University of California Riverside

UNIVERSITY OF CALIFORNIA RIVERSIDE

Characterization of Tumor Derived B56gamma Mutations and Their Effect on Tumor Suppressor Function of B56gamma PP2A

A Dissertation submitted in partial satisfaction of the requirements for the degree of

Doctor of Philosophy

Biochemistry and Molecular Biology

Yumiko Nobumori

June 2013

Dissertation Committee: Dr. Xuan Liu, Chairperson Dr. Frances M. Sladek Dr. Jolinda A. Traugh

The Dissertation of Yumiko Nobumori is approved:

______

______Committee Chairperson

University of California, Riverside

ACKNOWLEDGEMENTS

First and foremost, I would like to express my deep gratitude to Dr. Xuan Liu for her continued guidance, support, and encouragement to work in this project. I would like to thank the members of my committee Dr. Frances M. Sladek and Dr. Jolinda A. Traugh for valuable advice and encouragement. I would also like to thank past and present members of my laboratory for providing suggestions and assistance. I am grateful to Dr. Ernest Martinez, Dr. Frank U. Sauer, and members of their laboratories, and members of Dr. Frances M. Sladek’s laboratory who have provided research suggestions and experimental advice during laboratory meeting. Last but not least, I would like to thank my husband, Jonathan, for his love, kindness, and constant support throughout entire process.

The materials in Chapter 2 were originally published in Oncogene. 2010; 29:3933-3941.

Publishing Group, a division of Macmillan Publishers Limited. The materials in Chapter 3 were originally published in The journal of biological chemistry. 2012; 287:11030-11036. Y.

Nobumori, G.P. Shouse and X. Liu. HEAT repeat 1 motif is required for B56γ-containing protein phosphatase 2A (B56γ-PP2A) holoenzyme assembly and tumor-suppressive function. Copyright

ABSTRACT OF THE DISSERTATION

Characterization of Tumor Derived B56gamma Mutations and Their Effect on Tumor Suppressor Function of B56gamma PP2A

Yumiko Nobumori

Doctor of Philosophy, Graduate Program in Biochemistry and Molecular Biology University of California, Riverside, June 2013 Dr. Xuan Liu, Chairperson

The heterotrimeric PP2A holoenzyme consists of a scaffolding A, a catalytic C, and a variable regulatory B subunit. Previous study has reported that B56γ subunit mediates the tumor suppressive function of PP2A. B56γ-PP2A inhibits cell transformation by dephosphorylating and activating p53. In addition, B56γ-PP2A possesses the p53-independent tumor suppressive function although its mechanism is unclear. In the present study, we demonstrate the mechanism of B56γ tumor suppressive function inactivation through characterizing B56γ mutations identified in human cancer.

First, we demonstrate that the A383G and F395C mutations disrupt B56γ from interacting and activating p53. The mutants are unable to dephosphorylate p53 and thus lack the p53-dependent tumor suppressive function. Moreover, we identify the region adjacent to these mutations as the p53 binding domain. This finding implies that the bridging interaction between

B56γ and p53 is required for p53 dephosphorylation by PP2A.

Second, we show that the C39R mutation within the HEAT repeat 1 disrupts B56γ binding with the AC core and thus abolishes the B56γ tumor suppressive function completely.

We further demonstrate that the intact HEAT repeat 1 is required for B56γ to bind the AC core, providing structural insights into the B56γ-PP2A holoenzyme assembly.

Further characterization of additional B56γ mutations reveals two classes of mutations with different mechanisms of tumor suppressive function inactivation. Mutations in the first class disrupt the interaction between B56γ and the AC core and thus fail to regulate all substrates and completely lose the tumor suppressive function. Mutations in the second class prevent specific substrates from binding B56γ and partially reduce the tumor suppressive function. Although it remains to be investigated how frequently these mutations occur, our results underline the importance of B56γ tumor suppressive function in human cancer.

Finally, we identified the S220N mutation that specifically disrupts the p53-independent tumor suppressive function of B56γ, suggesting that it may disrupt the binding of unknown substrates. Identification of these proteins would contribute to the further understanding of the tumor suppressor role of B56γ in cells.

TABLE OF CONTENTS

Chapter 1: Introduction ...... 1

References ...... 18

Figure legend ...... 31

Figures ...... 32

Tables ...... 33

Chapter 2: A B56γ mutation in lung cancer disrupts the p53-dependent tumor suppressor function

of protein phosphatase 2A ...... 35

Abstract ...... 36

Introduction ...... 37

Materials and Methods ...... 39

Results ...... 41

Discussion ...... 48

References ...... 51

Acknowledgements ...... 54

Figure legend ...... 55

vii

Figures ...... 59

Chapter 3: HEAT repeat 1 is required for B56γ-PP2A holoenzyme assembly and tumor

suppressive function ...... 66

Abstract ...... 67

Introduction ...... 68

Materials and Methods ...... 71

Results ...... 74

Discussion ...... 79

References ...... 82

Acknowledgements ...... 84

Figure legend ...... 85

Figures...... 88

Chapter 4: Characterization of B56γ tumor-associated mutations reveals mechanisms for

inactivation of B56γ-PP2A ...... 94

Abstract ...... 95

Introduction ...... 96

viii

Materials and Methods ...... 98

Results ...... 100

Discussion ...... 108

References ...... 110

Acknowledgements ...... 112

Figure legend ...... 113

Figures...... 115

Tables ...... 120

Chapter 5: Identification of B56γ-PP2A substrates responsible for its p53-independent tumor

suppressive function ...... 122

Introduction ...... 123

Materials and Methods ...... 125

Results ...... 128

Discussion ...... 134

References ...... 136

Acknowledgements ...... 140

Figure legend ...... 141

Figures...... 143

Tables ...... 147

Chapter 6: Conclusion...... 154

References ...... 159

Figure legend ...... 160

Figure ...... 161

LIST OF FIGURES

Figure 1.1 Structure of PP2A. 19

Figure 2.1 Mapping of the B56γ domain required for interaction with p53. 59

Figure 2.2 Mutant B56γ3 is unable to promote p53 Thr55 dephosphorylation. 60

Figure 2.3 Overexpression of cancer derived mutant B56γ3 is unable to block 61

anchorage-independent growth in a p53-dependent manner.

Figure 2.4 Overexpression of cancer derived mutant B56γ3 is unable to block cell 62

proliferation in a p53-dependent manner.

Figure S2.1 HA-B56γ3 is unable to interact with endogenous B56γ. 63

Figure S2.2 Statistical analysis of cell growth and transformation data. 64

Figure S2.3 The A383G mutant is unable to promote p53-dependent tumor suppressive 65

functions.

Figure 3.1 HEAT repeat 1 motif of B56γ is required for B56γ-PP2A holoenzyme 88

assembly.

Figure 3.2 The C39R mutation disrupts interaction of B56γ with A and C subunits. 89

Figure 3.3 HEAT repeat 1 mutants of B56γ fail to inhibit cell proliferation. 90

Figure 3.4 HEAT repeat 1 mutants of B56γ fail to inhibit anchorage-independent cell 91

growth.

Figure S3.1 Comparison of CD spectra of wild type B56γ (WT) and C39R in the far UV 92

region.

Figure S3.2 Statistical analysis of cell growth and transformation data. 93

Figure 4.1 Distribution of B56γ mutations identified by AceView. 115

Figure 4.2 Effect of B56γ mutants on tumor suppressor function. 116

Figure 4.3 Interaction of B56γ mutants with PP2A A and C and p53. 117

Figure 4.4 S220N interacts with the AC core and p53. 118

Figure 4.5 Inactivation of B56γ-PP2A tumor suppressive function by B56γ mutations. 119

Figure 5.1 Plasmid maps for pcDNA4/TO-HA-B56γ and pcDNA6/TR. 143

Figure 5.2 Tetracycline-inducible B56γ protein expression and p21 induction. 144

Figure 5.3 Testing the pulldown efficiency of WT and S220N by HA 145

immunoprecipitation.

Figure 5.4 Analysis of WT B56γ and S220N complex proteins purified by HA- 146

immunoprecipitation.

Figure 6.1 A model for the tumor suppressive function mediated by the interaction 161

between B56 and substrates.

xii

LIST OF TABLES

Table 1.1 Nomenclature and subcellular localization of PP2A B subunits. 33

Table 1.2 Summary of proteins dephosphorylated by PP2A, classified by B subunits. 34

Table 4.1 List of tumor-derived B56γ mutations. 120

Table 4.2 Summary of fourteen tumor-derived B56γ mutations’ tumor suppressive 121

functional assays.

Table 5.1 List of potential B56γ interacting proteins identified by LC-MS/MS. 147

Table 5.2 List of potential B56γ interacting proteins detected in both WT B56γ and 150

S220N mutant complexes.

Table 5.3 List of potential B56γ interacting proteins detected in WT B56γ but not in 151

S220N mutant protein complex.

Table 5.4 List of potential B56γ interacting proteins detected in S220N mutant but not in 153

WT B56γ protein complex.

xiii

LIST OF ABBREVIATIONS

ATM: ataxia telangiectasia mutated

ATR: ataxia telangiectasia and Rad3-related protein

BMP: bone morphogenetic protein

CASK: calcium/calmodulin-dependent serine protein kinase

CD: circular dichroism

Cdc25: cell division cycle mutant 25

CDK: cyclin dependent kinase

CMV: cytomegalovirus c-Myc: cellular myelocytomatosis

CREB: cAMP responsive element binding protein

DMEM: Dulbecco's Modified Eagle Medium

DMSO: dimethyl sulfoxide

DNA: deoxyribonucleic acid

DSP: dual-specificity phosphatase emPAI: exponentially modified protein abundance index

ERK: extracellular signal regulated kinase

GAP: GTPase-activating protein

GEF: guanine nucleotide exchange factors

GIT1: ARF GTPase-activating protein

xiv

GSK3β: glycogen synthase kinase 3β

GST: glutathione-S-transferase

HA: hemagluttinin

HBV: hepatitis B virus

HCV: hepatitis C virus

HEAT: huntington-elongation-A subunits-TOR-like

IB: immunoblotting

IEX-1: immediate early response gene X-1

IL: intra-loop

IP: immunoprecipitation

IR: ionizing radiation

LAR: leukocyte common antigen-related protein tyrosine phosphatase

LC-MS/MS: liquid chromatography-tandem mass spectrometry

LCMT: leucine carboxyl methyltransferase 1

M: mock

MAPK: mitogen-activated protein kinase

Mdm2: mouse double minute 2

NCBI: National Center for Biotechnology Information

OA: okadaic acid

PBS: phosphate buffered saline

PCR: polymerase chain reaction

Phos: phosphorylation

PKA: protein kinase A

PKB: protein kinase B

PKC: protein kinase C

PKR: protein kinase RNA-activated

PME-1: PP2A methylesterase 1

PP1: protein phosphatase 1

PP2A: protein phosphatase 2A

PPM: metal-dependent protein phosphatase

PPP: phosphoprotein phosphatase

PSTP: protein serine/threonine phosphatase

PTP: protein tyrosine phosphatase

PTPA: peptidyl prolyl isomerase

PyMT: polyoma middle T antigen

PyST: polyoma small t antigen

RNA: ribonucleic acid

RPMI media: Roswell Park Memorial Institute media

RT-PCR: reverse transcription polymerase chain reaction

SBP: streptavidin-binding peptide

SDS-PAGE: sodium dodecyl sulfate polyacrylamide gel electrophoresis

SEM: standard error of the mean

xvi

SV40: Simian virus 40

TGF-β: transforming growth factor β

Ubiq: ubiquitination

UV: ultraviolet

xvii

CHAPTER 1

Introduction

Protein phosphatases

Phosphorylation is the most common type of post-translational modification of proteins and is essential for the proper regulation of signaling pathways in cells. In response to environmental cues, cells quickly undergo dynamic adaptations involving reversible phosphorylation via complex networks of kinases and phosphatases. In eukaryotic cells, phosphorylation mainly occurs on three hydroxyl-containing amino acids, serine, threonine, and tyrosine residues which account for 86.4%, 11.8%, and 1.8%, respectively, of the phosphorylated residues (Olsen et al., 2006). According to the specific amino acids that they dephosphorylate, phosphatases can be categorized into either serine/threonine phosphatases (PSTP) or protein tyrosine phosphatases (PTP). There are approximately 30-40 PSTPs in eukaryotic cells. Based on the sequence, catalytic function, and structure, PSTPs can be further subdivided into three major families: phosphoprotein phosphatases (PPP), metal-dependent protein phosphatases (PPM), and aspartate-based phosphatases. The PPP family includes PP1, PP2A, PP2B (also known as calcineurin), PP4, PP5, PP6, and PP7. All members of the PPP family share conserved catalytic subunits and function as multimeric holoenzymes consisting of a catalytic subunit, and one or more of a number of regulatory and scaffolding subunits. The PPM family includes PP2C and pyruvate dehydrogenase phosphatase and its activity is dependent on manganese (Mg+2) or magnesium (Mn+2) ions. Unlike the PPP family, the PPM family has a single subunit with multiple isoforms encoded by different genes that contain regulatory sequences that determine the substrate specificity. The more recently identified aspartate-based phosphatases include FCP1 and

SCP1 and target only one substrate, the C-terminal domain of RNA polymerase II (Gallego and

Virshup, 2005; Shi, 2009). In comparison, phosphatases of the PTP superfamily are encoded by over 100 genes and are defined by a characteristic motif containing cysteine at the active site.

Classical PTPs possess a conserved catalytic domain and are subdivided based on their cellular

localization: transmembrane receptor and cytoplasmic non-receptor phosphatases. In addition, the

PTP superfamily includes the dual-specificity phosphatases (DSP). The DSPs have a similar catalytic mechanism as the classical PTPs but their active site is structured in a way that it can accommodate not only the tyrosine, but also serine and threonine residues (Tonks, 2006). Various phosphatases display exquisite substrate specificity and functions, hence control a diverse array of cell signaling responses playing a critical role in biological processes.

PP2A (Protein Phosphatase 2A)

PP2A is arguably the most important PSTP and is highly conserved from yeast to human.

PP2A accounts for 30-50% of all the serine/threonine dephosphorylation activity depending on the type of cells and tissues. PP2A is ubiquitously expressed and highly abundant, accounting for up to 1% of total cellular protein in some tissues. PP2A function has been implicated in a multitude of signaling pathways that control development, cell proliferation and death, cell mobility, cytoskeleton dynamics, and control of cell cycle (Schönthal, 2001; Sontag, 2001;

Janssens et al., 2008; Virshup and Shenolikar, 2009). As such, deregulation of PP2A has a profound effect on cell function and ultimately can lead to cancer, Alzheimer’s disease, and viral and parasitic diseases (Lim and Lu, 2005; Eichhorn et al., 2009).

PP2A is not a single protein but a collection of multiple phosphatases. PP2A can exist either as a heterodimeric core enzyme or a heterotrimeric holoenzyme. The heterodimeric core enzyme consists of a scaffolding A subunit and a catalytic C subunit (AC core). The A and C subunits are each encoded by two ubiquitously expressed genes, α and β, where the α isoform is

10-fold more abundant than the β isoform in most tissues (Khew-Goodall and Hemmings, 1988;

Hendrix et al., 1993b; Bosch et al., 1995). With one of the variable B subunits bound to the AC

core, the heterotrimeric PP2A holoenzyme is believed to exhibit distinct substrate specificity as well as spatially and temporally determined functions. The B subunits are comprised of four gene families based on sequence homology: B (B55 or PR55), B’ (B56 or PR61), B”

(PR48/59/72/130), and B”’ (PR93/110). Each B subunit family consists of two to five isoforms, which many contain alternative splice variants. Isoforms within the same family share highly conserved sequence, however there is no sequence similarity across different gene families.

Various regulatory B subunits display differential expression depending on tissue, cell types, and developmental stages (Li and Virshup, 2002). Cellular and tissue localizations for the representative B subunits are listed in Table 1.1. A large number of PP2A holoenzyme combinations are possible and this most likely accounts for the diverse range of phospho-proteins and signaling pathways targeted by PP2A.

An increasing number of reports now demonstrate that PP2A holoenzymes with different

B subunits exert diverse effects on many signaling pathways, emphasizing the importance of B subunits in PP2A regulation. Intriguingly, PP2A holoenzymes with varying B subunits often display antagonizing effects while targeting the same pathway as seen in the TGF-β signaling

(Batut et al., 2008), MAPK pathway (Sieburth et al., 1999; Abraham et al., 2000; Silverstein et al.,

2002; Ory et al., 2003; Adams et al., 2005; Junttila et al., 2008), and Wnt signaling (Seeling et al.,

1999; Li et al., 2001). Furthermore, PP2A has been implicated to function both positively and negatively in the control of cell cycle by regulating multiple pocket proteins and CDK substrates

(Reviewed in Wurzenberger and Gerlich, 2011; Kurimchak and Graña, 2012). List of some of the known individual substrate targeted by PP2A holoenzymes with specific B subunits is shown in

Table 1.2. Given that the B subunit has a key regulatory role in determining PP2A holoenzyme activity, recent studies focus on the individual B subunits to characterize PP2A function.

Regulation of PP2A

The A and C subunits

Precise functions of PP2A in a broad spectrum of signaling pathways are achieved by its highly complex regulatory mechanisms. Since activities of PP2A are determined by the varying combinations of subunits, it is apparent that proper regulation of the subunits and their assembly is critical. Recent reports suggest that the distinct expression pattern of the A subunit contributes to the regulation of PP2A function. The scaffolding A subunit is encoded by two genes, Aα

(PPP2R1A) and Aβ (PPP2R1B), respectively. Although two isoforms share 87% sequence homology, they have different binding affinity for specific C subunit isoforms and various B subunits. At the mRNA level, the Aα isoform is approximately 40-fold more abundant than Aβ isoform (Hemmings et al., 1990). This difference can be partially explained by recent findings that the promoters of the Aα and Aβ genes are differentially regulated by multiple transcription factors including Ets-1, CREB, AP2α, SP1/SP3 and RXRα/β (Chen et al., 2009; Liu et al., 2012).

At the protein level, the Aα isoform is 10 times more abundant than the Aβ isoform in adult tissues. However, expression of the Aβ isoform is higher than the Aα isoform in the ovary and during the early stages of development in Xenopus (Hendrix et al., 1993b; Bosch et al., 1995;

Zhou et al., 2003). Changes in the protein expression levels of the Aα and Aβ isoforms have been associated with cancer and ocular diseases, underlining the importance of controlled A subunit expression in cells (Palczewski et al., 1990; Kantorow et al., 1998; Colella et al., 2001).

Nevertheless, much remains unknown about the alterations in the Aα and Aβ isoforms expression patterns and resulting consequences in cells.

The C subunit expression is also suggested to be regulated at the transcription level. A catalytic C subunit has two isoforms, Cα (PPP2CA) and Cβ (PPP2CB), which share 97%

sequence homology. Both isoforms are ubiquitously expressed, but the mRNA expression of the

Cα isoform is approximately 10-fold higher than the Cβ isoform because of its stronger promoter

(Khew-Goodall and Hemmings, 1988). Despite the high sequence homology, loss of the Cα isoform is lethal in yeast and mice and cannot be rescued by the Cβ isoform, suggesting that there are functions specific to each isoforms (Götz et al., 1998). In general, the expression of the C subunit is maintained at constant level via tight autoregulation mechanism through translational regulation (Baharians and Schönthal, 1998). The Cα isoform expression is regulated by SP1 and

CREB binding at its TATA-less promoter (Sunahori et al., 2010). And increased C subunit expression is associated with viral infections including HBV, HCV (Christen et al., 2007) and

CMV (Hakki and Geballe, 2008) which cause CREB activation. In contrast, the expression of the

Cβ isoform is unlikely to be regulated by CREB as its promoter lacks a CRE element for CREB to bind. Additional studies are required to fully understand the regulation of C subunit expression.

In addition to the protein expression level, the enzyme activity of PP2A is regulated through posttranslational modification of the C subunit. Posttranslational modification of the highly conserved C-terminal tail of the C subunit (304-TPDYFL-309) is believed to be involved in modulating PP2A holoenzyme formation and enzyme activity. Reversible carboxyl methylation on Leu309 of the C-terminal tail is a conserved mechanism that has been demonstrated to promote the binding of some, but not all families of the B subunits to the AC core enzyme. As such, Leu309 methylation may modulate substrate specificity and enzyme function of PP2A by affecting the B subunit association (Ogris et al., 1997; Bryant et al., 1999; Tolstykh et al., 2000;

Wei et al., 2001; Ikehara et al., 2007). Leu309 methylation of the C-terminal tail is regulated by the methyltransferase LCMT1 and methylesterase PME-1 (Ogris et al., 1999a; Longin et al.,

2007). When PME-1 forms complex with the C subunit, it inactivates PP2A activity by preventing LCMT1 from methylating Leu309 as well as by removing the catalytic metal ions

from the active site (Xing et al., 2008). In opposition, a cis-trans prolyl isomerase PTPA can bind the A subunit and disrupt the interaction between PME-1 and PP2A, leading to the activation of

PP2A (Longin et al., 2004; Hombauer et al., 2007). Moreover, PTPA changes conformation of the catalytic site of PP2A holoenzyme and thus may alter PP2A enzyme activity and substrate specificity (Jordens et al., 2006; Leulliot et al., 2006). Leu309 methylation status fluctuates throughout the cell cycle, suggesting that its control mechanism is dependent on the cell cycle

(Turowski et al., 1995; Lee and Pallas, 2007). Nevertheless, the precise cellular processes involved in maintaining the balance of LCMT1 and PME-1 activity and functional significance of

Leu309 methylation regulation in vivo remain to be investigated.

In addition to methylation, the C-terminal tail of the C subunit is transiently phosphorylated by the tyrosine kinases pp60v-src and pp56lck in response to growth factors and cell transformation signals (Chen et al., 1992), indicating that PP2A regulation is an important part of signal transduction by kinase cascades. Phosphorylation on Tyr307 of the C-terminal tail has been shown to affect the recruitment of B55 and B56α, β, and ε subunits to the AC core and also to inactivate PP2A activity by inhibiting methylation on Leu309. On the other hand, phosphorylation on Tyr304 selectively inhibits the binding of B55 but not B56 and B”’ subunits, and does not affect Leu309 methylation (Yu et al., 2001; Gentry et al., 2005; Longin et al., 2007).

While posttranslational modification on the C-terminal tail of the C subunit appears to be pivotal for PP2A function in vivo, it still requires further experiments to delineate the circumstances and physiological consequences of such posttranslational modifications on PP2A holoenzyme activity.

The B subunit

Binding of the B subunit to the AC core determines the subcellular localization, the substrate specificity, as well as the enzyme activity of PP2A holoenzyme. As such, a large number of the B subunits provides foundation for the multi-faced role of PP2A in cells. So far, 15

B subunit genes with multiple, distinct alternative transcripts and additional splice variants have been identified in human (Eichhorn et al., 2009). As shown in Table 1.1, the expressions of the B subunits are regulated in spatially and developmentally specific manner and diverse B subunits in different cell types, tissues, and developmental stages likely manipulate environmentally precise function of PP2A. One of the four gene families, the B56 family has at least five isoforms, α

(PPP2R5A), β (PPP2R5B), γ (PPP2R5C), δ (PPP2R5D), and ε (PPP2R5E) (Csortos et al., 1996;

McCright et al., 1996). The B56γ gene has at least four alternative splice transcripts (Muneer et al., 2002; Ortega-Lázaro and del Mazo, 2003) and the B56δ gene has three alternative splice variants (Csortos et al., 1996). All isoforms share a central region which is approximately 80% identical in sequence, but diverge more toward the C- and N-termini. While the B56α and B56γ isoforms are highly expressed in heart and skeletal muscle, the B56β, B56δ, and B56ε isoforms are expressed predominantly in brain (McCright and Virshup, 1995; Csortos et al., 1996; Tehrani et al., 1996). Consistent with this, recent studies have demonstrated the participation of B56β and

B56δ in neuronal functions (Saraf et al., 2007; Van Kanegan and Strack, 2009; Louis et al., 2011).

Evidence suggests that differential tissue and cellular expression of the B subunits is an important aspect in the functional control of PP2A. This is especially important because to certain extent there is functional redundancy among some members within the same family of B subunit. For instance, an overexpression of any one of the B56 isoforms downregulates β-catenin and reduces

Wnt signaling in vitro (Seeling et al., 1999). Surprisingly, during the development of Xenopus embryo in vivo, B56α and B56γ inhibit Wnt signaling, but B56ε acts as a positive regulator and is

required for the organogenesis (Seeling et al., 1999; Li et al., 2001). Discrepancy between the in vitro and in vivo functions among the B56 isoforms may be explained by the fact that these isoforms display dynamic differential expression patterns throughout the developmental stages.

Specific PP2A function is likely achieved through particular localization of the individual B56 isoforms in the embryo (Everett et al., 2002; Baek and Seeling, 2007). Abundant variety of the B subunits along with their spatially and temporally regulated differential expression is an important aspect of PP2A regulation mechanism.

In addition to the mRNA and protein expression level, the B subunit is also subjected to posttranslational modifications which lead to the altered protein stability, PP2A holoenzyme assembly, and enzyme activity. In particular, the B56 family members are known as phosphoproteins (McCright et al., 1996). For example, B56α is phosphorylated by PKR at Ser28, which promotes B56α localization in mitochondria and enhanced dephosphorylation of Bcl2 (Ruvolo et al., 2008). B56γ is phosphorylated at Ser327 by ERK in the presence of IEX-1, which results in dissociation of B56γ from the AC core and the PP2A inactivity (Letourneux et al., 2006). In response to DNA damage, ATM phosphorylates B56γ at Ser510, which stabilizes B56γ by preventing degradation and increases PP2A activity toward p53 (Shouse et al., 2011). Similarly,

B56δ is phosphorylated at Ser37 by Chk1, leading to increased PP2A activity toward Cdc25 upon replication checkpoint activation (Margolis et al., 2006). In neurons, B56δ is phosphorylated at

Ser566 by PKA, which augments B56δ –PP2A function (Ahn et al., 2007; Stipanovich et al.,

2009). These reports suggest that either the protein level or affinity for the AC core or substrates of particular B subunit is regulated temporally and spatially through posttranslational modifications to alter the overall function of PP2A under different conditions.

Structure of PP2A

Recent crystallographic studies provide a mechanistic basis for the PP2A holoenzyme assembly and phosphatase activity. As shown in Figure 1.1A, the monomeric A subunit forms an elongated, L-shaped structure, composed of 15 tandem repeated Huntington-elongation-A subunit-TOR (HEAT) sequences (Groves et al., 1999). Each HEAT motif consists of a conserved

39-residue sequence organized into a pair of antiparallel α-helices connected by the intrahelical loop, stacked in a consecutive array. Aligned intrahelical loops from 15 HEAT repeats form a contiguous ridge on one side of the A subunit.

Crystal structure of the dimeric AC core enzyme revealed interesting information in regards to the PP2A holoenzyme assembly. The catalytic C subunit has a compact ellipsoidal structure (Figure 1.1B). The C subunit interacts with part of the ridge (intrahelical loop 11-15) of the A subunit (Ruediger et al., 1992, 1994) where extensive hydrogen bonding makes association of the A and C subunits relatively stable (Xing et al., 2006). The catalytic cleft of the AC core is somewhat shallow compared to that of most protein kinases, explaining the absence of specific target substrate motif by the AC core alone. The most remarkable aspect of the AC core assembly is structural flexibility of the A subunit. Upon the C subunit binding, the A subunit undergoes dramatic conformational adjustments and bends to adopt a horse-shoe shape conformation (Xing et al., 2006; Xu et al., 2006; Cho and Xu, 2007). This conformational flexibility of the A subunit is likely the key for PP2A activity as it enables to accommodate not only the binding of the C subunit, but especially the structurally varying B subunits.

To date, heterotrimeric PP2A holoenzymes containing two families of the B subunit,

B56γ and B55α, have been successfully crystalized, revealing distinct structures for each. PP2A holoenzymes comprising the B” or B”’ subunits have not yet been crystalized. Four families of

the B subunit are unrelated by sequence except that they have a conserved sequence in common which is used to interact with the A subunit, making the binding of different B subunits to the AC core mutually exclusive (Ruediger et al., 1999). As such, each B subunit is presumably very different in their structures and manner of binding the AC core. Considering the role of B subunit in regulating the substrate specificity of PP2A, it is predicted that the association of specific B subunit with the AC core potentially creates a unique substrate binding site and facilitates the dephosphorylation of target proteins by directing them to be positioned in the catalytic site.

B56γ consists of eight pseudo HEAT repeats motifs, forming an elongated, solenoid- shaped structure as shown in Figure 1.1C (Xu et al., 2006; Cho and Xu, 2007). Interestingly,

B56γ structure closely resembles that of the A subunit even though there is no sequence similarity.

It is notable that once B56γ-PP2A holoenzyme assembles, the C-terminal tail of the C subunit resides at the interface between B56γ and the A subunit. Although its exact purpose has not been elucidated, methylation of Leu309 on the C-terminal tail of the C subunit has been suggested to be important for regulating the PP2A holoenzyme assembly and activity through controlling the

B subunit binding. Curiously, methylated Leu309 fits into the negatively charged interface between B56γ and the A subunit, providing an explanation that methylation may promote the binding of B56γ to the AC core by neutralizing the negatively charged carboxyl group of the C- terminal tail of the C subunit (Cho and Xu, 2007). Moreover, B56γ interacts with both the A and

C subunits extensively using the conserved central region. B56γ binding with the A subunit involves the convex side of B56γ (HEAT repeat 2-5) interacting with the ridge on the A subunit

(intrahelical loop 2-8), and B56γ binding with the C subunit involves the rim of B56γ (HEAT repeat 2, 5-8). This leaves an open concave surface of B56γ placed next to the catalytic pocket of

PP2A holoenzyme. Although further study is needed, this surface is predicted to be used for recognition and recruitment of the substrates to the active site for dephosphorylation.

B55α comprises a seven-bladed β propeller with a protruding β-hairpin arm as shown in

Figure 1.1D (Xu et al., 2008). As expected, B55α differs greatly in its structure and manner of holoenzyme assembly compared to B56γ. Notably, B55α lacks the stable association of the methylated C-terminal tail of the C subunit, which is consistent with the biochemical data that show methylation is not required for the in vitro assembly of B55-PP2A holoenzyme (Ikehara et al., 2007). Overall C subunit interaction with B55α involves much smaller area compared to its interaction with B56γ. On the other hand, B55α extensively interacts with the A subunit.

