tubulin 0.5 * sequencing and western blot) were unaltered in Ptpn23+/ Em-myc lymph nodes tumors, ARF protein levels were signifi0.0 cantly decreased (Figure S2C). Consistently, ARF expression was reduced in spleen and MEFs from Ptpn23+/ mice Ptpn23 Hemizygosity Accelerates Myc-Driven (Figures S2D and S2E), which resulted from decreased tranLymphomagenesis scriptional activity, as lower Arf mRNA levels and decreased To determine if Ptpn23+/ predisposes to oncogene-driven lym- Arf promoter activity were observed in Ptpn23+/ MEFs (Figures phomagenesis in vivo, Ptpn23+/ mice were crossed with the S2F and S2G). HD-PTP knockdown in wild-type MEFs similarly transgenic Em-myc B cell lymphoma mouse model (Figure 2A) reduced ARF protein levels, confirming that HD-PTP regulates (Adams et al., 1985). Reduced HD-PTP expression was ARF expression (Figure S2H). Commonly, ARF deletion leads confirmed in Ptpn23+/ Em-myc B cells (Figures 2B and S2A). to bypass of replicative senescence and spontaneous immortalInterestingly, the lymphoma onset of Ptpn23+/ Em-myc mice ization of primary MEFs cultured in standard conditions (Kamijo was accelerated compared to Ptpn23+/+ Em-myc (43 versus et al., 1997). Consistent with a partial ARF reduction, the 87 days) and correlated with reduced survival (73 versus Ptpn23+/ MEFs did not bypass senescence (Figure S2I), but

** P=0.01

Cell Reports 15, 1893–1900, May 31, 2016 1895

their immortalization rate was accelerated (Figures S2I and S2J). Our results suggest that hemizygous loss of Ptpn23 might promote tumorigenesis by reducing ARF levels and enhancing immortalization rate. However, the specific mechanism involved remains elusive. Ptpn23 Hemizygous Loss Promotes Integrin-Dependent B Cell Lymphoma Survival and Spreading Overcoming Myc-induced apoptosis is a prerequisite for Mycdriven lymphomagenesis (Eischen et al., 1999). Consistent with the accelerated lymphomagenesis, TUNEL staining indicated reduced cell death levels in Ptpn23+/ Em-myc compared to Em-myc tumors (Figures 3A and S3A). In parallel, cell viability of freshly isolated Ptpn23+/ Em-myc lymph node tumors (LNT) and bone marrow-derived cells was increased (Figures 3B and 3C). We recently reported that loss of HD-PTP stimulates b1 integrin cell surface stabilization, turnover, and promotes FAK downstream signaling (Kharitidi et al., 2015), both involved in cell survival and tumorigenesis in different cancers, including lymphoma (Alanko et al., 2015; Bosch et al., 2011; Desgrosellier and Cheresh, 2010). Consistent with our previous work, Ptpn23+/ Em-myc B cells exhibited elevated cell surface integrin levels (Figures 3D and S3B). Moreover, increased total integrin and FAK phosphorylation levels (pFAK) were observed in Ptpn23+/ Em-myc tumors (Figure 3E) as well as in Em-myc tumor-derived B cells downregulated for HD-PTP expression (Figure 3F). In agreement with integrin-dependent downstream signaling, treatment of the Em-myc B cells with an integrin b1-blocking antibody reduced pFAK levels to the control levels (Figure 3G). Furthermore, HD-PTP downregulation in Em-myc B cells conferred a survival advantage upon serum starvation ( FBS), which is abolished by FAK inhibitor (PND-1186) or integrin b1-blocking antibody (Figures 3G, 3H, and S3C). Thus, our data reveal that loss of HD-PTP enhances cell survival by increasing integrin b1 and pFAK levels. Our previous work indicated that HD-PTP depletion favors integrin-dependent cell migration and invasion, known as drivers of tumor progression and metastasis in various cancers, including lymphoma (Dozynkiewicz et al., 2012; Kharitidi et al., 2015; Stroeken et al., 1998, 2000). To determine if a potentiated pro-invasive phenotype contributes to PTPN23 TSG function, we used MEFs and confirmed that Ptpn23+/ MEFs exhibited increased integrin b1 and pFAK levels and integrin-dependent invasion rate (Figures 3I and 3J). Consistently, we observed a higher incidence of Em-myc lymphoma dissemination in multiple tissues upon Ptpn23+/ loss (Figures 3K and S3D). Altogether, our data imply that hemizygous loss of Ptpn23 stimulates tumorigenesis by promoting integrin-dependent cell survival and invasion. Loss of PTPN23 Is Frequent in Human Cancer In agreement with a PTPN23 haploinsufficient TSG function, extensive bioinformatic analyses using public cancer databases revealed frequent PTPN23 hemizygous deletions, mutations (Figures 4A and S4A), and reduced HD-PTP expression (Figure 4B) in different human cancers. In addition, PTPN23 mRNA levels were significantly reduced in human lymphoma (Figure 4C) and lung cancer (Figure 4D), when compared to normal tissue. 1896 Cell Reports 15, 1893–1900, May 31, 2016

