The Role of Caspase-2 and Pidd1 in Lung Tumorigenesis

THE ROLE OF CASPASE-2 AND PIDD1 IN LUNG TUMORIGENESIS

AND RESPONSE TO GENOTOXIC STRESS

Matthew Robinson Terry

A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

Department of Oncological Sciences

The University of Utah

December 2014

The University of Utah Graduate School

STATEMENT OF DISSERTATION APPROVAL

The dissertation of Matthew Robinson Terry has been approved by the following supervisory committee members:

Trudy G. Oliver , Chair 10/23/14 Date Approved

Donald E. Ayer , Member 10/23/14 Date Approved

Srividya Bhaskara , Member 10/23/14 Date Approved

Sheri L. Holmen , Member 10/24/14 Date Approved

David A. Jones , Member Date Approved

and by Bradley R. Cairns , Chair/Dean of the Department/College/School of Oncological Sciences and by David B. Kieda, Dean of The Graduate School.

ABSTRACT

Platinum-based chemotherapy is the standard-of-care for non-small cell lung

cancer. As with other chemotherapies, not all patients respond positively to treatment and

almost all patients will develop chemotherapy-resistant disease. Using the KrasG12D-

driven mouse model of lung adenocarcinoma, our lab identified upregulation of p53-

induced with a death domain isoform 1 (Pidd1) in resistant tumors in response to long-

term cisplatin chemotherapy treatment. Remarkably, PIDD1 expression in vitro induces

p53-dependent resistance to cisplatin as well as other DNA, damaging

chemotherapeutics.

PIDD1 expression leads to assembly of large molecular weight complexes named

PIDDosomes. The Caspase-2-PIDDosome is critical for PIDD1-induced chemotherapy

resistance. Our lab demonstrated that the Caspase-2-PIDDosome dynamically regulates

p53 activity via proteolytic modification of MDM2, the master negative regulator of p53.

In unstressed cells, MDM2 binds to and ubiquitinates p53, targeting it for proteasomal

degradation. Upon stress, Pidd1 is upregulated, ultimately leading to formation of the

Caspase-2-PIDDosome, and subsequent Caspase-2 activation. Activated Caspase-2 cleaves MDM2 into two fragments: p60 and p30. p60 maintains the ability to bind p53, but cannot ubiquitinate p53, resulting in increased p53 protein stability. The role of the

Caspase-2-PIDDosome in regulating the p53-MDM2 feedback loop is not well characterized in tumorigenesis or chemotherapy response.

In this study, I utilize the KrasG12D-driven mouse model of lung adenocarcinoma

to investigate tumorigenesis and chemotherapy response in Caspase-2-deficient and

Pidd1-deficient mice. These data demonstrate that Caspase-2 is a tumor suppressor in

lung adenocarcinoma primarily by inhibiting cell proliferation. Caspase-2-deficient tumors respond to chemotherapy; however, rapid tumor cell proliferation following treatment reduces the long-term therapeutic efficacy of chemotherapy. Unexpectedly,

Pidd1-deficiency does not impact lung tumorigenesis or chemotherapy response.

Mechanistic investigation in vitro revealed that ATM phosphorylation of PIDD1 is

critical for Caspase-2-PIDDosome assembly, Caspase-2-mediated MDM2 cleavage, cell

cycle arrest, and resistance to DNA damaging agents. This pathway is not modulated by exogenous MDM2 or its cleavage product p60. Further, pharmacological inhibition of the p53-MDM2 negative feedback-loop using nutlin-3 does not alter PIDD1-induced growth arrest or cisplatin resistance. Together these findings demonstrate that Caspase-2 and

PIDD1 have distinct functions in vivo and elucidate the Caspase-2-PIDDosome signaling network in p53-dependent response to DNA damage.

iv TABLE OF CONTENTS

ABSTRACT ...... iii

LIST OF ABBREVIATIONS ...... vii

ACKNOWLEDGEMENTS ...... viii

Chapters

1 INTRODUCTION ...... 1

Lung Cancer and Therapy ...... 1 PIDD1 Protein Interaction Domains ...... 4 PIDD1 as Molecular Switch: NF-kB vs Caspase-2 ...... 5 MDM2: p53 Regulator and Oncoprotein ...... 7 Caspase-2: The Enigma ...... 9 Caspase-2 and PIDD1 in Apoptosis ...... 11 Caspase-2 and Cell Cycle Regulation ...... 12 Caspase-2 and PIDD1 in Cancer ...... 14 p53 and Chemotherapy Response ...... 16 Summary ...... 19 References ...... 19

2 CASPASE-2 IMPACTS LUNG TUMORIGENESIS AND CHEMOTHERAPY RESPONSE IN VIVO ...... 32

Preface ...... 32 Introduction ...... 32 Results ...... 34 Discussion ...... 43 References ...... 46

3 A ROLE FOR PIDD1 IN P53-DEPENDENT GENOTOXIC STRESS RESPONSE, BUT NOT LUNG TUMORIGENSIS ...... 66

Introduction ...... 66 Results ...... 69 Discussion ...... 77 Concluding Remarks ...... 80 References ...... 80

4 MDM2 CLEAVAGE AND THE CASPASE-2-PIDDOSOME ...... 91

Introduction ...... 91 Results ...... 94 Discussion ...... 102 Concluding Remarks ...... 106 References ...... 107

5 CONCLUSION ...... 123

Summary ...... 123 Perspectives and Future Directions ...... 124 References ...... 130

APPENDIX: MATERIALS AND METHODS ...... 132

vi LIST OF ABBREVIATIONS

AD Adenocarcinoma

BrdU Bromodeoxyuridine

CARD Caspase activation and recruitment domain

DD Death domain

DISC Death-inducing signaling complex gIR Gamma irradiation

LRR Leucine-rich repeats

MEF Mouse embryonic fibroblast

MMEJ Microhomology-mediated end joining

NHEJ Nonhomologous end joining

NSCLC Non-small cell lung cancer

PBS Phosphate buffered saline

RING Really interesting new gene domain

ROS Reactive oxygen species

SCLC Small cell lung cancer

SNP Single nucleotide polymorphism

TKI Tyrosine kinase inhibitor

TRE Tetracycline response element ACKNOWLEDGEMENTS

First, I would like to thank my Thesis Advisor, Dr. Trudy Oliver. Without her, I

would not have been able to complete this doctoral program. Thankfully, she provided

me an opportunity allowing me to push forward and achieve my dream of obtaining a

Ph.D. I have learned many important lessons from her that I’m sure will benefit me in the

future.

I would also like to thank a previous mentor, Dr. David Grunwald. I learned a great deal from David about science as well as life in general. I am appreciative of the

time I spent under his mentorship and how it has positively impacted my world view.

I would like to thank my thesis committee and faculty reviewers for their guidance throughout my Ph.D. Their constructive feedback has been critical to my growth as a scientist.

My research would not have been possible without a large cast of people helping me along the way. Special thanks to current and previous members of the Oliver Lab for their support: Colin Russell, David McClellan, Kristofer Berrett, Philip Clair, Julio

Hidalgo, Stelian Pop, Rahul Arya, Chunhua Wu, Peter Hale, Jeff Clegg, Christine Lin,

Ushma Kc. Collaborating pathologist Mohamed Salama and Benjamin Witt were very helpful with histological analyses. It has been especially fun to work with and bounce ideas of Anandaroop Mukhopadhyay. He has been a tremendous help in my graduate career. I would also like to thank the scientific community at the University of Utah. It

has been a pleasure to work with so many smart and caring individuals. I would like to thank Trudy Oliver, Katie Ullman, Sergio Flores, Kigen Curtice, and Anandaroop

Mukhopadhyay for their help editing this dissertation.

I would like to thank my family for making me who I am. My father William

Terry, brother Rob Terry, and my grandparents George and Audrey Robinson had a large impact on my life. They taught me to be a well-rounded individual and to never give up on what you want in life. I am especially thankful of my number one fan, my mother Jill

Robinson Terry. She has always encouraged me to follow my passions and I cannot thank her enough for all the love and support she has provided throughout my life. She fostered my natural curiosity to seek out new experiences and learn new things. I would not be where I am today without her. I am forever grateful and blessed to have her in my life.

I am also extremely grateful to my partner Lauren Johnson who has been with me throughout this endeavor. She has supported me immensely and helped me grow as a person.

Financial support for this research was provided in part by National Institute of

Health Training Program in Genetics as well as the American Cancer Society as a

Research Scholar Award to Trudy G. Oliver.

CHAPTER 1

INTRODUCTION

Lung Cancer and Therapy

Lung cancer is responsible for more cancer deaths worldwide than the next three

most common cancers combined. The overall five-year survival rate for individuals with

lung cancer is a staggeringly low 16% 24. Many patients are asymptomatic until late stage disease has developed, accounting for the high mortality rate. The standard-of-care treatment for lung cancer remains a platinum-doublet regimen established over 30 years ago. However, other treatments are becoming available. For example, recent genomic analyses have identified key oncogenic driver mutations, sparking development of tyrosine kinase inhibitors (TKIs) for personalized targeted therapy. Most notable, TKIs targeting EGFR and EML4-ALK mutations have been successfully used in the clinic.

TKIs are not curative, however, and additional secondary mutations eliminating the therapeutic effect of these drugs are commonly observed. Furthermore, only a small subset of patients harbor oncogenic driver mutations in targetable genes, limiting widespread clinical value. Thus, combination chemotherapy remains the standard of care for the majority of lung cancer patients, underscoring the need for further investigation to advance our understanding of lung tumor cell biology and the biological response to chemotherapy. 2

Lung cancer is historically split into two categories: small cell lung cancer (SCLC

~20%) and non-small cell lung cancer (NSCLC ~80%) for the purpose of therapeutic

treatment. SCLC is highly associated with a smoking history and is exquisitely sensitive

to cisplatin and etoposide combination therapy 49. NSCLC consists of diverse histologic

subtypes primarily categorized as large cell carcinoma (LCC), squamous cell carcinoma

(SCC), and adenocarcinomas (AD). Adenocarcinomas are often associated with

oncogenic driver mutation in genes such as KRAS, EGFR, ALK, BRAF, MEK, etc. and

will be discussed further in this dissertation 81.

Constitutively activating mutations in the small GTPase KRAS are the most

frequent mutation observed in human lung AD 81. Between 15%-25% of all NSCLC

patients harbor a KRAS mutation. Furthermore, KRAS mutation is associated with poor

prognosis and there are no approved targeted therapies available 64. Mouse models of

Kras-driven lung tumorigenesis, pioneered in Tyler Jacks’ lab, have been utilized for

over a decade 40. However, the complexity of tumorigenic signaling activated by oncogenic Kras has proved a colossal challenge to developing novel therapeutic

strategies.

The oncogenic KrasG12D-model of adenocarcinoma recapitulates our understanding of the human disease accurately. Similar to human disease progression, mouse lung tumors in this model progress from hyperplasia to early stage adenomas to late stage adenocarcinomas 40. Additionally, mice harboring mutation in p53, a

commonly mutated tumor suppressor, develop metastases similar to those seen in

advanced stage human patients 41. Finally, comparative analysis of gene expression profiles from resected human ADs and mouse ADs demonstrates that murine oncogenic

Kras induces a similar transcriptional profile 95. Recently, synchronous co-clinical trials 3

in humans and the murine KrasG12D-model have been developed to test genetic modifiers of chemotherapy response in real-time, highlighting the use of this mouse model to advance understanding and treatment of the human disease 17.

The KrasG12D mouse model was used to investigate response to long-term

cisplatin treatment by our lab. Lung tumors in this model initially respond to cisplatin

treatment by inducing cell cycle arrest and apoptosis, resulting in tumor regression 74.

However, lung tumors become unresponsive following multiple rounds of cisplatin

treatment, as determined by increased tumor growth during the treatment phase 74. Gene

expression profiling of cisplatin response in naïve tumors treated for the first time

compared to pretreated resistant tumors revealed enrichment of gene pathways involved in cell cycle control and DNA damage response, among others 74. One interesting gene upregulated in resistant tumors was Pidd1, p53-induced with a death domain isoform 1 74.

Pidd1 is a p53 transcriptional target activated in response to cellular stress 57.

PIDD1 protein expression leads to the formation of a large molecular weight complex

termed the Caspase-2-PIDDosome 97. The Caspase-2-PIDDosome leads to proximity-

induced activation of Caspase-2, an atypical cysteine-aspartic protease 97. Caspase-2

cleaves MDM2, a negative p53 regulator, in the highly conserved DVPD amino acid

motif 75. Cleavage of full-length MDM2 into the p60 and p30 products uncouples the p53

binding domain from the E3 ligase containing RING domain, thereby preventing MDM2

from targeting p53 for proteasomal degradation 75. Preventing p53 degradation through

MDM2 cleavage leads to p53 accumulation and expression of Pidd1 completing a

positive feedback loop (Figure 1.1) 75.

Importantly, Caspase-2 activity does not promote apoptosis and cleaved MDM2 is

prevalent in nonapoptotic tumor cells 75, 83. The remainder of this review will focus on 4

current knowledge of Caspase-2 and Pidd1 function in tumorigenesis, DNA damage

response, and regulation of the p53-MDM2 negative feedback loop.

PIDD1 Protein Interaction Domains

PIDD1 is also known as LRDD, leucine-rich repeats with a death domain. PIDD1

contains seven N-terminal leucine-rich repeats (LRRs), two centrally located ZU-5 domains, and a C-terminal death domain (DD). LRRs from other proteins have been found to form structural motifs involved in protein-protein interactions 6. To date, ATM

is the only protein known to interact with the LRR domain of PIDD1 2. The function of

ZU-5 domains is not well understood, but limited studies of the ZU-5 domain containing

proteins Unc5b and Ankyrins suggest it is important for protein-protein interactions as well as auto-inhibition 106, 111. Both proliferating cell nuclear antigen (PCNA) and RCF5,

DNA replication factors, interact with PIDD1 through its ZU-5 domains 58. Furthermore,

PIDD1 processing described below removes one or both ZU-5 domains during PIDD1

activation, supporting an auto-inhibitor function of the ZU-5 domains. The C-terminal

DD of PIDD1 is a protein-interacting motif that interacts with the DD of RIP-associated

ICH-1 homologous protein with a death domain (RAIDD). RAIDD also contains a

caspase recruitment domain (CARD) that interacts with pro-Caspase-2 and is essential

for the formation of the Caspase-2-PIDDosome 23, 97. Thus, all notable PIDD1 domains

are involved in protein-protein interactions, supporting its role as a scaffold protein.

To gain additional insight into PIDD1 functions, Logette et al. used an unbiased mass spectrometry-based proteomics approach to identify novel PIDD1 interacting proteins 58. Interestingly, of the 21 interaction partners identified, 13 proteins are directly

involved in transcription and/or DNA damage repair. Most significantly, they 5

demonstrate that PIDD1 binds to PCNA, and in doing so prevents p21-mediated

repression of PCNA. Increased PCNA activity leads to recruitment of error-prone

translesion DNA polymerases to reduce replication stress in the presence of DNA

damage. Thus, PIDD1 contains several protein-protein interaction domains that function to regulate downstream effectors, suggesting it is involved in cell cycle regulation and

DNA damage repair.

PIDD1 as a Molecular Switch: NF-kB vs Caspase-2

PIDD1 activity is regulated by auto-proteolytic activity via an intein-like mechanism. Full-length PIDD1 is processed into three fragments. The first auto-cleavage event at S446 yields PIDD1-N and PIDD-C and further processing of PIDD1-C at S588 yields PIDD1-CC 97. PIDD1-C and PIDD1-CC have distinct functions in Nuclear Factor

kB (NF-kB) signaling and Caspase-2 activation. Under basal conditions, PIDD1 is

rapidly processed into PIDD1-C, and in the presence of genotoxic stress, PIDD1-C is

further processed into PIDD1-CC 98.

PIDD1 functions as a molecular switch, triggering cell survival by activating the

NF-kB pathway or cell death by activating Caspase-2 2, 98. PIDD1-induced activation of the NF-kB pathway has been reviewed in detail 10. In brief, DNA damage induces

PIDD1-C translocation to the nucleus where it forms a complex with receptor interacting

protein 1 (RIP1) and NF-kB essential modulator (NEMO). This complex promotes

posttranslational modification of NEMO, ultimately leading NEMO to be exported to the

cytoplasm. In the cytoplasm, NEMO activates the IKK complex, thereby repressing the

IkB complex and allowing for NF-kB to translocate to the nucleus, where it functions as a

transcription factor to induce expression of genes that promote cell survival. 6

In the absence of PIDD1, NF-kB signaling is significantly reduced in response to

DNA damage, signifying the importance of PIDD1 in regulating this pathway 2, 11, 44.

Interestingly, loss of Caspase-2 results in higher activation of the NF-kB pathway 31, 87. It is presumed that PIDD1 interaction with RIP is increased in the absence of Caspase-2-

PIDDosome formation. However, this presumption is potentially flawed because the terminal PIDD1-CC fragment produced under these conditions (described below) cannot interact with RIP/NEMO and thus cannot activate the NF-kB pathway. Although evidence is limiting, perhaps there is cross-talk between PIDD1-C and Caspase-2 within the nucleus that regulates PIDD1 processing and function.

Processing of PIDD1-C to PIDD1-CC is required for assembly of the Caspase-2-

PIDDosome 98. Formation of this high molecular weight complex activates Caspase-2 97.

Caspase-2 cleaves full-length MDM2 (p90) into two cleavage products: p60 and p30 75.

MDM2 contains three notable protein motifs: a N-terminal p53 binding domain, a centrally located acidic domain involved in protein-protein interactions, and a C-terminal

RING domain containing E3 ubiquitin ligase activity. MDM2 is cleaved between the acidic domain and the RING domain. The p60 cleavage product contains the p53 DNA binding domain and the acidic domain, whereas the p30 cleavage product contains only the RING domain. Moreover, increased p60 expression stabilizes p53 protein levels leading to increased p53 activity. No studies have examined the role of the p30 cleavage product. Although p53 regulation can clearly be altered by MDM2 cleavage, few studies have attempted to distinguish between full length MDM2 and the cleavage products.

The endogenous Caspase-2-PIDDosome-mediated positive feedback loop is activated in response to DNA damage in several NSCLC, osteosarcoma, and colorectal cell lines 75. This suggests the Caspase-2-PIDDosome pathway is a general p53 stress 7

response. Activation of this feedback loop leads to p53-dependent growth arrest and

increased cell viability in the presence of genotoxic agents, including cisplatin, etoposide, and doxorubicin 75. Again, PIDD1 was shown to be upregulated in cisplatin resistant lung tumors and may be a viable therapeutic target to re-sensitize tumors 74.

MDM2: p53 Regulator and Oncoprotein

MDM2 is the principal negative regulator of p53. The role of MDM2 in regulating p53 is exemplified by p53-dependent embryonic lethality in Mdm2 null mice

69. As mentioned, MDM2 is an E3 ubiquitin ligase that binds directly to p53, leading to

poly-ubiquitination of p53 and subsequent proteasomal degradation 36. Also, Mdm2 is a p53 transcriptional target forming a negative feedback loop. At the transcriptional level, p53 regulates Mdm2 via two different promoters termed p1 and p2 4. The p1 promoter is

active under basal conditions and the p2 promoter is utilized by p53 in response to stress.

While the p2 promoter is dispensable for viability, the p2 promoter mutant mice are

sensitized to p53-dependent apoptosis in response to sub-lethal irradiation, a standard

assay of p53 functional activity 80.

In addition to targeting p53 for proteasomal degradation, MDM2 regulates p53

through a variety of other mechanisms. First, a major endogenous regulator of the p53-

MDM2 interaction is the tumor suppressor p14ARF (ARF) in humans (p19ARF in mice)

encoded by an alternative reading frame of CDKN2A. ARF binds to MDM2, inhibiting

the p53-MDM2 interaction, leading to increased p53 stability and activity 37, 108. In this

regard, ARF acts as a tumor suppressor by activating p53. Consistent with this, ARF loss and p53 mutation are inversely correlated 70. Additionally, ARF-mediated disruption of the p53-MDM2 interaction prevents histone methyl transferase recruitment to p53 targets 8

and promotes p300-CBP acetylation of p53, which enhances p53 activity 39. Disruption of

the p53-MDM2 interaction is a strong inducer of p53 activity and has been a target of therapeutic intervention, which will be discussed later. Second, MDM2 binds to p53 mRNA, leading to increased mRNA stability and translation during cellular stress 27. This further increases p53 protein levels via increased translation of p53 mRNA, and an inhibition of MDM2 E3 ligase activity 27. Third, Mdm2 recruits histone modifying

enzymes to p53 target genes. Histone methyl transferases EHMT1 and SUV39HI are

recruited by MDM2 to p53 target genes and methylate histone H3-K9, reducing

expression of p53 targets genes 16. Last, Mdm2 binds to the p53 transactivation domain inhibiting p53 transcriptional activity 68, 73. However, the biological relevance of MDM2-

mediated transcriptional repression is not well understood.

MDM2 E3 ligase activity is not restricted to p53 and several proteins have been

shown to be regulated by MDM2 that are also involved in tumorigenesis 56. Noteworthy, tumor suppressor pRb and p53 target gene p21 are targeted by MDM2 for degradation,

promoting enhanced cell proliferation 101, 114. The transmembrane protein E-cadherin,

important for cell-cell adhesion, is also targeted for destruction by MDM2 110.

Downregulation of E-cadherin is strongly associated with the epithelial to mesenchymal

transition, which is critical for metastasis 78. Also, MDM2 enhances stability and translation of x-linked inhibitor of apoptosis (XIAP) mRNA, promoting cell survival 30.

As briefly described here, several p53-independent MDM2 substrates contribute to

MDM2 oncogenic function.

In a variety of human cancers, MDM2 is genomically amplified or upregulated at

the transcriptional and/or translational level 86. A single nucleotide polymorphism in the

MDM2 promoter (SNP309) naturally occurring in the human population increases MDM2 9

transcription and contributes to increased tumor susceptibility in NSCLC 12, 116. Indeed,

over-expression of Mdm2 in mice leads to chromosomal aberrations and promotes spontaneous tumor formation in vivo independent of p53 14, 47.

The studies discussed here describe both p53-dependent and p53-independent oncogenic functions of MDM2. It is unclear, however, how Caspase-2 cleavage of full- length MDM2 impacts these regulatory mechanisms. p60 protein levels accumulate several-fold higher than full-length MDM2, suggesting p60 may compete with full-length

MDM2 for access to p53.

Caspase-2: The Enigma

Caspases are commonly known for their role in programmed cell death; however, many have nonapoptotic functions as well 28. Importantly, caspases are not indiscriminate

proteases, but rather proteolytically cleave specific targets to alter protein function and/or

reduce protein abundance. The lack of validated Caspase-2 target proteins is currently a

major impediment to understanding the role of Caspase-2 25.

Caspase-2 is evolutionarily well conserved but unique among caspases 51.

Caspase-2 is the only known caspase that localizes to the nucleus, although its function within the nucleus is still controversial 18. It contains a long prodomain containing a

caspase activation and recruitment domain (CARD) that is involved in dimerization.

Dimerization facilitates proximity-induced activation of initiator caspases, as opposed to

proteolytic activation of executioner caspases such as Caspase-3 84. Activated initiator

caspases cleave inactive executioner caspases, thereby activating executioner caspases to induce apoptotic activity. In this regard, Caspase-2 differs from initiator caspases in that it cannot cleave executioner caspases 102. 10

Caspase-2-deficient mouse models have been developed to better understand the

enigmatic role of Caspase-2; however, much remains unknown. Genetic deletion of

Caspase-2 in mice does not elicit an overt phenotype; Caspase-2 null mice are viable and born in Mendelian ratios 7. Caspase-2-deficient female mice have an excess number of

oocytes due to apoptotic resistance, demonstrating a mild defect in apoptosis 7.

Furthermore, Caspase-2-deficient mice do not develop spontaneous tumors and are not sensitized to DNA damaging agents, suggesting Caspase-2 has a minor role in apoptosis

60. Functional redundancy with other caspase family members may account for the unremarkable phenotype in Caspase-2-deficient mice.

In vitro analyses testing Caspase-2 function in chemically-induced cell death are

contradictory. Several studies have observed that Caspase-2 loss sensitizes cells to

apoptosis in response to gamma irradiation (gIR), etoposide, stauroporine, and taxol 34, 65,

93. However, a majority of studies have found Caspase-2 loss reduces apoptosis in

response to similar agents 34, 45, 62, 87. Noteworthy, these studies tested the role of Caspase-

2 in chemically-induced apoptosis in vitro. How cells respond to DNA damage in vivo

remains to be thoroughly addressed. Shalini et al. identified increased death in the liver of

Caspase-2 null mice fed a high ethanol diet92. Whether increased cell death is a result of

impairing detoxification or an anti-apoptotic role for Caspase-2 in the liver is unresolved.

Interestingly, Caspase-2 deficient mice display characteristic signs of premature

aging and have a reduced maximum lifespan 113. Many of the aging symptoms in

Caspase-2 null mice are a result of increased levels of reactive oxygen species (ROS) 92.

Interestingly, ROS induces oxidative damage to protein and DNA alike, which can lead to genome instability. As a compensatory mechanism, elevated levels of p53 and p21 are

observed in older Caspase-2 null mice92. Conversely, late passage Caspase-2 null MEFs 11

have elevated ROS but reduced expression of p53 targets p21, Puma and Noxa 22, 35. The

interconnection between ROS, Caspase-2, and p53 signaling in regulating genome

stability, aging, and carcinogenesis remains to be determined.

Caspase-2 and PIDD1 in Apoptosis

Caspase-2 has unique features of both initiator and executioner caspases that

appear to be nonfunctional in the context of wildtype p53. Studies utilizing p53-deficient

cells clearly demonstrate a role for Caspase-2 in apoptosis. Either delay or absence of

apoptosis has been observed in Caspase-2-deficient cells 53, 89, 100. Furthermore, several

Caspase-2 substrates have been identified that contribute to apoptosis. The proapoptotic

protein BID, BH3 interacting-domain death agonist, is cleaved and activated by Caspase-

2 32. BID cleavage leads to cytochrome release and mitochondrial outer membrane

permeablization 59. Caspase-2 cleaves DeltaNp63, an inhibitor of the proapoptotic transcription factor TAp63 46. Caspase-2 also cleaves RIP1, reducing activation of the

NF-kB pathway 31. Thus, several Caspase-2 substrates activate apoptotic functions or

reduce prosurvival signals.

Caspase-2 activation in Pidd1 null MEFs suggests that the PIDDosome is not the

only mechanism for Caspase-2 activation 50, 60. Indeed, other studies have suggested

Caspase-2 is activated during apoptosis through the DISC complex containing CD95, the

TNFR1 signaling complex containing RAIDD, RIP, and TRADD, and other unidentified

high molecular weight complexes 23, 38, 77. However, the biological relevance of these

complexes has not been confirmed. Functional redundancy may allow Caspase-2 to be

activated in the absence of the PIDDosome by other signaling platforms 60.

Counter to a role in apoptosis, high levels of activated Caspase-2 in response to 12

PIDD1 expression does not induce cell death in p53 wildtype cells 8, 75, 97. One

explanation for the dichotomy in apoptotic response based on p53 status is sensitization

do to de-regulation of the cell cycle. Cells deficient in p53 fail to induce a G1 arrest and

Caspase-2 is important for the G2/M arrest. Thus, combined loss of both checkpoints

may potentially increase susceptibility to apoptosis. In fact, this mechanism is the basis

for the development of therapeutic inhibitors to checkpoint kinases such as CHK1,

CHK2, and ATM 19, 29.