Specifically, the β propeller of B55α interacts with the ridge on the A subunit (intrahelical loop 3-

7) plus the β-hairpin arm of B55α interacts with the intrahelical loop 1-2 of the A subunit. Similar to B56γ, B55α binding with the AC core creates an open surface adjacent to the catalytic pocket, suggesting a potential substrate binding site. Taken together, these studies offer structural explanation in regards to the distinct role of individual B subunit in determining the substrate specificity and regulating PP2A phosphatase activity.

PP2A in cancer

Deregulation of cell growth is a hallmark of tumorigenic cell transformation. PP2A is implicated in regulating many critical signaling pathways and is considered to be an important tumor suppressor against carcinogenesis. Regulation of protein kinase cascades is one of the major targets of PP2A. PP2A serves to reverse the action of kinases in major signaling pathways including the MAPK signaling (Sonoda et al., 1997; Junttila et al., 2008), PKC signaling

(Nunbhakdi-Craig et al., 2002; Adams et al., 2005; Xu et al., 2008), and Wnt signaling (Seeling et al., 1999; Li et al., 2001; Creyghton et al., 2005). In the classical model of reversible phosphorylation regulation, it was thought that phosphatases reverse the effects of kinases by

dephosphorylating the substrates of kinases. However, it is now apparent that PP2A also dephosphorylates kinases themselves and in many cases kinases are negatively regulated by dephosphorylation. For example, PP2A directly dephosphorylates ERK, Ca2+/calmodulin- dependent kinases, PKA, PKB, PKC, and IκB kinases and modulates their enzyme activities

(Reviewed in Millward et al., 1999). Similarly, various kinases directly phosphorylate and regulate PP2A as well. This feedback signaling mechanism of kinases and phosphatases allows them to regulate each other and maintain cellular phosphorylation in balance. As such, PP2A function impacts a wide range of signaling pathways and deregulation of PP2A can contribute to multiple types of cancer (Eichhorn et al., 2009).

The tumor suppressor role of PP2A was first confirmed when okadaic acid, a PP2A inhibitor, was shown to promote skin tumor growth in mouse (Suganuma et al., 1988, 1990).

Subsequently, several tumor-promoting viral proteins including polyomavirus middle (PyMT) and small (PyST) tumor antigens and simian virus SV40 small (SVST) tumor antigen were found to alter the enzyme activity of PP2A by binding to the AC core and displacing the B subunit

(Pallas et al., 1990; Walter et al., 1990; Yang et al., 1991; Cayla et al., 1993; Kamibayashi et al.,

1994; Campbell et al., 1995; Shtrichman et al., 1999). In spite of their similar AC core binding ability, SV40 and polyomavirus antigens affect PP2A function and cellular signaling pathways by different mechanisms. SVST binds the Aα isoform of the A subunit and displaces the B55 and

B56 subunit from PP2A holoenzyme (Chen et al., 2004; Moule et al., 2004). SVST expression leads to the activation of Akt kinase pathway (Gillet et al., 2001; Yuan et al., 2002) and NF-κB signaling (Sontag et al., 1997; Moreno et al., 2004), promoting cell growth and survival. On the other hand, Py antigens displace the B subunit from the AC core by binding to either Aα and Aβ isoforms of the A subunit (Rodriguez-Viciana et al., 2006). PyST and PyMT are derived from the same genomic DNA as a result of alternative splicing and they have different cellular

localizations; PyST is located in both the cytoplasm and nucleus while PyMT is located in the membrane. Both PyST and PyMT promote cell proliferation but modulate different signaling pathways. PyMT forms complex with Src family tyrosine kinases and mediates cell transformation through activation of numerous growth stimulatory pathways including MAPK and PI3-kinase pathways (Brewster et al., 1997; Ogris et al., 1999b). PyST activates MAPK, but not PI3-kinase pathway (Yuan et al., 2002; Zhao et al., 2003a). Altogether, this indicates the complexity of the effects of PP2A activity in signaling pathways involved in cell growth.

Deregulated expression of tumor suppressor genes is a common phenomenon in cancer.

Aberrant expression patterns of the A subunit have been associated with cancer, suggesting that the overall function of PP2A may be disturbed during tumorigenesis. Decreased Aα expression has been observed in brain tumors and primary human gliomas (Colella et al., 2001) and MCF-7 human breast cancer cells (Suzuki and Takahashi, 2003). Meanwhile, many of the established cancer cell lines do not show reduction in the expression level of Aα isoform, suggesting that the

Aα isoform may be an important selection factor for cell immortalization. On the other hand, expression of the Aβ isoform was detected at lower levels in approximately 50% of 34 tumor cell lines tested in one study (Zhou et al., 2003). In addition to the A subunit, suppression of the C subunit, particularly the Cα isoform, has been observed in androgen receptor signaling independent prostate cancer cells (Bhardwaj et al., 2011). And suppressed protein expression of the B56γ subunit has been reported in transformed melanomas compared to melanocytic naevi

(Deichmann et al., 2001). Significance of such reduced expression of individual PP2A subunits and whether it contributes to the loss of tumor suppressive activity of PP2A in cells requires further investigation.

Given the importance of the B subunit in regulating PP2A substrate specificity and enzyme activity, it is generally perceived that a subset of B subunit is responsible for the PP2A tumor suppressive functions by directing PP2A activity towards proteins involved in tumorigenesis. In particular, members of the B56 family have been implicated in governing the tumor suppressive functions of PP2A (Mumby, 2007; Westermarck and Hahn, 2008; Eichhorn et al., 2009; Virshup and Shenolikar, 2009). B56α-PP2A has been shown to inactivate the oncogene c-Myc by dephosphorylating Ser62 (Arnold and Sears, 2006). B56β-PP2A negatively regulates

Pim1 kinase which cooperates with c-Myc in lymphomagenesis, suggesting that the regulation of c-Myc and Pim1 is a coordinated act of B56α- and B56β-PP2A (Ma et al., 2007). B56γ-PP2A inhibits cell proliferation and anchorage-independent growth by dephosphorylating tumor suppressor protein p53 and is required for SVST mediated cell transformation (Chen et al., 2004;

Li et al., 2007; Lee et al., 2010). In addition, B56δ-PP2A has been demonstrated to dephosphorylate and inactivate Cdc25 phosphatase to block mitotic cell cycle progression

(Margolis et al., 2006; Forester et al., 2007). These studies suggest that B56-PP2A exerts its tumor suppressive function by targeting particular substrates and regulating their cellular function.

To date, there are several lines of evidence suggesting that B56-PP2A is involved in the regulation of p53 tumor suppressor pathway. p53 is a major tumor suppressor protein that controls a variety of cellular processes including DNA repair, cell senescence, apoptosis, cell cycle arrest, and metabolism. In response to cellular stresses, p53 is activated as a transcription factor through a series of posttranslational modifications including phosphorylation, and regulates a vast number of genes (Bode and Dong, 2004; Menendez et al., 2009; Vousden and Prives,

2009). Upon DNA damage, B56γ-PP2A directly dephosphorylates p53 at Thr55, leading to p53 stabilization and activation. This results in the induction of CDK inhibitor p21 and G1 cell cycle arrest (Li et al., 2007). PP2A indirectly affects p53 function through regulating other proteins

involved in p53 signaling pathway as well. A recent study suggests that B56-PP2A dephosphorylates the antiapoptotic regulator Bcl2 which is a transcriptional target of p53 and modulates apoptosis (Ruvolo et al., 2002; Lin et al., 2006). In addition, B56-PP2A has been shown to form a complex with Mdm2, an E3 ubiquitin ligase of p53, and cyclin G, a transcriptional target of p53 (Okamoto et al., 1996, 2002). This interaction promotes dephosphorylation of Mdm2 at Ser166 and Thr216. The interpretation of this, however, is complicated since phosphorylation on Ser166 and Thr216 have opposing effects on Mdm2 activity; phosphorylation on Ser166 stimulates p53 ubiquitination while phosphorylation on

Thr216 weakens Mdm2 association with p53 (Meek and Knippschild, 2004). Cellular function of cyclin G is ambiguous and conflicting positive and negative effects on cell growth have been previously found (Kimura et al., 2001; Zhao et al., 2003b). As such, it is possible that the interaction of B56-PP2A and Mdm2 mediated by cyclin G results in different effects on p53 under varying circumstances.

It is widely accepted that tumorigenesis is a multistep process, the progression of which depends on a sequential accumulation of mutations in the genes that regulate cell growth and differentiation, apoptosis, and DNA repair. Loss of tumor suppressive activities is a critical step for cancer progression and mutations are commonly identified in tumor suppressor genes. It is well known that mutations in the Aα and Aβ genes are associated with cancer. In one study, Aβ mutations were identified in 15% of primary lung tumors, 15% of colorectal carcinomas, 13% of breast cancers, and 6% of lung tumor-derived cell lines (Wang et al., 1998). Most of these are missense mutations and the A subunit with these mutations display decreased binding with either the C subunit or various B subunits, leading to altered PP2A activity (Esplin et al., 2006; Sablina et al., 2010). Although occurring at lower frequency, Aα mutations have been also reported in human melanoma, breast, lung, and ovarian cancers (Calin et al., 2000; Ruediger et al., 2001b;

Jones et al., 2010). Four Aα mutations have been characterized, and two of them disrupt the C subunit binding and the other two disrupt the B56 subunit and at lesser extent B55 subunit binding (Ruediger et al., 2001b), suggesting the particular significance of B56 subunit in the tumor suppressor function of PP2A.

Even though there is accumulating evidence suggesting that B56-PP2A is involved in the cancer-associated pathways, mutations in the B56 subunit are less well known. For the first time, this study presents the loss of function B56γ mutations in cancer. Mechanism of mutations inhibiting the tumor suppressor ability of B56γ-PP2A is discussed in the following chapters. In

Chapter 2, mechanisms behind the loss of p53 interaction and consequential loss of p53- dependent function associated with mutations F395C and A383G are described. In Chapter 3, characterization of the C39R mutation and the importance of HEAT repeat 1 in B56γ function and PP2A holoenzyme assembly is discussed. In Chapter 4, results summarizing the panel of tumor-derived B56γ mutations and the model of inactivation mechanisms of B56γ-PP2A function are presented. Lastly, proteomic analysis of B56γ protein complex in an attempt to elucidate the cause of loss of p53-independent function in S220N mutant is described in Chapter 5. Taken together, this study illustrates the complex mechanisms underlying the B56γ tumor suppressive activity and further strengthens the link between B56γ-PP2A function and tumor progression.

References

Abraham, D., Podar, K., Pacher, M., Kubicek, M., Welzel, N., Hemmings, B.A., Dilworth, S.M., Mischak, H., Kolch, W., and Baccarini, M. (2000). Raf-1-associated protein phosphatase 2A as a positive regulator of kinase activation. J. Biol. Chem. 275, 22300–22304.

Adams, D.G., Coffee, R.L., Zhang, H., Pelech, S., Strack, S., and Wadzinski, B.E. (2005). Positive regulation of Raf1-MEK1/2-ERK1/2 signaling by protein serine/threonine phosphatase 2A holoenzymes. J. Biol. Chem. 280, 42644–42654.

Ahn, J.-H., McAvoy, T., Rakhilin, S. V, Nishi, A., Greengard, P., and Nairn, A.C. (2007). Protein kinase A activates protein phosphatase 2A by phosphorylation of the B56δ subunit. Proc. Natl. Acad. Sci. U.S.A. 104, 2979–2984.

Akiyama, N., Shima, H., Hatano, Y., Osawa, Y., Sugimura, T., and Nagao, M. (1995). cDNA cloning of BRγ, a novel brain-specific isoform of the B regulatory subunit of type-2A protein phosphatase. Eur J Biochem 230, 766–772.

Arnold, H.K., and Sears, R.C. (2006). Protein Phosphatase 2A Regulatory Subunit B56α Associates with c-Myc and Negatively Regulates c-Myc Accumulation. Mol. Cell. Biol. 26, 2832–2844.

Baek, S., and Seeling, J.M. (2007). Identification of a novel conserved mixed-isoform B56 regulatory subunit and spatiotemporal regulation of protein phosphatase 2A during Xenopus laevis development. BMC Dev. Biol. 7, 139.

Baharians, Z., and Schönthal, A.H. (1998). Autoregulation of protein phosphatase type 2A expression. J. Biol. Chem. 273, 19019–19024.

Batut, J., Schmierer, B., Cao, J., Raftery, L.A., Hill, C.S., and Howell, M. (2008). Two highly related regulatory subunits of PP2A exert opposite effects on TGF-β/Activin/Nodal signalling. Development 135, 2927–2937.

Bode, A.M., and Dong, Z. (2004). Post-translational modification of p53 in tumorigenesis. Nat. Rev. Cancer 4, 793–805.

Bosch, M., Cayla, X., Van Hoof, C., Hemmings, B.A., Ozon, R., Merlevede, W., and Goris, J. (1995). The PR55 and PR65 subunits of protein phosphatase 2A from Xenopus laevis. Molecular cloning and developmental regulation of expression. Eur J Biochem 230, 1037–1045.

Brewster, C.E.P., Glover, H.R., and Dilworth, S.M. (1997). pp60 c-src Binding to Polyomavirus Middle T-Antigen (MT) Requires Residues 185 to 210 of the MT Sequence. J. Virol. 71, 5512– 5520.

Bryant, J.C., Westphal, R.S., and Wadzinski, B.E. (1999). Methylated C-terminal leucine residue of PP2A catalytic subunit is important for binding of regulatory Bα subunit. Biochem. J. 246, 241–246.

Calin, G.A., Di Iasio, M.G., Caprini, E., Vorechovsky, I., Natali, P.G., Sozzi, G., Croce, C.M., Barbanti-Brodano, G., Russo, G., and Negrini, M. (2000). Low frequency of alterations of the α(PPP2R1A) and β(PPP2R1B) isoforms of the subunit A of the serine-threonine phosphatase 2A in human neoplasms. Oncogene 19, 1191–1195.

Campbell, K.S., Auger, K.R., Hemmings, B.A., Roberts, T.M., and Pallas, D.C. (1995). Identification of regions in polyomavirus middle T and small t antigens important for association with protein phosphatase 2A. J. Virol. 69, 3721–3728.

Castets, F., Rakitina, T., Gaillard, S., Moqrich, A., Mattei, M.G., and Monneron, A. (2000). Zinedin, SG2NA, and striatin are calmodulin-binding, WD repeat proteins principally expressed in the brain. J. Biol. Chem. 275, 19970–19977.

Cayla, X., Ballmer-Hofer, K., Merlevede, W., and Goris, J. (1993). Phosphatase 2A associated with polyomavirus small-T or middle-T antigen is an okadaic acid-sensitive tyrosyl phosphatase. Eur J Biochem 214, 281–286.

Chen, H.-G., Han, W.-J., Deng, M., Qin, J., Yuan, D., Liu, J.-P., Xiao, L., Gong, L., Liang, S., Zhang, J., et al. (2009). Transcriptional regulation of PP2A-Aα is mediated by multiple factors including AP-2α, CREB, ETS-1, and SP-1. PLoS ONE 4, e7019.

Chen, J., Martin, B.L., and Brautigan, D.L. (1992). Regulation of protein serine-threonine phosphatase type-2A by tyrosine phosphorylation. Science (80- ) 257, 1261–1264.

Chen, W., Possemato, R., Campbell, K.T., Plattner, C.A., Pallas, D.C., and Hahn, W.C. (2004). Identification of specific PP2A complexes involved in human cell transformation. Cancer Cell 5, 127–136.

Cho, U.S., and Xu, W. (2007). Crystal structure of a protein phosphatase 2A heterotrimeric holoenzyme. Nature 445, 53–57.

Christen, V., Treves, S., Duong, F.H.T., and Heim, M.H. (2007). Activation of endoplasmic reticulum stress response by hepatitis viruses up-regulates protein phosphatase 2A. Hepatology 46, 558–565.

Colella, S., Ohgaki, H., Ruediger, R., Yang, F., Nakamura, M., Fujisawa, H., Kleihues, P., and Walter, G. (2001). Reduced expression of the Aα subunit of protein phosphatase 2A in human gliomas in the absence of mutations in the Aα and Aβ subunit genes. Int. J. Cancer 93, 798–804.

Creyghton, M.P., Roël, G., Eichhorn, P.J.A., Hijmans, E.M., Maurer, I., Destrée, O., and Bernards, R. (2005). PR72 , a novel regulator of Wnt signaling required for Naked cuticle function. Genes Dev. 2, 376–386.

Csortos, C., Zolnierowicz, S., Bako, E., Durbin, S.D., and Depaoli-Roach, A.A. (1996). High Complexity in the Expression of the B Subunit of Protein Phosphatase 2A. J. Biol. Chem. 271, 2578–2588.

Deichmann, M., Polychronidis, M., and Wacker, J. (2001). The protein phosphatase 2A subunit Bγ gene is identified to be differentially expressed in malignant melanomas by subtractive suppression hybridization. Melanoma Res. 11, 577–585.

Eichhorn, P.J.A., Creyghton, M.P., and Bernards, R. (2009). Protein phosphatase 2A regulatory subunits and cancer. Biochim. Biophys. Acta 1795, 1–15.

Eichhorn, P.J.A., Creyghton, M.P., Wilhelmsen, K., Van Dam, H., and Bernards, R. (2007). A RNA interference screen identifies the protein phosphatase 2A subunit PR55γ as a stress-sensitive inhibitor of c-SRC. PLoS Genet. 3, e218.

Esplin, E.D., Ramos, P., Martinez, B., Tomlinson, G.E., Mumby, M.C., and Evans, G.A. (2006). The Glycine 90 to Aspartate Alteration in the Aβ Subunit of PP2A (PPP2R1B) Associates with Breast Cancer and Causes a Deficit in Protein Function. Genes Chromosomes Cancer 45, 182– 190.

Forester, C.M., Maddox, J., Louis, J. V, Goris, J., and Virshup, D.M. (2007). Control of mitotic exit by PP2A regulation of Cdc25C and Cdk1. Proc. Natl. Acad. Sci. U.S.A. 104, 19867–19872.

Gallego, M., and Virshup, D.M. (2005). Protein serine/threonine phosphatases: life, death, and sleeping. Curr. Opin. Cell Biol. 17, 197–202.

Gentry, M.S., Li, Y., Wei, H., Syed, F.F., Patel, S.H., Hallberg, R.L., and Pallas, D.C. (2005). A Novel Assay for Protein Phosphatase 2A (PP2A) Complexes In Vivo Reveals Differential Effects of Covalent Modifications on Different Saccharomyces cerevisiae PP2A Heterotrimers. Eukaryotic Cell 4, 1029–1040.

Gillet, R., Cavard, C., Grimber, G., Briand, P., and Joulin, V. (2001). Hepatic expression of SV40 small-T antigen blocks the in vivo CD95-mediated apoptosis. Biochem. Biophys. Res. Commun. 284, 369–376.

Götz, J., Probst, A., Ehler, E., Hemmings, B., and Kues, W. (1998). Delayed embryonic lethality in mice lacking protein phosphatase 2A catalytic subunit Cα. Proc. Natl. Acad. Sci. U.S.A. 95, 12370–12375.

Groves, M.R., Hanlon, N., Turowski, P., Hemmings, B.A., and Barford, D. (1999). The structure of the protein phosphatase 2A PR65/A subunit reveals the conformation of its 15 tandemly repeated HEAT motifs. Cell 96, 99–110.

Hakki, M., and Geballe, A.P. (2008). Cellular serine/threonine phosphatase activity during human cytomegalovirus infection. Virology 380, 255–263.

Hatano, Y., Shima, H., Haneji, T., Miura, A.B., Sugimura, T., and Nagao, M. (1993). Expression of PP2A B regulatory subunit β isotype in rat testis. FEBS Lett. 324, 71–75.

Hemmings, B.A., Adams-Pearson, C., Maurer, F., Müller, P., Goris, J., Merlevede, W., Hofsteenge, J., and Stone, S.. (1990). α- and β-forms of the 65-kDa subunit of protein phosphatase 2A have a similar 39 amino acid repeating structure. Biochemistry 29, 3166–3173.

Hendrix, P., Mayer-Jaekel, R.E., Cron, P., Goris, J., Hofsteenge, J., Merlevede, W., and Hemmings, B.A. (1993a). Structure and Expression of a 72-kDa Regulatory Subunit of Protein Phosphatase 2A. J. Biochem. 268, 15267–15276.

Hendrix, P., Turowski, P., Mayer-Jaekel, R.E., Goris, J., Hofsteenge, J., Merlevede, W., and Hemmings, B.A. (1993b). Analysis of subunit isoforms in protein phosphatase 2A holoenzymes from rabbit and Xenopus. J. Biol. Chem. 268, 7330–7337.

Hombauer, H., Weismann, D., Mudrak, I., Stanzel, C., Fellner, T., Lackner, D.H., and Ogris, E. (2007). Generation of active protein phosphatase 2A is coupled to holoenzyme assembly. PLoS Biol. 5, e155.

Ikehara, T., Ikehara, S., Imamura, S., Shinjo, F., and Yasumoto, T. (2007). Methylation of the C- terminal leucine residue of the PP2A catalytic subunit is unnecessary for the catalytic activity and the binding of regulatory subunit (PR55/B). Biochem. Biophys. Res. Commun. 354, 1052–1057.

Ito, A., Kataoka, T.R., Watanabe, M., Nishiyama, K., Mazaki, Y., Sabe, H., Kitamura, Y., and Nojima, H. (2000). A truncated isoform of the PP2A B56 subunit promotes cell motility through paxillin phosphorylation. EMBO J. 19, 562–571.

Janssens, V., Longin, S., and Goris, J. (2008). PP2A holoenzyme assembly: in cauda venenum (the sting is in the tail). Trends Biochem. Sci. 33, 113–121.

Jones, S., Wang, T.-L., Shih, I.-M., Mao, T.-L., Nakayama, K., Roden, R., Glas, R., Slamon, D., Diaz, L.A., Vogelstein, B., et al. (2010). Frequent mutations of chromatin remodeling gene ARID1A in ovarian clear cell carcinoma. Science (80- ) 330, 228–231.

Jordens, J., Janssens, V., Longin, S., Stevens, I., Martens, E., Bultynck, G., Engelborghs, Y., Lescrinier, E., Waelkens, E., Goris, J., et al. (2006). The protein phosphatase 2A phosphatase activator is a novel peptidyl-prolyl cis/trans-isomerase. J. Biol. Chem. 281, 6349–6357.

Junttila, M.R., Li, S.-P., and Westermarck, J. (2008). Phosphatase-mediated crosstalk between MAPK signaling pathways in the regulation of cell survival. FASEB Journal 22, 954–965.

Kalev, P., Simicek, M., Vazquez, I., Munck, S., Chen, L., Soin, T., Danda, N., Chen, W., and Sablina, A. (2012). Loss of PPP2R2A inhibits homologous recombination DNA repair and predicts tumor sensitivity to PARP inhibition. Cancer Res. 72, 6414–6424.

Kamibayashi, C., Estes, R., Lickteig, L.R., Yang, S.-I., Craft, C., and Mumby, M.C. (1994). Comparison of Heterotrimeric Protein Phosphatase 2A Containing Different B Subunits. J. Biol. Chem. 269, 20139–20148.

Van Kanegan, M.J., and Strack, S. (2009). The protein phosphatase 2A regulatory subunits B’β and B'δ mediate sustained TrkA neurotrophin receptor autophosphorylation and neuronal differentiation. Mol. Cell. Biol. 29, 662–674.

Kantorow, M., Kays, T., Horwitz, J., Huang, Q., Sun, J., Piatigorsky, J., and Carper, D. (1998). Differential display detects altered gene expression between cataractous and normal human lenses. Invest. Ophthalmol. Vis. Sci. 39, 2344–2354.

Khew-Goodall, Y., and Hemmings, B.A. (1988). Tissue-specific expression of mRNAs encoding alpha- and beta-catalytic subunits of protein phosphatase 2A. FEBS Lett. 238, 265–268.

Kimura, S.H., Ikawa, M., Ito, A., Okabe, M., and Nojima, H. (2001). Cyclin G1 is involved in G2/M arrest in response to DNA damage and in growth control after damage recovery. Oncogene 20, 3290–3300.

Kono, Y., Maeda, K., Kuwahara, K., Yamamoto, H., Miyamoto, E., Yonezawa, K., Takagi, K., and Sakaguchi, N. (2002). MCM3-binding GANP DNA-primase is associated with a novel phosphatase component G5PR. Genes Cells 7, 821–834.

Kuo, Y.-C., Huang, K.-Y., Yang, C.-H., Yang, Y.-S., Lee, W.-Y., and Chiang, C.-W. (2008). Regulation of phosphorylation of Thr-308 of Akt, cell proliferation, and survival by the B55α regulatory subunit targeting of the protein phosphatase 2A holoenzyme to Akt. J. Biol. Chem. 283, 1882–1892.

Kurimchak, A., and Graña, X. (2012). PP2A holoenzymes negatively and positively regulate cell cycle progression by dephosphorylating pocket proteins and multiple CDK substrates. Gene 499, 1–7.

Lee, J.A., and Pallas, D.C. (2007). Leucine carboxyl methyltransferase-1 is necessary for normal progression through mitosis in mammalian cells. J. Biol. Chem. 282, 30974–30984.

Lee, T.-Y., Lai, T.-Y., Lin, S.-C., Wu, C.-W., Ni, I.-F., Yang, Y.-S., Hung, L.-Y., Law, B.K., and Chiang, C.-W. (2010). The B56γ3 regulatory subunit of protein phosphatase 2A (PP2A) regulates S phase-specific nuclear accumulation of PP2A and the G1 to S transition. J. Biol. Chem. 285, 21567–21580.

Letourneux, C., Rocher, G., and Porteu, F. (2006). B56-containing PP2A dephosphorylate ERK and their activity is controlled by the early gene IEX-1 and ERK. EMBO J. 25, 727–738.

Leulliot, N., Vicentini, G., Jordens, J., Quevillon-Cheruel, S., Schiltz, M., Barford, D., Van Tilbeurgh, H., and Goris, J. (2006). Crystal structure of the PP2A phosphatase activator: implications for its PP2A-specific PPIase activity. Mol. Cell 23, 413–424.

Li, H.H., Cai, X., Shouse, G.P., Piluso, L.G., and Liu, X. (2007). A specific PP2A regulatory subunit, B56γ, mediates DNA damage-induced dephosphorylation of p53 at Thr55. EMBO J. 26, 402–411.

Li, S., Wang, L., Berman, M. a, Zhang, Y., and Dorf, M.E. (2006). RNAi screen in mouse astrocytes identifies phosphatases that regulate NF-κB signaling. Mol. Cell 24, 497–509.

Li, X., and Virshup, D.M. (2002). Two conserved domains in regulatory B subunits mediate binding to the A subunit of protein phosphatase 2A. Eur J Biochem 269, 546–552.

Li, X., Yost, H.J., Virshup, D.M., and Seeling, J.M. (2001). Protein phosphatase 2A and its B56 regulatory subunit inhibit Wnt signaling in Xenopus. EMBO J. 20, 4122–4131.

Lim, J., and Lu, K.P. (2005). Pinning down phosphorylated tau and tauopathies. Biochim. Biophys. Acta 1739, 311–322.

Lin, S.S., Bassik, M.C., Suh, H., Nishino, M., Arroyo, J.D., Hahn, W.C., Korsmeyer, S.J., and Roberts, T.M. (2006). PP2A regulates BCL-2 phosphorylation and proteasome-mediated degradation at the endoplasmic reticulum. J. Biol. Chem. 281, 23003–23012.

Liu, J., Ji, W., Sun, S., Zhang, L., Chen, H.-G., Mao, Y., Liu, L., Zhang, X., Gong, L., Deng, M., et al. (2012). The PP2A-Aβ Gene is Regulated by Multiple Transcriptional Factors Including Ets- 1, SP1/SP3, and RXRα/β. Curr. Mol. Med. 12, 982–994.

Longin, S., Jordens, J., Martens, E., Stevens, I., Janssens, V., Rondelez, E., De Baere, I., Derua, R., Waelkens, E., Goris, J., et al. (2004). An inactive protein phosphatase 2A population is associated with methylesterase and can be re-activated by the phosphotyrosyl phosphatase activator. Biochem. J. 380, 111–119.

Longin, S., Zwaenepoel, K., Louis, J. V, Dilworth, S., Goris, J., and Janssens, V. (2007). Selection of protein phosphatase 2A regulatory subunits is mediated by the C terminus of the catalytic Subunit. J. Biol. Chem. 282, 26971–26980.

Louis, J. V, Martens, E., Borghgraef, P., Lambrecht, C., Sents, W., Longin, S., Zwaenepoel, K., Pijnenborg, R., Landrieu, I., Lippens, G., et al. (2011). Mice lacking phosphatase PP2A subunit PR61/B’δ (Ppp2r5d) develop spatially restricted tauopathy by deregulation of CDK5 and GSK3β. Proc. Natl. Acad. Sci. U.S.A. 108, 6957–6962.

Ma, J., Arnold, H.K., Lilly, M.B., Sears, R.C., and Kraft, A.S. (2007). Negative regulation of Pim-1 protein kinase levels by the B56β subunit of PP2A. Oncogene 26, 5145–5153.

Margolis, S.S., Perry, J.A., Forester, C.M., Nutt, L.K., Guo, Y., Jardim, M.J., Thomenius, M.J., Freel, C.D., Darbandi, R., Ahn, J., et al. (2006). Role for the PP2A/ B56δ Phosphatase in Regulating 14-3-3 Release from Cdc25 to Control Mitosis. Cell 3, 759–773.

Martens, E., Stevens, I., Janssens, V., Vermeesch, J., Götz, J., Goris, J., and Van Hoof, C. (2004). Genomic organisation, chromosomal localisation tissue distribution and developmental regulation

of the PR61/B’ regulatory subunits of protein phosphatase 2A in mice. J. Mol. Biol. 336, 971– 986.

Mayer, R.E., Hendrix, P., Cron, P., Matthies, R., Stone, S.R., Goris, J., Merlevede, W., Hofsteenge, J., and Hemmings, B.A. (1991). Structure of the 55-kDa regulatory subunit of protein phosphatase 2A: evidence for a neuronal-specific isoform. Biochemistry 30, 3589–3597.