Finally, PTPN23 mRNA downregulation in lung and breast cancer (Figures 4E and 4F) and loss of PTPN23 allele number in breast cancer (Figure S4B) correlate with reduced survival. Different PTPN23 point mutations were reported in human cancers (Figure S4A), including lymphoma (Schmitz et al., 2012; Cerami et al., 2012; Gao et al., 2013). The effect of some of these missense mutations on HD-PTP expression and stability were addressed via their ectopic expression in HeLa cells followed by cycloheximide chase. Interestingly, two missense mutations identified in lymphoma significantly destabilized HD-PTP protein (Figures S4C and S4D), implying that these mutations contribute to PTPN23 loss of function. Altogether, our data revealed that PTPN23 acts as a prominent haploinsufficient TSG in humans. DISCUSSION While the TSG function of HD-PTP was previously proposed by Tanaka et al. (2013) based on HD-PTP overexpression studies, the evidence that PTPN23 loss of function drives oncogenic transformation, in vitro and in vivo, was lacking. Taking advantage of our Ptpn23+/ mouse model, here, we establish PTPN23 as a haploinsufficient TSG in vivo. Indeed, we demonstrate that Ptpn23+/ deficiency predisposes mice to sporadic B cell lymphoma, lung tumors, and accelerated Myc-driven lymphomagenesis. Moreover, extensive bioinformatic analyses reveal recurrent hemizygous PTPN23 loss in many human tumors, which correlates with poor outcome. Consistently, hemizygous deletions in the 3p21.3 region are very frequent in human tumors and were described as the earliest genetic alteration in several types of lung cancer (Hung et al., 1995; Wistuba et al., 1999) while mutations are infrequent (Lerman and Minna, 2000). PTPN23 mutation rate is low across human tumors (<7%) (cBioPortal), however, some mutations were reported in human cancers. Interestingly, by revealing the destabilizing nature of some of these mutations, our data suggest an alternative loss of function mechanism for HD-PTP. Importantly, this study presents evidences linking ESCRT and TSG function. Indeed, despite the essential role of ESCRT components in cell surface receptor degradation leading to downstream signaling termination, their TSG function was not clearly established, since their homozygous deletion lead to embryonic lethality in mice. Here, we report that loss of one Ptpn23 allele is sufficient to drive tumorigenesis in vivo in mice but also in humans. Based on this data, we hypothesize that other ESCRT proteins might also acts as haploinsufficient TSGs. We previously reported that HD-PTP depletion accelerates integrin recycling and boosts integrin-dependent cell migration and invasion (Kharitidi et al., 2015). Interestingly, potentiated integrin recycling and signaling were reported as drivers of tumorigenesis by enhancing cell survival, invasion, and metastasis formation in vivo (Caswell et al., 2008; Delamarre et al., 2009; Dozynkiewicz et al., 2012). In agreement, we report here that Ptpn23 TSG mechanism is related to integrin-dependent pro-survival and pro-invasive properties in vitro and in vivo, which contributes to tumor initiation, progression, and dissemination. Moreover, we demonstrate using a knockout animal that integrin trafficking controls tumor progression in vivo.

