Advertisement
Molecular and Cellular Pathobiology

PP2A Inactivation Mediated by PPP2R4 Haploinsufficiency Promotes Cancer Development

Ward Sents, Bob Meeusen, Petar Kalev, Enrico Radaelli, Xavier Sagaert, Eline Miermans, Dorien Haesen, Caroline Lambrecht, Mieke Dewerchin, Peter Carmeliet, Jukka Westermarck, Anna Sablina and Veerle Janssens
DOI: 10.1158/0008-5472.CAN-16-2911 Published December 2017

Abstract

Protein phosphatase 2A (PP2A) complexes counteract many oncogenic kinase pathways. In cancer cells, PP2A function can be compromised by several mechanisms, including sporadic mutations in its scaffolding A and regulatory B subunits or more frequently through overexpression of cellular PP2A inhibitors. Here, we identify a novel genetic mechanism by which PP2A function is recurrently affected in human cancer, involving haploinsufficiency of PPP2R4, a gene encoding the cellular PP2A activator PTPA. Notably, up to 70% of cancer patients showed a heterozygous deletion or missense mutations in PPP2R4. Cancer-associated PTPA mutants exhibited decreased abilities to bind the PP2A-C subunit or activate PP2A and failed to reverse the tumorigenic phenotype induced by PTPA suppression, indicating they function as null alleles. In Ppp2r4 gene-trapped (gt) mice showing residual PTPA expression, total PP2A activity and methylation were reduced, selectively affecting specific PP2A holoenzymes. Both PTPAgt/gt and PTPA+/gt mice showed higher rates of spontaneous tumors, mainly hematologic malignancies and hepatocellular adenomas and carcinomas. These tumors exhibited increased c-Myc phosphorylation and increased Wnt or Hedgehog signaling. We observed a significant reduction in lifespan in PTPA+/gt mice compared with wild-type mice. In addition, chemical-induced skin carcinogenesis was accelerated in PTPA+/gt compared with wild-type mice. Our results provide evidence for PPP2R4 as a haploinsufficient tumor suppressor gene, defining a high-penetrance genetic mechanism for PP2A inhibition in human cancer. Cancer Res; 77(24); 6825–37. ©2017 AACR.

Introduction

Protein phosphatase 2A (PP2A) enzymes comprise a large family of Ser/Thr-specific phosphatases, the majority of which are heterotrimers, consisting of a catalytic C, a scaffolding A and a regulatory B subunit (1). On the basis of the number of known A, B, and C subunits, nearly 100 structurally different PP2A complexes theoretically exist, each characterized by its specific substrate specificity, regulation, tissue-specific expression, and subcellular localization, all largely determined by the regulatory B subunit (2). Many PP2A holoenzymes are involved in the negative control of kinase-driven oncogenic pathways (3, 4), and their inactivation contributes to tumor initiation or progression in many different tissues and cells (5).

The most convincing data supporting the antioncogenic role of PP2A derive from models of transformation of human epithelial cells by a defined set of genetic elements (6). These studies revealed that expression of telomerase catalytic subunit (hTERT) and inhibition of tumor suppressors p53 and pRb are sufficient for immortalization, whereas additional expression of oncogenic H-RasV12 and inhibition of PP2A are required to achieve full transformation (7–11). Among the various B subunits, specifically depletion of B'α, B'γ, and B”α contributed to the PP2A inhibition–mediated cell transformation process in HEK293 cells (7, 10), and multiple relevant molecular targets of these tumor suppressive PP2A complexes were identified, including c-Myc, Akt, p70 S6K, β-catenin, RalA, and p27/KIP (8–13).

In addition, there is ample evidence for functional impairment of PP2A in solid cancers and hematologic malignancies (5, 14, 15). Cancer-specific inactivating mutations or deletions have been found in both genes encoding A subunit isoforms (Aα, Aβ; refs. 16–19), and to some extent in B subunit encoding genes (15, 20–22). Increased expression of PP2A inhibitory oncoproteins, such as SET (Suvar/Enhancer-of-zeste/Trithorax; ref. 23), cancerous inhibitor of PP2A (CIP2A; ref. 24), and cAMP-regulated phosphoprotein of 19 kDa (ARPP-19; ref. 25), or of PP2A modulators, such as PP2A methylesterase-1 (PME-1) and α4 (26–28), represent other recurrent PP2A-inactivating events in human cancer, often associated with poor prognosis and therapeutic resistance (5, 14, 15, 24).

In this study, we reveal a novel and highly penetrant genetic mechanism by which PP2A activity can be compromised in cancer: heterozygous loss or inactivating mutation of PPP2R4. PPP2R4 encodes a cellular PP2A chaperone and activator, denoted phosphatase 2A phosphatase activator (PTPA). PTPA is one of several essential modulators (29–32) of PP2A holoenzyme biogenesis, a poorly understood process that serves to control promiscuous activity of free PP2A C subunit (PP2A-C) until its assembly into holoenzymes (33). Although originally identified as an enzymatic stimulator of phospho-Tyr phosphatase activity of PP2A in vitro (34), PTPA appears essential for maintaining phospho-Ser/Thr phosphatase activity of PP2A in vivo (35) and restores phospho-Ser/Thr phosphatase activity of a native, PME-1-bound inactive PP2A form in vitro (36). PTPA activity requires ATP/Mg2+ (34), the binding of which occurs at a composite PP2A–PTPA binding pocket, and serves to modulate incorporation of catalytic metal ions into the PP2A-C active site (37). In addition, PTPA may affect PP2A conformation through its ATP/Mg2+-dependent prolyl-peptidyl cis/trans isomerase activity towards a unique PP2A-C Proline residue (38, 39). Consistent with its role in generating active PP2A holoenzymes, knockdown of PTPA resembles a PP2A-deficient state (35), which, depending on the context and remaining PTPA levels, fully transforms immortalized human epithelial cells (10), causes apoptosis (35), or protects from apoptosis and mediates chemoresistance (40). PTPA deletion in vivo is lethal under normal growth conditions in fruitfly (41) and yeast (35, 42–45), further underscoring its essential role in PP2A regulation.

Here, our genetic and functional analyses of cancer-associated PTPA alterations reveal that impaired PTPA function and resulting PP2A inhibition are of relevance in a strikingly large fraction of human cancers. In addition, we show that Ppp2r4 gene–trapped mice are predisposed to spontaneous tumor development and are more susceptible to chemical-induced carcinogenesis, providing compelling in vivo evidence for the tumor suppressive role of PTPA and PP2A.

Materials and Methods

Site-directed mutagenesis

Wild-type (WT) PTPA cDNA (isoform α; ref. 46) was cloned into GST-tag eukaryotic expression vector (pGMEX-T), (His)6-tag bacterial expression vector (pET15b) and FLAG-tag lentiviral expression vector (pLA CMV-N-Flag). PCR-based site-directed mutagenesis (Stratagene) to generate PTPA point mutants (R41H, E110K, T118M, R134W, G243V, N312K, S314C, V316F) was performed with Pwo polymerase (Roche) and complementary DNA primers (Sigma Genosys) containing the desired mutations (Supplementary Table S1).

Purification of recombinant PTPA

(His)6-tagged WT/mutant PTPA were expressed in E. coli BL21-pLys cells at 37°C, and purified from the soluble lysate via metal affinity chromatography (Ni-NTA, Affiland). After five washes in 50 mmol/L Tris-HCl pH 8.0, 300 mmol/L NaCl, and 13 mmol/L imidazole, and elution in 250 mmol/L imidazole, pooled eluates were dialyzed against 20 mmol/L Tris-HCl pH 7.4/PEG 10,000. Concentration was determined by BCA (Pierce).