McCright, B., Rivers, a M., Audlin, S., and Virshup, D.M. (1996). The B56 family of protein phosphatase 2A (PP2A) regulatory subunits encodes differentiation-induced phosphoproteins that target PP2A to both nucleus and cytoplasm. J. Biol. Chem. 271, 22081–22089.

McCright, B., and Virshup, D.M. (1995). Identification of a new family of protein phosphatase 2A regulatory subunits. J. Biol. Chem. 270, 26123–26128.

Meek, D.W., and Knippschild, U. (2004). Posttranslational Modification of MDM2. Mol. Cancer Res. 1, 1017–1026.

Menendez, D., Inga, A., and Resnick, M.A. (2009). The expanding universe of p53 targets. Nat. Rev. Cancer 9, 724–737.

Merrick, S.E., Demoise, D.C., and Lee, V.M. (1996). Site-specific dephosphorylation of tau protein at Ser202/Thr205 in response to microtubule depolymerization in cultured human neurons involves protein phosphatase 2A. J. Biol. Chem. 271, 5589–5594.

Millward, T.A., Zolnierowicz, S., and Hemmings, B.A. (1999). Regulation of protein kinase cascades by protein phosphatase 2A. Trends Biochem. Sci. 24, 186–191.

Moreno, C.S., Lane, W.S., and Pallas, D.C. (2001). A mammalian homolog of yeast MOB1 is both a member and a putative substrate of striatin family-protein phosphatase 2A complexes. J. Biol. Chem. 276, 24253–24260.

Moreno, C.S., Ramachandran, S., Ashby, D.G., Laycock, N., Plattner, C.A., Chen, W., Hahn, W.C., and Pallas, D.C. (2004). Signaling and transcriptional changes critical for transformation of human cells by simian virus 40 small tumor antigen or protein phosphatase 2A B56γ knockdown. Cancer Res. 64, 6978–6988.

Moule, M.G., Collins, C.H., McCormick, F., and Fried, M. (2004). Role for PP2A in ARF signaling to p53. Proc. Natl. Acad. Sci. U.S.A. 101, 14063–14066.

Mumby, M.C. (2007). PP2A: unveiling a reluctant tumor suppressor. Cell 130, 21–24.

Muneer, S., Ramalingam, V., Wyatt, R., Schultz, R.A., Minna, J.D., and Kamibayashi, C. (2002). Genomic organization and mapping of the gene encoding the PP2A B56γ regulatory subunit. Genomics 79, 344–348.

Nagase, T., Murakami, T., Nozaki, H., Inoue, R., Nishito, Y., Tanabe, O., Usui, H., and Takeda, M. (1997). Tissue and subcellular distributions, and characterization of rat brain protein phosphatase 2A containing a 72-kDa δ/B" subunit. J. Biochem. 122, 178–187.

Nunbhakdi-Craig, V., Machleidt, T., Ogris, E., Bellotto, D., White, C.L., and Sontag, E. (2002). Protein phosphatase 2A associates with and regulates atypical PKC and the epithelial tight junction complex. J. Cell Biol. 158, 967–978.

Ogris, E., Du, X., Nelson, K.C., Mak, E.K., Yu, X.X., Lane, W.S., and Pallas, D.C. (1999a). A protein phosphatase methylesterase (PME-1) is one of several novel proteins stably associating with two inactive mutants of protein phosphatase 2A. J. Biol. Chem. 274, 14382–14391.

Ogris, E., Gibson, D.M., and Pallas, D.C. (1997). Protein phosphatase 2A subunit assembly: the catalytic subunit carboxy terminus is important for binding cellular B subunit but not polyomavirus middle tumor antigen. Oncogene 15, 911–917.

Ogris, E., Mudrak, I., Mak, E., Gibson, D.M., and Pallas, D.C. (1999b). Catalytically Inactive Protein Phosphatase 2A Can Bind to Polyomavirus Middle Tumor Antigen and Support Complex Formation with pp60 c-src. J. Virol. 73, 7390.

Okamoto, K., Kamibayashi, C., Serrano, M., Prives, C., Mumby, M.C., and Beach, D. (1996). p53-dependent association between cyclin G and the B’ subunit of protein phosphatase 2A. Mol. Cell. Biol. 16, 6593.

Okamoto, K., Li, H., Jensen, M.R., Zhang, T., Taya, Y., Thorgeirsson, S.S., and Prives, C. (2002). Cyclin G recruits PP2A to dephosphorylate Mdm2. Mol. Cell 9, 761–771.

Olsen, J. V, Blagoev, B., Gnad, F., Macek, B., Kumar, C., Mortensen, P., and Mann, M. (2006). Global, in vivo, and site-specific phosphorylation dynamics in signaling networks. Cell 127, 635– 648.

Ortega-Lázaro, J., and Del Mazo, J. (2003). Expression of the B56δ subunit of protein phosphatase 2A and Mea1 in mouse spermatogenesis. Identification of a new B56γ subunit (B56γ4) specifically expressed in testis. Cytogenet. Genome Res. 103, 345–351.

Ory, S., Zhou, M., Conrads, T.P., Veenstra, T.D., and Morrison, D.K. (2003). Protein Phosphatase 2A Positively Regulates Ras Signaling by Dephosphorylating KSR1 and Raf-1 on Critical 14-3-3 Binding Sites. Curr. Biol. 13, 1356–1364.

Palczewski, K., Farber, D.B., and Hargrave, P.A. (1990). Elevated Level of PP2A Activity in Retinas of rd Mice. Exp. Eye Res. 53, 101–105.

Pallas, D.C., Shahrik, L.K., Martin, B.L., Jaspers, S., Miller, T.B., Brautigan, D.L., and Roberts, T.M. (1990). Polyoma small and middle T antigens and SV40 small t antigen form stable complexes with protein phosphatase 2A. Cell 60, 167–176.

Paroni, G., Cernotta, N., Russo, C. Dello, Gallinari, P., Pallaoro, M., Foti, C., Talamo, F., Orsatti, L., Brancolini, C., Udine, U., et al. (2008). PP2A Regulates HDAC4 Nuclear Import. Mol. Biol. Cell 19, 655–667.

Pitman, M.R., Barr, R.K., Gliddon, B.L., Magarey, A.M., Moretti, P.A.B., and Pitson, S.M. (2011). A critical role for the protein phosphatase 2A B’α regulatory subunit in dephosphorylation of sphingosine kinase 1. Int. J. Biochem. Cell Biol. 43, 342–347.

Ricotta, D., Hansen, J., Preiss, C., Teichert, D., and Höning, S. (2008). Characterization of a protein phosphatase 2A holoenzyme that dephosphorylates the clathrin adaptors AP-1 and AP-2. J. Biol. Chem. 283, 5510–5517.

Rodgers, J.T., Vogel, R.O., and Puigserver, P. (2011). Clk2 and B56β mediate insulin-regulated assembly of the PP2A phosphatase holoenzyme complex on Akt. Mol. Cell 41, 471–479.

Rodriguez-Viciana, P., Collins, C., and Fried, M. (2006). Polyoma and SV40 proteins differentially regulate PP2A to activate distinct cellular signaling pathways involved in growth control. Proc. Natl. Acad. Sci. U.S.A. 103, 19290–19295.

Ruediger, R., Fields, K., and Walter, G. (1999). Binding specificity of protein phosphatase 2A core enzyme for regulatory B subunits and T antigens. J. Virol. 73, 839–842.

Ruediger, R., Hentz, M., Fait, J., Mumby, M., and Walter, G. (1994). Molecular model of the A subunit of protein phosphatase 2A: interaction with other subunits and tumor antigens. J. Virol. 68, 123–129.

Ruediger, R., Pham, H.T., and Walter, G. (2001). Disruption of protein phosphatase 2A subunit interaction in human cancers with mutations in the Aα subunit gene. Oncogene 20, 10–15.

Ruediger, R., Roeckel, D., Fait, J., and Bergqvist, A. (1992). Identification of binding sites on the regulatory A subunit of protein phosphatase 2A for the catalytic C subunit and for tumor antigens of Simian Virus 40 and Polyomavirus. Mol. Cell. Biol. 12, 4872–4882.

Ruvolo, P.P., Clark, W., Mumby, M.C., Gao, F., and May, W.S. (2002). A functional role for the B56α-subunit of protein phosphatase 2A in ceramide-mediated regulation of Bcl2 phosphorylation status and function. J. Biol. Chem. 277, 22847–22852.

Ruvolo, P.P., Deng, X., and May, W.S. (2001). Phosphorylation of Bcl2 and regulation of apoptosis. Leukemia 15, 515–522.

Ruvolo, V.R., Kurinna, S.M., Karanjeet, K.B., Schuster, T.F., Martelli, A.M., McCubrey, J.A., and Ruvolo, P.P. (2008). PKR regulates B56(α)-mediated BCL2 phosphatase activity in acute lymphoblastic leukemia-derived REH cells. J. Biol. Chem. 283, 35474–35485.

Sablina, A.A., Hector, M., Colpaert, N., and Hahn, W.C. (2010). Identification of PP2A complexes and pathways involved in cell transformation. Cancer Res. 70, 10474–10484.

Saraf, A., Virshup, D.M., and Strack, S. (2007). Differential expression of the B’β regulatory subunit of protein phosphatase 2A modulates tyrosine hydroxylase phosphorylation and catecholamine synthesis. J. Biol. Chem. 282, 573–580.

Schönthal, A.H. (2001). Role of serine/threonine protein phosphatase 2A in cancer. Cancer Lett. 170, 1–13.

Seeling, J.M., Miller, J.R., Gil, R., Moon, R.T., White, R., and Virshup, D.M. (1999). Regulation of β-Catenin Signaling by the B56 Subunit of Protein Phosphatase 2A. Science (80- ) 283, 2089– 2091.

Shi, Y. (2009). Serine / Threonine Phosphatases : Mechanism through Structure. Cell 139, 468– 484.

Shouse, G.P., Nobumori, Y., Panowicz, M.J., and Liu, X. (2011). ATM-mediated phosphorylation activates the tumor-suppressive function of B56γ – PP2A. Oncogene 30, 3755– 3765.

Shtrichman, R., Sharf, R., Barr, H., Dobner, T., and Kleinberger, T. (1999). Induction of apoptosis by adenovirus E4orf4 protein is specific to transformed cells and requires an interaction with protein phosphatase 2A. Proc. Natl. Acad. Sci. U.S.A. 96, 10080–10085.

Sieburth, D.S., Sundaram, M., Howard, R.M., and Han, M. (1999). A PP2A regulatory subunit positively regulates Ras-mediated signaling during Caenorhabditis elegans vulval induction. Genes Dev. 13, 2562–2569.

Silverstein, A.M., Barrow, C.A., Davis, A.J., and Mumby, M.C. (2002). Actions of PP2A on the MAP kinase pathway and apoptosis are mediated by distinct regulatory subunits. Proc. Natl. Acad. Sci. U.S.A. 99, 4221–4226.

Sonoda, Y., Kasahara, T., Yamaguchi, Y., Kuno, K., Matsushima, K., and Mukaida, N. (1997). Stimulation of Interleukin-8 Production by Okadaic Acid and Vanadate in a Human Promyelocyte Cell Line , an HL-60 Subline. J. Biol. Chem. 272, 15366–15372.

Sontag, E. (2001). Protein phosphatase 2A: the Trojan Horse of cellular signaling. Cell. Signal. 13, 7–16.

Sontag, E., Nunbhakdi-Craig, V., Lee, G., Bloom, G.S., and Mumby, M.C. (1996). Regulation of the phosphorylation state and microtubule-binding activity of Tau by protein phosphatase 2A. Neuron 17, 1201–1207.

Sontag, E., Nunbhakdi-Craig, V., Lee, G., Brandt, R., Kamibayashi, C., Kuret, J., White, C.L., Mumby, M.C., and Bloom, G.S. (1999). Molecular interactions among protein phosphatase 2A, tau, and microtubules. Implications for the regulation of tau phosphorylation and the development of tauopathies. J. Biol. Chem. 274, 25490–25498.

Sontag, E., Sontag, J.M., and Garcia, A. (1997). Protein phosphatase 2A is a critical regulator of protein kinase C zeta signaling targeted by SV40 small t to promote cell growth and NF-κB activation. EMBO J. 16, 5662–5671.

Stevens, I., Janssens, V., Martens, E., Dilworth, S., Goris, J., and Van Hoof, C. (2003). Identification and characterization of B"-subunits of protein phosphatase 2A in Xenopus laevis oocytes and adult tissues. Evidence for an independent N-terminal splice variant of PR130 and an extended human PR48 protein. Eur J Biochem 270, 376–387.

Stipanovich, A., Valjent, E., Matamales, M., Ahn, J.-H., Maroteaux, M., Bertran-Gonzalez-, J., Enslen, H., Corbillé, A.-G., Filhol, O., Angus, C., et al. (2009). A phosphatase cascade by which natural rewards and drugs of abuse regulate nucleosomal response in the mouse. Nature 453, 879–884.

Strack, S., Chang, D., Zaucha, J.A., Colbran, R.J., and Wadzinski, B.E. (1999). Cloning and characterization of Bδ, a novel regulatory subunit of protein phosphatase 2A. FEBS Lett. 460, 462–466.

Strack, S., Zaucha, J.A., Ebner, F.F., Colbran, R.J., and Wadzinski, B.E. (1998). Brain protein phosphatase 2A: developmental regulation and distinct cellular and subcellular localization by B subunits. J. Comp. Neurol. 392, 515–527.

Suganuma, M., Fujiki, H., Furuya-suguri, H., Yoshizawa, S., Yasumoto, S., and Kato, Y. (1990). Calyculin A, an Inhibitor of Protein Phosphatases, a Potent Tumor Promoter on CD-1 Mouse Skin. Cancer Res. 50, 3521–3525.

Suganuma, M., Fujiki, H., Suguri, H., Yoshizawa, S., Hirota, M., Nakayasu, M., Ojika, M., Wakamatsu, K., Yamada, K., and Sugimura, T. (1988). Okadaic acid: an additional non-phorbol- 12-tetradecanoate-13-acetate-type tumor promoter. Proc. Natl. Acad. Sci. U.S.A. 85, 1768–1771.

Sunahori, K., Juang, Y.-T., and Tsokos, G.C. (2010). Methylation Status of CpG Islands Flanking a CRE Motif on the Protein Phosphatase 2ACα Promoter Determines CREB Binding and Activity. J Immunol 182, 1500–1508.

Suzuki, K., and Takahashi, K. (2003). Reduced expression of the regulatory A subunit of serine/threonine protein phosphatase 2A in human breast cancer MCF-7 cells. J Oncol 5, 1263– 1268.

Tanabe, O., Nagase, T., Murakami, T., Nozaki, H., Usui, H., Nishito, Y., Hayashi, H., Kagamiyama, H., and Takeda, M. (1996). Molecular cloning of a 74-kDa regulatory subunit (B" or δ) of human protein phosphatase 2A. FEBS Lett. 379, 107–111.

Tehrani, M.A., Mumby, M.C., and Kamibayashi, C. (1996). Identification of a Novel Protein Phosphatase 2A Regulatory Subunit Highly Expressed in Muscle. J. Biol. Chem. 271, 5164–5170.

Tolstykh, T., Lee, J., Vafai, S., and Stock, J.B. (2000). Carboxyl methylation regulates phosphoprotein phosphatase 2A by controlling the association of regulatory B subunits. EMBO J. 19, 5682–5691.

Tonks, N.K. (2006). Protein tyrosine phosphatases: from genes, to function, to disease. Nat. Rev. Mol. Cell Biol. 7, 833–846.

Turowski, P., Fernandez, A., Favre, B., Lamb, N.J., and Hemmings, B.A. (1995). Differential methylation and altered conformation of cytoplasmic and nuclear forms of protein phosphatase 2A during cell cycle progression. J. Cell Biol. 129, 397–410.

Virshup, D.M., and Shenolikar, S. (2009). From promiscuity to precision: protein phosphatases get a makeover. Mol. Cell 33, 537–545.

Vousden, K.H., and Prives, C. (2009). Blinded by the Light: The Growing Complexity of p53. Cell 137, 413–431.

Walter, G., Ruediger, R., Slaughter, C., and Mumby, M.C. (1990). Association of protein phosphatase 2A with polyoma virus medium tumor antigen. Proc. Natl. Acad. Sci. U.S.A. 87, 2521–2525.

Wang, S.S., Esplin, E.D., Li, J.L., Huang, L., Gazdar, A., Minna, J., and Evans, G.A. (1998). Alterations of the PPP2R1B Gene in Human Lung and Colon Cancer. Science (80- ) 282, 284– 287.

Wei, H., Ashby, D.G., Moreno, C.S., Ogris, E., Yeong, F.M., Corbett, A.H., and Pallas, D.C. (2001). Carboxymethylation of the PP2A catalytic subunit in Saccharomyces cerevisiae is required for efficient interaction with the B-type subunits Cdc55p and Rts1p. J. Biol. Chem. 276, 1570–1577.

Westermarck, J., and Hahn, W.C. (2008). Multiple pathways regulated by the tumor suppressor PP2A in transformation. Trends Mol Med 14, 152–160.

Wurzenberger, C., and Gerlich, D.W. (2011). Phosphatases: providing safe passage through mitotic exit. Nat. Rev. Mol. Cell Biol. 12, 469–482.

Xing, Y., Li, Z., Chen, Y., Stock, J.B., Jeffrey, P.D., and Shi, Y. (2008). Structural mechanism of demethylation and inactivation of protein phosphatase 2A. Cell 133, 154–163.

Xing, Y., Xu, Y., Chen, Y., Jeffrey, P.D., Chao, Y., Lin, Z., Li, Z., Strack, S., Stock, J.B., and Shi, Y. (2006). Structure of Protein Phosphatase 2A Core Enzyme Bound to Tumor-Inducing Toxins. Cell 127, 341–353.

Xu, Y., Chen, Y., Zhang, P., Jeffrey, P.D., and Shi, Y. (2008). Structure of a protein phosphatase 2A holoenzyme: insights into B55-mediated Tau dephosphorylation. Mol. Cell 31, 873–885.

Xu, Y., Xing, Y., Chen, Y., Chao, Y., Lin, Z., Fan, E., Yu, J.W., Strack, S., Jeffrey, P.D., and Shi, Y. (2006). Structure of the protein phosphatase 2A holoenzyme. Cell 127, 1239–1251.

Yan, Z., Fedorov, S.A., Mumby, M.C., and Williams, R.S. (2000). PR48, a Novel Regulatory Subunit of Protein Phosphatase 2A, Interacts with Cdc6 and Modulates DNA Replication in Human Cells. Mol. Cell. Biol. 20, 1021–1029.

Yang, S.I., Lickteig, R.L., Estes, R., Rundell, K., Walter, G., and Mumby, M.C. (1991). Control of protein phosphatase 2A by simian virus 40 small-t antigen. Mol. Cell. Biol. 11, 1988–1995.

Yu, X.X., Du, X., Moreno, C.S., Green, R.E., Ogris, E., Feng, Q., Chou, L., McQuoid, M.J., and Pallas, D.C. (2001). Methylation of the protein phosphatase 2A catalytic subunit is essential for association of Bα regulatory subunit but not SG2NA, striatin, or polyomavirus middle tumor antigen. Mol. Biol. Cell 12, 185–199.

Yuan, H., Veldman, T., Rundell, K., and Schlegel, R. (2002). Simian virus 40 small tumor antigen activates AKT and telomerase and induces anchorage-independent growth of human epithelial cells. J. Virol. 76, 10685–10691.

Zhao, J.J., Gjoerup, O. V, Subramanian, R.R., Cheng, Y., Chen, W., Roberts, T.M., and Hahn, W.C. (2003a). Human mammary epithelial cell transformation through the activation of phosphatidylinositol 3-kinase. Cancer Cell 3, 483–495.

Zhao, L., Samuels, T., Winckler, S., Korgaonkar, C., Tompkins, V., Horne, M.C., and Quelle, D.E. (2003b). Cyclin G1 Has Growth Inhibitory Activity Linked to the ARF-Mdm2-p53 and pRb Tumor Suppressor Pathways. Mol. Cancer Res. 1, 195–206.

Zhou, J., Pham, H.T., Ruediger, R., and Walter, G. (2003). Characterization of Aα and Aβ subunit isoforms of PP2A: differences in expression, subunit interaction, and evolution. Biochem. J. 369, 387–398.

Zolnierowicz, S., Csortos, C., Bondor, J., Verin, A., Mumby, M.C., and DePaoli-Roach, A.A. (1994). Diversity in the regulatory B-subunits of protein phosphatase 2A: identification of a novel isoform highly expressed in brain. Biochemistry 33, 11858–11867.

Zwaenepoel, K., Louis, J. V, Goris, J., and Janssens, V. (2008). Diversity in genomic organisation, developmental regulation and distribution of the murine PR72/B" subunits of protein phosphatase 2A. BMC Genomics 9,.

Figure legend

Figure 1.1 Structure of PP2A.

Using PyMOL, the crystal structures of PP2A are prepared for (A) the monomeric A subunit

(adapted from Protein Data Bank, accession code 1B3U), (B) the heterodimeric AC core enzyme

(adapted from Protein Data Bank, accession code 2IE3), (C) the heterotrimeric PP2A holoenzyme involving the B56 subunit (adapted from Protein Data Bank, accession code 2NYM), and (D) the heterotrimeric PP2A holoenzyme involving the B55α subunit (adapted from Protein Data

Bank, accession code 3DW8).

Figures

Figure 1.1 Structure of PP2A.

Tables

B Tissue Subcellular Gene name Alternate names Reference subunit localization localization

PPP2R2A B55α PR55α Ubiquitous Ubiquitous Mayer et al., 1991; Hatano et al., 1993; PPP2R2B B55β PR55β Brain and testes Cytosol Hendrix et al., 1993b; Zolnierowicz B PPP2R2C B55γ PR55γ Brain Cytoskeleton et al., 1994; Akiyama et al., PPP2R2D B55δ PR55δ Widespread Cytosol 1995; Strack et al., 1998, 1999

Heart and PPP2R5A B56α PR61α Cytoplasm skeletal muscle

PPP2R5B B56β PR61β Brain Cytoplasm McCright and Heart and Mainly B56γ1 PR61γ1 Virshup, 1995; skeletal muscle cytoplasmic Csortos et al., 1996; McCright et al., Heart and PPP2R5C B56γ2 PR61γ2 Mainly nuclear 1996; Tanabe et al., skeletal muscle B’ 1996; Tehrani et al., Heart and 1996; Nagase et al., B56γ3 PR61γ3 Mainly nuclear skeletal muscle 1997; Ito et al., 2000; Muneer et al., Nucleus, 2002; Martens et al., cytosol, 2004 PPP2R5D B56δ PR61δ Brain mitochondria, microsome

PPP2R5E B56ε PR61ε Brain Cytoplasm

Heart, brain, Centrosome, B”α1 PR130 lung, muscle, Golgi kidney Hendrix et al., PPP2R3A 1993a; Yan et al., Heart, skeletal 2000; Kono et al., B” B”α2 PR72 Cytosol, nucleus muscle 2002; Stevens et al., 2003; Zwaenepoel et PPP2R3B B”β PR70/48 Placenta Nucleus al., 2008 C14orf10 B”γ G5PR Ubiquitous Cytosol, nucleus

Brain Cytoplasm STRN Striatin PR110 Castets et al., 2000; B”’ Moreno et al., 2001 STRN3 SGN2A PR93 Brain Cytoplasm

Table 1.1 Nomenclature and subcellular localization of PP2A B subunits.

B subunit Interacting protein Phos residue References B55α Akt Thr308 Kuo et al. 2008 Ap-1/Ap-2 Thr156 Ricotta et al. 2008 ATM Ser367/1893/1981 Kalev et al. 2012 HDAC4 Ser298 Paroni et al. 2008 Raf1 Ser259 Adams et al. 2005 Tau Ser202/Thr205 Merrick et al. 1996; Sontag et al. 1996, 1999 B55γ Src Ser12 Eichhorn et al. 2007 B55δ Raf1 Ser259 Adams et al. 2005 B56α c-Myc Ser62 Arnold and Sears 2006 Bcl2 Ser70 Ruvolo et al. 2001 SK1 Ser225 Pitman et al. 2011 B56β Pim1 Ma et al. 2007 Akt Ser473/T308 Rodgers et al. 2011 B56γ p53 Thr55 Li et al. 2007 TRAF2 Thr117 Li et al. 2006 B56δ Cdc25 Thr138 Margolis et al. 2006 PR70 Rb Thr826 Margolis et al. 2006 PR72 DARPP-32 Thr75 Ahn et al. 2007

Table 1.2 Summary of proteins dephosphorylated by PP2A, classified by B subunits.

CHAPTER 2

A B56γ mutation in lung cancer disrupts the p53-dependent tumor suppressor function of protein phosphatase 2A

This research was originally published in Oncogene. G.P. Shouse, Y. Nobumori, and X. Liu. A

B56γ mutation in lung cancer disrupts the p53-dependent tumor-suppressor function of protein phosphatase 2A. Oncogene. 2010; 29:3933-3941. © Nature Publishing Group. I am responsible for generating B56γ mutant plasmid A383G and Figure S2.3B, D, and E.

Abstract

Earlier studies have shown both p53-dependent and -independent tumor suppressive functions of

B56γ-specific protein phosphatase 2A (B56γ-PP2A). In the absence of p53, B56γ-PP2A can inhibit cell proliferation and cell transformation by an unknown mechanism. In the presence of p53, on DNA damage, a complex including B56γ-PP2A and p53 is formed which leads to Thr55 dephosphorylation of p53, induction of the p53 transcriptional target p21, and inhibition of cell proliferation. In spite of its significance in inhibition of cell proliferation, no B56γ mutations have been linked to human cancer to date. In this study, we first differentiate between the p53- dependent and -independent functions of B56γ-PP2A by identifying a domain of the B56γ protein required for interaction with p53. Within this region we identify a B56γ mutation, F395C, in lung cancer that disrupts the B56γ-p53 interaction. More importantly, we show that F395C is unable to promote p53 Thr55 dephosphorylation, transcriptional activation of p21, and the p53-dependent tumor suppressive function of PP2A. This finding provides a mechanistic basis for the p53- dependent and -independent functions of B56γ-PP2A and establishes a critical link between

B56γ-PP2A p53-dependent tumor suppressive function and tumorigenesis.

Introduction

Protein phosphatase 2A (PP2A) is a family of serine/threonine phosphoprotein phosphatases that has been implicated as a potential tumor suppressor. PP2A consists of a heterotrimeric complex composed of a scaffolding (A) subunit, catalytic (C) subunit, and regulatory (B) subunit. The B subunit is postulated to confer substrate specificity to the PP2A holoenzyme (Janssens and Goris, 2001; Schönthal, 2001; Eichhorn et al., 2009; Virshup and

Shenolikar, 2009). The role of PP2A in control of tumor progression is thought to be governed by a small subset of specific B subunits directing PP2A to dephosphorylate and regulate key tumor suppressors or oncogenes (Eichhorn et al., 2009; Virshup and Shenolikar, 2009). Indeed, several members of the B56 family have been described as having a role in directing PP2A potential tumor suppressive activity. It has been shown that B56α-PP2A (PPP2R5A) dephosphorylates the c-Myc oncogene at Ser62 leading to its inactivation (Arnold and Sears, 2008), B56δ-PP2A

(PPP2R5D) dephosphorylates Cdc25 blocking cell cycle progression (Margolis et al., 2006;

Forester et al., 2007), and B56γ-PP2A (PPP2R5C) was shown to be important in SV40 mediated cell transformation (Chen et al., 2004). Interestingly, mutational analyses of PP2A genes have yielded data suggesting PP2A Aα and Aβ mutations are present in some cancers (Wang et al.,

1998; Ruediger et al., 2001a, 2001b; Chen et al., 2005b; Esplin et al., 2006). Importantly, the unifying effect of these mutations is loss of interaction with B56 subunit family members.

Although one mutation in B56ε (PPP2R5E) gene of the B56 family has been reported to associate with increased risk of soft tissue sarcoma (Grochola et al., 2009), no mutations in the B56α, B56δ or B56γ tumor suppressor have been linked to human cancer to date.

p53 is a very important highly studied tumor suppressor protein that functions primarily as a transcription factor (Kruse and Gu, 2009; Vousden and Prives, 2009). In response to DNA damage, p53 is highly post-translationally modified promoting its activation and nuclear

localization, whereby it acts as a transcription factor turning on genes responsible for apoptosis or cell cycle arrest including the CDK inhibitor, p21 (Bode and Dong, 2004). Our laboratory has previously demonstrated that in response to DNA damage, B56γ-PP2A dephosphorylates p53 at

Thr55 and promotes its transcriptional activation. In addition, we were able to describe both a p53-dependent and –independent tumor suppressive function of B56γ-PP2A (Li et al., 2007;

Shouse et al., 2008). These results, along with recent crystallography studies of the B56γ-PP2A complex (Xu et al., 2006; Cho and Xu, 2007), led us to focus our investigations toward identifying the region of B56γ required for interaction with p53 as an attempt to better understand its p53-dependent and –independent functions.

In this study, we show that a small region of B56γ (aa391-401) is critical for interaction with the tumor suppressor protein p53. Within this region we identify a cancer-derived mutation,

F395C which disrupts the B56γ-p53 interaction. Although this mutation has been found from only one human cancer sample thus far, we were able to further show that F395C lacks the ability to promote p53 Thr55 dephosphorylation and transcriptional activation of the p21 gene.

Importantly, we demonstrate that F395C is unable to drive the previously described p53- dependent tumor suppressive functions of PP2A either by inhibiting cell proliferation or by decreasing anchorage-independent growth of tumor-derived cells, while retaining its p53- independent functions. Taken together, these results provide evidence that PP2A-mediated activation of p53 may have an important role in prevention of tumor progression.

Materials and Methods

Identification of cancer-derived mutation

The AceView program, developed at NCBI, provides a strictly cDNA-supported view of the human transcriptome and the genes by summarizing all quality-filtered human cDNA data from

GenBank, dbEST and RefSeq. Using this program, we looked for B56γ mutations present within our identified p53 interaction domain where the annotated sequencing data was taken from tumors, tumor samples, and cancer cell lines. Our specific mutant was identified from a lung tumor sample and the corresponding gene accession number is BE829999.