P=0.04

*

10

60 40 20 0

LNT

D

Ptpn23 Eμ-myc (n=5) Ptpn23+/- Eμ-myc (n=4)

*

sh

-S

C R -H D

D

R C +/ +

pn

23

**

Pt

***

I Pt

Fold viability

P=0.0002

Eµ-myc B-cells

pn

sh-SCR shHD-PTP P=0.005

1.0

actin

actin Eµ-myc B-Cells

H 1.5

pFAK

pFAK

actin Eµ-myc (LNT)

-S

Int. β1

pFAK

sh

Int. β1

Int. β1

CTL HD-PTP

sh

D

R

G

-H

C -S

+/-

sh

Ptpn23

50

Eµ-myc (B-cells)

F

E Ptpn23

**

0

0 48 Time (hrs)

+/+

P=0.008

(A) Quantification of TUNEL positive cells in the indicated lymph node tumors (LNT). (B and C) Percent cell viability determined by trypan blue exclusion in freshly isolated LNT (B) or bone marrow-derived B cells (6-week-old mice) and cultured 48 hr in vitro (C). (D) Flow cytometry quantification of the cell surface integrin b1 levels in B cells (B220+ gated population) derived from LNT (8-week-old mice). (E–G) Representative western blot analysis of primary LNT (E) and Em-myc B cells depleted (sh-HD) or not (sh-SCR) for HD-PTP expression (F and G), treated or not with integrin b1-blocking antibody (5 mg/ml) for 18 hr when indicated (G). (H) Fold viability measured in the indicated B cells, pretreated with DMSO, PND-1186, IgM, or integrin b1-blocking antibody for 18 hr and serum-starved ( FBS) for 5 hr. Data represent the mean ± SEM of greater than or equal to five independent experiments. (I) Representative western blot analysis of MEFs plated on fibronectin. (J) Fold cell invasion of MEFs, treated with integrin b1-blocking or control (IgM) antibodies assessed by xCELLigence technology. (K) Incidence of lymphoma dissemination in the indicated strains. Data represent the mean ± SEM of greater than or equal to three independent experiments. See also Figure S3.

+/ -

0

Ptpn23+/+ Eμ-myc (n=5) Ptpn23+/- Eμ-myc (n=9)

150

100

P=0.04 50

LN T

Figure 3. Ptpn23+/ Loss Promotes Integrin-Dependent Survival and Invasion

23

100

Cell surface integrin β1

+/+

-H

0

**

sh

20

Ptpn23+/+ Eμ-myc (n=4) Ptpn23+/- Eμ-myc (n=4) P=0.01

80 Cell viability (%)

30

C Cell viability (%)

B

Ptpn23+/+ Eμ-myc (n=3) Ptpn23+/- Eμ-myc (n=3)

sh

TUNEL positive cells (%)

A

HD-PTP Int. β1

0.5 pFAK actin

0.0 DMSO PND-1186

IgM

-FBS Eµ-myc B-cells P=0.002

J

P=0.005

**

Int.β1

MEFs

K

**

P=0.002

Fold invasion

1.5

Lymphoma dissemination

**

Ptpn23+/+ Ptpn23+/-

1.0

0.5

Tissue

Ptpn23+/+

Spleen Liver Kidney Heart Lung Brain

66 (n=6) 50 (n=6) 33 (n=6) 16 (n=6) 25 (n=4) 0 (n=3)

Ptpn23+/100 85 85 85 50 33

(n=7) (n=7) (n=7) (n=7) (n=4) (n=3)

0.0 IgM

Int. β1 MEFs

Cell Reports 15, 1893–1900, May 31, 2016 1897

A

D 0.15

80

PTPN23 mRNA (Log2 expression)

Alteration frequency (%)

100

60 40 20

n=108 n=980 0.10 0.05 0.00

**

-0.05

Normal Tumor 1.0

E

0.8 Overall survival

100

50

PTPN23 mRNA levels high n=842 low n=303 P=0.0054

0.6 0.4 0.2 0.0

0

re a R s en Pr a os l C tate ar ci no id Lu ng Sk C in er U vica r o H ea the l d lia & l ne ck En Bre do as m t et ria Te l st is M Liv el e an r om G a Ly lio m ma ph om a

0

50

Pa

100

150

200

250

Time (months)

nc

B

Downregulation incidence (%)