Cell lines

HEK293, HEK293T, and A375 cells (ATCC), characterized by short tandem repeat profiling, were used at low passage number (<15) immediately after receipt or after resuscitation from early stocks. Mycoplasma contamination was checked bimonthly with Venor GeM Classic kit (Minerva Biolabs). Plasmid transfections were achieved with PEI. For lentiviral transductions, HEK293T cells were transfected with shPTPA targeting 3′UTR (10), WT or mutant PTPA, or empty vector (pLA CMV-N-Flag), in the presence of pCMV-deltaR8.91 and pMD.G-VSVG, using Turbofect (Thermo Scientific). Twenty-four hours later, supernatants were used to transduce immortalized HEK-TER (10) or A375 cells.

Anchorage-independent growth and xenografting

For soft agar assays, 105 HEK-TER or A375 cells were plated in triplicate into 0.35% Noble agar (Sigma) on top of a 0.5% agar base layer + 10% FCS. 10-14 days (HEK-TER) or 21 days (A375) after seeding, colonies were stained with 0.005% Crystal violet and counted (Motic-AE31 microscope for HEK-TER; ColCount colony (>200 μm) counter (Oxford Optronix) for A375). For tumorigenicity assays, 2 × 106 HEK-TER cells were subcutaneously injected into female BALB/AnNTac-Foxn1nu/nu immunodeficient mice. Per condition, three mice were injected in both flanks (n = 6). Tumor number and size (width2 × (length/2)) were determined 40 days later.

Ppp2r4 gene–trapped (gt) mice

ES cell line XH627 (129P2/OlaHsd; Baygenomics) was used for morula aggregation. Chimeric animals, identified by coat pigmentation, were mated with C57BL/6N females to obtain germline offspring. All mice used for experiments (PTPA+/+, PTPA+/gt and PTPAgt/gt) derived from PTPA+/gt intercrosses, and showed a mixed 129P2/OlaHsd:C57BL/6N genetic background (F2, F3, F4, F6). Genotyping was performed on genomic DNA isolated from tail biopsies. Ppp2r4 WT and gt allele-specific bands (each 500 bp) were amplified using Forw-1/Rev-2 and Forw-1/Rev-1 primer pairs, respectively (Supplementary Table S1). A gt-specific band (850 bp) was obtained with Forw-2/Rev-3 primers (Supplementary Table S1). All mice procedures were approved by KU Leuven ECD (P036-2008 and P109-2013).

5′RACE

Total XH627 cell RNA was extracted (RNeasy; Qiagen) and 5′RACE performed on 1 μg RNA with AMV-reverse transcriptase (Roche) and β-Geo-rev. primer (Supplementary Table S1); a nested β-Geo-rev. (Supplementary Table S1) and oligo dT-anchor primer (Roche) were used for further amplification by Pwo polymerase (Roche). Products were cloned in pBluescript and sequenced.

Vectorette analysis and qPCR

Three libraries (SmaI, ClaI, and EcoRV/SmaI) were generated from tail or ES cell genomic DNA using Vectorette Kit (Sigma), and used as templates for PCR amplification with Vectorette-unit primer/gt forw. (or gt rev.), and Vectorette-unit primer/nested gt forw. (or nested gt rev.; Supplementary Table S1). All fragments were sequenced (LGC Genomics). qPCR was performed in duplicate on 500 ng tail genomic DNA using QPCR SYBR-Green mix (Thermo Scientific) and Forw-2/Rev-4 primers (Supplementary Table S1) in an ABI PRISM-7000 device (Applied Biosystems). Internal control primers were intron1-forw/intron1-rev (Supplementary Table S1).

β-Galactosidase staining

Whole-mount embryos, fixed in 1% formaldehyde (PBS) for 20 minutes, were stained overnight at 30°C in PBS containing 0.5 mg X-gal/mL (Fermentas), 2 mmol/L MgCl2, 5 mmol/L K4Fe(CN)6, and 5 mmol/L K3Fe(CN)6. Following three washes in PBS, embryos were fixed in 4% paraformaldehyde (PBS) overnight at 4°C, and stored in 70% ethanol.

Tissue sample preparation

Mice were anesthetized (Nembutal) and transcardially perfused with saline (NaCl 0.9% Baxter). Biopsies of liver, spleen, enlarged lymph nodes, kidney, heart, and sporadically, other organs with apparent abnormalities were taken, fixed in 4% paraformaldehyde (PBS), and paraffin embedded. Hematoxylin/eosin (H&E)-stained sections (4 μm) of these organs were independently examined by two experienced pathologists. When appropriate, additional peroxidase immunostainings were performed with B220/CD45R (BD Pharmingen), CD3-ε (Santa Cruz Biotechnology), Iba1 (Abcam), β-catenin (1:200, BD Biosciences 610153), and glutamine synthetase antibodies (1:400, BD Biosciences). Some samples were frozen in liquid N2 and kept at −80°C until further use. For PP2A assays, tissues were always used freshly but not frozen.

Analysis of protein extracts

Tissues were homogenized (douncer) in 25 mmol/L Tris-HCl pH 7.6, 150 mmol/L NaCl, 1 mmol/L EDTA, 1 mmol/L EGTA, and protease/phosphatase inhibitors (Roche) for 15 minutes on ice. After clearance (13,000 × g; 20 minutes, 4°C), supernatants were further analyzed. For immunoblotting, samples were resolved on 4%–12% or 3%–8% gels (Bio-Rad) and transferred to polyvinylidene difluoride (PVDF; GE Healthcare). Membranes were blocked in 5% milk in TBS/0.1% Tween-20, developed with primary antibodies (Supplementary Table S2), horseradish peroxidase–coupled secondary antibodies (DAKO) and chemiluminescence (ECL, Perkin Elmer) on the ImageQuant LAS-4000 (GE Healthcare). Quantifications were done with ImageQuantTL software.

Phosphatase assays

PTPA-mediated stimulation of PP2A dimer Tyr phosphatase activity was measured on p-nitrophenyl-phosphate (pNPP) as described previously (47). Ser/Thr phosphatase activity was measured on [32P]-labeled phosphorylase-a for both inactive and active PP2A. Inactive PP2A was collected by GST pulldown from GST-PME-1 expressing HEK293 cells, and in vitro stimulated with PTPA as described previously (48). PP2A activity assays in brain lysates (10–20 mg/mL) were executed in four conditions to enable deduction of PP2A activity: untreated (total activity), with addition of 3 μmol/L recombinant NIPP1 (PP1 activity specifically abolished), with 20 nmol/L Okadaic acid (OA; PP2A activity specifically abolished), and with both NIPP1 and OA (residual activity). Lysates ± NIPP1, ± OA, or ± NIPP1/OA were incubated at 30°C for 10 to 20 minutes with [32P]-phosphorylase-a, and the released [32P]-phosphate measured following CCl3COOH precipitation. Pilot experiments (Supplementary Fig. S1A and S1B) indicated that under the conditions used, NIPP1-inhibited phosphatase activity minus the residual activity could be entirely contributed to PP2A(-like) activity. For activity measurements in B-type subunit immunoprecipitates, brain lysates were precleared with 20 μL protein-A Sepharose (GE Healthcare) for 30 minutes, incubated with 1 μg primary antibody (Supplementary Table S2) for 2 hours at 4°C, and with 20 μL protein-G Sepharose (GE Healthcare) for 30 minutes. After two washes in 0.1% NP-40 (TBS) and two washes in 20 mmol/L Tris-HCl pH 7.4, 1 mmol/L DTT, beads were resuspended in the latter and assayed for [32P]-phosphorylase-a phosphatase activity. All activities were normalized for PP2A-C levels (determined by immunoblotting).

Two-stage DMBA/TPA chemical carcinogenesis

The shaven back skin of 7-week-old mice (F6) was treated with a single dose of dimethylbenz[a]anthracene (DMBA, Acros Organics, 50 μg in 200 μL acetone). Two weeks later, mice were shaved again and treated with 12-O-tetradecanoylphorbol-13-acetate (TPA, Acros Organics, 6 μg in 200 μL acetone) twice a week for 15 weeks. Neoplastic development was evaluated each time before the next TPA treatment.