Caspase-2 and Cell Cycle Regulation

The cell cycle and programmed cell death are intimately connected to maintain

tissue homeostasis. Through genetic and knockdown approaches, several studies have

shown that Caspase-2 deficiency leads to an increase in cellular proliferation 35, 75, 82, 87.

This often contributes to genome stability. Furthermore, increased cell proliferation and

colony formation following gamma irradiation demonstrates that Caspase-2 regulation of the cell cycle is important for its tumor suppressor function 34, 35, 87. Discussed below are

three key mechanisms by which Caspase-2 functions as a negative regulator of the cell

cycle.

First, Caspase-2 is important for G2/M growth arrest. Caspase-2 activity is

downregulated during normal mitosis by cdk1/cyclinB1 phosphorylation of Caspase-2

S340 preventing Caspase-2 activation 1. In the presence of DNA damage, phosphatase

PP1 is activated and de-phosphorylates S340, allowing to be activated and Caspase-2-

induced mitotic arrest 13. Cells exposed to gamma irradiation undergo G2/M cell-cycle

arrest and knockdown of Caspase-2 significantly inhibited mitotic arrest 15. Without

mitotic arrest, cells continue through the cell cycle, leading to improper cell divisions 13

associated with increased genome instability.

Second, Caspase-2 is involved in cell cycle arrest due to nutrient deprivation. In

the presence of sufficient energy availability, high levels of nicotinamide adenine

dinucleotide phosphate (NADPH) activate CaMKII, Ca2+ /calmodulin-dependent protein kinase II. CaMKII phosphorylates Caspase-2 S135 in the prodomain, inhibiting Caspase-

2 processing 71. This phosphorylation event is further enforced by 14-3-3 recognizing the

S135 modification 72. Under metabolic stress, 14-3-3 disengages from Caspase-2 and the

phosphatase PP1 de-phosphorylates S135, allowing Caspase-2 to be activated 72.

Supporting a role for Caspase-2 in regulating metabolic stress, Caspase-2-deficient MEFs are resistant to serum deprivation-induced cell death 34.

Third, negative feedback between Caspase-2, p53, and p21 has been identified

(Figure 2.1). Dorstyn et al. demonstrated that the initial elevation of p53 and p21 protein

levels following DNA damage is Caspase-2-dependent 22. The interaction of Caspase-2

with the 3’ UTR of p21 mRNA promotes p21 mRNA stability and translation 93. Both

p53 and p21 inhibit Caspase-2 gene expression forming a negative feedback loop. p53

represses transcriptional activation of Caspase-2, while p21 interacts with E2F1 at the

Caspase-2 promoter to inhibit transcriptional activation 3, 21. In contrast, our lab and

others have observed a normal acute p53 response in the absence of Caspase-2 75, 93. One

explanation is the difference in DNA damage type; gamma irradiation inducing double-

stranded DNA breaks versus topoisomerase II inhibitors doxorubicin and etoposide.

Future studies addressing stimuli-specific Caspase-2 activity may clarify this

discrepancy.

To conclude, Caspase-2 utilizes multiple mechanisms to regulate the cell cycle.

These mechanisms are also critical for regulating cell response to DNA damaging agents. 14

In response to DNA damage, Caspase-2 deficient cells fail to induce proper growth arrest, promoting tumorigenesis.

Caspase-2 and PIDD1 in Cancer

Caspase-2 has been implicated as a tumor suppressor in a variety of human cancers. As the first hint, the Caspase-2 locus is on chromosome 7q, a location where

alterations have been described in a variety of cancers, including lung, breast, and

hematological cancers 9, 55, 90. More specifically, Caspase-2 gene loss or downregulation

has been observed in several lymphomas as well as gastric cancer, metastatic brain

tumors, and several adenocarcinomas 87, 112, 117. However, mutations in Caspase-2 in

human cancers are rare 82.

Several studies using mouse models have been utilized to address the role of

Caspase-2 in cancer. Genetic deletion of Caspase-2 accelerates tumorigenesis in Myc-

driven lymphoma, c-Neu-driven mammary carcinoma, and xenograft models 35, 61, 82, 87. In

agreement with its role in regulation of the cell cycle, loss of Caspase-2 in these genetic

models is associated with increased rates of proliferation. As a consequence, genome

instability indicators such as increased aneuploidy, micronuclei, and binucleate cells have been observed in Caspase-2 null tumor cells 22, 82.

Genome instability is often associated with perturbations in the p53 pathway.

Accelerated tumorigenesis in Myc-driven lymphoma is p53-dependent and the selective pressure to lose p53 in this model was significantly reduced in mice lacking Caspase-2 61.

Furthermore, no p53 mutations were observed in c-Neu-driven mammary carcinoma 82.

Interestingly, expression of p53 and its transcriptional target p21 are under-expressed in

Caspase-2 null MEFs 35, 82, 93. Attenuated activation of the p53 pathway was also 15

observed in Caspase-2 null MEFs in response to DNA damage 82, 93. These findings can

be explained in part by our data demonstrating an increase in full-length MDM2 in the absence of Caspase-2, ultimately leading to increased MDM2-mediated p53 proteasomal degradation 75.

In contrast to Caspase-2, the role of Pidd1 in mouse models of cancer is less well

established. In Myc-driven lymphoma, loss of Pidd1 strongly delays the onset of tumorigenesis in a p53-dependent manner, the opposite of that observed in Caspase-2- deficient mice 61. However, the mechanism by which Pidd1 loss prevents tumorigenesis

remains a mystery. Outside of this finding, research from the same group did not observe

any differential effects due to Pidd1-deficiency in common surrogate assays for

tumorigenic potential, such as proliferation rate, apoptosis rate, clonal survival, or DNA

damage-induced tumorigenesis 11, 61. Our lab previously identified upregulation of Pidd1

in cisplatin-resistant Kras-driven lung tumors 74. Importantly, reduced sensitivity to

genotoxic agents is recapitulated in a variety of human cell lines that exogenously

expressed PIDD1 75.

PIDD1 perturbation has not been clearly implicated in any human cancers to date.

While its cell survival functions promoting the prosurvival NF-kB pathway may be protumorigenic, its role in inducing p53-mediated growth arrest or apoptosis in p53- deficient cells is antitumorigenic and would be strongly selected against. Therefore, a mechanism to induce PIDD1 expression in response to genotoxic stress, but not under normal conditions would be selected for. Indeed, PIDD1 is upregulated as a mechanism of resistance in cisplatin-resistant mouse lung tumors 74. The lack of postchemotherapy

biopsies and/or surgeries has limited most genomic studies in human cancer to operable

primary tumors where PIDD1 is not thought to be dysregulated. 16

Altogether, the literature supports a model whereby PIDD1 and Caspase-2 have distinct and independent functions in cancer. Interestingly, PIDD1 and Caspase-2 have

opposing effects on Eu-myc-driven lymphoma, yet both are functionally dependent on

p53 61. This finding highlights the unpredictable nature of p53-dependent responses to

cellular stress and warrants further investigation into the context-specific nature of each of these proteins.

p53 and Chemotherapy Response

p53 was coined guardian of the genome in 1992 by David Lane 52. p53 is a highly

studied tumor suppressor that is mutated in half of all lung cancers 96. These mutations are predominantly missense mutations occurring in the DNA binding domain and thereby prevent p53 transcriptional activity 76. Differences in tumor susceptibility have been

associated with different p53 mutant alleles, especially when compared to p53 null

tumors, suggesting dominant negative or neomorphic functions of mutant 53 33, 76, 79.

Further complicating the role of p53 in oncology, p53 status has been only loosely correlated to prognosis and chemotherapy response. However, in lung cancer, p53 mutant status is correlated with poor prognosis 67.

The role of p53 in tumor suppression has been attributed to both transcription-

dependent and -independent functions. Transcription-dependent functions are critical to

p53-dependent response in the context of a wide variety of stimuli. p53 functions

predominantly as a transcriptional activator of target genes involved in cell cycle arrest,

DNA damage repair, apoptosis, and senescence. One mechanism regulating appropriate

target gene activation is the utilization of high and low affinity p53 consensus DNA

sequences 88. High affinity binding sites allow rapid gene expression following p53 17

activation, while low affinity binding sites associated with apoptotic genes are expressed

at later time points due to the requirement for high and/or prolonged exposure to p53 85,

115. Genes critically important to Caspase-2-PIDDosome-induced cell cycle arrest, such

as Pidd1, Mdm2, and p21, fall into the rapidly expressed gene category.

Early induction of cell cycle arrest and DNA damage repair allows cells to repair any DNA damage before entering mitosis. Cells that fail to arrest and prematurely enter mitosis with DNA damage have two potential fates. Cells with extensive DNA damage that cannot be repaired undergo mitotic cell death, commonly referred to as mitotic catastrophe. These cells are of little consequence to an organism as a whole. However, cells that harbor DNA damage and survive through mitosis often contain chromosomal aberration that could potentially lead to oncogenic transformation.

The classical early response p53 target gene primarily responsible for p53- dependent growth arrest is the cyclin-dependent kinase inhibitor CDKN1A, encoding p21.

Mice deficient in p21 have increased chromosomal aberrations, suggesting genome integrity is compromised 5. Conversely, over-expression of p21 in vitro enhances cell viability in the presence of genotoxic agents such as cisplatin and etoposide, suggesting p21-induced growth arrest protects against genomic insults 107. An elegant study by

Jackson et al. ’12 tested the role of growth arrest induced by p53 and p21 in response to

chemotherapy in a mouse model of mammary carcinomas 43. They found that mammary tumors deficient for p53 or p21, which fail to arrest in response to chemotherapy, had improved tumor regression and response duration. By contrast, tumors with wildtype p53 and p21 exhibited growth arrest following treatment, followed by relapse. Similarly, our lab identified a strong correlation between cisplatin resistance and wildtype p53 status in

cisplatin resistant lung tumors in mice 74. These studies support the long-standing dogma 18

that rapidly dividing cells are more prone to cell death in response to cytotoxic agents

compared to terminally differentiated cells.

In addition to promoting cell viability through growth arrest, p53 has a well-

documented role in inducing cell death through transcription-dependent and -independent

mechanisms to promote therapeutic response. Several classical proapoptotic genes are

p53 transcriptional targets. p53 can also translocate to the mitochondria during apoptosis

where it interacts with Bcl-2 family members, leading to mitochondrial outer membrane

permeablization, cytochrome c release, and ultimately apoptosis 103. The diversity of functions associated with p53 and the context-specific nature of its function have precluded oncologists from developing a definitive connection between p53 status and

chemotherapy response 42.

The high prevalence of p53 mutation in human cancers suggests p53 is highly

detrimental to tumor cells. Indeed, genetic restoration of p53 in mouse models of

lymphoma, sarcoma, and liver carcinoma have demonstrated p53 activation induces

tumor regression 63, 105, 109. More specifically in lung cancer, regression of

adenocarcinomas and reversion of high grade adenocarcinomas to low grade adenomas

was observed in oncogenic Kras mouse models 26, 48. This findings support further

research into activating p53 as a general therapeutic intervention.

Several classes of compounds have been developed to reactivate mutant p53 or

induce wildtype p53 activity. One of the most notable compounds is nutlin-3, a small

molecule that binds to the p53 binding pocket of MDM2, preventing its interaction with

p53. As a consequence of nutlin-3 treatment, p53 promoter occupancy and expression of

p53 target genes increases 20, 91. Several studies have shown that nutlin-3-induced apoptosis in a p53-depenendet manner 66, 91, 94. Importantly, doses of nutlin-3 that are 19

sufficient to reduce growth of tumor xenografts are well tolerated by mice 54, 99, 104.

Summary

In the studies presented in this dissertation, I evaluate the role of the Caspase-2-

PIDDosome and its effector proteins in lung tumorigenesis, and chemotherapy response

in vivo and pathway regulation in vitro. My data demonstrate that Caspase-2 functions as

a tumor suppressor in a mouse model of oncogenic Kras-driven lung cancer, negatively

impacting chemotherapeutic response to cisplatin (Chapter 2); Loss of Pidd1 has no

direct impact on oncogenic Kras-driven lung tumorigenesis or chemotherapy response

(Chapter 3); and exogenous MDM2 is not sufficient to regulate of p53 functional activity

and PIDD1-induced cellular functions are not dependent on the p53:MDM2 interaction

(Chapter 4). In total, these findings contribute to the field of oncology by elucidating genetic determinants impacting tumorigenesis and their influence on chemotherapy response.

References

1 Andersen JL, Johnson CE, Freel CD, Parrish AB, Day JL, Buchakjian MR et al. Restraint of apoptosis during mitosis through interdomain phosphorylation of caspase-2. EMBO J 2009; 28: 3216-3227.

2 Ando K, Kernan JL, Liu PH, Sanda T, Logette E, Tschopp J et al. PIDD death- domain phosphorylation by ATM controls prodeath versus prosurvival PIDDosome signaling. Mol Cell 2012; 47: 681-693.

3 Baptiste-Okoh N, Barsotti AM, Prives C. Caspase 2 is both required for p53- mediated apoptosis and downregulated by p53 in a p21-dependent manner. Cell Cycle 2008; 7: 1133-1138.

4 Barak Y, Gottlieb E, Juven-Gershon T, Oren M. Regulation of mdm2 expression by p53: alternative promoters produce transcripts with nonidentical translation potential. Genes Dev 1994; 8: 1739-1749. 20

5 Barboza JA, Liu G, Ju Z, El-Naggar AK, Lozano G. p21 delays tumor onset by preservation of chromosomal stability. Proc Natl Acad Sci U S A 2006; 103: 19842-19847.

6 Bella J, Hindle KL, McEwan PA, Lovell SC. The leucine-rich repeat structure. Cell Mol Life Sci 2008; 65: 2307-2333.

7 Bergeron L, Perez GI, Macdonald G, Shi L, Sun Y, Jurisicova A et al. Defects in regulation of apoptosis in caspase-2-deficient mice. Genes Dev 1998; 12: 1304- 1314.

8 Berube C, Boucher LM, Ma W, Wakeham A, Salmena L, Hakem R et al. Apoptosis caused by p53-induced protein with death domain (PIDD) depends on the death adapter protein RAIDD. Proc Natl Acad Sci U S A 2005; 102: 14314- 14320.

9 Bieche I, Khodja A, Driouch K, Lidereau R. Genetic alteration mapping on chromosome 7 in primary breast cancer. Clin Cancer Res 1997; 3: 1009-1016.

10 Bock FJ, Peintner L, Tanzer M, Manzl C, Villunger A. P53-induced protein with a death domain (PIDD): master of puppets? Oncogene 2012; 31: 4733-4739.

11 Bock FJ, Krumschnabel G, Manzl C, Peintner L, Tanzer MC, Hermann-Kleiter N et al. Loss of PIDD limits NF-kappaB activation and cytokine production but not cell survival or transformation after DNA damage. Cell Death Differ 2013; 20: 546-557.

12 Bond GL, Hu W, Bond EE, Robins H, Lutzker SG, Arva NC et al. A single nucleotide polymorphism in the MDM2 promoter attenuates the p53 tumor suppressor pathway and accelerates tumor formation in humans. Cell 2004; 119: 591-602.

13 Bouchier-Hayes L, Green DR. Caspase-2: the orphan caspase. Cell Death Differ 2012; 19: 51-57.

14 Bouska A, Lushnikova T, Plaza S, Eischen CM. Mdm2 promotes genetic instability and transformation independent of p53. Mol Cell Biol 2008; 28: 4862- 4874.

15 Castedo M, Perfettini JL, Roumier T, Valent A, Raslova H, Yakushijin K et al. Mitotic catastrophe constitutes a special case of apoptosis whose suppression entails aneuploidy. Oncogene 2004; 23: 4362-4370.

16 Chen L, Li Z, Zwolinska AK, Smith MA, Cross B, Koomen J et al. MDM2 recruitment of lysine methyltransferases regulates p53 transcriptional output. EMBO J 2010; 29: 2538-2552.

17 Chen Z, Cheng K, Walton Z, Wang Y, Ebi H, Shimamura T et al. A murine lung cancer co-clinical trial identifies genetic modifiers of therapeutic response. Nature 2012; 483: 613-617.

18 Colussi PA, Harvey NL, Kumar S. Prodomain-dependent nuclear localization of the caspase-2 (Nedd2) precursor. A novel function for a caspase prodomain. J Biol Chem 1998; 273: 24535-24542.

19 Cremona CA, Behrens A. ATM signalling and cancer. Oncogene 2014; 33: 3351- 3360.

20 Cross B, Chen L, Cheng Q, Li B, Yuan ZM, Chen J. Inhibition of p53 DNA binding function by the MDM2 protein acidic domain. J Biol Chem 2011; 286: 16018-16029.

21 Delavaine L, La Thangue NB. Control of E2F activity by p21Waf1/Cip1. Oncogene 1999; 18: 5381-5392.

22 Dorstyn L, Puccini J, Wilson CH, Shalini S, Nicola M, Moore S et al. Caspase-2 deficiency promotes aberrant DNA-damage response and genetic instability. Cell Death Differ 2012; 19: 1288-1298.

23 Duan H, Dixit VM. RAIDD is a new 'death' adaptor molecule. Nature 1997; 385: 86-89.

24 Ettinger DS, Akerley W, Borghaei H, Chang AC, Cheney RT, Chirieac LR et al. Non-small cell lung cancer, version 2.2013. J Natl Compr Canc Netw 2013; 11: 645-653; quiz 653.

25 Fava LL, Bock FJ, Geley S, Villunger A. Caspase-2 at a glance. J Cell Sci 2012; 125: 5911-5915.

26 Feldser DM, Kostova KK, Winslow MM, Taylor SE, Cashman C, Whittaker CA et al. Stage-specific sensitivity to p53 restoration during lung cancer progression. Nature 2010; 468: 572-575.

27 Gajjar M, Candeias MM, Malbert-Colas L, Mazars A, Fujita J, Olivares-Illana V et al. The p53 mRNA-Mdm2 interaction controls Mdm2 nuclear trafficking and is required for p53 activation following DNA damage. Cancer Cell 2012; 21: 25-35.

28 Galluzzi L, Joza N, Tasdemir E, Maiuri MC, Hengartner M, Abrams JM et al. No death without life: vital functions of apoptotic effectors. Cell Death Differ 2008; 15: 1113-1123.

29 Garrett MD, Collins I. Anticancer therapy with checkpoint inhibitors: what, where and when? Trends Pharmacol Sci 2011; 32: 308-316.

30 Gu L, Zhu N, Zhang H, Durden DL, Feng Y, Zhou M. Regulation of XIAP translation and induction by MDM2 following irradiation. Cancer Cell 2009; 15: 363-375.

31 Guha M, Xia F, Raskett CM, Altieri DC. Caspase 2-mediated tumor suppression involves survivin gene silencing. Oncogene 2010; 29: 1280-1292.

32 Guo Y, Srinivasula SM, Druilhe A, Fernandes-Alnemri T, Alnemri ES. Caspase-2 induces apoptosis by releasing proapoptotic proteins from mitochondria. J Biol Chem 2002; 277: 13430-13437.

33 Hanel W, Marchenko N, Xu S, Yu SX, Weng W, Moll U. Two hot spot mutant p53 mouse models display differential gain of function in tumorigenesis. Cell Death Differ 2013; 20: 898-909.

34 Ho LH, Read SH, Dorstyn L, Lambrusco L, Kumar S. Caspase-2 is required for cell death induced by cytoskeletal disruption. Oncogene 2008; 27: 3393-3404.

35 Ho LH, Taylor R, Dorstyn L, Cakouros D, Bouillet P, Kumar S. A tumor suppressor function for caspase-2. Proc Natl Acad Sci U S A 2009; 106: 5336- 5341.

36 Honda R, Tanaka H, Yasuda H. Oncoprotein MDM2 is a ubiquitin ligase E3 for tumor suppressor p53. FEBS Lett 1997; 420: 25-27.

37 Honda R, Yasuda H. Association of p19(ARF) with Mdm2 inhibits ubiquitin ligase activity of Mdm2 for tumor suppressor p53. EMBO J 1999; 18: 22-27.

38 Imre G, Heering J, Takeda AN, Husmann M, Thiede B, zu Heringdorf DM et al. Caspase-2 is an initiator caspase responsible for pore-forming toxin-mediated apoptosis. EMBO J 2012; 31: 2615-2628.

39 Ito A, Lai CH, Zhao X, Saito S, Hamilton MH, Appella E et al. p300/CBP- mediated p53 acetylation is commonly induced by p53-activating agents and inhibited by MDM2. EMBO J 2001; 20: 1331-1340.

40 Jackson EL, Willis N, Mercer K, Bronson RT, Crowley D, Montoya R et al. Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev 2001; 15: 3243-3248.

41 Jackson EL, Olive KP, Tuveson DA, Bronson R, Crowley D, Brown M et al. The differential effects of mutant p53 alleles on advanced murine lung cancer. Cancer Res 2005; 65: 10280-10288.

42 Jackson JG, Post SM, Lozano G. Regulation of tissue- and stimulus-specific cell fate decisions by p53 in vivo. J Pathol 2011; 223: 127-136.

43 Jackson JG, Pant V, Li Q, Chang LL, Quintas-Cardama A, Garza D et al. p53- mediated senescence impairs the apoptotic response to chemotherapy and clinical outcome in breast cancer. Cancer Cell 2012; 21: 793-806.

44 Janssens S, Tinel A, Lippens S, Tschopp J. PIDD mediates NF-kappaB activation in response to DNA damage. Cell 2005; 123: 1079-1092.

45 Jelinek M, Balusikova K, Kopperova D, Nemcova-Furstova V, Sramek J, Fidlerova J et al. Caspase-2 is involved in cell death induction by taxanes in breast cancer cells. Cancer Cell Int 2013; 13: 42.

46 Jeon YJ, Jo MG, Yoo HM, Hong SH, Park JM, Ka SH et al. Chemosensitivity is controlled by p63 modification with ubiquitin-like protein ISG15. J Clin Invest 2012; 122: 2622-2636.

47 Jones SN, Hancock AR, Vogel H, Donehower LA, Bradley A. Overexpression of Mdm2 in mice reveals a p53-independent role for Mdm2 in tumorigenesis. Proc Natl Acad Sci U S A 1998; 95: 15608-15612.

48 Junttila MR, Karnezis AN, Garcia D, Madriles F, Kortlever RM, Rostker F et al. Selective activation of p53-mediated tumour suppression in high-grade tumours. Nature 2010; 468: 567-571.

49 Kalemkerian GP, Akerley W, Bogner P, Borghaei H, Chow LQ, Downey RJ et al. Small cell lung cancer. J Natl Compr Canc Netw 2013; 11: 78-98.

50 Kim IR, Murakami K, Chen NJ, Saibil SD, Matysiak-Zablocki E, Elford AR et al. DNA damage- and stress-induced apoptosis occurs independently of PIDD. Apoptosis 2009; 14: 1039-1049.

51 Lamkanfi M, Declercq W, Kalai M, Saelens X, Vandenabeele P. Alice in caspase land. A phylogenetic analysis of caspases from worm to man. Cell Death Differ 2002; 9: 358-361.

52 Lane DP. Cancer. p53, guardian of the genome. Nature 1992; 358: 15-16.

53 Lassus P, Opitz-Araya X, Lazebnik Y. Requirement for caspase-2 in stress- induced apoptosis before mitochondrial permeabilization. Science 2002; 297: 1352-1354.

54 Laurie NA, Donovan SL, Shih CS, Zhang J, Mills N, Fuller C et al. Inactivation of the p53 pathway in retinoblastoma. Nature 2006; 444: 61-66.

55 Lee JS, Pathak S, Hopwood V, Tomasovic B, Mullins TD, Baker FL et al. Involvement of chromosome 7 in primary lung tumor and nonmalignant normal lung tissue. Cancer Res 1987; 47: 6349-6352.

56 Li Q, Lozano G. Molecular pathways: targeting Mdm2 and Mdm4 in cancer therapy. Clin Cancer Res 2013; 19: 34-41.

57 Lin Y, Ma W, Benchimol S. Pidd, a new death-domain-containing protein, is induced by p53 and promotes apoptosis. Nat Genet 2000; 26: 122-127.

58 Logette E, Schuepbach-Mallepell S, Eckert MJ, Leo XH, Jaccard B, Manzl C et al. PIDD orchestrates translesion DNA synthesis in response to UV irradiation. Cell Death Differ 2011; 18: 1036-1045.

59 Luo X, Budihardjo I, Zou H, Slaughter C, Wang X. Bid, a Bcl2 interacting protein, mediates cytochrome c release from mitochondria in response to activation of cell surface death receptors. Cell 1998; 94: 481-490.

60 Manzl C, Krumschnabel G, Bock F, Sohm B, Labi V, Baumgartner F et al. Caspase-2 activation in the absence of PIDDosome formation. J Cell Biol 2009; 185: 291-303.

61 Manzl C, Peintner L, Krumschnabel G, Bock F, Labi V, Drach M et al. PIDDosome-independent tumor suppression by Caspase-2. Cell Death Differ 2012; 19: 1722-1732.

62 Manzl C, Fava LL, Krumschnabel G, Peintner L, Tanzer MC, Soratroi C et al. Death of p53-defective cells triggered by forced mitotic entry in the presence of DNA damage is not uniquely dependent on Caspase-2 or the PIDDosome. Cell Death Dis 2013; 4: e942.

63 Martins CP, Brown-Swigart L, Evan GI. Modeling the therapeutic efficacy of p53 restoration in tumors. Cell 2006; 127: 1323-1334.

64 Mascaux C, Iannino N, Martin B, Paesmans M, Berghmans T, Dusart M et al. The role of RAS oncogene in survival of patients with lung cancer: a systematic review of the literature with meta-analysis. Br J Cancer 2005; 92: 131-139.

65 Mhaidat NM, Wang Y, Kiejda KA, Zhang XD, Hersey P. Docetaxel-induced apoptosis in melanoma cells is dependent on activation of caspase-2. Mol Cancer Ther 2007; 6: 752-761.

66 Mir R, Tortosa A, Martinez-Soler F, Vidal A, Condom E, Perez-Perarnau A et al. Mdm2 antagonists induce apoptosis and synergize with cisplatin overcoming chemoresistance in TP53 wild-type ovarian cancer cells. Int J Cancer 2013; 132: 1525-1536.

67 Mitsudomi T, Hamajima N, Ogawa M, Takahashi T. Prognostic significance of p53 alterations in patients with non-small cell lung cancer: a meta-analysis. Clin Cancer Res 2000; 6: 4055-4063.

68 Momand J, Zambetti GP, Olson DC, George D, Levine AJ. The mdm-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell 1992; 69: 1237-1245.

69 Montes de Oca Luna R, Wagner DS, Lozano G. Rescue of early embryonic lethality in mdm2-deficient mice by deletion of p53. Nature 1995; 378: 203-206.

70 Nicholson SA, Okby NT, Khan MA, Welsh JA, McMenamin MG, Travis WD et al. Alterations of p14ARF, p53, and p73 genes involved in the E2F-1-mediated apoptotic pathways in non-small cell lung carcinoma. Cancer Res 2001; 61: 5636- 5643.

71 Nutt LK, Margolis SS, Jensen M, Herman CE, Dunphy WG, Rathmell JC et al. Metabolic regulation of oocyte cell death through the CaMKII-mediated phosphorylation of caspase-2. Cell 2005; 123: 89-103.

72 Nutt LK, Buchakjian MR, Gan E, Darbandi R, Yoon SY, Wu JQ et al. Metabolic control of oocyte apoptosis mediated by 14-3-3zeta-regulated dephosphorylation of caspase-2. Dev Cell 2009; 16: 856-866.

73 Oliner JD, Pietenpol JA, Thiagalingam S, Gyuris J, Kinzler KW, Vogelstein B. Oncoprotein MDM2 conceals the activation domain of tumour suppressor p53. Nature 1993; 362: 857-860.