Cell culture and plasmids

U2OS, HCT116 or H1299 cells were cultured in McCoy’s 5A, DMEM or RPMI 1640 medium supplemented with 10% fetal calf serum. The B56γ3 deletion mutants, ΔN17, ΔN156, ΔC347,

ΔC376, and ΔC401 were generated by PCR. The B56γ3 point mutants, Q392G, C398L,

Q392G/C398L, QQ400/401SS, F395C, were generated using the QuikChange Site-Directed

Mutagenesis Kit (Stratagene). All plasmids were verified by sequencing.

Western blot and immunoprecipitation

Whole cell extract was prepared by lysing the cells in a buffer containing 50 mM Tris–HCl (pH

8.0), 120 mM NaCl, 0.5% NP-40, 1 mM dithiothreitol, 2 μg/ml aprotinin and 2 μg/ml leupeptin.

Cell lysates were subjected to SDS–PAGE, followed by immunoblotting analysis with anti-p53

(DO1, Santa Cruz Biotechnology), anti-p21 (C-19, Santa Cruz), anti-PP2A A subunit (Upstate), anti-PP2A C subunit (1D6, Upstate), anti-PP2A B56γ3 (against full-length B56γ3), anti-ERK

(Santa Cruz Biotechnology), anti-Cyclin G (Santa Cruz Biotechnology), anti-SGOL1

(ABNOVA), or anti-vinculin (VIN-11-5, Sigma) antibodies. For Thr55 dephosphorylation, the

cell lysate was immunoprecipitated with phospho-specific antibody for Thr55 (Ab202) and immunoblotted with anti-p53 antibody (Li et al., 2007). For GST pull down experiments, GST fusion proteins were expressed in BL21 bacteria and purified using glutathione sepharose beads.

U2OS cell lysate or 35S-labeled p53 was then incubated with the GST protein. Precipitated proteins were then analyzed by SDS-PAGE followed by immunoblotting or autoradiography. The peptide competition experiments were performed by addition of different peptides in the reactions at 100- to 1000-fold molar excess. For interaction of endogenous p53 with the transfected B56 proteins, U2OS cells were transfected with various B56 plasmids using FuGene (Roche) and lysed 28 h after transfection. Immunoprecipitation was performed using either anti-p53 polyclonal antibody (FL393, Santa Cruz) or anti-HA 12CA5 monoclonal antibody. The amounts of co-precipitated proteins were determined by immunoblotting.

Cell proliferation and anchorage-independent growth assays

To generate proliferation curves for HCT116 cells, cells were transfected with wild type, QC,

F395C, Q392G, C398L, or QQ mutant B56γ3 or a control CMV empty vector using FuGene, seeded in triplicate and counted at 0, 24, 48, 72, 96 and 120 h post seeding. To generate proliferation curves for H1299 cells, cells were cotransfected with wild type, QC, F395C, Q392G,

C398L, or QQ mutant B56γ3 or a control CMV empty vector along with either wild type p53 or

T55D mutant p53, using FuGene, seeded in triplicate and counted at 0, 24, 48, 72, 96 and 120 h post seeding. For anchorage-independent growth assays of HCT116 cells, cells were transfected with wild type, Q392G, C398L, QC, QQ, or F395C mutant B56γ3 or a control CMV empty vector seeded in triplicate in 0.35% Noble Agar (Fisher) and counted 4 weeks post seeding.

Results

Identification of a domain of B56γ required for interaction with p53

The B56γ gene product directs PP2A phosphatase activity toward p53, promoting p53

Thr55 dephosphorylation, activation, and tumor suppressive function. We sought to better understand the region of the B56γ protein required for interaction with p53. To this end, we compared the amino acid sequences of two different spliced forms of B56γ protein, B56γ1 and

B56γ3, able to interact with p53, and the highly related B56α protein, which is unable to interact with p53 (Li et al., 2007). We posited that the more variable regions found toward the C- and N- terminus (Figure 2.1A) of B56γ were most likely the areas of potential interaction with p53 because of their variation from the sequence of the noninteracting B56α protein and their position in the potential substrate binding area of the B56γ-PP2A holoenzyme complex as described in crystallographic studies (Xu et al., 2006; Cho and Xu, 2007). To test this, we constructed several

B56γ deletion mutants as glutathione S-transferase fusion proteins and assayed their ability to interact with the PP2A core and p53 (Figure 2.1A). Deletion of the N-terminal 156 amino acids

(ΔN156) or the C-terminus after amino acid 401 (ΔC401) disrupted the interaction between B56γ and the AC core of PP2A. Interestingly, although the ΔN156 and ΔC401 mutants were no longer able to interact with the PP2A core, they were still able to bind p53 to a level similar to that of wild type B56γ3 and B56γ1. Significantly, the ΔC347 and ΔC376 mutants lost p53 interaction, suggesting that the deleted region may be required for interaction with p53. To confirm this result, we show that the ΔN156, ΔC376, and ΔC401 mutants also disrupted interaction with the PP2A core in vivo (Figure 2.1B). In addition, the ΔC376 mutant lost interaction with p53 in vivo, whereas ΔC401 and ΔN156 mutants still maintained interaction (Figure 2.1B).

To exclude the possibility that our result could be explained by a loss of proper conformation because of the deletion, we conducted peptide competition experiments using a

peptide consisting of the amino acids of B56γ from 376 to 402 (peptide B56γ, Figure 2.1C). The assay showed that the B56γ peptide was clearly able to compete out the p53 interaction. An additional smaller peptide containing amino acids 391 to 402 (peptide 391) was also able to block the interaction, whereas a peptide containing amino acids 376 to 388 (peptide 376) was unable to do so. A combination of both smaller peptides showed results similar to those of peptide 391 only.

As a control, a peptide consisting of amino acids from the corresponding region of homology from B56α (peptide B56α) was unable to block the p53 interaction, as expected. These data suggest that the amino acids on B56γ from 391 to 402 constitute a domain required for interaction with p53.

To provide further evidence, we designed and generated four B56γ3 point mutants,

Q392G, C398L, the double mutant Q392G/C398L (QC), and the double mutant QQ400/401SS

(QQ), in which B56γ amino acids were mutated to the corresponding amino acids present on

B56α (Figure 2.1C). We assayed for the effects of the mutations on interaction with p53 and the

PP2A core in vivo (Figure 2.1D). Although the interaction was constant between the PP2A core and each of the B subunit constructs expressed, p53 interaction with the QC mutant was almost undetectable and no interaction was detected with the B56α negative control. Interestingly, both

Q392 and C398 single mutants retained their ability to interact with p53. To confirm these results, we performed p53 IP followed by immunoblotting with anti-HA antibody and showed that the

QC mutant and B56α are indeed unable to interact with p53 (Figure 2.1D, right). We previously showed an endogenous interaction between PP2A and p53 that can be strongly enhanced by

B56γ3 overexpression (Li et al., 2007). Blotting with the PP2A C subunit showed that the enhanced interaction between p53 and PP2A observed in the presence of overexpressed B56γ3 was lost in the presence of the QC mutant, suggesting that the QC mutant lost its ability to

interact with p53. Overall, these results provide additional evidence of the importance of the region of B56γ between amino acids 391 and 402 for p53 interaction.

A cancer-derived B56γ3 mutant, F395C, is unable to interact with p53 or promote Thr55 dephosphorylation

To continue to investigate the importance of this region in tumor progression, we analyzed annotated cDNA sequences within our p53 interaction domain present in public databases that were derived from human cancer cell lines and tumor samples and identified a mutation of a phenylalanine residue at amino acid position 395 of B56γ3 that was mutated to cysteine present in a lung tumor sample (Dias Neto et al., 2000). Since this mutation was present within our identified p53 interaction domain (Figure 2.2A), we tested whether this mutation could disrupt interaction with p53. Although the F395C mutant was able to interact normally with the

PP2A A and C subunits, it specifically lost its ability to interact with p53 (Figure 2.2A). The interaction with the QC mutant was shown as a negative control. To provide further evidence for specificity, we tested the ability of F395C to bind several other PP2A substrates, such as Cyclin

G2 (Bennin et al., 2002), ERK (Letourneux et al., 2006), Shugoshin (Riedel et al., 2006). As shown in Figure 2.2E, F395C, although defect in p53 binding, retains its ability to bind Cyclin G2,

ERK and Shugoshin. Together, these results suggest that the F395C mutant was unable to specifically interact with p53 and thus unable to recruit the PP2A core to p53.

To test this directly, we assayed the ability of the F395C mutant to promote p53 Thr55 dephosphorylation and p21 induction. As shown in Figure 2.2A, overexpression of wild type

B56γ3 promoted efficient dephosphorylation of p53 at Thr55, compared with the empty vector control. This dephosphorylation correlated with an induction of the p53 transcriptional target p21, underscoring the functional significance of the dephosphorylation. Overexpression of the F395C

mutant, however, was unable to promote p53 Thr55 dephosphorylation or p21 induction, suggesting that this mutant has lost its ability to direct PP2A phosphatase activity toward p53. In addition, the QC mutant was also unable to promote Thr55 dephosphorylation or p21 induction, providing evidence that interaction with p53 is required for the tumor suppressive function of

B56γ-PP2A. In addition, as shown in Figure 2.2B, mutants able to interact with p53, including

Q392G, C398L, and the QQ double mutant, promoted p53 Thr55 dephosphorylation and induction of p21 protein levels similar to that of wild type B56γ3. Taken together with the interaction results, we have shown that a cancer-derived mutation of B56γ3 may disrupt a physiologically significant function of PP2A in preventing tumor progression through activation of p53.

To investigate the potential of the F395C mutant to block the function of the endogenous

B56γ protein, we assayed for p53 Thr55 dephosphorylation and p21 induction after DNA damage in the presence of exogenous wild type B56γ3, F395C mutant B56γ3 or an empty vector control

(Figure 2.2C). In the case of controls, efficient Thr55 dephosphorylation and induction of p21 protein occur after DNA damage. In the presence of exogenous wild type B56γ3, Thr55 phosphorylation levels are already lower under mock conditions, and further dephosphorylation is observed after IR, whereas p21 levels are higher in the mock lane and increase further after IR.

Importantly, in the presence of the F395C mutant, Thr55 phosphorylation levels do not decrease after DNA damage and induction of the p53 transcriptional target p21 is also blocked. This result questions whether the F395C mutant could potentially function as a dominant negative mutant when present at a level similar to that of endogenous wild type protein. To test this possibility, we titrated the amount of overexpressed protein and assayed for the effect on Thr55 phosphorylation after DNA damage (Figure 2.2D). Decreasing the level of F395C expression to a level similar to that of endogenous protein (0.25-0.5 μg) led to some Thr55 dephosphorylation after DNA

damage, whereas higher levels of F395C expression completely blocked this function. This finding suggests that F395C may be able to function as a dominant negative mutant by competing with wild type B56γ for binding to the PP2A core, thereby blocking C subunit-substrate interactions. To further support this view, we assayed for the ability of either wild type or F395C mutant B56γ3 to interact with the endogenous B56γ protein (Figure S2.1). No interaction was detected between exogenous protein and the endogenous B56γ, suggesting that the inhibitory mechanism of F395C overexpression is probably through competitive binding to the AC core of

PP2A or due to some unknown factors.

Cancer-derived mutant B56γ3 lacks p53-dependent tumor suppressive functions

Although the F395C mutation was found in a tumor sample, we wanted to elucidate whether the mutation could directly contribute to a cancer phenotype. We previously showed that wild type B56γ overexpression decreased anchorage-independent growth and cell proliferation rates utilizing a p53-dependent mechanism (Li et al., 2007). Based on our findings that F395C mutant B56γ3 lacks interaction with p53 and ability to promote p53 activation, we first tested whether it could function to block anchorage-independent cell growth in a p53-dependent manner.

HCT116 cells with either a p53-/- or p53+/+ background were transfected with each of the B56γ3 constructs, including F395C, QC, wild type, Q392G, C398L, and QQ, or an empty vector control and seeded in soft agar. As shown in Figure 2.3, overexpression of each of the B56γ3 constructs into the HCT116 cells lacking p53 led to a similar decrease in the number of colonies present in the agar from approximately 280 colonies in the empty vector control to approximately 250 colonies. In the presence of p53, however, only those B56γ3 constructs able to interact with p53, namely wild type, Q392G, C398L, and QQ, showed a more dramatic decrease in colony number from about 250 colonies in the empty vector control to around 75 colonies. Expression of the

F395C mutant led to a decrease in colony number, but only to an extent similar to that observed in the absence of p53 corresponding to around 210 colonies. Statistical analysis of the data is shown in Figure S2.2. This finding suggests that F395C mutant B56γ3 is no longer able to function as a p53-dependent tumor suppressor protein. Samples from transfected cells were analyzed for p21 induction. A strong correlation was shown between the ability of a B56γ3 construct to interact with p53, promote p21 induction, and the ability of that construct to block anchorage-independent growth in a p53-dependent manner.

To provide further evidence that F395C mutant B56γ3 lacks p53-dependent tumor suppressive function, we tested whether expression of this mutant protein could inhibit cell proliferation in a p53-dependent manner. As shown in Figure 2.4A, expression of wild type

B56γ3 in the presence of p53 led to a dramatic decrease in cell proliferation corresponding to a

45% reduction in cell number by 120 h. Interestingly, the F395C mutant was unable to block cell proliferation to a similar extent with only a 20% decrease in cell number. The reduced inhibition of cell proliferation observed in the presence of the F395C mutant can be attributed to its p53- independent function as expression of the mutant in an isogenic p53 null background (HCT116 p53-/-) led to a decrease that showed no significant difference from wild type B56γ3 (P = 0.08)

(Figure 2.4A: Figure S2.2). To provide further data correlating p53 interaction with p53- dependent functions of B56γ3, we expressed the non-interacting QC mutant, as well as the interacting Q392G, C398L, and QQ mutants in the same assay. The non-interacting QC mutant showed results similar to those of the F395C mutant, whereas those mutants that retained p53 interaction also retained similar p53-dependent functions comparable to wild type B56γ3 (Figure

2.4A). As a control, lysates of the cell proliferation experiment were analyzed. As expected, increase in p21 protein levels occurred in the case of wild type B56γ3 but not F395C

overexpression in HCT116 p53+/+ and HCT116 p53-/- cells, which correlated with the p53- dependent inhibition of cell proliferation (Figure 2.4A).

To provide evidence that p53 Thr55 phosphorylation status contributes to the p53- dependent tumor suppressive function of B56γ, similar cell proliferation experiments were conducted in H1299 cells that lack functional wild type p53. These cells were transfected with either wild type p53 or a Thr55 to Asp mutant (T55D) that cannot be dephosphorylated at Thr55.

As shown in Figure 2.4B, expression of wild type B56γ3 along with wild type p53 led to a significant inhibition of cell proliferation corresponding to a 50% decrease in cell number at 120 h. Expression of F395C mutant B56γ3, however, led to only a small inhibition of cell proliferation corresponding to approximately a 10% decrease in cell number. This small inhibition can be attributed to the Thr55 dephosphorylation-independent function, as a similar inhibition was observed in the presence of T55D mutant p53. Similarly, expression of the noninteracting QC mutant also led to a small inhibition of cell proliferation supporting the notion that p53 interaction and Thr55 dephosphorylation are required for the p53-dependent tumor suppressive function of B56γ3 (Figure 2.4B). Statistical analysis of the data is shown in Figure

S2.2. Cell lysates from the experiment were subjected to western blot to assess p21 protein levels

(Figure 2.4B). The correlation between p21 induction and decrease in cell proliferation was again documented in the presence of wild type p53 and B56γ3. The lack of p21 induction in the presence of the F395C mutant B56γ3 provides further evidence that this protein has lost its Thr55 dephosphorylation-dependent tumor suppressive function. Taken together our findings show for the first time that a cancer-derived mutant of B56γ3 lacks the physiologically significant p53- dependent tumor suppressive function, providing evidence for the importance of this pathway in protecting cells from a cancerous phenotype.

Discussion

In this study, we describe a novel molecular mechanism behind a cancer-derived loss of function mutation in the PP2A B56γ regulatory subunit gene. This mutation was previously described in a lung tumor sample, although the functional significance of this mutation in tumorigenesis was not previously characterized. The missense mutation changes a phenylalanine residue at amino acid 395 to a cysteine within a region we mapped as being required for B56γ-

PP2A-p53 interaction. According to recent crystal structure data, this region maps to an area of high structural variability positioned in such a manner as to allow docking of the PP2A core with the substrate (Xu et al., 2006; Cho and Xu, 2007). This provides further support to the notion that this region directly bridges the p53-PP2A interaction. In terms of tumorigenic function, we showed that the mutant protein has lost its ability to mediate the p53-dependent tumor suppressive function of PP2A. Specifically we showed that F395C is no longer able to interact with p53 or inhibit cultured cancer cell proliferation or growth on soft agar in a p53-dependent manner. Unfortunately, we are unable to assess p53 status in this identified cancer sample.

However, recent data suggest that as many as 61% of lung tumors have wild type p53 (Petitjean et al., 2007), suggesting the importance of the B56γ3-p53 interaction in regulating wild type p53 tumor suppression function.

Although the F395C mutant lacks p53 interaction, it maintains its ability to interact with the PP2A core similar to the wild type protein (Figure 2.2A). As such, we were able to show a competitive effect of F395C expression affecting the activity of the endogenous protein. It seems that even a single copy of this mutation may affect the function of the wild type protein, because when levels of F395C expression are comparable to endogenous levels, there is a decrease in p53

Thr55 dephosphorylation after DNA damage (Figure 2.2D). The mechanism for this interference seems to be due to competition between the mutant and wild type proteins for some other

molecules required for function, as increasing the amount of expressed F395C protein enhances the effect. The molecule being titrated out may be the PP2A core complex, as both F395C and wild type B56γ can bind the PP2A core to similar levels. In addition, it does not seem as though the B56γ protein can dimerize because HA-B56γ3 was unable to interact with endogenous protein

(Figure S2.1). These findings suggest that the mechanism for F395C interference with wild type function is not dominant negative in the classical sense but is more because of competitive binding to other molecules.

Our investigation of the functional consequences of a previously described cancer- derived mutation in the B56γ gene provides additional evidence as to the importance of B56γ in tumor suppression. Although the mutation was identified in a lung tumor sample complementary

DNA library, the mutation was not present in the B56γ gene sequence of normal lung tissue libraries present in the NIH database, suggesting that this mutation may have contributed to the phenotype of the tumor sample. In addition, since the mutation was not present in the NCBI SNP database, it is not likely to be a polymorphism. Although there was only one mutation found on the NIH database directly within the mapped p53 interaction domain, several mutations have been identified immediately outside this region that could potentially disrupt the p53-B56γ interaction.

We thus tested one of the mutants, A383G, which is immediately adjacent to the mapped p53 interaction domain (aa391-401) (Suzuki et al., 2004). Our results suggest that A383G also specifically lost its ability to interact with p53 (Figure S2.3B). Furthermore, we showed that

A383G is unable to promote p53 Thr55 dephosphorylation (Figure S2.3C) and p53-dependent tumor suppressive function of PP2A (Figure S2.3D). This finding supports our conclusion that interaction with p53 plays an important role in PP2A-mediated activation of p53 and prevention of tumor progression. Perhaps it is worthwhile to mention that additional mutations have been detected in the B56γ gene. For example, mutations that disrupt binding to the A and C subunits

would potentially disrupt both the p53-dependent and independent tumor suppressive functions of this protein and may therefore be more common. Furthermore, the previously described mutations in the A subunit genes that contribute to tumorigenesis may function to disrupt the same pathway as this B56γ mutation, therefore making B56γ mutations in these cancer types less likely.

Nevertheless, our study provides compelling biochemical data suggesting the importance of the

F395C mutation and potentially other mutants in that expression of the mutant proteins no longer promotes PP2A-dependent activation of the tumor suppressor protein p53 and actually disrupts the normal p53 response to DNA damage stress.

Although the cancer-derived F395C mutant protein no longer promotes a p53-dependent tumor suppressive function, it still demonstrates a less dramatic p53-independent function that has been previously described (Li et al., 2007). Interestingly, this mutant protein is able to distinguish between these two and therefore provides compelling evidence as to the ability of the

B subunits of PP2A to direct the holoenzyme to specific substrate targets and the potential for the presence of multiple substrate binding domains within a single B subunit. Clearly, this subunit must promote PP2A-mediated dephosphorylation of other target proteins inside the cell that mediate this p53-independent function and further study may provide insight into these protein targets and their role in PP2A-dependent tumor suppression as well as into the domain of B56γ responsible for mediating interaction with the PP2A substrate. Overall, although the regulation of

PP2A functions is complex, our study adds significant support to the role of B56γ-PP2A in tumor suppression specifically through activation of the tumor suppressor protein, p53.

References

Arnold, H.K., and Sears, R.C. (2008). A tumor suppressor role for PP2A-B56α through negative regulation of c-Myc and other key oncoproteins. Cancer Metastasis Rev. 27, 147–158.

Bennin, D.A., Don, A.S.A., Brake, T., McKenzie, J.L., Rosenbaum, H., Ortiz, L., DePaoli-Roach, A.A., and Horne, M.C. (2002). Cyclin G2 associates with protein phosphatase 2A catalytic and regulatory B’ subunits in active complexes and induces nuclear aberrations and a G1/S phase cell cycle arrest. J. Biol. Chem. 277, 27449–27467.

Bode, A.M., and Dong, Z. (2004). Post-translational modification of p53 in tumorigenesis. Nat. Rev. Cancer 4, 793–805.

Chen, W., Arroyo, J.D., Timmons, J.C., Possemato, R., and Hahn, W.C. (2005). Cancer- associated PP2A Aα subunits induce functional haploinsufficiency and tumorigenicity. Cancer Res. 65, 8183–8192.

Cho, U.S., and Xu, W. (2007). Crystal structure of a protein phosphatase 2A heterotrimeric holoenzyme. Nature 445, 53–57.

Dias Neto, E., Correa, R.G., Verjovski-Almeida, S., Briones, M.R., Nagai, M.A., Da Silva, W., Zago, M.A., Bordin, S., Costa, F.F., Goldman, G.H., et al. (2000). Shotgun sequencing of the human transcriptome with ORF expressed sequence tags. Proc. Natl. Acad. Sci. U.S.A. 97, 3491– 3496.

Eichhorn, P.J.A., Creyghton, M.P., and Bernards, R. (2009). Protein phosphatase 2A regulatory subunits and cancer. Biochim. Biophys. Acta 1795, 1–15.

Forester, C.M., Maddox, J., Louis, J. V, Goris, J., and Virshup, D.M. (2007). Control of mitotic exit by PP2A regulation of Cdc25C and Cdk1. Proc. Natl. Acad. Sci. U.S.A. 104, 19867–19872.

Grochola, L.F., Vazquez, A., Bond, E.E., Würl, P., Taubert, H., Müller, T.H., Levine, A.J., and Bond, G.L. (2009). Recent natural selection identifies a genetic variant in a regulatory subunit of protein phosphatase 2A that associates with altered cancer risk and survival. Clin. Cancer Res. 15, 6301–6308.

Janssens, V., and Goris, J. (2001). Protein phosphatase 2A: a highly regulated family of serine/threonine phosphatases implicated in cell growth and signalling. Biochem. J. 353, 417–439.

Kruse, J., and Gu, W. (2009). Modes of p53 regulation. Cell 137, 609–622.

Letourneux, C., Rocher, G., and Porteu, F. (2006). B56-containing PP2A dephosphorylate ERK and their activity is controlled by the early gene IEX-1 and ERK. EMBO J. 25, 727–738.

Li, H.H., Cai, X., Shouse, G.P., Piluso, L.G., and Liu, X. (2007). A specific PP2A regulatory subunit, B56γ, mediates DNA damage-induced dephosphorylation of p53 at Thr55. EMBO J. 26, 402–411.

Petitjean, A., Mathe, E., Kato, S., Ishioka, C., Sean, V., Hainaut, P., and Olivier, M. (2007). Impact of Mutant p53 Functional Properties on TP53 Mutation Patterns and Tumor Phenotype : Lessons from Recent Developments in the IARC TP53 Database. Hum. Mutat. 28, 622–629.

Riedel, C.G., Katis, V.L., Katou, Y., Mori, S., Itoh, T., Helmhart, W., Gálová, M., Petronczki, M., Gregan, J., Cetin, B., et al. (2006). Protein phosphatase 2A protects centromeric sister chromatid cohesion during meiosis I. Nature 441, 53–61.

Ruediger, R., Pham, H.T., and Walter, G. (2001a). Alterations in protein phosphatase 2A subunit interaction in human carcinomas of the lung and colon with mutations in the A β subunit gene. Oncogene 20, 1892–1899.

Ruediger, R., Pham, H.T., and Walter, G. (2001b). Disruption of protein phosphatase 2A subunit interaction in human cancers with mutations in the Aα subunit gene. Oncogene 20, 10–15.

Schönthal, A.H. (2001). Role of serine/threonine protein phosphatase 2A in cancer. Cancer Lett. 170, 1–13.

Shouse, G.P., Cai, X., and Liu, X. (2008). Serine 15 phosphorylation of p53 directs its interaction with B56γ and the tumor suppressor activity of B56γ-specific protein phosphatase 2A. Mol. Cell. Biol. 28, 448–456.

Suzuki, Y., Yamashita, R., Shirota, M., Sakakibara, Y., Chiba, J., Mizushima-Sugano, J., Nakai, K., and Sugano, S. (2004). Sequence Comparison of Human and Mouse Genes Reveals a Homologous Block Structure in the Promoter Regions. Genome Res. 14, 1711–1718.

Virshup, D.M., and Shenolikar, S. (2009). From promiscuity to precision: protein phosphatases get a makeover. Mol. Cell 33, 537–545.

Vousden, K.H., and Prives, C. (2009). Blinded by the Light: The Growing Complexity of p53. Cell 137, 413–431.

Wang, S.S., Esplin, E.D., Li, J.L., Huang, L., Gazdar, A., Minna, J., and Evans, G.A. (1998). Alterations of the PPP2R1B Gene in Human Lung and Colon Cancer. Science (80- ) 282, 284– 287.

Xu, Y., Xing, Y., Chen, Y., Chao, Y., Lin, Z., Fan, E., Yu, J.W., Strack, S., Jeffrey, P.D., and Shi, Y. (2006). Structure of the protein phosphatase 2A holoenzyme. Cell 127, 1239–1251.

Acknowledgements

We are grateful to Dr. B. Shen at City of Hope for assistance searching public databases for mutations in the B56γ gene and to P. Podlesny for generating B56γ mutant plasmid constructs.

We thank all members of our laboratory for many helpful discussions. This work was supported by NIH grant CA075180 to X. Liu.

Figure legend

Figure 2.1 Mapping of the B56γ domain required for interaction with p53.

(A) Left: Diagram of amino acid sequences of B56α, B56γ and various B56γ deletion constructs aligned based on sequence homology. Shades of gray indicate increasing percentage of identical amino acids within that region. Right: Each of the B56γ constructs were expressed in bacteria as

GST-fusion proteins, purified, and analyzed by SDS-PAGE (upper). U2OS cell lysates were incubated with the purified GST-fusion proteins indicated, then analyzed by western blot against p53, PP2A A and C, and vinculin (vinc) (lower). (B) U2OS cells were transfected with the HA- tagged B56γ constructs listed, lysed, and subjected to immunoprecipitation with anti-HA antibody. The precipitated proteins were analyzed by western blot against p53, PP2A A and C,

HA, and vinculin (vinc). (C) 35S-p53 was incubated with GST protein (GST), GST-B56γ3 protein, or GST-B56γ3 with peptides corresponding to amino acids 376 to 402 (B56γ), 376 to 388 (376), or 391 to 402 of B56γ3 (391), or amino acids 411 to 437 of B56α (B56α). Amino acid sequence of the p53-interaction domain of B56γ aligned with the homologous sequence of B56α is shown below. (D) Lysates of U2OS cells transfected with HA-tagged B56γ3, Q392G, C398L,

Q392G/C398L (QC), Q400S/Q401S (QQ) mutant B56γ3, or B56α were either immunoprecipitated with anti-HA antibody, then analyzed by western blot against p53, PP2A A and C, HA, and vinculin (vinc) (left) or with anti-p53 antibody, then analyzed by western blot against HA, PP2A C, p53, and vinculin (vinc) (right).

Figure 2.2 Mutant B56γ3 is unable to promote p53 Thr55 dephosphorylation.

(A) Lysates of U2OS cells transfected with HA-tagged B56γ3, F395C, or Q392G/C398L (QC) mutant B56γ3, were either immunoprecipitated with anti-HA antibody, then analyzed by western blot against p53, PP2A A and C, HA, and vinculin (vinc) (left); or analyzed for p53 Thr55

phosphorylation, p21 protein levels, HA, p53, and vinculin (vinc) (right). Amino acid sequence of the p53-interaction domain of B56γ showing the cancer-derived F395C mutation is present on the top. (B) Lysates of U2OS cells transfected with empty vector control (EV), HA-tagged B56γ3,

Q392G, C398L, Q392G/C398L (QC), or Q400S/Q401S (QQ) mutant B56γ3 were analyzed for p53 Thr55 phosphorylation, p21 protein levels, HA, p53, and vinculin (vinc). (C) Lysates of

U2OS cells transfected with empty vector control (EV), HA-tagged B56γ3, or F395C mutant

B56γ3 and either mock treated (M) or treated with ionizing radiation (IR) were analyzed for p53

Thr55 phosphorylation, HA, p53, and vinculin (vinc). (D) Lysates of U2OS cells transfected with varying amounts of HA-tagged F395C mutant B56γ3, were either mock (M) treated or treated with ionizing radiation (IR), then analyzed by western blot against endogenous (endo) and exogenous (exo) B56γ, as well as p53 and Thr55 phosphorylation. (E) Lysates of U2OS cells transfected with HA-tagged B56γ3 or F395C were immunoprecipitated with anti-HA antibody, then analyzed by western blot against p53, Cyclin G, ERK, Sgo, PP2A A, HA, and vinculin

(vinc) antibodies.

Figure 2.3 Overexpression of cancer-derived mutant B56γ3 is unable to block anchorage- independent growth in a p53-dependent manner.

(A) Anchorage-independent growth of HCT116 cells transfected with wild type (WT), F395C,

Q392G/C398L (QC), Q392G, C398L, Q400S/Q401S (QQ), or an empty CMV vector control

(EV). (B) Colony numbers. The values are the averages +/- standard deviations of the results of three representative experiments. (C) Immunoblots of the transfected HA-B56γ3 and endogenous

B56γ3/B56γ2, p53, p21, and vinculin (vinc) proteins. -: HCT116 -/-; +: HCT116 +/+.

Figure 2.4 Overexpression of cancer-derived mutant B56γ3 is unable to block cell proliferation in a p53-dependent manner.

Representatives of cell proliferation of the HCT116 human colon cancer cell lines (A) or the

H1299 human lung cancer cell line (B) transfected with either wild type (WT), F395C,

Q392G/C398L (QC), Q392G, C398L, Q400S/Q401S (QQ) mutant B56γ3, or a control CMV empty vector. Numbers of cells present at the 120 h time point were normalized against the representative empty vector controls and plotted in a bar graph. Error bars show average +/- s.d. from triplicate plates in one representative experiment. Cell lysates were analyzed by immunoblotting of the transfected HA-B56γ3, endogenous B56γ3/B56γ2, p53 and p21 proteins.

H1299 + WT: H1299 transfected with wild type p53; H1299 + T55D: H1299 transfected with

T55D. M: empty vector-transfected cell lysate at seeding.

Figure S2.1 HA-B56γ3 is unable to interact with endogenous B56γ.

Lysates of U2OS cells transfected with either wild type or F395C mutant B56γ3, were immunoprecipitated with HA antibody, then analyzed by western blot against B56γ and vinculin

(vinc).

Figure S2.2 Statistical analysis of cell growth and transformation data.

P-values for anchorage-independent cell growth experiments from Figure 3 (A); for HCT116 cells growth experiments from Figure 4A (B); and for H1299 cell cells growth experiments from

Figure 4B (C) were calculated by comparing the effect of expressing the various mutants either versus the EV control or versus the wild type protein.

Figure S2.3 The A383G mutant is unable to promote p53-dependent tumor suppressive functions.

(A) Amino acid sequence of B56γ showing the p53-interaction domain, adjacent regions and the cancer-derived mutations. Lysates of U2OS cells transfected with HA-tagged B56γ3, A383G, or an empty vector control, were either immunoprecipitated with anti-HA antibody, then analyzed by western blot against p53, PP2A A and C, HA, and vinculin (B); or analyzed for p53 Thr55 phosphorylation, p21 protein levels, HA, p53, and vinculin (C). (D) Representatives of cell proliferation of HCT116 cells were transfected with either wild type, or A383G, or an empty vector control. Numbers of cells present at the 120 h time point were normalized against the representative empty vector controls and plotted in a bar graph. Error bars show average +/- s.d. from triplicate plates in one representative experiment. Cell lysates were analyzed by immunoblotting of the transfected HA-B56γ3, endogenous B56γ3/B56γ2, and p53 proteins. M: empty vector-transfected cell lysate at seeding. (E) Statistical analysis of cell growth from (D).

Figures

Figure 2.1 Mapping of the B56γ domain required for interaction with p53.

Figure 2.2 Mutant B56γ3 is unable to promote p53 Thr55 dephosphorylation.

Figure 2.3 Overexpression of cancer-derived mutant B56γ3 is unable to block anchorage- independent growth in a p53-dependent manner.

Figure 2.4 Overexpression of cancer-derived mutant B56γ3 is unable to block cell proliferation in a p53-dependent manner.

Figure S2.1 HA-B56γ3 is unable to interact with endogenous B56γ.

Figure S2.2 Statistical analysis of cell growth and transformation data.

Figure S2.3 The A383G mutant is unable to promote p53-dependent tumor suppressive functions.

CHAPTER 3

HEAT repeat 1 is required for B56γ-PP2A holoenzyme assembly and tumor suppressive function

This research was originally published in The journal of biological chemistry. Y. Nobumori, G.P.

Shouse and X. Liu. HEAT repeat 1 motif is required for B56γ-containing protein phosphatase 2A

(B56γ-PP2A) holoenzyme assembly and tumor-suppressive function. The journal of biological chemistry. 2012; 287:11030-11036. © the American Society for Biochemistry and Molecular

Biology. I am responsible for all of the figures in this chapter except Figure 3.3A which was prepared by G.P. Shouse.

Abstract

Protein phosphatase 2A (PP2A) enzyme consists of a heterodimeric core (AC core) comprising a scaffolding subunit (A), a catalytic subunit (C), and a variable regulatory subunit (B). Earlier studies suggest that upon DNA damage, a specific B subunit, B56γ, bridges the PP2A AC core to p53, leading to dephosphorylation of p53 at Thr55, induction of the p53 transcriptional target p21 and the inhibition of cell proliferation and transformation. In addition to dephosphorylation of p53, B56γ–PP2A also inhibits cell proliferation and transformation by an unknown mechanism.

B56γ contains eighteen α-helices that are organized into eight HEAT (Huntington-elongation-

A subunit-TOR) repeat motifs. Although previous crystal structure study has revealed the residues of B56γ that directly contact the A and C subunits, the contribution of HEAT repeats to holoenzyme assembly and to B56γ-PP2A tumor suppressive function remains to be elucidated.

Here, we show that HEAT repeat 1 is required for the interaction of B56γ with the PP2A AC core and, more importantly, for B56γ-PP2A tumor suppressive function. Within this region, we identified a tumor-associated mutation, C39R, which disrupts the interaction of B56γ with the AC core and thus was unable to mediate dephosphorylation of p53 by PP2A. Furthermore, due to its lack of AC interaction, C39R was also unable to promote the p53-independent tumor suppressive function of B56γ-PP2A. This study provides structural insight into the PP2A holoenzyme assembly and emphasizes the importance of HEAT repeat 1 in B56γ-PP2A tumor suppressive function.

Introduction

The protein phosphatase 2A (PP2A) family of serine/threonine phosphoprotein phosphatases is involved in a multitude of cell-signaling pathways. The PP2A holoenzyme is a heterotrimeric complex that consists of scaffolding A subunit, catalytic C subunit, and regulatory

B subunit. Each subunit is thought to have its own distinct function: the C subunit catalyzes the dephosphorylation of specific serine/threonine residues on target substrates and the A subunit acts as a scaffold holding the complex together. Together, A and C subunits form a PP2A core enzyme (AC core). The AC core is associated with one of the variable B subunits, which determines the diverse cellular localization, substrate specificities, and enzymatic activity of

PP2A holoenzyme (Eichhorn et al., 2009; Virshup and Shenolikar, 2009).

Recent evidence suggested that a subset of B56-containing PP2A holoenzyme (B56-

PP2A) exhibits tumor suppressive functions. The B56 family consists of five different genes, α

(PPP2R5A), β (PPP2R5B), γ (PPP2R5C), δ (PPP2R5D), and ε (PPP2R5E) (Csortos et al., 1996;

McCright et al., 1996). B56γ-PP2A has been reported to dephosphorylate tumor suppressor p53 at

Thr55, leading to p53 activation, induction of cdk inhibitor p21, and inhibition of cell proliferation (Li et al., 2007); B56γ-PP2A to dephosphorylate the c-Myc oncogene, resulting in c-Myc inactivation (Yeh et al., 2004; Arnold and Sears, 2006), and B56γ-PP2A to dephosphorylate Cdc25c, blocking cell cycle progression (Margolis et al., 2006; Forester et al.,

2007). These studies suggest that B56-PP2A exerts its tumor suppressive function by bridging the

PP2A AC core to the substrate proteins involved in cell growth and proliferation. In support of this view, some viral oncoproteins have been shown to function by displacing B56 subunit from the AC core binding (Chen et al., 2004, 2007). In addition, among seven cancer-derived mutations (E64D in lung carcinoma, E64G in breast carcinoma, R418W in malignant melanoma,

Δ171-589 also in breast carcinoma, R182W, R183W and R183G in ovarian carcinoma) reported in PP2A Aα gene (PPP2R1A) to date, four of them have been characterized and the unifying effect of them is loss of interaction with either C subunit or B56 subunits (Ruediger et al., 2001b;

Chen et al., 2005b; Jones et al., 2010). We have also reported two tumor-associated mutations in

B56γ gene that specifically block interaction with p53 and thus p53-dependent, but not p53- independent, tumor suppressive activity of B56γ-PP2A (Shouse et al., 2010). Together those data suggest that interaction of B56 subunit with either the PP2A AC core or its substrate is critical for its tumor suppressive function. However, no B56 mutation has been identified to specifically block the PP2A AC core interaction in human cancer to date.

The crystal structure of B56γ subunit revealed that it has an elongated, superhelical structure comprising eighteen α-helices that are organized into eight HEAT-like (Huntington- elongation-A subunit-TOR-like) repeat motifs (Xu et al., 2006; Cho and Xu, 2007). Each HEAT repeat consists of two antiparallel α-helices connected by an intra-loop, and adjacent HEAT repeats are connected by short inter-repeat turns. Upon holoenzyme assembly, B56γ is placed into a position close to the active site of the C subunit and forms the substrate docking site of the holoenzyme, supporting the view that B56γ controls PP2A specificity by bridging the PP2A AC core and phosphorylated protein substrates. The structure of the B56γ-PP2A holoenzyme reveals a number of conserved amino acid residues located mainly on the intra-loops which mediate interaction of B56γ with either A or C subunit. However, how each HEAT repeat modulates holoenzyme formation and, importantly, contributes to B56γ-PP2A activity remains unknown.

In this study, we show that deletion of HEAT repeat 1, although distant from the A and C subunits, prevents B56γ from binding to the AC core, suggesting that this motif is critical for the assembly of the B56γ-PP2A holoenzyme. To further investigate the importance of HEAT repeat 1,

we characterized a mutation, C39R, within this motif that was previously identified from a pooled glandular tumor sample. Our data reveal that, like the HEAT repeat 1 deletion mutants, the C39R mutant was unable to interact with the PP2A A and C subunits. Importantly, we show that, although retaining binding to p53, all HEAT repeat 1 mutants tested fail to promote p53 Thr55 dephosphorylation and transcriptional activation of the p21 gene. As a consequence, they abolished the p53-dependent tumor suppressive function of PP2A. Furthermore, due to their missing interaction with the A and C subunit, those mutants also lost the p53-independent tumor suppressive function of B56γ-PP2A. This study thus provides structural insight into the PP2A holoenzyme assembly and suggests an additional mechanism to inactivate tumor suppressive function of B56γ-PP2A.

Materials and Methods

Cell culture and plasmids

U2OS and HCT116 cells were cultured in McCoy’s 5A medium supplemented with 10% fetal calf serum. The B56γ deletion mutants, ΔN40, ΔN53, and ΔN73 were generated by PCR from the wild type B56γ3 gene. The B56γ point mutants, C39R, C39A, and C39S, were generated using the QuikChange Site-Directed Mutagenesis Kit (Stratagene). All plasmids were verified by sequencing.

Western blot and immunoprecipitation

Whole cell extract was prepared by lysing the cells in a buffer containing 50 mM Tris-HCl (pH

8.0), 120 mM NaCl, 0.5% NP-40, 1 mM dithiothreitol, 2 μg/ml aprotinin and 2 μg/ml leupeptin.

Cell lysates were subjected to SDS-PAGE, followed by western blot analysis with anti-p53 (DO1,

Santa Cruz Biotechnology), anti-PP2A A subunit (Upstate), anti-PP2A C subunit (1D6, Upstate), anti-p21 (Santa Cruz Biotechnology), anti-PP2A B56γ (Shouse et al., 2008), anti-ERK (Santa

Cruz Biotechnology), anti-cyclin G (Santa Cruz Biotechnology), anti-HA (12CA5), or anti- vinculin (VIN-11-5, Sigma) antibodies. For Thr55 dephosphorylation, the cell lysate was immunoprecipitated with phospho-specific antibody for Thr55 (Ab202) and immunoblotted with anti-p53 antibody (Li et al., 2007). For interaction of PP2A A and C subunits, p53, ERK, and cyclin G with B56γ3 proteins, U2OS cells were transfected with various B56γ plasmids using

FuGene (Roche) or BioT (Bioland Scientific) and lysed 28 h after transfection.

Immunoprecipitation was performed using anti-HA monoclonal antibody. The amounts of co-

precipitated proteins were determined by immunoblotting. For microcystin binding assay, U2OS cell lysate was incubated with microcystin agarose beads (Upstate).

Identification of cancer-derived mutation

The NCBI AceView program (http://www.ncbi.nlm.nih.gov/IEB/Research/Acembly/index.html) provides a comprehensive sequence of the human transcriptome and genes of all quality-filtered human complementary DNA data from GenBank, RefSeq, dbEST, and Trace in a strictly complementary DNA-supported manner. Using this program, we looked for B56γ mutations within HEAT repeat 1 region from the annotated sequencing data taken from samples of tumor sample and cancer cell lines. C39R mutation was identified from a glandular tumor sample and the corresponding gene accession number is CB956287.

Circular dichroism (CD) analysis

GST fusion proteins were expressed in BL21 bacteria and purified using glutathione sepharose beads (GE Healthcare) in 50 mM Tris-HCl (pH=8.0) and 100 mM NaCl. Circular dichroism (CD) measurements were performed on a Jasco J-815 CD spectrophotometer (Essex, United Kingdom).

Cell proliferation and anchorage-independent growth assays

To generate proliferation curves for HCT116 cells, cells were transfected with wild type, C39R,

ΔN40, and ΔN73 mutant B56γ or a control CMV empty vector using BioT, seeded in triplicate and counted at 120 h after seeding. For anchorage-independent growth assays of HCT116 cells,

cells were transfected with wild type, C39R, and ΔN73 mutant B56γ or a control CMV empty vector seeded in triplicate in 0.35% Noble Agar (Fisher) and colony numbers were counted 4 weeks after seeding.

Results

HEAT repeat 1 of B56γ is required for B56γ-PP2A holoenzyme assembly

Previously, we have mapped the p53-binding domain at a varying region in the C- terminus of B56γ and shown this domain is required for p53-dependent tumor suppressive function of B56γ-PP2A (Shouse et al., 2010). To better understand the role of the N-terminal domain in B56γ function, we first constructed a HEAT repeat 1 deletion mutant (ΔN73; Figure

3.1A) as glutathione S-transferase fusion protein and assayed its ability to interact with the PP2A

AC core and p53 in U2OS cell lysates (Figure 3.1B). The assay shows that although the wild type

B56γ protein was able to interact with both the AC core and p53, the ΔN73 mutant lost interaction with the PP2A A and C subunits. This is a surprising result because, according to

B56γ-PP2A crystal structure data, the HEAT repeat 1 motif is distant from the AC core (Figure

3.1E) and thus should be less likely to affect the interaction. To exclude the possibility that our result could be explained by a loss of proper conformation because of the deletion, we conducted circular dichroism (CD) analysis (Figure 3.1C and Figure S3.1). The wild type B56γ protein is expected to have characteristic CD spectra of α-helical proteins with negative bands at 222 and

208 nm and a positive band at 193 nm. The CD spectrum of our bacterially expressed and purified B56γ protein indeed displays this feature. Importantly, the CD spectrum of purified

ΔN73 protein was identical to that of the wild type protein over the entire recorded spectrum.

These data, taken together with the ΔN73-p53 interaction result (Figure 3.1B), suggest that mutation of HEAT repeat 1 is unlikely to cause overall conformational change of the protein.

HEAT repeat 1 motif consists of two antiparallel α-helices (helix-1 and -2) connected by an intra-loop (IL1) (Figure 3.1A). To study their role in the interaction with the PP2A AC core, we generated two smaller deletion mutants: ΔN40 for deleting helix-1 only and ΔN53 for deleting

helix-1 plus IL1, and tested their ability to interact with the PP2A AC core in vivo (Figure 3.1D).

Although the wild type B56γ protein binds both the PP2A AC core and p53 effectively in U2OS cells, all HEAT repeat 1 mutants lost their interaction with the PP2A A and C subunits, suggesting that the entire HEAT repeat 1 domain is required for the PP2A AC core interaction.

All deletion mutants were able to bind p53 in vivo. Together, these data suggest that the HEAT repeat 1 motif, although distant from the A and C subunits on the crystal structure, is required for

B56γ-PP2A holoenzyme assembly.

A cancer-associated HEAT repeat 1 mutant, C39R, is unable to form B56γ-PP2A holoenzyme

Because B56γ-PP2A functions as a tumor suppressor (Chen et al., 2004; Li et al., 2007), our finding that HEAT repeat 1 is required for PP2A holoenzyme assembly prompted us to search for cancer-associated mutation within this region. We analyzed annotated complementary DNA sequences in public databases that were derived from human cancer cell lines and tumor samples and identified a mutation of a cysteine residue at amino acid position 39 that was mutated to arginine in a pooled glandular tumor sample. Next, we tested whether the C39R mutation could affect interaction of B56γ with the PP2A A and C subunits as well as with several reported substrates for B56γ-PP2A including p53 (Li et al., 2007), ERK (Letourneux et al., 2006), and

Cyclin G2 (Bennin et al., 2002) by either immunoprecipitation (Figure 3.2A) or microcystin bead pull down (Figure 3.2B). Both assays indicate the C39R mutant while retaining binding to all substrates tested, lost interaction with the PP2A A and C subunits (Figure 3.2A and 3.2B). The

CD analysis of C39R confirms that the cysteine to arginine mutation has no effect on overall protein conformation (Figure S3.1). Together, these results suggest that the cancer-associated

C39R mutant disrupts B56γ-PP2A holoenzyme assembly.

We consider the possibility that cysteine to arginine mutation results in a larger side chain at the position of residue 39, which may clash with nearby helixes and disrupt the HEAT repeat 1 structure. As a consequence, structural change in HEAT repeat 1 may affect the stacking of the rest of the HEAT repeats and thus prevent the interaction of B56γ with the PP2A A and C subunits (Figure 3.2C), To test this hypothesis, we generated two mutations with smaller side chains, C39S and C39A, and tested whether these changes could rescue the interaction of B56γ with the AC core in vivo (Figure 3.2A and 3.2B). The assays show that both C39S and C39A mutants only slightly rescued the PP2A core interaction in U2OS cells, suggesting that the cysteine residue, but not just the size of its side chain, is specifically required for maintenance of proper HEAT repeat 1 conformation and for the PP2A AC core interaction. Because Cys39 is located in a hydrophobic region and is not involved in any S-S bridges it may play a role in holding adjacent helices together through van der Waals interactions with hydrophobic residues from helix-2 (Leu63 and Met66), -3 (Val85 and Met88) and -4 (Val127 and Phe130) (Figure

3.2C). Nevertheless, these data further support our finding of the importance of HEAT repeat 1 in

PP2A holoenzyme.

B56γ HEAT repeat 1 mutants abate B56γ-PP2A tumor suppressive function

Next, we tested the impact of HEAT repeat 1 mutants on B56γ-PP2A tumor suppressive function. We previously showed that B56γ-PP2A dephosphorylates p53 at Thr55, leading to p53 activation, induction of p53 transcriptional target p21, and cell growth arrest (Li et al., 2007). We thus first assessed the ability of the C39R mutant to promote p53 Thr55 dephosphorylation and p21 induction. As shown in Figure 3.3A, overexpression of wild type B56γ led to an efficient dephosphorylation of p53 at Thr55. As consequence of this dephosphorylation, p21 induction was

clearly observed. Overexpression of C39R, however, was unable to promote p53 Thr55 dephosphorylation and p21 induction, indicating that this mutant lost its ability to direct PP2A phosphatase activity toward substrate p53. Similarly, none of HEAT repeat 1 deletion mutants tested were able to promote Thr55 dephosphorylation or p21 induction, suggesting the importance of HEAT repeat 1 in directing B56γ-PP2A phosphatase activity toward p53. Because the C39R mutation was identified in a tumor sample, those results also indicated that this mutant might potentially contribute to a cancer phenotype by blocking B56γ-PP2A tumor suppressive function.

To test this directly, we assessed the effect of C39R on cell proliferation. Overexpression of wild type B56γ has been previously shown to inhibit cell proliferation in HCT116 cells in either p53-/- or p53+/+ background (Li et al., 2007). We thus transfected those cells with the C39R construct and compared its ability to inhibit cell growth to the wild type protein. As shown in

Figure 3.3B and S3.2, overexpression of wild type B56γ in the presence of p53 led to an approximately 40% decrease in cell number as compared with control empty vector after 120 h of cell growth, whereas in the absence of p53, overexpression of wild type B56γ had a decreased effect on cell proliferation, with a 20% decrease in cell number. By comparison, overexpression of C39R showed no significant difference from control (P = 0.75 and P = 0.57 respectively), suggesting that this mutant had no effect on blocking cell proliferation in both p53-/- and p53+/+ background. The presence of B56γ and p53 at the analysis was verified by immunoblotting

(Figure 3.3B). These results indicate that the C39R mutant was no longer able to inhibit cell proliferation regardless of p53 status. To further assess the role of the HEAT repeat 1 domain in inhibition of cell proliferation, we also overexpressed the HEAT repeat 1 deletion mutant ΔN40 and ΔN73 in HCT116 cells (Figure 3.3B and S3.2). Our results show that overexpression of those

mutants also led to no inhibition of cell proliferation, suggesting that the importance of the HEAT repeat 1 domain in PP2A tumor suppressive function.

In addition to blocking cell proliferation, overexpression of wild type B56γ has also been shown to inhibit anchorage-independent cell growth (Li et al., 2007). We thus tested the effect of

C39R on this activity of B56γ. As shown in Figure 3.4 and S3.2, overexpression of wild type

B56γ3 in HCT116 cells with p53 significantly decreased the number of colonies from approximately 375 colonies to 120 colonies, whereas in HCT116 cells lacking p53 overexpression of wild type B56γ3 decreased the number of colonies from 450 to 400 on the agar.

By comparison, overexpression of C39R showed no significant difference from control empty vector (P = 0.28 and P = 0.65 respectively), suggesting this mutant had no effect on colony formation in both cell lines. The presence of B56γ and p53 at the analysis was verified by immunoblotting (Figure 3.4B). These results indicate that C39R lost both p53-dependent and - independent tumor suppressive activity. Further, overexpression of the smallest HEAT repeat 1 deletion mutant ΔN40 also had no effect on colony formation in both cell lines (Figure 3.4 and

S3.2). Together these data confirm the importance of the HEAT repeat 1 domain in B56γ-PP2A tumor suppressive function. Our results of the functional consequences of a cancer-associated mutation in the B56γ gene also provide an additional mechanism to inactivate B56γ-PP2A in cancer.

Discussion

In this study, we show that the HEAT repeat 1 motif of B56γ plays a critical role in assembly of the B56γ-PP2A holoenzyme. Importantly, we identified a tumor-associated missense mutation, C39R, within the HEAT repeat 1 motif. Although this mutation was found in a pooled glandular tumor sample its effect on B56γ-PP2A tumor suppressive function had not been previously determined. We showed that the mutant protein, although retaining binding to its substrates, is no longer able to interact with the AC core. As a consequence, C39R lost its ability to support both p53-dependent and p53-independent B56γ-PP2A tumor suppressive function.

These results suggest a novel mechanism behind a cancer-associated loss of function mutation in the PP2A B56γ subunit gene and provide evidence for the importance of the HEAT repeat 1 motif in B56γ-PP2A tumor suppressive function.

Study of the crystal structure has revealed that the regulatory B56γ subunit has eighteen

α-helices with pairs of antiparallel α-helices forming eight pseudo HEAT repeat motifs stacking against each other. Previous studies have shown that the B56γ subunit is involved in extensive interactions with both A and C subunits mediated by a number of conserved residues mainly located in the loops connecting these HEAT repeats (Xu et al., 2006; Cho and Xu, 2007). In this study, we examined the role of HEAT repeat 1 in PP2A holoenzyme formation. Although the crystal structures have shown that HEAT repeat 1 is not directly involved in binding with either the A or C subunits, deletion of this region prevented the holoenzyme assembly both in vivo and in vitro (Figure 3.1). The importance of HEAT repeat 1 is also supported by the result that a single point mutation, C39R, within this region has also lost the AC core interaction. One possible explanation for this disruption is that HEAT repeat 1 (helix-1 and -2) may play a role in stabilizing HEAT repeat 2 (helix-3 and -4) which contains loop IL2 that is involved in interacting

with the A and C subunits. Examining the helices of the B56γ subunit indicates interactions among the helices within each HEAT repeat and between two adjacent HEAT repeats including

HEAT repeat 1 and HEAT repeat 2. Therefore, when HEAT repeat 1 is disrupted by mutations, the stacking of HEAT repeat 1 with HEAT repeat 2 may change, leading to a position shift of the interacting motif at the tip of IL2 in HEAT repeat 2 and therefore affecting the interaction of

B56γ with the PP2A AC core. In the case of C39R mutation, the cysteine residue at position 39 may play a role in holding adjacent helices together so that HEAT repeat 1 is in a proper position to stack with HEAT repeat 2 through van der Waals interactions with residues from helix-3 and -

4. This cysteine specific interaction network seems to be indispensable for the proper coordination of HEAT repeat 2, as mutation of cysteine to another residue also caused significant weakening of the interaction with the AC core. Indeed, our study reveals that neither alanine nor serine substitution at residue 39 could rescue the binding deficiency despite their side chain size being comparable with cysteine. These results suggest that point mutation at Cys39 rearranges the alignment of the helices between HEAT repeats 1 and 2, resulting in the position shift of the interacting motif at the tip of the extended IL2 loop and therefore weakening its interaction with the AC core. Naturally, arginine substitution introduces two dramatic changes as arginine side chain is larger in size and is positively charged. This has likely led to a much more significant repositioning of helix-3, forcing the tip of the extended IL2 to be pushed out of position. Together, those results suggest that Cys39 is important for stable HEAT repeat 1 and the proper positioning of HEAT repeat 2 helices, allowing B56γ to form a stable complex with the PP2A AC core.

Perhaps it is worthwhile to mention that the cysteine residue is conserved among all B56 family members, suggesting its importance in maintaining the rigid alignment of the helices for all B56-

PP2A holoenzyme assembly.

B56γ has been described to inhibit cell transformation in both p53-dependent and p53- independent manner (Li et al., 2007). Previously, we identified two B56γ mutations, F395C from lung cancer and A383G from intestinal tumor, which specifically disrupted B56γ-p53 interaction and thus specifically inactivated p53-dependent tumor suppressive function of B56γ (Shouse et al., 2010). However, because these mutants remain their ability to interact with the PP2A AC core, they continue to support p53-independent tumor suppressive function of B56γ (Shouse et al.,

2010). In this study, we identified a new cancer-associated mutant in the B56γ gene, C39R, which, unlike F395C and A383G, prevented B56γ from binding to the PP2A AC core. As a consequence, this mutant is no longer able to mediate any B56γ-mediated PP2A dephosphorylation and thus inactivated both p53-dependent and -independent tumor suppressive function of B56γ-PP2A.

These data thus provide a new, and perhaps more important, molecular basis for inactivating tumor suppressor B56γ in cancer. Given the fact that p53 is highly mutated in human cancers, this mechanism may have a broader impact on tumor suppressive function of B56γ. Interestingly, several mutations in PP2A A subunit have been reported to abolish interaction with B56 subunit in human carcinoma (Ruediger et al., 2001b; Chen et al., 2005b). Furthermore, several viral proteins such as simian virus SV40 ST antigen as well as polyoma virus ST and MT antigens have been reported to cause cancer by dissociating B56 subunits from the PP2A AC core (Chen et al., 2004, 2007). These data suggest that displacement of B56γ subunit from PP2A holoenzyme probably represents a more common mechanism for inactivation of tumor suppressive function

B56γ in cancer. It will be interesting to further investigate the role of HEAT repeat 1 in the displacement of B56γ from PP2A by oncoproteins.

References

Arnold, H.K., and Sears, R.C. (2006). Protein Phosphatase 2A Regulatory Subunit B56α Associates with c-Myc and Negatively Regulates c-Myc Accumulation. Mol. Cell. Biol. 26, 2832–2844.

Chen, W., Arroyo, J.D., Timmons, J.C., Possemato, R., and Hahn, W.C. (2005). Cancer- associated PP2A Aα subunits induce functional haploinsufficiency and tumorigenicity. Cancer Res. 65, 8183–8192.

Chen, Y., Xu, Y., Bao, Q., Xing, Y., Li, Z., Lin, Z., Stock, J.B., Jeffrey, P.D., and Shi, Y. (2007). Structural and biochemical insights into the regulation of protein phosphatase 2A by small t antigen of SV40. Nat. Struct. Mol. Biol. 14, 527–534.

Cho, U.S., and Xu, W. (2007). Crystal structure of a protein phosphatase 2A heterotrimeric holoenzyme. Nature 445, 53–57.

Csortos, C., Zolnierowicz, S., Bako, E., Durbin, S.D., and Depaoli-Roach, A.A. (1996). High Complexity in the Expression of the B Subunit of Protein Phosphatase 2A. J. Biol. Chem. 271, 2578–2588.

Eichhorn, P.J.A., Creyghton, M.P., and Bernards, R. (2009). Protein phosphatase 2A regulatory subunits and cancer. Biochim. Biophys. Acta 1795, 1–15.

Forester, C.M., Maddox, J., Louis, J. V, Goris, J., and Virshup, D.M. (2007). Control of mitotic exit by PP2A regulation of Cdc25C and Cdk1. Proc. Natl. Acad. Sci. U.S.A. 104, 19867–19872.

Letourneux, C., Rocher, G., and Porteu, F. (2006). B56-containing PP2A dephosphorylate ERK and their activity is controlled by the early gene IEX-1 and ERK. EMBO J. 25, 727–738.

Li, H.H., Cai, X., Shouse, G.P., Piluso, L.G., and Liu, X. (2007). A specific PP2A regulatory subunit, B56γ, mediates DNA damage-induced dephosphorylation of p53 at Thr55. EMBO J. 26, 402–411.

Ruediger, R., Pham, H.T., and Walter, G. (2001). Disruption of protein phosphatase 2A subunit interaction in human cancers with mutations in the Aα subunit gene. Oncogene 20, 10–15.

Shouse, G.P., Nobumori, Y., and Liu, X. (2010). A B56γ mutation in lung cancer disrupts the p53-dependent tumor-suppressor function of protein phosphatase 2A. Oncogene 29, 3933–3941.

Virshup, D.M., and Shenolikar, S. (2009). From promiscuity to precision: protein phosphatases get a makeover. Mol. Cell 33, 537–545.

Xu, Y., Xing, Y., Chen, Y., Chao, Y., Lin, Z., Fan, E., Yu, J.W., Strack, S., Jeffrey, P.D., and Shi, Y. (2006). Structure of the protein phosphatase 2A holoenzyme. Cell 127, 1239–1251.

Yeh, E., Cunningham, M., Arnold, H., Chasse, D., Monteith, T., Ivaldi, G., Hahn, W.C., Stukenberg, P.T., Shenolikar, S., Uchida, T., et al. (2004). A signalling pathway controlling c- Myc degradation that impacts oncogenic transformation of human cells. Nat. Cell Biol. 6, 308– 318.

Acknowledgements

We are very grateful to Dr. J.A. Traugh, Dr. F.M. Sladek and members of our laboratory for valuable suggestion and discussion. We thank Bob Zhao for assistance on constructing C39A plasmid and Dr. D. Borchardt for assistance on performing CD analysis. This work was supported by NIH grant CA0751 to X. Liu.

Figure legend

Figure 3.1 HEAT repeat 1 motif of B56γ is required for B56γ-PP2A holoenzyme assembly

(A) Diagram of amino acid sequence and eight HEAT repeats of B56γ. (B) Bacterially expressed and purified GST-wild type B56γ (WT) or HEAT repeat 1 deletion mutant ΔN73 was incubated with U2OS cell lysates and bound proteins were analyzed by western blot using PP2A A and C, p53, B56γ, and vinculin antibodies. (C) Comparison of CD spectra of wild type B56γ (WT) and

ΔN73 in the far UV region. (D) Lysates of U2OS cells transfected with HA-tagged wild type

B56γ or HEAT repeat 1 deletion mutants were immunoprecipitated with anti-HA antibody, then analyzed by western blot using PP2A A and C, HA, p53 and vinculin antibodies. All experiments were repeated three times and a representative of the data was shown. (E) The crystal structure of

B56γ-PP2A holoenzyme (adapted from Protein Data Bank, accession code 2NYM) is prepared using PyMOL. Parts of PP2A are displayed by spheres with subunit A in grey and subunit C in orange. Helices of B56γ are displayed by colored cylinders with HEAT repeat 1 in green, HEAT repeat 2 in red, and rest of HEAT repeats in blue.

Figure 3.2 The C39R mutation disrupts interaction of B56γ with A and C subunits.

(A) Lysates of U2OS cells transfected with empty-vector control (EV), HA-tagged wild type

B56γ (WT), C39R, C39A or C39S were immunoprecipitated with anti-HA antibody, then analyzed by western blot using PP2A A and C, HA, p53, Cyclin G, ERK and vinculin antibodies.

(B) Lysates of U2OS cells transfected with empty-vector control, HA-tagged wild type B56γ,

C39R, C39A, or C39S were bound to microcystin beads and analyzed by western blot using HA,

PP2A A and C, p53 and vinculin antibodies. All experiments were repeated three times and a

representative of the data was shown. (C) Stereo view of the interface among the first four helices of B56γ (adapted from Protein Data Bank accession code 2NYM) in which Cys39 (red) is mutated to arginine, alanine, or serine, respectively. The dots represent van der Waals spheres of residue 39. The figure is prepared using PyMOL.

Figure 3.3 HEAT repeat 1 mutants of B56γ fail to inhibit cell proliferation.

(A) Lysates of U2OS cells transfected with HA-tagged wild type B56γ (WT), C39R, C39A, C39S,

ΔN40, ΔN53 and ΔN73, or a control empty vector (EV) were analyzed by western blot for p53

Thr55 dephosphorylation, p21, HA-B56γ, p53, and vinculin. (B) Representatives of cell proliferation of the HCT116 human colon cancer cell lines transfected with HA-tagged wild type

B56γ, C39R, ΔN40, or ΔN73, or a control empty vector (EV). Number of cells present at the 120 h time point were normalized against the empty vector controls and plotted in a bar graph. Error bars show average ± SEM from triplicate plates in one representative experiment. Cell lysates were analyzed by immunoblotting of transfected HA-B56γ and p53 proteins.

Figure 3.4 HEAT repeat 1 mutants of B56γ fail to inhibit anchorage-independent cell growth.

(A) Representatives of anchorage-independent growth of HCT116 human colon cancer cell lines transfected with HA-tagged wild type B56γ (WT), C39R or ΔN40, or a control empty vector

(EV). (B) Colony numbers for anchorage-independent growth were counted and represented on a bar graph. The values are the averages ± SEM from three experiments. Cell lysates were analyzed by immunoblotting of transfected or endogenous B56γ, p53, and vinculin proteins.

Figure S3.1 Comparison of CD spectra of wild type B56γ (WT) and C39R in the far UV region.

(A) CD spectra of wild type B56γ3 (WT) or C39R mutant in the far UV region. (B) Bacterially expressed and purified wild type B56γ3, C39R and ΔN73 were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

Figure S3.2 Statistical analysis of cell growth and transformation data.

P-values for HCT116 cell growth experiments from Figure 3.3B (A) and anchorage-independent cell growth experiments from Figure 3.4B (B) were calculated by comparing the effect of expressing the various mutants either versus the empty vector (EV) control or versus the wild type B56γ3 (WT). Statistical significance was defined as a P < 0.01.

Figures

Figure 3.1 HEAT repeat 1 motif of B56γ is required for B56γ-PP2A holoenzyme assembly

Figure 3.2 The C39R mutation disrupts interaction of B56γ with A and C subunits.

Figure 3.3 HEAT repeat 1 mutants of B56γ fail to inhibit cell proliferation.

Figure 3.4 HEAT repeat 1 mutants of B56γ fail to inhibit anchorage-independent cell growth.

Figure S3.1 Comparison of CD spectra of wild type B56γ (WT) and C39R in the far UV region.

Figure S3.2 Statistical analysis of cell growth and transformation data.

CHAPTER 4

Characterization of B56γ tumor-associated mutations reveals mechanisms for inactivation of B56γ-PP2A

I am responsible for all of the figures in this chapter, except Figure 4.3C which was prepared by

Y. Wu.

Abstract

A subset of the hetero-trimeric PP2A serine/threonine phosphatases that contain B56, and in particular B56γ can function as tumor suppressors. In response to DNA damage, the B56γ subunit complexes with the PP2A AC core (B56γ–PP2A), and binds p53. This event promotes PP2A- mediated dephosphorylation of p53 at Thr55, which induces expression of p21, and the subsequent inhibition of cell proliferation and transformation. In addition to dephosphorylation of p53, B56γ–PP2A also inhibits cell proliferation and transformation by a second, as yet unknown, p53-independent mechanism. Here, we characterized a panel of B56γ mutations found in human cancer samples and cancer cell lines and showed that the mutations lost B56γ tumor suppressive activity by two distinct mechanisms; one is by disrupting interaction with the PP2A AC core and the other with B56γ–PP2A substrates (p53 and unknown proteins). For the first mechanism, due to the absence of the C catalytic subunit in the complex, the mutants would be unable to mediate dephosphorylation of any substrate and thus failed to promote both p53-dependent and p53– independent tumor suppressive function of B56γ-PP2A. For the second mechanism, the mutants lack specific substrate interactions and thus partially lost tumor suppressive function, i.e. either p53-dependent or p53-independent depending on which substrate-binding was affected. Overall the data provide new insights into the mechanisms for inactivation of tumor suppressive function of B56γ and further indicate the importance of B56γ-PP2A in tumorigenesis.

Introduction

The protein phosphatase 2A (PP2A) is a family of serine/threonine phosphatases that is involved in a multitude of cell-signaling pathways. PP2A exists either in the cell as a heterodimer of scaffolding A subunit and catalytic C subunit (the AC core), or as a heterotrimeric complex where the AC core additionally associates with one of the variable B subunits. The B subunits have four gene families based on sequence homology: the B (B55 or PR55), B’ (B56 or PR61), B”

(PR48/59/72/130), and B”’ (PR93/110). Each B subunit family contains two to five isoforms and many contain alternative splice variants. Binding of a specific B subunit determines diverse cellular localization and substrate specificities, allowing PP2A holoenzyme to have a diverse enzymatic activity in the cell (Eichhorn et al., 2009; Virshup and Shenolikar, 2009).

Recent evidence suggested that a subset of PP2A holoenzymes that contain B56 (B56-

PP2A), in particular B56γ (PPP2R5C) can function as tumor suppressor (Chen et al., 2004; Li et al., 2007). Although the underlying mechanism is not fully understood, B56γ-PP2A is known to dephosphorylate and regulate specific substrates involved in cellular functions. For example, dephosphorylation of tumor suppressor p53 at Thr55 activates p53, resulting in the induction of the CDK inhibitor p21, and inhibition of cell growth. However, in the absence of p53, B56γ-

PP2A can still reduce cell growth, suggesting that it dephosphorylates other unknown substrates that also play roles in tumor suppression (Li et al., 2007). Although the mechanism of the p53- independent function is unknown, additional proteins have been shown to interact with B56γ and potentially could be dephosphorylated by B56γ-PP2A. These proteins include the mitogen- activated kinase ERK (Letourneux et al., 2006), transcription co-factor p300 (Chen et al., 2005a), and centromeric cohesion recruited by Sgo1 (Kitajima et al., 2006; Riedel et al., 2006). Overall these studies suggest that B56γ-PP2A acts as a tumor suppressor by dephosphorylating specific target substrates to regulate their effects on cellular functions. In support of this view, some viral

oncoproteins function by displacing the B56 subunits from the AC core (Chen et al., 2004, 2007).

In addition, mutations in PP2A Aα gene (PPP2R1A) and Aβ gene (PPP2R1B) identified in cancers are known to lose interaction with either the C subunit or the B56 subunits (Calin et al.,

2000; Ruediger et al., 2001b; Tamaki et al., 2004; Chen et al., 2005b). Despite its importance in tumorigenesis, mechanisms for inactivation of B56γ-PP2A tumor-suppression by B56γ mutations are not well studied.

In this study, we characterized a panel of B56γ mutations previously identified in human tumor samples and cancer cell lines. Our results revealed three classes of mutations in the B56γ gene. The class I mutants, which includes A61V, A212T, S251R, E266R, P274T, H287Q and

P289S, could bind to both the AC core and p53, and displayed a tumor suppressive activity similar to wild type (WT) B56γ, suggesting these mutations have no effect on the B56γ tumor suppressor function. The class II mutants (C39R, E164K, Q256R and L257R) lost all B56γ-PP2A tumor suppressor activity, although still able to bind p53, we showed they could not complex with the AC core. In contrast, the class III mutants (S220N, A383G and F395C) had a partial tumor suppressor activity compared to WT B56γ and could complex with the AC core. Two mutations, A383G and F395C, fail to bind to p53, thus explaining their loss of p53-dependent function (Shouse et al., 2010). In contrast, S220N still bound and dephosphorylated p53 but reduced B56γ–PP2A tumor suppressive activity, indicating that it disrupts p53-independent mechanism although the involved substrate is unknown. These results provide mechanistic insight into the inactivation of tumor suppressive function of B56γ and further support the notion that multiple pathways are involved in B56γ-PP2A mediated tumor suppression.

Materials and Methods

Cell culture and plasmids

U2OS and HCT116 cells were cultured in McCoy’s 5A medium supplemented with 10% fetal calf serum. The B56γ point mutants were generated using the QuikChange Site-Directed

Mutagenesis Kit (Stratagene).

Western blot and immunoprecipitation

Whole cell extract was prepared by lysing the cells in a buffer containing 50 mM Tris-HCl (pH

8.0), 120 mM NaCl, 0.5% NP-40, 1 mM dithiothreitol, 2 μg/ml aprotinin and 2 μg/ml leupeptin.

Cell lysates were subjected to SDS-PAGE, then analyzed by western blot analysis using anti-p53

(DO1, Santa Cruz Biotechnology), anti-PP2A A subunit (Upstate), anti-PP2A C subunit (1D6,

Upstate), anti-p21 (Santa Cruz Biotechnology), anti-PP2A B56γ (Shouse et al., 2010), anti-HA

(12CA5), anti-ERK (Santa Cruz Biotechnology), anti-SGOL1 (ABNOVA), anti-cyclin G (Santa

Cruz Biotechnology), or anti-vinculin (VIN-11-5, Sigma) antibodies. For Thr55 dephosphorylation experiments, the cell lysate was immunoprecipitated with a phospho-specific antibody for phos-Thr55 (Ab202) and then immunoblotted with anti-p53 antibody (Li et al.,

2007). For interaction of endogenous proteins with transfected B56γ proteins, U2OS cells were transfected with various B56γ plasmids using Fugene 6 (Roche) or BioT (Bioland Scientific) and lysed 28 h after transfection. Immunoprecipitation was performed using anti-HA monoclonal antibody. The amounts of co-precipitated proteins were determined by immunoblotting.

RT-PCR

Total RNA was extracted using Trizol reagent (Sigma), and RT-PCR was performed using

SuperScript One-Step RT-PCR kit (Invitrogen) according to manufacturer’s protocol. RT-PCR

for p21 mRNA was performed with (F) 5’-CGACTGTGATGCGCTAATGG–3’ and (R) 5’-

GGCGTTTGGAGTGGTAGAAAT-3’, and for GAPDH mRNA was performed with (F) 5’-

AGGTGAAGGTCGGAGTCAAC-3’ and (R) 5’- GACAAGCTTCCCGTTCTCAG-3’.

Identification of cancer-derived mutation

The NCBI AceView program (http://www.ncbi.nlm.nih.gov/IEB/Research/Acembly/index.html) provides a comprehensive sequence of the human transcriptome and genes of all quality-filtered human complementary DNA data from GenBank, RefSeq, dbEST, and Trace in a strictly complementary DNA-supported manner. Using this program, we looked for B56γ mutations in the annotated sequences of tumor samples and cancer cell lines.

Cell proliferation and anchorage-independent growth assays

To generate proliferation curves for HCT116 cells, cells were transfected with WT, mutant B56γ or a control cytomegalovirus (CMV) empty vector using BioT. Transfected cells were seeded in triplicate, and then counted at 120 h post seeding. The presence of overexpressed B56γ protein in the cell was verified by immunoblotting. For anchorage-independent growth assays of HCT116 cells, cells transfected with WT, mutant B56γ or a control empty vector were seeded in triplicate in 0.35% Noble agar (Fisher) and colony numbers were counted 4 weeks post seeding.

Results

Identification of potential tumor-derived mutations in B56γ gene

To better understand the role of B56γ in human cancers, we performed an AceView search for all of the known, tumor-associated mutations in the B56γ gene. AceView regularly downloads the whole set of cDNA sequences from the public databases, aligns them on the current genome available at NCBI, and clusters them into reference transcripts. Because all of the identified alternative transcripts were originally cloned from cancer samples or cancer cell lines, they represent potential tumor-inducing mutations. We identified twenty-four point mutations scattered throughout the B56γ coding region (Table 4.1; Figure 4.1), whereas a search of the

NCBI single-nucleotide polymorphism (SNP) database did not yield any of these mutations. The fact that B56γ mutations are found in several different types of tumors (Table 4.1) suggests that

B56γ may play a broad role in tumor suppression. The B56γ protein consists of eight pseudo

Huntington-elongation-A subunit-TOR (HEAT) repeats. Interestingly, although the mutations are spread across the entire B56γ sequence, they cluster more frequently toward the center of the gene, notably on the HEAT-repeat 4, 5, 6 and in a very small domain (aa 383-410) that contains the p53 binding domain (Figure 4.1), suggesting that there are potential cancer mutation hot spots in the gene. The human B56γ transcript has at least three long splice variants known as 1, 2, and 3 (Muneer et al., 2002; Ortega-Lázaro and del Mazo, 2003) and most of the mutations we identified are common to all three variants. To begin investigating their function in tumor suppression, we used site directed mutagenesis to generate eleven of the new mutations (A61V,

E164K, A212T, S220N, S251R, Q256R, L257R, E266R, P274T, H287Q, and P289S; Figure 4.1).

All of these mutations, plus three previously reported C39R, A383G and F395C mutations, are shared by the three spliced isoforms and represent different clusters located in the B56γ gene

(Figure 4.1).

100

Effect of identified mutations on the tumor suppressive functions of B56γ

Because the mutations were identified in tumor samples, we assessed their effect on B56γ tumor suppressor activity. Previously, we showed that overexpression of WT

B56γ inhibits cell proliferation and anchorage-independent cell growth in both p53- dependent and p53-independent manner (Li et al., 2007). To evaluate the effect of the mutants, we first tested whether mutations affect the ability of B56γ to inhibit cell proliferation. Human colon cancer cells, HCT116 cells with either a p53-/- or p53+/+ background, were transfected with WT B56γ or each of the mutants. As shown in Figure

4.2A, overexpression of WT B56γ in the presence of p53 (HCT116 p53+/+ cells) led to approximately 45% decrease in cell number compared to the vector control after 120 h of cell growth. This decrease represented both p53-dependent and p53–independent inhibition. In contrast, in the p53-/- cells, overexpression of WT B56γ had a reduced effect on cell proliferation, with a 20% decrease in cell number, which represented the level of p53-independent inhibition.

In comparison, overexpression of B56γ mutants led to cell growth inhibition ranged from similar to WT, to partial reduction, to no inhibition at all. Based on their growth inhibition property, we classified all 14 mutants tested into one of three classes

(Table 4.2). Class I, including A61V, A212T, S251R, E266R, P274T, H287Q and P289S, had little or no effect on B56γ-mediated growth inhibition in both HCT116 p53+/+ cells and p53-/- cells, suggesting those individual single mutations have no effect on the B56γ tumor suppressor function. In contrast, Class II, including C39R, E164K, Q256R and

101

L257R, were unable to inhibit cell growth in both p53-/- and p53+/+ HCT116 cells

(highlighted in blue), suggesting these mutants lost their ability to block cell proliferation regardless of p53 status. Class III, including S220N, A383G and F395C, only partially inhibited cell proliferation compared to wild type B56γ (highlighted in red). As previously described, A383G and F395C show reduced inhibitory effect in p53+/+ cells but not in p53-/- cells. This can be explained by their inability to bind and dephosphorylate p53 (Shouse et al., 2010). Interestingly, S220N showed a partial inhibitory effect in p53+/+ cells, but not in p53-/- cells, suggesting that this mutant specifically lost the p53-independent tumor suppressor activity of B56γ-PP2A.

To provide further evidence, we tested the effect of the mutations on anchorage- independent cell growth. Based on the result from cell proliferation assay (Figure 4.2A), we assayed A212T and P274T from Class I, E164K, Q256R, L257R from Class II, and

S220N from Class III. HCT116 p53+/+ cells and p53-/- cells were transfected with WT

B56γ or the mutants, and seeded in soft agar. As shown in Figure 4.2B, overexpression of

WT B56γ in p53+/+ cells led to 65% reduction in the number of colonies compared to empty vector control, which represents both p53-dependent and p53-independent inhibition. In contrast, overexpression of WT B56γ in p53-/- cells only led to 18% decrease in colony numbers, which represents the p53-independent inhibition.

When Class I A212T and P274T mutants were overexpressed, no significant changes in the number of colonies were observed compared to WT, suggesting those mutants have no effect on anchorage-independent growth suppression of B56γ.

Overexpression of Class II E164K, Q256R and L257R mutants, however, completely

102

abolished WT B56γ-mediated anchorage-independent growth suppression in both p53-/- and p53+/+ cells, indicating that those mutations lost their ability to suppress anchorage- independent cell growth in both p53-dependent and –independent manner. Compared to

WT, Class III S220N mutant partially lost its ability to inhibit anchorage-independent growth in p53+/+ cells and completely lost its ability in p53-/- cells, suggesting that it specifically disrupts p53-independent function of B56γ. A383G and F395C from Class III were previously shown to specifically block p53-dependent function of B56γ (Shouse et al., 2010). Taken together, our results demonstrate that Class II and III cancer-associated

B56γ mutations disrupt, either completely or partially, tumor suppressive activity of B56γ.

B56γ mutations interfere with interaction with either the AC core or substrates

To understand the mechanisms for inactivation of B56γ tumor suppressive function, we next assayed the ability of the mutants to interact with the AC core and with p53. WT or mutant B56γ was expressed in U2OS cells, and their interaction with the AC core and p53 was assayed by immunoprecipitation. As shown in Figure 4.3A and summarized in Table 4.2, none of the eleven new mutations tested affect interaction of

B56γ with p53. This is perhaps not surprising because most of the mutations are not located near the mapped p53-binding domain (aa391-401). Furthermore, all Class I mutants (A61V, A212T, S251R, E266R, P274T, H287Q and P289S) showed little or no effect on the AC interaction (Figure 4.3), supporting the notion that these individual mutations do not affect the B56γ tumor suppressor function (Figure 4.2 and Table 4.2).

To further prove this, we examined the effect of these mutations on p53 Thr55

103

dephosphorylation and function. Results of four representatives (A61V, A212T, E266R and P274T) are shown in Figure 4.3B. Overexpression of these mutants led to efficient dephosphorylation of p53 at Thr55 and activation of the p53 transcription target p21 at levels similar to WT B563 (summarized in Table 4.2). Because p53 Thr55 is the only known residue that is directly dephosphorylated by B56γ-PP2A, we were unable to assess the effect of the mutations on other potential dephosphorylation by B56γ-PP2A. However, given their ability to fully support p53-independent tumor suppressive function (Figure

4.2), it is likely that Class I mutants fully support B56γ-PP2A dephosphorylation.

In contrast, Class II mutants (E164K, Q256R and L257R) remained bound to p53, but lost their ability to interact with the AC core (Figure 4.3A). In addition, C39R was also unable to bind to the AC core (Nobumori et al., 2012). We note that all residues in this class, C39, E164, Q256, and L257, are not making any direct contact to A or C subunits according to B56γ–PP2A crystal structure (Xu et al., 2006; Cho and Xu, 2007).

However, they are located in close proximity from the interaction interface (Figure 4.3D).

E164 residue is located within intra-loop of HEAT-repeat 3 and its negatively charged side chain is important to form hydrogen bond to E118 and R167. Mutation of E to K in this position would abolish these hydrogen bonds and thus destabilize intra-loop of

HEAT-repeat 2 that mediates interaction with A and C subunits. Q256 and L257 residues are located within the second helix of HEAT-repeat 5. The polar side chain of Q256 points toward a helix of HEAT-repeat 4 and interacts with E216. In Q256R mutation, arginine has a positively charged side chain that is larger than glutamine and additionally contacts E213. Alternatively, the hydrophobic side chain of L257 points toward the first

104

helix of HEAT-repeat 5. In L257R mutation, a larger side chain of arginine would protrude into the adjacent helix and a positive charge of arginine would induce further alterations in the environment. Both cases would result in displacement of helices, leading to rearrange the location of intra-loops that mediate the interaction between B56γ and the AC core. Interestingly, all four residues are conserved among the B56 family isoforms, indicating the importance of these residues for maintaining the interaction between B56 and the AC core.

Because the C subunit is required for PP2A catalytic activity, our data explain why the Class II mutants completely abolished all tumor suppressive function of B56γ-

PP2A (Figure 4.2 and Table 4.2). To support this view, we assayed their effect on p53

Thr55 dephosphorylation and induction of p21 (Figure 4.3B, 4.3C and Table 4.2). The assays showed that, unlike WT B56γ, overexpression of Class II mutants fails to induce dephosphorylation of p53 at Thr55 and activation of p53 transcription target p21. These results demonstrate that the interaction of B56γ with the AC core is required for B56γ-

PP2A to dephosphorylate and activate p53. Consequently, Class II mutants have lost their tumor suppressive function through disruption of the AC core interaction and loss of catalytic activity.

Previous study has shown that two mutants in Class III, A383G and F395C, remain bound to the AC core, but have lost their ability to bind and dephosphorylate p53, leading to disruption of the p53-dependent tumor suppressor activity of B56γ -PP2A

(Shouse et al., 2010). Interestingly, unlike A383G and F395C, S220N still interacted with p53 at a level comparable to WT B56γ. Importantly, the protein also promoted

105

dephosphorylation of p53 at Thr55 and induced p21 expression to levels similar to WT

B56γ (Figure 4.3B, 4.3C, and Table 4.2). These results suggest that p53 is unlikely responsible for partial loss of tumor suppressor activity of the S220N protein (Figure 4.2).

Together, our results show that Class III mutations inactivate the tumor suppressor activity of B56γ-PP2A by preventing B56γ from binding to its substrates, thereby indicating the importance of B56γ in recruiting the AC core to substrate, so that B56γ-

PP2A can function correctly.

S220N binds B56γ interacting proteins

Because S220N specifically abolished the p53-independent tumor suppressor activity of

B56γ-PP2A, we hypothesize that this may be due to its lack of interaction with another unknown substrate. Interestingly, S220 is located on a large concave surface of B56γ that is unoccupied by

A and C subunits and leans toward the catalytic pocket of the C subunit (Figure 4.4A). It has been previously suggested that this open area may be important for recruiting substrates (Xu et al.,

2006). To identify potential substrate that may bind to wild type B56γ but not S220N, we assayed the ability of the S220N mutant protein to interact with several known B56γ interacting proteins including ERK, Cyclin G2, Shugoshin and p300 (Figure 4.4B and data not shown). The assay showed that the interaction of S220N with all of these proteins was similar to WT B56γ (Figure

4.4B), suggesting that interactions with these proteins are unlikely responsible for loss of p53- independent tumor suppressor activity of S220N. Nevertheless, our results indicate that Class III mutations specifically abolished individual substrate interaction, thus partially disrupting B56γ tumor suppressor activity. Furthermore, our results also indicate that B56γ contains multiple substrate-binding domains, implying the role of a specific B subunit in multiple pathways.

106

Further study will provide insight into these pathways and their role in PP2A-dependent tumor suppression.

107

Discussion

In this study, we characterized a panel of B56γ mutations that were previously identified in human cancers and defined the molecular mechanisms behind loss of B56γ-PP2A tumor suppression. In general, the mutations could be categorized into three groups, “no effect”, “loss of the AC core interaction” and “loss of substrate interaction”, based on how they inactivated B56γ function (Figure 4.5). The “loss of the AC core interaction” group or Class II mutations (C39R,

E164K, Q256R and L257R) failed to bind to the AC core and thus disrupted the B56γ-PP2A complex in the cells. This leads to the loss of all B56γ-PP2A tumor suppressor-related functions.

As a result, overexpression of these mutants failed to promote p53-dependent and p53- independent tumor-suppression in HCT116 p53+/+ cells and p53-independent tumor-suppression in HCT116 p53-/- cells. Interestingly, the “loss of the AC core interaction” mechanism has been observed with several previously described A subunit tumor-associated mutations and SV40 ST antigen, emphasizing the importance of the B56γ-PP2A complex in cancer suppression. In contrast, the “loss of substrate interaction” group or Class III mutations (S220N, A383G and

F395C) remain bound to the AC core, but failed to bind B56γ-PP2A substrates. The A383G and

F395C mutations are located at the p53-binding domain, and thus caused the loss of p53 interaction, which specifically abolished p53-dependent B56γ tumor suppressor function. In contrast, the S220N mutation is located at another potential substrate binding domain. Although we have not identified the substrate(s) involved in this loss of B56γ-PP2A function, we hypothesize that an unknown protein(s) plays a role in p53-independent tumor suppression.

Taken together, our data identified detailed mechanisms for the inactivation of B56γ-PP2A in cancer, and suggest possible avenues for targeted therapy against cancers caused by loss of B56γ-

PP2A function.

108

Although the B56γ mutations identified so far spread across the entire B56γ sequence, they appear more frequently in two regions: the center of the B56γ gene (HEAT-repeat 4, 5, and

6) and a small region toward the C-terminus (Figure 4.1). These clusters suggest possible existence of hot spots that are more susceptible to tumor-associated mutations. The existence of hot spots for somatic mutations often indicates that the region is essential for the function of protein. Indeed, the small region at the C-terminus (aa 383-401) contains a domain crucial for p53 binding (Shouse et al., 2010), arguing that p53 is an important substrate for B56γ-PP2A function.

Interestingly, although eleven mutations are located on HEAT-repeat 4, 5, and 6 alone, the majority of them did not affect the B56γ ability to inhibit cell growth. Aside from sequencing errors, it is possible that more than one point mutation is needed to change the protein function supported by those repeats. In fact, multiple interacting interfaces have been suggested for the complex formed between B56γ and the A and C subunits (Xu et al., 2006; Cho and Xu, 2007).

Further study of this region may lead to a better understanding of those interfaces and their roles in PP2A function.

109

References

Chen, J., St-Germain, J.R., and Li, Q. (2005a). B56 Regulatory Subunit of Protein Phosphatase 2A Mediates Valproic Acid-Induced p300 Degradation. Mol. Cell. Biol. 25, 525–532.

Chen, W., Arroyo, J.D., Timmons, J.C., Possemato, R., and Hahn, W.C. (2005b). Cancer- associated PP2A Aα subunits induce functional haploinsufficiency and tumorigenicity. Cancer Res. 65, 8183–8192.

Cho, U.S., and Xu, W. (2007). Crystal structure of a protein phosphatase 2A heterotrimeric holoenzyme. Nature 445, 53–57.

Eichhorn, P.J.A., Creyghton, M.P., and Bernards, R. (2009). Protein phosphatase 2A regulatory subunits and cancer. Biochim. Biophys. Acta 1795, 1–15.

Kitajima, T.S., Sakuno, T., Ishiguro, K., Iemura, S., Natsume, T., Kawashima, S.A., and Watanabe, Y. (2006). Shugoshin collaborates with protein phosphatase 2A to protect cohesin. Nature 441, 46–52.

Letourneux, C., Rocher, G., and Porteu, F. (2006). B56-containing PP2A dephosphorylate ERK and their activity is controlled by the early gene IEX-1 and ERK. EMBO J. 25, 727–738.

Li, H.H., Cai, X., Shouse, G.P., Piluso, L.G., and Liu, X. (2007). A specific PP2A regulatory subunit, B56γ, mediates DNA damage-induced dephosphorylation of p53 at Thr55. EMBO J. 26, 402–411.

Nobumori, Y., Shouse, G.P., Fan, L., and Liu, X. (2012). HEAT repeat 1 motif is required for B56γ-containing protein phosphatase 2A (B56γ-PP2A) holoenzyme assembly and tumor- suppressive function. J. Biol. Chem. 287, 11030–11036.

110

Ruediger, R., Pham, H.T., and Walter, G. (2001). Disruption of protein phosphatase 2A subunit interaction in human cancers with mutations in the Aα subunit gene. Oncogene 20, 10–15.

Shouse, G.P., Nobumori, Y., and Liu, X. (2010). A B56γ mutation in lung cancer disrupts the p53-dependent tumor-suppressor function of protein phosphatase 2A. Oncogene 29, 3933–3941.

Tamaki, M., Goi, T., Hirono, Y., Katayama, K., and Yamaguchi, A. (2004). PPP2R1B gene alterations inhibit interaction of PP2A-Aβ and PP2A-C proteins in colorectal cancers. Oncology Reports 11, 655–659.

Virshup, D.M., and Shenolikar, S. (2009). From promiscuity to precision: protein phosphatases get a makeover. Mol. Cell 33, 537–545.

Xu, Y., Xing, Y., Chen, Y., Chao, Y., Lin, Z., Fan, E., Yu, J.W., Strack, S., Jeffrey, P.D., and Shi, Y. (2006). Structure of the protein phosphatase 2A holoenzyme. Cell 127, 1239–1251.

111

Acknowledgements

We thank Dr. Margaret Morgan for careful editing of this manuscript. This work was supported by the NIH grant R01CA075180 to X. Liu.

112

Figure legend

Figure 4.1 Distribution of B56γ mutations identified by AceView.

Distribution of twenty-four mutations on three splice variants, known as γ1, 2, and 3, of B56γ.

The structure of B56γ is shown below. Each α-helix is indicated by a box and HEAT-repeat is represented by a shaded box. Mutants in bold were chosen for characterization.

Figure 4.2 Effect of B56γ mutants on tumor suppressor function.

HCT116 human colon cancer cells with p53+/+ or p53-/- background were transfected with HA- tagged WT or B56γ mutants. For the control (EV), cells were transfected with an empty cytomegalovirus (CMV) vector. (A) Representatives of cell proliferation assay where transfected cells were seeded and harvested and counted after 120 h of growth. Numbers of cells were normalized against the representative empty vector controls and plotted in a bar graph. Error bars show average ± s.d. from triplicate plates in one representative experiment. The experiments were repeated at least three times. Cells harvested were lysed and protein expression for B56, p53, and vinculin (vinc) were analyzed by western blot. (B) Representatives of anchorage- independent growth assay where transfected cells were seeded in soft agar and number of colonies were counted. Error bars show average ± s.d. from triplicate plates in one representative experiment. Cells at initial seeding were lysed and analyzed for B56 protein expression.

Figure 4.3 Interaction of B56γ mutants with PP2A A and C and p53.

U2OS cells were transfected with empty vector control (EV), HA-tagged WT or mutant B56γ.

(A) WT and mutant B56γ were immunoprecipitated and interacting proteins were analyzed by western blot using antibodies listed. (B) p53 Thr55 dephosphorylation, p21 protein levels, p53, and vinculin (vinc) were analyzed by western blot. p53 Thr55 phosphorylation levels were

113

analyzed by phospho-specific antibody for Thr55 in presence of MG132. The p21 protein levels were tested in the absence of MG132. (C) p21 mRNA levels were analyzed by RT-PCR. (D)

Class II mutations are shown on the crystal structure of B56γ-PP2A holoenzyme (adapted from

Protein Data Bank, accession code 2NYM), prepared by PyMOL. The PP2A holoenzyme is displayed with subunit A in green, subunit C in yellow, and subunit B56γ in blue. Microcystin-LR

(MCLR) is shown in magenta. Mutation residues are indicated by orange for C39, lime-green for

Q256, and red for L257. Dashed lines indicate hydrogen bonds.

Figure 4.4 S220N interacts with the AC core and p53.

(A) S220N mutation residue is indicated by the color red on the crystal structure of B56γ-PP2A holoenzyme. Is displayed with(B) Lysates of U2OS cells that were transfected with empty vector control (EV), HA-tagged WT or S220N B56γ were immunoprecipitated with anti-HA antibody, then analyzed by western blot against PP2A A and C, ERK, p53, Sgo, Cyclin G, HA, and vinculin (vinc).

Figure 4.5 Inactivation of B56γ-PP2A tumor suppressive function by B56γ mutations.

The Class II mutations C39R, E164K, Q256R and L257R failed to bind to the AC core and thus disrupted all B56γ-PP2A tumor suppressor-related functions. The Class III mutations (S220N,

A383G and F395C) failed to bind B56γ-PP2A substrates and thus partially lost tumor suppressive function of B56γ-PP2A, i.e. either p53-dependent or p53-independent depending on which substrate-binding was affected.

114

Figures

Figure 4.1 Distribution of B56γ mutations identified by AceView.

115

Figure 4.2 Effect of B56γ mutants on tumor suppressor function.

116

Figure 4.3 Interaction of B56γ mutants with PP2A A and C and p53.

117

Figure 4.4 S220N interacts with the AC core and p53.

118

Figure 4.5 Inactivation of B56γ-PP2A tumor suppressive function by B56γ mutations.

119

Tables

Amino acid B56γ HEAT- Tumor origin Reference (protein) isoforms repeat* C39R γ1, 2, 3 Pooled glandular 1 Nobumori et al., 2012 A61V γ1, 2, 3 Pooled glandular 1 This study L157P γ1, 2, 3 Pooled glandular 3

E164K γ1, 2, 3 Pooled glandular 3 This study R169K γ1, 2, 3 Pooled glandular 3

A212T γ1, 2, 3 Melanotic melanoma skin 4 This study S220N γ1, 2, 3 Leiomyosarcoma uterus 4 This study H233P γ1, 2, 3 Embryonal carcinoma 5

S251R γ1, 2, 3 Embryonal carcinoma 5 This study Q256R γ1, 2, 3 Melanotic melanoma skin 5 This study L257R γ1, 2, 3 Melanotic melanoma skin 5 This study L265F γ1, 2, 3 Embryonal carcinoma 5

E266R γ1, 2, 3 Embryonal carcinoma 5 This study P274T γ1, 2, 3 Large cell carcinoma lung 6 This study H287Q γ1, 2, 3 Pooled glandular 6 This study P289S γ1, 2, 3 Pooled glandular 6 This study A383G γ1, 2, 3 Ilea mucosa 8 Shouse et al., 2010 F395C γ1, 2, 3 Lung carcinoma 9 Shouse et al., 2010 Q401R γ1, 2, 3 Pooled germ cell tumor 9

E409R γ1, 2, 3 Pooled germ cell tumor 9

K410R γ1, 2, 3 Pooled germ cell tumor 9

S440I γ1, 2, 3 Lung carcinoma

H494P γ2, 3 Embryonal carcinoma

T495H γ2, 3 Embryonal carcinoma

*location on the B56 protein

Table 4.1 List of tumor-derived B56γ mutations.

120

Interaction Cell Growth Inhibition p53 Thr55 p21 B563 PP2A p53 p53 p53 dephosphorylation Induction AC core dependent independent WT Y Y Y Y Y Y C39R N Y N N N N A61V Y Y Y Y Y Y E164K N Y N N N N A212T Y Y Y Y Y Y S220N Y Y Y Y Y N S251R Y Y Y Y Y Y Q256R N Y N N N N L257R N Y N N N N E266R Y Y Y Y Y Y P274T Y Y Y Y Y Y H287Q Y Y Y Y Y Y P289S Y Y Y Y Y Y A383G Y N N N N Y F395C Y N N N N Y

Table 4.2 Summary of fourteen tumor-derived B56γ mutations’ tumor suppressive functional assays.

121

CHAPTER 5

Identification of B56γ-PP2A substrates responsible for its p53-independent tumor suppressive function

122

Introduction

Protein phosphatase 2A (PP2A) is a family of serine/threonine phosphatases that plays a role in the regulation of diverse signal transduction cascades and cellular processes. PP2A exists in the cell as a heterodimer consisting of scaffolding A subunit and catalytic C subunit (AC core), or as a heterotrimeric complex where the AC core additionally associates with one of the variable

B subunits. The multitude of PP2A function is believed to be accomplished by members of various regulatory B subunits, which regulate the substrate specificity, cellular localization, and enzymatic activity of the PP2A holoenzymes (Virshup and Shenolikar, 2009).

Several evidence indicates that a subset of the PP2A holoenzymes that contain B56 subunit (B56-PP2A), and in particular B56 (PPP2R5C), has tumor suppressive functions.

Although the underlying mechanism has not been fully elucidated, B56 is thought to recruit particular oncogenes and tumor suppressors to the AC core for dephosphorylation and regulate their cellular function (Eichhorn et al., 2009). For example, B56γ inhibits cell transformation by recruiting p53 to the AC core and dephosphorylating p53 at Thr55, leading to p53 activation, induction of Cdk inhibitor p21, and inhibition of cell proliferation. In addition, B56γ-PP2A can inhibit cell proliferation in the absence of p53, suggesting that it may dephosphorylate additional substrates that also play roles in tumor suppression (Li et al., 2007). The mechanism and proteins associated with this p53-independent function are currently unknown.

In chapter 4, we described the characterization of a panel of tumor-associated B56γ mutations. During the course of study, we found the S220N mutation which retained its p53- dependent tumor suppressive function, but lost its p53-independent tumor suppressive function.

Because this mutant can bind the AC core and p53 comparable to the wild type (WT) B56, we hypothesize that its loss of tumor suppressive activity is due to its inability to bind other unknown substrate(s).

123

In this study, we attempted to identify the unknown substrate(s) of B56γ–PP2A involved in p53-independent tumor suppressive function. We first established tetracycline-inducible U2OS stable cell lines that express either HA-WT B56γ or HA-S220N. We then carried out affinity purification of WT B56γ or S220N complexes and precipitated proteins were identified by LC-

MS/MS analysis. In addition to known proteins that interact with both WT B56γ and S220N, our analysis identified a series of proteins that associate with WT B56γ but not S220N. As such, they may potentially mediate p53-independent tumor suppressive function of B56γ–PP2A.

Interestingly, some proteins are known to function in cellular processes involved in cell transformation and here we discuss some of the proteins’ functions and implications in tumor suppressive function of B56γ–PP2A.

124

Materials and Methods

Cell culture and plasmids

U2OS cells were cultured in McCoy's 5A supplemented with 10% fetal calf serum. pcDNA6/TR and pcDNA4/TO vectors were obtained from Invitrogen. B56γ3 with N-terminus 4x HA epitope tag plasmid (pCEP4-HA-B56γ3) has been previously described (Li et al., 2007). S220N point mutant was generated using the QuikChange Site-Directed Mutagenesis Kit (Stratagene). pcDNA4/TO-HA-B56γ3 plasmid was obtained by excising HA-B56γ3 DNA fragment from pCEP4-HA-B56γ3, and then inserting into pcDNA4/TO vector by digestion with KpnI and XbaI restriction enzymes (Figure 5.1). All plasmids were verified by sequencing.

Cell lines

U2OS cells were transfected with pcDNA6/TR and pcDNA4/TO-HA-B56γ3 or S220N plasmids using BioT (Bioland Scientific). Stably transfected cell lines resistant to 5 μg/ml blasticidin

(A.G.Scientific) and 200 µg/ml zeocin (Invitrogen) were selected, and individual colonies were isolated and expanded. To induce HA-B56γ protein expression, cells were treated with 0.5 μg/ml tetracycline (Sigma) for 12 h before harvesting.

Purification of HA-B56γ complexes by immunoprecipitation

Whole cell extract was prepared by lysing cells in a lysis buffer containing 50 mM Tris–HCl (pH

8.0), 120 mM NaCl, 0.5% NP-40, 1 mM dithiothreitol, 2 μg/ml aprotinin and 2 μg/ml leupeptin.

For HA IP, U2OS lysate was incubated with anti-HA (12CA5) monoclonal antibody conjugated

125

Protein A sepharose beads (Pierce) for 6 h at 4 °C. Double affinity purification of HA-B56γ was performed by HA-IP followed by B56γ IP. After HA-IP, beads were washed once in lysis buffer and 100 mM KCl D buffer containing 20 mM HEPES (pH=7.9), 20% glycerol, 100 mM KCl, 0.2 mM EDTA, and 1 mM dithiothreitol. Bound proteins were eluted with twice the beads volume of

1 mg/ml HA peptide in 100 mM KCl D buffer for 30 m at room temperature, repeated twice. The elutes were combined and diluted 10-fold in lysis buffer, and incubated with B56γ antibody and

Protein A beads for 4 h at 4 °C. Beads were washed twice with lysis buffer, boiled in SDS- loading buffer, and analyzed by immunoblotting. For LC-MS/MS analysis, U2OS lysate from ten

10-cm-diameter dishes yielding approximately 4 mg of total protein was incubated with anti-HA antibody conjugated beads for 6 h at 4 °C. Beads were washed three times with lysis buffer and twice with 300 mM KCl D buffer. The co-precipitated proteins were either visualized by silver staining, detected by immunoblotting, or analyzed by MS.

Western blot

Cell lysates were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-

PAGE), followed by immunoblotting analysis with anti-p53 (DO1, Santa Cruz Biotechnology), anti-p21 (c-19, Santa Cruz Biotechnology), anti-PP2A A subunit (Upstate), anti-PP2A C subunit

(1D6, Upstate), anti-PP2A B56γ (against full-length B56γ3), anti-HA (12CA5), or anti-vinculin

(VIN-11-5, Sigma) antibodies.

Mass spectrometry analysis

126

Tryptic digests of proteins co-purified with HA-B56γ3 were identified by LC/ESI/MS/MS and search in the NCBI database was done using Mascot MS/MS ion search program (Matrix

Sciences Ltd.). Lists of significant “hit” proteins that have at least one peptide ion scoring higher than 20 (p < 0.05) were generated. Proteins identified in control mock-purified samples were considered as background and were not included in the interaction results. Protein abundance index (PAI) is the number of detected peptides per protein normalized by the theoretical number of peptides. The exponentially modified protein abundance index (emPAI) indicates approximate, relative quantitation of the proteins in a mixture based on protein coverage by the peptide matches in a database search result (Ishihama et al., 2005).

127

Results

Establishment of the tetracycline-inducible B56γ U2OS cell lines

To identify proteins that specifically lost interaction with S220N, we created tetracycline- inducible U2OS stable cell lines that express high levels of either HA-WT B56γ or HA-S220N protein upon tetracycline treatment. Tetracycline-inducible system allows efficient and rapid induction of target protein and facilitates preparing samples in larger scale. The cell lines were prepared by stable transfection of U2OS cells with tetR expression plasmid and either HA-WT or

HA-S220N plasmid. tetR expression plasmid contains tetR sequence downstream of cytomegalovirus (CMV) promoter along with the blasticidin resistance gene under control of EM-

7 promoter. B56γ expression plasmids contained B56γ3 gene with HA epitope tag on the N- terminus downstream of CMV promoter containing two tetO sequence as well as the zeocin resistance gene under control of EM-7 promoter (Figure 5.1). Individual colonies of stably transfected cell lines resistant to blasticidin and zeocin were selected and expanded. All of the selected colonies of WT and S220N induced high level of HA-B56γ protein. In order to confirm that expressed B56γ protein is functional, p21 expression level was analyzed for several of the isolated colonies (Figure 5.2A). Upon western blotting analysis, WT #4 and S220N #6 colonies were selected for the following experiments. For those cells, we also observed similar levels of dephosphorylation of p53 at Thr55 (data not shown). Next, induction efficiency was tested for various tetracycline dosage and treatment time. As shown in Figure 5.2B, 1 μg/ml treatment promoted slightly higher expression of B56γ and p21 induction compared to 0.5 μg/ml treatment.

Detection of efficient upregulation of p21 protein suggests that tetracycline-induced B56γ protein is functional and presumably capable of interacting with B56γ binding proteins. When cells were treated with 0.5 μg/ml of tetracycline, gradual increase in the protein expression level of B56γ was observed in association with increasing treatment time in the cell and reached its maximum

128

approximately at 10 h. On the other hand, increase in the p21 protein expression is not apparent until 18 h treatment (Figure 5.2C). All together, the result suggests that 0.5 μg/ml of tetracycline treatment for 18 h is sufficient for B56γ protein to be induced, to dephosphorylate p53 at Thr55 and promote p21 upregulation.

Optimizing conditions for affinity purification of B56γ-PP2A

We next sought to find the optimal immunoprecipitation conditions to purify B56γ binding proteins using the established cell lines. First, we assessed the adequate amount of HA- antibody-conjugated beads needed to immunoprecipitate B56γ from the cell lysate. We prepared the whole cell lysate from tetracycline treated cells and performed HA IP with varying volume of beads and detected the amount of B56γ protein bound to the beads as well as the co-precipitating

A and C subunits. The assay indicates that 5 to 10 μl of conjugated beads would achieve the most efficient immunoprecipitation of B56γ-PP2A from cell lysate of a 10-cm-diameter dish (Figure

5.3A). Second, we tested different wash conditions with increased amount of salt in an attempt to reduce the nonspecific protein binding while maintaining specific protein interaction of B56γ intact. B56γ bound HA-beads were washed with either the lysis buffer containing 0.5% NP-40 and 120 mM NaCl (Wash 1), D buffer containing 300 mM KCl (Wash 2), or D buffer containing

500 mM KCl (Wash 3). High concentration of salt in Wash 3 allowed the precipitate with less nonspecific proteins as shown by the silver staining (Figure 5.3B). However, although the B56γ association between A and C subunits was not affected, its interaction with p53 was significantly compromised for Wash 3 (Figure 5.3C). Therefore, D buffer containing 300 mM KCl was used for the wash for the following experiments in order to avoid losing B56γ specific interaction.

129

We also attempted to purify B56γ complexes using double affinity purification where HA

IP and B56γ IP were performed in tandem. As shown in Figure 5.3D, B56γ complexes were first immunoprecipitated from cell lysate with HA-beads and then eluted from HA-beads using HA peptides. The eluted B56γ complexes were further purified by immunoprecipitation with anti-

B56γ antibody. The assay shows that significant portion of B56γ protein remained bound to HA- beads after eluting with HA peptides (Figure 5.3D, left panel). Moreover, overall protein yield of

B56γ and C subunit after double affinity purification was small (Figure 5.3D, right panel). This suggests that majority of HA-B56γ protein complex is lost during the elution step with HA peptides and thus a very low amount of B56γ protein complex is recovered. Since LC-MS/MS analysis requires a large quantity of purified proteins, it would be necessary to modify double affinity purification method and increase the overall protein recovery.

Mass Spectrometry Analysis

Based on the above results, we performed HA IP with optimized conditions (Figure 5.4) and subjected WT and S220N complexes to LC-MS/MS (minimum of one peptide with a

MASCOT score of >20 and p < 0.05). In parallel, mock purification and LC-MS/MS analysis were performed from parental U2OS cells treated with tetracycline as a control. The list of identified proteins containing those with highest number of spectral counts indicated by emPAI is summarized in Table 5.1. Among those proteins, 17 proteins were detected in both WT and

S220N protein complexes (Table 5.2). Interestingly we successfully detected A and C subunits with both WT and S220N, however, did not detect p53. This is likely due to low amount of endogenous p53 protein in the sample and/or low ionization efficiency of the sample peptides

130

could have caused them to be undetected. Nevertheless, a list of 26 proteins that interact with only WT but not S220N is summarized in Table 5.3 and several of them are discussed below.

BMPR-II is a serine/threonine receptor kinase of the BMP signaling pathway. BMP ligands are members of the TGF-β superfamily that bind to two types of transmembrane receptor kinases, BMPR-I and BMPR-II. Upon BMP binding, BMPR-II phosphorylates and activates

BMPR-I, leading to the activation of downstream signaling molecules through Smad-dependent and –independent pathways (Miyazono et al., 2005). Originally discovered as an inducer of bone formation, following studies have revealed the multipurpose role of BMPs as cytokines that regulate cell growth, apoptosis and differentiation in a number of cell types (Von Bubnoff and

Cho, 2001; Zwijsen et al., 2003). It has been recently shown that the BMB pathway is disrupted in majority of colorectal (Kodach et al., 2008) and prostate cancer (Kim et al., 2004). Moreover, tumor suppressive function of BMPR-II to inhibit cell growth inhibition has been described, indicating a link between the BMP signaling and tumorigenesis (Pouliot et al., 2003; Beppu et al.,

2008; Owens et al., 2012). Several phosphorylation sites on BMPR-II have been reported although the mechanism of BMPR-II regulation remains largely unknown. If B56γ indeed interacts with BMPR-II, B56γ–PP2A may directly dephosphorylate BMPR-II to affect its function. Interestingly, another PP2A B subunit, B55β, has been shown to interact with BMPR-II and dephosphorylates Smads, positively regulating the BMP signaling (Bengtsson et al., 2009). It will be interesting to investigate in the future whether B56γ also plays a role in BMP signaling and affect the rate of cell proliferation.

Liprin α-1 and α-2 have been previously reported to interact with B56γ although functional significance of the interaction remains to be elucidated (Arroyo et al., 2008;

Goudreault et al., 2009). Liprins are cytoplasmic dimeric scaffolding proteins that have two

131

isoforms, α and β. In humans, liprin α has four genes where α-1 is ubiquitously expressed and α-2,

3, and 4 are predominantly expressed in neuronal tissues (Serra-Pagès et al. 1998). In neuronal cells, liprins have been shown to function in regulation of nervous system including lymphatic vessel development, synaptic development and activity (Astigarraga et al. 2010; Choe et al. 2006;

Dai et al. 2006; Olsen et al. 2005; Wyszynski et al. 2002; Zhen and Jin 1999). In non-neuronal cells, liprin α is involved in cell adhesion, cell motility, and cytoskeleton organization and has been shown to bind proteins such as LAR, GIT1, and CASK (de Curtis 2011; Ko et al. 2003; Shin et al. 2003; Stryker and Johnson 2007). Phosphorylation of several serine residues of liprin α has been shown to promote its association with LAR, leading to translocation of LAR and disassembly of focal adhesions (Serra-Pagès et al. 2005). Whether phosphorylation also affects the binding of liprin α with other proteins is unclear, but this suggests the possible regulation of liprin α activity through dephosphorylation by B56γ-PP2A. Alternatively, liprins may act as scaffold, forming complex and recruiting B56γ-PP2A to specific proteins at the plasma membrane. Liprin α is has been shown to regulate tumor cell invasion and migration in breast cancer cells and liprin α-1 gene is often amplified in several human cancers (Shen et al., 2007;

Astro et al., 2011). As such, B56γ potentially have a tumor suppressive function through regulating liprin α activity.

Rho GEF-H1 and Rho GAP-p190A are the regulators of small GTPase oncogene RhoA that influences the cell proliferation, cytoskeleton reorganization, and cellular invasion (Sahai and

Marshall, 2002; Aznar et al., 2004). GEF-H1, a positive regulator of RhoA, is a conserved, ubiquitously expressed protein. GEF-H1 activity is regulated through microtubule association, thus linking the microtubular cytoskeletal dynamics with Rho dependent regulation (Krendel et al., 2002; Matsuzawa et al., 2004). Emphasizing its role as an oncogene, cancer-associated mutant p53 increases GEF-H1 transcription, promoting cell viability (Mizuarai et al., 2006). Post-

132

translational phosphorylation is believed to regulate GEF-H1 expression and function, and several kinases including ERK1/2, Aurora A/B and CDK1 are known to phosphorylate GEF-H1

(Birkenfeld et al., 2008; Fujishiro et al., 2008). In contrast, ubiquitously expressed GAP-p190A negatively regulates RhoA activity and inhibits RhoA induced cell transformation (Bernards,

2003; Wang et al., 2003; Kusama et al., 2006). GAPs are also regulated by various kinases including GSK-3β (Bernards and Settleman, 2004; Jiang et al., 2008). If B56γ indeed binds and dephosphorylates GEF-H1 and GAP-p190A, it may block tumorigenesis by negatively regulating the activation of oncogenic RhoA.

In summary, our LC-MS/MS analysis provided a variety of proteins that interact and thus potentially possess B56γ-mediated function in addition to previously known proteins. Several proteins were detected with only WT but not with S220N, implying that they could potentially be responsible for the p53-independent function of B56γ-PP2A. However, data have to be analyzed with caution considering that binding of many of these proteins actually may not be specific for

B56γ. Indeed, we detected 41 proteins only in S220N but not in WT protein complex (Table 5.4).

These proteins are probably not specific to mutant S220N but rather MS failed to detect them in other samples because of their low amount. Hence additional experiments are absolutely needed to verify their interaction with WT B56γ or S220N.

133

Discussion

Identification and characterization of signaling molecules interacting with B56γ will provide important clues in understanding the tumor suppressive function and regulation of B56γ-

PP2A. Earlier, we showed that S220N mutation results in the loss of p53-independent tumor suppressive function of B56γ and hypothesized that S220N mutation specifically disrupts B56γ interaction with an unknown substrate. In this study, we sought to find proteins that interact with

WT but not with S220N protein through affinity purification coupled with LC-MS/MS analysis.

MS analysis has detected a number of potential interacting proteins, including those that have not been previously reported to interact with B56γ. Interestingly, some of these proteins are known to function in various cellular processes including cell motility, focal adhesion, metastasis, and cell transformation. Once we verify the interactions of these proteins with B56γ, we will further examine their role in p53-independent tumor suppressive function of B56γ in the future.

Even though our result provides multiple proteins that potentially interact with B56γ, MS has its limitations and it is difficult to detect proteins of less abundance and to distinguish between specific and nonspecific interactions. In this study, LC-MS/MS analysis did not detect p53 in B56γ protein complex. Since p53 interaction was confirmed by immunoblotting (Figure

5.4A), lack of detection by MS was most likely due to low amount of proteins in the sample and/or low ionization efficiency of the sample peptides. In addition, a large number of proteins detected unique in S220N protein complex suggests the presence of many background proteins

(Table 5.4). Peptide peaks in the MS spectra are more likely to overlap in presence of abundant proteins, making it difficult to extract proteins in scarcity with low signal intensities. Thus, enrichment and purification of B56γ protein complex are crucial for increasing the likelihood of detecting B56γ specific associating proteins by LC-MS/MS. For reducing the number of

134

background proteins, one possibility is to purify B56γ complexes by double affinity purification method. In our early attempt, we tried to purify B56γ complexes by HA IP followed by B56γ IP.

However, a large amount of B56γ protein was lost during HA IP elution because majority of the

HA-B56γ remained tightly bound on HA-beads (Figure 5.3D). Low purification yields require large amounts of starting cell material which imposes challenge. To address this issue, alternative tags for B56γ may be used in tandem affinity purification to improve the protein elution efficiency. For example, a recent paper has shown a highly efficient method of purifying protein from mammalian cells using a double-affinity tag containing a streptavidin-binding peptide (SBP) and a HA-tag (Glatter et al., 2009). In this method, proteins were first bound to streptavidin beads with high affinity and eluted under mild conditions with 2mM biotin with almost 90% recovery, and then subjected to further purification by HA IP. This purification method may reduce nonspecific protein while minimizing loss of sample during purification processes. Another option is to perform subcellular fractionation and then separately analyze B56γ protein complexes.

B56γ expression is mostly ubiquitous but becomes enriched in the nucleus during S phase in the cell (Lee et al., 2010). Thus, purifying B56γ complex from different subcellular fractionation may also provide clues to understanding B56γ functions specific to cellular localizations in addition to reducing background proteins. Taken together, further improvements on B56γ protein complex purification are clearly needed to facilitate identification of B56γ interacting proteins by LC-

MS/MS.

135

References

Arroyo, J.D., Lee, G.M., and Hahn, W.C. (2008). Liprin α1 interacts with PP2A B56γ. Cell Cycle 7, 525–532.

Astigarraga, S., Hofmeyer, K., Farajian, R., and Treisman, J.E. (2010). Three Drosophila liprins interact to control synapse formation. J. Neurosci. 30, 15358–15368.

Astro, V., Asperti, C., Cangi, M.G., Cangi, G., Doglioni, C., and De Curtis, I. (2011). Liprin-α1 regulates breast cancer cell invasion by affecting cell motility, invadopodia and extracellular matrix degradation. Oncogene 30, 1841–1849.

Aznar, S., Fernández-Valerón, P., Espina, C., and Lacal, J.C. (2004). Rho GTPases: potential candidates for anticancer therapy. Cancer Lett. 206, 181–191.

Bengtsson, L., Schwappacher, R., Roth, M., Boergermann, J.H., Hassel, S., and Knaus, P. (2009). PP2A regulates BMP signalling by interacting with BMP receptor complexes and by dephosphorylating both the C-terminus and the linker region of Smad1. J. Cell. Sci. 122, 1248– 1257.

Beppu, H., Mwizerwa, O.N., Beppu, Y., Dattwyler, M.P., Lauwers, G.Y., Bloch, K.D., and Goldstein, A.M. (2008). Stromal inactivation of BMPRII leads to colorectal epithelial overgrowth and polyp formation. Oncogene 27, 1063–1070.

Bernards, A. (2003). GAPs galore! A survey of putative Ras superfamily GTPase activating proteins in man and Drosophila. Biochim. Biophys. Acta 1603, 47–82.

Bernards, A., and Settleman, J. (2004). GAP control: regulating the regulators of small GTPases. Trends Cell Biol. 14, 377–385.

Birkenfeld, J., Nalbant, P., Bohl, B.P., Pertz, O., Hahn, K.M., and Bokoch, G.M. (2008). GEF-H1 modulates localized RhoA activation during cytokinesis under the control of mitotic kinases. Dev. Cell 12, 699–712.

Von Bubnoff, A., and Cho, K.W. (2001). Intracellular BMP signaling regulation in vertebrates: pathway or network? Dev. Biol. 239, 1–14.

Choe, K.-M., Prakash, S., Bright, A., and Clandinin, T.R. (2006). Liprin-α is required for photoreceptor target selection in Drosophila. Proc. Natl. Acad. Sci. U.S.A. 103, 11601–11606.

De Curtis, I. (2011). Function of liprins in cell motility. Exp. Cell Res. 317, 1–8.

Dai, Y., Taru, H., Deken, S.L., Grill, B., Ackley, B., Nonet, M.L., and Jin, Y. (2006). SYD-2 Liprin-α organizes presynaptic active zone formation through ELKS. Nat. Neurosci. 9, 1479– 1487.

136

Eichhorn, P.J.A., Creyghton, M.P., and Bernards, R. (2009). Protein phosphatase 2A regulatory subunits and cancer. Biochim. Biophys. Acta 1795, 1–15.

Fujishiro, S.-H., Tanimura, S., Mure, S., Kashimoto, Y., Watanabe, K., and Kohno, M. (2008). ERK1/2 phosphorylate GEF-H1 to enhance its guanine nucleotide exchange activity toward RhoA. Biochem. Biophys. Res. Commun. 368, 162–167.

Glatter, T., Wepf, A., Aebersold, R., and Gstaiger, M. (2009). An integrated workflow for charting the human interaction proteome: insights into the PP2A system. Mol. Syst. Biol. 5, 237.

Goudreault, M., D’Ambrosio, L.M., Kean, M.J., Mullin, M.J., Larsen, B.G., Sanchez, A., Chaudhry, S., Chen, G.I., Sicheri, F., Nesvizhskii, A.I., et al. (2009). A PP2A phosphatase high density interaction network identifies a novel striatin-interacting phosphatase and kinase complex linked to the cerebral cavernous malformation 3 (CCM3) protein. Mol Cell Proteomics 8, 157– 171.

Ishihama, Y., Oda, Y., Tabata, T., Sato, T., Nagasu, T., Rappsilber, J., and Mann, M. (2005). Exponentially modified protein abundance index (emPAI) for estimation of absolute protein amount in proteomics by the number of sequenced peptides per protein. Mol Cell Proteomics 4, 1265–1272.

Jiang, W., Betson, M., Mulloy, R., Foster, R., Lévay, M., Ligeti, E., and Settleman, J. (2008). p190A RhoGAP is a glycogen synthase kinase-3β substrate required for polarized cell migration. J. Biol. Chem. 283, 20978–20988.

Kim, I.Y., Lee, D.-H., Lee, D.K., Ahn, H.-J., Kim, M.M., Kim, S.J., and Morton, R.A. (2004). Loss of expression of bone morphogenetic protein receptor type II in human prostate cancer cells. Oncogene 23, 7651–7659.

Ko, J., Kim, S., Valtschanoff, J.G., Shin, H., Lee, J.-R., Sheng, M., Premont, R.T., Weinberg, R.J., and Kim, E. (2003). Interaction between liprin-α and GIT1 is required for AMPA receptor targeting. J. Neurosci. 23, 1667–1677.

Kodach, L.L., Wiercinska, E., De Miranda, N.F.C.C., Bleuming, S.A., Musler, A.R., Peppelenbosch, M.P., Dekker, E., Van den Brink, G.R., Van Noesel, C.J.M., Morreau, H., et al. (2008). The bone morphogenetic protein pathway is inactivated in the majority of sporadic colorectal cancers. Gastroenterology 134, 1332–1341.

Krendel, M., Zenke, F.T., and Bokoch, G.M. (2002). Nucleotide exchange factor GEF-H1 mediates cross-talk between microtubules and the actin cytoskeleton. Nat. Cell Biol. 4, 294–301.

Kusama, T., Mukai, M., Endo, H., Ishikawa, O., Tatsuta, M., Nakamura, H., and Inoue, M. (2006). Inactivation of Rho GTPases by p190 RhoGAP reduces human pancreatic cancer cell invasion and metastasis. Cancer Sci. 97, 848–853.

Lee, T.-Y., Lai, T.-Y., Lin, S.-C., Wu, C.-W., Ni, I.-F., Yang, Y.-S., Hung, L.-Y., Law, B.K., and Chiang, C.-W. (2010). The B56γ3 regulatory subunit of protein phosphatase 2A (PP2A) regulates

137

S phase-specific nuclear accumulation of PP2A and the G1 to S transition. J. Biol. Chem. 285, 21567–21580.

Li, H.H., Cai, X., Shouse, G.P., Piluso, L.G., and Liu, X. (2007). A specific PP2A regulatory subunit, B56γ, mediates DNA damage-induced dephosphorylation of p53 at Thr55. EMBO J. 26, 402–411.

Matsuzawa, T., Kuwae, A., Yoshida, S., Sasakawa, C., and Abe, A. (2004). Enteropathogenic Escherichia coli activates the RhoA signaling pathway via the stimulation of GEF-H1. EMBO J. 23, 3570–3582.

Miyazono, K., Maeda, S., and Imamura, T. (2005). BMP receptor signaling: transcriptional targets, regulation of signals, and signaling cross-talk. Cytokine Growth Factor Rev. 16, 251–263.

Mizuarai, S., Yamanaka, K., and Kotani, H. (2006). Mutant p53 induces the GEF-H1 oncogene, a guanine nucleotide exchange factor-H1 for RhoA, resulting in accelerated cell proliferation in tumor cells. Cancer Res. 66, 6319–6326.

Olsen, O., Moore, K. a, Fukata, M., Kazuta, T., Trinidad, J.C., Kauer, F.W., Streuli, M., Misawa, H., Burlingame, A.L., Nicoll, R.A., et al. (2005). Neurotransmitter release regulated by a MALS- liprin-α presynaptic complex. J. Cell Biol. 170, 1127–1134.

Owens, P., Pickup, M.W., Novitskiy, S. V, Chytil, A., Gorska, A.E., and Aakre, M.E. (2012). Disruption of bone morphogenetic protein receptor 2 (BMPR2) in mammary tumors promotes metastases through cell autonomous and paracrine mediators. PNAS 109, 2814–2819.

Pouliot, F., Blais, A., and Labrie, C. (2003). Overexpression of a Dominant Negative Type II Bone Morphogenetic Protein Receptor Inhibits the Growth of Human Breast Cancer Cells. Cancer Res. 63, 277–281.

Sahai, E., and Marshall, C.J. (2002). RHO–GTPases and cancer. Nat. Rev. Cancer 2, 133–142.

Serra-Pagès, C., Medley, Q.G., Tang, M., Hart, A., and Streuli, M. (1998). Liprins, a family of LAR transmembrane protein-tyrosine phosphatase-interacting proteins. J. Biol. Chem. 273, 15611–15620.

Serra-Pagès, C., Streuli, M., and Medley, Q.G. (2005). Liprin Phosphorylation Regulates Binding to LAR : Evidence for Liprin Autophosphorylation. Biochemistry 44, 15715–15724.

Shen, J.-C., Unoki, M., Ythier, D., Duperray, A., Varticovski, L., Kumamoto, K., Pedeux, R., and Harris, C.C. (2007). Inhibitor of growth 4 suppresses cell spreading and cell migration by interacting with a novel binding partner, liprin alpha1. Cancer Res. 67, 2552–2558.

Shin, H., Wyszynski, M., Huh, K.-H., Valtschanoff, J.G., Lee, J.-R., Ko, J., Streuli, M., Weinberg, R.J., Sheng, M., and Kim, E. (2003). Association of the kinesin motor KIF1A with the multimodular protein liprin-alpha. J. Biol. Chem. 278, 11393–11401.

138

Wang, L., Yang, L., Luo, Y., and Zheng, Y. (2003). A novel strategy for specifically down- regulating individual Rho GTPase activity in tumor cells. J. Biol. Chem. 278, 44617–44625.

Wyszynski, M., Kim, E., Dunah, A.W., Passafaro, M., Valtschanoff, J.G., Serra-page, C., Streuli, M., Weinberg, R.J., Sheng, M., Hill, C., et al. (2002). Interaction between GRIP and Liprin- α/SYD2 Is Required for AMPA Receptor Targeting. Neuron 34, 39–52.

Zhen, M., and Jin, Y. (1999). The liprin protein SYD-2 regulates the differentiation of presynaptic termini in C. elegans. Nature 401, 371–375.

Zwijsen, A., Verschueren, K., and Huylebroeck, D. (2003). New intracellular components of bone morphogenetic protein/Smad signaling cascades. FEBS Letters 546, 133–139.

139

Acknowledgements

We thank Dr. S. Pan for running LC-MS/MS analysis.

140

Figure legend

Figure 5.1 Plasmid maps for pcDNA4/TO-HA-B56γ and pcDNA6/TR.

HA-tagged B56γ gene was isolated from the plasmid pCEP4-HA-B56γ and inserted into pcDNA4/TO using KpnI and XbaI restriction enzymes.

Figure 5.2 Tetracycline-inducible B56γ protein expression and p21 induction.

The extra band below HA-B56γ3 marked by * indicates the endogenous B56γ3 protein. (A)

Selected isolated colonies of the U2OS cells that were stably transfected with either B56γ WT (#1,

2, 3, 4, 5) or S220N (#6, 7, 8), or parental U2OS cells (M: mock) were treated with tetracycline.

Cells were harvested after 24 h, and cell lysates were subjected to SDS-PAGE and analyzed by

WB for proteins listed. (B) WT expressing cells were harvested after the incubation with either

0.5 or 1 μg/ml tetracycline for the indicated time period. (C) WT expressing cells were harvested after the incubation with 0.5 μg/ml tetracycline for the indicated time period.

Figure 5.3 Testing the pulldown efficiency of B56γ-PP2A by HA immunoprecipitation.

Immunoprecipitated proteins were subjected to SDS-PAGE and analyzed by WB using antibodies listed. (A) Precipitates from one 10-cm-diameter dish with varying amount of HA-beads (2.5-20

μl) were analyzed by WB. Precipitates from four 10-cm-diameter dishes with different wash conditions were analyzed by (B) silverstaining and (C) WB. “Wash 1” consists of four washes with lysis buffer, “Wash 2” consists of three washes with 300 mM KCl D buffer and once with lysis buffer, and “Wash 3” consists of three washes with 500 mM KCl D buffer and once with lysis buffer. Double affinity purification of HA-B56γ was performed by HA IP followed by B56γ

IP, and precipitates were analyzed by (D) WB. U2OS cell lysates were incubated with indicated amount of HA-beads (5-40 μl). Bound B56γ protein was eluted by HA peptide, and the elute was

141

further purified by B56γ IP. Purified B56γ complexes after B56γ IP were analyzed by WB for proteins listed (right). Amount of proteins left bound on the used HA-beads was detected by WB as well (left: Beads after elution).

Figure 5.4 Analysis of WT B56γ and S220N mutant complex proteins purified by HA immunoprecipitation.

Precipitates were subjected to SDS-PAGE and analyzed by (A) WB for proteins listed or (B) silver staining. The numbers indicated on the right correspond to the molecular weight (kDa) of marker proteins.

142

Figures

Figure 5.1 Plasmid maps for pcDNA4/TO-HA-B56γ and pcDNA6/TR.

143

Figure 5.2 Tetracycline-inducible B56γ protein expression and p21 induction.

144

Figure 5.3 Testing the pulldown efficiency of B56γ-PP2A by HA immunoprecipitation.

145

Figure 5.4 Analysis of WT B56γ and S220N mutant complex proteins purified by HA immunoprecipitation.

146

Tables

emPAI Protein Name NCBI GI no. WT S220N α isoform of scaffolding A subunit PR65, PP2A 4558258 2.67 2 γ isoform of regulatory B subunit B56, PP2A 31083259 1.86 1.23 ribosomal protein S29 isoform 1 4506717 -- 0.54 ribosomal protein S12 36146 0.52 0.23 ribosomal protein S17 4506693 0.22 0.48 ribosomal protein S11 4506681 -- 0.4 ribosomal protein L21 18104948 -- 0.4 ribosomal protein S6 337514 -- 0.39 β isoform of scaffolding A subunit PR65, PP2A 3603418 0.37 0.23 small nuclear ribonucleic protein (snRNP) 338259 0.36 -- ribosomal protein L37a 4506643 -- 0.34 histone H1.1 4885373 0.33 -- histone H1t 184084 0.33 0.33 human leukocyte antigen B 10799656 -- 0.33 ribosomal protein L19 4506609 -- 0.31 Tat binding protein 7 (TBP-7) 263099 0.28 0.28 ribosomal protein S26 296452 -- 0.27 histone H2B type 3-B 28173554 -- 0.25 DNA-binding protein 189299 -- 0.23 ribosomal protein S16 4506691 0.21 0.21 tubulin β-2C chain 5174735 -- 0.21 BBF2H7/FUS protein 34481861 -- 0.21 ribosomal protein S19 4506695 0.2 0.21 ribosomal protein L23a 1574942 0.2 -- ribosomal protein L24 4506619 -- 0.19 ribosomal protein L21 808090 -- 0.18 hnRNP U protein 16041796 0.08 0.18 ribosomal protein L11 495126 0.17 -- histone H1.0 4885371 -- 0.16 β'-actin 28336 -- 0.16 ribosomal protein L7a 55958183 -- 0.16 ATP/GTP binding protein 1 55661112 -- 0.16

147

emPAI Protein Name NCBI GI no. WT S220N ribosomal protein S5 550021 0.15 -- histone H1x 5174449 0.15 0.15 proteasome 26S subunit ATPase 1, isoform CRA_b 119601826 -- 0.15 ribosomal protein S9 550023 -- 0.15 ribosomal protein S8 4506743 0.14 0.14 β-tubulin 338695 -- 0.14 ribosomal protein L15 12006350 -- 0.14 ras-related protein Ral-A 4506405 -- 0.13 lung cancer oncogene 3 (HLC3); Kanadaptin 32394404 0.12 -- brain-specific angiogenesis inhibitor 1-associated protein 2 61680505 -- 0.12 ribosomal protein S3a 4506723 -- 0.11 heterogeneous nuclear ribonucleoprotein A0 5803036 -- 0.11 emerin 4557553 -- 0.11 ribosomal protein L3 337580 0.1 -- polymerase I and transcript release factor 42734430 0.1 -- hnRNP C protein 337455 0.1 -- leucine zipper transcription factor-like 1 8980369 0.1 -- HSP70-1 4529893 0.1 0.1 proteasome 26S subunit, ATPase, 1, isoform CRA_a 119603737 -- 0.1 transcription factor IIE 228298 -- 0.1 ribosomal protein P0 4506667 -- 0.1 MHC class I 118026426 -- 0.1 homeobox-containing protein 1 47059040 0.09 -- catalytic C subunit, PP2A 17149125 0.09 0.09 HSP70-6 34419635 -- 0.09 bone morphogenetic protein receptor type II (BMPR-II) 15451916 0.08 -- glutathione-insulin transhydrogenase 31746 -- 0.08 leukocyte antigen (HLA) class I molecule 32179 -- 0.08 liprin-α-1 4505983 0.07 -- cell division cycle-associated protein 2 44681484 0.07 -- Rho GAP p190-A 8886143 0.07 -- α tubulin 2843123 0.07 -- tubulin β-6 chain 14210536 0.07 0.21 xenotropic and polytropic retrovirus receptor 1 19923272 0.07 0.04

148

emPAI Protein Name NCBI GI no. WT S220N DNA-binding protein A 181484 -- 0.07 GPATCH8 protein 39644463 0.06 -- UBX domain-containing protein 4 24307965 0.06 -- nucleolin 189306 0.06 -- transmembrane protein 209 66348165 0.05 -- zinc finger CCCH domain-containing protein 4 126723060 -- 0.05 RAB GTPase activating protein 1-like 56202421 -- 0.05 coiled-coil domain-containing protein 81 isoform 1 226493512 -- 0.05 nucleophosmin-retinoic acid receptor α fusion protein 1314308 -- 0.05 NHS-like protein 1 isoform 1 254692997 0.04 -- Na+, K+ activated adenosine triphosphatase α subunit 179227 0.04 -- leucine-rich repeat and calponin homology domain-containing 14249428 0.04 0.04 protein 3 ATP-dependent RNA helicase 46 2696613 -- 0.04 splicing factor 3 subunit 1 isoform 1 5032087 -- 0.04 liprin-α-2 29171755 0.03 -- Na+, K+ -ATPase catalytic subunit 497763 0.03 -- kinesin family member 13A 57208793 0.03 -- Rho GEF-H1 15011974 0.03 --

Table 5.1 List of potential B56γ interacting proteins identified by LC-MS/MS.

149

emPAI Protein Name NCBI GI no. WT S220N α isoform of scaffolding A subunit PR65, PP2A 4558258 2.67 2.00 γ isoform of regulatory B subunit B56, PP2A 31083259 1.86 1.23 ribosomal protein S12 36146 0.52 0.23 β isoform of scaffolding A subunit PR65, PP2A 3603418 0.37 0.23 histone H1t 184084 0.33 0.33 Tat binding protein 7 (TBP-7) 263099 0.28 0.28 ribosomal protein S17 4506693 0.22 0.48 ribosomal protein S16 4506691 0.21 0.21 ribosomal protein S19 4506695 0.2 0.21 histone H1x 5174449 0.15 0.15 ribosomal protein S8 4506743 0.14 0.14 HSP70-1 4529893 0.10 0.10 catalytic C subunit, PP2A 17149125 0.09 0.09 hnRNP U protein 16041796 0.08 0.18 tubulin β-6 chain 14210536 0.07 0.21 xenotropic and polytropic retrovirus receptor 1 19923272 0.07 0.04 leucine-rich repeat and CH domain-containing protein 3 14249428 0.04 0.04

Table 5.2 List of potential B56γ interacting proteins detected in both WT B56γ and S220N mutant complexes.

150

emPAI Protein Name NCBI GI no. WT S220N small nuclear ribonucleic protein (snRNP) 338259 0.36 -- histone H1.1 4885373 0.33 -- ribosomal protein L23a 1574942 0.20 -- ribosomal protein L11 495126 0.17 -- ribosomal protein S5 550021 0.15 -- lung cancer oncogene 3 (HLC3); Kanadaptin 32394404 0.12 -- ribosomal protein L3 337580 0.10 -- polymerase I and transcript release factor 42734430 0.10 -- hnRNP C protein 337455 0.10 -- leucine zipper transcription factor-like 1 8980369 0.10 -- homeobox-containing protein 1 47059040 0.09 -- bone morphogenetic protein receptor type II (BMPR-II) 15451916 0.08 -- liprin-α-1 4505983 0.07 -- cell division cycle-associated protein 2 44681484 0.07 -- Rho GAP p190-A 8886143 0.07 -- α tubulin 2843123 0.07 -- GPATCH8 protein 39644463 0.06 -- UBX domain-containing protein 4 24307965 0.06 -- nucleolin 189306 0.06 -- transmembrane protein 209 66348165 0.05 -- NHS-like protein 1 isoform 1 254692997 0.04 -- Na+, K+ activated adenosine triphosphatase α subunit 179227 0.04 -- liprin-α-2 29171755 0.03 -- Na+, K+ -ATPase catalytic subunit 497763 0.03 -- kinesin family member 13A 57208793 0.03 -- Rho GEF-H1 15011974 0.03 --

Table 5.3 List of potential B56γ interacting proteins detected in WT B56γ but not in S220N mutant protein complex.

151

emPAI Protein Name NCBI GI no. WT S220N ribosomal protein S29 isoform 1 4506717 -- 0.54 ribosomal protein S11 4506681 -- 0.40 ribosomal protein L21 18104948 -- 0.40 ribosomal protein S6 337514 -- 0.39 ribosomal protein L37a 4506643 -- 0.34 human leukocyte antigen B 10799656 -- 0.33 ribosomal protein L19 4506609 -- 0.31 ribosomal protein S26 296452 -- 0.27 histone H2B type 3-B 28173554 -- 0.25 DNA-binding protein 189299 -- 0.23 tubulin β-2C chain 5174735 -- 0.21 BBF2H7/FUS protein 34481861 -- 0.21 ribosomal protein L24 4506619 -- 0.19 ribosomal protein L21 808090 -- 0.18 histone H1.0 4885371 -- 0.16 β'-actin 28336 -- 0.16 ribosomal protein L7a 55958183 -- 0.16 ATP/GTP binding protein 1 55661112 -- 0.16 proteasome 26S subunit, ATPase, 1, isoform CRA_b 119601826 -- 0.15 ribosomal protein S9 550023 -- 0.15 β-tubulin 338695 -- 0.14 ribosomal protein L15 12006350 -- 0.14 ras-related protein Ral-A 4506405 -- 0.13 brain-specific angiogenesis inhibitor 1-associated protein 2 61680505 -- 0.12 ribosomal protein S3a 4506723 -- 0.11 heterogeneous nuclear ribonucleoprotein A0 5803036 -- 0.11 emerin 4557553 -- 0.11 proteasome 26S subunit ATPase 1, isoform CRA_a 119603737 -- 0.10 transcription factor IIE 228298 -- 0.10 ribosomal protein P0 4506667 -- 0.10 MHC class I 118026426 -- 0.10 HSP70-6 34419635 -- 0.09 glutathione-insulin transhydrogenase 31746 -- 0.08 leukocyte antigen (HLA) class I molecule 32179 -- 0.08

152

emPAI Protein Name NCBI GI no. WT S220N DNA-binding protein A 181484 -- 0.07 zinc finger CCCH domain-containing protein 4 126723060 -- 0.05 RAB GTPase activating protein 1-like 56202421 -- 0.05 coiled-coil domain-containing protein 81 isoform 1 226493512 -- 0.05 nucleophosmin-retinoic acid receptor α fusion protein 1314308 -- 0.05 ATP-dependent RNA helicase 46 2696613 -- 0.04 splicing factor 3 subunit 1 isoform 1 5032087 -- 0.04

Table 5.4 List of potential B56γ interacting proteins detected in S220N mutant but not in WT

B56γ protein complex.

153

CHAPTER 6

Conclusion

154

Protein phosphatase 2A (PP2A) is a family of serine/threonine phosphoprotein phosphatases that are involved in the regulation of diverse signal transduction cascades and cellular processes. Accumulating evidence now highlights the role of PP2A as a guardian against tumorigenic transformation of cells. PP2A family consists of numerous PP2A heterotrimeric holoenzymes with dynamic subunits compositions where one of the various regulatory B subunits binds the AC core, which consists of a scaffolding A and a catalytic C subunit. Different B subunits confer diverse substrate specificity, enzyme activity, and cellular localization of PP2A holoenzyme. As such, particular B subunits are believed to govern the tumor suppressive function of PP2A by targeting tumor suppressor proteins or oncogenes for dephosphorylation and regulating cellular functions (Eichhorn et al., 2009). Interestingly, it was shown that tumor promoting SV40 ST antigen targets B56γ and suppression of B56γ leads to cell transformation

(Chen et al., 2004), suggesting that B56γ is a key factor in the tumor suppressive function of

PP2A. In support of this, our lab has previously revealed that the B56γ subunit associated PP2A

(B56γ-PP2A) dephosphorylates the tumor suppressor p53 at Thr55 and activates p53 function, promoting transcription of genes involved in cell cycle arrest and apoptosis in response to DNA damage. Moreover, we observed that B56γ-PP2A inhibits cell transformation in absence of p53

(Li et al., 2007), indicating that other unknown substrates are also involved in p53-independent function of B56γ. In an attempt to gain further understanding of tumor suppressive activity of

B56γ, we characterized a series of B56γ mutations originally identified in cancer samples. Our results present the underlying mechanism of B56γ regulatory function in directing tumor suppressive function of PP2A and provide a model of inactivation mechanism of B56γ-PP2A tumor suppressive activity in human cancer.

In this study, we demonstrated that the tumor suppressive ability of B56γ-PP2A requires

B56γ to bridge interactions between substrates and the AC core. Within B56γ, there are specific

155

interaction domains that mediate different substrate interactions with B56γ. For example, p53- dependent tumor suppressive function of B56γ-PP2A requires interaction between B56γ and p53

(Chapter 2). We showed that the B56γ mutants A383G and F395C, which do not interact with p53, lose p53-dependent tumor suppressive function of B56γ-PP2A. Specifically, they were unable to promote p53 Thr55 dephosphorylation and transcriptional activation of p21.

Importantly, we showed that a specific domain adjacent to these mutations is essential for B56γ binding with p53. In addition, we identified another B56γ mutation S220N inhibits p53- independent tumor suppressive function of B56γ-PP2A (Chapter 4). Ser220 is located on the open surface of B56γ adjacent to the catalytic pocket of B56γ-PP2A holoenzyme. This region of B56γ is likely responsible for the binding of one or more unknown substrates that conveys p53- independent tumor suppressive function of B56γ-PP2A. As such, our results provide a model where individual domains on B56γ mediate interaction with different substrates (Figure 6.1). This model supports the hypothesis that the B subunit provides a binding site such that a substrate can be recruited and positioned in the correct orientation for fitting of phosphorylated residues in the

PP2A active site and thus determines the substrate specificity of PP2A holoenzyme. In the future, elucidation of the structure of these domains together with bound substrates will provide useful information on how PP2A recognizes and interacts with its substrates.

In addition, our study provides structural insights into the B56γ-PP2A holoenzyme assembly that the integrity of B56γ structure is important for association of B56γ with the AC core. B56γ is an elongated, superhelical protein formed by eight pseudo HEAT repeats where each HEAT repeat consists of a pair of antiparallel α-helices connected by an intrahelical loop.

Crystal structure of B56γ-PP2A holoenzyme shows a number of residues in the conserved central domain (HEAT repeat 2-8) of B56γ that interact with A and C subunits (Xu et al., 2006; Cho and

Xu, 2007). We identified several B56γ mutations within this central domain that disrupt

156

interaction between B56γ and the AC core even though none of these residues directly contact with either A or C subunit (Chapter 4). This suggests that ordered alignment of helices as a whole is critical in stabilizing B56γ and maintaining B56γ-PP2A holoenzyme assembly. Although B56γ directly interacts extensively with both the A and C subunits at multiple interfaces, our data indicate that B56γ mutants lost interaction with A and C subunits altogether. Similarly, previously described A subunit mutants that cannot bind to C subunit simultaneously lost B56γ interaction (Ruediger et al., 2001b; Sablina et al., 2007). These data suggest that B56γ binds stably only to the AC core in cells. Dynamic regulatory B subunit exchanges of PP2A in response to different physiological circumstances are believed to be the basis for diverse PP2A activities.

Additional knowledge on the B56γ-PP2A structure will provide clues to the underlying mechanism of PP2A holoenzyme assembly and regulation of B56γ-PP2A function.

To date, we have identified several B56γ mutations that abolish tumor suppressive activity of B56γ-PP2A, which provides evidence for the tumor suppressive function of B56γ in cells. Because these mutations were originally identified in tumor samples, loss of B56γ-PP2A functions may be a critical event during tumorigenesis. We have demonstrated that mutations inactivate B56γ functions via several mechanisms resulting in the simultaneous deregulation of

B56γ-PP2A tumor suppressive activities in cells. Nonetheless, an important question remains as to what frequency does loss or disruption of B56γ function occur in human cancer. If tumor samples indeed commonly contain loss of function mutations in the B56γ gene, this will substantiate the causal role of B56γ as a tumor suppressor during cancer progression. Because different cell signaling pathways have distinct roles in different tissues, deregulation of certain molecular pathways can contribute to progression of different types of tumor. As such, it will be interesting to screen samples from various tumor types to investigate whether they possibly reveal additional mutations that have negative effects on B56γ function. So far, studies suggest that

157

tumor suppressive function of B56γ-PP2A is exerted through direct dephosphorylation of p53, however it is yet unknown what additional substrates of B56γ-PP2A play role in tumor suppression. A better understanding of B56γ-PP2A tumor suppressive function will depend largely on the identification of specific target of B56γ-PP2A and relevance of B56γ-PP2A dephosphorylation. Together, it is important to investigate the specific functions and cellular processes of which B56γ-PP2A is involved and the overall molecular and biological consequences of loss of B56γ-PP2A function.

Lastly and most intriguingly, our study may provide insight into the molecular aspects of observed differences in the cancer pathways between human and mouse. Perturbation of PP2A pathway in addition to five other pathways including p53, Raf, pRb, telomerase and Ral-GEFs is required for the tumorigenic transformation of human fibroblasts. On the other hand, perturbation of only two pathways involving p53 and Raf is sufficient to immortalize mouse fibroblasts

(Rangarajan et al., 2004). Curiously, while dephosphorylation modification of Thr55 of p53 appears to be essential for p53 activity in humans, Thr55 residue is absent in mouse p53. As such, impact of tumor suppressive function of B56γ-PP2A is presumably less in mice compared to humans. This tumor suppressive function of B56γ specific to humans perhaps offers a partial explanation why deregulation of PP2A function is detrimental to cancer prevention in humans.

Although other B subunits are also suggested to contribute to the tumor suppressive function of

PP2A (Eichhorn et al., 2009), our study underlines the importance of B56γ in regulating cell proliferation and transformation. Future studies on the relevance of B56γ-PP2A function within diverse cell signaling cascades will lead to a more complete understanding of how B56γ functions as a tumor suppressor. Given the potential role of PP2A in human cancer, unraveling the molecular mechanism of B56γ tumor suppressive function will ultimately aid in designing therapeutic agents for cancer diagnosis and counteracting PP2A deregulation in cancer.

158

References

Cho, U.S., and Xu, W. (2007). Crystal structure of a protein phosphatase 2A heterotrimeric holoenzyme. Nature 445, 53–57.

Eichhorn, P.J.A., Creyghton, M.P., and Bernards, R. (2009). Protein phosphatase 2A regulatory subunits and cancer. Biochim. Biophys. Acta 1795, 1–15.

Li, H.H., Cai, X., Shouse, G.P., Piluso, L.G., and Liu, X. (2007). A specific PP2A regulatory subunit, B56γ, mediates DNA damage-induced dephosphorylation of p53 at Thr55. EMBO J. 26, 402–411.

Rangarajan, A., Hong, S.J., Gifford, A., and Weinberg, R.A. (2004). Species- and cell type- specific requirements for cellular transformation. Cancer Cell 6, 171–183.

Ruediger, R., Pham, H.T., and Walter, G. (2001). Disruption of protein phosphatase 2A subunit interaction in human cancers with mutations in the Aα subunit gene. Oncogene 20, 10–15.

Sablina, A.A., Chen, W., Arroyo, J.D., Corral, L., Hector, M., Bulmer, S.E., DeCaprio, J.A., and Hahn, W.C. (2007). The tumor suppressor PP2A Abeta regulates the RalA GTPase. Cell 129, 969–982.

Xu, Y., Xing, Y., Chen, Y., Chao, Y., Lin, Z., Fan, E., Yu, J.W., Strack, S., Jeffrey, P.D., and Shi, Y. (2006). Structure of the protein phosphatase 2A holoenzyme. Cell 127, 1239–1251.

159

Figure legend

Figure 6.1 A model for the tumor suppressive function mediated by the interaction between

B56 and substrates.

The mutations A383G and F395C prevented B56γ interaction with p53 and thus disrupted p53- dependent tumor suppressive function of B56γ-PP2A. The mutation S220N disrupted p53- independent tumor suppressive function of B56γ-PP2A suggesting that it likely prevented B56γ interaction with an unknown substrate.

160

Figure

Figure 6.1 A model for the tumor suppressive function mediated by the interaction between B56 and substrates.

161