Results

High-frequency PPP2R4 aberrations in human cancers

cBioportal data analysis (http://www.cbioportal.org; ref. 49) revealed a strikingly high frequency (up to 70%) of heterozygous PPP2R4 deletion in different human cancer types (Fig. 1A). Kaplan–Meier curves showed a significant correlation between heterozygous loss of PPP2R4 and decreased overall survival in several of these cancers, including sarcoma, several types of renal carcinoma and glioma (Fig. 1B), suggesting that heterozygous loss of PPP2R4 is associated with more aggressive tumors and poor prognosis in some tumor types. Importantly, loss of one PPP2R4 allele was overall associated with decreased mRNA expression (Fig. 1C).

Figure 1.
Figure 1.

PPP2R4 heterozygous loss frequently occurs in human cancer. A, Heterozygous PPP2R4 deletion in several human cancers according to cBioportal (dd. 18 July 2017). Homozygous PPP2R4 deletion appeared very rare. Only cancers with ≥10% heterozygous PPP2R4 loss are displayed. B, Kaplan–Meier curves of cBioportal cohorts of ccRCC (n = 448), hepatocellular carcinoma (liver HCC; n = 365), uterine corpus endometrial carcinoma (uterine; n = 242), sarcoma (n = 242), chromophobe renal carcinoma (chRCC, n = 64), “diffuse” glioma (LGG-GBM, n = 729), and papillary renal carcinoma (pRCC, n = 278) showing decreased overall survival in case of heterozygous loss of PPP2R4 (P-values, log-rank test). C, Heterozygous PPP2R4 loss correlates well with decreased mRNA expression, as illustrated, for example, for sarcoma (mRNA scale: log 2; CNA, copy number alteration).

In addition to heterozygous loss of PPP2R4, we found several PPP2R4 exon mutations (Fig. 2A; Supplementary Table S3) in the cBioportal and COSMIC (Catalogue of Somatic Mutations in Cancer; http://cancer.sanger.ac.uk/cosmic) databases. Biochemical characterization of eight cancer-associated PTPA mutants demonstrated that PTPA mutants G243V, N312K, V316F, and S314C lost the ability to interact with the PP2A-C subunit, whereas R41H, E110K, T118M, and R134W mutants retained PP2A-C binding (Fig. 2B).

Figure 2.
Figure 2.

Biochemical characterization of cancer-associated PTPA mutants. A, Overview of PPP2R4 mutants reported in cBioportal and COSMIC databases (dd. 18 July 2017), superposed on human PTPAβ. Region “a” indicates residues Glu72-Ala108 (exon 3B) that are absent in PTPAα; region “b” encompasses residues Tyr234-Leu253, denoted as “active site deep pocket” (39) or “finger loop” (37); region “c” covers residues Ala304-Tyr330, denoted as “lid-linker border” (54). Mutants whose functional impact was tested are italicized and underlined: R41H (head and neck squamous cell carcinoma), E110K (breast carcinoma), R134W (breast carcinoma; colorectal cancer), T118M (ovarian cancer, CLL), N312K (pancreatic cancer), S314C (melanoma), V316F (lung adenocarcinoma), and G243V (lung squamous carcinoma). Full list of PPP2R4 exon mutations is given in Supplementary Table S3. B, Cellular PP2A-C binding assays. GST-tagged WT or mutant PTPAα were expressed in HEK293 cells and endogenous PP2A-C in the GST pulldowns was determined by immunoblotting (IB). C, Phosphorylase-a assay of PTPAα mutants on inactive, PME-1–associated PP2A. Results represent the mean of three independent experiments (*, P < 0.05; ***, P < 0.001 vs. WT). Bottom, Coomassie-stained SDS-PAGE gel of the mutants (lane 1, Mw marker; lanes 2–9, undiluted proteins; lanes 10–17, 1/3 dilution). D, pNPP assay of WT PTPAα, T118M, and N312K mutants on active PP2A dimer. Results represent the average of three independent experiments (***, P < 0.001 vs. WT). Bottom, Coomassie-stained gel of the recombinant proteins (lane 1, Mw marker; lanes 2–4, undiluted proteins; lanes 5–7, 1/3 dilution).

We also monitored PTPA activity of the mutants by analyzing their ability to reactivate Ser/Thr phosphatase activity of PME-1–associated inactive PP2A (36, 48). PP2A-reactivating ability was dramatically impaired for the PP2A-C nonbinding N312K, S314C, V316F, and G243V mutants, as well as for the PP2A-C binding R134W mutant. In contrast, R41H, E110K, and T118M showed no, or at best, a very modest decrease in activity compared with WT PTPA (Fig. 2C). We confirmed these data by analyzing activity of a functional (T118M) and a loss-of-function mutant (N312K) in a pNPP assay, which measures Tyr phosphatase activating ability of PTPA on an active PP2A dimer (47). Consistent with phosphorylase-a assay data, the N312K mutant showed decreased activation ability, whereas T118M had the same activity as WT PTPA (Fig. 2D).

We next used human kidney epithelial cells overexpressing hTERT, SV40 Large T, and H-RasV12 (HEK-TER cells), to assess whether expression of inactive PTPA mutants could rescue the transformed phenotype triggered by PTPA suppression (10). Restoration of WT PTPA expression suppressed anchorage-independent (AI) growth induced by loss of PTPA, whereas PTPA mutants N312K, S314C, V316F, and R134W, were all unable to functionally replace WT PTPA in this assay (Fig. 3A–C). WT PTPA expression also suppressed xenograft formation and growth in immunodeficient mice, whereas S314C and V316F mutants entirely and R134W partially failed to rescue tumor formation and growth, further confirming that these mutants are loss-of-function (Fig. 3D). These data were further underscored in the human melanoma cell line A375, where loss of PTPA, again, significantly increased AI growth, and WT PTPA, but not the melanoma-associated S314C mutant, rescued this effect (Fig. 3E–G).

Figure 3.
Figure 3.

PTPA loss-of-function mutants contribute to human cell transformation. A and B, AI growth of HEK-TER-shGFP, and HEK-TER-shPTPA cells expressing empty vector, WT PTPAα, or cancer-related PTPAα mutants. AI growth experiments (n = 3/condition) were done in duplicate (*, P < 0.05 vs. shPTPA). C, PTPA immunoblots confirm PTPA knockdown and overexpression of PTPA WT/mutants. D, Tumorigenicity of xenografted HEK-TER-shPTPA cells expressing empty vector or PTPA WT/mutants. The graph shows average tumor size (***, P < 0.001 vs. HEK-TER-shPTPA); tumor number is indicated at the bottom. E and F, AI growth of A375 human melanoma cells in the presence of shGFP control, or shPTPA cotransfected with an empty (EV), WT PTPA or S314C mutant PTPA expression vector. Experiments (n = 2/condition) were done in triplicate (*, P < 0.05 vs. shPTPA+EV). G, PTPA immunoblot confirms PTPA knockdown and expression of the FLAG-tagged rescue constructs in A375 cells.

Summarized, five of eight cancer-associated PTPA mutants tested showed impaired biochemical activity in vitro, and transformation assays confirmed that all of four biochemically impaired PTPA mutants tested are loss-of-function. Our results underscore that PTPA is commonly downregulated in human cancer, associated with more aggressive disease in some cancer types.

Generation of Ppp2r4 gene–trapped mice

To further corroborate the role of PTPA in tumorigenic transformation, we generated PTPA-deficient mice. Human PTPA is encoded by a single gene from which multiple splice variants arise. Four of these can be translated into functional proteins, with the α isoform being the most abundant one (46). Likewise, in mice, a single PTPA-encoding gene is present, which lacks however an equivalent exon 3B (Fig. 4A). Using similar methodologies as described for PPP2R4 (46), we identified only one alternative murine transcript (isoform η), which does not give rise to a functional protein due to a frame shift and premature translation stop. Ppp2r4 therefore encodes a single protein, corresponding to human PTPAα.

Figure 4.
Figure 4.