74 Oliver TG, Mercer KL, Sayles LC, Burke JR, Mendus D, Lovejoy KS et al. Chronic cisplatin treatment promotes enhanced damage repair and tumor progression in a mouse model of lung cancer. Genes Dev 2010; 24: 837-852.

75 Oliver TG, Meylan E, Chang GP, Xue W, Burke JR, Humpton TJ et al. Caspase- 2-mediated cleavage of Mdm2 creates a p53-induced positive feedback loop. Mol Cell 2011; 43: 57-71.

76 Olivier M, Hollstein M, Hainaut P. TP53 mutations in human cancers: origins, consequences, and clinical use. Cold Spring Harb Perspect Biol 2010; 2: a001008.

77 Olsson M, Vakifahmetoglu H, Abruzzo PM, Hogstrand K, Grandien A, Zhivotovsky B. DISC-mediated activation of caspase-2 in DNA damage-induced apoptosis. Oncogene 2009; 28: 1949-1959.

78 Onder TT, Gupta PB, Mani SA, Yang J, Lander ES, Weinberg RA. Loss of E- cadherin promotes metastasis via multiple downstream transcriptional pathways. Cancer Res 2008; 68: 3645-3654.

79 Oren M, Rotter V. Mutant p53 gain-of-function in cancer. Cold Spring Harb Perspect Biol 2010; 2: a001107.

80 Pant V, Xiong S, Jackson JG, Post SM, Abbas HA, Quintas-Cardama A et al. The p53-Mdm2 feedback loop protects against DNA damage by inhibiting p53 activity but is dispensable for p53 stability, development, and longevity. Genes Dev 2013; 27: 1857-1867.

81 Pao W, Girard N. New driver mutations in non-small-cell lung cancer. Lancet Oncol 2011; 12: 175-180.

82 Parsons MJ, McCormick L, Janke L, Howard A, Bouchier-Hayes L, Green DR. Genetic deletion of caspase-2 accelerates MMTV/c-neu-driven mammary carcinogenesis in mice. Cell Death Differ 2013; 20: 1174-1182.

83 Pochampally R, Fodera B, Chen L, Shao W, Levine EA, Chen J. A 60 kd MDM2 isoform is produced by caspase cleavage in non-apoptotic tumor cells. Oncogene 1998; 17: 2629-2636.

84 Pop C, Salvesen GS. Human caspases: activation, specificity, and regulation. J Biol Chem 2009; 284: 21777-21781.

85 Purvis JE, Karhohs KW, Mock C, Batchelor E, Loewer A, Lahav G. p53 dynamics control cell fate. Science 2012; 336: 1440-1444.

86 Rayburn E, Zhang R, He J, Wang H. MDM2 and human malignancies: expression, clinical pathology, prognostic markers, and implications for chemotherapy. Curr Cancer Drug Targets 2005; 5: 27-41.

87 Ren K, Lu J, Porollo A, Du C. Tumor-suppressing function of caspase-2 requires catalytic site Cys-320 and site Ser-139 in mice. J Biol Chem 2012; 287: 14792- 14802.

88 Riley T, Sontag E, Chen P, Levine A. Transcriptional control of human p53- regulated genes. Nat Rev Mol Cell Biol 2008; 9: 402-412.

89 Robertson JD, Enoksson M, Suomela M, Zhivotovsky B, Orrenius S. Caspase-2 acts upstream of mitochondria to promote cytochrome c release during etoposide- induced apoptosis. J Biol Chem 2002; 277: 29803-29809.

90 Rowley JD. Nonrandom chromosomal abnormalities in hematologic disorders of man. Proc Natl Acad Sci U S A 1975; 72: 152-156.

91 Saha MN, Jiang H, Chang H. Molecular mechanisms of nutlin-induced apoptosis in multiple myeloma: evidence for p53-transcription-dependent and -independent pathways. Cancer Biol Ther 2010; 10: 567-578.

92 Shalini S, Dorstyn L, Wilson C, Puccini J, Ho L, Kumar S. Impaired antioxidant defence and accumulation of oxidative stress in caspase-2-deficient mice. Cell Death Differ 2012; 19: 1370-1380. 27

93 Sohn D, Budach W, Janicke RU. Caspase-2 is required for DNA damage-induced expression of the CDK inhibitor p21(WAF1/CIP1). Cell Death Differ 2011; 18: 1664-1674.

94 Stuhmer T, Chatterjee M, Hildebrandt M, Herrmann P, Gollasch H, Gerecke C et al. Nongenotoxic activation of the p53 pathway as a therapeutic strategy for multiple myeloma. Blood 2005; 106: 3609-3617.

95 Sweet-Cordero A, Mukherjee S, Subramanian A, You H, Roix JJ, Ladd-Acosta C et al. An oncogenic KRAS2 expression signature identified by cross-species gene-expression analysis. Nat Genet 2005; 37: 48-55.

96 Takahashi T, Nau MM, Chiba I, Birrer MJ, Rosenberg RK, Vinocour M et al. p53: a frequent target for genetic abnormalities in lung cancer. Science 1989; 246: 491-494.

97 Tinel A, Tschopp J. The PIDDosome, a protein complex implicated in activation of caspase-2 in response to genotoxic stress. Science 2004; 304: 843-846.

98 Tinel A, Janssens S, Lippens S, Cuenin S, Logette E, Jaccard B et al. Autoproteolysis of PIDD marks the bifurcation between pro-death caspase-2 and pro-survival NF-kappaB pathway. EMBO J 2007; 26: 197-208.

99 Tovar C, Graves B, Packman K, Filipovic Z, Higgins B, Xia M et al. MDM2 small-molecule antagonist RG7112 activates p53 signaling and regresses human tumors in preclinical cancer models. Cancer Res 2013; 73: 2587-2597.

100 Tu S, McStay GP, Boucher LM, Mak T, Beere HM, Green DR. In situ trapping of activated initiator caspases reveals a role for caspase-2 in heat shock-induced apoptosis. Nat Cell Biol 2006; 8: 72-77.

101 Uchida C, Miwa S, Kitagawa K, Hattori T, Isobe T, Otani S et al. Enhanced Mdm2 activity inhibits pRB function via ubiquitin-dependent degradation. EMBO J 2005; 24: 160-169.

102 Van de Craen M, Declercq W, Van den brande I, Fiers W, Vandenabeele P. The proteolytic procaspase activation network: an in vitro analysis. Cell Death Differ 1999; 6: 1117-1124.

103 Vaseva AV, Moll UM. The mitochondrial p53 pathway. Biochim Biophys Acta 2009; 1787: 414-420.

104 Vassilev LT, Vu BT, Graves B, Carvajal D, Podlaski F, Filipovic Z et al. In vivo activation of the p53 pathway by small-molecule antagonists of MDM2. Science 2004; 303: 844-848.

105 Ventura A, Kirsch DG, McLaughlin ME, Tuveson DA, Grimm J, Lintault L et al. Restoration of p53 function leads to tumour regression in vivo. Nature 2007; 445: 661-665.

106 Wang R, Wei Z, Jin H, Wu H, Yu C, Wen W et al. Autoinhibition of UNC5b revealed by the cytoplasmic domain structure of the receptor. Mol Cell 2009; 33: 692-703.

107 Wang Y, Blandino G, Oren M, Givol D. Induced p53 expression in lung cancer cell line promotes cell senescence and differentially modifies the cytotoxicity of anti-cancer drugs. Oncogene 1998; 17: 1923-1930.

108 Xirodimas D, Saville MK, Edling C, Lane DP, Lain S. Different effects of p14ARF on the levels of ubiquitinated p53 and Mdm2 in vivo. Oncogene 2001; 20: 4972-4983.

109 Xue W, Zender L, Miething C, Dickins RA, Hernando E, Krizhanovsky V et al. Senescence and tumour clearance is triggered by p53 restoration in murine liver carcinomas. Nature 2007; 445: 656-660.

110 Yang JY, Zong CS, Xia W, Wei Y, Ali-Seyed M, Li Z et al. MDM2 promotes cell motility and invasiveness by regulating E-cadherin degradation. Mol Cell Biol 2006; 26: 7269-7282.

111 Yasunaga M, Ipsaro JJ, Mondragon A. Structurally similar but functionally diverse ZU5 domains in human erythrocyte ankyrin. J Mol Biol 2012; 417: 336- 350.

112 Yoo NJ, Lee JW, Kim YJ, Soung YH, Kim SY, Nam SW et al. Loss of caspase-2, -6 and -7 expression in gastric cancers. APMIS 2004; 112: 330-335.

113 Zhang Y, Padalecki SS, Chaudhuri AR, De Waal E, Goins BA, Grubbs B et al. Caspase-2 deficiency enhances aging-related traits in mice. Mech Ageing Dev 2007; 128: 213-221.

114 Zhang Z, Wang H, Li M, Agrawal S, Chen X, Zhang R. MDM2 is a negative regulator of p21WAF1/CIP1, independent of p53. J Biol Chem 2004; 279: 16000- 16006.

115 Zhao R, Gish K, Murphy M, Yin Y, Notterman D, Hoffman WH et al. Analysis of p53-regulated gene expression patterns using oligonucleotide arrays. Genes Dev 2000; 14: 981-993.

116 Zhuo W, Zhang L, Zhu B, Ling J, Chen Z. Association of MDM2 SNP309 variation with lung cancer risk: evidence from 7196 cases and 8456 controls. PLoS One 2012; 7: e41546.

117 Zohrabian VM, Nandu H, Gulati N, Khitrov G, Zhao C, Mohan A et al. Gene expression profiling of metastatic brain cancer. Oncol Rep 2007; 18: 321-328.

Figure 1.1: Caspase-2-PIDDosome positive feedback loop.

(A) Under unstressed conditions, MDM2 and p53 form a negative feedback loop. 1) p53 transcriptionally activates gene expression of Mdm2, an E3 ubiquitin ligase. 2) MDM2 binds p53 directly and ubiquitinates p53. 3) Poly-ubiquitinated p53 is targeted for proteasomal degradation. Basal expression of p53 and MDM2 are low in unstressed cells. (B) In response to various stimuli, p53 is activated. 1) Activated p53 transcriptionally activates gene expression of Pidd1. 2) PIDD1 interacts with RAIDD and Caspase-2 to form the Caspase-2-PIDDosome, thus activating Caspase-2 enzymatic activity. 3) Activated Caspase-2 cleaves MDM2 into two fragments: p60 and p30. The p60 fragment binds to p53, but lacks the E3 ubiquitin ligase function of the RING domain separated with the p30 fragment. 4) Relieved MDM2 inhibition leads to increased p53 stability and activation of p53 target genes such as Pidd1, thus completing the positive feedback loop.

Figure 1.2: p53 and p21 negatively feedback on Caspase-2 expression. p53 transcriptionally activates expression of CDKN1A, encoding p21. Caspase-2 binds to the 3’ UTR of p21 mRNA promoting its stability and enhances translation. Both p53 and p21 protein induce transcriptional repression of Caspase-2. CHAPTER 2

CASPASE-2 IMPACTS LUNG TUMORIGENESIS AND CHEMOTHERAPY

RESPONSE IN VIVO

Preface

This chapter is a re-formatted version of a manuscript accepted, but not yet published at Cell Death and Differentiation in September 2014. The work was performed with the following co-authors: Rahul Arya, Anandaroop Mukhopadhyay, Kristofer C.

Berrett, Phillip M. Clair, Benjamin Witt, Mohamed E. Salama, Arjun Bhutkar, and Trudy

G. Oliver. I made significant contributions to the manuscript, including by not limited to conceptual design, executation of experiments, interpretation of results, and preparation of written material.

Introduction

Caspase-2 is an atypical member of the cysteine-aspartic acid protease family.

Although Caspase-2 is among the first caspases identified and is highly conserved across species, its function is not fully understood5, 16. Studies have focused on the role of

Caspase-2 in apoptosis; however, mice lacking Caspase-2 are viable, fertile, and undergo

normal levels of apoptosis, with the exception of a limited number of cell types3. Few

endogenous Caspase-2 cleavage targets have been identified, making it difficult to fully 33

understand Caspase-2 function 15. More recently, it has become appreciated that Caspase-

2 has nonapoptotic roles in cell cycle arrest, genome stability, and tumor suppression9, 10,

25, yet the mechanisms behind these processes are incompletely understood. Loss-of-

function mutations and downregulation of Caspase-2 have been observed in a variety of hematological and solid malignancies14, 26. In mouse models of Eμ-Myc-driven lymphoma and c-Neu-driven mammary carcinoma, Caspase-2 functions as a tumor suppressor10, 23.

In SV40/Kras-transformed mouse embryonic fibroblasts, Caspase-2-mediated tumor suppression requires catalytic activity26. Given that Myc, c-Neu, and Kras function in diverse signaling pathways, Caspase-2 may act as a general tumor suppressor that

cooperates nonspecifically with oncogenes much like the tumor suppressor p53. The

tumor suppressor function of Caspase-2 is dependent on p53, and p53 induction is

impaired in Caspase-2-deficient cells, suggesting Caspase-2 and p53 act in the same

pathway 9, 1911.

p53 is a transcription factor that mediates the response to a wide variety of

cellular stresses and promotes diverse responses, including cell cycle arrest, DNA

damage repair, or apoptosis 17. Because of these crucial roles, p53 is the most commonly mutated gene across all cancer types, including lung cancer 7. Loss of p53 accelerates tumor development, promotes genomic instability and is associated with poor outcome11,

20. Although the role of p53 in chemotherapy response is controversial, in both mice and

humans, p53 loss can improve response to chemotherapy in multiple tissues4, 6, 12. Our

previous work in Kras-driven lung tumors suggests that cancer cells utilize wildtype p53

as a resistance mechanism to chronic cisplatin treatment21.

Upon DNA damage, p53 induces expression of p53-induced protein with a death

domain 1 (PIDD1)18. PIDD1 promotes the assembly of a complex called the Caspase-2- 34

PIDDosome, leading to autocatalytic cleavage and activation of Caspase-227. We

previously showed that Mdm2, a master regulator and inhibitor of p53, is a cleavage

target of Caspase-222. Caspase-2 cleaves the C-terminal E3-ubiquitin ligase domain of

Mdm2, preventing Mdm2 from targeting p53 for proteasomal degradation22. Cleavage of

Mdm2 increases protein levels of p53, elevates expression of p53 target genes including

the cyclin-dependent kinase inhibitor p21, and induces cell cycle arrest22. It is predicted

that Caspase-2 loss leads to increased Mdm2-mediated degradation of p53, ultimately

serving as a direct mechanism to explain how Caspase-2 acts as a p53-dependent tumor

suppressor. Interestingly, in response to DNA damage, Pidd1 is upregulated in chemo- resistant Kras-driven lung tumors, presumably leading to higher Caspase-2 activity and p53-mediated cell cycle arrest21. However, the role of Caspase-2 in Kras-driven lung

tumor development and chemotherapy response has not yet been addressed.

Here, we investigate the role of Caspase-2 in KrasG12D-driven lung tumorigenesis and chemotherapy response in vivo.

Results

Caspase-2 levels are significantly reduced in human lung

adenocarcinoma with wildtype p53

Caspase-2 has been implicated as a tumor suppressor in mouse models of breast

cancer and lymphoma10, 23, but its role in lung cancer has been largely unexplored. Our

previous work in lung cancer demonstrated that Caspase-2 directly enhances p53 function

through Mdm2 cleavage22. Based on this model, we reasoned that selective pressure to

lose Caspase-2 may occur preferentially in p53-wildtype lung tumors. To address this, we analyzed data from a large cohort of human lung adenocarcinomas from The Cancer 35

Genome Atlas (TCGA) publicly available data portal, which contains both somatic

mutation calls and gene expression profiles. Caspase-2 protein-altering mutations were

detected in 2% of human lung adenocarcinomas (11/538 tumors with somatic mutations).

Protein-altering mutations in p53 were observed in 53% of tumors and Caspase-2 mutant

tumors were marginally enriched for p53 mutations (9/11 tumors, p = 0.0398,

hypergeometric test). In addition to Caspase-2 mutations, however, global

downregulation of Caspase-2 mRNA has been observed in human cancer26. To probe the

association between Caspase-2 expression levels and p53 mutation status, lung

adenocarcinomas (459 tumors with mutation and gene expression data) were stratified

based on p53 mutation status. We observed that 47% of tumors (217 of 459) were p53-

wildtype versus 53% of tumors (242 of 459) that harbored protein-altering p53 mutations.

Caspase-2 levels were significantly lower in p53-wildtype tumors compared to p53

mutant tumors (Figure 2.1A). As controls, we observed that p53 target genes, p21 and

Mdm2, were significantly enriched in p53-wildtype tumors (Figure 2.1BC). We also

investigated expression of components of the Caspase-2-PIDDosome, Pidd1 and

Raidd/Cradd. Pidd1 expression was not associated with p53 status, whereas Raidd

expression was enriched in p53-wildtype tumors (Supplementary Figure 2.S1). These

data suggest that Caspase-2 mRNA downregulation is selectively enriched in p53- wildtype tumors, consistent with the model that Caspase-2 functions as a p53-dependent tumor suppressor through Mdm2 cleavage.

Caspase-2 is a tumor suppressor in Kras-driven lung cancer

Next, we sought to determine the impact of Caspase-2 loss using a mouse model

of lung adenocarcinoma, the Lox-Stop-Lox (LSL)-KrasG12D/+ mice28. Lung tumors were 36

generated in LSL-KrasG12D/+ mice that were Casp2-/- (KC2-/-), Casp2+/- (KC2+/-), or

Casp2+/+ (KC2+/+) by intranasal administration of adenovirus carrying Cre recombinase

(AdCre). Mice were sacrificed 12 weeks after tumor initiation and the average tumor

size, number, and percent tumor burden (tumor area/total lung area) were determined by

quantifying hematoxylin and eosin (H&E)-stained lung sections. KC2-/- mice harbored more lung tumors with an increased average size, contributing to a significantly higher tumor burden compared to KC2+/- and KC2+/+ littermates (Figure 2.2A-C, G-I). In addition, tumors in KC2-/- mice were histologically more advanced compared to KC2+/- and KC2+/+ tumors (Figure 2.2D-F). KC2-/- tumors were mostly high grade (>60% grade 3) as opposed to KC2+/- and KC2+/+ tumors, which were mostly low grade

(100% and >96% grade 1/2, respectively) (Figure 2.3A). Interestingly, we observed large tumors with necrotic centers specifically in KC2-/- lungs, which we have not previously observed in KrasG12D-only lungs. We compared tumors of similar size from each

genotype as well as KrasG12D/p53 null lungs that produce higher grade tumors, and

Caspase-2 null tumors had significantly more necrotic centers than either genotype

(Figure 2.3B-F). In total, nine tumors with necrotic centers were identified in 20 KC2-/-

animals (45% frequency). To determine why tumors were larger and more advanced in

the absence of Caspase-2, we quantified tumors for cellular proliferation using Ki67

antibodies. Tumors in KC2-/- mice exhibited increased proliferation compared to KC2+/-

and KC2+/+ tumors (Figure 2.3G-J). Similar analyses using BrdU incorporation

demonstrated a trend of increased proliferation in KC2-/- lungs, but these data were not

quite significant (p = 0.056, Supplementary Figure 2.S3A). To test whether loss of

Caspase-2 impacts basal apoptotic rates, we stained tumor sections using antibodies to

cleaved Caspase-3 (CC3) or via TUNEL. We observed low basal apoptotic rates that 37

were not significantly altered upon Caspase-2 loss (Supplementary Figure 2.S3BC). High

lung tumor burden in KC2-/- mice was associated with a reduction in overall survival

compared to heterozygous but not wildtype littermates (Supplementary Figure 2.S3D).

Taken together, these data suggest that Caspase-2 functions as a tumor suppressor in lung cancer by regulating cellular proliferation.

Caspase-2-deficient tumors have a reduction in p53 target genes

but respond to chemotherapy in vivo

We previously demonstrated that activation of the Caspase-2-PIDDosome

promotes chemotherapy resistance in human lung cancer cells, suggesting that loss of

Caspase-2 may enhance chemo-sensitivity. To test whether Caspase-2 impacts the response to acute genotoxic stress, mice were treated with a single dose of cisplatin

(7mg/kg) 12 weeks after tumor initiation. We have observed that BrdU incorporation is a sensitive measure of cell cycle arrest following chemotherapy treatment, whereas Ki67 staining is not since Ki67 label can be retained in nonproliferating cells for a considerable time period 29. Ki67 staining, however, is more sensitive at detecting basal changes in

proliferation, likely because BrdU labels considerably fewer cells in this protocol. To assess proliferation changes following cisplatin treatment, we analyzed BrdU incorporation and found that cisplatin treatment significantly reduced cellular proliferation regardless of Caspase-2 status (Figure 2.4A). Whereas BrdU analysis did not detect significant differences in basal tumor proliferation based on Caspase-2 genotype, Ki67 staining of tumors from the same animals revealed a significant increase in tumor proliferation in KC2-/- mice (Supplementary Figure 2.S4A). Consistently, we observed a significant induction of p21 in tumors from mice treated with cisplatin 38

regardless of genotype (Figure 2.4B). In response to cisplatin, we also observed an increase in CC3-positive cells in KC2-/- mice that was not observed in KC2+/+ or

KC2+/- mice at this time point (Figure 2.4C). Expression of the pro-apoptotic gene Bax

was significantly induced in both KC2+/+ and KC2-/- tumors (Figure 2.4D). These data

demonstrate that Caspase-2-deficiency does not hinder initial cell cycle arrest and

apoptosis in response to chemotherapy, and may even sensitize tumors to apoptosis.

Our previous work demonstrated that the Caspase-2-PIDDosome promotes p53

activity in a positive feedback loop22. Therefore, we predicted that loss of Caspase-2 may

be associated with a reduction in p53 activity. To address this, we analyzed expression of

additional p53 target genes in KC2+/+ and KC2-/- tumors treated with or without a

single dose of cisplatin by real-time RT-PCR. Apoptotic genes including Puma and Noxa

were induced by cisplatin treatment, and to a significant level in KC2-/- tumors. Although

basal levels of apoptotic genes tended to be lower in KC2-/- tumors, these data were not

significant (Figure 2.4EF). Genes involved in cell cycle arrest and DNA damage repair,

including p21, cyclin G1, and Msh2, were significantly lower in KC2-/- tumors at the

basal level (Figure 2.4BGH). However, p53 target genes controlling cell cycle arrest and

DNA repair were induced upon cisplatin treatment in KC2-/- tumors. Together, these data

suggest that basal p53 activity is significantly dampened in the absence of Caspase-2, but

that upon DNA damage, KC2-/- tumors are capable of inducing p53 target genes to levels

similar to KC2+/+ tumors.

These data suggest that Caspase-2-deficient tumors are histogically advanced but

maintain p53. To test this directly, we examined levels of p53 mRNA, which did not

change upon cisplatin treatment or by Caspase-2 genotype (Figure 2.4I). To determine

whether p53 had acquired point mutations, we sequenced p53 mRNA from macro- 39

dissected lung tumors (n = 9-11 tumors per genotype) from KC2+/+ and KC2-/- mice or normal lung control. We did not detect p53 mutations in tumors regardless of Caspase-2 genotype suggesting that Caspase-2-deficient tumors maintain wildtype p53

(Supplementary Figure 2.S4B).

Caspase-2-deficient tumors rapidly resume proliferation

following chemotherapy

To further investigate cisplatin response in KC2-/- mice, we used micro-computed tomography (microCT) imaging to analyze the response of individual tumors longitudinally. Mice were imaged before and after two doses of cisplatin (one dose per week) and individual tumor volumes were quantified (representative 3D reconstructions,

Figure 2.5A). Following two doses of cisplatin treatment, lung tumors demonstrated significant growth inhibition irrespective of genotype (Figure 2.5B). Thus, while loss of

Caspase-2 accelerates tumorigenesis, Caspase-2-deficient tumors maintain chemo-

sensitivity similar to mice lacking the tumor suppressor p53 or p2112, 21.

To determine the effect of Caspase-2 on long-term cisplatin response, we used a

previously established regimen of cisplatin treatment whereby mice are treated with two

doses of cisplatin followed by a two-week recovery period before receiving two

additional doses of cisplatin (four total doses)21. We analyzed tumor volume change

before and after the second cycle of cisplatin (doses 3 and 4) by microCT imaging

(Figure 2.5B). Cisplatin-treated tumors in KC2-/- mice displayed a significant reduction in tumor volume in response to the second cycle of cisplatin similar to KC2+/+ tumors, demonstrating that KC2-/- tumors maintain sensitivity to chemotherapy over repeated treatment (Figure 2.5B). Next, we analyzed tumor growth during the two-week drug-free 40

recovery period in between doses 2 and 3 (Figure 2.5B). This period is similar to the

“drug holiday” given to cancer patients to alleviate the negative side effects associated

with therapy. Surprisingly, KC2-/- tumors had a significant increase in tumor volume

during the recovery phase, whereas no change was detected during this time period in

KC2+/+ littermates (Figure 2.5B). Consistently, lung tumors in cisplatin-treated KC2-/-

mice proliferated significantly more than KC2+/+ and KC2+/- tumors during the

recovery period as determined by BrdU incorporation (Figure 2.5C). These findings

demonstrate that tumors in KC2-/- mice rapidly re-enter the cell cycle following cisplatin

treatment. Loss of Caspase-2, therefore, may limit long-term therapeutic benefit due to

enhanced proliferation and cellular recovery following chemotherapy.

Loss of Caspase-2 impacts long-term chemotherapy treatment

Since Caspase-2 loss promoted a rapid rebound following chemotherapy

response, we sought to determine the long-term impact of Caspase-2 loss on tumor

burden. We treated KC2+/+ and KC2-/- mice with four doses of PBS control or

cisplatin, and then sacrificed mice after the fourth and final treatment. Lung tumor

volume was quantified by microCT imaging prior to the first treatment and 5 days

following the fourth treatment (Figure 2.6A). Untreated tumors from both KC2+/+ and

KC2-/- mice grew over time, and cisplatin treatment caused a significant reduction in

tumor growth in both genotypes (Figure 2.6B). To take into consideration the total tumor

burden, H&E-stained sections were analyzed for total tumor burden in an independent

cohort of mice sacrificed 72 hours after the fourth dose of cisplatin. Cisplatin caused a

reduction in tumor burden irrespective of genotype (Figure 2.6CD). However, KC2-/- mice exhibited significantly greater tumor burden following four doses of cisplatin 41

compared to KC2+/+ and KC2+/- littermates (10% versus 2%, respectively; Figure

2.6D). In addition, tumors remaining in KC2-/- mice following cisplatin treatment were

significantly larger than KC2+/+ tumors (Figure 2.6E). Together, these data demonstrate

that Caspase-2 loss impedes the long-term benefit of cisplatin treatment in Kras-driven

lung tumors.