H Lu ea n g d s & qu n E s cc ec o p RC k U h C te a g rin u s L u Ce e C n g rv S a ica O den l va o Br rian St e om a s B l ac t Pa a d h nc d e re r a Sa AC s M rc C el om an a om Li a c C hR ver ol C or C M ect P a Pr NS l os T U tat te e G rin lio e m G a BM AC yC

0

1.0

F Relapse free survival

C

Log rank ratio

0.5

0.0

-0.5

0.8 0.6 0.4

PTPN23 mRNA levels high n=828 low n=832 P=2e-05

0.2

P=0.0123 -1.0

0.0 0

50

100

150

200

250

300

Time (months) -1.5 Figure 4. PTPN23 Loss of Function Is Frequent in Human Cancers (A) Frequency of PTPN23 hemizygous deletions in human cancers (cBioPortal). (B) Percent human tumor samples downregulated for HD-PTP over total samples (Protein Atlas). (C) Relative PTPN23 mRNA levels quantified by qRT-PCR in cDNA tissue scan array (33 lymphoma and 5 normal lymph node samples; Origene), normalized to actin transcript levels and expressed as the log rank ratio of PTPN23 expression in lymphoma samples compared to normal lymph nodes. The horizontal lines represents SEM. (D) PTPN23 mRNA levels in human lung normal and tumor samples. Data are expressed as mean ± SEM (p = 0.002). (E and F) Kaplan-Meier analysis of lung (E) and breast cancer patient survival (F) according to PTPN23 mRNA levels classified as low or high PTPN23 mRNA levels. See also Figure S4.

In addition, we observed that HD-PTP depletion correlates with reduced ARF levels via an unknown mechanism currently under investigation. The INK4a/ARF locus is frequently silenced or deleted in human cancers and has been described as a mech1898 Cell Reports 15, 1893–1900, May 31, 2016

anism to bypass oncogene-induced apoptosis and senescence (Kamijo et al., 1997; Ruas and Peters, 1998). In agreement, decreased apoptosis levels in Ptpn23+/ tumors and increased rate of immortalization in Ptpn23+/ MEFs were observed.

However, our data suggest that hemizygous loss of Ptpn23 is not sufficient to fully transform cells and that other genetic alterations are required. Altogether, it is likely that PTPN23 hemizygous loss affects multiple signaling cascades including integrin and ARF that together potentiate tumor initiation and progression in vivo. Therefore, identification of additional signaling receptors or pathways targeted by HD-PTP will be the subject of future investigations.

the intensity of staining in tumor samples (graded as high medium, low, or undetectable) to normal tissue (Table S2).

EXPERIMENTAL PROCEDURES

S.M., M.C.G., and A.P. carried out the study conceptualization and experimental design. S.M., M.C.G., D.K., L.G., M.Y., R.C., and F.R. performed experiments and analysis. M.P. carried out pathology diagnosis. S.M. and M.M. performed bioinformatic analyses. S.M., D.K., M.C.G., J.P., M.W., and A.P. wrote and/or edited the manuscript.

For additional experimental procedures see the Supplemental Experimental Procedures. Animal Studies Experimental procedures were approved by the animal care committee of McGill University and performed according the committee rules and regulations. The Ptpn23 mouse model was described (Gingras et al., 2009a). Ptpn23+/ mice were backcrossed to C57BL/6 or 129/OlaHsd backgrounds over eight generations and the Em-myc offspring were obtained by crossing C57BL/6 Ptpn23+/ (N10) with C57BL/6 Em-myc mice (gift from R. Jones, McGill University). Mice were genotyped twice (weaning and sacrifice), using primers listed in Table S1. Mice were sacrificed at specific time points (3, 6, 8, 10, 12, 16, 18, and 20 months) and lung or lymphoma tumors were monitored upon dissection. The onset of lymphomagenesis in Em-myc mice was monitored bi-weekly by palpation and mice were sacrificed upon any sign of distress. Tissues were collected, snap-frozen in liquid nitrogen for DNA, RNA, and protein extraction or fixed in 10% neutral-buffered formalin for histological analysis. Cell-Based Assays Cell viability was determined by Trypan blue exclusion method or using CellTiter-Glo (Promega: G755B) according to manufacturer’s instructions. Briefly, 1.5 3 104 B cells were seeded in 12-well plate and treated with 1 mM FAK inhibitor (PND-1186, Selleck Chemicals: 7653), DMSO, or 5 mg/ml IgM (Biolegend: 401002) or integrin b1-blocking antibody (BD: 555003) for 18 hr. Then cells were resuspended in FBS-free media supplemented with either DMSO, PND-1186, IgM, or integrin b1-blocking antibody, and 1 3 104 cells were seeded in 96-well plate and incubated 5 hr at 37 C. Following CellTiter-Glo addition, luminescence was measured using a PHERAstar FS HTS microplate reader and cell viability was expressed as the fold of control. Invasion assays were performed as described (Kharitidi et al., 2015). Briefly, 2 3 104 MEFs cells, treated with IgM control or integrin b1-blocking antibody, were employed in triplicate in Matrigel invasion assays, using xCELLigence technology. Data were expressed as the fold invasion rate (cell index/hr) calculated for a 5-hr interval. Human Data Analysis TCGA data including lung cancer gene expression RNA sequencing (RNA-seq) (Illumina HiSeq: 1,124 samples), breast invasive carcinoma copy number (gistic2-thresholded: 1,079 samples with survival data) were downloaded from cancer genomics browser website (https://genome-cancer.ucsc.edu/proj/ site/hgHeatmap/). Kaplan-Meier curves for the analysis of PTPN23 copy number were generated using diploid or haploid samples. Overall survival analysis of lung cancer patients using PTPN23 mRNA levels (probe 223150_s_at and split patients by auto select best cutoff option) and relapse free survival analysis of breast cancer patients (probe 223150_s_at and split patients by median) were extracted from online KM-Plotter database and graphed with } rffy et al., 2013). Data on hemizygous SPSS software (Gyo¨rffy et al., 2010; Gyo deletion of PTPN23 and mutations in cancers were extracted from the cBioPortal public database. Protein expression data were extracted from the Human Protein Atlas database (ENSG0000076201-PTPN23/cancer), and the relative incidence of HD-PTP downregulation was obtained by comparing

SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, four figures, and two tables and can be found with this article online at http://dx.doi.org/10.1016/j.celrep.2016.04.076. AUTHOR CONTRIBUTIONS

ACKNOWLEDGMENTS We thank K. McDonald and D. Ethier from the McGill Flow Cytometry Core Facility and the Histology facilities of the McGill Goodman Cancer Center and IRIC-Universite´ de Montre´al for technical support as well as E. Vincent and A. Malina for critical review of the manuscript. This work was supported by grants from the Cancer Research Society (CRS) (to A.P.), the Canadian Cancer Society Research Institute (CCSRI) (700525/702500 to A.P. and 702778 to J.P.). S.M. was supported by a Canadian Institutes of Health Research/Fonds de la recherche en sante´ du Que´bec (CIHR/FRSQ) (FRN53888) of the McGill Integrated Cancer Research Training Program (MICRTP), Canderel, George G. Harris, and Biochemistry Graduate Excellence studentship awards. D.K. was supported by a FRSQ and M.Y. by a CIHR Doctoral Research Award. Received: October 17, 2015 Revised: February 26, 2016 Accepted: April 19, 2016 Published: May 19, 2016 REFERENCES Adams, J.M., Harris, A.W., Pinkert, C.A., Corcoran, L.M., Alexander, W.S., Cory, S., Palmiter, R.D., and Brinster, R.L. (1985). The c-myc oncogene driven by immunoglobulin enhancers induces lymphoid malignancy in transgenic mice. Nature 318, 533–538. Alanko, J., Mai, A., Jacquemet, G., Schauer, K., Kaukonen, R., Saari, M., Goud, B., and Ivaska, J. (2015). Integrin endosomal signalling suppresses anoikis. Nat. Cell Biol. 17, 1412–1421. Angeloni, D. (2007). Molecular analysis of deletions in human chromosome 3p21 and the role of resident cancer genes in disease. Brief. Funct. Genomics Proteomics 6, 19–39. Bosch, R., Dieguez-Gonzalez, R., Ce´spedes, M.V., Parren˜o, M., Pavo´n, M.A´., Gran˜ena, A., Sierra, J., Mangues, R., and Casanova, I. (2011). A novel inhibitor of focal adhesion signaling induces caspase-independent cell death in diffuse large B-cell lymphoma. Blood 118, 4411–4420. Castiglioni, S., Maier, J.A., and Mariotti, M. (2007). The tyrosine phosphatase HD-PTP: A novel player in endothelial migration. Biochem. Biophys. Res. Commun. 364, 534–539. Caswell, P.T., Chan, M., Lindsay, A.J., McCaffrey, M.W., Boettiger, D., and Norman, J.C. (2008). Rab-coupling protein coordinates recycling of a5b1 integrin and EGFR1 to promote cell migration in 3D microenvironments. J. Cell Biol. 183, 143–155. Cerami, E., Gao, J., Dogrusoz, U., Gross, B.E., Sumer, S.O., Aksoy, B.A., Jacobsen, A., Byrne, C.J., Heuer, M.L., Larsson, E., et al. (2012). The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2, 401–404.

Cell Reports 15, 1893–1900, May 31, 2016 1899

Delamarre, E., Taboubi, S., Mathieu, S., Be´renguer, C., Rigot, V., Lissitzky, J.-C., Figarella-Branger, D., Ouafik, L., and Luis, J. (2009). Expression of integrin a6b1 enhances tumorigenesis in glioma cells. Am. J. Pathol. 175, 844–855. Desgrosellier, J.S., and Cheresh, D.A. (2010). Integrins in cancer: biological implications and therapeutic opportunities. Nat. Rev. Cancer 10, 9–22. Dozynkiewicz, M.A., Jamieson, N.B., Macpherson, I., Grindlay, J., van den Berghe, P.V., von Thun, A., Morton, J.P., Gourley, C., Timpson, P., Nixon, C., et al. (2012). Rab25 and CLIC3 collaborate to promote integrin recycling from late endosomes/lysosomes and drive cancer progression. Dev. Cell 22, 131–145. Eischen, C.M., Weber, J.D., Roussel, M.F., Sherr, C.J., and Cleveland, J.L. (1999). Disruption of the ARF-Mdm2-p53 tumor suppressor pathway in Mycinduced lymphomagenesis. Genes Dev. 13, 2658–2669. Gao, J., Aksoy, B.A., Dogrusoz, U., Dresdner, G., Gross, B., Sumer, S.O., Sun, Y., Jacobsen, A., Sinha, R., Larsson, E., et al. (2013). Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci. Signal. 6, pl1. Gingras, M.-C., Kharitidi, D., Che´nard, V., Uetani, N., Bouchard, M., Tremblay, M.L., and Pause, A. (2009a). Expression analysis and essential role of the putative tyrosine phosphatase His-domain-containing protein tyrosine phosphatase (HD-PTP). Int. J. Dev. Biol. 53, 1069–1074. Gingras, M.-C., Zhang, Y.L., Kharitidi, D., Barr, A.J., Knapp, S., Tremblay, M.L., and Pause, A. (2009b). HD-PTP is a catalytically inactive tyrosine phosphatase due to a conserved divergence in its phosphatase domain. PLoS ONE 4, e5105.

Lerman, M.I., and Minna, J.D. (2000). The 630-kb lung cancer homozygous deletion region on human chromosome 3p21.3: identification and evaluation of the resident candidate tumor suppressor genes. The International Lung Cancer Chromosome 3p21.3 Tumor Suppressor Gene Consortium. Cancer Res. 60, 6116–6133. Lin, G., Aranda, V., Muthuswamy, S.K., and Tonks, N.K. (2011). Identification of PTPN23 as a novel regulator of cell invasion in mammary epithelial cells from a loss-of-function screen of the ‘PTP-ome’. Genes Dev. 25, 1412–1425. Mariotti, M., Castiglioni, S., Garcia-Manteiga, J.M., Beguinot, L., and Maier, J.A. (2009). HD-PTP inhibits endothelial migration through its interaction with Src. Int. J. Biochem. Cell Biol. 41, 687–693. Mattissek, C., and Teis, D. (2014). The role of the endosomal sorting complexes required for transport (ESCRT) in tumorigenesis. Mol. Membr. Biol. 31, 111–119. Mellman, I., and Yarden, Y. (2013). Endocytosis and cancer. Cold Spring Harb. Perspect. Biol. 5, a016949. Raiborg, C., and Stenmark, H. (2009). The ESCRT machinery in endosomal sorting of ubiquitylated membrane proteins. Nature 458, 445–452. Ruas, M., and Peters, G. (1998). The p16 INK4a/CDKN2A tumor suppressor and its relatives. Biochim. Biophys. 1378, F115–F177. Schmitz, R., Young, R.M., Ceribelli, M., Jhavar, S., Xiao, W., Zhang, M., Wright, G., Shaffer, A.L., Hodson, D.J., Buras, E., et al. (2012). Burkitt lymphoma pathogenesis and therapeutic targets from structural and functional genomics. Nature 490, 116–120.

Gyo¨rffy, B., Lanczky, A., Eklund, A.C., Denkert, C., Budczies, J., Li, Q., and Szallasi, Z. (2010). An online survival analysis tool to rapidly assess the effect of 22,277 genes on breast cancer prognosis using microarray data of 1,809 patients. Breast Cancer Res. Treat. 123, 725–731.

Stroeken, P.J., van Rijthoven, E.A., van der Valk, M.A., and Roos, E. (1998). Targeted disruption of the b1 integrin gene in a lymphoma cell line greatly reduces metastatic capacity. Cancer Res. 58, 1569–1577.

} rffy, B., Surowiak, P., Budczies, J., and La´nczky, A. (2013). Online survival Gyo analysis software to assess the prognostic value of biomarkers using transcriptomic data in non-small-cell lung cancer. PLoS ONE 8, e82241.

Stroeken, P.J., van Rijthoven, E.A., Boer, E., Geerts, D., and Roos, E. (2000). Cytoplasmic domain mutants of beta1 integrin, expressed in beta 1-knockout lymphoma cells, have distinct effects on adhesion, invasion and metastasis. Oncogene 19, 1232–1238.

Hung, J., Kishimoto, Y., Sugio, K., Virmani, A., McIntire, D.D., Minna, J.D., and Gazdar, A.F. (1995). Allele-specific chromosome 3p deletions occur at an early stage in the pathogenesis of lung carcinoma. JAMA 273, 558–563. Ji, L., Minna, J.D., and Roth, J.A. (2005). 3p21.3 tumor suppressor cluster: prospects for translational applications. Future Oncol. 1, 79–92.

Tanaka, K., Kondo, K., Kitajima, K., Muraoka, M., Nozawa, A., and Hara, T. (2013). Tumor-suppressive function of protein-tyrosine phosphatase nonreceptor type 23 in testicular germ cell tumors is lost upon overexpression of miR142-3p microRNA. J. Biol. Chem. 288, 23990–23999.

Kamijo, T., Zindy, F., Roussel, M.F., Quelle, D.E., Downing, J.R., Ashmun, R.A., Grosveld, G., and Sherr, C.J. (1997). Tumor suppression at the mouse INK4a locus mediated by the alternative reading frame product p19ARF. Cell 91, 649–659.

Wistuba, I.I., Behrens, C., Milchgrub, S., Bryant, D., Hung, J., Minna, J.D., and Gazdar, A.F. (1999). Sequential molecular abnormalities are involved in the multistage development of squamous cell lung carcinoma. Oncogene 18, 643–650.

Kharitidi, D., Apaja, P.M., Manteghi, S., Suzuki, K., Malitskaya, E., Roldan, A., Gingras, M.-C., Takagi, J., Lukacs, G.L., and Pause, A. (2015). Interplay of endosomal pH and ligand occupancy in integrin a5b1 ubiquitination, endocytic sorting, and cell migration. Cell Rep. 13, 599–609.

Wittinger, M., Vanhara, P., El-Gazzar, A., Savarese-Brenner, B., Pils, D., Anees, M., Grunt, T.W., Sibilia, M., Holcmann, M., Horvat, R., et al. (2011). hVps37A Status affects prognosis and cetuximab sensitivity in ovarian cancer. Clin. Cancer Res. 17, 7816–7827.

1900 Cell Reports 15, 1893–1900, May 31, 2016