Characterization of Ppp2r4 gt mice. A, Human and mouse PTPA encoding genes with the gt cassette, consisting of a splice acceptor site (SA), β-Geo (encoding β-galactosidase:neomycin phosphotransferase-II fusion protein), and polyadenylation sequence (pA), located in intron 1 of Ppp2r4. The predicted mRNA and fusion protein resulting from an efficient gene-trapping event are indicated. B, Sequence of XH627 ES cell DNA surrounding one genomic insertion site of the gt cassette. C, Genotyping with primers indicated in A to distinguish WT PTPA+/+, heterozygous PTPA+/gt, and homozygous PTPAgt/gt mice. D, Confirmation of a single genomic gt insertion by qPCR on genomic XH627 ES cell DNA with β-Geo–specific primers. The positive control (+) contains exactly one gt copy (51). E, β-Galactosidase staining on whole-mount embryos. While all WT embryos (n = 10) lacked staining, all PTPA+/gt (n = 33) and PTPAgt/gt (n = 13) embryos stained blue, indicative for high and ubiquitous embryonic PTPA expression. F–J, PTPAgt is a hypomorphic allele, characterized by residual PTPA expression in most tissues. PTPA immunoblotting (IB) in PTPA+/+ (n = 5) and PTPAgt/gt (n = 5) spleen (F), brain (G), liver (H) and heart (I), and in PTPA+/+ (n = 3) and PTPAgt/gt (n = 4) testis (J; ***, P < 0.001). Vinculin (VINC) immunoblots served as loading controls. K, Genotypic analysis of PTPA+/gt × PTPA+/gt progeny (n = 357).

To generate Ppp2r4 knockout mice, we retrieved ES cell line XH627 from the International Gene Trap Consortium (www.genetrap.com; ref. 50) as a candidate clone harboring a gene-trapping cassette (gt) in Ppp2r4. 5′-cDNA rapid amplification PCR (5′-RACE) confirmed gt location in Ppp2r4 intron 1 (Fig. 4A). PCR screening and sequencing determined its exact location, 8,381-bp downstream of exon 1 (Fig. 4B). After aggregation of XH627 with acceptor morulae, chimeric mice were bred with C57BL/6N mice to produce germline animals carrying the trapped allele. An allele-specific genotyping protocol was designed to discriminate WT (PTPA+/+), heterozygous (PTPA+/gt), and homozygous (PTPAgt/gt) mice (Fig. 4C). Of initially 95 genotyped mice, all gt-positive mice also showed a Ppp2r4 gt-specific PCR product. To further exclude multiple genomic gt insertions, we performed Vectorette PCR and qPCR on genomic DNA. In all of three Vectorette libraries generated, we obtained only sequences from Ppp2r4 intron 1. qPCR analysis on genomic DNA of PTPA+/+, PTPA+/gt, and PTPAgt/gt mice confirmed a single genomic gt insertion (Fig. 4D). A positive control with exactly one gt insertion (51) was included for quantification (Fig. 4D). In Ppp2r4 gt mice, β-Geo is fused in frame to Ppp2r4 exon 1, implying that expression of the resulting fusion protein is under transcriptional control of the endogenous Ppp2r4 promoter. Upon staining of embryos for β-galactosidase, we found all PTPA+/+ embryos (n = 10) were negative for β-Geo expression, whereas PTPA+/gt (n = 33) and PTPAgt/gt (n = 13) embryos all stained blue (Fig. 4E). These observations again confirm a single genomic gt insertion, and indicate high and ubiquitous embryonic expression of Ppp2r4.

To evaluate gene-trapping efficiency, we quantified PTPA expression in protein lysates from liver, brain, heart, spleen, and testis of homozygous PTPAgt/gt mice. Compared with WT tissues, immunoblots showed barely detectable PTPA expression in heart, significantly reduced expression in spleen, liver and brain (18% ± 5%, 26% ± 2%, and 21% ± 4% residual PTPA), and high remaining expression in testis (84% ± 36%; Fig. 4F–J). Thus, we concluded that PTPAgt constitutes a hypomorphic allele. Heterozygous PTPA+/gt intercrossing resulted in a non-Mendelian distribution of progeny with 36% PTPA+/+, 49% PTPA+/gt and 15% PTPAgt/gt in the offspring (n = 357; Fig. 4K). This suggested impaired neonatal viability or increased embryonic lethality of PTPAgt/gt mice, consistent with observed lethality of PTPA deficiency in yeast (35, 42–45) and Drosophila (41).

PP2A activity is significantly impaired in PTPA-deficient tissues

To measure effects of decreased PTPA expression on PP2A activity, we analyzed phosphorylase-a phosphatase activity in brain extracts from PTPA+/+, PTPA+/gt, and PTPAgt/gt mice, using PP1 and PP2A specificity controls. Brain was the preferred tissue for these assays because of its size, easiness of homogenization and high PP2A expression/activity. Decreased PTPA expression was confirmed in both PTPA+/gt and PTPAgt/gt samples (Fig. 5A and B). Immunoblotting with PP2A-C (de)methylation-sensitive antibodies also revealed significantly decreased PP2A-C methylation (Fig. 5A and B). Specific PP2A-like activity was about 50% and 20% reduced in PTPAgt/gt and PTPA+/gt brain, respectively (Fig. 5C–E). Thus, PP2A activity and methylation are significantly impaired in PTPA-deficient mice, confirming the role of PTPA as a major PP2A activator.

Figure 5.
Figure 5.

Significantly altered PP2A methylation and activity in PTPA-deficient tissue. A, Representative immunoblots (IB) of WT, PTPA+/gt, and PTPAgt/gt brain lysates (dilutions 1/2, 1/4, 1/8) with indicated antibodies. B, Average quantifications (ImageQuant) of immunoblot signals (n = 3). PTPA and PP2A-C were normalized to vinculin; methyl and dimethyl were normalized to PP2A-C (*, P < 0.05; **, P < 0.01; ***, P < 0.001 vs. WT). C, Decreased PP2A-like activity in PTPA+/gt and PTPAgt/gt brain lysates (***, P < 0.001 vs. WT). Assays were performed on freshly prepared lysates (n = 3/genotype), with PP1/PP2A specificity controls and phosphorylase-a as substrate. Activities were normalized to PP2A-C expression levels. D and E Time (D) and concentration curves (E) show results in a linear range. PP2A activity in immunoprecipitates of B55/Bα (F), B56/B'γ (G), and B56/B'ε subunits (H) isolated from brain lysates of WT, PTPA+/gt, and PTPAgt/gt mice (n = 3/genotype). Activities were normalized to coprecipitating PP2A-C levels (***, P < 0.001 vs. WT).

Total PP2A activity in a given tissue represents the overall Ser/Thr phosphatase activity of the complete set of different PP2A complexes present in a given stoichiometric ratio (33). To determine whether PTPA deficiency might preferentially affect specific holoenzymes, we immunoprecipitated B55/Bα, B56/B'γ, and B56/B'ε from PTPA+/gt and PTPAgt/gt brain lysates, and measured associated PP2A activity in the immunoprecipitates (IPs), relative to the amount of coprecipitating PP2A-C. We found significantly decreased PP2A activity in B56/B'ε IPs of both PTPA+/gt and PTPAgt/gt mice, and in B56/B'γ IPs of PTPAgt/gt mice, whereas PP2A activity remained unaffected in B55/Bα IPs (Fig. 5F–H;Supplementary Fig. S2). Thus, our in vivo data demonstrate that decreased PTPA expression selectively affects activity of specific PP2A holoenzymes.

Spontaneous cancer development in PTPA+/gt and PTPAgt/gt mice

To assess whether decreased PTPA expression might affect tumorigenesis, a cohort of PTPA+/+ (n = 29), PTPA+/gt (n = 30), and PTPAgt/gt mice (n = 37) was aged up to 24 months. At different ages, mice were sacrificed and representative samples from major organs and grossly detectable lesions were collected for histopathologic examination. A spontaneous tumor phenotype was observed in PTPA+/gt and PTPAgt/gt mice with age ≥12 months (Fig. 6A). The most common neoplasms encountered were hematologic malignancies, in particular histiocytic sarcoma and lymphomas, originating from spleen, (mesenteric) lymph nodes, and Gut-associated lymphoid tissue (GALT), and with secondary involvement of liver and kidney (Supplementary Fig. S3A–S3D). Hepatocellular adenoma or carcinoma was less common (Fig. 6A; Supplementary Fig. S3B). Notably, survival curves underscored that PTPA+/gt mice showed a significantly decreased life span when compared with WT mice (Fig. 6B), further corroborating the tumor phenotype in these mice, and human cBioportal survival data (Fig. 1B–D).