PIDD1 phosphorylation regulates Mdm2 cleavage and p53 activity

Our previous work demonstrated that high basal phosphorylation of the

checkpoint kinase Chk2 and genomic instability were associated with cisplatin-resistant

lung tumors in this model21. High basal phospho-Chk2 suggests that resistant tumors may

have increased ATM activity, potentially as a result of genomic instability. Recently,

Ando et al. showed that ATM phosphorylates PIDD1 on Thr788 and this phosphorylation

is critical for Caspase-2-PIDDosome formation1. Activation of the Caspase-2-

PIDDosome promotes Mdm2 cleavage and resistance to a variety of genotoxic agents in a p53-dependent manner22. Together, this suggests that an ATM-Caspase-2 signaling pathway may regulate chemo-resistance in p53-wildtype lung tumor cells. Whether

PIDD1 phosphorylation impacts Caspase-2 cleavage targets and the biological response

of p53-wildtype cells has not been tested. To investigate the impact of PIDD1

phosphorylation on Mdm2 cleavage, we stably expressed PIDD1 mutants that either

mimic or prevent T788 phosphorylation in p53-wildtype lung cancer cell lines A549 and

SW1573 using retroviral infection (Figure 2.7A). Wildtype and phospho-mimetic PIDD1

T788D, but not phospho-mutant PIDD1 T788A, were sufficient to activate Caspase-2 as

demonstrated by the absence of pro-Caspase-2 (Figure 2.7A). Importantly, only wildtype

PIDD1 and PIDD T788D led to accumulation of the Mdm2 cleavage product, p60 42

(Figure 2.7A). As predicted, Mdm2 cleavage was associated with increased p53 and p21 protein levels (Figure 2.7A). Upon PIDD1 expression, cells exhibited undetectable or low levels of PARP cleavage (Figure 2.7A), but this did not lead to apoptosis as assessed by subG1 cell cycle analysis (Supplementary Figure 2.S7A). Consistent with p21 induction, expression of wildtype and T788D PIDD1, but not T788A, led to G1-cell cycle arrest as measured by propidium iodide (PI) staining and flow cytometry (Figure 2.7B). Finally, expression of PIDD1 T788D led to enhanced cell viability in the presence of cisplatin similar to wildtype PIDD1 (Figure 2.7C). These data demonstrate that PIDD1 phosphorylation directly impacts Mdm2 cleavage, p53 activity, and chemotherapy response.

To investigate the requirement for p53 in chemo-resistance induced by Caspase-2-

PIDDosome activation, we performed the same experiments in the p53 isogenic cell lines, HCT116 p53+/+ and p53-/-. Wildtype and T788D PIDD1 promoted Caspase-2 activation and Mdm2 cleavage regardless of p53 status. In the absence of p53, cleaved

Mdm2 cannot promote p53 activation and thus, p21 induction is abolished (Figure 2.7D).

These findings support a critical role for PIDD1 T788 phosphorylation in Caspase-2-

mediated Mdm2 cleavage and p53-dependent drug resistance.

DNA damage activates p53, leading to induction of PIDD1 expression, activation of the Caspase-2-PIDDosome, and Mdm2 cleavage22, 27. We hypothesized that ATM inhibition should block PIDD1-induced activation of Caspase-2 and reduce Mdm2 cleavage and p53 levels. To test whether ATM regulates endogenous Mdm2 cleavage, we treated cell lines with the ATM inhibitor KU55933 in the presence of a sublethal dose of cisplatin. KU55933 treatment attenuated Caspase-2 activity as demonstrated by reduced accumulation of Mdm2 cleavage product, p60 (Figure 2.7E). Furthermore, KU55933 43

treatment led to corresponding reductions in p53 and p21 levels (Figure 2.7E). KU55933

treatment did not affect Mdm2 cleavage in the context of forced overexpression of PIDD,

possibly due to the high levels of PIDD expression (Supplementary Figure 2.S7B).

Combining KU55933 with cisplatin significantly increased cell death compared to

cisplatin alone at multiple concentrations of cisplatin (Figure 2.7F). Together, these

results suggest a model whereby PIDD1 phosphorylation promotes Caspase-2-

PIDDosome-mediated cleavage of Mdm2, leading to enhanced p53 activity, cell cycle

arrest, and drug resistance (Figure 2.7G).

Discussion

Here we show that Caspase-2 acts as a tumor suppressor in lung cancer by

blocking tumor cell proliferation and progression. Loss of Caspase-2 does not hinder the

initial response of lung tumors to chemotherapy. However, Caspase-2-deficient tumors

rapidly re-enter the cell cycle following chemotherapy-induced growth arrest, ultimately

impeding the therapeutic benefit of long-term treatment. Caspase-2 thus acts similar to

p53 in lung cancer, in that loss of p53 accelerates cell proliferation and tumor progression

but is not required for chemo-sensitivity21. Our previous work demonstrated that loss of

Caspase-2 leads to increased levels of its cleavage target, Mdm2, with corresponding

decreases in p53, ultimately establishing a direct mechanistic link between loss of

Caspase-2 and decreased p53 activity 22. Caspase-2-deficient tumors have reduced

expression of p53 target genes involved in cell cycle arrest and DNA damage repair,

consistent with a reduction in p53 activity. We further show that phosphorylation of

PIDD1 is a critical upstream regulator of Mdm2 cleavage, p53 activation, and cell survival following DNA damage. 44

Caspase-2 acts as a tumor suppressor in lymphoma and mammary tumor models

driven by expression of Myc and Neu oncogenes, respectively10, 23. Our studies show that

loss of Caspase-2 also enhances tumorigenesis in the context of oncogenic Kras in the

lung. Caspase-2 mutations have been found in lung cancer and decreased Caspase-2

expression is found in multiple cancer types13, 26. Consistent with the model that Caspase-

2 acts as a p53-dependent tumor suppressor, we observe that Caspase-2 levels are

significantly reduced in p53-wildtype human adenocarcinomas compared to tumors with

p53 mutations. An alternative explanation for this association is that p53 can repress

Caspase-2 expression2. Despite the mechanism by which Caspase-2 expression is reduced in p53-wildtype tumors, this reduction may contribute to reduced p53 activity by abolishing the Caspase-2-Mdm2-p53 positive feedback loop22.

Together, this suggests that Caspase-2 acts as a general tumor suppressor,

reminiscent of p53, which cooperates with numerous oncogenes to prevent tumorigenesis

in multiple tissues. Our data suggest that the primary mechanism of tumorigenesis upon

Caspase-2 loss in vivo is increased cellular proliferation, consistent with in vitro studies

where knockdown of Caspase-2 accelerates cell growth10, 24. We observed significant

reductions in expression of basal p53 target genes, including p21, cyclin G1, and Msh2, providing a potential mechanistic link to the loss of growth control and advanced histopathology observed in Caspase-2 null tumors. Loss of expression of cell cycle and

DNA damage repair genes may be involved in the genome maintenance functions of

Caspase-2 as well9, 25.

Our previous work implicated the Caspase-2-PIDDosome in drug resistance in

p53-wildtype lung tumors. Based on this, we predicted that loss of Caspase-2 may confer

drug sensitivity. While loss of Caspase-2 did not dramatically enhance chemo-sensitivity 45

compared to Caspase-2-wildtype tumors per se, it did not hamper initial drug sensitivity

or drug sensitivity following repeated doses, despite the fact that Caspase-2-deficient tumors were more advanced than wildtype tumors. Kras-driven p53-deficient lung

tumors also respond significantly to chemotherapy despite being more aggressive than

their p53-wildtype counterparts 21. MicroCT imaging and tumor proliferation studies

indicated that Caspase-2 loss inhibits the long-term response to chemotherapy by accelerating tumor recovery following DNA damage. Similar to our lung tumor model, mammary tumors with wildtype p53 also arrest in response to chemotherapy, limiting their therapeutic benefit compared to p53-null tumors that continuously respond to treatment 12. Furthermore, a p53-like gene expression signature is associated with

platinum-based chemotherapy resistance in human bladder cancer 6.

Our previous studies demonstrated that chemo-resistant Kras-driven lung tumors have high basal phospho-Chk2, suggesting that ATM activity may be increased in

chemo-resistant tumors 21. Recent studies demonstrated that ATM phosphorylates PIDD1

and that this modification is required for formation of the Caspase-2-PIDDosome.

However, the impact of PIDD1 phosphorylation on Caspase-2 cleavage targets was not

investigated 1. Here, we show that phosphorylation of PIDD1 is sufficient to induce

Mdm2 cleavage, and enhance p53 activity, cell cycle arrest, and drug resistance in human

lung cancer cells. Taken together with our previous report of increased Pidd1 expression

in resistant tumors responding to DNA damage, these findings suggest that ATM-induced

PIDD1 activity may contribute to cell cycle arrest and chemo-resistance in vivo. While our data suggest that PIDD1 phosphorylation serves as one mechanism of Caspase-2 activation, PIDD1-independent regulation of Caspase-2 has also been reported19. Future studies will address whether chemo-resistant p53-wildtype lung tumors are sensitized to 46

chemotherapy when combined with ATM inhibitors.

In contrast to previous studies, our results implicate PIDD1 phosphorylation and

Caspase-2 activation in cell cycle arrest as opposed to apoptosis 1. An important

difference between these studies is that PIDD1-induced apoptosis was observed in the

context of p53-deficient cells with Chk1 inhibition 1. We observe that PIDD1

phosphorylation dictates Caspase-2 activity and Mdm2 cleavage in p53-wildtype cells in

the absence of Chk1 inhibition. This suggests that activation of the Caspase-2-

PIDDosome is potentially relevant to a much broader cellular context than previously

described and may serve to protect cells from reparable amounts of DNA damage.

Indeed, our data suggest that the Caspase-2-PIDDosome primarily induces p53-

dependent cell cycle arrest upon DNA damage, and that apoptosis may arise in the

presence of other cellular perturbations or genetic changes. Genomic sequencing studies

have suggested that ATM and p53 mutations are mutually exclusive in lung cancer 8. Our

data suggest that Caspase-2 may mechanistically link ATM and p53 via Caspase-2-

mediated Mdm2 cleavage. Recent studies, however, suggest that ATM and Caspase-2

may also play nonredundant signaling roles 25. Our data suggest that pharmacological modulation of ATM or Caspase-2 in an acute manner may enhance chemotherapy

response in the proper context.

References

1 Ando K, Kernan JL, Liu PH, Sanda T, Logette E, Tschopp J et al. PIDD death- domain phosphorylation by ATM controls prodeath versus prosurvival PIDDosome signaling. Mol Cell 2012; 47: 681-693.

2 Baptiste-Okoh N, Barsotti AM, Prives C. Caspase 2 is both required for p53- mediated apoptosis and downregulated by p53 in a p21-dependent manner. Cell Cycle 2008; 7: 1133-1138. 47

3 Bergeron L, Perez GI, Macdonald G, Shi L, Sun Y, Jurisicova A et al. Defects in regulation of apoptosis in caspase-2-deficient mice. Genes Dev 1998; 12: 1304- 1314.

4 Bertheau P, Turpin E, Rickman DS, Espie M, de Reynies A, Feugeas JP et al. Exquisite sensitivity of TP53 mutant and basal breast cancers to a dose-dense epirubicin-cyclophosphamide regimen. PLoS Med 2007; 4: e90.

5 Bouchier-Hayes L, Green DR. Caspase-2: the orphan caspase. Cell Death Differ 2012; 19: 51-57.

6 Choi W, Porten S, Suengchan K, Willis D. Identification of Distinct Basal and Luminal Subtypes of Muscle-Invasive Bladder Cancer with Different Sensitivities to Frontline Chemotherapy. Cancer Cell 2014; 25: 152-165.

7 Clinical Lung Cancer Genome Project. A genomics-based classification of human lung tumors. Sci Transl Med 2013; 5: 209ra153.

8 Ding L, Getz G, Wheeler DA, Mardis ER, McLellan MD, Cibulskis K et al. Somatic mutations affect key pathways in lung adenocarcinoma. Nature 2008; 455: 1069-1075.

9 Dorstyn L, Puccini J, Wilson CH, Shalini S, Nicola M, Moore S et al. Caspase-2 deficiency promotes aberrant DNA-damage response and genetic instability. Cell Death Differ 2012; 19: 1288-1298.

10 Ho LH, Taylor R, Dorstyn L, Cakouros D, Bouillet P, Kumar S. A tumor suppressor function for caspase-2. Proc Natl Acad Sci U S A 2009; 106: 5336- 5341.

11 Jackson EL, Olive KP, Tuveson DA, Bronson R, Crowley D, Brown M et al. The differential effects of mutant p53 alleles on advanced murine lung cancer. Cancer Res 2005; 65: 10280-10288.

12 Jackson JG, Pant V, Li Q, Chang LL, Quintas-Cardama A, Garza D et al. p53- mediated senescence impairs the apoptotic response to chemotherapy and clinical outcome in breast cancer. Cancer Cell 2012; 21: 793-806.

13 Kan Z, Jaiswal BS, Stinson J, Janakiraman V, Bhatt D, Stern HM et al. Diverse somatic mutation patterns and pathway alterations in human cancers. Nature 2010; 466: 869-873.

14 Kim MS, Kim HS, Jeong EG, Soung YH, Yoo NJ, Lee SH. Somatic mutations of caspase-2 gene in gastric and colorectal cancers. Pathol Res Pract 2011; 207: 640- 644.

15 Krumschnabel G, Sohm B, Bock F, Manzl C, Villunger A. The enigma of caspase-2: the laymen's view. Cell Death Differ 2009; 16: 195-207.

16 Kumar S, Kinoshita M, Noda M, Copeland NG, Jenkins NA. Induction of apoptosis by the mouse Nedd2 gene, which encodes a protein similar to the product of the Caenorhabditis elegans cell death gene ced-3 and the mammalian IL-1 beta-converting enzyme. Genes Dev 1994; 8: 1613-1626.

17 Levine AJ, Oren M. The first 30 years of p53: growing ever more complex. Nat Rev Cancer 2009; 9: 749-758.

18 Lin Y, Ma W, Benchimol S. Pidd, a new death-domain-containing protein, is induced by p53 and promotes apoptosis. Nat Genet 2000; 26: 122-127.

19 Manzl C, Peintner L, Krumschnabel G, Bock F, Labi V, Drach M et al. PIDDosome-independent tumor suppression by Caspase-2. Cell Death Differ 2012; 19: 1722-1732.

20 Meek DW. Tumour suppression by p53: a role for the DNA damage response? Nat Rev Cancer 2009; 9: 714-723.

21 Oliver TG, Mercer KL, Sayles LC, Burke JR, Mendus D, Lovejoy KS et al. Chronic cisplatin treatment promotes enhanced damage repair and tumor progression in a mouse model of lung cancer. Genes Dev 2010; 24: 837-852.

22 Oliver TG, Meylan E, Chang GP, Xue W, Burke JR, Humpton TJ et al. Caspase- 2-mediated cleavage of Mdm2 creates a p53-induced positive feedback loop. Mol Cell 2011; 43: 57-71.

23 Parsons MJ, McCormick L, Janke L, Howard A, Bouchier-Hayes L, Green DR. Genetic deletion of caspase-2 accelerates MMTV/c-neu-driven mammary carcinogenesis in mice. Cell Death Differ 2013; 20: 1174-1182.

24 Puccini J, Dorstyn L, Kumar S. Caspase-2 as a tumour suppressor. Cell Death Differ 2013; 20: 1133-1139.

25 Puccini J, Shalini S, Voss AK, Gatei M, Wilson CH, Hiwase DK et al. Loss of caspase-2 augments lymphomagenesis and enhances genomic instability in Atm- deficient mice. Proc Natl Acad Sci U S A 2013; 110: 19920-19925.

26 Ren K, Lu J, Porollo A, Du C. Tumor-suppressing function of caspase-2 requires catalytic site Cys-320 and site Ser-139 in mice. J Biol Chem 2012; 287: 14792- 14802.

27 Tinel A, Tschopp J. The PIDDosome, a protein complex implicated in activation of caspase-2 in response to genotoxic stress. Science 2004; 304: 843-846.

28 Tuveson DA, Shaw AT, Willis NA, Silver DP, Jackson EL, Chang S et al. Endogenous oncogenic K-ras(G12D) stimulates proliferation and widespread neoplastic and developmental defects. Cancer Cell 2004; 5: 375-387.

29 van Dierendonck JH, Keijzer R, van de Velde CJ, Cornelisse CJ. Nuclear distribution of the Ki-67 antigen during the cell cycle: comparison with growth fraction in human breast cancer cells. Cancer Res 1989; 49: 2999-3006.

Figure 2.1: Caspase-2 levels are significantly reduced in human lung cancers with wildtype p53.

Empirical cumulative density functions (ECDFs) of normalized and standardized expression values for Caspase-2 (A), p21 (B) and Mdm2 (C) are shown for TCGA lung adenocarcinoma tumors with protein-altering p53 mutations (blue) or tumors with wildtype p53 (red). In general, ECDF (y-axis) estimates the fraction of tumors at or below a given value of gene expression. X-axis represents normalized expression of indicated gene. P values indicated in the figures. Left shift in A indicates lower expression of Caspase-2 in p53-wildtype tumors. Right shift in B and C indicate higher expression levels of p21 and Mdm2 in p53-wildtype tumors.

Figure 2.2: Caspase-2 is a tumor suppressor in Kras-driven lung cancer.

(A-F) Representative H&E histology of KC2+/+ (A, D), KC2+/- (B, E), and KC2-/- (C, F) lungs 12 weeks after induction of tumorigenesis at 1x and 40x magnification, respectively. Scale bar in A-C represents 2 mm. Scale bar in D-F represents 50 μm. (G) Average lung tumor size in KC2+/+, KC2+/-, and KC2-/- mice, * P < 0.01, *** P < 0.0001. (H) Average tumor number per lung area in cross-section in KC2+/+, KC2+/-, and KC2-/- mice, ** P < 0.001. (I) Tumor area / total lung area (% tumor burden) in KC2+/+, KC2+/-, and KC2-/- mice, * P < 0.04. Error bars represent standard error of the mean (SEM). Numbers in each bar indicate the number of tumors quantified per genotype.

Figure 2.3: Caspase-2 deficiency enhances tumor proliferation and progression.

(A) Percent low grade (grade 1/2) and advanced grade (grade 3) tumors in KC2+/+, KC2+/-, and KC2-/- mice 12 weeks after tumor induction, ** P < 0.001 compared to KC2+/+. (B) Number of necrotic lung tumors of largest tumors present in KC2 +/+ and KC2+/- (n = 23), KC2-/- (n = 20), and Kp53fl/fl (KP53) mice (n = 7). ** P < 0.0069, *** P < 0.0001 using Fisher’s Exact Test, two-tailed. (C-F) Two representative tumors with necrotic centers observed exclusively in KC2-/- mice. Insets indicated by black squares in C and E. Scale bars represent 2 mm (C), 200 μm (D), 400 μm (E), and 100 μm (F). (G-I) Representative Ki67 staining in lung sections from KC2+/+ (G), KC2+/- (H), or KC2-/- (I) mice 12 weeks after tumor induction, as quantified in J. Scale bars represent 50 μm (G-I). (J) Number of Ki67-positive cells per 40x image field. ** P < 0.0045 using unpaired Student’s t-test compared to KC2+/+. For panels A, B, and J, numbers in each bar indicate the number of tumors quantified per genotype.

Figure 2.4: Caspase-2-deficient tumors respond to chemotherapy treatment and exhibit reduced p53 activity.

(A) Number of BrdU-positive cells per mm2 tumor area in KC2+/+, KC2+/-, and KC2-/- mice treated with PBS control (white bars) or cisplatin (7 mg/kg) (black bars) and analyzed after 72 hours. (B) Expression of p21 in KC2+/+ and KC2-/- mice treated with PBS or cisplatin and analyzed by real time RT-PCR after 72 hours. (C) Number of CC3- positive cells per mm2 tumor area in KC2+/+, KC2+/-, and KC2-/- mice treated with PBS or cisplatin and analyzed after 72 hours. (D-I) Expression of Bax (D), Puma (E), Noxa (F), Cyclin G1 (G), Msh2 (H), and p53 (I) in KC2+/+ (n = 6 mice each) and KC2-/- mice (n = 5 mice each) treated with PBS or cisplatin and analyzed by real-time RT-PCR after 72 hours. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001 using Student’s unpaired t-test. n.s. = not significant. Numbers in each bar indicate the number of tumors quantified per genotype. 54

Figure 2.5: Caspase-2-deficient tumors rapidly resume proliferation following cisplatin treatment.

(A) Representative microCT 3D reconstructions of lungs 12 days before (Baseline), one day before (Before Tx) and five days after two doses of PBS or cisplatin (Cis) (After Tx). Individual tumors on the lung surface are pseudo-colored for visualization. Below is schematic of longitudinal mCT analysis during acute cisplatin treatment. Black arrow indicates AdCre infection. Grey triangles indicate Cis treatments; black arrowheads indicate mCT scans. “Basal” (green) corresponds to period of Baseline to Before Tx. “Cisplatin” (yellow) corresponds to period of Before Tx to After Tx. (B) Log2 tumor volume fold-change of individual tumors in KC2+/+ (white bars) and KC2-/- mice (black bars). Individual tumor volumes were quantified before treatment (basal), before and after doses 1-2 (Tx1), during the recovery period (Recovery), and before and after doses 3-4 (Tx2), * P < 0.01. ** P < 0.005. Below is schematic of long-term cisplatin treatment and longitudinal mCT analysis. Mice were infected with AdCre at Day 0 (black arrow). Lung tumors were quantified by mCT two weeks prior or one day prior to treatment and then five days after treatments 1-2 and 3-4 (black arrowheads). Mice were treated with PBS control or cisplatin at weeks 12, 13, 15, and 16 (grey arrowheads). (C) Number of BrdU-positive cells per tumor area in KC2+/+, KC2+/-, and KC2-/- mice treated with two doses of PBS (white bars) or two doses of cisplatin (black bars) and analyzed after 10 days, equivalent to the recovery period, *** P < 0.0002, ** P < 0.002, * P < 0.012, ns = not significant. Error bars represent SEM. Numbers in each bar graph indicate the number of tumors quantified per genotype.

Figure 2.6: Loss of Caspase-2 impacts long-term chemotherapy treatment.

(A) Schematic of long-term cisplatin treatment and longitudinal microCT analysis. Individual tumors were quantified by mCT one day prior to, and five days after, four doses of PBS or cisplatin (black arrowheads). Mice were infected with AdCre at Day 0 (black arrow) and treated with PBS control or cisplatin (7mg/kg) at weeks 12, 13, 15, and 16 (grey arrowheads). (B) Log2 tumor volume fold change of individual tumors in KC2+/+ and KC2-/- mice treated with PBS control (white bars) or cisplatin (black bars), as indicated in A, ** P < 0.001. (C) Representative images of indicated mice treated with four doses of PBS or cisplatin (Cis). Green outlines lung area; yellow outlines tumor. Scale bar represents 1 mm. (D) Quantification of tumor area/total lung area (% tumor burden) of H&E-stained sections in KC2+/+, KC2 +/-, and KC2-/- mice treated with four doses of PBS or cisplatin, ** P < 0.003, *** P < 0.001, **** P < 0.0001. (E) Average tumor size in cross-section from KC2+/+, KC2 +/- and KC2-/- mice treated with four doses of PBS or cisplatin. * P < 0.04, ** P < 0.001, **** P < 0.0001. Error bars represent SEM. Numbers in each bar graph indicate the number of tumors (B) or mice (D, E) analyzed per genotype.

Figure 2.7: PIDD1 phosphorylation regulates Mdm2 cleavage and p53 activity.

(A) Immunoblot for Flag-tag (PIDD1), pro-Caspase-2 (proC2), Mdm2, p53, p21, and Parp in A549 and SW1573 cells with retroviral overexpression of control (Puro) or indicated PIDD1 constructs. Actin serves as a loading control. Arrowhead indicates cleaved Mdm2, p60; arrow indicates cleaved Parp. (B) Percent of A549 cells in G0/G1 phase of the cell cycle as analyzed by PI staining by flow cytometry in three independent experiments, **** P < .0001. (C) Cisplatin IC50s of A549 cells over-expressing indicated constructs in triplicate in three independent experiments, * P < .05 and **** P < .0001. (D) Immunoblot for Flag-tag (PIDD1), proC2, Mdm2, p53, and p21 in HCT116 p53 +/+ and p53-/- cells with retroviral overexpression of control (Puro) or indicated PIDD1 constructs. Actin serves as a loading control. Arrowhead indicates cleaved Mdm2, p60. (E) Immunoblot analysis of A549 and SW1573 cells treated +/- cisplatin (10 μM and 9 μM, respectively) +/- KU55933 (10 μM) for 72 hours. Whole cell lysates were blotted for Mdm2, p53, and p21. Actin serves as a loading control and pChk2 serves as functional readout of KU55933 activity. Arrowhead indicates cleaved Mdm2, p60. (F) Raw luminescent values of SW1573 cells treated in triplicate with increasing doses of cisplatin +/- KU55933 or vehicle control, analyzed for cell viability by CTG assay after 72 hours. Statistical significance determined using Student’s unpaired t-test. (G) PIDD1 phosphorylation regulates Caspase-2-PIDDosome cleavage of Mdm2 and p53 activity. Under conditions of DNA damage, PIDD1 phosphorylation promotes assembly of the Caspase-2-PIDDosome, leading to Caspase-2 activation. Activated Caspase-2 cleaves and inhibits Mdm2, promoting p53 stability and activity. In p53-wildtype cells, our data suggest that p53 activity promotes cell cycle arrest and resistance to DNA damage.

Supplemental Figure 2.S1: p53 status does not affect expression of Raidd and Pidd1.

(A) Empirical cumulative density functions (ECDFs) of normalized and standardized expression values for Pidd1/Lrdd (A) and RAIDD/Cradd (B) are shown for TCGA lung adenocarcinoma tumors with protein-altering p53 mutations (blue) or tumors with wildtype p53 (red). In general, ECDF (y-axis) estimates the fraction of tumors at or below a given value of gene expression. X-axis represents normalized expression of indicated gene. P values indicated in the figures.

Supplemental Figure 2.S3: Surival and cell death are unaffected by Caspase-2 loss.

(A) Number of BrdU-positive cells per tumor area in KC2+/+, KC2+/-, and KC2-/- mice from Fig 3J. * P < 0.038, Error bars represent SEM. (B) Number of CC3-positive cells per mm2 tumor analyzed from cross-sections of tumor-bearing lungs from KC2+/+ and KC2-/- mice. Numbers in each bar graph indicate the number of tumors quantified per genotype. Error bars represent SEM. (C) Representative images of TUNEL staining in indicated tissues. Negative control = no TdT; Positive control = DNaseI-treated lung tumor tissue. (D) Kaplan-Meier survival curve for KC2+/+, KC2 +/-, and KC2-/- mice following AdCre infections. Time is indicated in days since the AdCre infection (Day 0). Number of mice per cohort is indicated in the legend. Log-rank (Mantel-Dox) test indicates that KC2 +/- mice survive significantly longer than KC2+/+ (p < 0.04) and KC2-/- mice (p < 0.002). Survival differences between KC2+/+ and KC2-/- were not significant.

Supplemental Figure 2.S4: Elevated proliferation in Caspase-2-deficient tumors.

(A) Number of Ki67-positive cells per mm2 tumor area in lung tumor sections from PBS- treated KC2+/+, KC2 +/- and KC2-/- mice in Figure 4A. Error bars represent SEM. Number of tumors analyzed per genotype is indicated in the bar graph. **** p < 0.0001, ** p < 0.001. (B) Table with number of mutations detected in p53 cDNA from microdissected lung tumors from KC2+/+ and KC2-/- mice.

Supplemental Figure 2.S7: Pidd1 expression is not associated with cell death.

(A) Percentage of lung cancer cells in subG1 as analyzed by flow cytometry of A549 or SW1573 cells stained with PI for cell cycle analysis. Cells expressing Puro (control) or indicated PIDD1 constructs were analyzed after 48 hours of expression. (B) Immunoblot for Mdm2, p53, pChk2 (T68), Flag (PIDD1), and Actin in SW1573 cell lysates from cells treated with (+) or without (-) 8 hours of pretreatment with KU55933 (10 μM) followed by 48 hours of treatment with cisplatin (9 μM). Cells were stably expressing vector control (Puro) or indicated PIDD1 constructs. Arrowhead indicates cleaved Mdm2.