Figure 6.
Figure 6.

Tumor development in PTPA+/gt and PTPAgt/gt mice. A, Incidence of spontaneous tumorigenesis in WT, PTPA+/gt, and PTPAgt/gt mice as a function of age, based on systematic analysis of spleen, lymph nodes, liver, and kidney. B, Survival curves for PTPA+/+ (n = 13) and PTPA+/gt (n = 13) mice indicate reduced life-span for PTPA+/gt (log-rank test, P = 0.013). Survival was followed up to 104 weeks, with death taken as endpoint. C, DMBA/TPA induced papilloma incidence in WT (n = 9) and PTPA+/gt (n = 16) mice. At 12.5 weeks, the largest difference is seen (Chi-square test, P = 0.16). D, Average onset of papilloma formation (unpaired t test, P = 0.15). E, Number of lesions per mouse (unpaired t test, P = 0.08 at 12.5 weeks). F, Average volume/lesion in function of time.

Increased chemical-induced carcinogenesis in PTPA+/gt mice

To provide further support for the tumor suppressive role of PTPA, we subjected WT and PTPA+/gt mice to the well-established DMBA/TPA-induced skin carcinogenesis procedure (52). Interestingly, papilloma formation showed a trend to occur earlier in PTPA+/gt mice, on average 11.5 weeks versus 14 weeks after TPA treatment (Fig. 6C and D). In accordance, the average number of lesions per mouse appeared higher in PTPA+/gt mice up to 13 weeks after TPA treatment (Fig. 6E). Moreover, the average lesion size was overall increased in PTPA+/gt mice (Fig. 6F), despite the fact that some lesions regressed (Fig. 6C and E), as previously described for this model (52). Notably, during the 15-week-follow-up-period, progression to squamous cell carcinoma was only observed in PTPA+/gt mice (in one case). These data further support a major role for heterozygous loss of Ppp2r4 in tumor initiation.

Alterations of protein phosphorylation associated with decreased PTPA expression

Previous work demonstrated activation of Wnt signaling, Akt and c-Myc in transformed HEK-TER cells depleted from PTPA (10). Importantly, these oncogenes or oncogenic pathways are all suppressed by PP2A complexes of the B56/B' family (3–5, 15) whose very activities are selectively suppressed in PTPA+/gt and PTPAgt/gt mice (Fig. 5F–H). To check whether phosphorylation of these PP2A targets is affected in vivo, we performed targeted immunoblot analysis of WT, and noncancerous and neoplastic PTPA+/gt and PTPAgt/gt spleens. Although we confirmed reduced PTPA expression in PTPA+/gt (54% ± 9%) and PTPAgt/gt (9% ± 3%) samples, we found that expression of some PP2A subunits was slightly upregulated, particularly in PTPAgt/gt neoplastic samples (Supplementary Fig. S4A and S4B). We did not find consistent differences in phospho-Akt levels, whereas increased phosphorylation of c-Myc (Ser62) was observed in most PTPAgt/gt neoplastic spleens (Fig. 7A and B). Sporadically, we observed increased expression of the Hedgehog transcription factor Gli1 and the Wnt transcription coactivator β-catenin (Fig. 7A). Additional IHC analysis of spontaneous hepatocellular carcinomas from PTPA+/gt and PTPAgt/gt mice, using β-catenin and glutamine synthetase antibodies, confirmed activated Wnt signaling, specifically in the tumors, in one out of three samples of each genotype (Fig. 7C). These findings suggest a contributing role of c-Myc, Wnt, and/or Hedgehog activation to tumor initiation/progression in PTPA-deficient mice.

Figure 7.
Figure 7.

Oncogenic phosphorylation alterations of known PP2A targets in PTPA-deficient tissues. A, Immunoblots (IB) of WT (n = 5), noncancerous PTPA+/gt (n = 5) and PTPAgt/gt (n = 5), and neoplastic (indicated by ‘°’) PTPA+/gt (n = 6) and PTPAgt/gt (n = 4) spleens with indicated antibodies. Numbers indicate quantifications of a given band relative to an internal control (total protein or vinculin); underlined, increased levels for an individual sample (difference >2 SD towards WT average). Wnt and Hedgehog activation were scored positive (+) or negative (−) through expression of β-catenin and Gli1, respectively. B, Graphs indicate average values (±SD) of the quantified band intensities of all samples within a given condition, normalized to an internal control (vinculin or total protein), and calculated relative to WT (*, P < 0.05; **, P < 0.01 and |fold change| >1.5). C, Immunohistochemical staining of PTPA-deficient livers with occurrence of hepatocellular adenomas/adenocarcinomas using β-catenin and glutamine synthetase antibodies revealed activated Wnt signaling in one of three PTPA+/gt and one of three PTPAgt/gt samples, specifically in the tumor.

Discussion

Inactivation of tumor suppressive PP2A complexes is a recurrent event in human cancer. The current study highlights a novel mechanism of PP2A inactivation, involving heterozygous loss or loss-of-function mutation of PPP2R4, a gene encoding the PP2A activator PTPA. Our work also provides direct in vivo evidence for PP2A's tumor suppressor function through demonstration of a spontaneous cancer phenotype and increased chemical-induced tumor initiation in hypomorphic Ppp2r4 gene-trapped mice, which represent a model of inactivation of select PP2A holoenzymes.

Our biochemical and functional analyses showed that at least five of the cancer-associated PPP2R4 mutations are loss-of-function. Analysis of reported crystal structures (37, 39, 53, 54) can rationalize these observations. For instance, N312, S314, and V316 residues cluster near a major PTPA–PP2A-C interface, whereas R134 and G243 directly neighbor two crucial residues in another PTPA–PP2A-C interface (37). In yeast, the residue corresponding to R134 appears crucial for PTPA functionality (39). V316 directly binds to the PP2A-C carboxyterminal tail (53). Accordingly, a PTPA V316D mutant lost binding to PP2A-C (37, 54), whereas reintroduction of a V316A mutant in a yeast PTPA deletion strain could not restore PTPA function (39). F135D and V244D mutants showed reduced PP2A-C binding and diminished PP2A-C activating ability (37). G243 and V244 are located in the best-conserved part of PTPA, deletion of which severely reduced phospho-Tyr (55) and phospho-Ser/Thr phosphatase activation of PP2A (38) and diminished prolyl-peptidyl isomerase activity of PTPA (38). Thus, these structure–function data underscore that many cancer-related PPP2R4 mutations affect crucially important PTPA domains.

cBioportal data also indicated a strikingly high incidence of heterozygous PPP2R4 loss, associated with poor patient survival in several cancers. The spontaneous cancer phenotype, encountered in both PTPA+/gt and PTPAgt/gt mice, the increased chemical-induced skin carcinogenesis in PTPA+/gt mice, together with significantly decreased survival of PTPA+/gt mice as opposed to WT mice, further underscore these human cancer data, and suggest a major contribution of PPP2R4 heterozygous loss to tumor initiation or progression. Although hematologic cancers and hepatocellular carcinoma were the predominant neoplasms in PTPA-deficient mice, other tumor types may appear upon further systematic analysis of additional organs, or upon coexpression of specific oncogenes.

These in vivo findings confirm and extend studies in cellular models of human transformation, where PTPA, B56/B'γ, and PP2A-C methylation, which are all affected in Ppp2r4 gt mice, proved crucially important to prevent H-RasV12-induced cell transformation to varying degrees (7, 10, 11). Notably, in the DMBA/TPA skin carcinogenesis model, activated H-Ras is the likely oncogenic factor as well (52). Although suppression of a single tumor suppressive B-type subunit, such as B56/B'α or B56/B'γ, resulted in partial transformation (7, 10), suppression of PTPA mimicked knockdown of PP2A-C, or expression of SV40 small t much better, showing the strongest transformation phenotype (10). Likewise, cosuppression of B56/B'γ and the PP2A methyltransferase LCMT-1 was required before an effect of LCMT-1 knockdown on transformation could be observed (11). Consequently, more than one tumor suppressive PP2A complex may need to be impaired to allow full transformation, and the contribution of PP2A inactivation to tumor development/progression may involve dysregulation of more than one (proto)oncogenic substrate.

Our analysis of several signaling pathways known to be regulated by PP2A underscores this view, as we observed increased phosphorylation/activation of different established PP2A-B56/B' targets in PTPA-deficient neoplastic spleens (c-Myc S62, Hedgehog, and canonical Wnt signaling). Exactly which oncogenic signaling dysfunctions might occur in a given PTPA-deficient tumor is expected to be determined by the random, “spontaneous” oncogenic alteration(s) involved.

Although we initially predicted a lethal outcome of Ppp2r4 loss, PTPAgt/gt mice turned out to be viable due to the hypomorphic nature of the gt allele. In PTPA+/gt and PTPAgt/gt brain, expressing around 60% and 20% of original PTPA levels, total PP2A activity was decreased by 20% and 50%, respectively, concomitantly with decreased PP2A-C methylation. As only an active PP2A-C conformation is efficiently methylated (32), the latter observation indirectly underscores that PTPA is a PP2A activator. Moreover, PTPA deficiency resulted in inactivation of a specific subset of PP2A holoenzymes (all B56/B' trimers), whereas another subset remained catalytically competent (B55/B trimers). Conflicting phenotypic data from several studies (10, 35, 40) might be explained like that, as different levels of PTPA deficiency, affecting different PP2A holoenzymes, could indeed yield different phenotypic outcomes. As such, the functional consequences of PTPA “deficiency” become reminiscent of the effects of progressively decreasing PP2A-C methylation (56) and cancer-associated PPP2R1A haploinsufficiency (8), which selectively affect binding of particular B-type subunits.

In conclusion, this study presents compelling in vivo evidence for the critical tumor suppressor function of PP2A, and establishes PPP2R4 as a novel haploinsufficient tumor suppressor gene that is frequently impaired by heterozygous loss or loss-of-function mutation in cancer. The fact that PPP2R4 heterozygous loss correlates with decreased overall survival in distinct cancer types opens interesting perspectives for its clinical use as a predictive or prognostic biomarker.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Authors' Contributions

Conception and design: W. Sents, P. Kalev, D. Haesen, J. Westermarck, V. Janssens

Development of methodology: W. Sents, B. Meeusen, P. Kalev, E. Radaelli

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): B. Meeusen, P. Kalev, E. Radaelli, X. Sagaert, C. Lambrecht, M. Dewerchin, P. Carmeliet

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): W. Sents, B. Meeusen, P. Kalev, E. Radaelli, X. Sagaert, A. Sablina, V. Janssens

Writing, review, and/or revision of the manuscript: W. Sents, B. Meeusen, E. Radaelli, D. Haesen, M. Dewerchin, P. Carmeliet, J. Westermarck, A. Sablina, V. Janssens

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): P. Kalev

Study supervision: V. Janssens

Other (masters thesis): E. Miermans

Grant Support

This work was supported by KU Leuven Research Fund (GOA/08/016 to V. Janssens; OT/13/094 to V. Janssens and A. Sablina), Research Foundation-Flanders (G.0582.11 to V. Janssens; G.0598.12N to M. Dewerchin; G.0690.12N to P. Carmeliet), Methusalem funding (to P. Carmeliet), Belgian IAP program (P7/13 to V. Janssens), Flemish Agency for Innovation by Science and Technology (to C. Lambrecht and D. Haesen), Belgian Foundation Against Cancer (FAF-F/2016/822 to V. Janssens), Sigrid Juselius Foundation and Foundation of Finnish Cancer Institute (to J. Westermarck). W. Sents received an Emmanuel van der Schueren fellowship of the Flemish Cancer League.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Acknowledgments

We thank Prof. Beullens and N. Sente for phosphorylase-a and recombinant NIPP1; Profs. Dilworth, Hemmings, and Ogris for antibodies; Prof. Zwijsen for gt control, Prof. Agostinis for A375 cells, and LPPP members for inspiring discussions.

Footnotes

  • Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

  • Received October 25, 2016.
  • Revision received August 19, 2017.
  • Accepted October 13, 2017.