CHAPTER 3

A ROLE FOR PIDD1 IN P53-DEPENDENT GENOTOXIC STRESS RESPONSE,

BUT NOT LUNG TUMORIGENESIS

Introduction

PIDD1, also known as leucine rich with a death domain (LRDD), is a scaffold

protein involved in formation of multimeric signaling complexes termed PIDDosomes 19.

Pidd1 was initially discovered while screening for novel proteins containing death domains that were involved in apoptosis 18. It was also identified independently as a gene

induced by activation of tumor suppressor and transcription factor p53 9. PIDD1 regulates

a diverse set of cellular processes such as cell survival, DNA damage repair, cell

proliferation, and apoptosis. Dysregulation of these core biological processes allows

tumor cells to thrive and is a hallmark of cancer.

PIDD1 regulates two distinct pathways that are involved in determining cell fate.

It is sequentially auto-processed by an intein-like mechanism of self-splicing to generate

two fragments: PIDD1-C and PIDD1-CC that are involved in activation of different pathways. PIDD1-C is involved in activation of the NF-kB signaling pathway, thus

promoting cell survival 6, whereas, PIDD1-CC has a mutually exclusive role in assembly

of the Caspase-2-PIDDosome 19. Interestingly, phosphorylation of PIDD1 amino acid

T788 by the DNA damage response kinase ATM is required for formation of the 67

Caspase-2-PIDDosome 2. The molecular switch created by this phosphorylation event ensures Caspase-2 is only activated in response to cellular stress. Thus, PIDD1 dynamically regulates cell fate by inducing diverse biological functions.

Within the last decade, it has become appreciated that p53 and its transcriptional targets (such as MDM2, PIDD2, and CDKN1A) are expressed in oscillatory waves

(pulses). p53 exhibits pulsatile expression and the number of pulses determines the cell response rather than absolute p53 levels 8. The p53-MDM2 negative feedback loop is a critical component of p53 pulses 8. MDM2 also exhibits pulsatile behavior, but with a

phase delay compared to p53 8. MDM2 creates a negative feedback-loop by targeting p53

for proteasomal degradation. Once p53 levels are reduced via degradation, MDM2 levels

begin to decrease due to a lack of p53-dependent transcriptional activation. When the

levels of MDM2 are sufficiently low for p53 protein to re-accumulate, the pulse repeats 3.

Under repairable levels of DNA damage, p53 pulses serve to induce growth arrest and facilitate repair 10. After DNA lesions have been repaired and the damage signal de-

activated, p53 is turned off and expression returns to basal state 10. However, irreparable

DNA damage sustains the damage signal, resulting in continuous p53 pulsing events.

Long-term p53 activation induces senescence and/or apoptosis 16.

In response to DNA damage, p53 also transcriptionally activates PIDD1 and

CDKN1A, encoding p21. First, p21 is a cyclin-dependent kinase inhibitor responsible for

p53-induced growth arrest in response to a variety stimuli 1. Growth arrest prevents the

cell from proceeding through the cell cycle with excess damage. Second, PIDD1

expression leads to assembly of the Caspase-2-PIDDosome, activating Caspase-2 19.

Activated Caspse-2 cleaves MDM2, preventing MDM2 from targeting p53 for

proteasomal degradation 15. MDM2 cleavage prevents the negative feedback loop with 68

p53, leading to p53 stability. In this regard, the Capase-2-PIDDosome creates a positive

feed-forward loop with p53 15. How this feed-forward loop affects the regulation of p53

pulsing expression has not been investigated.

Genetic mouse models involving loss of Pidd1 have been generated. These mice

are born at Mendelian ratios and have no overt phenotype 7, 12. The role of Pidd1 in

cancer has also been investigated. Pidd1 null mice do not develop spontaneous tumors

and are not sensitive to DNA damage-induced tumorigenesis 12. However, Pidd1-

deficiency does delay the onset of Myc-driven lymphoma in a p53-dependent manner 13.

Pidd1 loss reduces activation of the NF-kB pro-survival pathway; however, activation of

Caspase-2 is unaffected 4, 7. Furthermore, no difference in sensitivity to DNA damage-

induced apoptosis was observed in Pidd1-deficient MEFs 4, 13. Taken together, Pidd1-

deficient cells may be more sensitive to oncogenic stress-induced apoptosis, explaining

the delay in tumor initiation in Myc-driven lymphoma.

Our lab identified increased expression of Pidd1 in chemo-resistant lung tumors

with wildtype p53 14. Furthermore, PIDD1 reduces sensitivity to DNA damage agents in

vitro by inducing p53-dependent growth arrest 15. In this study, we investigate the role of

PIDD1 in lung tumorigenesis and chemotherapy response in vivo for the first time in a solid tumor using the oncogenic KrasG12D mouse model. In addition, we investigate

downstream effects of DNA damage on the Caspase-2-PIDDosome signaling pathway.

Our research focuses on the p53-dependent functions of the Caspase-2-PIDDosome, an

overlooked aspect of the pathway.

Results

Pidd1 is not required for KrasG12D-induced tumorigenesis

or chemotherapy response

Pidd1 loss in Myc-driven lymphoma significantly delays tumor onset 13; however,

its role in the development of solid tumors has not been addressed. To determine the role

of Pidd1 in lung tumorigenesis, we used a well-characterized mouse model of lung

adenocarcinoma, Lox-Stop-Lox (LSL)-KrasG12D/+ mice 5. Intranasal administration of

adenovirus carrying Cre recombinase (AdCre) was used to generate tumors in LSL-

KrasG12D/+ mice that were Pidd1-/- (KPidd-/-), Pidd1+/- (KPidd+/-), or Pidd1+/+

(KPidd+/+). Surprisingly, no significant difference in overall survival rate was observed with median survival rates of KPidd+/+ 110 days, KPidd+/- 122 days, and KPidd-/- 104 days (p = 0.24) (Figure 3.1A). To determine if there is difference in tumor development, mice were sacrificed 13 weeks following tumor initiation and tumor burden (tumor area / total lung area) was quantified using hematoxylin and eosin (H&E)-stained lung sections.

Consistent with survival rates, no difference in tumor burden was observed among genotypes (Figure 3.1B). To determine if Pidd1 loss alters the proliferative rate of tumor

cells, we analyzed tumors for the proliferative marker BrdU. As anticipated based on tumor burden, no difference in BrdU staining was observed in KPidd+/+, KPidd+/-, or

KPidd-/- mice (Figure 3.1C).

To determine the role of Pidd1 in long-term chemotherapy response, mice were treated seven weeks after tumor initiation with four doses of cisplatin, standard-of-care chemotherapy, following a previously established therapeutic regimen. First, KPidd+/- and KPidd-/- mice treated with four doses of cisplatin (7 mg/kg) or PBS control were monitored for overall survival. Unfortunately, several mice irrespective of genotype 70

succumbed to chemotherapy-related toxicities. In the surviving mice, no difference was

observed in overall survival, with both KPidd+/- and KPidd-/- mice succumbing to

disease by 185 days (Figure 3.2A). Notably, untreated mice had a maximum life span of

152 days, 18% shorter than treated mice with a maximum life span of 185 days (p = 0.16,

Figure 3.2A). In addition, an independent cohort 13 weeks post tumor initiation was

treated with four doses of cisplatin and the number of tumors were analyzed from H&E

lung sections. Again, no difference in the number of tumors remaining after four doses of

cisplatin was observed (Figure 3.2B). These findings suggest PIDD1 is dispensable for the therapeutic effect of cisplatin in lung cancer.

p21 reduces cisplatin sensitivity independent of p53

PIDD1-induced cell cycle arrest and cisplatin resistance is dependent on wildtype

p53 status and is associated with expression of the cyclin-dependent kinase inhibitor p21

15. Previously, it was demonstrated that p21 is sufficient to confer resistance to cisplatin

20. As a downstream target of p53, p21-induced cisplatin resistance is predicted to be

independent of p53 status. To test this prediction, cell lines inducibly expressing p21

(TRE-p21) were generated in NSCLC cell lines harboring wildtype p53 (A549 and

SW1573) or a p53 point mutation (H23).

TRE-p21 cell lines along with control cell lines TRE-GFP and TRE-PIDD1 were

validated for inducible expression by western blot. As a control, the p53-wildtype cell line SW1573 expressing PIDD1 increased expression of p53 and p21 (Figure 3.3A).

However, no increase in p21 expression was observed in the p53 mutant cell line H23 as

expected (Figure 3.3A). PIDD1 expression induces MDM2 cleavage in H23, but p21

expression is not induced because p53 is incapable of transcriptional activity in these 71

cells. Importantly, all TRE-p21 cell lines expressed p21 in the presence of doxycycline.

In agreement with the known role of p21 in G1 cell arrest, p21 expression in all cells

lines resulted in a flattened cell structure and enlarged cytoplasmic area, a morphological

phenotype associated with senescence (data not shown).

To test p21 dependence on p53 status in response to cisplatin, cell lines A549,

SW1573, and H23 harboring TRE-GFP, TRE-PIDD1, and TRE-p21 were treated with or

without doxycycline in the presence of increasing concentrations of cisplatin. As a

negative control, no difference in cisplatin IC50 was observed in any cell lines expressing

GFP (Figure 3.3B). As a positive control, PIDD1 expression in A549 and SW1573 cells

showed a significant increase in cisplatin IC50, whereas p53 mutant cell line H23 did not

(Figure 3.3B). In all cell lines, p21 expression was sufficient to confer resistance to

cisplatin treatment (Figure 3.3B). These findings demonstrate that p21 is downstream of

p53 in the context of cisplatin response and supports a model in which p21 is a key driver

of PIDD1-induced drug resistance downstream of p53.

PIDD1-induced cisplatin resistance is p21-independent

To test the requirement of p21 in PIDD1-induced cisplatin resistance, cell lines

stably expressing short hairpin RNAs targeted to p21 (shp21) were generated to reduce

p21 expression levels. p21 knockdown was validated in A549 and SW1573 TRE-PIDD1

cells lines treated with or without doxycycline and analyzed by western blot. Elevated

p21 expression was observed in doxycycline-treated TRE-PIDD1 cells co-expressing

shLuc control shRNA, but not in cells expressing shp21 shRNA (Figure 3.4AB).

To test the requirement for p21 in PIDD1-mediated cisplatin resistance, A549 and

SW1573 TRE-PIDD1 cells expressing shLuc control or shp21 were treated with 72

increasing concentrations of cisplatin. As a positive control, PIDD1 expression in both

lines induced a significant increase in cisplatin IC50 in cells expressing shLuc (Figure

3.4CD). Surprisingly, a similarly significant increase in cisplatin IC50 was observed in

cells expressing PIDD1 in combination with shp21 (Figure 3.4CD). Noteworthy, shp21

did not inhibit PIDD1-induced G1 cell cycle arrest (Figure 3.4EF). However, the

mechanism by which PIDD1 induces growth arrest in the absence of p21 is unknown.

This unexpected finding suggests PIDD1-mediated cisplatin resistance is p21-

independent.

Caspase-2-PIDDosome pathway is functional in MCF7 cells

The Caspase-2-PIDDosome pathway is functionally active in a variety of wildtype p53 cell lines established from different tumor types 15. The Lahav lab, our

collaborator, identified pulsatile expression of p53 and MDM2 in response to gamma

irradiation in the breast cancer cell line MCF7. We hypothesized that the Caspase-2-

PIDDosome-mediated positive feedback loop disrupting MDM2-mediated p53 degradation is involved in regulating p53 expression kinetics. To verify the Capase-2-

PIDDosome pathway is intact in MCF7, cells were infected with control MSCV-Puro or

MSCV-PIDD1 retroviruses, selected with a puromycin resistance cassette, and harvested after 48 hours. The Caspase-2-PIDDosome pathway integrity was analyzed by western blot. As expected, PIDD1 expression resulted in accumulation of p60 and p53 protein products (Figure 3.5A). To further characterize the pathway, TRE-GFP, TRE-PIDD1, and TRE-p21 stable cell lines were generated in the MCF7 cell line. Consistent with retroviral over-expression, inducible PIDD1 expression increased MDM2 cleavage and expression of p53 and p21 (Figure 3.5B). Additionally, cell viability in response to 73

cisplatin was tested as previously described. Again, as expected, MCF7 cells inducibly

expressing PIDD1 and p21, but not GFP, were significantly more resistant to cisplatin

treatment (Figure 3.5C). These results confirm that PIDD1-induced growth arrest and

cisplatin resistance pathways are intact in MCF7 cells.

Double-strand DNA breaks induce the Caspase-2-PIDDosome

pathway in MCF7 cells

MDM2 was previously shown to be cleaved in response to treatment with

chemotherapeutic agents cisplatin and doxorubicin 15. Both induce genotoxic stress;

however, their mechanisms of action are different. Cisplatin covalently binds to DNA,

forming bulky adducts that are predominantly repaired via nucleotide excision repair 17.

On the other hand, doxorubicin is a DNA intercalating agent that induces double-stranded

DNA breaks (DSBs) that are resolved by DSB repair pathways 21. To characterize

MDM2 cleavage and activation of the p53-p21 signaling pathway in response to other

chemotherapeutics, MCF7 cells were treated with neocarzinostatin (radiomimetic –

induces DSBs), etoposide (topoisomerase inhibitor – induces DSBs), doxorubicin, and

taxol (mitotic inhibitor that induces mitotic catastrophe). Whole cell extracts harvested

48 hours following drug treatment were analyzed by western blot. Surprisingly, MDM2

cleavage was detected in cells treated with neocarzinostatin, but not etoposide or

doxorubicin as observed in other cell lines (Figure 3.6A). This discrepancy may be a specific difference in MCF7 as it is a breast cancer cell line. Notably, all chemotherapeutics induced p53 and p21 relative to untreated control cells. These results suggest that MCF7 may have a cell-line-specific response to genotoxic stressors in the context of MDM2 cleavage; however, the p53-p21 pathway is activated, suggesting a 74

stress response is intact.

Next, we further investigated DSB-induced activation of the Caspase-2-

PIDDosome and subsequent MDM2 cleavage in MCF7 cells. MCF7 cells were treated with 500ng/ml neocarzinostatin and harvested at time points 24, 48 and 72 hours. MDM2 cleavage was most prominent 24 hours following treatment and elevated levels were maintained through 48 and 72 hours (Figure 3.6B). A reduction in pro-Caspase-2 was

associated with neocarzinostatin treatment, suggesting Caspase-2-PIDDosome assembly

in MCF7 cells occurs in response to radiation (Figure 3.6B). Next, we tested the response

of MCF7 cells to gamma irradiation (gIR). Gamma irradiation is high-energy radiation

that induces DNA damage lesions similar to neocarzinostatin and is a clinically approved

treatment modality for a range of cancer types. MCF7 cells were exposed to 10 gray gIR

and were harvested at time points 0, 20, 28, 44, 52, and 72 hours. Similar to

neocarzinostatin, gIR-induced MDM2 cleavage increased in a time-dependent manner

and was associated with increased p53 and p21 as well as a reduction in pro-Caspase-2

(Figure 3.7A). To determine if radiation-induced MDM2 cleavage was dose-dependent,

MCF7 cells were exposed to increasing levels of gIR (0, 0.5, 1, 2, 4, 8, and 16 gray).

MDM2 cleavage and p53 expression increased in a dose-dependent manner (Figure

3.7B). Taken together, these results suggest the Caspase-2-PIDDosome pathway in

MCF7 cells is intact and is induced in response to irradiation in a time- and dose-

dependent manner. These findings are consistent with the DNA damage response in lung

cancer cell lines15.

PIDD1 expression in response to irradiation

Previously, our collaborator demonstrated p53 expression pulses in response to

gamma irradiation rather than sustain elevated expression 8. This pulsing expression

pattern was observed at the mRNA and protein levels. As a p53 target gene, we

hypothesized that PIDD1 gene expression would also exhibit expression similar to p53.

To investigate PIDD1 expression kinetics in response to irradiation, MCF7 cells were

treated with gIR and harvested hourly for ten hours following 10gy gIR exposure. We observed an increase in PIDD1 expression as early as three hours after treatment, peaking at four hours and returning to basal levels at seven hours (Figure 3.9C). Similarly, Purvis

et al. ’12 observed similar kinetics of MDM2 expression follow 10 gray gIR exposure 16.

At 24 hours posttreatment, PIDD1 expression levels were slightly elevated compared to

controls; however, it is unclear if PIDD1 expression is maintained at a lower level or if the 24 hour time-point was taken during a trough in pulsatile expression. Interestingly,

PIDD1 expression remains elevated 48 hours after NCS treatment relative to controls, suggesting p53 is active at this later time-point (Figure 3.7D). These results suggest

PIDD1 expression is regulated by p53 transcriptional activity at the onset of the p53- mediated DNA damage response. However, Caspase-2-PIDDosome assembly and subsequent MDM2 cleavage occurs at later time points (Figure 3.7A). These findings are consistent with the molecular switch model whereby PIDD1 early leads to activation of the NF-kB pathway and later formation of the Caspase-2-PIDDosome.

PIDD1-mediated DNA damage response and repair

Previous studies identified direct interaction of PIDD1 with proteins involved in

DNA damage response and repair such as DNA-PKcs, DDB1, and PCNA 11. 76

Furthermore, PIDD1 expression is sufficient to confer resistance to genotoxic stress-

induced cell death. Taken together, these findings suggest PIDD1 is involved in DNA

damage repair. To determine if PIDD1 alters DNA damage response, pH2Ax, a marker of

double-strand DNA breaks, positive cells were analyzed by flow cytometry following

exposure to cisplatin or neocarzinostatin. In two independent studies using H460 cells, sub-lethal doses of cisplatin (2 uM or 4 uM) increased the levels of pH2AX positive cells slightly from 1% to 1.8% (Figure 3.8A). Notably, the percentage of PIDD1 expressing cells that are positive for pH2AX is higher following Cisplatin treatment. However, the biological relevance of an absolute change of 1 % in the pH2AX positive cells is unclear.

A sub-lethal dose of neocarzinostatin (500 ug/ml), however, induced a strong increase in

the number of pH2Ax cells peaking one hour posttreatment (Figure 3.8B). Subsequent

reduction in the number of pH2Ax positive cells after four hours suggests resolution of

the DNA damage following treatment (Figure 3.8B). Interestingly, PIDD1 expressing

cells had an average of 17.6% more pH2AX positive cells than control at one and two

hour time points (Figure 3.8B). These data suggests that PIDD1 may be involved in

recognition of DNA damage lesions.

Next, I used fluorescence-based DNA damage repair assays developed in the

Hiom laboratory to test the role of PIDD1 in specific DNA damage repair pathways 22.

To test the role of PIDD1 in microhomology-mediated end joining (MMEJ), I transfected

H460 cells with intact pCMV/I-Sce1/GFP or pCMV/I-Sce1/GFP linearized with the restriction enzyme Bmt1, and used flow cytometry to count the number of GFP positive cells. The GFP sequence in the pCMV/I-Sce1/GFP construct contains a 28 bp insertion flanked by 7 bp microhomology sequence. DNA repair through MMEJ is required to generate a functionally intact GFP. No difference in the number of GFP positive cells 77

were observed in cells expressing PIDD1 or Mdm2-p60 control (Figure 3.8C). Notably,

the number of GFP positive cells is increased upon PIDD1 expression in cells transfected

with both intact and linearized constructs (Figure 3.8CD). The increase in GFP

expression may be a result of increased transfection efficiency in PIDD1 expressing cells

due to growth arrest and/or reduced dilution of the constructs from cell divisions.

However, this issue is accounted for by only comparing the fold change of PIDD

expressing cells with intact and linearized constructs. A similar assay was used to

determine the role of PIDD1 in accurate nonhomologous end joining (NHEJ). Cells were transfected with intact pCMV/myc/cyto/GFP or pCMV/myc/cyto/GFP linearized with blunt cutting restriction enzyme Hinc2. The restriction site is within the GFP coding region and accurate ligation of the two DNA ends is required to generate functionally intact GFP. PIDD1 induced a significant increase in the number of GFP-positive cells transfected with linearized pCMV/myc/cyto/GFP compared to PIDD1-expressing cells transfected with intact plasmid (fold change 4.9 vs 3.2, p <0.05) (Figure 3.8D). No difference was observed in cells expressing p60 control (Figure 3.8D). Taken together, these preliminary findings suggest PIDD1 is involved in recognizing double-strand DNA lesions and may facilitate repair via the NHEJ pathway.

Discussion

The findings presented here failed to detect a role for Pidd1 in oncogenic Kras- induced lung tumorigensis. These findings are in contrast to Manzl et al. ’12 that observed a strong delay in tumor onset in a Myc-driven lymphoma model. It is unclear why PIDD1 has such a strong effect in a mouse model of lymphoma, but not in our lung cancer model. One explanation is that cells of the hematopoietic lineage are more prone 78

to apoptosis and therefore are more sensitive to loss of cell survival proteins. Several

studies have observed reduced NF-kB signaling in Pidd1 -/- mice 2, 4. Lack of this pro-

survival signal may promote death in response to the oncogenic stress associated with

Myc expression in lymphocytes, whereas long-lived but slow-cycling cell types such as

the cells of the lung are less prone to apoptosis.

More interestingly, no difference in survival or tumor response to long-term

cisplatin was observed in KPidd-/- mice. Our lab previously identified PIDD1 as a

resistance mechanism to long-term cisplatin treatment 14. However, resistance to cisplatin

was only noted in tumors treated more than four times. This study does not explore

cisplatin response beyond the fourth dose and thus does not address the role of PIDD1 in

cisplatin resistance as previously identified. Future studies addressing the role of PIDD1

in development of cisplatin resistance in long-term cisplatin-treated mice will test the

requirement of PIDD1 to acquire cisplatin resistance and/or identify alternative resistance

mechanisms.

The current dogma of DNA damage repair proposes that cells undergo cell cycle

arrest in response to DNA damage to allow time for the lesions to be repaired prior to

proceeding through the cell cycle. This prevents cells from entering mitosis with DNA

damage. Chemotherapy is most effective against rapidly dividing cells that proceed

through the cell cycle with DNA damage and undergo apoptosis. However, the direct

connection between growth arrest and drug resistance remains to be explored in detail.

Previous studies from our lab identified growth arrest and reduced sensitivity to genotoxic stress (cisplatin and doxorubicin) as primary functional consequences of

PIDD1 expression in the context of wildtype p53 15. In agreement with other reports, here

we show that p53 target gene p21 is sufficient to induce both growth arrest and resistance 79

to cisplatin. Conversely, p21 knockdown sensitized cells to cisplatin. However, p21

knockdown did not inhibit PIDD1-induced growth arrest or drug resistance in vitro.

Furthermore, KrasG12D tumors from both p53 null and p21 null mice treated with cisplatin

undergo arrest, further validating this phenomenon in vivo 14. Future studies

characterizing p53- and p21-independent mechanisms of growth arrest should be

investigated as downstream effectors of the Caspase-2-PIDDosome pathway.

PIDD1 increased the number of pH2AX positive cells in response to DNA

damage and facilitated increase NHEJ. Interactions with several proteins involved in the

DNA damage response pathway suggest a role for PIDD1 in this capacity. Future studies

to identify which protein interactions are critical for PIDD1-induced DNA damage repair will be informative. Alternatively, PIDD1-induced DNA lesion recognition and repair may be a direct result of growth arrest. Similar to p21 induce growth arrest and resistance, the role of p21 in DNA damage lesion recognition and DNA damage repair assay should be tested. It is important for future studies to dissociate unique functions of

PIDD1 from growth arrest.

p21-independent growth arrest and cisplatin resistance in PIDD1 expressing cells was unexpected. The current study does not explore the role of Caspase-2-mediated growth arrest as a downstream mechanism of PIDD1. PIDD1 induces apoptosis in lieu of growth arrest in p53-deficient cells, suggesting that Caspase-2-induced growth arrest

should also be p53-dependent. Identification of a p21-independent mechanism of PIDD1-

induced growth arrest, potentially mediated by Caspase-2, will be informative and merits

investigation as a therapeutic target to sensitize resistant tumors or promote senescence.

Other studies have identified PIDD1 functional mechanisms that promote cell

survival and DNA damage repair in response to genotoxic stress. The contribution of 80

these functions to cisplatin resistance is unknown. Given that growth arrest induced by

p21 is sufficient to confer resistance to cisplatin, it may be important to extend

experimental observations over a longer time-frame to investigate the long-term effect of

PIDD1 on chemotherapy response. Growth arrest induced by a variety of mechanisms may promote temporary cell viability in response DNA damage agents during the short

72 hour time-period tested in our assays. However, PIDD1-induced growth arrest and

DNA damage repair may enhance long-term cell viability in response to genotoxic insults due to lower rates of mutation accumulation. Importantly, longer treatment duration in

human patients, measured in months, warrants long assay times in vitro or use of models

to better model the disease.

Concluding Remarks

In this study, we demonstrated that Pidd1 is dispensable for Kras-driven lung

tumorigenesis and is irrelevant in long-term response to chemotherapy. However, these

findings are in contrast to reports identifying an oncogenic role for PIDD1 in B-cell

lymphoma. Additionally, we observed PIDD1 expression in vitro enhances the DNA

damage response by promoting DNA lesion recognition and repair via the NHEJ

pathway. Taken together, PIDD1’s role in tumorigenesis and DNA damage response may

be context-specific and further studies in additional mouse models should be pursued to

clarify these discrepancy.

References

1 Abbas T, Dutta A. p21 in cancer: intricate networks and multiple activities. Nat Rev Cancer 2009; 9: 400-414.

2 Ando K, Kernan JL, Liu PH, Sanda T, Logette E, Tschopp J et al. PIDD death- domain phosphorylation by ATM controls prodeath versus prosurvival PIDDosome signaling. Mol Cell 2012; 47: 681-693.

3 Batchelor E, Mock CS, Bhan I, Loewer A, Lahav G. Recurrent initiation: a mechanism for triggering p53 pulses in response to DNA damage. Mol Cell 2008; 30: 277-289.

4 Bock FJ, Krumschnabel G, Manzl C, Peintner L, Tanzer MC, Hermann-Kleiter N et al. Loss of PIDD limits NF-kappaB activation and cytokine production but not cell survival or transformation after DNA damage. Cell Death Differ 2013; 20: 546-557.

5 Jackson EL, Willis N, Mercer K, Bronson RT, Crowley D, Montoya R et al. Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev 2001; 15: 3243-3248.

6 Janssens S, Tinel A, Lippens S, Tschopp J. PIDD mediates NF-kappaB activation in response to DNA damage. Cell 2005; 123: 1079-1092.

7 Kim IR, Murakami K, Chen NJ, Saibil SD, Matysiak-Zablocki E, Elford AR et al. DNA damage- and stress-induced apoptosis occurs independently of PIDD. Apoptosis 2009; 14: 1039-1049.

8 Lahav G, Rosenfeld N, Sigal A, Geva-Zatorsky N, Levine AJ, Elowitz MB et al. Dynamics of the p53-Mdm2 feedback loop in individual cells. Nat Genet 2004; 36: 147-150.

9 Lin Y, Ma W, Benchimol S. Pidd, a new death-domain-containing protein, is induced by p53 and promotes apoptosis. Nat Genet 2000; 26: 122-127.

10 Loewer A, Batchelor E, Gaglia G, Lahav G. Basal dynamics of p53 reveal transcriptionally attenuated pulses in cycling cells. Cell 2010; 142: 89-100.

11 Logette E, Schuepbach-Mallepell S, Eckert MJ, Leo XH, Jaccard B, Manzl C et al. PIDD orchestrates translesion DNA synthesis in response to UV irradiation. Cell Death Differ 2011; 18: 1036-1045.

12 Manzl C, Krumschnabel G, Bock F, Sohm B, Labi V, Baumgartner F et al. Caspase-2 activation in the absence of PIDDosome formation. J Cell Biol 2009; 185: 291-303.

13 Manzl C, Peintner L, Krumschnabel G, Bock F, Labi V, Drach M et al. PIDDosome-independent tumor suppression by Caspase-2. Cell Death Differ 2012; 19: 1722-1732.

14 Oliver TG, Mercer KL, Sayles LC, Burke JR, Mendus D, Lovejoy KS et al. Chronic cisplatin treatment promotes enhanced damage repair and tumor progression in a mouse model of lung cancer. Genes Dev 2010; 24: 837-852.

15 Oliver TG, Meylan E, Chang GP, Xue W, Burke JR, Humpton TJ et al. Caspase- 2-mediated cleavage of Mdm2 creates a p53-induced positive feedback loop. Mol Cell 2011; 43: 57-71.

16 Purvis JE, Karhohs KW, Mock C, Batchelor E, Loewer A, Lahav G. p53 dynamics control cell fate. Science 2012; 336: 1440-1444.

17 Siddik ZH. Cisplatin: mode of cytotoxic action and molecular basis of resistance. Oncogene 2003; 22: 7265-7279.

18 Telliez JB, Bean KM, Lin LL. LRDD, a novel leucine rich repeat and death domain containing protein. Biochim Biophys Acta 2000; 1478: 280-288.

19 Tinel A, Tschopp J. The PIDDosome, a protein complex implicated in activation of caspase-2 in response to genotoxic stress. Science 2004; 304: 843-846.

20 Wang Y, Blandino G, Oren M, Givol D. Induced p53 expression in lung cancer cell line promotes cell senescence and differentially modifies the cytotoxicity of anti-cancer drugs. Oncogene 1998; 17: 1923-1930.

21 Yang F, Teves SS, Kemp CJ, Henikoff S. Doxorubicin, DNA torsion, and chromatin dynamics. Biochim Biophys Acta 2014; 1845: 84-89.

22 Yun MH, Hiom K. CtIP-BRCA1 modulates the choice of DNA double-strand- break repair pathway throughout the cell cycle. Nature 2009; 459: 460-463.

Figure 3.1: Normal Kras-driven tumorigenesis in Pidd1-deficient mice.

(A) Kaplan-Meier survival curve of KPidd-/-, KPidd+/-, and KPidd+/+ mice following AdCre infection at time point zero. (B) Tumor area / total lung area (% tumor burden) in KPidd+/+, KPidd+/-, and KPidd-/- mice. (C) Number of BrdU positive cells / total number of cells (% BrdU positive cells) in KPidd+/+, KPidd+/-, and KPidd-/- tumors. Error bars represent standard error of the mean (SEM). Numbers in each bar indicate the number of mice quantified per genotype. (D-F) Representative H&E histology of KPidd+/+ (D), KPidd+/- (E), and KC2-/- (F) lungs 13 weeks after induction of tumorigenesis at 2.5x magnification. Scale bar in D-E represents 1 mm. (G-I) Representative BrdU staining of KPidd+/+ (G), KPidd+/- (H), and KC2-/- (I) lungs 13 weeks after induction of tumorigenesis at 40x magnification. Scale bar in G-I represents 50 um.

Figure 3.2: Kras-driven tumors in Pidd1-deficient mice respond to chemotherapy.

(A) Kaplan-Meier survival curve of KPidd-/-, KPidd+/-, and KPidd+/+ mice following AdCre infection. Mice were treated with four doses (dotted line) of PBS control of 7 mg/kg cisplatin. (B) Number of tumors per H&E stained slide in KPidd+/+, KPidd+/-, and KPidd-/- mice treated with four doses 7 mg/kg cisplatin and sacrificed five days following the fourth treatment. Error bars represent standard error of the mean (SEM). Numbers in each bar indicate the number of tumors quantified per genotype. (C-E) Representative H&E histology of KPidd+/+ (C), KPidd+/- (D), and KC2-/- (E) lungs following four treatments of 7 mg/kg cisplatin beginning 13 weeks after induction of tumorigenesis. Images are at 2.5x magnification. Scale bar in C-E represents 1 mm.

Figure 3.3: p21-induced p53-independent cisplatin resistance.

(A) Immunoblot of lysates from H23 (mutant p53C246G), A549 and SW1573 inducible GFP control (TRE-GFP), Pidd1 (TRE-PIDD), and p21 (TRE-p21) cells treated with PBS control (-) or 500 ng/ml doxycycline (+). Blot is probed with Flag-tag (Pidd1), Mdm2, p53, and p21 antibodies. Actin serves as a loading control. (B) IC50 of cell lines from A, 48 hours after cisplatin. Representative of two independent experiments performed in triplicate. * P < .05, *** P < .001, **** P < .0001. Error bars represent standard error of the mean (SEM). Peter Hale performed SW1573 immunoblot. 86

Figure 3.4: p21-independent PIDD1-induced G1 arrest and cisplatin resistance.

Inducible Pidd1 (TRE-Pidd) A549 and SW1573 cells stability integrated with control Luciferase (shLuc) or p21 (shp21) short hairpin RNAs were used for the following assays. (A, B) Immunoblot for p21 and Actin in A549 (A) and SW1573 (B) treated with PBS control (-) or 0.5uM doxycycline (+). Actin serves as a loading control. (C, D) Cisplatin IC50 of cells treated as in A and B, respectively. *** P < .001, **** P < .0001. Representative of two independent experiments. (E, F) Percentage of cells in G0/G1 as analyzed by flow cytometry, A549 (E) and SW1573 (F) cells were stained with PI for cell cycle analysis. Representative of two independent experiments. All experiments carried out by Rahul Arya.

Figure 3.5: Caspase-2-PIDDosome pathway is intact in MCF7 cell line.

(A) Immunoblot for Mdm2, p53, and Actin in MCF7 cells with retroviral overexpression of MSCV-Puro control (-) or MSCV-PIDD1 (+). Actin serves as a loading control. (B) Immunoblot for Flag-tag (Pidd1), Mdm2, p53, p21, and Actin in inducible GFP (TRE- GRP), Pidd1 (TRE-Pidd), and p21 (TRE-p21) MCF7 cell lines treated with PBS control or 500ng/ml doxycycline (+). Actin serves as a loading control. (C) IC50 of cells in B 48 hours after cisplatin. * P < .05, **, *** P < .001. Representative of two independent experiments performed in triplicate. Error bars represent standard error of the mean (SEM).

Figure 3.6: DNA damage-specific MDM2 cleavage in MCF7 cells.

(A) Immunoblot of lysate from MCF7 cells treated with PBS control, neocarzinostatin (NCS), etoposide (Etop), doxorubicin (Doxo), or Taxol. Blot is probed for Mdm2, p53, and p21. Arrowheads indicate full-length (top) and cleave (bottom) Mdm2. n=1 (B) immunoblot of lysates from MCF7 cells treated with 500 ng/ml neocarzinostatin (NCS) for 24, 48, and 72 hours. n=1

Figure 3.7: Time- and dose-dependent Caspse-2-PIDDosome activation.

(A) Immunoblot of lysates from MCF7 cells harvested at the indicated time points in untreated (-) or following exposure to 10 gray gamma irradiation (+) for Mdm2, p53, p21, pro-Caspase-2, and Actin. (B) Immunoblot of lysates from MCF7 cells exposed to increasing levels of gamma irradiation for MDM2, p53, p21, and ACTIN. (C) Expression of PIDD1 normalized to ACTIN at the indicated time point following exposure to 10 gray gamma irradiation. Two independent experiments are represented. (D) Expression of PIDD1 normalized to ACTIN in MCF7 cell treated with 500 ug/ml neocarzinostatin after 48 hours.

Figure 3.8: Enhanced DNA damage response in PIDD1 expressing cells.

(A) Percentage of pH2Ax positively labeled H460 control cells (red) or inducible PIDD1 expressing cells (black) at the indicated time points following 4 uM cisplatin treatment. Representative of two independent experiments. (B) Percentage of pH2Ax positively labeled H460 control cells (red) or inducible PIDD1 expressing cells (black) at the indicated time points following 500 ug/ml neocarzinostatin treatment. Representative of two independent experiments. (C) MMEJ assay in H460 cells inducible p60 control cells and inducible PIDD1 cells treated with PBS control or 500 ng/ml doxycycline. Cells were transfect with circular or linearized pCMV/I-Sce1/GFP and the percentage of GFP positive cells as quantified by FACS. (D) NHEJ assay in H460 cells inducible p60 control cells and inducible PIDD1 cells treated with PBS control or 500 ng/ml doxycycline. Cells were transfect with circular or linearized pCMV/myc/cyto/GFP and the percentage of GFP positive cells as quantified by FACS. * P < 0.05. Error bars represent standard error of the mean (SEM).

CHAPTER 4

MDM2 CLEAVEAGE AND THE CASPASE-2-PIDDOSOME

Introduction

Context-specific response to cellular stress, especially in chemotherapy, continues

to challenge clinicians and scientists alike. Identification and treatment of patients with a

high response rate while sparing nonresponders negative side effects is the goal.

Currently, less than 30% of non-small cell lung cancer (NSCLC) patients treated with

cisplatin-based combination therapy experience a therapeutic response 21. Almost all

responsive patients will invariably relapse and be unresponsive to further treatment,

leading to a dismal 16% five-year survival rate 4. It remains unclear why the majority of patients fail to respond and what targetable mechanisms are critical in unresponsive or relapsed tumors. While each tumor is unique, investigation of commonly mutated pathways provides insight into the mechanisms regulating the chemotherapy response.

The mutational landscape is a key contextual determinant that influences cellular response to chemotherapy dictating whether a cell will live or die. The transcription

factor and tumor suppressor p53 is heavily involved in the cellular response to a wide variety of cell stressors. Interestingly, p53 is mutated in half of all lung cancers and often associated with poor outcome 9, 26. However, p53 status is an unreliable marker of

therapeutic response due to its highly context-dependent functions. Various forms of 92 cellular stress-induced transcription-dependent and -independent functions of p53 impact a range of cellular responses such as growth arrest, DNA damage repair, or apoptosis.

p53 activity and function is tightly regulated by the E3 ubiquitin ligase MDM2.

MDM2 is characterized by three protein motifs: a p53 binding domain, an acidic domain, and a RING domain. First, the N-terminal p53 binding domain interacts with the transactivation domain (TAD) of p53. Some reports suggest MDM2 can inhibit p53 transcriptional activity 13, 14; however, many reports demonstrate the presence of MDM2 and p53 complexes at the promoter of actively expressed p53 target genes. Second, the centrally located acidic domain also contains a distinct p53 binding motif. This second p53 interaction domain is important for MDM2-mediated p53 degradation 12. In addition, the acidic domain is important for several protein-protein interactions that inhibit the p53-

MDM2 interaction such as alternative reading frame of CDKN2A (p19ARF) or are proteins recruited to p53 target gene promoters (p300/CBP, SUV39H1, EHMT1, etc.) to influence p53 transcriptional activity 1, 2, 32. Third, the C-terminal RING (really interesting new gene domain) contains the E3 ubiquitin ligase activity responsible for ubiquitinating p53 and targeting it for proteasomal degradation 6. The RING domain is also important for protein-protein interaction with MDM4, which promotes p53 poly-ubiquitination as well as protein-mRNA interaction with p53 mRNA to enhance p53 translation following DNA damage 5.

MDM2 and p53 form a negative feedback loop. p53 transcriptionally activates expression of MDM2, and MDM2 protein binds directly to p53, leading to its poly- ubiquitination and proteasomal degradation. This negative feedback loop is disrupted during stress response by multiple mechanisms to prevent p53 degradation and promote p53 activity. Two well-characterized mechanisms that inhibit p53 degradation are (1) 93

ATM phosphorylates MDM2 and p53, inhibiting their physical interaction 11, 24 and (2)

p19ARF binds to MDM2, sequestering MDM2 in the nucleolus 32. While these regulatory

events are reversible, proteolytic cleavage of MDM2 into two fragments, p60 and p30, is

a permanent mechanism utilized to stabilize p53 protein levels. p60 retains the p53

binding domain and the acidic domain, but lacks the RING domain. Without the RING

domain, p60 is E3 ligase-deficient and thus cannot target p53 for degradation.

MDM2 is processed into p60 and p30 by the serine protease Caspase-2 in

nonapoptotic cells 16, 20. Caspase-2 is activated predominantly by the large molecular weight complex called the Caspase-2-PIDDosome 27. A key structural component of the

Caspase-2-PIDDosome is PIDD1. PIDD1 is a p53 transcriptional target that forms a positive feedback loop with p53 via facilitating Caspase-2 activation and consequently enhanced MDM2 cleavage, p53 activity, and PIDD1 transcription 16. Interestingly,

PIDD1 induces growth arrest and reduces sensitivity to a variety of genotoxic agents in

cells with wildtype p5316. Conversely, PIDD1 induces apoptosis in p53-deficient cells

due to canonical initiator caspase functions of Caspase-2 17, 29. The dichotomy of cellular

response induced by PIDD1 based on p53 status suggests p53 and its interaction with

MDM2 is an important determinant of cell fate, growth arrest, or apoptosis.

The role of MDM2 cleavage product p60 in regulation of p53 activity and DNA damage response has not been thoroughly investigated. Nuclear localization and an intact p53 binding domain suggest p60 may have a role in regulation of gene expression. We

hypothesized that the p53-p60 interaction and p60-dependent recruitment of additional

co-factors regulates p53 target gene selection, transcriptional activity, and ultimately cellular outcome.

Results

MDM2-p60 is localized in the chromatin

Previous studies revealed a major fraction of p60 is localized to the nucleus 16. To

determine if nuclear p60 is associated with the chromatin fraction in response to PIDD1-

induced cleavage, chromatin fractionation was performed on two NSCLC cell lines,

H460 and A549. Chromatin fractionation of whole cell extracts was performed and each

fraction was analyzed by western blot for the presence of MDM2 and p53. Histone H3

serves as a control for proper fractionation as it is exclusively localized to the chromatin

(Figure 4.1A). Low levels of MDM2 were detected in control cells as anticipated. To

induce expression of both full-length MDM2 and p60, PIDD1 was expressed via

retroviral infection using previously established methods 15. Full-length MDM2 was

detected at low levels; however, p60 was markedly increased in cytoplasmic and

chromatin fractions (Figure 4.1A-C). Although differences in fraction volumes preclude direct comparison of protein levels between fractions, the abundance of p60 localized in the chromatin fraction supports a role for the cleavage product in transcriptional regulation.

MDM2 functional variant validation

MDM2 cleavage product p60 is associated with chromatin, suggesting it may

influence transcriptional regulation and thereby cellular response through regulation of

p53 target gene selection and transcriptional activity. To test the function of MDM2 in

regulating p53 function, the following MDM2 constructs with 3x N-terminal myc-tags

were generated: A) wildtype MDM2, B) p60 truncated at amino acid residue D367 to

mimic the cleavage product, C) MDM2 D367A containing a mutation in the cleavage 95

motif such that it should not be cleaved by caspases16, and D) MDM2 D367A/C470A containing a mutation in the cleavage motif and in the RING domain inhibiting

ubiquitination function6 (Figure 4.2). Mutation C470A also disrupts RING-RING domain

interactions, preventing MDM2 homodimers and MDM2-MDM4 heterodimers8.

To validate the function of the MDM2 constructs, Human Embryonic Kidney

293T (HEK-293T) cells were co-transfected with MDM2 constructs in combination with

either activated Caspase-2 or Caspase-2 C320A harboring a mutation in the catalytic

domain, abolishing protease activity. Activated Caspase-2 serves as a control to verify

disruption of the MDM2 cleavage motif in D367A mutants. As expected, western blot

analysis of whole cell extracts revealed wildtype MDM2 is completely processed into

p60 in cells co-transfected with active Caspase-2 and cleavage is significantly reduced in

cells co-transfected with Caspase-2 C320A (Figure 4.3). The presence of endogenous

Caspase-2 in HEK-293T cells likely accounts for the low level of MDM2 cleavage observed in cells transfected with Caspase-2 C320A (Figure 4.3). Importantly, neither

MDM2 D367A nor MDM2 D367A/C470A containing the D367A mutation in the

cleavage motif was cleaved by Caspase-2 to generate p60 (Figure 4.3).

Ubiquitin ligase activity of the MDM2 constructs was also analyzed in transfected

HEK-293T cells by blotting for p53 protein. MDM2 ubiquitinates p53, targeting it for

proteasomal degradation. Ubiquitinated p53 is rapidly degraded and does not

accumulated to detectable levels under normal conditions. However, HEK-293T cells

express SV40 Large T antigen, which stabilizes p53 protein and prevents degradation of

poly-ubiquitinated p53 10. The p53 ubiquitin modifications appear as a laddering pattern

on an immunoblot. In cells co-transfected with wildtype MDM2 and Caspase-2 C320A where full-length MDM2 is present, the characteristic p53 banding pattern is observed 96

(Figure 4.3). This banding pattern is not as abundant in cells expressing wildtype MDM2

and active Caspase-2 as no full-length MDM2 was observed (Figure 4.3). As expected,

MDM2 D367A and MDM2 D367A/C470A construct activity was not influenced by co- transfection of Caspase-2 constructs. A strong p53 banding pattern was observed in all cells transfected with MDM2 D367A and no banding pattern was observed in MDM2

D367A/C470A in agreement with the functional status of RING domain for each respective protein (Figure 4.3). Together, these results confirm the functional activity of

MDM2 and/or the disruption of these functions in the various MDM2 mutant constructs.

Exogenous MDM2 does not regulate p53 activity

We hypothesized that p60 stabilizes p53 protein levels and would phenocopy

PIDD1-induced cell cycle arrest and resistance to cisplatin treatment. To test the

functional consequence of exogenous Mdm2 expression, the MDM2 constructs discussed

in Figure 4.2 as well as PIDD1 were cloned into a tetracycline inducible expression

vector (TET-on) (Figure 4.4) and stably integrated into cells using lentiviral infection.

The optimal concentration of doxycycline to induce maximal expression of the

gene of interest in the TET-on system was determined in NCI-H460 TRE-p60 cells by

treating with escalating doxycycline dosages (0, 0.5, 1, 2, 4, and 8 ug/ml). Expression of

myc-tagged p60 was analyzed 48 hours following doxycycline treatment. All

concentrations of doxycycline tested were sufficient to induce expression of p60 with

maximal expression observed at 0.5 ug/ml (Figure 4.5). Increased concentration of

doxycycline beyond 0.5 ug/ml reduced expression in a dose-dependent manner (Figure

4.5). All further experiments were performed using 0.5 ug/ml doxycycline to achieve maximal expression and were analyzed at 48 hours unless otherwise noted. 97

To determine the molecular response to expression of the MDM2 constructs,

whole cell extracts from NCI-H460 and A549 TRE cell lines treated with or without

doxycycline were analyzed by western blot for p53 and p21 expression. As a control,

inducible PIDD1 expression in the TET-on system led to increased p53 and p21 expression, mimicking constitutive retroviral PIDD1 expression (Figure 4.6AB). In both

A549 and NCI-H460 TRE-p60 cell lines, treatment with doxycycline led to accumulation of higher levels of p60 than are observed in TRE-Pidd1 expressing cells (Figure 4.6AB).

Surprisingly, higher levels of p60 did not lead to appreciable stabilization of p53 or p21 levels as anticipated (Figure 4.6AB). Inducible expression of wildtype MDM2, MDM2

D367A, or MDM2 D367A/C470A did not lead to a consistent difference in p53 or p21 protein levels (Figure 6AB). Taken together, these results suggest that exogenous MDM2

is not sufficient to regulate p53 protein levels under basal conditions.

To investigate p53 target genes potentially regulated by MDM2, RNA was

isolated from A549 inducible cell lines expressing wildtype MDM2, p60, MDM2

D367A, MDM2 D367A/C470A, and PIDD1 for downstream analysis using real-time qPCR. The p53 target genes PIDD1, p21, ZMAT3, APAF, XPC, DCR, BAX, and PML were analyzed. As expected for the positive control, PIDD1 induced higher levels of all p53 target genes (Figure 4.7). No differential expression of p53 targets was observed in cells expressing wildtype MDM2, p60, MDM2 D367A, or MDM2 D367A/C470A

(Figure 4.7), consistent with the immunoblot analysis, suggesting that exogenous MDM2 does not regulate p53 levels or activity.

To investigate the biological consequence of exogenous MDM2, growth rate and cell viability in response to the chemotherapeutic agent cisplatin was analyzed in A549 and NCI-H460 TRE inducible cell lines. First, cell line growth rate was analyzed with or 98

without doxycycline treatment for up to five days. All cell lines grown in the absence of

doxycycline grew at an exponential rate as expected (Figure 4.8). In contrast, a stagnant

growth rate was observed in TRE-PIDD1 cells treated with doxycycline due to G1 cell

cycle arrest as previously described 16(Figure 4.8A-D). Expression of wildtype MDM2,

p60, MDM2 D367A, or MDM2 D367A/C470A did not alter the exponential growth rate

of A549 or NCI-H460 cells (Figure 4.8A-D). These findings suggest neither exogenous

MDM2 nor p60 regulates the cell cycle under these conditions.

Second, cell viability in response to increasing cisplatin concentrations was

measured 72 hours after cisplatin treatment using a luciferase-based ATP indicator (Cell-

Titer glo, Promega). The positive control TRE-PIDD1 cells displayed a greater than two- fold increase in cisplatin inhibitory concentration 50 (IC50) following doxycycline treatment, in agreement with previous studies (Oliver et al ’11) (Figure 4.9AB). No difference in cisplatin IC50 was observed in MDM2, p60, MDM2 D367A, or MDM2

D367A/C470A inducible cell lines (Figure 4.9AB). Taken together, these findings suggest exogenous MDM2 does not regulate the cellular response to genotoxic stressors, likely due to lack of induction of p53 levels and activity.

Engagement of the endogenous p53-MDM2 negative feedback loop at 48 hours may mask the effects of acute exogenous p60 expression on p53 activity. The surprising lack of p60 involvement in p53 transcriptional activity and cellular outcome warranted further characterization of the onset of p60 expression in the TET-on system. To capture cells prior to activation of the negative feedback loop, MCF7 TRE-p60 cells were harvested at 0, 2, 4, 8, 16, and 24 hours after induction. Expression of p60 was first detected four hours after doxycycline treatment and expression levels increased through

24 hours (Figure 4.10). Although no detectable difference in p53 levels was observed, 99

increased p21 expression was observed four hours after doxycycline treatment and p21

expression remained elevated through 24 hours to controls (Figure 4.10). This finding suggests that p60 may regulate p53 transcriptional activity independent of protein levels.

However, analyses of exogenous p60 are confounded by activation of the endogenous

MDM2–p53 negative feedback loop. Exogenous p60 is expected to stabilize p53 protein levels and increase activity. Increased p53 activity will activate expression of p53 target gene MDM2. In turn, MDM2 will target p53 for degradation. Thus, exogenous p60 will activate the p53-MDM2 negative feedback loop masking the potential effects of p60

expression.

MDM2 cleavage disrupts MDM2-MDM4 dimerization,

but not the p53-MDM2 interaction

PIDD1-induced growth arrest and drug resistance is p53-dependent and

downstream activation of p21 is predicted to be a critical component 16. PIDD1 in this

context functions predominately through formation of the Caspase-2-PIDDosome,

leading to Caspase-2 activation and subsequent cleavage of MDM2, altering the nature of

the p53-MDM2 interaction 16. To investigate the consequence of MDM2 cleavage on the

p53-MDM2 interaction, NCI-H460 TRE-PIDD1 and TRE-p60 cells were treated with or

without doxycycline and harvested after 48 hours of induction. Western blot analysis was performed on cell lysates subjected to immunoprecipitation (IP) of p53 or MDM2.

Endogenous MDM2 was recovered with p53 IP in both TRE-PIDD1 and TRE-p60 un- induced cell lines (Figure 4.11A). Following treatment with doxycycline, both p60 and full-length MDM2 bound to p53 (Figure 4.11A). Notably, similar levels of full-length

MDM2 and p60 were bound to p53 in doxycycline-treated cells despite a high ratio of 100

p60 protein compared to full-length MDM2 in whole cell extracts observed in previous

experiments (Figure 4.6A and 4.11A). These data demonstrate that p60 retains the ability

to interact with p53, but also suggests p60 may have a lower binding affinity for p53 than

full-length MDM2.

MDM2 and MDM4, an MDM2 homolog, interact through the RING domains and

this interaction is required for efficient MDM2 ubiquitination of p53 31. To test if the

MDM2-MDM4 interaction is abolished as a result of MDM2 cleavage, IP of MDM2 from TRE-PIDD1 cells treated with or without doxycycline was analyzed. MDM2 was bound by MDM4 in untreated TRE-PIDD1 cells; however, this interaction was significantly reduced in cells treated with doxycycline (Figure 4.11B). Since, MDM4 is also a cleavage target of Caspase-2, the loss of MDM2-MDM4 interaction may be a

result of MDM2 and/or MDM4 cleavage. These data confirm PIDD1-mediated disruption of the MDM2-MDM4 interaction and its correlation with p53 stabilization.

p53-independent PIDD1 function impacts cisplatin resistance

Given the dependence of PIDD1 function on wildtype p53 and the importance of

the p53-MDM2 interaction, I hypothesized inhibiting the p53-MDM2 interaction would reduce PIDD1-induced growth arrest and drug resistance in favor of apoptosis in response to cisplatin. To test this, the p53-MDM2 interaction was disrupted pharmacologically using the small molecule nutlin-3, an MDM2 antagonist that binds the p53 binding pocket of MDM2, preventing the p53-MDM2 interaction 30. Disruption of

the interaction prevents p53 degradation, leading to increased expression of p53 and p53

target genes. To verify disruption of the p53-MDM2 interaction, IP of p53 was performed

on A549 TRE-PIDD1 cells treated with nutlin-3. Both full-length MDM2 and p60 were 101 recovered upon p53 IP in TRE-PIDD1 cells treated with doxycycline as expected (Figure

4.12A). Importantly, MDM2 was not recovered with p53 in TRE-PIDD1 cells treated with both doxycycline and nutlin-3, verifying inhibition of the p53-MDM2 interaction

(Figure 4.12A).

To distinguish PIDD1 specific functions from general p53 functions, A549 cells were treated with increasing concentrations of nutlin-3 (0, 0.5, 1, 2, and 5 uM) and analyzed by western blot to determine a nutlin-3 concentration inducing similar p53 and p21 protein levels as observed in PIDD1 expressing cells. A dose-dependent increase in p53 and p21 protein levels was observed in response to increasing nutlin-3 concentration, with 5 uM nutlin-3 inducing similar levels of p53 and p21 compared to A549 TRE-

PIDD1 cells treated with doxycycline (Figure 4.12B).

To investigate whether pharmacologically increasing p53 protein levels using nutlin-3 is sufficient to phenocopy PIDD1 expression, cell viability in response to increasing concentrations of cisplatin was assayed. A reduction in cisplatin sensitivity was observed in cells treated with 4 uM nutlin-3, and sensitivity to cisplatin was significantly reduced in cells treated with 13 uM nutlin-3 (p = 0.071 and p = 0.0015, respectively) (Figure 4.13A). This finding suggests increasing p53 levels alone is sufficient to confer cisplatin resistance in cell lines maintaining a functionally intact p53 pathway. Furthermore, these findings suggest the p53-MDM2 interaction is not necessary to regulate p53 transcriptional activity to promote cisplatin resistance.

To determine if the p53-MDM2 interaction is required for PIDD1-induced cisplatin resistance, H460 TRE-PIDD1 cells were treated with or without doxycycline in addition to 0, 4, or 13uM nutlin-3 and cell viability was measured in response to increasing cisplatin concentration. As observed before, nutlin-3 treatment alone led to an 102

increase in cisplatin IC50 in TRE-PIDD1 prior to PIDD1 induction (Figure 4.13B).

Surprisingly, nutlin-3 and PIDD1 functioned additively to further increase cisplatin IC50 in H460 TRE-PIDD1 cells treated with doxycycline. Irrespective of nutlin-3, PIDD1- induced an approximately two-fold increase in cisplatin IC50 (cisplatin IC50 2.7 uM to

5.5 uM, p = 0.0015; 4.7 uM to 10.5 uM, p = 0.0073; 11.5 uM to 22.2 uM, p=0.0068; respectively) (Figure 13C). Furthermore, nutlin-3 treatment significantly increased cisplatin resistance among PIDD1 expressing cells (cisplatin IC50 5.5 uM, 10.5 uM, and

22.2 uM, respectively) (Figure 13B).These findings create an interesting paradox.

PIDD1-induced cisplatin resistance is p53-dependent, yet, pharmacologically stabilizing p53 using nutlin-3, which also promotes resistance, has an additive effect.

Discussion

Despite decades of research on p53 signaling, regulation of p53 context-specific functions is not well understood. Although half of all cancers harbor mutations in p53, no consistent role for p53 in response to chemotherapy has been identified and p53 is not widely used as prognostic marker to predict chemotherapy response 4. This study

investigates p53 signaling via the negative feedback loop with MDM2 as regulated by the

Caspase-2-PIDDosome in the context of chemotherapy. Growth arrest and chemo-

resistance induced by the Caspase-2-PIDDosome is dependent on p53 and subsequent

activation of downstream target genes. Here we further characterized the role of MDM2

regulation of p53 in the context of drug resistance.

103

Exogenous p60 function

Previous studies identified that p60 is associated with cell viability and that

MDM2 influences p53 transcriptional activity by recruiting transcriptional co-factors and chromatin remodelers to p53 target genes 1, 2, 20. The data presented here did not detect a

measurable difference in response to exogenous expression of MDM2, p60, or other

MDM2 variants in four different cell culture-based analyses (immunoblot, qPCR, growth

rate, and cell viability in response to genotoxic stress) in which PIDD1 demonstrated an

ability to elicit differential response. Importantly, p60 expression combined with cisplatin

treatment, a condition promoting Caspase-2 activation inhibiting the endogenous

MDM2/p53 negative feedback loop, provided no benefit in cell viability. Thus, the most

likely role of p60 is to prevent full-length MDM2 binding to and targeting of p53 for

degradation. Alternatively, MDM2 cleavage may simply be a mechanism to inhibit

MDM2 function and increase p53 levels. In either case, the function of MDM2 cleavage

is redundant with posttranslational modifications and protein interactions that also

stabilize p53.

While exogenous p60 does not significantly stabilize p53, subtle increases in p53 and p21 expression were observed. It remains reasonable that cleavage of endogenous full-length MDM2 regulates p53 activity to a larger extent. This is supported by the increased p21 at early time-points following p60 expression. A major limitation of the current study is the presence of the endogenous p53-MDM2 feedback loop. The requirement of endogenous MDM2 for cell survival in p53-wildtype cells adds complexity to the conclusions drawn for these analyses. Mdm2 null mice are embryonic

lethal in a p53-dependent manner and MDM2 knockdown in cell culture induces cell

death 7. Furthermore, the regulatory ability of basal MDM2 is exemplified in mice 104

engineered with a mutation in the Mdm2 P2 promoter such that basal Mdm2 expression is normal, but unresponsive to DNA damage. These mice have only a slight delay in p53

degradation kinetics following gIR, suggesting basal Mdm2 is sufficient to regulate p53

in response to DNA damage 18. Acute loss of endogenous MDM2 using shRNA induces

cell death within 48 hours (personal observation). Future investigation using knockdown

and rescue approaches to eliminate the endogenous feedback loop will yield insight into

the role of MDM2 and its cleavage product p60.

MDM2 binding affinity

The observation that full-length MDM2 and p60 are recovered in p53 IP

experiments at equal ratios in cells expressing PIDD1, which have high levels of p60

compared to full-length MDM2, suggests that MDM2 cleavage reduces its p53 binding

affinity. The potential increased binding affinity of full-length MDM2 could facilitate

MDM2 driving of the negative feedback loop in the presence of the more stable p60

protein. Furthermore, full-length MDM2 may have a higher binding affinity for p53 due

to formation of MDM2-MDM2 homodimers and MDM2-MDM4 heterodimers. As a

target of Caspases-2, MDM4 is cleaved similarly to MDM2, disrupting the MDM2-

MDM4 interaction that occurs through the RING domains (Rahul Arya and Chunhua

Wu). Recently, the MDM2-MDM4 interaction was shown to be critical for inhibiting p53

activity independent of p53 degradation in vivo 28. Taken together, these findings suggest

disrupting the MDM2-MDM4 interaction can influence p53 regulation at multiple levels.

Which of these modes of regulation are critical to PIDD1-induced drug resistance is an

open question.

Alternatively, there may be different pools of p53 available to full-length MDM2 105

compared to p60. This is supported by the observation that large amounts of p60 are

bound to p53 in p60 expressing cells, yet the relative level of full-length MDM2 bound to p53 is unaffected by p60 expression. Thus, p60 regulation of p53 may be distinct from full-length MDM2, which could explain the lack of p53 activity in the presence of exogenous p60 in MDM2-proficient cells. Future studies addressing the compartmentalization or gene target association of p60 compared to full-length MDM2 warrant attention.

Disrupting the p53-MDM2 interaction

Previous studies have identified several transcriptional regulators and chromatin

remodelers that are recruited to p53 DNA binding sites through interactions with MDM2

22. How these interactions are influenced by MDM2 cleavage and if there is a functional

consequence has not been addressed. Here we show in the context of cisplatin treatment

that disruption of the p53-MDM2 interaction via pharmacological intervention with nutlin-3 resulted in increased resistance to cisplatin. Conversely, nutlin-3 alone induces

apoptosis in a variety of cancers of the hematopoietic linage 23, 25. These studies highlight

the context-specific nature of p53 activation. Taken together, these results demonstrate

that in solid tumor cell lines with wildtype p53, the p53-MDM2 interaction is not required to mediate growth arrest and drug resistance as a opposed to apoptosis.

Therefore, MDM2 cleavage may be a redundant mechanism to promote elevation of p53 protein levels by inhibiting full-length MDM2-mediated p53 degradation.

Nutlin-3 and PIDD1 both independently induce p53 and p21, leading to growth arrest and drug resistance. Unexpectedly, nutlin-3 and PIDD1 function additively in cisplatin resistance assays. PIDD1-induced a two-fold increase in cisplatin resistance 106

independent of nutlin-3 concentration. This finding further provides support for a model

in which PIDD1 has multiple functional roles that contribute to cisplatin resistance.

Using an experimental approach combining PIDD1 expression and nutlin-3 treatment,

critical DNA damage repair and cell survival functions of PIDD1 independent of p53

regulation can be analyzed. Previous studies addressing p53-independent functions of

PIDD1 have utilized p53-deficient cell lines where PIDD1 induces cell death rather than

growth arrest. Such models have severely limited the ability to investigate the pro-

survival mechanisms, on which our lab has focused. Alternatively, nutlin-3 may also induce oncogenic functions of MDM2 as discussed in Chapter 1. These functions may contribute to increased cisplatin resistance independent of p53. Currently, the role of nutlin-3 in the absence of p53 is controversial 3, 19. Future studies to identify nutlin-3

mechanisms of action independent of p53 are warranted.

Concluding Remarks

Targeting the p53-MDM2 pathway as a therapeutic option has garnered much

attention. However, the context-specific nature of p53 function is a challenge to

development of therapeutically beneficial agents. Contrary to reports demonstrating

therapeutic response to p53 activation, here we provide evidence that stimulation of the

p53 pathway actually promotes resistance to chemotherapy. Thus, caution should be

heeded when targeting the complex p53 signaling pathway that is highly context-

dependent.

107

References

1 Chen L, Li Z, Zwolinska AK, Smith MA, Cross B, Koomen J et al. MDM2 recruitment of lysine methyltransferases regulates p53 transcriptional output. EMBO J 2010; 29: 2538-2552.

2 Cross B, Chen L, Cheng Q, Li B, Yuan ZM, Chen J. Inhibition of p53 DNA binding function by the MDM2 protein acidic domain. J Biol Chem 2011; 286: 16018-16029.

3 Du W, Wu J, Walsh EM, Zhang Y, Chen CY, Xiao ZX. Nutlin-3 affects expression and function of retinoblastoma protein: role of retinoblastoma protein in cellular response to nutlin-3. J Biol Chem 2009; 284: 26315-26321.

4 Ettinger DS, Akerley W, Borghaei H, Chang AC, Cheney RT, Chirieac LR et al. Non-small cell lung cancer, version 2.2013. J Natl Compr Canc Netw 2013; 11: 645-653; quiz 653.

5 Gajjar M, Candeias MM, Malbert-Colas L, Mazars A, Fujita J, Olivares-Illana V et al. The p53 mRNA-Mdm2 interaction controls Mdm2 nuclear trafficking and is required for p53 activation following DNA damage. Cancer Cell 2012; 21: 25-35.

6 Honda R, Yasuda H. Activity of MDM2, a ubiquitin ligase, toward p53 or itself is dependent on the RING finger domain of the ligase. Oncogene 2000; 19: 1473- 1476.

7 Jones SN, Roe AE, Donehower LA, Bradley A. Rescue of embryonic lethality in Mdm2-deficient mice by absence of p53. Nature 1995; 378: 206-208.

8 Kawai H, Lopez-Pajares V, Kim MM, Wiederschain D, Yuan ZM. RING domain- mediated interaction is a requirement for MDM2's E3 ligase activity. Cancer Res 2007; 67: 6026-6030.

9 Lai SL, Perng RP, Hwang J. p53 gene status modulates the chemosensitivity of non-small cell lung cancer cells. J Biomed Sci 2000; 7: 64-70.

10 Lane DP, Crawford LV. T antigen is bound to a host protein in SV40-transformed cells. Nature 1979; 278: 261-263.

11 Maya R, Balass M, Kim ST, Shkedy D, Leal JF, Shifman O et al. ATM- dependent phosphorylation of Mdm2 on serine 395: role in p53 activation by DNA damage. Genes Dev 2001; 15: 1067-1077.

12 Meulmeester E, Frenk R, Stad R, de Graaf P, Marine JC, Vousden KH et al. Critical role for a central part of Mdm2 in the ubiquitylation of p53. Mol Cell Biol 2003; 23: 4929-4938.

108

13 Momand J, Zambetti GP, Olson DC, George D, Levine AJ. The mdm-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell 1992; 69: 1237-1245.

14 Oliner JD, Pietenpol JA, Thiagalingam S, Gyuris J, Kinzler KW, Vogelstein B. Oncoprotein MDM2 conceals the activation domain of tumour suppressor p53. Nature 1993; 362: 857-860.

15 Oliver TG, Mercer KL, Sayles LC, Burke JR, Mendus D, Lovejoy KS et al. Chronic cisplatin treatment promotes enhanced damage repair and tumor progression in a mouse model of lung cancer. Genes Dev 2010; 24: 837-852.

16 Oliver TG, Meylan E, Chang GP, Xue W, Burke JR, Humpton TJ et al. Caspase- 2-mediated cleavage of Mdm2 creates a p53-induced positive feedback loop. Mol Cell 2011; 43: 57-71.

17 Olsson M, Vakifahmetoglu H, Abruzzo PM, Hogstrand K, Grandien A, Zhivotovsky B. DISC-mediated activation of caspase-2 in DNA damage-induced apoptosis. Oncogene 2009; 28: 1949-1959.

18 Pant V, Xiong S, Jackson JG, Post SM, Abbas HA, Quintas-Cardama A et al. The p53-Mdm2 feedback loop protects against DNA damage by inhibiting p53 activity but is dispensable for p53 stability, development, and longevity. Genes Dev 2013; 27: 1857-1867.

19 Peirce SK, Findley HW. The MDM2 antagonist nutlin-3 sensitizes p53-null neuroblastoma cells to doxorubicin via E2F1 and TAp73. Int J Oncol 2009; 34: 1395-1402.

20 Pochampally R, Fodera B, Chen L, Shao W, Levine EA, Chen J. A 60 kd MDM2 isoform is produced by caspase cleavage in non-apoptotic tumor cells. Oncogene 1998; 17: 2629-2636.

21 Ramalingam S, Belani C. Systemic chemotherapy for advanced non-small cell lung cancer: recent advances and future directions. Oncologist 2008; 13 Suppl 1: 5-13.

22 Riley MF, Lozano G. The Many Faces of MDM2 Binding Partners. Genes Cancer 2012; 3: 226-239.

23 Saha MN, Jiang H, Chang H. Molecular mechanisms of nutlin-induced apoptosis in multiple myeloma: evidence for p53-transcription-dependent and -independent pathways. Cancer Biol Ther 2010; 10: 567-578.

24 Shieh SY, Ikeda M, Taya Y, Prives C. DNA damage-induced phosphorylation of p53 alleviates inhibition by MDM2. Cell 1997; 91: 325-334.

109

25 Stuhmer T, Chatterjee M, Hildebrandt M, Herrmann P, Gollasch H, Gerecke C et al. Nongenotoxic activation of the p53 pathway as a therapeutic strategy for multiple myeloma. Blood 2005; 106: 3609-3617.

26 Takahashi T, Nau MM, Chiba I, Birrer MJ, Rosenberg RK, Vinocour M et al. p53: a frequent target for genetic abnormalities in lung cancer. Science 1989; 246: 491-494.

27 Tinel A, Tschopp J. The PIDDosome, a protein complex implicated in activation of caspase-2 in response to genotoxic stress. Science 2004; 304: 843-846.

28 Tollini LA, Jin A, Park J, Zhang Y. Regulation of p53 by Mdm2 E3 Ligase Function Is Dispensable in Embryogenesis and Development, but Essential in Response to DNA Damage. Cancer Cell 2014; 26: 235-247.

29 Vakifahmetoglu H, Olsson M, Orrenius S, Zhivotovsky B. Functional connection between p53 and caspase-2 is essential for apoptosis induced by DNA damage. Oncogene 2006; 25: 5683-5692.

30 Vassilev LT, Vu BT, Graves B, Carvajal D, Podlaski F, Filipovic Z et al. In vivo activation of the p53 pathway by small-molecule antagonists of MDM2. Science 2004; 303: 844-848.

31 Wang X, Wang J, Jiang X. MdmX protein is essential for Mdm2 protein-mediated p53 polyubiquitination. J Biol Chem 2011; 286: 23725-23734.

32 Weber JD, Taylor LJ, Roussel MF, Sherr CJ, Bar-Sagi D. Nucleolar Arf sequesters Mdm2 and activates p53. Nat Cell Biol 1999; 1: 20-26.

110

Figure 4.1: MDM2 cleavage product p60 is localized to the chromatin.

(A) Immunoblot for Mdm2, p53, and histone H3 in A549 with retroviral overexpression of MSCV-Puro control (-) or MSCV-PIDD1 (+). Histone H3 serves as a fractionation control. WCE – whole cell extract. (B) Immunoblot for p53, Mdm2, and Flag-tag (PIDD1) in A549 cells with retroviral overexpression of MSCV-Puro control (-) or MSCV-PIDD1 (+). (C) Immunoblot for p53, Mdm2, and Actin in H460 cells with retroviral over-expression of MSCV-Puro control (-) or MSCV-PIDD1 (+). Actin serves as a loading control.

111

Figure 4.2: Schematic of Mdm2 constructs.

(A) Full-length Mdm2 contains a p53 DNA binding domain, acidic domain (AD), zinc finger domain (ZnF), conserved Caspase-2 cleavage motif DVPD (cleavage site), and a really interesting new gene domain (RING). (B) p60 is Mdm2 truncated at Asp367, mimicking the cleaved product p60. (C) D367A is Mdm2 harboring a point mutation in the Caspase-2 cleavage motif, inhibiting Caspase-2 mediated cleavage. (D) D367A/C470A is Mdm2 harboring point mutations in the cleavage motif, and in the RING domain, disrupting ubiquitin ligase activity.

112

Figure 4.3: MDM2 cleavage mutant and ligase mutant validation.

Immunoblot for Myc-tag (Mdm2), Mdm2, p53, HA-tag (Caspase-2), and Actin in HEK- 293T cells transfected with wildtype Mdm2 (WT), Mdm2 point mutant D367A (D367A), or Mdm2 double mutant (D367A C470A) in combination with activated Caspase-2 (A) or catalytically dead Capase-2 (D). Actin serves as a loading control. FL – full-length MDM2. Ub – ubiquitin modification to p53.

113

Figure 4.4: TET-On inducible gene expression system.

A tetracycline response element (TRE) in the promoter regulates expression of the gene of interest. The plasmid also contains a constitutively expressed “reverse” tetracycline- controlled transactivator (rtTA) protein, which requires the antibiotic doxycycline (or tetracycline) to bind the TRE. In the absence of doxycycline, rtTA does not bind the TRE and therefore, the gene of interest is not expressed. Upon addition of doxycycline, rtTA bound by doxycycline subsequently binds the TRE in the promoter, inducing expression of the cloned gene of interest. Use of the TET-on system in stably integrated cells allows timing and expression level of the gene of interest to be controlled.

114

Figure 4.5: TET-On doxycycline dose-responses.

Immunoblot for Myc-tag (Mdm2), Mdm2, and Actin in A549 TRE-p60 cells treated with escalating doxycycline (doxy) dosage.

115

Figure 4.6: Exogenous MDM2 does not alter p53-p21 pathway.

Immunoblot for Flag-tag (Pidd1), Mdm2, Myc-tag (Mdm2), p53, p21, and Actin in A549 (A) or H460 (B) inducible cell lines PBS control (-) or 500 ng/ml doxycycline. Inducible cell lines tested express empty vector (control), Pidd1, Mdm2, p60, Mdm2 D367A, or Mdm2 D367A/C470A. 116

Figure 4.7: p53 target gene expression is unaffected by exogenous MDM2.

Fold change in p53 target genes PIDD1, p21, ZMAT3, APAF1, XPC, DCR, BAX, and PML mRNA levels in A549 inducible cell-lines treated with PBS control vs 500 ng/ml doxycycline. n=1, normalized to ACTIN. Performed by Anandaroop Mukhopadhyay. 117

Figure 4.8: Exogenous MDM2 does not control cell growth.

Stably integrated inducible Mdm2 (A), p60 (B), Mdm2 D367A (C), or Mdm2 D367A/C470A (D) A549 cells treated with PBS control or 500 ng/ml doxycycline (+ Doxy). Cells counted in triplicate every 24 hours, representative of two independent cell lines.

118

Figure 4.9: Cisplatin sensitivity is not affected by exogenous MDM2.

(A) IC50 for inducible vector control, Pidd, p60, Mdm2, Mdm2 D367A, and Mdm2 D367A/C470A doxycycline-treated A549 cell lines compared to PBS control after 48 hours of cisplatin treatment. Average data from three independent experiments performed in triplicate. (B) IC50 for inducible vector control, Pidd, p60, Mdm2, Mdm2 D367A, and Mdm2 D367A/C470A doxycycline-treated A549 cell lines compared to PBS control after 48 hours of cisplatin treatment. Average data from two independent experiments performed in triplicate. * P < .05. Error bars represent SEM.

119

Figure 4.10: p60 expression induces p21 at early time points following induction.

Immunoblot for Mdm2, p53, p21, and Actin in A549 TRE-p60 inducible cells treated with PBS control (-) or 500 ng/ml doxycycline and harvested at the indicated time points. A549 cells with retroviral overexpression of control (MSCV-Puro) or PIDD1 (MSCV- PIDD1) serve as pathway activation controls.

120

Figure 4.11: MDM2 cleavage disrupts MDMX, but not p53 interaction.

(A) Immunoblot of control Nrf2 or p53 immunoprecipitated (IP) lysates from A549 TRE- p60 or TRE-Pidd1 cells treated with PBS control (-) or 500 ng/ml doxycycline (+). Immunoblot is probed for p53 and Mdm2. (B) Immunoblot of control Ercc1 or p53 immunoprecipitated (IP) lysates from H460 TRE-Pidd1 cells treated with PBS control (-) or 500 ng/ml doxycycline (+). Immunoblot is probed (IB) for MdmX, p53, and Mdm2. (B) was performed by Colin Russell.

121

Figure 4.12: Nutlin-3 disrupts p53-MDM2 interaction enhancing p53 activity.

(A) Immunoblot of control IRAK or p53 immunoprecipitated (IP) lysates from A549 TRE-p60 or TRE-Pidd1 cells treated with PBS control (-), 500 ng/ml doxycycline (+), or 500 ng/ml doxycycline and 5 uM nultin-3a (++) . Immunoblot is probed for p53 and Mdm2. (B) Immunoblot of whole cell lysates from A549 TRE-Pidd1 cells treated with PBS control (-) or 500 ng/ml doxycycline (+), in addition to escalating concentrations (0, 0.5 uM, 1 uM, 2 uM, or 5 uM) nultin-3a. Immunoblot is probed for p53, p21, and Actin. These experiments were performed by David McClellan.

122

Figure 4.13: The p53-MDM2 interaction is dispensable for PIDD1-induced cisplatin resistance

(A) IC50 for H460 cells treated with 0, 4 uM, or 13 uM nutlin-3 after 48 hours of cisplatin treatment. Average data from two independent experiments performed in triplicate. * P < .05, ** P < .01, *** P < .001. (B) IC50 for H460 TRE-Pidd1 cells treated with PBS control or 5 uM doxycycline in addition to 0, 4 uM, or 13 uM nutlin-3 after 48 hours cisplatin treatment. n=1, ** P < .01. (C) Alternative graphical display of (B). * P < .05, ** P < .01, *** P < .001. Error bars represent SEM.

CHAPTER 5

CONCLUSION

Summary

In this dissertation, I investigated the function of the Caspase-2-PIDDosome with

regard to tumorigenesis, chemotherapy, DNA damage response, and p53 activity via

MDM2 cleavage.

In Chapter 2, my data demonstrate that Caspase-2 is a tumor suppressor in the

KrasG12D mouse model of lung adenocarcinoma. Loss of Caspase-2 accelerates tumor

growth primarily through increased cell proliferation, which impairs long-term

chemotherapy efficacy. Human datasets revealed low Caspase-2 expression is correlated with wildtype p53 status, providing support for Caspase-2 function via regulation of the p53-MDM2 negative feedback loop. We also were able to identify ATM as a key regulator of Caspase-2-PIDDosome assembly and p53 activity. These results link ATM,

Caspase-2, and p53 in a previously underappreciated DNA damage signaling network.

In Chapter 3, my work demonstrated that Pidd1-deficiency is not a major

contributor to oncogenic KrasG12D-induced lung adenocarcinoma and does not influence

response to acute or long-term chemotherapy treatment. However, in cultured cells,

Pidd1 is induced in response to DNA damage in an oscillatory nature similar to p53.

These results indicate that PIDD1 may promote DNA damage repair via the non- 124

homologous end-joining pathway. Additionally, PIDD1 induces p21-independent G1

growth arrest, causing reduced sensitivity to DNA damage-induced apoptosis through an

undetermined mechanism.

In Chapter 4, my results demonstrated that MDM2 cleavage product p60 is

localized to chromatin, and interacts with the transcription factor p53. No transcriptional

or biological function was observed in p60 expressing cell lines, but the role of p60 was

likely masked by expression of endogenous Mdm2. Additionally, pharmacological disruption of the p53-MDM2 interaction did not affect PIDD1-induced growth arrest or

drug resistance, supporting a role for p60 or at least loss of full-length MDM2 in

promoting p53 activity.

Taken together, these findings support a model in which Caspase-2 functions as a

tumor suppressor by cleaving MDM2 and relieving inhibition of p53, the master tumor

suppressor. Although the PIDDosome is the primary activating platform for Caspase-2,

Caspase-2 can be activated via redundant mechanisms, which make PIDD1 dispensable

for Caspase-2 function. Furthermore, activating p53 using small molecules recapitulates

PIDD1 function, supporting p53 as the central hub critical for Caspase-2-mediated tumor

suppression.

Perspectives and Future Directions

Growth arrest as a therapeutic target

Growth arrest can be a mechanism of resistance to chemotherapy; however, growth arrest can also lead to progression free survival, a favorable outcome in a clinical setting. Inhibitors of several key checkpoint kinases, such as Chk1, Chk2, and ATM, have been developed under the paradigm that inhibition of growth arrest will allow cells 125

to proliferate more freely, thus affecting tumor cells with high proliferation rates and genome instability more so than nondividing normal cells. Presumably, rapidly proliferating tumor cells undergo apoptosis due to further loss of cell-cycle regulation.

This approach of adding fuel to the fire to burn the fire out has proven effective in some

instances5, 17. However, inhibition of checkpoint kinases, which are also tumor

suppressors, may promote malignant transformation of non-tumor cells. Indeed, Chk1-

deficient and ATM-deficient mice are more susceptible to spontaneous tumorigenesis11,

21.

The interconnection between growth control, DNA damage repair, and apoptosis

provides an interesting mechanism by which both sides of the balance can be targeted

therapeutically. Akin to re-activating p533, 18, inducing growth arrest or senescence is also

therapeutically relevant1, 14. However, identification of reliable biomarkers of senescence

and molecular targets inducing irreversible growth arrest has been challenging.

Additionally, the trade-off between tumor suppressive functions of DNA damage

checkpoints and premature aging due to stem cell depletion is well-established 16. By

inducing arrest in aggressive tumors, DNA damage machinery could potentially

recognize elevated DNA damage often associated with high grade tumors and mediate

clearance of these cells. A better understanding of the genetic determinants that influence

response to checkpoint inhibition or activation will be critical to improving patient

outcomes.

Loss of Caspase-2 tumor suppression, as well as other tumor suppressors, does

not preclude response to chemotherapy; however, long-term therapeutic efficacy is

hindered by the lack of growth inhibition following treatment. Activation of growth arrest

as maintenance therapy with checkpoint activators may hinder proliferation following 126

chemotherapy treatment. Additionally, activation of DNA damage checkpoints could

potentially sensitize putative chemo-resistant cancer stem cells or maintain them in a

quiescent state. While extensive time and resources have been poured into development

of checkpoint kinase inhibitors, the other side of the complex balance of cell cycle

regulation in tissue homeostasis warrants equal attention.

Protein variants in the Caspase-2-PIDDosome pathway

and the functional consequence

The current study of the Caspase-2, PIDD1, and the role of the Caspase-2-

PIDDosome in p53 regulation is complicated by expression of different protein isoforms.

In addition to the simplistic Caspase-2-PIDDosome regulatory pathway described in this

dissertation, all of the components (Caspase-2, MDM2, p53, and PIDD1) of the pathway

are expressed from alternative promoters within the same gene, alternatively spliced, and

are subject to extensive posttranslational modification. Currently, the field lacks the

experimental sensitivity to distinguish these complex forms of regulation, contributing to

inconsistent reporting of the roles of these genes in different contexts.

Caspase-2, for instance, has 2 different isoforms: Caspase-2S and Caspase-2L.

These two splicing variants have opposing functions in apoptosis. Caspase-2S represses

apoptosis whereas Caspase-2L promotes it2, 19. Interestingly, Caspase-2S is expressed in response to a variety of genotoxic agents, promoting cell survival6. High levels of

activated Caspase-2 via the Caspase-2-PIDDosome are observed in viable PIDD1 expressing cells, supporting a nonapoptotic Caspase-2 function. Presumably, these cell

are expressing activated Caspase-2S and not Caspase-2L; however, this hypothesis has

not been thoroughly validated7. Additionally, how Caspase-2S and Caspase-2L impact 127

Caspase-2-PIDDosome assembly and proteolytic cleavage of target substrates are

exciting and unexplored questions.

Most studies of Mdm2 function fail to distinguish between full-length MDM2 and

the cleavage product p60. In this dissertation, I have taken the first steps to understand

the biological implications of these two different products. Future studies characterizing

the regulatory role of p60 will be informative. Mouse models harboring a mutation in the

conserved MDM2 cleavage site DVPD could be used to test the role of Mdm2 in vivo. I hypothesize Mdm2 cleavage mutant mice will have impaired response to cellular stress in

a p53-dependent manner.

Not explored in this dissertation is the role of MDM4 cleavage. Similar to

MDM2, MDM4 is also cleaved by Caspase-2 activated via the PIDDosome. MDM4 is

critical for p53 poly-ubiquitination and Mdm4-deficient mice are p53-dependent

embryonic lethal similar to Mdm2 null mice 9, 12, 13, 20. In combination with MDM2

cleavage, MDM4 cleavage in vivo may have a profound effect on p53 stability and

activity in response to various stimuli. Importantly, under conditions that induce Caspase-

2 cleavage of MDM2, MDM4, and any other Caspase-2 targets can potentially contribute

to the growth arrest and drug resistance phenotype. Using the Mdm2 or Mdm4 cleavage

mutant mice, the correlation between Caspase-2-mediated cleavage of the MDM proteins

and p53 activity in response to DNA damage could be tested directly.

Caspase-2-PIDDosome unresolved questions

While our study and previous reports establish a clear link between the Caspase-

2-PIDDsome and regulation of p53 activity, a meticulous investigation of the pathway is

still warranted. It is clear that Caspase-2 is responsible for MDM2 cleavage; however, the 128

requirement of Caspase-2 for PIDD1-induced growth arrest and drug resistance in

wildtype p53 cells remains unknown. Findings presented in this dissertation that p21 is

not required for PIDD1-induced growth arrest suggest that the canonical p53-p21

response pathway may be dispensable for PIDD1 function. Furthermore, both p53- and

p21-deficient mice induce cell cycle arrest in response to DNA damage, suggesting that

additional mechanisms regulate stress response 15. However, PIDD1 does not induce

growth arrest or cisplatin resistance in p53-null cells. It is unclear if or how PIDD1

utilizes these alternative cell cycle regulators in a p53-dependent manner to induce

growth arrest and drug resistance. Future studies investigating these alternative pathways

will provide insight into PIDD1 functions, as well as identify potential therapeutic

targets.

MDM2 can regulate p53 activity through mechanisms other than targeting p53 for

proteasomal degradation. Presumably, many of these regulatory mechanisms remain

intact in the MDM2 cleavage product p60. Although findings in this dissertation failed to

observe a regulatory role for p60, the presence of endogenous MDM2 most likely

hampered our ability to identify its functions. Reducing endogenous MDM2 using RNAi

strategies induces p53-dependent cell-death, a phenotype which can be rescued by

expression of RNAi resistant MDM2 constructs, an approach frequently used in

developmental biology. Using MDM2 constructs similar to those generated in Chapter 4,

the role of MDM2 and its cleavage product p60 in regulating p53 activity can be studied

in finer detail.

The role of PIDD1 function primarily revolves around activation of NF-kB or

Caspase-2 signaling. Activation of Caspase-2 either results in apoptosis in the context of p53-deficient cells or growth arrest in wildtype p53 cells. However, it is currently 129

unknown if PIDD1 functions through Caspase-2 in either of these situations. The role of

Caspase-2 in apoptosis is controversial and its role in growth control via p21 is moot,

given that PIDD1-induced growth arrest is p21-independent. Expression of PIDD1 in

Caspase-2-deficient cells would be extremely informative to the field’s understanding of

Caspase-2-PIDDosome downstream functions. A single study showed Caspase-2

knockdown increases cell proliferation independent of PIDD1 expression, questioning the

role of Caspase-2 in PIDD1-induced growth arrest. However, data on Pidd1 function in

the absence of Caspase-2 are limited. Future studies investigating the role of PIDD1 in

the absence of Caspase-2 will elucidate mechanisms of Pidd and Caspase-2 that are

associated with p53, but may not be directly linked to Caspase-2-mediated MDM2

cleavage.

Pidd1 upregulation identified in mouse lung tumors treated with more than four

doses of cisplatin chemotherapy inspired the research described in this dissertation. The

role of PIDD1 and Caspase-2 in the developing of resistance to chemotherapy in vivo remains to be tested. Given the role of the Caspase-2-PIDDosome in cisplatin-resistant tumors, loss of Pidd1 or Caspase-2 is predicted to prevent or delay the development of chemotherapy resistance. However, several different mechanisms of chemotherapy resistance have been identified4. Therefore, in the absence of Pidd1 or Caspase-2, other

mechanisms of resistance may develop instead. Furthermore, Caspase-2-deficient tumor

cells proliferate soon after chemotherapy treatment, similar to loss of the tumor

suppressors p53 or p218, 10, 15. While this does not confer resistance per se, it does

ultimately lead to reduced overall response to chemotherapy. Rapid growth may reduce

the selective pressure for tumor cells to develop mechanisms of resistance. In this case,

Caspase-2-deficient tumors would not develop cisplatin resistance because of the 130 inherent advantages associated with the lack of a tumor suppressor. Reconstitution of

Caspase-2 tumor suppressor activity in human tumors may be a viable therapy option.

This strategy would be especially efficacious in tumors with wildtype p53 and/or overexpression of MDM2.

References

1 Acosta JC, Gil J. Senescence: a new weapon for cancer therapy. Trends Cell Biol 2012; 22: 211-219.

2 Droin N, Beauchemin M, Solary E, Bertrand R. Identification of a caspase-2 isoform that behaves as an endogenous inhibitor of the caspase cascade. Cancer Res 2000; 60: 7039-7047.

3 Feldser DM, Kostova KK, Winslow MM, Taylor SE, Cashman C, Whittaker CA et al. Stage-specific sensitivity to p53 restoration during lung cancer progression. Nature 2010; 468: 572-575.

4 Galluzzi L, Senovilla L, Vitale I, Michels J, Martins I, Kepp O et al. Molecular mechanisms of cisplatin resistance. Oncogene 2012; 31: 1869-1883.

5 Garrett MD, Collins I. Anticancer therapy with checkpoint inhibitors: what, where and when? Trends Pharmacol Sci 2011; 32: 308-316.

6 Han C, Zhao R, Kroger J, Qu M, Wani AA, Wang QE. Caspase-2 short isoform interacts with membrane-associated cytoskeleton proteins to inhibit apoptosis. PLoS One 2013; 8: e67033.

7 Heikaus S, Pejin I, Gabbert HE, Ramp U, Mahotka C. PIDDosome expression and the role of caspase-2 activation for chemotherapy-induced apoptosis in RCCs. Cell Oncol 2010; 32: 29-42.

8 Jackson JG, Pant V, Li Q, Chang LL, Quintas-Cardama A, Garza D et al. p53- mediated senescence impairs the apoptotic response to chemotherapy and clinical outcome in breast cancer. Cancer Cell 2012; 21: 793-806.

9 Jones SN, Roe AE, Donehower LA, Bradley A. Rescue of embryonic lethality in Mdm2-deficient mice by absence of p53. Nature 1995; 378: 206-208.

10 Komarova EA, Christov K, Faerman AI, Gudkov AV. Different impact of p53 and p21 on the radiation response of mouse tissues. Oncogene 2000; 19: 3791- 3798. 131

11 Lam MH, Liu Q, Elledge SJ, Rosen JM. Chk1 is haploinsufficient for multiple functions critical to tumor suppression. Cancer Cell 2004; 6: 45-59.

12 Migliorini D, Lazzerini Denchi E, Danovi D, Jochemsen A, Capillo M, Gobbi A et al. Mdm4 (Mdmx) regulates p53-induced growth arrest and neuronal cell death during early embryonic mouse development. Mol Cell Biol 2002; 22: 5527-5538.

13 Montes de Oca Luna R, Wagner DS, Lozano G. Rescue of early embryonic lethality in mdm2-deficient mice by deletion of p53. Nature 1995; 378: 203-206.

14 Nardella C, Clohessy JG, Alimonti A, Pandolfi PP. Pro-senescence therapy for cancer treatment. Nat Rev Cancer 2011; 11: 503-511.

16 Sperka T, Wang J, Rudolph KL. DNA damage checkpoints in stem cells, ageing and cancer. Nat Rev Mol Cell Biol 2012; 13: 579-590.

17 Thompson R, Eastman A. The cancer therapeutic potential of Chk1 inhibitors: how mechanistic studies impact on clinical trial design. Br J Clin Pharmacol 2013; 76: 358-369.

18 Ventura A, Kirsch DG, McLaughlin ME, Tuveson DA, Grimm J, Lintault L et al. Restoration of p53 function leads to tumour regression in vivo. Nature 2007; 445: 661-665.

19 Wang L, Miura M, Bergeron L, Zhu H, Yuan J. Ich-1, an Ice/ced-3-related gene, encodes both positive and negative regulators of programmed cell death. Cell 1994; 78: 739-750.

20 Wang X, Wang J, Jiang X. MdmX protein is essential for Mdm2 protein-mediated p53 polyubiquitination. J Biol Chem 2011; 286: 23725-23734.

21 Xu Y, Ashley T, Brainerd EE, Bronson RT, Meyn MS, Baltimore D. Targeted disruption of ATM leads to growth retardation, chromosomal fragmentation during meiosis, immune defects, and thymic lymphoma. Genes Dev 1996; 10: 2411-2422.

APPENDIX

MATERIALS AND METHODS

Antibodies

Flag-Tag (F3165, 1:1000) and Actin (A2066, 1:1000) were obtained from Sigma.

BrdU (B44, 1:200) and Ki67 antibodies (556528, 1:300) were obtained from BD

Pharmingen. Caspase-2 (11B4, 1:500) was obtained from Alexis. Mdm2 N-terminal (IF2,

1:200) and p53 (DO-1, 1:2000) were obtained from Calbiochem. p21 (F-5, 1:100), p53

(FL393, 1:1000), NRF2 (H300), ERCC1 (3H11), IRAK (H-273), and Mdm2 (SMP14,

1:1000) were obtained from Santa Cruz. Cleaved Caspase-3 (9661, 1:500), Myc-tag

(2272, 1:1000), phospho-Chk2 (T68) (2661, 1:1000), Phospho-Histone H2Ax (Ser139), and Parp (9532, 1:1000) were obtained from Cell Signaling Technology. HA-tag

(1:1000) was obtained from Covance. Histone H3 (Millipore, 1:3000) was obtained from

Fisher Scientific.

Cell Culture and Viral Preparation

Cell lines HEK-293T, A549, SW1573, and HCT116 p53 isogenic lines were maintained in DMEM. Cell lines H23, H460, and MCF7 were maintained in RPMI. For generating viral supernatants, 293T cells were transfected using Mirus LT-1 transfection

reagent with gene plasmids as indicated along with Gag/pol and Env plasmids. 133

Supernatants were harvested after 48-72 hours and human cell lines were infected twice

with viral supernatant at 1:1 (media: supernatant) with 1000× polybrene (8 μg/mL).

Cellswere then selected for two days in puromycin or ten days in blasticidin. Tetracycline responsive cell lines were treated with 500 ng/ml doxycycline to induce gene expression.

Cells were treated in etoposide obtained from BioTang, neocarzinostatin and doxycycline were obtained from Sigma, and taxol was a gift from the Rosenblatt lab.

Cell Cycle Analysis

Cells were harvested with trypsin, washed twice in PBS, and 1 x 10^6 cells were fixed in ice-cold 70% ethanol overnight. Cells were washed 2X in PBS/1% BSA, treated with 500 μg/mL RNase for 30 minutes at 37C, then stained with 50 μg/mL propidium iodide overnight at 4C. Cells were analyzed on a BD FACScan flow cytometer and cell cycle analysis was performed using FlowJo software.

Cell Viability Assays

Cells were seeded in triplicate (5000/well) in opaque 96 well plates. The next day, cells were treated with increasing doses of cisplatin (0–200 μM). After 48 hours of treatment, cell viability was measured using Cell Titer Glo (Promega) on a luminometer.

Normalized, transformed dose response curves were generated and analyzed using

GraphPad Prism to determine IC50. KU55933 was obtained from Selleck Chemicals and used at 10 μM by addition 24 hours prior to cisplatin treatment in immunoblot and cell viability assays.

Tetracycline responsive cell lines were treated with PBS control or 500 ng/ml doxycycline to induce gene expression. Nutlin-3 was obtained from Sigma and used at 4 134

uM or 10 uM by addition 24 hours prior to cisplatin treatment in immunoblot and cell

viability assays.

Chromatin Fractionation

Cells were lysed on ice for 8 minutes in Buffer A containing 10 mM HEPES pH

7.9, 10 mM KCL, 1.5 mM MgCl, 340 mM sucrose, 10% glycerol, 1 mM DTT, and 0.1% triton x-100. After centrifugation (4C, 5 minutes, 1,300 g), the supernatant (cytoplasmic,

S1) was removed from the pellet (nuclei, P1). The S1 fraction was centrifuged (4C, 5 minutes, 20,000 g) and stored at -20C. The P1 was washed in Buffer A again. Then, the

P1 fraction lysed on ice for 30 minutes in Buffer B containing 3 mM EDTA, 0.2 mM

EGTA, and 0.1 mM DTT with intermittent agitation. After centrifugation (4C, 5 minutes,

1,700 g), the soluble fraction (nucleoplasm, S2) was removed from the pellet (chromatin,

P2). The P2 fraction was washed in Buffer G again and resuspended in sample buffer containing DTT for analysis by western blot.

DNA Damage Repair Assays

MMEJ and NHEJ assays are previously described in detail in Yun and Hiom 2009

22. For the MMEJ assay, cells were transfected using Mirus LTX Plus transfection

reagent with uncut pCMV/I-Sce1/GFP or Bmt1-digested pCMV/I-Sce1/GFP plasmid.

Cells were collected after 24 hours, GFP fluorescence was analyzed on a BD FACScan

flow cytometer, and analysis was performed using BD CellQuest Pro software. The

NHEJ assay was performed in a similar manner using uncut pCMV/cyto/myc/GFP or

HincII-digested pCMV/cyto/myc/GFP.

135

Growth Assay

Cells were seeded in duplicate (100k/well) in a 6 well cell culture plate. At time points Day 2-5, cells were trypsinized, resuspended in diluted Trypan blue solution, and

counted using a hemocytometer using standard quantification methods. Growth curves

were generated using GraphPad Prism.

Histology and Immunohistochemistry

Mice were sacrificed by carbon dioxide asphyxiation and lungs were inflated with

PBS, fixed overnight in normalized buffered formalin (NBF), and transferred to 70%

ethanol. Lung lobes were separated and embedded in paraffin according to standard

procedures. Grossly cutting the lungs served to increase the lung area examined

microscopically. Lungs were sectioned at 4 μm and stained with H&E for tumor

pathology. Percent tumor burden and all subsequent staining indices per tumor area were

quantified using whole slide imaging. H&E stained slides were digitally scanned using

Aperio XT instrument (Aperio Inc., Vista, CA). Image analysis was performed using

ImageScope software (Aperio) or AxioVision software (Zeiss). For staining with BrdU,

CC3, or Ki67 antibodies, 4% NBF-fixed lungs were paraffin embedded. 4 μm sections

were dewaxed, rehydrated, and subjected to high temperature antigen retrieval, 10

minutes boiling in a pressure cooker in 10 mM citrate buffer, pH 6.0. Slides were blocked

in 3% H2O2 for 15 minutes, blocked in 5% goat serum in PBS/0.1% Tween-20 for 1 hour,

and stained overnight in 5% goat serum in PBS/0.1%Tween-20 with BrdU (1:100), CC3

(1:100 in SignalStain Diluent, Cell Signaling Technology), or Ki67 (1:300) antibodies. A

goat anti-mouse or anti-rabbit HRP-conjugated secondary antibody (Vector Laboratories)

was used at 1:200 dilution in 5% goat serum in PBS-Triton, incubated for 45 minutes at 136 room temperature, followed by DAB staining (Vector Laboratories). For CC3 staining, secondary antibody staining was performed using the MOM kit (Vector Laboratories).

All staining was performed with Sequenza coverplate technology. IHC slides were visualized using Zeiss Axio Scope.A1 microscope and analyzed using AxioVision SE64 software.

Human Lung Adenocarcinoma Bioinformatics

Gene expression profiles and somatic mutation calls for a large collection of lung adenocarcinoma tumors (n = 459 primary tumors from the RNASeqV2 collection) were downloaded from “The Cancer Genome Atlas” (TCGA) data portal (https://tcga- data.nci.nih.gov/tcga/tcgaHome2.jsp). Expression levels of selected genes were analyzed between tumors with protein-altering mutations in the p53 gene (n=242 tumors) and tumors with wildtype p53 (n=217). Briefly, RNA-seq gene expression levels were normalized to Reads Per Kilobase Exon Per Million Mapped reads (RPKM). Normalized

RPKM expression values for the selected gene were standardized into z-scores (mean =

0, s.d. = 1) and the nonparametric Kolmogorov-Smirnov test was used to assess significance between the empirical distribution functions of the p53 mutant and wildtype sample sets.

Immunoprecipitation

Cells were lysed on ice for 10 minutes in a buffer containing 0.1% NP-40, 5 mM

EDTA, 50 mM Tris (pH 8.0), and 150 mM NaCl, followed by three cycles of freezing and thawing. After centrifugation (4C, 10 minutes, 16,000 g), the supernatants were precleared at 4C on a rotating wheel for 45 minutes in the presence of 20 ml sepharose 137

6B (Sigma). After centrifugation (4C, 1 minute, 1900 g), a small fraction of supernatant

was frozen for later analysis (cell extract), and the remaining fraction was incubated at

4C on a rotating wheel, overnight, in the presence of 10 ml sepharose 6B, 10 ml protein

G sepharose (GE Healthcare), and 0.5 mg of desired antibody. The IPs bound to protein

G were washed three times in lysis buffer, prior to boiling in sample buffer containing

DTT for analysis by western blot.

MicroCT Imaging

At indicated time points, mice were scanned for 2 minutes under isoflurane

anesthesia using a small animal Quantum FX microCT (Perkin Elmer) at 45 μm

resolution, 90 kV, with 160 μA current. Images were acquired using Perkin Elmer

Quantum FX software and processed with Analyze 11.0 (AnalyzeDirect).

Mouse Breeding and Drug Treatment

Mice were housed in an environmentally controlled room according to the

Committee of Animal Care. LSL-KrasG12D/+ mice were kindly provided by Tyler Jacks,

Casp2tm1Yuan-deficient mice were purchased from Jackson Labs and PIDD1 null mice lacking exons 1-5 were provided by Pamela Ohashi. Mice were backcrossed over six generations and bred onto a 129svJae background. Mice were infected with 6.47 × 10^7 plaque-forming units (PFU) of AdCre (University of Iowa) by intranasal instillation as described previously (Jackson et al. 2001) and allowed to develop tumors for 8-13 weeks prior to cisplatin treatment. Mice were given freshly prepared cisplatin (7 mg/kg body weight in PBS, Sigma) or PBS by intraperitoneal (i.p.) injection. For BrdU labeling experiments, BrdU (Sigma) was injected i.p. (30 mg/kg) 24 hours prior to sacrifice. 138

p53 mRNA Sequencing

The p53 DNA binding domain was sequenced from cDNA. RNA was isolated by

TRIzol (Invitrogen) from individual tumors as described (Oliver et al., 2010); DNase-1- treated and 1 µg of total RNA was converted to cDNA using iScript cDNA synthesis kit

(Bio-Rad). Two PCR reactions covering p53 exon 4-8 and exons 6-11 were purified using QIAquick gel extraction kit (Qiagen) and sequenced using Sanger sequencing. Base call and mutation detection analysis was performed manually with ApE software. NCBI

Reference Sequence XP_006533220.1 was used as the reference sequence for mutation analysis, and tumors were also compared to normal lung. Primers used to amplify the p53

DNA Binding Domain were: p53 Exon 4F - ATCTGTTGCTGCCCCAGGATGTTG, p53

Exon 8R - AGGCACAAACACGAACCTC, p53 Exon 6F –

AAGACAGGCAGACTTTTCG, and p53 Exon 11R -

AAAAAGGCAGCAGAAGGGACCG. Primers used for sequencing were: p53 Exon 4F

Seq GCCCCTGTCATCTTTTGTC, p53 Exon 7R Seq

GAGTCTTCCAGTGTGATGATG, p53 Exon 7F Seq AACCGCCGACCTATCCTTAC, and p53 Exon 11R Seq CCTGAAGTCATAAGACAGCAAG.

Plasmids

MSCV-Puro and MSCV-PIDD1 were kindly provided by A. Tinel and J.

Tschopp. TRE-PIDD1 and pCMV-Hdm2-D367A were cloned in Oliver et al. ’11 in the

Tyler Jacks lab. pCMV-myc3-HDM2 and pCMV-HDM2 C464A are available from

Addgene. Caspase-2 constructs (Ac152 wildtype and Ac152 C320A) were kindly provided by M. Guha and D. Altieri. MSCV-PIDD1 point mutants (T788A and T788D) were cloned by site-directed mutagenesis using the following primers: 139

Pidd_T788A_For – AGACCGGCTTTCTGGCGCAGAGCAACCTG,

Pidd_T788A_Rev – CAGGTTGCTCTGCGCCAGAAAGCCGGTCT,

Pidd_T788D_For – CCGAGACCGGCTTTCTGGATCAGAGCAACCTGCTGAG,

Pidd_T788D_Rev – CTCAGCAGGTTGCTCTGATCCAGAAAGCCGGTCTCGG. pCMV-Hdm2-D367A was used to clone pCMV-Hdm2-D367A-D by site-directed mutagenesis using the following primers:

Mdm2 D367A F -

GAAGAGGGCTTTGATGTTCCTGCTTGTAAAAAAACTATAGTGAATG,

Mdm2 D367A R -

CATTCACTATAGTTTTTTTACAAGCAGGAACATCAAAGCCCTCTTC. Cloning was performed according to the manufacturer’s instructions (QuikChange Lightening,

Stratagene) and verified by DNA sequencing.

The following primers were used to clone GFP, Mdm2, p60, Mdm2 D367A,

Mdm2 D367A C470A, p60, and p21 into the TET-on vector at the 5’ restriction site

Hpa1 and the 3’ restriction site Pac1:

Mdm2-F CCGGCCGTTAACATGGAGCAAAAGTTAATCTCAGAGG. p60-R GCGCGCTTAATTAACTAATCAGGAACATCAAAGCCCTCTTC

Mdm2-R GCGCGCTTAATTAACTAGGGGAAATAAGTTAGCAC

GFP-F ATTATAGTTAACATGGTGAGCAAGGGCGAGGAGCTGT

GFP-R GACTATTTAATTAACTAACTTGTACAGCTCGTCCATGCCG p21-F AGTCAT GTTAACATGTCAGAACCGGCTGGGGATGTCCGTC p21-R GCTTAATTAATTAGGGCTTCCTCTTGGAGAAGATCAGCC.

140

Real-Time RT-PCR

For gene expression analysis by real-time RT-PCR, RNA was isolated by TRIzol

(Invitrogen) as described (Oliver et al., 2010) and 1 µg of total RNA was converted to cDNA using iScript cDNA synthesis kit (Bio-Rad). Real-time RT-PCR was performed using gene-specific primers and Sybr Green Supermix (Bio-Rad) in a 20 μl reaction in triplicate on a Bio-Rad CFX96 Real-Time PCR machine. Analysis was performed using

Bio-Rad CFX Manager software and expression values were based on 10-fold serial dilutions of standards and normalized to Actin levels. Primers for real-time RT-PCR of mouse cDNA were:

Primer Sequence Size (bp)

Actin B2 Ex5 GGCATAGAGGTCTTTACGGATGTC 139

Actin F2 Ex4 TATTGGCAACGAGCGGTTCC 139

Bax-F1 TGCAGAGGATGATTGCTGAC 173

Bax-R1 GATCAGCTCGGGCACTTTAG 173

CCNG1 F3 CTCAGTTCTTTGGCTTTGACACG 160

CCNG1 B14 TGGGACATTCCTTTCCTCTTCAG 160

Msh2-F1 CAGGTGGAAAACCACGAGTT 226

Msh2-R1 TGTTGTTGCGAAGCACTTTC 226

Noxa (Pmaip1) F1 CCCACTCCTGGGAAAGTACA 139

Noxa (Pmaip1) R1 AATCCCTTCAGCCCTTGATT 139

Cdkn1a-F TCCAGACATTCAGAGCCACA 184

Cdkn1a-R ACGAAGTCAAAGTTCCACCG 184

Puma (bbc3) F GCCCAGCAGCACTTAGAGTC 191 141

Puma (bbc3) R TGTCGATGCTGCTCTTCTTG 191

p53 F TATCCTGCCATCACCTCACTGC 220

p53 R GAAGCCATAGTTGCCCTGGTAAG 220

Primers for real time RT-PCR of human cDNA were:

Primer Sequence Size (bp)

ACTIN-F CCAACCGCGAGAAGATGA 96bp

ACTIN-R CCAGAGGCGTACAGGGATAG 96bp

APAF-F ACCCTGCTGGCAACGGGAGAT 123bp

APAF-R CACCCAGCCTCCATGGGTAGCA 123bp

BAX-F TGCCTCAGGATGCGTCCACCAAG 292bp

BAX-R CGCTCCCGGAGGAAGTCCAATG 292bp

DCR-F ACCGTGTGTCAGTGTGAAAAAGG 226bp

DCR-R AGTGATAGGGAGAGGCAAGCATC 226bp

MDM2-F GGCTGCTTCTGGGGCCTGTG 120bp

MDM2-R CGGTGCTCCTGGCTGCGAAA 120bp

CDKN1A-F TGAGCTGCGCCAGCTGAGGT 148bp

CDKN1A-R TGCCGCATGGGTTCTGACGG 148bp

PIDD1-F TCTGACACGGTGGAGATGTTCG 124bp

PIDD1-R AGGTGCGAGTAGAAGACAAAGCAG 124bp

PML-F CACCCGCAAGACCAACAACATCTT 159bp

PML-R ATCTCTGCGCTGATGTCGCACTT 159bp

XPC-F CGCCAGAGCAGGCGAAGACAAG 226bp

XPC-R ACTCTGGTAAAGCGGGCTGGGA 226bp

ZMAT3-F GGTCCTTACTTCAATCCCCGCTCT 127bp 142

ZMAT3-R GCCGGAATTCCATCTCTTCGCC 127bp

Statistical Analysis

All statistical analyses were performed using Graphpad Prism 5.0. For column statistics to determine p-values, unpaired two-tailed Student's t-tests were performed. For

IC50 analysis, nonlinear fit-log (inhibitor) versus normalized response (variable slope)

was performed. Error bars represent standard error of the mean, unless otherwise noted.

Western Blotting

Protein lysates were prepared as previously described (Oliver et al., 2010). Protein

samples were separated via SDS-PAGE and transferred to a PVDF membrane.

Membranes were blocked for 1 hour, followed by overnight incubation with primary

antibodies at 4°C. Membranes were washed 6 × 5 minutes at room temperature in PBS-T.

HRP-conjugated secondary antibodies (Jackson ImmunoResearch, 1:10,000) were

incubated for 1 hour at room temperature. For detection, membranes were exposed to

WesternBright HRP Quantum substrate (Advansta) and detected on Hyblot CL film

(Denville Scientific Inc.).