References

  1. 1.↵
    1. Janssens V,
    2. Goris J
    . Protein phosphatase 2A: a highly regulated family of serine/threonine phosphatases implicated in cell growth and signalling. Biochem J 2001;353:417–39.
    OpenUrlAbstract/FREE Full Text
  2. 2.↵
    1. Lambrecht C,
    2. Haesen D,
    3. Sents W,
    4. Ivanova E,
    5. Janssens V
    . Structure, regulation, and pharmacological modulation of PP2A phosphatases. Methods Mol Biol 2013;1053:283–305.
    OpenUrlCrossRefPubMed
  3. 3.↵
    1. Janssens V,
    2. Rebollo A
    . The role and therapeutic potential of Ser/Thr phosphatase PP2A in apoptotic signaling networks in human cancer cells. Curr Mol Med 2012;12:268–87.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Eichhorn PJ,
    2. Creyghton MP,
    3. Bernards R
    . Protein phosphatase 2A regulatory subunits and cancer. Biochim Biophys Acta 2009;1795:1–15.
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. Perrotti D,
    2. Neviani P
    . Protein phosphatase 2A: a target for anticancer therapy. Lancet Oncol 2013;14:e229–38.
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. Hahn WC,
    2. Weinberg RA
    . Rules for making human tumor cells. N Engl J Med 2002;347:1593–603.
    OpenUrlCrossRefPubMed
  7. 7.↵
    1. Chen W,
    2. Possemato R,
    3. Campbell KT,
    4. Plattner CA,
    5. Pallas DC,
    6. Hahn WC
    . Identification of specific PP2A complexes involved in human cell transformation. Cancer Cell 2004;5:127–36.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Chen W,
    2. Arroyo JD,
    3. Timmons JC,
    4. Possemato R,
    5. Hahn WC
    . Cancer-associated PP2A Aalpha subunits induce functional haploinsufficiency and tumorigenicity. Cancer Res 2005;65:8183–92.
    OpenUrlAbstract/FREE Full Text
  9. 9.↵
    1. Sablina AA,
    2. Chen W,
    3. Arroyo JD,
    4. Corral L,
    5. Hector M,
    6. Bulmer SE,
    7. et al.
    The tumor suppressor PP2A Abeta regulates the RalA GTPase. Cell 2007;129:969–82.
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. Sablina AA,
    2. Hector M,
    3. Colpaert N,
    4. Hahn WC
    . Identification of PP2A complexes and pathways involved in cell transformation. Cancer Res 2010;70:10474–84.
    OpenUrlAbstract/FREE Full Text
  11. 11.↵
    1. Jackson JB,
    2. Pallas DC
    . Circumventing cellular control of PP2A by methylation promotes transformation in an Akt-dependent manner. Neoplasia 2012;14:585–99.
    OpenUrlCrossRefPubMed
  12. 12.↵
    1. Yeh E,
    2. Cunningham M,
    3. Arnold H,
    4. Chasse D,
    5. Monteith T,
    6. Ivaldi G,
    7. et al.
    A signaling pathway controlling c-myc degradation that impacts oncogenic transformation of human cells. Nat Cell Biol 2004;6:308–18.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Miller JP,
    2. Yeh N,
    3. Hofstetter CP,
    4. Keskin D,
    5. Goldstein AS,
    6. Koff A
    . p27kip1 levels reflect a nexus of oncogenic signaling during cell transformation. J Biol Chem 2012;287:19775–85.
    OpenUrlAbstract/FREE Full Text
  14. 14.↵
    1. Haesen D,
    2. Sents W,
    3. Lemaire K,
    4. Hoorne Y,
    5. Janssens V
    . The basic biology of PP2A in hematologic cells and malignancies. Front Oncol 2014;4:347.
    OpenUrlPubMed
  15. 15.↵
    1. Sangodkar J,
    2. Farrington CC,
    3. McClinch K,
    4. Galsky MD,
    5. Kastrinsky DB,
    6. Narla G
    . All roads lead to PP2A: exploiting the therapeutic potential of this phosphatase. FEBS J 2016;283:1004–24.
    OpenUrlCrossRefPubMed
  16. 16.↵
    1. Calin GA,
    2. di Iasio MG,
    3. Caprini E,
    4. Vorechovsky I,
    5. Natali PG,
    6. Sozzi G,
    7. et al.
    Low frequency of alterations of the alpha (PPP2R1A) and beta (PPP2R1B) isoforms of the subunit A of the serine-threonine phosphatase 2A in human neoplasms. Oncogene 2000;19:1191–5.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Shih IeM,
    2. Wang TL
    . Mutation of PPP2R1A: a new clue in unveiling the pathogenesis of uterine serous carcinoma. J Pathol 2011;224:1–4.
    OpenUrlCrossRefPubMed
  18. 18.↵
    1. Haesen D,
    2. Abbasi Asbagh L,
    3. Derua R,
    4. Hubert A,
    5. Schrauwen S,
    6. Hoorne Y,
    7. et al.
    Recurrent PPP2R1A mutations in uterine cancer act through a dominant-negative mechanism to increase malignant growth. Cancer Res 2016;76:5719–31.
    OpenUrlAbstract/FREE Full Text
  19. 19.↵
    1. Wang SS,
    2. Esplin ED,
    3. Li JL,
    4. Huang L,
    5. Gazdar A,
    6. Minna J,
    7. et al.
    Alterations of the PPP2R1B gene in human lung and colon cancer. Science 1998;282:284–7.
    OpenUrlAbstract/FREE Full Text
  20. 20.↵
    1. Cheng Y,
    2. Liu W,
    3. Kim ST,
    4. Sun J,
    5. Lu L,
    6. Sun J,
    7. et al.
    Evaluation of PPP2R2A as a prostate cancer susceptibility gene: a comprehensive germline and somatic study. Cancer Genet 2011;204:375–81.
    OpenUrlCrossRefPubMed
  21. 21.↵
    1. Curtis C,
    2. Shah SP,
    3. Chin SF,
    4. Turashvili G,
    5. Rueda OM,
    6. Dunning MJ,
    7. et al.
    The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature 2012;486:346–52.
    OpenUrlCrossRefPubMed
  22. 22.↵
    1. Kalev P,
    2. Simicek M,
    3. Vazquez I,
    4. Munck S,
    5. Chen L,
    6. Soin T,
    7. et al.
    Loss of PPP2R2A inhibits homologous recombination DNA repair and predicts tumor sensitivity to PARP inhibition. Cancer Res 2012;72:6414–24.
    OpenUrlAbstract/FREE Full Text
  23. 23.↵
    1. Haesen D,
    2. Sents W,
    3. Ivanova E,
    4. Lambrecht C,
    5. Janssens V
    . Cellular inhibitors of protein phosphatase PP2A in cancer. Biomed Res 2012;23:197–211.
    OpenUrl
  24. 24.↵
    1. Khanna A,
    2. Pimanda JA,
    3. Westermarck J
    . Cancerous inhibitor of protein phosphatase 2A, an emerging human oncoprotein and a potential cancer therapy target. Cancer Res 2013;73:6548–53.
    OpenUrlAbstract/FREE Full Text
  25. 25.↵
    1. Song H,
    2. Pan J,
    3. Liu Y,
    4. Wen H,
    5. Wang L,
    6. Cui J,
    7. et al.
    Increased ARPP-19 expression is associated with hepatocellular carcinoma. Int J Mol Sci 2014;16:178–92.
    OpenUrl
  26. 26.↵
    1. Puustinen P,
    2. Junttila MR,
    3. Vanhatupa S,
    4. Sablina AA,
    5. Hector ME,
    6. Teittinen K,
    7. et al.
    PME-1 protects extracellular signal-regulated kinase pathway activity from protein phosphatase 2A-mediated inactivation in human malignant glioma. Cancer Res 2009;69:2870–7.
    OpenUrlAbstract/FREE Full Text
  27. 27.↵
    1. Wandzioch E,
    2. Pusey M,
    3. Werda A,
    4. Bail S,
    5. Bhaskar A,
    6. Nestor M,
    7. et al.
    PME-1 modulates protein phosphatase 2A activity to promote the malignant phenotype of endometrial cancer cells. Cancer Res 2014;74:4295–305.
    OpenUrlAbstract/FREE Full Text
  28. 28.↵
    1. Chen LP,
    2. Lai YD,
    3. Li DC,
    4. Zhu XN,
    5. Yang P,
    6. Li WX,
    7. et al.
    α4 is highly expressed in carcinogen-transformed human cells and primary human cancers. Oncogene 2011;30:2943–53.
    OpenUrlCrossRefPubMed
  29. 29.↵
    1. Hombauer H,
    2. Weismann D,
    3. Mudrak I,
    4. Stanzel C,
    5. Fellner T,
    6. Lackner DH,
    7. et al.
    Generation of active protein phosphatase 2A is coupled to holoenzyme assembly. PLoS Biol 2007;5:e155.
    OpenUrlCrossRefPubMed
  30. 30.↵
    1. Kong M,
    2. Ditsworth D,
    3. Lindsten T,
    4. Thompson CB
    . Alpha4 is an essential regulator of PP2A phosphatase activity. Mol Cell 2009;36:51–60.
    OpenUrlCrossRefPubMed
  31. 31.↵
    1. Jiang L,
    2. Stanevich V,
    3. Satyshur KA,
    4. Kong M,
    5. Watkins GR,
    6. Wadzinski BE,
    7. et al.
    Structural basis of protein phosphatase 2A stable latency. Nat Commun 2013;4:1699.
    OpenUrlCrossRefPubMed
  32. 32.↵
    1. Stanevich V,
    2. Jiang L,
    3. Satyshur KA,
    4. Li Y,
    5. Jeffrey PD,
    6. Li Z,
    7. et al.
    The structural basis for tight control of PP2A methylation and function by LCMT-1. Mol Cell 2011;41:331–42.
    OpenUrlCrossRefPubMed
  33. 33.↵
    1. Sents W,
    2. Ivanova E,
    3. Lambrecht C,
    4. Haesen D,
    5. Janssens V
    . The biogenesis of active protein phosphatase 2A holoenzymes: a tightly regulated process creating phosphatase specificity. FEBS J 2013;280:644–61.
    OpenUrlCrossRefPubMed
  34. 34.↵
    1. Cayla X,
    2. Van Hoof C,
    3. Bosch M,
    4. Waelkens E,
    5. Vandekerckhove J,
    6. Peeters B,
    7. et al.
    Molecular cloning, expression, and characterization of PTPA, a protein that activates the tyrosyl phosphatase activity of protein phosphatase 2A. J Biol Chem 1994;269:15668–75.
    OpenUrlAbstract/FREE Full Text
  35. 35.↵
    1. Fellner T,
    2. Lackner DH,
    3. Hombauer H,
    4. Piribauer P,
    5. Mudrak I,
    6. Zaragoza K,
    7. et al.
    A novel and essential mechanism determining specificity and activity of protein phosphatase 2A (PP2A) in vivo. Genes Dev 2003;17:2138–50.
    OpenUrlAbstract/FREE Full Text
  36. 36.↵
    1. Longin S,
    2. Jordens J,
    3. Martens E,
    4. Stevens I,
    5. Janssens V,
    6. Rondelez E,
    7. et al.
    An inactive protein phosphatase 2A population is associated with methylesterase and can be re-activated by the phosphotyrosyl phosphatase activator. Biochem J 2004;380:111–9.
    OpenUrlAbstract/FREE Full Text
  37. 37.↵
    1. Guo F,
    2. Stanevich V,
    3. Wlodarchak N,
    4. Sengupta R,
    5. Jiang L,
    6. Satyshur KA,
    7. et al.
    Structural basis of PP2A activation by PTPA, an ATP-dependent activation chaperone. Cell Res 2014;24:190–203.
    OpenUrlCrossRefPubMed
  38. 38.↵
    1. Jordens J,
    2. Janssens V,
    3. Longin S,
    4. Stevens I,
    5. Martens E,
    6. Bultynck G,
    7. et al.
    The protein phosphatase 2A phosphatase activator is a novel peptidyl-prolyl cis/trans-isomerase. J Biol Chem 2006;281:6349–57.
    OpenUrlAbstract/FREE Full Text
  39. 39.↵
    1. Leulliot N,
    2. Vicentini G,
    3. Jordens J,
    4. Quevillon-Cheruel S,
    5. Schiltz M,
    6. Barford D,
    7. et al.
    Crystal structure of the PP2A phosphatase activator: implications for its PP2A-specific PPIase activity. Mol Cell 2006;23:413–24.
    OpenUrlCrossRefPubMed
  40. 40.↵
    1. MacKeigan JP,
    2. Murphy LO,
    3. Blenis J
    . Sensitized RNAi screen of human kinases and phosphatases identifies new regulators of apoptosis and chemoresistance. Nat Cell Biol 2005;7:591–600.
    OpenUrlCrossRefPubMed
  41. 41.↵
    1. Zhang F,
    2. Huang ZX,
    3. Bao H,
    4. Cong F,
    5. Wang H,
    6. Chai PC,
    7. et al.
    Phosphotyrosyl phosphatase activator facilitates localization of Miranda through dephosphorylation in dividing neuroblasts. Development 2016;143:35–44.
    OpenUrlAbstract/FREE Full Text
  42. 42.↵
    1. Rempola B,
    2. Kaniak A,
    3. Migdalski A,
    4. Rytka J,
    5. Slonimski PP,
    6. di Rago JP
    . Functional analysis of RRD1 (YIL153w) and RRD2 (YPL152w), which encode two putative activators of the phosphotyrosyl phosphatase activity of PP2A in Saccharomycescerevisiae. Mol Gen Genet 2000;262:1081–92.
    OpenUrlCrossRefPubMed
  43. 43.↵
    1. Van Hoof C,
    2. Janssens V,
    3. De Baere I,
    4. de Winde JH,
    5. Winderickx J,
    6. Dumortier F,
    7. et al.
    The Saccharomyces cerevisiae homologue YPA1 of the mammalian phosphotyrosyl phosphatase activator of protein phosphatase 2A controls progression through the G1 phase of the yeast cell cycle. J Mol Biol 2000;302:103–20.
    OpenUrlCrossRefPubMed
  44. 44.↵
    1. Bernal M,
    2. Sanchez-Romero MA,
    3. Salas-Pino S,
    4. Daga RR
    . Regulation of fission yeast morphogenesis by PP2A activator pta2. PLoS One 2012;7:e32823.
    OpenUrlCrossRefPubMed
  45. 45.↵
    1. Goyal A,
    2. Simanis V
    . Characterization of ypa1 and ypa2, the Schizosaccharomyces pombe orthologs of the peptidyl proyl isomerases that activate PP2A, reveals a role for Ypa2p in the regulation of cytokinesis. Genetics 2012;190:1235–50.
    OpenUrlAbstract/FREE Full Text
  46. 46.↵
    1. Janssens V,
    2. Van Hoof C,
    3. Martens E,
    4. De Baere I,
    5. Merlevede W,
    6. Goris J
    . Identification and characterization of alternative splice products encoded by the human phosphotyrosyl phosphatase activator gene. Eur J Biochem 2000;267:4406–13.
    OpenUrlPubMed
  47. 47.↵
    1. Van Hoof C,
    2. Cayla X,
    3. Bosch M,
    4. Merlevede W,
    5. Goris J
    . The phosphotyrosyl phosphatase activator of protein phosphatase 2A. A novel purification method, immunological, and enzymic characterization. Eur J Biochem 1994;226:899–907.
    OpenUrlPubMed
  48. 48.↵
    1. Longin S,
    2. Zwaenepoel K,
    3. Martens E,
    4. Louis JV,
    5. Rondelez E,
    6. Goris J,
    7. et al.
    Spatial control of protein phosphatase 2A (de)methylation. Exp Cell Res 2008;314:68–81.
    OpenUrlCrossRefPubMed
  49. 49.↵
    1. Cerami E,
    2. Gao J,
    3. Dogrusoz U,
    4. Gross BE,
    5. Onur Sumer S,
    6. Arman Aksoy B,
    7. et al.
    The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2012;2:401–4.
    OpenUrlAbstract/FREE Full Text
  50. 50.↵
    1. Nord AS,
    2. Chang PJ,
    3. Conklin BR,
    4. Cox AV,
    5. Harper CA,
    6. Hicks GG,
    7. et al.
    The international gene trap consortium website: a portal to all publicly available gene trap cell lines in mouse. Nucleic Acids Res 2006;34:D642–8.
    OpenUrlCrossRefPubMed
  51. 51.↵
    1. Soriano P
    . Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet 1999;21:70–1.
    OpenUrlCrossRefPubMed
  52. 52.↵
    1. Abel EL,
    2. Angel JM,
    3. Kiguchi K,
    4. DiGiovanni J
    . Multi-stage chemical carcinogenesis in mouse skin: fundamentals and applications. Nat Protoc 2009;4:1350–62.
    OpenUrlCrossRefPubMed
  53. 53.↵
    1. Löw C,
    2. Quistgaard EM,
    3. Kovermann M,
    4. Anandapadamanaban M,
    5. Balbach J,
    6. Nordlund P
    . Structural basis for PTPA interaction with the invariant C-terminal tail of PP2A. Biol Chem 2014;395:881–9.
    OpenUrl
  54. 54.↵
    1. Chao Y,
    2. Xing Y,
    3. Chen Y,
    4. Xu Y,
    5. Lin Z,
    6. Li Z,
    7. et al.
    Structure and mechanism of the phosphotyrosyl phosphatase activator. Mol Cell 2006;23:535–46.
    OpenUrlCrossRefPubMed
  55. 55.↵
    1. Van Hoof C,
    2. Janssens V,
    3. Dinishiotu A,
    4. Merlevede W,
    5. Goris J
    . Functional analysis of conserved domains in the phosphotyrosyl phosphatase activator. Molecular cloning of the homologues from Drosophila melanogaster and Saccharomyces cerevisiae. Biochemistry 1998;37:12899–908.
    OpenUrlCrossRefPubMed
  56. 56.↵
    1. Longin S,
    2. Zwaenepoel K,
    3. Louis JV,
    4. Dilworth S,
    5. Goris J,
    6. Janssens V
    . Selection of protein phosphatase 2A regulatory subunits is mediated by the C terminus of the catalytic Subunit. J Biol Chem 2007;282:26971–80.
    OpenUrlAbstract/FREE Full Text
Back to top
View this article with LENS

Open full page PDF
Article Alerts
Email Article
Citation Tools
PP2A Inactivation Mediated by PPP2R4 Haploinsufficiency Promotes Cancer Development
Ward Sents, Bob Meeusen, Petar Kalev, Enrico Radaelli, Xavier Sagaert, Eline Miermans, Dorien Haesen, Caroline Lambrecht, Mieke Dewerchin, Peter Carmeliet, Jukka Westermarck, Anna Sablina and Veerle Janssens
Cancer Res December 15 2017 (77) (24) 6825-6837; DOI: 10.1158/0008-5472.CAN-16-2911
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement