Post-Transcriptional Regulation of PML Protein by Distinct Mechanisms by DONGYIN GUAN Submitted in Partial Fulfillment of the Re

Post-transcriptional Regulation of PML protein by Distinct Mechanisms

DONGYIN GUAN

Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Thesis Advisor: Dr. Hung-Ying Kao

Department of Biochemistry CASE WESTERN RESERVE UNIVERSITY

January, 2016

CASE WESTERN RESERVE UNIVERSITY SCHOOL OF GRADUATE STUDIES

We hereby approve the dissertation of

Dongyin Guan _

Candidate for the Doctor of Philosophy degree*

(signed) ______Dr. David Samols ______(Chair of the committee)

______Dr. Hung-Ying Kao______

______Dr. Angela Ting ______

______Dr. Barbara Bedogni______

______Dr. Zhenghe Wang______

(data) JULY, 2015 _

*We also certify that written approval has been obtained for any proprietary material contained therein.

TABLE OF CONTENTS

LIST OF TABLES ...... iii

LIST OF FIGURES ...... iv

ACKNOWLEDGEMENTS ...... vii

LIST OF ABBREVIATIONS ...... ix

Post-transcriptional Regulation of PML protein by Distinct Mechanisms ...... 1

CHAPTER 1: INTRODUCTION ...... 2

ABSTRACT ...... 2

INTRODUCTION ...... 3

PML and PML-nuclear bodies (NBs) ...... 5

The role of nuclear PML and PML-NBs in tumor suppression ...... 8

Janus-faced role of cytoplasmic PML in tumorigenesis ...... 15

Regulation of PML expression and therapeutic opportunities ...... 17

Mechanisms underlying nucleocytoplasmic shuttling of PML ...... 21

CHAPTER 2: The epigenetic regulator UHRF1 promotes ubiquitination-mediated degradation of the tumor suppressor protein promyelocytic leukemia protein (PML)

...... 26

ABSTRACT ...... 26

INTRODUCTION ...... 27

MATERIALS AND METHODS ...... 29

RESULTS ...... 35

DISCUSSION ...... 65

CHAPTER 3: Deacetylation of the tumor suppressor protein PML regulates hydrogen peroxide-induced cell death ...... 70

ABSTRACT ...... 70

INTRODUCTION ...... 71

MATERIALS AND METHODS ...... 74

RESULTS ...... 79

DISCUSSION ...... 107

CHAPTER 4: microRNA-24 Regulates Endothelial Cell Angiogenesis by Targeting

PML Expression ...... 113

ABSTRACT ...... 113

MATERIALS AND METHODS ...... 117

RESULTS ...... 124

DISCUSSION ...... 141

CHAPTER 5: FUTURE DIRECTIONS ...... 144

References ...... 151

LIST OF TABLES

Table 1. Summary of PML expression and post-translational modification regulators in cancer...... 25

Table 2. Summary of Putative miRNA-Binding Sites on Human H19...... 148

III

LIST OF FIGURES

Figure 1. Summary of PML functions in diseases...... 4

Figure 2. The diagram of PML structure...... 6

Figure 3. PML interactome...... 7

Figure 4. PML NBs-mediated tumor suppression pathways...... 9

Figure 5. UHRF1 negatively regulates PML protein accumulation in HUVECs. ... 37

Figure 6. UHRF1 negatively regulates number of PML NBs and PML protein

accumulation in HUVECs...... 38

Figure 7. Ectopic overexpression of UHRF1 decreased PML protein levels in

mammalian cells...... 39

Figure 8. UHRF1 interacts with PML...... 42

Figure 9. UHRF1 negatively regulates PML protein accumulation in PC3 prostate

cancer cells...... 44

Figure 10. Mapping UHRF1 interacting domain in PML...... 48

Figure 11. PML (1-180) or PML (181-350) does not bind GST-FLNB (R11-15)...... 49

Figure 12. Knockdown of UHRF1 prolongs PML protein half-life...... 53

Figure 13. Knockdown efficiency of UHRF1 in HUVECs...... 55

Figure 14. The half-life of PML (∆B box 2) and PML (∆CC) in control and UHRF1

knockdown cells...... 57

Figure 15. The RING domain is essential for UHRF1-mediated PML ubiquitination

and degradation...... 61

Figure 16. Knockdown of UHRF1 inhibits HUVEC migration...... 62

Figure 17. Knockdown of PML alleviates UHRF1 depletion-mediated inhibition of

HUVEC migration, invasion and capillary tube formation...... 64

Figure 18. The effect of UHRF1 on p53 acetylation...... 68

Figure 19. Validation of the specificity of PML antibodies...... 75

Figure 20. H2O2 induces accumulation of PML-NBs and deacetylation of PML in

HeLa cells...... 81

Figure 21. PML deacetylase screening...... 82

Figure 22. SIRT1 and SIRT5 deacetylate and interact with PML...... 85

Figure 23. PML K487 is the major acetylation site and is critical for nuclear localization of PML in HeLa cells...... 87

Figure 24. SIRT1 and SIRT5 deacetylate PML isoforms...... 89

Figure 25. K487 is crucial for nuclear localization of PML...... 91

Figure 26. Accumulation of PML SUMO1 conjugation in the nucleus...... 94

Figure 27. K487 acetylation and K490 SUMOylation represses to each other...... 97

Figure 28. Deacetylation of PML is required for H2O2-induced accumulation of

PML-NBs...... 102

Figure 29. Knockdown SIRT1 or SIRT5 attenuates nuclear PML accumulation in response to H2O2 treatment...... 103

Figure 30. The role of PML in SIRT1-mediated, H2O2-induced cell death...... 106

Figure 31. A, An alignment of a putative sequence motif that are both acetylated and SUMO1 conjugated...... 112

Figure 32. miR-24 and miR-133 target PML1 mRNA, a major PML mRNA isoform in HUVECs...... 127

Figure 33. Overexpression of miR-24 enhances PML1 protein accumulation in high passage HUVECs or when cells are serum-starved...... 130

Figure 34. miR-24 enhances PML1 mRNA translation in starvation condition. .... 134

Figure 35. The role of FXR1 and miR-133 in miR-24-mediated induction of PML1 protein accumulation...... 136

Figure 36. miR-24 regulates the expression of genes involved in angiogenesis in a

PML-dependent manner...... 138

Figure 37. miR-24 inhibits capillary tube formation in a PML-dependent manner..

...... 139

Figure 38. PML is essential for the inhibitory effect of miR-24 on microvessel outgrowth of mouse aortic rings...... 140

Figure 39. Summary of three distinct mechnisms for post-transcriptional targeting

PML protein...... 147

ACKNOWLEDGEMENTS

First of all, I would like to thank my mentor, Dr. Hung-Ying Kao. He is wise, patient and positive advisor, who also provided challenges and shared life experience to me. These encouragements, challenges and guidance inspired me to another level in regarding to facing difficulties both in scientific and real life world. I feel more confident to be an independent scientist after receiving training under his mentorship. Without his support, I would not have the achievements today.

My committee members Dr. David Samols, Dr. Zhenghe Wang, Dr. Barbara Bedogni and

Dr. Angela Ting provide their constructive criticism and support to my research and career development. I am deeply indebted to my research advisor and the committee members for devoting their wisdom and quality time.

The studies of my Ph.D. stage are collaborated with Dr. Zhenghe Wang, Dr. Masaru

Miyagi, Dr. Benlian Wang, Dr. Edward Seto, Dr. Eckhard Jankowsky, Dr. Ganapati H.

Mahabaleshwar and Andrea Putnam. I also appreciate the support from my lab members,

Dr. Yu Liu, Dr. Jun Hee Lim, Dr. Daniel Factor, Dr. Simran Khurana, Dr. Yu-Ting Su,

Dr. Shuang Gou, Xuan Zhao and Nada Alhazmi. I would like to thank all the people in the biochemistry department and school of medicine for their help and assistance during my study here.

In this scientific journal, I would like to thanks my previous mentors and old friends who provided me the essential knowledge and strength, and their continuous support and encouragements in my Ph.D. study. They are Prof. Zonghou Shen, Prof. Zhonglai Wang,

Prof. Weida Huang, Dr. Hao Wang, Dr. Li Tan, Dr. Yingnan Bian, Dr. Siwei Tang, Dr.

VII

Sanjian Yu and Dan Liu. I also want to thank new friends in Cleveland. Dr. Yuanyuan

Chen and Dr. Tao Che gave me a lot of suggestions and encouragements, especially in some tough moments. I had happy time with Junjie Zhao, Liang Zhu, Xuan Ye, Haitao

Liu, Bolan Li and Ningzhou Zeng. Thanks to everyone who has played a part in my research and life. I truly appreciate all of your help although I did not mention your name herein.

Last but definitely not the least, I appreciate my parents, parents in law and other family members for their unconditional love, understanding and support, which let me focus on the research and chase my dream. My wife, Xiaolan Huang, tries her best to form a warm and cozy home for me and shares every good moments or bad. It is hard to image I can achieve these goals without her help and support.

VIII

LIST OF ABBREVIATIONS

ALT alternative lengthening of telomeres

APL acute promyelocytic leukemia

ARE AU-rich element

ATRA all-trans retinoic acid

ATO arsenic trioxide

BMK1 big MAP Kinase 1

CC domain coiled coil domain

CBP CREB-binding protein

CHX cycloheximide

CPS1 carbamoyl phosphate synthetase 1

DMNT1 DNA methyltransferase 1

HA hemagglutinin

HAEC Human Aorta Endothelial Cells

HIF-1α hypoxia-inducible factor-1

H2O2 hydrogen peroxide

HDACs histone deacetylases

HIPK2 homeodomain-interacting protein kinase-2

HMVECs Human Microvascular Endothelial Cells

HUVECs Human Umbilical Vein Endothelial Cells

INFs Interferons

IRG interferon-responsive gene

JAK/STAT Janus kinase/signal transducer and activator of transcription

MAM mitochondria-associated membranes miRNA microRNAs

NAD+ Nicotinamide adenine dinucleotide

NBS1 nijmegen breakage syndrome protein 1

NCOA2/GRIP1 nuclear receptor coactivator 2

NEM N-ethylmaleimide

NES nuclear export signal

NF-κB nuclear factor kappa-light-chain-enhancer of activated B cells

NJEM non-homologous recombination repair

NLS nuclear localization sequence

PARP1 poly (ADP-ribose) polymerase 1

PBS phosphate-buffered saline

Pin1 peptidyl-prolyl cis/trans isomerase 1

PML promyelocytic leukemia protein

PML-NBs promyelocytic leukemia protein-nuclear bodies pp1a protein phosphatase 1α

PP2a protein phosphatase 2α

POD PML oncogenic domains

PTEN Phosphatase and tensin homolog

PTM post-translational modification pRB retinoblastoma protein

RARα retinoic acid receptor alpha

Rb retinoblastoma

RBCC domain RING-finger, two B-boxes and alpha-helical coiled-coil domain

RING domain Really Interesting New Gene

SIM SUMO interacting motif

SIRTs Sirtuins

SOCS1 suppressor of cytokine signalling 1

STAT3 signal transducer and activator of transcription 3

SUMO small ubiquitin-like molecule

TGFβ transforming growth factor beta

TRIM tripartite motif

TSA trichostatin A

UHRF1 ubiquitin-like, containing PHD and RING finger domains 1

Ubc9 ubiquitin-conjugating enzyme 9

UTR untranslated region

XPC xeroderma pigmentosum C

XII

Post-transcriptional Regulation of PML protein by Distinct Mechanisms

Abstract by

Dongyin Guan

The promyelocytic leukemia (PML) protein is a tumor suppressor originally identified in acute promyelocytic leukemia and implicated in tumorigenesis in multiple forms of cancer. PML protein is frequently down-regulated in various cancers, but PML mRNA levels are relatively similar between normal and cancerous tissues (1). These observations indicate that PML protein levels are tightly regulated, in part, through post-transcriptional regulation. We previously demonstrated that the PML protein undergoes ubiquitination

(Ub)-mediated degradation facilitated by an E3 ligase UHRF1 (2), and that SIRT1/SIRT5 promotes deacetylation and SUMO1 conjugation of PML (3) in response to oxidative stress. Here, we found that the small noncoding RNA miR-24 and miR-133 target 3’

UTR of PML1 mRNA, the major PML isoform in primary human endothelial cells (ECs).

In normal culture condition, miR-24 and miR-133 down-regulate PML1 protein expression in primary human ECs. However, miR-24 but not miR-133 up-regulates

PML1 protein expression under starvation condition. Moreover, miR-24 inhibits ECs angiogenesis, while knocking down PML attenuates the inhibitory effect of miR-24 to angiogenesis. We further confirm miR-24 inhibitory effect on angiogenesis ex-vivo. We found that miR-24 inhibits angiogenesis of PML wild-type but not PML-/- mouse aortic

ECs as demonstrated by aorta ring assays. These findings revealed a new mechanism that

miR-24 switches from an inhibitor to an activator of PML1 mRNA translation under

serum starvation condition.

CHAPTER 1: INTRODUCTION

ABSTRACT

The tumor suppressor protein, promyelocytic leukemia protein (PML), was originally

identified in acute promyelocytic leukemia due to a chromosomal translocation between

chromosomes 15 and 17. PML is the core component of subnuclear structures called

PML nuclear bodies (PML-NBs), which are disrupted in acute promyelocytic leukemia

(APL) cells. PML plays important roles in cell cycle regulation, survival and apoptosis,

and inactivation or down-regulation of PML is frequently found in cancer cells. More than 120 proteins have been experimentally identified to physically associate with PML,

and most of them either transiently or constitutively co-localize with PML-NBs. These

interactions are associated with many cellular processes, including cell cycle arrest,

apoptosis, senescence, transcriptional regulation, DNA repair and intermediary

metabolism. Importantly, PML inactivation in cancer cells can occur at the

transcriptional-, translational- or post-translational- levels. However, only a few somatic mutations have been found in cancer cells. A better understanding of its regulation and its role in tumor suppression will provide potential therapeutic opportunities. Here, we discuss the role of PML in multiple tumor suppression pathways and summarize the players and stimuli that control PML protein expression or subcellular distribution.

INTRODUCTION

In the early 1990s, groundbreaking discoveries in PML research attracted the attention of

cancer researchers. The first breakthrough was the mapping of the breakpoint of a

reciprocal chromosomal translocation between chromosomes 15 and 17 involved in acute

promyelocytic leukemia (APL) (4,5). The promyelocytic leukemia gene (PML also known as MYL, RNF71, TRIM19 and PP8675) was first described as a fusion partner of the retinoic acid receptor alpha (RARα), generating the oncogenic protein (PML-RARα), which is present in > 98% of APL cases (6). Twenty-five years of intense study on the

PML protein from many laboratories has led to the conclusion that PML is a multi- faceted protein that plays pivotal roles in cellular events under physiological and pathological conditions (7-9). These include its role in tumor suppression, anti-viral and anti-bacterial responses, inflammatory responses, metabolism, aging, circadian rhythm and unfolded protein responses (Figure 1) (9-12). Understanding the mechanism by which PML participates in these processes will facilitate development of therapeutic strategies for the treatment of PML-related diseases. Here, we review the literature and highlight recent progress with a focus on our current understanding of the role of PML in tumor suppression.

Figure 1. Summary of PML functions in diseases.

PML plays vital roles in the indicated conditions including anti-inflammatory responses, metabolism, stem cell maintenance, circadian rhythms, aging and unfolded protein responses. PML protein exerts its tumor suppressive function by regulating the cell cycle, apoptosis, senescence, migration, angiogenesis, and DNA repair pathways.

PML and PML-nuclear bodies (NBs)

The Pml gene contains nine exons and spans approximately 53 kb in the genome. Due to

alternative splicing of its C-terminal exons, six nuclear and one cytoplasmic isoform have

been experimentally verified. PML I is the longest isoform and contains 882 amino acids,

while the shortest isoform, PML VII, has 435 amino acids (7,13). The N-terminal 418

amino acids are common to all isoforms and harbor several structurally conserved

domains that include a RING (Really Interesting New Gene) finger domain (R), two

cysteine/histidine-rich B-Box domains (B1 and B2) and an α-helical coiled-coil domain

(CC) (Figure 2). Collectively, these domains are referred as the RBCC domain or the

tripartite motif (TRIM) (14,15). The C-terminal SUMO interacting motif (SIM) is present

in several nuclear PML isoforms is required for the formation of PML NBs (16).

PML has been shown to be enriched in nuclear punctate structures that are interspersed between chromatin (17). These structures have been variably named PML nuclear bodies

(PML-NBs), Kremer bodies, ND10 (nuclear domain 10) or POD (PML oncogenic domains) (17). PML-NBs are heterogeneous and dynamic structures. The RBCC domain

is essential for PML-NB formation (16,18). The size of PML-NBs ranges from 0.1 to 1.0

µm and typically there are 5-30 bodies per nucleus, depending on the cell types, phase of

cell cycle, stress and nutritional condition (19). Loss of PML results in a loss of PML-

NBs, indicating that PML protein is an essential component of PML-NBs (8,20,21).

It has been estimated that PML functionally interacts with more than 160 proteins directly or indirectly (22). Based on information in BIOGRID (http://thebiogrid.org/), 120

proteins physically interact with PML, either transiently or constitutively, as

Figure 2. The diagram of PML structure. All PML isoforms share the same N-terminal

418 amino acids, which contain RING (R), B-Box1 (B1), B-Box1 (B2) and coiled coil

(CC) domains. Nuclear PML isoforms share the same N-terminal 552 amino acids, which in addition to RBCC contains a nuclear localization signal (NLS) and a SUMO- interacting motif (SIM) (23) present in PML isoforms I-V. Only PML1 contains a putative nuclear export signal (NES) at its C-terminus.

Figure 3. PML interactome. Based on the data from BIOGRID (http://thebiogrid.org/),

120 proteins transiently or constitutively physically interact with PML by affinity capture

following Western blot experiments. The PML-associating proteins identified by high- throughput methods are not included. The thicker of the line the more publications support the association.

demonstrated by affinity capture experiments followed by Western blotting (Figure 3).

These interactions suggest the possibility of mutual regulation between PML and its

interacting partners (24-26).

The role of nuclear PML and PML-NBs in tumor suppression

Five years after the discovery of the Pml gene, the tumor suppressive activity of PML

was demonstrated in several cancer types including breast, lung, colorectal, prostate and

bladder cancer (1,27-30). Overexpression of PML inhibits cell proliferation and leads to

cell cycle arrest, senescence and apoptosis whereas Pml knockout cells exhibit increased

proliferation and resistance to UV and cytokine-induced apoptosis (31-33). Moreover,

Pml knockout mice demonstrated elevated spontaneous and chemically-induced tumorigenesis (32). These data suggest that PML is a tumor suppressor. PML-NBs are thought to function as nuclear storage sites that accumulate or sequester proteins in order to release these proteins when required (34). Recent studies indicated that PML-NBs

mediate protein-protein interactions and functions as a platform that promotes protein

post-translational modification, for example, SUMOylaiton, acetylation, ubiquitination

and phosphorylation (35).

Several distinct mechanisms underlying PML-mediated tumor suppression activity have

been reported (Figure 4): (1) PML sequesters proteins in PML-NBs to repress their

functions, (2) PML recruits proteins to PML-NBs or mediates protein-protein interaction

to activate their function, (3) PML-NBs serve a post-translational modification hub to

regulate protein activity and function, (4) PML facilitates targeting of transcription

factors and co-regulators to specific region of genome to control gene expression, (5)

PML and PML-NBs are a part of complexes that regulate DNA damage repair and (6)

Figure 4. PML NBs-mediated tumor suppression pathways. PML NBs repress protein function by sequestration, mediate protein-protein interaction, or act as a post- translational modification hub to regulate diverse tumor suppressor pathways.

PML mediates alternative lengthening of telomeres (ALT) to maintain genome integrity.

These mechanisms influence important cellular pathways such as apoptosis, p53 stability,

Akt activity and gene regulation.

Caspase 3-dependent and -independent pathways in apoptosis

The activation of caspase 3 is a key event in apoptosis and is vital for the inhibition of

cancer cell growth (36). PML induces caspase 3 activation and mediates multiple

apoptotic pathways in response to various stimuli, including γ-irradiation, tumor necrosis

factor α (TNFα), Fas, type I and II interferon (INFs), and ceramide (37,38). The lethal effects of γ-irradiation and anti-Fas antibody are attenuated in Pml knockout mice and cells (38,39), indicating that Pml-mediated activation of caspase 3 is essential for apoptosis. However, PML can also recruit BAX and p27KIP1 to PML-NBs and can mediate apoptosis independently of caspase 3 activation (39). In summary, PML mediates apoptosis via both caspase 3-dependent and -independent pathways.

Regulation of p53

The tumor suppressor p53 is an extensively studied gene that is important for many aspects of tumor biology (40). PML is a critical regulator of p53 activity and p53- mediated cellular processes, such as apoptosis, cell cycle arrest, DNA repair and senescence. In response to cellular stress and DNA damage, PML enhances p53 protein stability by sequestering Mdm2 in NBs (41-43). Mdm2 is a major cellular p53 E3 ubiquitin ligase that destabilizes p53. The activated big MAP kinase 1 (BMK1) interacts

with PML and disrupts its association with Mdm2, thereby destabilizing p53 (44).

Furthermore, DNA damage promotes the recruitment of the DNA damage-induced kinase

Chk2 to PML-NBs where it phosphorylates p53 at serine 20, thereby blocking the

interaction between Mdm2 and p53, and subsequently alleviating p53 degradation (45).

In response to DNA damage or ultraviolet-induced apoptosis, the acetyltransferases

CBP/p300 and the homeodomain-interacting protein kinase-2 (HIPK2) along with the tumor suppressor AXIN are recruited to PML NBs where CBP/p300 and HIPK2 acetylate p53 at K382 and phosphorylate it at Ser46, respectively (46-49). Both of these modifications activate p53 transcriptional activity and induce cell apoptosis or senescence

(19,49). By contrast, the deacetylase SIRT1 is also recruited to PML-NBs upon overexpression of PML or activation of oncogenic Ras (Ha-ras V12) and deacetylation of p53 by SIRT1 represses p53 transcriptional activity (50). Thus, p53 can be stabilized or destabilized in PML-NB depending on the composition of the NBs. This may be cell type-specific and dependent on the conditions used in these studies. In sum, many PML is capable of regulating p53 protein abundance and activity by multiple mechanisms that include sequestration of Mdm2-dependent PTM of p53 and SIRT1- dependent deacetylation of p53.

Akt pathway

Activation of Akt results in phosphorylation of numerous substrates, which regulate metabolism, survival, migration and cell cycle progression (51). PML inhibits Akt activation by sequestering Akt and recruiting protein phosphatase 2a (PP2a) to PML-NBs

(52). In PML-NBs PP2a dephosphorylates Akt and inhibits its kinase activity (52).

Furthermore, PML can suppress Akt activity via the eIF4E-NBS1-PI3K-Akt axis (53).

PML directly interacts with and negatively regulates elF4E activity in PML NBs, thereby reducing eIF4E-dependent mRNA export, including mRNA for NBS1, an upstream

activator of the phosphoinositide-3 kinase-Akt pathway (53). PML also positively regulates PTEN (phosphatase and tensin homolog), a suppressor of PI3K/Akt activation.

Monoubiquitination of PTEN is required for its nuclear localization and tumor suppressor activity and deubiquitination by the deubiquitinase HAUSP blocks PTEN nuclear localization (54). Inactivation or loss of PML results in a decrease in nuclear PTEN (55).

In PC3 prostate cancer cells, overexpression of PML opposes HAUSP deubiquitination

activity. PML binds to and inhibits the death domain associated protein DAXX, which

stabilizes HAUSP (55). A recent study reported that cytoplasmic PML is also essential

for Akt- and PP2a-dependent activation of 1,4,5-triphosphate receptor (IP(3)R)

phosphorylation, which triggers calcium release from the endoplasmic reticulum to

initiate apoptosis (56). In summary, PML regulates cell proliferation and survival by

inhibiting Akt kinase activity through PP2a, eIF4E and HAUSP.

Potentiation of Rb activity

The retinoblastoma protein (RB) is a potent tumor suppressor through its inhibitory effect

on E2F transcription factors hosphorylation of RB (pRB) blocks its interaction with E2F

and promotes cell cycle progression. PML-NBs recruit protein phosphatase 1α (pp1a),

which dephosphorylates RB, thereby promoting the interaction between RB and E2Fs

and repressing E2F-driven transcription and cell cycle progression (57,58). Oncogenic

Ras induces PML protein expression in mouse embryonic fibroblasts (MEFs), which results in colocalization of RB to PML NBs, and hypophosphorylation of RB with subsequent cell senescence (59).

Transcriptional regulation by PML

PML-NBs can sequester the NF-κB subunit RelA/p65, and inhibit its transcriptional

activity in TNFα-induced apoptosis (60). A20 is a NF-κB target gene that inhibits TNFα- induced apoptosis in a negative feedback fashion. PML represses NF-κB-mediated A20

transcription by preventing NF-κB from binding to the A20 promoter (61). PML can also

sequester Sp1 and Nur77 to NBs and disrupt their binding to target promoters (62,63).

PML interacts and inhibits STAT3 by inhibiting STAT3 DNA binding activity (64). PML

also forms complexes with multiple corepressors (c-Ski, N-CoR, and mSin3A) and

histone deacetylase 1 (HDAC1), which are required for the tumor suppressor Mad to

mediate its transcriptional repression (65). Together, these studies support a model in

which PML represses gene transcription by sequestering transcription factors to PML-

NBs or by associating with transcriptionally repressive complexes.

In contract, several mechanisms have been proposed by which PML positively regulates transcription. PML and PML-NBs recruit DAXX, which also functions as a transcriptional co-repressor, thereby transcriptionally derepressing DAXX target genes, such as Pax3 (23) and GRα target genes (66-68). Such regulation depends on Sumo1 conjugation of PML and a SIM (sumo interacting motif) in DAXX. We have previously reported that in response to TNFα stimulation, PML NBs sequester HDAC7, reducing its association with the MMP-10 promoter, thereby inducing MMP-10 expression (69). In addition, PML blocks degradation of the class II transactivator (CIITA), thereby stabilizing the protein, and promotes the expression of its target genes that include the class II major histocompatibility complex (70). The PML II isoform associates with transcription factors NF-κB, STAT1 and CREB-binding protein (CBP) to facilitate transactivation complex formation and activate interferon beta (INFβ) and interferon-

responsive gene (IRG) expression (71). However, it is not known whether PML II is present at these promoters. PML, p300 and β-catenin form complexes and activate the transcription of a subset of β-catenin responsive genes that include ARF and Siamois (72).

Interestingly, PML collaborates with the known oncoprotein c-Fos and enhances AP-1 transcriptional activity in a transient transfection reporter assay (73) and is essential for c-

Jun DNA binding and transcriptional activation in response to UV irradiation (74). PML is also required for all-trans retinoic acid (AT-RA)-induced transactivation of the p21WAF1/CIP1 gene (32). Moreover, PML physically associates with GATA1 and GATA2,

the master transcriptional factors of hematopoietic stem cell development, facilitating

their transcriptional activities (75,76). In summary, PML activates gene transcription

through sequestration of transcriptional co-repressors, stabilization or post-translational

modification of transcriptional factors, and possibly other mechanisms yet to be

elucidated.

The role of PML in DNA damage repair

Recently it has been suggested that PML and PML-NBs play a critical role in DNA

damage repair and ALT (77-79). ALT is an alternative mechanism of telomere maintenance in immortalized human cells and cancer cells that is telomerase-independent

(80). In ALT cells, PML co-localizes with telomeric DNA, the telomere-binding proteins

TRF1 and TRF2 as well as proteins involved in DNA synthesis and recombination, such as NBS1, Mre11, Rad51 and Rad52 (78,79). By binding these proteins, PML and PML-

NBs play a role in DNA damage responses, which are is important for the maintenance of genomic stability and integrity in ALT cells (81).

PML also co-localizes, associates with and stabilizes the DNA damage response protein,

TopBP1 after ionizing radiation (IR) (82). Upon the induction of double strand breaks

(DSBs), NBS1, ATM, Chk2 and ATR facilitate biogenesis of PML-NBs (83). The 3’-> 5’

DNA helicase, BLM, is an important regulator of the maintenance of genomic stability

and has also been shown to reside in PML-NBs (84). Interestingly, loss of BLM or PML results in increased numbers of sister-chromatid exchanges (SCE). BLM, RAD51 and replication protein (RP)-A assemble in PML-NBs during late S/G2 phase in undamaged cells and again after DNA damage (85). The RAD50-Mre11-NBS1 complex is implicated in the maintenance of telomere length in the absence of telomerase and plays a role in repair of DSBs, including homologous and non-homologous recombination repair (NJEM)

(86). Following IR treatment, the RAD50-Mre11-NBS1 complex is co-localized in PML-

NBs at sites of DSBs, suggesting a role of PML in repair of DSBs (87,88). In summary,

PML has multiple roles in both DNA damage repair and maintenance of genomic stability.

Janus-faced role of cytoplasmic PML in tumorigenesis

Cytoplasmic PML has been reported to have both oncogenic and tumor suppressive functions in different biological contexts. A cytoplasmic isoform of PML that contains exons 1-4, 6 and 7 and part of exon 9, was identified in plasmacytoma J558 cells (89).

This isoform of PML contributes to MHC class I antigen presentation, and enables tumors to evade the immune defense of its host (89). In APL cells, the PML-RARα fusion protein can be cleaved after V420 or V432 of the PML protein by neutrophil elastase and form a truncated PML protein that does not have the NLS and localizes in the cytoplasm (90). Mutations in PML are not common but cytoplasmic PML can also

result from mutations. A small deletion (1272delAG) and a splice site mutation (IVS3-

1G→A) in the PML gene have been identified in aggressive from of APL. The mutant

PMLs generated are truncated and do not have a nuclear localization signal (NLS) (91).

They localize in the cytoplasm due to a premature stop codon before the NLS. Cells from the APL patients are resistant to retinoic acid treatments and have reduced levels of apoptosis and increased proliferation (90-93). In addition, the previously described truncated PML mutant derived from APL mutations (91) can sequester nuclear PML in the cytoplasm through dimerization and inhibits p53 tumor suppressive functions (94).

Additionally, increased expression and cytoplasmic localization of PML was observed in a hepatocellular carcinoma (95,96). However it was unclear whether the PML in this tissue contained mutations. Therefore, whether cytoplasmic wild-type PML promotes tumorigenesis is still debatable.

Emerging evidence suggests that cytoplasmic PML can also have tumor suppressor functions. The M2 type pyruvate kinase (PKM2) is overexpressed in many cancers (97).

A PML mutant, which harbors an NLS mutation and is constitutively cytoplasmic, interacts with and inhibits PKM2 activity and lactose production (98). The transforming growth factor beta (TGFβ) can promote or suppresses tumorigenesis, depending on the cellular context (99). Lin et al. reported TGFβ treatment for 24 hr specifically induces a cytoplasmic PML isoform, which contains exons 1-3, 7a, 8a and 8b and lacks the NLS.

This cytoplasmic PML isoform facilitates the assembly of the

TβRI/TβRII/SARA/Smad2/3 complex in endosomes and is required for Smad2/3- dependent transcription. Such transcription is critical for TGFβ-mediated inhibition of cell proliferation, apoptosis and cell senescence (100). Additionally, overexpression of

the homeodomain protein TGIF results in nuclear retention of such cytoplasmic PML and

blocks TGFβ signaling (101). Together, these reports conclude that cytoplasmic PML

regulates the TGFβ pathway to promote its tumor suppressor activity.

In MEFs, a fraction of PML localizes to the endoplasmic reticulum and to mitochondria-

associated membranes (MAM) (56). At these sites, PML forms a complex with IP(3)R,

Akt and PP2a. Overexpression of a fusion protein containing the entire PML protein that

was targeted to the outer surface of the ER in MEFs promotes apoptosis by stimulating

calcium release. In PML-/- MEFs, Akt-dependent phosphorylation of IP(3)R is enhanced and calcium release from ER is decreased, thereby impairing the apoptosis response to

H2O2 or menadione (56). These findings suggest that cytoplasmic PML possesses tumor

suppressive activity.

Regulation of PML expression and therapeutic opportunities

Inactivation of PML in cancer cells occurs through multiple mechanisms (1,102).

However, only few somatic mutations have been reported so other mechanisms must be involved (1,102). Studies have indicated that effects on PML accumulation occur are at the transcriptional and post-translational levels. Epigenetic regulation of PML expression and alternative splicing of PML mRNA are less well studied (35). In many types of cancers, down-regulation of PML protein, but not its mRNA, is observed. Thus, post- transcriptional regulatory mechanisms are involved (1,27-29). This observation provides therapeutic opportunities to target cancer cells with the goal of restoring PML protein

expression by altering PML translation, localization or post-translational modification.

Transcriptional and translational regulation

Several reports have suggested that inflammation-associated cytokines enhance PML

transcription. The PML promoter contains an IFNα/β stimulated response element and an

IFNγ binding site (103,104). Interferons (INFs) have been shown to induce senescence

(105), a key anti-cancer mechanism. IFNs induce PML transcription through activation of

the Janus kinase/signal transducer and activator of transcription (JAK/STAT) pathway

(103,104,106). Tumor necrosis factor alpha (TNFα) also activates PML transcription by promoting STAT1-dependent transactivation of the PML promoter (107,108). Moreover, interleukin 6 (IL-6) enhances PML transcription via NF-κB and JAK-STAT pathways

(109). In summary, PML transcription is tightly regulated by various cytokines.

In response to K-Ras-induced cellular senescence, p53 and its homolog p73 activate PML transcription, but this activation can be attenuated by Akt/PKB (110,111). Furthermore,

β-catenin and plakoglobin are capable of activating the PML promoter in a LEF/TCF- independent manner in p53-negative KTCTL60 renal carcinoma cells (72).

In addition to transcriptional regulation, PML mRNA translation can also be regulated. In rodent cells, oncogenic K-Ras activates Pml mRNA translation in an mTOR- and eIF4E- dependent manner, presumably by targeting the Pml 5’-untranslated region of its mRNA

(112). We have recently demonstrated that the 5’-UTR of the human PML mRNA harbors an internal ribosome entrance site (IRES) that can be activated in response to

TNFα. This IRES is conserved in most mammals except mouse (113).

Post-translational regulation

In most cancers, PML protein level is down-regulated. However, the PML transcript level is usually comparable between normal and cancerous tissue (1). These observations

suggest that PML protein abundance is controlled post-transcriptionally. PML protein abundance and its functions are regulated by multiple post-translational modifications

(PTMs), including ubiquitination, SUMOylation, phosphorylation, acetylation and

peptidyl-prolyl isomerization (114,115) (Table 1). Recent evidence indicates that there is

crosstalk among these PTMs, which adds a complex layer of regulation to the control of

the PML protein expression/function (3).

Inhibition of the proteasome pathway restores PML protein expression in select cancer

cell lines (1,116). This observation suggested the possibility that abnormal ubiquitination

and subsequent degradation of PML in cancer cells was involved. This prompted a search

for the relevant E3 ligases targeting the PML protein. So far, at least seven E3 ligases

have been identified that can ubiquitinate PML including RNF4, UHRF1, E6AP, KLHL1,

KLHL20 and SIAH1/2 (117-123). Interestingly, KLHL39 (kelch-like family member 39)

interacts with PML and disrupts the binding of KLHL20 to PML and blocks KLHL20-

mediated ubiquitination of PML (124). We have previously shown that the peptidyl-

prolyl cis-trans isomerase Pin1 binds phosphorylated PML at multiple sites that include

S403 and S518 and promotes its degradation in triple-negative MDA-MB-231 breast

cancer cells (138). Additionally, AT-RA promotes Pin1 degradation and potently inhibits

human triple-negative breast (TNB) cancer cell growth and tumor growth in TNB cancer animal models (125). Moreover, the phosphatases SCP1, SCP2 and SCP3 dephosphorylate PML at S518, thereby blocking Pin1- and CDK2-dependent PML ubiquitination as well as KLHL20-mediated degradation (126). By contrast, USP11 promotes deubiquitination and stabilization of PML (127).

PML is subject to SUMO1 monosumoylation on K490 and SUMO2/3 polysumoylation

on K65 and K165. The E3 ubiquitin ligase, RNF4, binds polysumoylated PML through its SIMs and promotes SUMOylation-dependent ubiquitination (117). Interestingly, depletion of SUMO-3 reduces the number and size of PML-NBs (128). SUMOylation of

PML facilitates the recruitment of partner proteins to PML-NBs through their SIMs (16).

RanBP2, SIRT1, HDAC7 and PIAS1 have been shown to promote PML SUMOylaiton, while MageA, a subfamily of the melanoma antigen genes, attenuates PML

SUMOylation (3,129-133). Arsenic trioxide (ATO) is cytotoxic and ATO-mediated degradation of the PML-RARα fusion protein contributes to its therapeutic effect for

APL patients (134-137). This process requires direct binding of ATO to PML protein

(138) and depends on SUMOylation-dependent, ubiquitin-mediated degradation by

RNF4 (117,118).

Phosphorylation of PML can also modulate PML protein stability. In response to growth factors, IGF-1 or EGF, hypoxia, ERK2 or CDK1/2 phosphorylation of PML is enhanced which in turn promotes the interaction between phospho-PML and Pin1 (132,139). This interaction facilitates Pin1-mediated protein isomerization (121,140,141) followed by ubiquitination-mediated protein degradation. By contrast, high doses of H2O2 disrupt the

PML and Pin1 interaction, thereby stabilizing PML (132,139). The CK2 kinase

phosphorylates PML S565 and promotes PIAS1-mediated degradation of PML, although

the identity of the putative ubquitin E3 ligase is unknown (131,142). Similarly, the Big

MAP Kinase 1 (BMK1) down-regulates PML protein levels by phosphorylating PML at

S403 and T409 and promoting its degradation, thereby disrupting the interaction between

PML and Mdm2 and suppressing p53 activity (44,143). Unlike CK2, CDK1/2 or BMK1,

DNA damage-activated HIPK2 promotes PML phosphorylation at S8 and S38, resulting

in stabilization of PML (144).

PML is also subjected to acetylation at K487 and K505 by the protein acetyltransferase

p300 (145). Through screening all 18 known HDACs, we demonstrated that SIRT1- and

SIRT5-mediate deacetylaiton of PML at K487 which is indispensible for H2O2-induced accumulation of nuclear PML and NBs and cell death in HeLa cells (3). Furthermore, nuclear localization of PML is essential for H2O2-induced cell death (3).

Accumulating evidence indicates that crosstalk between the PTMs controls PML function.

For example, the interaction between the ubiquitin E3 ligase, RNF4, and PML ubiquitination requires PML SUMOylaiton by Sumo2/3 (128). Phosphorylation of PML protein by CDK1/2 or ERK2 is essential for Pin1 binding and Pin1-mediated protein isomerization (121,140,141). CK2-mediated phosphorylation promotes proteasome-and ubiquitination-mediated degradation of PML (142) and the deacetylase SIRT1 promotes

PML sumoylation and increases PML and PML NB abundance (129). Lastly, we demonstrated that acetylation at K487 and sumoylation at K490 in PML are mutually exclusive, suggesting a negative crosstalk between these two modification (3).

Mechanisms underlying nucleocytoplasmic shuttling of PML

All nuclear PML isoforms harbor an NLS. Disruption of the NLS by mutation at K487 results in accumulation of PML in the cytoplasm (3,12,146). In addition, the longest isoform, PMLI, also contains a C-terminal putative NES (Figure 2). An early study suggested that this NES is functional, but inefficient (147). Currently, the mechanism by

which the C-terminal NES regulates nucleocytoplasmic trafficking of PML1 and how the

activity of the NES is regulated remain unknown.

In most studies, PML is localized both in the nucleus and cytoplasm. This can involve

active re-distribution of PML. For example, in response to high doses of H2O2, SIRT1

and PML move from the cytoplasm to the nucleus and promote cell death in HeLa cells

(3). The HDAC catalytic activity of SIRT1 is essential for this H2O2-induced accumulation of nuclear PML. Because SIRT1 promotes deacetylation of PML at K487, a residue lying in the center of NLS, acetylation of K487 may influence PML nuclear localization by blocking recognition of the NLS by importins.

Recently, we discovered that oxidative stress and antioxidants control the subcellular distribution of PML. The antioxidant sulforaphane (SFN) is a potent inducer of cytoprotective genes (148). The precursor of SFN, glucoraphanin, is abundant in cruciferous vegetables with its highest concentration found in broccoli (149). Recent studies indicate that SFN induces apoptosis in cancer cells, inhibits cancer cell proliferation (150) and suppresses tumorigenesis in various mouse models of cancer

(151). We have recently demonstrated that PML is essential for SFN-mediated inhibition of capillary tube formation and migration of endothelial cells (147). Notably, SFN induces an accumulation of cytoplasmic PML and a reduction in nuclear PML, although the underlying mechanism has not been elucidated. The role of PML nucleocytoplasmic trafficking in cellular activity remains an intriguing issue to address.

Conclusion and Perspective

One key direction for future study will be the role of PML in epigenetics and chromatin

organization. Many histone modifying enzymes and enzymatic components of chromatin

remodeling complexes interact with PML. For example, protein acetyltransferase (p300),

deacetylase (HDAC1, HDAC2, HDAC3, HDAC7, SIRT1 and SIRT5) (3,65,69,152,153),

methyltransferases (SETDB1 and SUV39H1) (154,155), component of polycomb

repressive complex (EZH2) (156), and epigenetic regulator UHRF1 physically associate

with PML (119). However, little is known about whether PML controls the activity of

these chromatin regulators. Understanding the epigenetic regulation by PML is a pivotal

step toward elucidating the mechanism of tumor suppression by PML and is reactivating

PML in cancer cells. Currently, γ-irradiation and chemical therapies IFN and IL6, have

been shown to stimulate accumulation of PML protein (49,103,104,109). PML protein

can also be induced by several small molecules, including MLN4924 (target KLHL20),

emodin (target CK2), XMD8-92 (target BMK1), TSA (HDAC inhibitor), as well as other

stimuli such as ROS/H2O2, EGF, sulforaphane, MG132, As2O3 and DNA damage regents

(115,146,157,158). It will be informative to see whether combinatorial treatment with

these reagents enhances potency of their anti-cancer activity by synergistically increasing

PML protein accumulation.

Finally, studies from Pandolfi’s group suggested that PML has an oncogenic function in

chronic myeloid leukemia due to its importance in the maintenance of hematopoietic stem cells (159). The same group also reported that PML is overexpressed in TNB cancer patients and suggested that PML is an oncoprotein in TNB (160). Our lab recently demonstrated that Pml KO mice exhibited increased fatty acid oxidation in liver, which

may contribute to a reduced incidence of Western diet-induced dysplastic hepatic nodules

(161). How PML may switch from a tumor suppressor in one tissue to an oncoprotein in another tissue is an outstanding question and warrants further investigation.

Table 1. Summary of PML expression and post-translational modification regulators in cancer. PML protein is controlled by several post-translational modifications, including ubquitination, SUMOylation, acetylation, phosphorylation and isomerization.

Regulators Effect on PML Target region/sites Reference

KLHL20 (MLN4924), Down-regulates PML (119-123) SIAH, E6AP, UHRF1 protein abundance by promoting ubiquitination RNF4 SUMO dependent (117,118) KLHL39, USP11 Up-regulates PML protein (124,127) by blocking ubiquitination MageA, HDAC7, Regulates PML K65, K160, K490 (130-132) PIAS1 SUMOylaiton K490 (3,129) SIRT1 p300, Acetylation K487, K515 (145) SIRT1, SIRT5 De-acetylation K487 (3,12) CDK1/2, Phosphorylation S518 (121) BMK1 (XMD9-92), S403, T409 (143) CK2 (Emodin), S565* (142) HIPK2, S8, S38 (144) ERK2 S403, S505 (141) Chk2 S117 (45) SCP1, SCP2 and SCP3 De-phosphorylation S518 (126) PIN1 Isomerization pS518-P519 (121,140,141)

* annotated as S517 in reference due to different PML isoform.

CHAPTER 2: The epigenetic regulator UHRF1 promotes ubiquitination-mediated degradation of the tumor suppressor protein promyelocytic leukemia protein (PML)

This chapter has been published in Oncogene (2).

ABSTRACT

The promyelocytic leukemia (PML) protein is a tumor suppressor originally identified in

acute promyelocytic leukemia (APL) and implicated in tumorigenesis in multiple forms

of cancer. Here, we demonstrate that the PML protein undergoes ubiquitination-mediated

degradation facilitated by an E3 ligase UHRF1 (ubiquitin-like with PHD and RING finger domains 1), which is commonly up-regulated in various human malignancies.

Furthermore, UHRF1 negatively regulates PML protein accumulation in primary human umbilical vein endothelial cells (HUVECs), HEK293 cells and cancer cells. Knockdown of UHRF1 up-regulates, whereas ectopic overexpression of UHRF1 down-regulates, protein abundance of endogenous or exogenous PML, doing so through its binding to the

N-terminus of PML. Overexpression of wild-type UHRF1 shortens PML protein half-life and promotes PML polyubiquitination, while deletion of the RING domain or co- expression of the dominant-negative E2 ubiquitin-conjugating enzyme, E2D2, attenuates this modification to PML. Finally, knockdown of UHRF1 prolongs PML half-life and increases PML protein accumulation, yet inhibits cell migration and in vitro capillary tube formation, while co-knockdown of PML compromises this inhibitory effect. These findings suggest that UHRF1 promotes the turnover of PML protein, and thus targeting

UHRF1 to restore PML-mediated tumor suppression represents a promising, novel anti- cancer strategy.

INTRODUCTION

The epigenetic regulator UHRF1 (ubiquitin-like with PHD and RING finger domains 1)

harbors multiple predicted functional elements, including N-terminal Ub-like, tudor, PHD,

SRA and RING domains. By binding to hemi-methylated DNA, UHRF1 is known to

maintain genomic DNA methylation by recruiting DNA methyltransferase, DNMT1, to

DNA replication forks (162-165). Moreover, UHRF1 epigenetically regulates gene transcription by coordinating with histone deacetylase 1 (HDAC1) (166), binding to methylated histones (167,168), and promoting histone ubiquitination (169). However, the role of UHRF1 in genome-independent processes has not been rigorously explored.

UHRF1 is up-regulated in several cancers, including breast (170,171), lung (172,173), colorectal (174-176), prostate (177,178) and bladder cancer (179,180), suggesting that

UHRF1 plays a role in carcinogenesis, and is a putative anti-cancer target. In tumor cells,

UHRF1 represses the transcription of several tumor suppressor genes including p16INK4A, hMLH1, p21 and RB (166,181,182). Down-regulation of UHRF1 is associated with re-expression of tumor suppressor genes (183-185). In addition to its role in epigenetics, we previously reported that UHRF1 promotes DNMT1 ubiquitination and degradation (175). Intriguingly, overexpression of a RING domain deletion mutant of

UHRF1 that lacks E3 ligase activity sensitizes A549 lung cancer cells to treatment with various chemotherapeutics (169). This finding suggests that the UHRF1 RING domain is a critical functional domain involved in cancer cell proliferation and anti-apoptotic activity, and that the RING domain targets tumor suppressor proteins for degradation.

However, the mechanism by which UHRF1 regulates tumor suppressor proteins remains unresolved.

The promyelocytic leukemia protein (PML) was first isolated as a fusion partner of the

retinoic acid receptor alpha (RARα) associated with acute promyelocytic leukemia (APL)

(186). It was later demonstrated that PML is a tumor suppressor protein capable of promoting apoptosis, inhibiting proliferation, and inducing senescence (35). Moreover, we recently identified PML as a negative regulator of cell migration (139). PML protein is frequently down-regulated in various cancers, but PML mRNA levels are relatively similar between normal and cancerous tissues (1). These observations indicate that PML protein levels are tightly regulated, particularly through post-transcriptional modification.

Indeed, several E3 ligases including Siah2 (122), Vmw110/ICP0 (187), E6AP (120),

KLHL20 (121) and RNF4 (117,118) promote PML ubiquitination and degradation. We have also reported that the peptidylprolyl cis/trans isomerase Pin1 promotes PML degradation through a phosphorylation-dependent mechanism (140).

The observations that UHRF1 is up-regulated, while PML protein is down-regulated in cancers, suggested that UHRF1 may regulate PML protein accumulation. In this study, we identify a novel interaction between PML and UHRF1, and demonstrate that UHRF1 negatively regulates PML by targeting it for ubiquitin-mediated degradation. We further demonstrate that UHRF1 promotes endothelial cell migration and capillary tube formation, in part by down-regulating PML protein levels. Our results support a model in which UHRF1’s E3 ligase activity promotes PML protein turnover to regulate endothelial cell migration and capillary tube formation.

MATERIALS AND METHODS

Cell Lines and Medium

Human Umbilical Vein Endothelial Cells (HUVEC, Lonza, C2519A) that had undergone

fewer than five passages were used in this study. Cells were maintained in Endothelial

cell Growth Medium-2 (EGM-2, Lonza, C2519A). HEK 293 and HeLa cells were maintained in DMEM (Cellgro) supplemented with 10% fetal bovine serum and 50 units/ml penicillin and streptomycin sulfate. Wild-type and HAUSP knockout DLD1 cells were grown as described previously (175). PC3 cells were maintained in F12K medium (Invitrogen) supplemented with 10% fetal bovine serum.

Plasmid Construction and Transfection

The UHRF1 expression construct was kindly provided by Dr. Christian Bronner and subcloned into the pCMX plasmid. GST-UHRF1 and GST-FLNB (Immunoglobulin-like domain 11-15, R11-15) were constructed by PCR using HA-UHRF1 and HA-FLNB as templates respectively, and subsequently subcloned into the pGEX4T-1 vector. PML1,

PML4, or PML6 were inserted into pCMX-HA or pCMX-FLAG plasmids as described previously (140). Flag-HAUSP was described previously (175). Full-length UHRF1 and different truncations of UHRF1 and PML were generated by PCR and subcloned into pCMX-HA or pCMX-FLAG plasmids. The dominant-negative E2D2, CMX-His-Myc-

dnD2E2, was generated by PCR using His-dnE2D2 (kindly provided by Dr. Cam

Patterson) as a template. The UHRF1 mutants, C741A and ∆Ring (in which amino acids

724 to 763 were deleted) were previously described in (177) and was generated by site- specific PCR mutagenesis and verified by sequencing. Lipofectamine 2000 (Invitrogen) was used for transfection according to the manufacturer’s instructions.

Immunofluorescence Microscopy

Immunofluorescence microscopy was carried out as described previously (141) with the following modifications. Primary antibody incubation with anti-PML, anti-UHRF1 was carried out at room temperature for 2 h. After washing, Alexa Fluor secondary antibodies

(Invitrogen) were added and incubated for 40 minutes in the dark. Nuclei were counterstained with DAPI (Vector Laboratories). All images were captured under the same settings using a Leica immunofluorescence microscope. siRNA Transfection and Antibodies

Non-targeting control (D-001810-01), PML (J-006547-05), and UHRF1 (J-006977-05 and J-006977-08) siRNAs and transfection regent DharmaFECT1 (T-2001) were purchased from Thermo Scientific. Anti-PML and UHRF rabbit polyclonal antibodies were purified in house. The following antibodies were purchased from: Sigma, β-actin

(A5441) and FLAG (F1084); Roche Applied Science: HA-HRP (12013819001); Santa

Cruz Biotechnology: rabbit polyclonal anti-PML (sc-5621) and anti-HA (sc-805) and mouse monoclonal anti-PML (sc-966) and anti-GFP (sc-9996), mouse monoclonal anti-

Myc (Cell signaling, 2276), and UHRF1 (BD Biosciences, 612264).

RNA isolation and Quantitative RT-PCR

HUVECs were transiently transfected with UHRF1 or scramble siRNAs for 72 h and total RNA was isolated using a PrepEase RNA Spin kit (Affymetrix). cDNAs were generated using iScript Reverse Transcription Supermix kit (Bio-Rad) according to the manufacturer’s protocol. The specific cDNAs of interest were amplified and quantified by real-time PCR using an iCycler (Bio-Rad) platform with 2X IQ SYBR Green

Supermix (Bio-Rad) and appropriate primers. The relative quantities of UHRF1 and PML

mRNA were normalized to 18S RNA, and shown as means ± SD from three independent

experiments. The primer sequences for qRT-PCR are described in Supplementary Table 1.

Immunoprecipitation and western analysis

HEK 293 cells, transfected with the indicated plasmids, or untreated HUVECs were

washed with 1X PBS and resuspended in NETN (20 mM Tris-HCl pH 8.0, 100 mM NaCl,

1mM EDTA, 10% glycerol, 1mM DTT, 0.1% NP-40) buffer along with a protease inhibitor cocktail (Roche Applied Science), followed by sonication to lyse the cells.

Lysed cells were centrifuged at 4 ̊C at 12,000 rpm for 10 min, and the supernatant was incubated with protein A conjugated beads for pre-clearing. To detect exogenous protein interactions, whole cell extracts were incubated with anti-HA antibody conjugated beads

(Sigma F2426) or anti-FLAG antibody conjugated beads (Sigma E6779). For endogenous protein interactions, whole cell extracts were incubated anti-PML antibodies for 3 h and then incubated with protein A conjugated beads for 2 h. The beads were washed with

NETN buffer five times, and supernatants were discarded. 2X sample buffer was added to the beads to elute immunoprecipitated protein, followed by SDS-PAGE and western blotting as previously described (141).

GST pulldown Assay

GST pulldown assays were performed according to a previously published protocol (188).

Purified immobilized GST-UHRF1 fusion protein was incubated with whole cell extracts from cells expressing the indicated proteins. After incubation, the beads were washed three times with NETN buffer and the supernatant was discarded. 2X sample buffer was added to elute protein from the beads followed by SDS-PAGE and western blotting.

Determination of endogenous and transfected PML protein half-lives

To determine the half-life of endogenous PML in HUVECs, cells were transiently

transfected with a non-targeting siRNA, or two independent UHRF1 siRNAs. Forty-eight

hours later, cells were trypsinized and equal numbers of cells was plated. Twelve hours

later, cells were treated with cycloheximide (20 µg/ml) for the indicated times. To

determine the half-life of transfected PML4, HeLa cells were transiently transfected with

UHRF1 with a non-targeting siRNA, or two independent UHRF1 siRNAs. Twenty-four

hours later, cells were transfected with wild-type or mutant HA-PML4 expression

plasmids. Another 24 h later, cells were trypsinized and equal number of cells was plated

for 12 h followed by cycloheximide (20 µg/ml) treatment as described above. The half- lives of endogenous and transfected wild-type or mutant PML were determined by

Western blotting and the intensity of the signals was quantified by ImageJ.

Ubiquitination Assay

The ubiquitination assay was performed as described (189). Briefly, HEK 293 cells were transfected with the indicated plasmids (Fig. 7B) and treated with 20 mM MG132 for 4 h before harvest. Cells were lysed with RIPA buffer (20 mM Tris-HCl pH 7.5, 200 mM

NaCl, 1 mM EDTA, 1% Triton X-100, 1 mM DTT, and 0.1% SDS), and lysates were immunoprecipitated using anti-HA antibody conjugated beads (Sigma E6779) followed by western blotting with anti-FLAG (ubiquitin) antibody to detect PML ubiquitination.

After washing, the PVDF membrane was stripped and incubated with anti-HA antibody to detect PML as the loading control.

Wound-healing assay

HUVECs were transfected with the indicated siRNAs. 60 h later, cells were trypsinized

and 106 cells were reseeded on a 6-well tissue culture plate. After 6 h, the attached cells

were scratched with a 200 µl pipette tip and 0 h images were captured using a Leica

Wetzlar microscope. The plates were placed back at 37 ̊C and 5% CO2 for 8 h, and another set of images were captured of the same wounds. The wound widths were measured by ImageJ (vision 1.44, NIH), normalized and presented as the percentage of wound measured at time 0 (Mean ± SD). Percent of migration was calculated as the width of a scratch divided by the initial width of the same scratch times 100. At least five fields were analyzed for each scratch and each sample was performed in duplicate.

Transwell migration assay

Sixty hours after transfection by the indicated siRNAs, HUVECs were trypsinized, and

105 cells seeded on transwell inserts with 8 µm micropore filters (Corning Costar) in 200

µL medium. Medium containing 10% FCS was added to the lower chamber as a

chemoattractant. After 12 h, cells on the upper side of the filter were removed with a

cotton swab. The remaining cells were fixed and stained using DAPI, followed by

imaging on a Leica immunofluorescence microscope. Five random fields were imaged

per transwell insert and the number of cells that had migrated to the bottom side of the

membrane was counted using the particle counting module of ImageJ. Each assay was

repeated in three independent experiments resulting in 15 fields for each sample.

In vitro capillary tube formation Assay

The assays were performed using a commercially available kit from Millipore Inc. (ECM

625). The gel matrix was prepared as recommended by the manufacturer, and allowed to solidify in 96 well plate. 64 h post-transfection by the indicated siRNAs, HUVECs were

trypsinized, and 5 ×104 cells seeded onto the surface of the polymerized matrix. The

o plate was incubated at 5% CO2, 37 C for 3 or 8 h, and images were captured using a

Leica Wetzlar microscope. The branch points were counted and presented as Mean ± SD.

Unpaired two-tail t-tests were used to determine significance.

RESULTS

UHRF1 negatively regulates PML protein levels

To determine whether UHRF1 regulates PML protein accumulation, UHRF1 was

transiently knocked down using two different UHRF1 siRNAs in human umbilical vein

endothelial cells (HUVECs). We found that knockdown of UHRF1 by siRNAs resulted

in an increase in endogenous PML protein accumulation (Figure 5A) without having an

effect on its mRNA levels (Figure 5B). The multiple bands present in the PML blot

represent the various isoforms of PML and their post-translationally modified derivatives

(18). When compared with the scramble siRNA, UHRF1 knockdown led to an increase in

most PML isoforms. This observation was accompanied by an increase in the number of

PML nuclear bodies (NBs) (Figure 5C, top row and 5D). In contrast, overexpression of

HA-UHRF1 significantly decreased endogenous PML levels (Figure 6), while the

enzymatically defective mutants, C741A or ∆Ring (177), had only a minor effect (Figure

5E.). These data indicate that UHRF1 negatively regulates endogenous PML protein

levels post-transcriptionally in HUVECs.

We further examined whether this regulation extends to other cell types. HEK 293 and

HeLa cells were cotransfected with FLAG-PML4 and GFP (as a loading control), with or

without HA-UHRF1. Cells transfected with HA-UHRF1 expressed lower levels of

FLAG-PML protein than cells transfected with vector (Figure. 7A and 7B). It has been

reported that the herpes virus-associated ubiquitin-specific protease HAUSP (or USP7) stabilizes UHRF1 and that knockdown of HAUSP decreases UHRF1 protein levels

(190,191). Indeed, we observed a modest decrease in endogenous UHRF1 protein accumulation when

Figure 5. UHRF1 negatively regulates PML protein accumulation in HUVECs. A-B,

HUVECs were transfected with scramble or two different UHRF1 siRNA for 72 h. An aliquot of cells were used to analyze protein (A) or mRNA (B) levels. A, Whole cell lysates prepared from HUVECs transfected with scramble or two different UHRF1 siRNA were analyzed by immunoblotting with anti-UHRF1, anti-PML, and anti-β-actin antibodies. β-actin served as a loading control. The intensity of the bands was quantified by Image J and normalized to β-actin, and the relative PML protein levels were presented as [PML]/[β-actin]. B, Total RNA was isolated and analyzed by RT-PCR and quantitative RT-PCR to analyze PML expression. 18S RNA was used as an internal control. There was no significant change in 18S RNA observed between scramble and

UHRF1 siRNA transfected cells. C, HUVECs were transfected with indicated siRNAs for 72 h. Cells were immunostained with anti-PML and anti-UHRF1 antibodies, and images were taken by fluorescence microscopy (Original magnification 200×). D,

Numbers of PML-NBs in each cell were counted. Over 100 cells in duplicated experiments were counted and presented as means ± SD. Unpaired two-tail t-tests (***, p<0.001) were used for statistical analyses. E, HUVECs were transfected with HA-

UHRF1 (wild-type) (a-d), or HA-UHRF1 (∆Ring) (e-h) or HA-UHRF1 (C741A) (i-l) for

48 h. Cells were immunostained with anti-PML and anti-HA antibodies and images were taken by fluorescence microscopy. DAPI was used to indicate nuclei.

Figure 6. UHRF1 negatively regulates number of PML NBs and PML protein accumulation in HUVECs. HUVECs were transfected with HA vector (a-d) or HA-

UHRF1 (wild-type) (e-h) for 48 h. Cells were immunostained with anti-PML and anti-

HA antibodies and images were taken by fluorescence microscopy (Original magnification 200X). DAPI was used to indicate nuclei.

Figure 7. Ectopic overexpression of UHRF1 decreased PML protein levels in

mammalian cells. A-B, HEK 293 cells (A) or HeLa cells (B) were co-transfected with

HA-UHRF1, FLAG-PML4, and GFP for 48 h. Cells were lysed and analyzed by immunoblotting with anti-FLAG, anti-HA, and anti-GFP antibodies. GFP served as a transfection and loading control. C, The effect of overexpressing HA-UHRF1 on endogenous PML protein levels. Wild-type DLD1 cells were transfected with HA empty vector, whereas HAUSP knockout DLD1 cells were transfected with HA empty vector or

HA-UHRF1. Cell lysates were analyzed by immunoblotting using the indicated antibodies.

HAUSP is knocked out in DLD1 colon cancer cells. This observation is accompanied by an increase in endogenous PML protein abundance (Figure 7C). Furthermore, ectopic expression of HA-UHRF1 partially abrogates this effect on PML accumulation. We therefore conclude that UHRF1 is capable of down-regulating PML protein levels in several cell types.

UHRF1 interacts with PML

To elucidate the potential mechanism by which UHRF1 negatively regulates PML, we characterized the association of UHRF1 and PML using GST pull-down assays.

Immobilized, purified GST-UHRF1 fusion protein was incubated with cell extracts expressing three commonly studied PML splice isoforms, PML1, PML4 or PML6. We found that GST-UHRF1 was capable of pulling down all three PML isoforms (Figure

8A-C). This interaction was validated by reciprocal coimmunoprecipitation when FLAG-

PML6 was transfected with or without HA-UHRF1 (Figure 8D). Furthermore, we carried out coimmunoprecipitation with extracts prepared from HUVECs and found that endogenous PML and endogenous UHRF1 interact (Figure 8E). Therefore, we conclude that UHRF1 and PML physically associate in mammalian cells.

We also determined whether UHRF1 negatively regulates PML protein levels in other cancer cells. We carried out coimmunoprecipitation and found that PML antibodies co- precipitate UHRF1 in PC3 prostate cancer cells, indicating endogenous PML and UHRF1 interact (Figure 9A). Moreover, knockdown of endogenous UHRF1 increased PML protein accumulation (Figure 9B), but did not alter PML mRNA levels (Figure 9C).

Figure 8. UHRF1 interacts with PML. A-C, HEK 293 cells were transfected with HA-

PML1 (A), PML4 (B), or PML6 (C). Whole cell lysates were prepared and incubated with GST, or GST-UHRF1 for 1 h. The pulldown fractions were analyzed by immunoblotting with anti-HA antibodies. Full-length GST-UHRF1 is marked as asterisk.

D, HEK 293 cells were transfected with FLAG-PML with or without HA-UHRF1, and whole cell lysates were immunoprecipitated with anti-HA beads (lane 3 and lane 4), or anti-FLAG conjugated beads (lane 7 and lane 8). The immune pellets were analyzed by western blotting with anti-FLAG and anti-HA antibodies. (E) Whole cell lysates from

HUVECs were prepared and immunoprecipitated with anti-HA (lane 3) or anti-PML

(lane 4) antibodies. The immune pellets were separated by SDS-PAGE, and immunoblotting was performed with anti-UHRF1 and anti-PML antibodies. 2.5% of input (lanes 1 & 2) was loaded. L.E., longer exposure.

Figure 9. UHRF1 negatively regulates PML protein accumulation in PC3 prostate

cancer cells. A, Endogenous UHRF1 and PML interact. Whole cell extracts prepared from PC3 cells were immunoprecipitated with anti-PML antibodies followed by

immunoblotting with anti-PML and anti-UHRF1 antibodies. Five percent of input was

loaded in lane 1. L.E., longer exposure. B, Knockdown of UHRF1 increases PML protein

levels. C, Knockdown of UHRF1 does not significantly alter PML mRNA accumulation.

D, The UHRF1 E3 ligase activity is essential for UHRF1-mediated down-regulation of

PML protein levels. PC3 cells were transiently transfected with FLAG-PML4, GFP and

HA vector, HA-tagged wild-type, ∆Ring mutant UHRF1. Whole cell extracts were analyzed by immunoblotting with GFP, anti-HA and anti-FLAG antibodies.

Similar to that observed in HEK 293 cells, overexpression of wild-type UHRF1

significantly decreased PML protein levels. However, the ligase activity-defective

UHRF1 (∆Ring) mutant only modestly decreased PML protein levels. These data indicate

that UHRF1 negatively regulates PML protein accumulation in PC3 prostate cancer cells

(Figure 9D).

Mapping of interaction Domains

PML contains RING, B box1, B box2 and coiled coil (CC) domains (Figure 10A). To

map the interaction of UHRF1 and PML, we performed immunoprecipitation

experiments. We found that individual deletions of each domain of PML did not

significantly reduce the interaction between PML and UHRF1 in HEK293 cells (Figure

10B). This observation was further confirmed by GST pulldown assays, although

deletion of the B box 2 or CC domains displayed somewhat decreased PML binding to

UHRF1 (Figure 10C). Further mapping by coimmunoprecipitation and GST pulldown

assays demonstrated that UHRF1 interacts independently with amino acids 2-180 and

181-350 of PML (Figure 10D-E). In contrast, GST-FLNB (R11-15) fusion protein does not interact with these two domains (Figure 11). Together, these data indicate that

UHRF1 binds two separate regions of the PML protein.

UHRF1-mediated down-regulation of PML protein levels requires B box 2 and CC domains of PML

To further elucidate the mechanism by which UHRF1 down-regulates PML protein levels, we tested whether UHRF1 regulates PML protein half-life. Our data show that knockdown

Figure 10. Mapping UHRF1 interacting domain in PML. A, A summary of domain mapping results. PML possesses Ring, B box 1, B box 2 and coiled coil (CC) domains.

The summary of domain mapping study is based on GST pulldown and coimmunoprecipitation experiments. B, Cells were transiently transfected with the indicated plasmids, cell extracts prepared and immunoprecipitated with anti-FLAG antibodies. The immune pellets were immunoblotted with anti-FLAG or anti-HA antibodies. C, Immobilized, purified GST-UHRF1 was incubated with cell extracts expressing HA-tagged PML deletion mutants followed by immunoblotting with anti-HA antibodies. D-E, The experiments were performed as described in B and C respectively, except that a different set of PML expression constructs were used.

Figure 11. PML (1-180) or PML (181-350) does not bind GST-FLNB (R11-15).

Immobilized, purified GST, GST-UHRF1, or GST-FLNB (R11-15) fusion proteins were

incubated with extracts expressing HA-PML (2-180) or PML (181-350) for pulldown assays followed by immunoblotting with anti-HA antibody.

of UHRF1 resulted in a longer half-life of endogenous PML protein in HUVECs (Figure

12A-D and Figure 13). As shown in Figure 10C, deletion of B box 2 or the CC domain of

PML resulted in a decreased association of PML with UHRF1. To determine whether B

box 2 or the CC domain plays a role in UHRF1-mediated PML protein degradation, we examined whether deletion of B box 2 or the CC domain has an effect on PML protein stability. Wild-type PML, PML (∆ B box 2) or PML (∆ CC) were transiently transfected

into HeLa cells followed by half-life measurements. We found that deletion of B box 2 or

CC domain led to a longer half-life of PML protein, though B box 2 deletion has a much

stronger effect than the CC domain deletion (Figure 12E-H).

We reasoned that if the physical association between PML and UHRF1 plays a role in

UHRF1-mediated PML degradation, we anticipate that the half-lives of PML (∆ B box 2)

and PML (∆ B box 2) should not be affected by UHRF1 knockdown. Indeed, while

transfected wild-type PML protein is significantly stabilized in UHRF1 knockdown cells,

the half-lives of PML (∆ B box 2) or PML (∆ CC) are comparable in control and UHRF1

knockdown cells (Figure 12I-K and Figure 14). These data indicate that the interaction

between PML and UHRF1 is critical for UHRF1-mediated PML protein turnover.

UHRF1 promotes degradation of PML via polyubiquitination

UHRF1 possess a C-terminal RING domain that is thought to promote protein

ubiquitination. Indeed, we have previously shown that UHRF1 promotes ubiquitination-

mediated DMNT1 degradation (175). To determine whether the RING domain of UHRF1

is essential for the UHRF1 mediated decrease of PML protein levels, HEK 293 cells were

Figure 12. Knockdown of UHRF1 prolongs PML protein half-life. The half-life of endogenous (HUVECs A-D) and transfected PML (HeLa cells E-K). A-D, HUVECs were transfected with a non-targeting siRNA (A) or two independent UHRF1 siRNAs (B-

C) and the half-life of endogenous PML was determined as described in “MATERIALS

AND METHODS”. D, Quantification of the immunoblots from A-C. Knockdown efficiency of UHRF1 is shown in Figure 13. E-H, The half-life of wild-type and mutant

PML. HeLa cells were transiently transfected with wild-type PML (E), PML (∆B box 2)

(F), or PML (∆CC) (G) and the half-life of transfected PML was determined as described

in “MATERIALS AND METHODS”. H, Quantification of the immunoblots from E-G.

I-K, The half-life of transfected wild-type or mutant PML in UHRF1 knockdown cells.

HeLa cells were transiently transfected with a non-targeting siRNA or a UHRF1 siRNA.

Cells were trypsinized and equal numbers of cells were transiently transfected with wild- type PML (I), PML (∆B box 2) (J) or PML (∆CC) (K) expression plasmid. UHRF1 knockdown efficiency and the immunoblos are shown in Figure 14.

Figure 13. Knockdown efficiency of UHRF1 in HUVECs. The experiments were performed similar to that of Figure 5A.

Figure 14. The half-life of PML (∆B box 2) and PML (∆CC) in control and UHRF1

knockdown cells. A, Knockdown efficiency of UHRF1 in HeLa cells. B-D, Control

(lanes 1-6) and UHRF1 knockdown cells (lanes 7-12) were transiently transfected with

wild-type HA-tagged PML (B), PML (∆B box 2) (C) or PML (∆CC) (D). Determination

of the transfected PML protein half-life was performed as described in “MATERIALS

AND METHODS”.

57 co-transfected with HA-PML4 and GFP, with or without wild-type FLAG-UHRF1, or a

RING domain deletion mutant. As shown in Figure. 15A, wild-type UHRF1 (lane 1) was capable of decreasing HA-PML4 protein abundance. However, E3 ligase-defective mutants, ∆Ring, (lane 3) or the point mutant C741A (lane 4) did not or only slightly decrease PML protein levels, respectively. We further determined whether UHRF1 promotes PML ubiquitination. Cells were transiently transfected with HA-PML4, wild type or RING domain deletion mutant with or without FLAG-tagged ubiquitin (FLAG-

Ub) or a dominant-negative E2D2. The ubiquitin-conjugating enzyme E2D2 has been shown to play a role in UHRF1-mediated ubiquitination (169,192). Cells were treated with MG132 4 h prior to harvest followed by immunoprecipitation with anti-HA antibody-conjugated beads and immunoblotting with anti-FLAG or anti-HA antibodies.

As shown in Figure. 15B, in the absence of exogenous UHRF1, modest levels of PML ubiquitination was detected. This basal level is due to endogenous UHRF1 and other

PML E3 ubiquitin ligases. However, ectopically expressed wild-type Myc-UHRF1 (lane

2 vs lane 1), but not the RING domain deletion mutant (lane 3 vs lane 1) potently increased HA-PML4 ubiquitination. Consistent with this notion, the dominant-negative

Myc-tagged E2D2 (Myc-dnE2D2) abrogated UHRF1-mediated PML4 ubiquitination.

Together, these data indicate that E2D2 and UHRF1 promote ubquitination-mediated

PML degradation.

UHRF1 promotes HUVEC migration and capillary tube formation

We have previously demonstrated that PML negatively regulates cell migration (139).

Therefore, we hypothesize that knockdown of UHRF1 will inhibit cell migration, in part,

as a result of increased PML protein accumulation. Indeed, siRNA-mediated UHRF1

knockdown resulted in a significant inhibition of HUVEC cell migration as determined

by a wound-healing assay (Fig. 16). To further evaluate the role of PML in UHRF1 knockdown-mediated inhibition of cell migration, we determined the effects of a double knockdown of UHRF1 and PML on cell migration. We found that similar to that shown in Figure 5, PML protein accumulated in UHRF1 knockdown cells (Fig. 17A).

Accordingly, wound-healing assays showed that cell migration was inhibited upon

UHRF1 knockdown (Fig. 17B, samples siU-1 and siU-2), mirroring the results shown in

Fig. 16. However, this inhibition was largely relieved by co-knockdown of PML (Fig.

17B, samples siU-1/siPML and siU-2/siPML). This result was further confirmed by transwell migration assays (Fig. 17C). Additionally, we carried out in vitro capillary tube formation assays, and found that knockdown of UHRF1 significantly reduced tube formation, while further knockdown of PML ablated the inhibitory effect of UHRF1 knockdown (Fig. 17D). In summary, these data strongly suggest that UHRF1 promotes

HUVEC migration and capillary tube formation, in part by down-regulating PML protein accumulation.

Figure 15. The RING domain is essential for UHRF1-mediated PML ubiquitination

and degradation. A, HEK293T cells were co-transfected with HA-PML4 and GFP in

the presence of wild-type UHRF1 (lane 1), an empty vector (lane 2), a RING deletion

mutant, ∆RING (lane 3) or a point mutant C741A (lane 4). Whole cell lysates were analyzed by immunoblotting with the indicated antibodies. Note that the full-length

UHRF1 has two bands. The slower migrating band is likely a modified form of UHRF1

and its intensity is much lower than the faster migrating species. The ∆RING mutant also

has two bands. Interestingly, the intensity of the slower migrating band is slightly lower than that of the faster migrating band. B, UHRF1 and E2D2 promote PML polyubiquitination. HEK293 cells were co-transfected with indicated plasmids. Whole cell extracts were prepared and immunopreciptated with anti-HA antibodies followed by

immunoblotting with anti-FLAG or anti-HA antibodies.

Figure 16. Knockdown of UHRF1 inhibits HUVEC migration. A, Representative images of wound-healing assays. HUVECs were transiently transfected with a non- targeting or two different UHRF1 siRNAs. Sixty hours after transfection, wound-healing assays were performed as described in “MATERIALS AND METHODS”. B, Statistical analyses of wound-healing assays. C, UHRF1 is efficiently knocked down by siRNAs.

Whole cell extracts were prepared from HUVECs 60 h post siRNA transfection.

Immunoblotting was performed to determine UHRF1 knockdown efficiency. Wound- healing assays were performed as described in “MATERIALS AND METHODS”.

Unpaired two-tail t-tests (*, p<0.01) were used for statistical analyses.

Figure 17. Knockdown of PML alleviates UHRF1 depletion-mediated inhibition of

HUVEC migration, invasion and capillary tube formation. A, HUVECs were transiently transfected with the siRNAs targeting UHRF1 and/or PML. An aliquot of the cells was subjected to immunoblotting with anti- UHRF1 and anti-PML antibodies. SE, shorter exposure; LE, longer exposure. The intensities of the bands were quantified by

Image J and normalized to β-actin, which presented as [PML]/[β-actin]. An aliquot of the cells was seeded 60 h after knockdown followed by wound-healing (B), transwell migration (C) or capillary tube formation (D) assays as described in “MATERIALS AND

METHODS”. Unpaired two-tail t-tests (*, p<0.01; ns, not significant) were used for statistical analyses.

DISCUSSION

UHRF1 is best known as an epigenetic regulator due to its ability to modulate heterochromatin by binding hemi-methylated CpG dinucleotides (165), unmethylated histone H3R2 (193) and H3K4me0/K9me3 (194). Our present study identifies a novel

UHRF1 target, PML, and, conversely, a novel E3 ubiquitin ligase for PML. We demonstrate that UHRF1 interacts with PML to modulate its protein stability.

Furthermore, we show that the C-terminal RING domain is essential to promote ubiquitination-mediated degradation of PML, thereby providing a novel mechanism by

which UHRF1 might promote tumorigenesis.

Recent immunohistochemical staining and microarray analyses from various cancer

patients indicate that UHRF1 is overexpressed in several cancer types but not in normal

cells (172,173,179,195,196). While its role in tumorigenesis is thought to involve

promotion of cell proliferation and metastasis (173,174), the underlying mechanism is

still elusive. Conversely, the tumor suppressor protein PML, which inhibits cell cycle

progression and promotes apoptosis, is found to be down-regulated in cancer cells, while

its mRNA level remains unchanged (1). Interestingly, it was reported that a RING domain deletion UHRF1 mutant that lacks E3 ligase activity inhibits A549 lung cancer cell growth (177). This observation implies a critical role of the UHRF1 E3 ligase activity in the promotion of tumorigenesis. Therefore, identification of UHRF1 E3 ligase targets provides a promising avenue for elucidating the mechanism by which UHRF1 promotes tumorigenesis. In this study, we show that overexpression of the E3 ligase-

defective (RING domain deletion) UHRF1 mutant, or a dominant negative E2D2

ubiquitin conjugating enzyme significantly inhibited the ability of UHRF1 to promote

PML ubiquitination (Figure 15B). PML is known to be capable of promoting p53 protein

acetylation (47,197). Consistent with this finding, knockdown of UHRF1 increases p53

acetylation (Figure 18A) and p21 and GADD45 mRNA levels (Figure 18). However, this

increase is ablated by subsequent PML knockdown. Thus, our data support the hypothesis

that PML is a critical cellular target of UHRF1 and that overexpression of UHRF1 results

in down-regulation of PML protein accumulation.

PML is highly expressed in endothelial cells (ECs) (95), but its role in ECs remains largely unexplored. We have previously shown that TNFa induces PML protein

expression and promotes sequestration of HDAC7 to PML NBs, thereby de-repressing

HDAC7 target genes (69). We have also recently demonstrated an inhibitory role of PML in EC migration and capillary tube formation (198). In contrast, the role of UHRF1 in

ECs has not been explored. We show here that knockdown of UHRF1 in HUVECs results in an increase in PML protein accumulation without changing its mRNA levels and this increase is associated with reduced cell migration and capillary tube formation.

This inhibitory effect is significantly alleviated by further knockdown of PML. These data support a model in which UHRF1 promotes EC migration, invasion and capillary tube formation, in part, by decreasing PML protein abundance.

The herpesvirus-associated ubiquitin-specific protease HAUSP positively regulates

UHRF1 protein stability by protecting UHRF1 from autoubiquitination (190), while it

negatively regulates PML protein accumulation (199). These reports raise the possibility

Figure 18. The effect of UHRF1 on p53 acetylation. A, Knockdown of UHRF1 does not affect p53 acetylation. HUVEC were transiently transfected with the siRNA targeting

UHRF1 and/or PML for 72h. An aliquot of the cells was subjected to immunoblotting with anti-p53 acetyl-K382, p53, UHRF1, PML and β-actin. The intensities of the bands were quantified by Image J and normalized to β-actin. B, Knockdown of UHRF1 enhances p53 target gene expression. An aliquot of the cells were used to analyze mRNA level as indicated.

that HAUSP decreases PML protein accumulation indirectly, by up-regulating UHRF1 protein abundance. Our data show that ectopically transfected UHRF1 is capable of decreasing PML protein levels, regardless of the status of HAUSP (Figure 7C).

Intriguingly, HAUSP interacts with Vmw110/ICP0, the herpes simplex virus type 1 immediate-early protein (187), another PML E3 ligase (199). Similarly, HAUSP blocks

ICP0 autoubiquitination (200) and enhances ICP0 protein stability. Taken together, our data suggest that Vmw110/ICP0 is functionally a viral counterpart of the UHRF1 E3 ligase.

In summary, we have identified UHRF1 as a novel PML E3 ubiquitin ligase and therefore provided the first example by which UHRF1 negatively regulates tumor suppressor at the post-transcriptional level. Identification of other tumor suppressor proteins, which are down-regulated and ubiquitinated by UHRF1, will be essential to fully understand the role of UHRF1 in cancer and to facilitate UHRF1 as a target for cancer therapy.

CHAPTER 3: Deacetylation of the tumor suppressor protein PML regulates hydrogen peroxide-induced cell death

This chapter has been published in Cell Death and Disease (3).

ABSTRACT

The promyelocytic leukemia protein (PML) is a tumor suppressor that is expressed at a low level in various cancers. Although post-translational modifications including

SUMOylation, phosphorylation, ubiquitination have been found to regulate the stability or activity of PML, little is known about the role of its acetylation in the control of cell survival. Here we demonstrate that acetylation of lysine 487 (K487) and SUMO1 conjugation of K490 at PML protein are mutually exclusive. We found that hydrogen peroxide (H2O2) promotes PML deacetylation and identified SIRT1 and SIRT5 as PML

deacetylases. Both SIRT1 and SIRT5 are required for H2O2-mediated deacetylation of

PML and accumulation of nuclear PML protein in HeLa cells. Knockdown of SIRT1

reduces the number of H2O2-induced PML-nuclear bodies (NBs) and increases the survival of HeLa cells. Ectopic expression of wild-type PML but not the K487R mutant rescues H2O2-induced cell death in SIRT1 knockdown cells. Furthermore, ectopic

expression of wild-type SIRT5 but not a catalytic defective mutant can also restore H2O2-

induced cell death in SIRT1 knockdown cells. Taken together, our findings reveal a novel

regulatory mechanism in which SIRT1/SIRT5-mediated PML deacetylation plays a role in the regulation of cancer cell survival.

INTRODUCTION

The tumor suppressor PML protein, first identified in a t(15;17) chromosomal

translocation in patients with acute promyelocytic leukemia(4), is the essential component of a macromolecular nuclear substructure, called PML-nuclear bodies (PML-

NBs)(201). PML protein levels are frequently down-regulated (complete or partial loss)

in several types of human cancer and often correlate with tumor progression(1).

Overexpression of PML inhibits cell proliferation(31), while pml-/- cells grow faster than

their wild-type counterparts(32). Moreover, pml-/- cells are resistant to multiple apoptotic

stimuli, e.g. H2O2, tumor necrosis factor-α and ionizing radiation(38). There are multiple

PML spliced isoforms. All PML isoforms contain the N-terminal RBCC domain that is

followed by alternatively spliced C-terminus(115). In response to stress signals, PML-

NBs alter subnuclear localization and/or mediate post-translational modification (PTM) of target proteins in a spatiotemporal manner to control apoptosis, cell proliferation, and senescence(19). Importantly, PML itself is also the subject of PTM, including ubiquitination, phosphorylation, SUMOylation, and acetylation, which add a complex layer of regulation to the activity and stability of PML-NBs(115). We have previously reported that UHRF1 promotes ubiquitination-mediated degradation of PML(2); that Pin1 promotes PML degradation through a phosphorylation-dependent mechanism(140,141); and that HDAC7 stimulates PML SUMOylation by associating with the E2 SUMO ligase,

Ubc9(132). Recently, it was reported that deacetylation of PML by SIRT1 affects virus infections and circadian function(12,129). However, the role of PML acetylation in tumorigenesis, the effects of other Sirtuins or histone deacetylases (HDACs) on PML

protein, and the regulation of PML acetylation status in response to oxidative stress are

largely unknown.

Lysine acetylation and deacetylation have been recognized as crucial events for

regulating activity, stability, and subcellular localization of proteins.

Acetylation/deacetylation can function as an on/off switch by changing charge status, or

through crosstalk with other PTMs(202,203). Sirtuins (SIRTs1-7) are a family of protein

deacetylases that catalyze NAD+ dependent removal of acetyl groups from modified

lysine side chains in various proteins(204). SIRT1 is the most well characterized member of the sirtuin family and has a wide spectrum of substrates with important functions in aging, metabolism, and cancer(205,206). Emerging evidence has shown that SIRT1 positively or negatively modulates tumorigenesis, depending on the context(207). The differential subcellular localization of SIRT1 in normal and cancer cells may affect substrate accessibility and partially account for the contrasting roles of SIRT1 in tumorigenesis(208,209). For example, SIRT1 is mainly localized in the nucleus of normal cells, while it is predominately localized in the cytoplasm in cancer or transformed cells(210,211). Nuclear SIRT1 deacetylates and inactivates transcription factors, including NF-κB, STAT3, and HIF-1α(212-214), exerting anti-inflammatory and anti-carcinogenic effects. Moreover, nuclear SIRT1 reduces DNA damage and maintains genomic integrity by deacetylating DNA repair proteins, e.g. PARP1, XPC, and NBS1

(215-217). In contrast, cytoplasmic SIRT1 deacetylates and activates the oncoprotein

Akt(218) and stabilizes c-Myc protein(219). SIRT1 has also been reported to deacetylate p53, retinoblastoma (Rb)(220,221) and PTEN(222), inactivating their tumor suppressive activity. Unlike SIRT1, SIRT5 is less well characterized and only carbamoyl phosphate

synthetase 1 (CPS1) has been functionally identified as a SIRT5 substrate(223). A recent

study showed that SIRT5 is significantly down-regulated in neck squamous cancerous

tissues compared with noncancerous tissues. It is also down-regulated in advanced stages

compared to early stages of the disease(224), implying that it acts as a tumor suppressor.

PTMs, such as SUMOylation, phosphorylation and ubiquitination, have been found to

regulate the tumor suppressor function of PML; however, little is known about its

acetylation. In addition to SIRT1, we have identified SIRT5 as a novel interacting partner

of PML and demonstrated that nuclear SIRT1 and SIRT5 increase accumulation of PML-

NBs and that this activity is dependent on their deacetylase activity. This regulation may

play an important role in H2O2-induced cell death. Thus, our current work has elucidated a novel regulation of PML protein and uncovered potential opportunities for therapeutic intervention by targeting PML regulators, SIRT1 and SIRT5.

MATERIALS AND METHODS

Cell culture and transfection

HeLa cells were maintained in DMEM (Dulbecco's Modified Eagle Medium) (Cellgro)

supplemented with 10% fetal bovine serum and 50 units/ml penicillin and streptomycin

sulfate. HCT 116 p53−/− cells were maintained in in McCoy's 5A (Cellgro) supplemented

with 10% fetal bovine serum and 50 units/ml penicillin and streptomycin sulfate. Control

(scramble shRNA) and SIRT1 shRNA stable HeLa cell lines were cultured under the

same conditions with the addition of 0.25 μg/ml of puromycin as previously

described(225). Transfections were performed with Lipofectamine 2000 (Invitrogen).

Antibodies, siRNA and plasmids

Anti-PML, SIRT1 and SUMO1 rabbit polyclonal antibodies were purified in-house. The

specificity of in-house anti-PML antibodies was validated previously in HUVECs(2) and

in HeLa cells (Figure 19). The following antibodies were purchased: β-actin (A5441), α-

tubulin (T6074), and FLAG (F1084) from Sigma (St Louis, MO, USA); HA-HRP

(12013819001) from Roche Applied Science (Indianapolis, IN, USA), Myc (no. 2276)

and Acetylated-Lysine (no. 9441) from Cell Signaling Technology (Danvers, MA, USA);

lamin B (sc-6216) and mouse monoclonal anti-PML (sc-966, for immunofluorescence) from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Non-targeting control (D-

001810-01) and PML (J-006547-03 and J-006547-05) siRNAs and transfection regent

DharmaFECT1 (T-2001) were purchased from Thermo Scientific. Non-targeting control

(SHC002), SIRT1 (TRCN0000018981 and TRCN0000018983) and SIRT5

Figure 19. Validation of the specificity of PML antibodies. Hela cells were transiently transfected with two PML siRNAs. Whole cell extracts were prepared and subjected to

Western blotting with indicated antibodies.

(TRCN0000018545 and TRCN0000018546) shRNA plasmids were purchased from

Sigma (St Louis, MO, USA). FLAG-SIRTs1-7(225), Myc-SIRT1, Myc-SIRT1 (H363Y),

FLAG-SIRT5 and FLAG-SIRT5 were subcloned into the pcDNA 3.1 plasmid. Other expression plasmids were subcloned into pCMX plasmid as previously described

(2,132,140,141) and mutants were generated by site-specific PCR mutagenesis and verified by sequencing. To construct NLS fusion PML protein, NLS derived from the simian virus 40 (SV40) large tumor antigen (PKKKRKV) was added at the N-terminus of hemagglutinin (HA) tagged wild-type and mutant PML protein. CMX-HA-PML1, 2, 3, and 5 were generated by PCR using pcDNA3-PML1, 2, 3 and 5 (kindly provided by

Kun-Sang Chang) as templates and subcloned into CMX-1H vector.

Immunofluorescence microscopy

Immunofluorescence microscopy was carried out as described previously (2,141) with minor modifications. Primary antibody incubation with anti-PML and anti-HA was carried out at room temperature for 2 h. After washing, Alexa Fluor secondary antibodies

(Invitrogen) were added and incubated for 40 minutes in the dark. Nuclei were counterstained with DAPI (Vector Laboratories). All fluorescence images were acquired using a Leica DMI 6000B inverted microscope or a confocal system (PerkinElmer,

Waltham, MA, USA).

Immunoprecipitation and Western blotting analysis

HeLa cells at 70-80% confluency, transfected with the indicated plasmids or treated with

H2O2, were washed with 1X phosphate-buffered saline (PBS) and resuspended in RIPA

buffer (1 X PBS, 1% NP-40, 0.5% sodium deoxycholate, and 0.1% SDS) along with 1X

protease inhibitor cocktail and phosphatases inhibitor cocktail (Roche Applied Science).

To detect PML SUMOylation in HeLa cells, whole-cell extracts in the presence of N-

ethylmaleimide (NEM) were prepared. Lysed cells were centrifuged at 4 ̊C at 12,000 rpm for 10 min, and the supernatant was incubated with protein A conjugated beads for pre-

clearing. To detect protein-protein interactions or acetylation and SUMO1 modification

of PML, whole cell extracts were incubated with anti-HA antibody conjugated beads

(Sigma F2426) or anti-FLAG antibody conjugated beads (Sigma E6779) for 2 h. The

beads were washed with NETN buffer (20mM Tris–HCl, pH 8.0, 100mM NaCl, 1mM

EDTA, 10% glycerol, 1mM dithiothreitol and 0.1% NP-40) five times, and supernatants

were discarded. 2X sample buffer was added to the beads, followed by SDS-PAGE and

Western blotting as previously described (2,141).

Fractionation of cytoplasmic and nuclear extracts

Nuclear and cytoplasmic fractionation has been described previously (226) with minor modifications. Briefly, cytoplasmic extracts were made by resuspending whole cell pellets with cytosolic extraction buffer: CEBN (10 mM HEPES 7.8, 10 mM KCl, 2 mM

MgCl2, 0.34 M Sucrose, 10% Glycerol, 0.2% NP40/IPEGAL) for 10 min on ice followed

by centrifugation at 2,000 X g for 5 min at 4 °C. The nuclear pellets were then washed

once with CEB buffer (CEBN buffer without NP-40), pelleted, resuspended in 2X sample

buffer and sheared by sonication. The cytoplasmic and nuclear extracts were analyzed by

Western blotting with the indicated antibodies.

Cell death and colony formation assays in cells treated with H2O2

Control or SIRT1 knockdown HeLa cells were transfected with the indicated plasmids.

After 24 h, cells were trypsinized and 1x104 cells were reseeded on a 96-well tissue

culture plate. After 12 h, the cells treated with 8 mM H2O2 for 1 h, washed and incubated

with fresh media for another 8 hours and another set of cells without H2O2 treatment

served as the control. Total cell number was determined using a Cyquant cell

proliferation assay kit (C7026, Molecular Probes, Eugene, OR, USA). Cell viability was

calculated as the number of H2O2 treated cells divided by the number of non-H2O2 treated cells in five independent experiments.

For colony formation assay, 2 × 103 cells were seeded into 100-mm tissue culture plate.

After 12 h, the cells were treated with 8 mM H2O2 for 1 h, washed and incubated with fresh media. 9 days later, when macroscopic colonies became detectable, the cells were washed with PBS, and stained with crystal violet and the number of colonies scored.

Each experiment was carried out in triplicate.

Statistical Analysis—Statistical analysis was performed using two-tailed Student’s t test, using p < 0.05 as a criterion of significance.

RESULTS

H2O2 stimulates PML-NB accumulation and PML deacetylation

H2O2 has attracted increasing attention as a molecule that regulates fundamental

biological processes and pathological progression, including angiogenesis, oxidative

stress, aging, and cancer (226,227). To study the effects of H2O2 on PML, we treated

HeLa cells with H2O2 and examined subcellular distribution of PML by

immunofluorescence microscopy. We found that H2O2 induced nuclear accumulation of

PML and increased PML NB number (Figures 20A and 20B). The increase in nuclear

PML and PML-NBs in response to H2O2 is accompanied by a decrease in cytoplasmic

PML in H2O2 treated cells (Figure 20C).

To determine whether acetylation status of PML was altered in response to H2O2, HeLa

cells were treated with H2O2, harvested and followed by immunoprecipitation with anti-

PML antibody and immunoblotting with anti-acetyl-lysine and anti-PML antibodies. We found that acetylation of endogenous and transfected PML was decreased in H2O2 –

treated cells (Figure 20D-E). Taken together, we conclude that H2O2 promotes accumulation of nuclear PML and PML-NBs and decreases PML acetylation.

SIRT1 and SIRT5 interact and deacelylate PML at lysine 487

We next determined which deacetylase is capable of deacetylating PML. We examined

PML acetylation status after co-transfecting epitope tagged PML4 with plasmids expressing class I, II or III histone deacetylases, all of which also deacetylate non-histone proteins. Through this screening, SIRT1 and SIRT5 were found to promote PML4 deacetylation (Figure 20F and Figure 21). We further determined that both SIRT1 and

Figure 20. H2O2 induces accumulation of PML-NBs and deacetylation of PML in

HeLa cells. A, HeLa cells were treated with 4 or 8 mM H2O2 for 1 h. Cells were

immunostained with an anti-PML antibody followed by fluorescence microscopy

(Original magnification 200×). B, Numbers of PML-NBs in each cell were counted. Over

100 cells in duplicate experiments were counted and presented as the mean ± SD.

Unpaired two-tail t-tests (**, p<0.01 and ***, p<0.001) were used for statistical analyses.

C, HeLa cells were treated with the indicated concentration of H2O2 for 1 h. Nuclear and

cytoplasmic fractions were prepared and subjected to Western blotting with the indicated

antibodies. D-E, HeLa cells (D) or the cells transfected with HA-tagged PML4 (E) were

treated with 0, 4 or 8 mM of H2O2 for 0.5 h. Whole cell extracts were prepared and

immunopreciptated with anti-PML (D) or anti-HA (E) antibodies followed by

immunoblotting with anti-acetyl-lysine and anti-PML(D) or anti-HA (E) antibodies. F,

HeLa cells were transfected with HA-PML4 and FLAG-SIRTs. Whole cell extracts

(WCEs) were analyzed by immunoprecipitation with anti-HA antibodies followed by immunoblotting with anti-acetyl-lysine and anti-HA antibodies. The expression of SIRTs were examined by immunoblotting with anti-FLAG antibodies. The composite image in panel F was spliced from a Western blot.

Figure 21. PML deacetylase screening. A, HeLa cells were transiently transfected with

FLAG-PML4 and HA-HDACs. Whole cell extracts (WCEs) were analyzed by immunoblotting with anti-FLAG and anti-HA antibodies (lower panel) and immunoprecipitation with anti-FLAG antibodies followed by immunoblotting with either anti-acetyl-lysine and anti-FLAG antibodies (upper panels). B-C. Due to low expression of HDAC 6 (Fig. S1A lane 6) and SIRT4, we separately determined the effects of

HDAC6 (B) and SIRT4 (C) on decatylaiton of PML.

SIRT5 promote deacetylation of other nuclear PML isoforms in HeLa cells (Figure 23A-

E) and deacetylation of PML4 in HCT116 p53-/- cells (Figure 23F). To determine

whether PML deacetylation is dependent SIRT1/SIRT5 catalytic activity, HeLa cells

were co-transfected with HA-PML4 and wild-type SIRT1, SIRT5 or catalytically

impaired mutants, SIRT1 (H363Y) or SIRT5 (H158Y). We found that PML acetylation

was significantly abolished by co-expression with the wild-type SIRT1 or SIRT5, but not

catalytically defective mutants, SIRT1 (H363Y) or SIRT5 (H158Y) (Figure 22A and

22B). Conversely, knockdown of SIRT1 or SIRT5 modestly increased PML4 acetylation

(Figure 22C-D and Supplementary Figure 22G). Moreover, double knockdown of SIRT1

and SIRT5 dramatically increased PML acetylation (Figure 22E). We further

demonstrated that either endogenous or transfected SIRT1 and SIRT5 associate with

PML (Figures 22F-I).

PML has two potential acetylation sites, K487 and K515(12,145). To determine which

residues are deacetylated by SIRT1, we generated single and double PML mutants,

K487R, K515R, and K487/515R, in which lysine was substituted by arginine. Compared

to wild-type PML, the K487R and K487/515R mutants were barely acetylated (Figure

23A). By contrast, there was no significant change in acetylation in the K515R mutant.

We co-transfected PML (K515R) with wild-type SIRT1 or the catalytically impaired

mutant SIRT1, H363Y, and found that the acetylation level of PML (K515R) was

significant decreased by wild-type SIRT1 but not by the catalytically impaired mutant

SIRT1 (H363Y) (Figure 23B). These date indicate that lysine 487 of PML is a target for

SIRT1 deacetylation. K487 is located within a functional nuclear localization sequence

(NLS) in PML. To determine the effect of K487 on PML subcellular distribution, we

Figure 22. SIRT1 and SIRT5 deacetylate and interact with PML. A-B, HeLa cells were transfected with HA-PML4 and Myc-SIRT1 (wild-type or H363Y mutant, A) or

FLAG-SIRT5 (wild-type or H158Y mutant, B). Whole cell extracts (WCEs) were

prepared and analyzed by immunoblotting with anti-HA and anti-Myc or anti-FLAG antibodies (upper panels). The whole cell extracts were analyzed by immunoprecipitation with anti-HA antibody followed by immunoblotting with anti-acetyl-lysine and anti-HA or anti-FLAG antibodies (lower panels). C-D, HeLa cells stably expressing indicated shRNA were transfected with HA-PML4. WECs were analyzed by immunoblotting with indicated antibodies (upper panels) and by immunoprecipitation with anti-HA antibody followed by immunoblotting with anti-acetyl-lysine and anti-HA antibodies (lower panels). E, WECs of HeLa cells stably expressing indicated shRNAs were analyzed by immunoblotting with indicated antibodies (upper panels) and by immunoprecipitation with anti-PML antibody followed by immunoblotting with anti-acetyl-lysine and anti-

PML antibodies. F and H, HeLa cells stably expressing SIRT1 (F) or SIRT5 (H) shRNA were grown, harvested and analyzed by immunoblotting with indicated antibodies (upper panels) and by immunoprecipitation with indicated antibodies followed by immunoblotting with indicated antibodies. G and I, HeLa cells were transfected with

HA-PML4 and with or without FLAG-SIRT1 (G) or FLAG-SIRT5 (I). Whole cell extracts were prepared and immunoprecipitated with anti-FLAG antibodies followed by immunoblotting with indicated antibodies.

Figure 23. PML K487 is the major acetylation site and is critical for nuclear

localization of PML in HeLa cells. A, HeLa cells were transfected with HA-PML4

(wild-type, K487R, K515R, and K487/515R) and WCEs were analyzed by

immunoprecipitation with an anti-HA antibody followed by immunoblotting with anti- acetyl-lysine and anti-HA antibodies. B, HeLa cells were transfected with HA-PML4

(K515R) and Myc-SIRT1 (wild-type or H363Y mutant). WCEs were prepared and analyzed by immunoblotting with anti-HA and anti-Myc antibodies (upper panels). The whole cell extracts were analyzed by immunoprecipitation with anti-HA antibody followed by immunoblotting with anti-acetyl-lysine and anti-HA antibodies (lower panels). C, HeLa cells were transfected with HA-PML4 (wild-type, K487R, K515R, and

K487/515R mutants). Cells were immunostained with anti-HA antibody and images were taken by fluorescence microscope. DAPI (4, 6-diamidino-2-phenylindole) was used to indicate nuclei.

Figure 24. SIRT1 and SIRT5 deacetylate PML isoforms. A-E, HeLa cells were

transfected with indicated HA-PML isoforms and FLAG-SIRT1 or FLAG-SIRT5. Whole

cell extracts (WCEs) were prepared and analyzed by immunoblotting with anti-FLAG

antibody. The whole cell extracts were analyzed by immunoprecipitation with anti-HA

antibody followed by immunoblotting with anti-acetyl-lysine and anti-HA. F, HCT116 p53-/- cells were transiently transfected with HA-PML4 and FLAG-SIRT1 or FLAG-

SIRT5. HA-PML4 acetylation was analyzed similarly to that in (A-E). G, HCT116 p53-/-

cells stably expressing indicated shRNA were transfected with HA-PML4. WECs were

analyzed by immunoblotting with indicated antibodies and by immunoprecipitation with

anti-HA antibody followed by immunoblotting with anti-acetyl-lysine and anti-HA antibodies.

transfected HeLa cells with wild-type, K487R, K515R and K487/515R PML, and visualized PML subcellular distribution by immunofluorescence microscopy. We found that PML4 (K487R) and (K487/515R) mutants were mostly located in the cytoplasm

(Figure 23C). To determine whether the cytoplasmic localization of PML4 (K487R) is isoform-specific, we introduced K487R and K515R mutations into two other commonly studied PML isoforms, PML1 and PML6. Similar to PML4 (K487R), PML1 (K487R)

and PML6 (K487R) showed exclusive cytoplasmic localization (Figure 25). These results

indicate that K487 is an important acetylation site in PML, which can be targeted by

SIRT1, and is essential for nuclear localization of PML.

Crosstalk between K487 acetylation and K490 SUMOylation

SUMOylation of PML primarily occurs at K65, K160 and K490 (115). To examine the

SUMOylation status of each site, we constructed HA-tagged PML4 mutants in which

only a single lysine is available for SUMOylation, namely K65/160R, K65/490R, and

K160/490R. In order to study protein SUMOylation, N-ethylmaleimide (NEM) was added in the lysis buffer to inhibit SUMO peptidase activity. We found that the SUMO1 conjugation at K490 was significantly increased in mutant PML (K65 and K160) (Figure

26A lane1 vs lane 2), while SUMO1 conjugation at K65 and K160 was almost undetectable. Our observation that K487 is acetylated raised the possibility that acetylation at lysine 487 affects SUMO1 conjugation at K490 or vice versa. Because

PML (K487R) is exclusively cytoplasmic, we first rescued the nuclear localization of

PML (K487R) by adding a NLS sequence derived from SV40 T antigen at the N-

terminus of wild-type and K487R mutant. As expected, PML (NLS-K487R) forms

similar PML-NBs as the wild-type protein (Figure 26B), and exhibits no detectable

Figure 25. K487 is crucial for nuclear localization of PML. HeLa cells were transfected with HA-PML1 (A) or PML6 (B) (wild-type, K487R, K515R, and

K487/515R mutants). Cells were immunostained with anti-HA antibody and DAPI.

Images were taken by fluorescence microscope.

acetylation (Figure 26C). Interestingly, we found that mutant PML (K487R) is devoid of

SUMO1 conjugation, but NLS-K487R is SUMO1 conjugated (Figure 26D). Taken together, we conclude that K490 is an important SUMOylation site and that SUMO1 conjugation occurs in the nucleus.To further dissect the crosstalk between K487 acetylation and K490 SUMO1 conjugation, we focused on nuclear PML because SUMO1 conjugation at K490 only occurs in the nucleus (Figure 26D). We generated constitutive nuclear PML by adding an NLS at the N-terminus. We co-transfected PML (NLS-HA-

K65/160R) and FLAG-SUMO1 with empty vector, wild-type Myc-SIRT1, or the catalytically impaired mutant Myc-SIRT1 (H363Y) into HeLa cells. As shown in Figure

27A, expression of wild-type SIRT1 abolished acetylation, but increased SUMO1

conjugation of PML (NLS-HA-K65/160R) (lane 2 vs lane 1). In contrast, co-transfected

Myc-SIRT1 (H363Y) slightly increased acetylation, but decreased SUMO1 conjugation of PML (NLS-HA-K65/160R) (lane 3 vs lane 1), possibly due to its dominant negative

effect. Furthermore, we examined endogenous SUMO1 modification on PML (NLS-HA-

K65/160R) co-transfected with empty vector, Myc-SIRT1, or Myc-SIRT1 (H363Y), and

obtained similar results (Figure 27B). These data suggest that deacetylation of PML

K487 by SIRT1 increases SUMO1 modification on PML K490.

We further examined whether SIRT1 plays a role in the crosstalk between K487

acetylation and K490 SUMO1 conjugation. As shown in Figure 5C, we found that no

significant differences of acetylation or SUMO1 modification of PML (NLS-K65/160R)

in control and SIRT1 knockdown cells. We next asked whether this lack of regulation of

PML acetylation and SUMOylation by SIRT1 was due to different subcellular

Figure 26. Accumulation of PML SUMO1 conjugation in the nucleus. A, K490 is the

major SUMOylated site of PML protein in HeLa cells. HeLa cells were transfected with

HA-PML4 (wild-type, K65/160R, K65/490R, K160/490R, and 3KR mutants) and the

resulting WCEs were immunoprecipitated with anti-HA antibody conjugated beads

followed by immunoblotting with anti-acetyl-lysine and anti-HA antibodies. B, The SV-

40 nuclear localization sequence was introduced into HA-PML4 constructs (wild-type

and K487R mutant) (upper panel). HeLa cells were transfected with NLS-HA-PML4

(wild-type and K487R mutant) and immunostained with anti-HA antibody (lower panels).

C, HeLa cells were transfected with HA-PML4 (wild-type and K487R mutant), and NLS-

HA-PML4 (wild-type and K487R mutant), WCEs prepared and immunoprecipitated with anti-HA antibody followed by immunoblotting with anti-acetyl-lysine and anti-HA antibodies. Note that NLS-HA-PML4 proteins migrate slower than HA- PML proteins due to the addition of NLS at the N-terminus. D, SUMO1 modification on PML occurs in the nucleus. HeLA cells were transfected with NLS-PML4, PML4 (487R), or NLS-PML4

(K487R). Note that 20 µM of NEM was added in the lysis buffer in panel B. The resulting WCEs were immunoprecipitated with anti-HA antibody followed by immunoblotting with anti-acetyl-lysine and anti-HA antibodies.

Figure 27. K487 acetylation and K490 SUMOylation represses to each other. A and

B, Wild-type SIRT1 but not a SIRT1catalytic defective mutant (H363Y) deacetylates

NLS-HA-PML (K65/160R) and enhances SUMOylation at K490. HeLa cells were co-

transfected with NLS-HA-PML (K65/160R) and FLAG-SUMO1 (A), and empty vector,

or wild-type SIRT1, or SIRT1 H363Y. After 48h, the cells were harvested. The WCEs

were immunoprecipitated anti-HA antibody conjugated beads followed by

immunoblotting with anti-FLAG (A) or anti-SUMO1 (B), anti-acetyl-lysine and anti-HA

antibodies. C-E, The effect of SIRT1 knockdown on the SUMOylation of PML protein

at K490 with (D and E) or without (C) H2O2 treatment. HeLa cells were transfected with

the indicated plasmids and the resulting WCEs were immunoprecipitated with anti-HA

antibody conjugated beads followed by immunoblotting with anti-FLAG (C and D) or

anti-SUMO1 (E) and anti-HA antibodies. F-G, The effect of K487 mutation on the

SUMOylation at K490.HeLa cells were transfected with the indicated PML mutation with (F) or without (G) FLAG-SUMO1. The resulting WCEs were immunoprecipitated with anti-HA antibody conjugated beads followed by immunoblotting with anti-FLAG (F) or anti-SUMO1 (G) and anti-HA antibodies. H, The effect of PML SUMOylation site mutation on its acetylation status. HeLa cells were transfected with indicated HA-PML4 mutants and the resulted whole cell extracts were immunoprecipitated with anti-HA

antibody conjugated beads followed by immunoblotting with anti-acetyl-lysine and anti-

HA antibodies. Note that 20 µM of NEM was added in the lysis buffer except in panel H.

S.E., shorter exposure; L.E., longer exposure.

localizations of SIRT1 and PML (NLS-K65/160R) in HeLa cells. SIRT1 is primarily localized in the cytoplasm of HeLa cells, while PML (NLS-K65/160R) is exclusively nuclear. It has been reported that H2O2 induces nuclear translocation of SIRT1(226).

Indeed, we observed an increase in acetylation and a decrease in SUMO1 conjugation of

PML (NLS-K65/160R) in SIRT1 knockdown HeLa cells upon H2O2 treatment (Figures

27D and E).

Lysine acetylation functions by generating a site for specific recognition by cellular

factors or by neutralizing positive charges. The lysine-to-arginine (K/R) substitution

prevents acetylation but maintains the same positive charge, thus mimicking the non-

acetylated form. In contrast, lysine-to-glutamine (K/Q) substitutions mimic the

constitutively acetylated form through neutralization of positive charge(228,229). To elucidate the mechanism by which K487 acetylation inhibits K490 SUMO1 conjugation,

HeLa cells were transfected with HA tagged PML (NLS-K65/160R), PML (NLS-

K65/160/487R), PML (NLS-K65/160R/K487Q), or PML (K65/160/490R). The latter was used as a negative control for SUMOylation, with (Figure 27F) or without (Figure

27G) co-expression of FLAG-SUMO1. Compared to PML (NLS-K65/160R), there was no significant change in SUMO1 modification of PML (NLS-K65/160/487R) (Figure

27F and 27G, lane 2 vs lane 1). However, we did observe a decrease in SUMO1 modification of PML (NLS-K65/160R/K487Q) (Figure 27F and 27G, lane 3 vs lane 1).

These results suggest that neutralization of the positively charged lysine by the negatively charged glutamine at 487 decreased SUMO1 conjugation at K490.

We also examined whether K490 SUMO modification affects K487 acetylation. We transfected HeLa cells with the HA-tagged PML4 mutants, K65/160R, K65/490R, and

K160/490R, in which only single lysine is available for SUMO conjugation. We found

that mutants bearing K490R, which abolishes SUMO1 modification, exhibited increased

acetylation (Figure 26H, lanes 3-5). These data suggest that the K490 SUMO1

modification prevents K487 from acetylation. In sum, these results indicate that K487

acetylation and K490 SUMOylation mutually inhibit each other.

Depletion of SIRT1 attenuates H2O2-induced accumulation of PML-NBs and cell death

To test whether H2O2 promotes nuclear accumulation of SIRT1, we performed a subcellular fractionation experiment. As expected, western blot analysis of the whole cell extracts and nuclear extracts revealed that H2O2 treatment led to an accumulation of

nuclear SIRT1 in a dose-dependent manner but did not affect total SIRT1 protein level

(Figure 28A). Because H2O2 promotes nuclear accumulation of both SIRT1 and PML, we

sought to determine whether SIRT1 is required for H2O2-induced nuclear accumulation of PML. To test this hypothesis, we treated control and SIRT1 knockdown HeLa cell lines with H2O2. Notably, knockdown of SIRT1 attenuated H2O2-induced accumulation

of nuclear PML (Figure 28B, lane 7 vs lane 8 and Figure 29A). We further confirmed this

result by immunofluorescence microscopy. As shown in Figure 28C, H2O2 stimulates

accumulation of PML-NBs in control HeLa cells in a dose-dependent manner. However,

this accumulation was abolished in SIRT1 knockdown cells. Expression of shRNA- resistant SIRT1 rescued nuclear PML accumulation (Figure 28D), although not completely. These data indicate that SIRT1 in required for H2O2 -induced nuclear PML

accumulation.

Similarly, knockdown of another PML deacetylase, SIRT5, also abolished the

accumulation of nuclear PML in response to H2O2 (Figure 28E lane 10 vs lane 11 and

Figure 29B-29C). Double knockdown SIRT1 and SIRT5 further decreased the

accumulation of nuclear PML under H2O2 treatment (Figure 28E lane 10 vs lane 12).

Because both SIRT1 and SIRT5 are both PML deacetylases, we hypothesize that

overexpression of SIRT5 may rescue PML-NBs in SIRT1 knockdown cells in response to

H2O2 treatment. As shown in Figure 28F, acetylation of PML increased in SIRT1

knockdown cells (lane2 vs lane 1), and was significantly decreased by overexpression of

wild-type SIRT5 but not a SIRT5 catalytic impaired mutant (lane 3 vs lane 2 and lane 4

vs lane 2). Indeed, overexpression of wild-type SIRT5 but not the SIRT5 catalytic

impaired mutant rescued PML-NBs formation in SIRT1 knockdown cells in response to

H2O2 treatment (Figure 28G and 28H).

Evidence, including ours, has shown that down-regulation of PML reduces sensitivity to

H2O2 -induced cell death (56,140). Because knockdown of SIRT1 reduces the

accumulation of PML-NBs, we speculated that SIRT1 knockdown HeLa cells will be

resistant to H2O2-induced cell death. To test this, we measured cell viability of control

and SIRT1 knockdown HeLa cells in response to H2O2 (Figures 30A). We found that

H2O2 treatment led to significant cell death of control cells but not SIRT1 knockdown

cells (Figure 30A, lane 4 vs lane 1). Furthermore, ectopically expressed wild-type PML,

but not PML (K487R) were capable of partially rescuing the sensitivity of HeLa cells to

H2O2 -induced cell death (Figure 30A lane 5 vs lane 4 and lane 6 vs lane 4). We further verified the above data by colony formation assays and obtained similar results (Figure

30B and 30C). However, expression of PML (K487R) only slightly alleviated H2O2-

100

101

Figure 28. Deacetylation of PML is required for H2O2-induced accumulation of PML-

NBs. A, H2O2 induces SIRT1 nuclear accumulation. HeLa cells were treated with or

without 4 or 8 mM of H2O2 for 0.5 h. Nuclear and cytoplasmic fractions were prepared

and subjected to Western blotting. B, HeLa cells stably expressing shCtrl and shSIRT1

were treated with H2O2 for 1 h. Nuclear and cytoplasmic fractions were prepared and

subjected to Western blotting. S.E., shorter exposure; L.E., longer exposure. C, HeLa

cells expressing shCtrl and shSIRT1 stably were treated with H2O2 for 1 h followed by immunofluorescence microscopy probed with anti-PML (Original magnification 200×).

D, HeLa cells stably expressing the indicated shRNAs were stably transfected shRNA resistant SIRT1. Nuclear fractions were prepared and subjected to Western blotting. E,

HeLa cells stably expressing the indicated shRNAs were treated with H2O2 for 1 h.

Nuclear and cytoplasmic fractions were prepared and subjected to Western blotting. F,

HeLa cells expressing shCtrl or shSIRT1 stably were transfected with HA-PML4 and

Myc-SIRT5 (wild-type or H158Y mutant). Whole cell extracts were prepared and

immunoprecipitated with anti-HA antibody. WCEs and immunopellets were subjected to

immunoblotting with anti-Myc, anti-acetyl-lysine and anti-HA antibodies. G, HeLa cells stably expressing the indicated SIRT1 shRNA were stably transfected wide type or mutant SIRT5. Nuclear fractions were prepared and analyzed as in (D). H, HeLa cells

stably expressing shCtrl or shSIRT1 were transfected with Myc-SIRT5 (wild-type or

H158Y mutant). Cells were immunostained with anti-PML and anti-Myc antibodies and images were taken by fluorescence microscope.

102

Figure 29. Knockdown SIRT1 or SIRT5 attenuates nuclear PML accumulation in

-/- response to H2O2 treatment. HeLa cells (A) or HCT116 p53 cells (B-C) stably expressing the indicated shRNA were treated with H2O2 for 1 h. Nuclear fractions were prepared and subjected to Western blotting with indicated antibodies.

103

mediated inhibition of colony formation (Figure 30C, lane 7 vs lane 5). Notably, restoration of an NLS on K487R (NLS-K487R) rescued the H2O2 sensitivity similarly to

the wild-type protein. Together, these data demonstrated that loss of nuclear PML in

SIRT1 knockdown HeLa cells is responsible for resistance to H2O2-induced cell death.

Both SIRT1 and SIRT5 are capable of deacetylating PML. We therefore asked whether

overexpression of SIRT5 is capable of rescuing sensitivity of H2O2-induced cell death in

SIRT1 knockdown cells. Indeed, overexpression of wild-type SIRT5, but not a catalytic

defective mutant, partially restored the sensitivity of SIRT1 knockdown HeLa cells to

H2O2 treatment (Figure 7D, lane 1-4). However, this rescue effect of overexpression of

SIRT5 was largely dependent on the presence of of PML (Figure 7D, lane 5-8). As a

control for the experiment, the expression levels of PML, SIRT1 and SIRT5 proteins

were examined by western blotting (Figure 7D). In summary, these data strongly suggest

that resistant to H2O2 treatment in SIRT1 depleted cells is due in part to change

acetylation status of PML.

104

105

Figure 30. The role of PML in SIRT1-mediated, H2O2-induced cell death. A, HeLa

cells stably expressing shCtrl or shSIRT1 were transfected with empty vector, or wild-

type, or K487R mutated PML4. After 24 h of transfection, 1 × 104 cells were seeded into

a 96-well plate. After 12 hours, the cells were treated with 8 mM H2O2 for 1 h, washed

and incubated with fresh media for another 8 hours. Cell number was determined by a

Cyquant cell proliferation assay kit according to the manufacturer’s instructions. Cell

viability was calculated as the number of H2O2 treated cells divided by the number of

non-H2O2 treated cells in five independent experiments. Data are displayed as the mean±

SD. Unpaired two-tail t-tests (*, p<0.05, **, p<0.01 and ***, p<0.001) were used for

statistical analyses. An aliquot of cells was also used to prepare whole cell lysates to

example expression levels of endogenous SIRT1 and exogenous HA-PML4 (left panel).

B, 2 × 103 cells of the treated cells in (A) were seeded into a 100 mm tissue culture plate.

After 12 h, the cells were treated with 8 mM H2O2 for 1 h, washed and incubated with fresh media. 9 days later, when macroscopic colonies became detectable, the cells were washed with PBS, stained with crystal violet (B) and the number of colonies scored (C).

D, HeLa cells stably expressing shCtrl or shSIRT1 were transfected with control or PML siRNA, and 24 h later were transfected with wild-type or H158Y mutant SIRT5. After 24 h of transfection, 104 cells were seeded into a 96-well plate. The cells were treated and measured same as indicated in (A).

106

DISCUSSION

In response to diverse extracellular stimuli, PML protein is post-translationally modified

to control its activity, stability and sub-cellular localization. In the present study, we

found that the acetylation of PML was down-regulated in response to H2O2 treatment, and identified SIRT1 and SIRT5 as deacetylases that are capable of deacetylating PML at

K487. When lysine 487 is substituted by arginine (K487R), PML-NBs are restricted to the cytoplasm (Figure 3C), an observation similar to previous reports (93,94). The oncogenic function of cytoplasmic PML was reported to correlate with redistributing nuclear PML to the cytoplasm, thus reducing the number of PML-NBs and inhibiting cell apoptosis, and promoting proliferation(92,94,145,230). Similarly, we found that SIRT1 knockdown HeLa cells, where the numbers of PML-NBs are markedly down-regulated, are resistant to H2O2-induced cell death. Ectopic overexpression of wild-type PML,

which forms PML-NBs, significantly rescued the resistance to H2O2-induced cell death in

SIRT1 knock-down HeLa cells, but not the cytoplasmic mutant PML (K487R) (Figure

30A-C). Consistently, wild-type SIRT5, which deacetylates PML in SIRT1 knockdown

HeLa cells and promotes accumulation of PML-NBs, also partially rescued the resistance

to H2O2-induced cell death of SIRT1 knockdown cells (Figure 28F and Figure 30D).

Our work also highlights a novel post-translational crosstalk between PML acetylation at

K487 and SUMO1 conjugation at K490 (Figure 30E). The interplay of different PTMs has emerged as key mechanism for dynamic control of cellular signaling(202).

Acetylation neutralizes the positive charge of lysine, which may disrupt protein

interaction(231), including enzymes that affect modification of neighboring sites, and

NLS binding to the translocation machinery. Several lines of evidence suggest that K487

107

acetylation blocks K490 SUMO1 conjugation. First, overexpression of SIRT1, the

deacetylase for PML K487, enhanced SUMO1 modification at K490 (Figure 27A-B),

whereas knockdown of SIRT1 reduced it. We also noted that the sizes of SUMO1

modified PML and acetylated PML are different (Figure 27A-B). Furthermore, SUMO1

conjugation at K490 in PML (NLS-K65/160R) is reduced in K487Q mutant, which is a

negatively charged amino acid that mimics the acetylated form (Figure 27C-G). Taken

together, these observations indicated that K487 acetylation and K490 SUMO1

conjugation are mutually exclusive. Acetylation blocking SUMOlyation is not due to

same site exclusion, although K487 and K490 are close to each other. Perhaps proximity

of the two modification sites is important. Interestingly, SUMOylation of K490 also

inhibits PML acetylation, presumably at K487 (Figure 27H). SUMOylation of

histone/non-histone proteins can lead to the recruitment of HDACs (232,233). However,

we did not observe significant differences between the interactions of GST-SIRT1/5 with

non-SUMO1-conjugated and SUMO1-conjugated PML (data not shown). Our findings

raise the possibility that whether such regulation is unique to PML. To address this issue,

we performed sequence analysis of known acetylated peptides (203). Combining the known acetylated peptide sequences and the well-established SUMO conjugation motif,

KXE, we hypothesize that a putative sequence motif, KXXKXE, may be regulated by the crosstalk described in this study (Figure 31A). For example, the nuclear receptor coactivator 2 (NCOA2/GRIP1) has been reported to be acetylated at K785(203) and

SUMO1 conjugated at K788(234). We have also identified several proteins that are acetylated within this putative motif. It would be interesting to see whether the second

108

lysine in this motif is SUMOylated and whether there is a similar crosstalk between

acetylation and SUMOylation in these proteins.

Since the identification of CPS1 as a SIRT5 deacetylation target from mouse

mitochondria matrix lysates(223), most studies have been focused on the function of

SIRT5 in mitochondria. From our biochemical deacetylase screen, we functionally

identified, PML, as the first non-mitochondria protein substrate for SIRT5 deacetylation.

One recent study also showed that SIRT5 is present extramitochondrially(235), which supports our observation that SIRT5 functionally regulates non-mitochondrial substrates such as PML. Consistent with these observations, a significant fraction of exogenously expressed SIRT5 is localized in the nucleus (Figure 28G-H). Although SIRT1 and SIRT5 have distinct substrate specificities for p53-related substrates and CPS1, both can deacetylate PML. Indeed, overexpression of SIRT5 partially rescued PML function in

SIRT1 knockdown cells when treated with H2O2 (Figure 30D). Interestingly, both SIRT1

and SIRT5 express and behave similarly in response to caloric restriction(236) and are

significantly down-regulated in head and neck squamous cell carcinomas(224),

suggesting some level of common regulation and function for these two deacetylases.

We have previously shown that PML is required for H2O2-mediated cell death in MDA-

MB-231 breast cancer cells (139). In this study, we also demonstrated that SIRT1 is

essential for H2O2-induecd cell death in HeLa cervical cancer cells. Notably, H2O2

treatment promotes nuclear translocation of PML and formation of PML-NBs. While the

mechanism underlying H2O2-mediated PML nuclear translocation remains largely

unknown, this activity is essential for H2O2-induced cell death because the loss of

sensitivity to H2O2-induced cell death in SIRT1 knockdown cells can be rescued by

109

overexpression of wild-type PML or NLS-K487R, but not the constitutively cytoplasmic mutant, K487R (Figure30A). These results imply that loss of nuclear PML contributes to the insensitivity of SIRT1 knockdown cells to H2O2-mediated cell death. In summary, we have identified SIRT1 and SIRT5 as PML deacetylases and established a novel post- translational crosstalk between K487 acetylation and K490 SUMO1 modification in PML.

Finally, we describe a role of PML deacetylaiton in H2O2-induced cell death.

110

111

Figure 31. A, An alignment of a putative sequence motif that are both acetylated and SUMO1 conjugated. This sequence motif is based on this study on the crosstalk between acetylation of K487 and SUMO1 conjugation at K490 of PML, the known acetylated peptide sequences that contain KXXK(203), and the well-established

SUMOylation motif, KXE. Note that some acetylated peptides contain KXK sequence

(data not shown), so potentially KXKXE is a loosely conserved motif. B, Proposed model summarizing the findings of this study. In HeLa cervical cancer cells, PML,

SIRT1 and SIRT5 are predominantly localized in the cytoplasm, where PML is acetylated by HATs and deacetylated by SIRT1 and SIRT5. H2O2 induce nuclear translocation of PML, SIRT1 and SIRT5. Deacetylation of PML ac-K487 by SIRT1 or

SIRT5 is required for H2O2-induced SUMO1 conjugation of PML K490, formation of

PML-NBs and cell death.

112

CHAPTER 4: microRNA-24 Regulates Endothelial Cell Angiogenesis by Targeting PML Expression

ABSTRACT

The tumor suppressor promyelocytic leukemia (PML) protein potently inhibits angiogenesis in vitro, but how PML is regulated in endothelial cells is not clear. Here, we demonstrate that microRNA-24 (miR-24) and microRNA-133 (miR-133) target the 3’-

UTR of PML1 mRNA, the largest PML spliced isoform. Under normal culture conditions, miR-24 and miR-133 down-regulate PML1 protein expression in primary human endothelial cells (ECs). However, under serum starvation conditions miR-24 but not miR-133, up-regulates PML1 protein expression by enhancing PML1 mRNA translation.

We further show that miR-24 inhibits EC capillary tube formation, and that knocking down PML attenuates the inhibitory effect of miR-24 on tube formation. In addition, miR-24 inhibits outgrowth of ECs of aortic rings isolated from wild type but not from

Pml knockout mice. Our findings provide the first evidence that under serum starvation conditions miR-24 activates PML1 mRNA translation and thereby inhibits endothelial cell angiogenesis.

113

INTRODUCTION

Angiogenesis, the growth of new blood vessels from pre-existing ones, is an essential process for normal development, tissue repair, as well as variety of inflammatory and pathological conditions, including tumor growth and invasion. microRNAs (miRNA) are a class of small (~21-25 nucleotides), non-coding RNAs that are important post-

transcriptional regulators in a variety of cellular events, including cell proliferation,

organogenesis, invasion and angiogenesis during carcinogenesis (237). miRNAs are initially transcribed as longer precursors (pri-miRNAs), which are initially processed by

Drosha-DGCR8, the RNase enzyme complex, into hairpin oligos named pre-miRNAs,

and later by Dicer and TRBP to generate mature miRNAs. miRNAs are thought to bind

in the 3’ UTR of target mRNA to reduce their stability or repress their translation.

However, several studies have suggested that miRNAs can target mRNA 3’-UTR to stimulate mRNA translation. This can occur under conditions of serum starvation (238), cell post-confluence (239), and in specific cell types, notably smooth muscle cells (240)

cardiomyocytes (241) and quiescent cells (242). Occasionally, miRNAs target the 5’-

UTR of an mRNA rather than only the 3’-UTR and stimulate mRNA translation

(243,244). Recently, the ability of miRNA to activate translation was observed in mitochondria during muscle differentiation (245).

Our studies focus on miR-24 and miR-133, two well-studied miRNAs. miR-24 has dual roles in tumorigenesis. miR-24 inhibits apoptosis by targeting the mRNA encoding the tumor suppressor Bim (246), HNF4α (247), XIAP (248) and DND1 (249), and promotes tumor invasion and metastasis by targeting PTPN9 (Protein Tyrosine Phosphatase, Non-

Receptor Type 9) and PTPRF (Protein Tyrosine Phosphatase, Receptor Type, F) (250).

114

Interestingly, miR-24 was also reported to inhibit cell proliferation by targeting

oncogenes including E2F2, MYC, CCNB1 and CDC2 (251).

The role of miR-24 in endothelial cell (EC) migration and angiogenesis has also been

explored (252,253). Another miRNA, miR-133, has been reported to play a pivotal role

in cardiovascular development and cardiac pathology (237,254). However, the role of

miR-24 and miR-133 in serum starvation is largely unknown.

The promyelocytic leukemia protein (PML) was originally discovered in acute

promyelocytic leukemia (APL), a disease associated with a t(15;17) chromosomal

translocation, which generates a fusion protein between PML and retinoic acid receptor

alpha (RARα) (4,5). PML is considered to be a tumor suppressor based on its ability to

inhibit cell growth, cell migration and angiogenesis and to promote apoptosis

(2,9,19,198). Mechanistically, PML is capable of regulating transcription, translation and post-translational modification. There are six experimentally verified nuclear isoforms of

PML in human, all of which harbor the N-terminal 552 amino acids in common (18).

Although PML mRNA is expressed in most cell types and the PML gene is rarely

mutated or deleted from the genome, PML protein expression is frequently down-

regulated in human cancers, and often correlates with tumor grade and progression

(1,28,29). Most studies of regulation of PML expression focus on how post-translational

modification affects PML protein stability. We have previously reported that UHRF1

(ubiquitin-like, containing PHD and RING finger domains 1) (2), SIRT1, SIRT5 (3),

peptidyl-prolyl cis/trans isomerase 1 (Pin1) (140) and Keap1 (146) regulate PML protein

stability and activity via post-translational modification. However, whether PML mRNA

is also subjected to regulation remains largely unknown.

115

In this study, we discovered that both miR-24 and miR-133 target PML1 mRNA 3’-UTR

and reduce PML1 protein abundance in primary ECs under normal culture condition.

Unexpectedly, miR-24 but not miR-133, activates PML1 mRNA translation after starvation. Lastly, we demonstrate that miR-24 inhibits angiogenesis in vitro and ex vivo

in a PML-dependent manner.

116

MATERIALS AND METHODS

Cell lines and culture conditions

Human Umbilical Vein Endothelial Cells (HUVECs) and Human Aorta Endothelial Cells

(HAEC,) were maintained in endothelial cell growth medium-2 (EGM-2) (Lonza Inc,

Allendale, NJ, USA) supplemented with FBS (2%) and growth factors. Human

Microvascular Endothelial Cells (HMVECs) were purchased from Lonza and grown in

endothelial cell basal medium (EBM-2) (Lonza Inc, Allendale, NJ, USA) supplemented

with FBS (2%) and growth factors. For serum starvation condition, HUVECs were

cultured in EGM-2 media with 0.1% FBS. If not specific indicated, passage 4 endothelial

cells were used in this study.

Animals

Pml+/+ and Pml-/- mice were maintained in the C57BL/6 genetic background in the Health

Science Animal Facility of Case Western Reserve University. Animal experimentation

was performed under the approved protocols of the Case Western Reserve University

Animal Care and Use Committee following National Institutes of Health guidelines.

Antibodies, siRNAs and miRNAs

The commercial antibodies used in this study were: anti-β-actin (Sigma, A5441), anti-

mouse PML (Chemicon, MAB3738), anti-human FXR1 (Santa Cruz, sc-10554), anti-

mouse IgG conjugated with HRP (Santa Cruz, sc-2005), anti-rabbit IgG conjugated with

HRP (Millipore, 12-348). Anti-PML1 and anti-panPML rabbit polyclonal antibodies were purified in-house. The antigen sequences used for generating antibodies against

PML1 and panPML were N-LRVLDENLADPQAEDRPLVF-C and N-

EARLARSSPEQPRPS-C, respectively. The specificity of anti-PML antibodies was

117

validated previously in HUVECs (Figure 2A) and in HeLa cells (2,3). Non-targeting siRNA control (siCtrl, D-001810-01), PML siRNA (J-006547-03 and J-006547-05), miRNA mimic negative control (miR-Ctrl, CN-001000-01), hsa-miR-24-3p mimic (c-

300497-03), hsa-miR-133a-3p mimic (c-300600-05) and transfection reagent

DharmaFECT1 (T-2001) were purchased from Thermo Scientific (Rockford, IL, USA).

Human miRNA antagonist negative control (HSTUD001) and hsa-miR-24-3p antagonist

(HSTUD0410) were purchase from Sigma (St Louis, MO, USA). shRNA knockdown and lentiviral infection

Non-targeting control (SHC002) and FXR1 (TRCN0000160812, TRCN0000160901 and

TRCN0000412665) shRNA expression plasmids were purchased from Sigma. Lentiviral particles were packaged in HEK293T cells following cotransfection of shRNA vector, psPAX2, and pMGM2 according to a published protocol (255). After incubation with lentiviral particles overnight, the medium was replaced and ECs were selected in the presence of puromycin (0.5 µg/ml).

Western blot analysis

ECs were washed with 1 × phosphate-buffered saline (PBS) and resuspended in RIPA buffer (1 × PBS, 1% NP-40, 0.5% sodium deoxycholate, and 0.1% SDS) containing a protease inhibitor cocktail and a phosphatase inhibitor cocktail (Roche Applied Science).

Lysed cells were centrifuged at 4 ̊C at 12,000 rpm for 10 min and the supernatant was added with 2 × sample buffer followed by SDS-PAGE and Western blotting as previously described (3).

Polyribosome analysis and PML mRNA distribution

118

1 × 107 HUVECs were seeded in 150-mm dishes and transfected with miR-control (miR-

Ctrl) or miR-24 mimic (miR-24). After 48 hours of serum starvation, HUVECs were

treated with 100 µg/ml cycloheximide for 15 min at 37 . Plates were transferred to ice

and cells were rinsed twice in ice cold PBS containing℃ 100 µg/ml cycloheximide. All

traces of PBS were removed by aspiration and 1 mL of lysis buffer (20 mM HEPES-

KOH, pH 7.4, 15 mM MgCl2, 200 mM KCl, 1% Triton X-100 (v/v), 100 μg/mL cycloheximide, 2 mM DTT, 1 mg/mL heparin, 200 unit/ml RNase inhibitor (RNaseOUT, invitrogen) and EDTA-free protease inhibitor (Roche Applied Science)) was added followed by cell scraping and transfer to a pre-chilled 1.5 mL tube. Cells were lysed by

passage through a 27 gauge pre-chilled syringe needle twice. Lysates were spun at 14,000

× g for 5 min in a refrigerated microcentrifuge. After determining the concentration of nucleic acid in the lysate, 20 OD260 units of the lysate were gently layered on the top of

15 - 45% cold sucrose gradients in buffer (50 mM Tris-acetate pH 7.0, 50 mM NH4Cl, 12

mM MgCl2, 1 mM DTT). Ribosomes and polysomes in the lysate were separated by

centrifuging at 41,000 rpm for 2 h and 26 min at 4°C in a Sw41Ti rotor. After centrifuge,

14 fractions (0.6 ml/fraction) were collected using an ISCO density fractionator. Two

adjacent fractions were combined for RNA extraction using TRIzol LS reagent

(Invitrogen). The relative quantities of total and PML1 mRNA were determined by

Northern blotting and quantitative RT-PCR (RT-qPCR). For the latter, an equal volume of RNA from each of the fractions was used for cDNA synthesis.

Northern analysis

Total RNA isolated from polysome fractions was suspended in a final concentration of 1%

SDS and 50 µg/mL proteinase K and stored overnight at -80 ̊C. Fractions were then

119

extracted once with phenol/LET (25 mM Tris pH 8.0, 100 mM LiCl, 20 mM EDTA),

once with phenol/chloroform/LET, and then precipitated with 95% ethanol. RNA pellets

were recovered by centrifugation at 14,000 rpm for 30 min. Pellets were washed once

with 700 µL 75% ethanol, air dried, and resuspended in 1 × sample buffer (200 mM

MOPS pH = 7.0, 50 mM sodium acetate, 12.5 mM EDTA, 3.33% formaldehyde, 0.4

mg/mL ethidium bromide). Samples were then heated to 65°C for 10 min to denature

RNA and loaded on 1.4% agarose-formaldehyde gels, and transferred to a HyBond N+

membrane (Amersham) overnight at room temperature. Membranes were pre-blocked in

hybridization buffer (6 × SSC, 10 × Denhardt’s solution, 0.1% SDS) and then probed

with 32P end-labeled deoxyoligonucletides (5’-GCCGACTTCTGGTGCTTTG-3’) at

42°C. The oligo probes sequence is complementary to PML mRNA (nt. 406->424 of

PML mRNA). Blots were washed twice with 6 × SSC/0.1%SDS for 15 min at room

temperature then once at 50°C in 6 × SSC/0.1%SDS for 15 min. Blots were exposed to a phosphorimager screen and scanned with a Typhoon imager and quantified with Image J.

RNA isolation and quantitative RT-PCR (RT-qPCR)

Total cellular RNA was isolated using a PrepEase RNA Spin kit (Affymetrix, Santa Clara,

CA, USA). Complementary DNAs (cDNAs) were generated using iScript Reverse

Transcription Supermix kit (Bio-Rad, Richmond, CA, USA). The specific cDNAs of interest were amplified and quantified by real-time PCR using an iCycler (Bio-Rad) platform with 2 × IQ SYBR Green Supermix (Bio-Rad) and appropriate primers. Relative

RNA levels were normalized to the β-actin mRNA level. Primer sequences are shown below: Pml forward, 5`-GCCGACTTCTGGTGCTTTG-3` and Pml reverse, `-

GTTGTTGGTCTTGCGGGTG-3`; Pml1 3’UTR forward, 5`-

120

GGAATTCCGTAGGGTCTTGTT-3` and Pml1 3’UTR reverse 5`-

CTCCTTCCATCTGCATCAATCT-3`; Notch4 forward, 5`-

CTGCCAGACTCTGATGGACTTA-3` and Notch4 reverse 5`-

TGTGGCAAAGGGAAGAGACG-3`; Ccl2 forward, 5`-

AGCAGCAAGTGTCCCAAAGA-3` and Ccl2 reverse, 5`-

TTGGGTTTGCTTGTCCAGGT-3`; Serprine1 forward, 5`-

TGGCACGGTGGCCTCCTCAT-3` and Serprine1 reverse, 5`-

ACTGTTCCTGTGGGGTTGTGCC-3`; Eaps1 forward, 5`-

TACACAGGTGGAGCTAACAGGA-3` and Eaps1 reverse, 5`-

AAGAAGTCCCGCTCTGTGGA-3`; β`-AAGA forward, 5`-

GCGCGGCTACAGCTTCA-3` and β-actin reverse 5`-

CTTAATGTCACGCACGATTTCC3`.

Luciferase Reporter Assays

The full-length 3’-UTR of human PML1 was purchased from SwitchGear Genomics

(Carlsbad, CA, USA) and subcloned into the pmiRGLO vector (Promega, Madison, WI,

USA). The extra sequence of the pmiRGLO construct between the firefly luciferase stop

codon and the 3’-UTR of PML1 was removed by site-specific PCR mutagenesis and

verified by sequencing. PML1 3’-UTR reporters were generated from the pmiRGLO-

PML1 3’-UTR construct described above. HUVECs were transfected with the wild type

or the indicated site-specific mutants and miR-Ctrl or miR-24. Two days after

transfection, cells with or without serum starvation were harvested and luciferase activity

determined using the Dual-GloTM Luciferase Assay system (Promega).

In vitro capillary tube formation assay

121

The in vitro capillary tube formation assay was performed as described previously using a kit from Millipore (ECM, Billerica, MA, USA). The gel matrix was prepared as indicated by the manufacturer and allowed to solidify in a 96-well plate. Forty-three

hours after transfection with the indicated miRNA with serum starvation, HUVECs were

typsinized, and 5 X 104 cells seeded onto the surface of the polymerized matrix. For PML

knockdown cells, HUVECs were transfected with the indicated siRNA and one day later

the cells were transfected with the indicated miRNA and serum starved. The plate was

incubated at 5% CO2 for 4 hours, and images were captured using a Leica Wetzlar

microscope. The branch points were counted and presented as the mean±s.d. Two-tail t-

tests were used to determine statistical significance.

Aortic ring angiogenesis assay

Ex vivo angiogenesis assay was performed by culturing 1-mm aortic rings in Matrigel

(BD Biosciences) as previously described with minor modification (256). Briefly, thoracic aortas were removed from euthanized wild-type or Pml knockout mice and gently stripped of periaortic fibroadipose tissue. Aortas were sectioned into 1-mm length rings ( 12 per aorta), and embedded in Matrigel. Gels containing the aortic rings were polymerized∼ in 6-well plates and incubated at 37 °C. Endothelial Cell Growth Medium-2

(EGM-2) supplemented with FBS (2%) and growth factors (Lonza), 100 units/ml penicillin, and 100 ng/ml streptomycin were added to the Matrigel-containing explants. miR-Ctrl or miR-24 was transfected by adding them to the growth medium to a final concentration at 80 nM. ECs sprouting and neovessel formation were assessed every 2 days through day 6 using a Leica Wetzlar microscope with bright-field optics. Aortic ring

122

outgrowth was quantified by counting neovessel numbers within the corresponding

regions of interest and mean number of sprouts per ring±s.d.

Statistical Analysis

Statistical analysis was performed using two-tailed Student’s t test, using p < 0.05 as a criterion of significance.

123

RESULTS

miR-24 and miR-133 target PML1 mRNA, a major PML spliced isoform in

HUVECs

The PML gene contains nine exons and the primary transcript is subject to extensive alternative splicing which generates six major nuclear isoforms (Figure 32A). To determine whether PML1 is the major isoform in HUVECs, we carried out immunoblotting using an antibody that recognizes all PML nuclear isoforms (panPML) and an antibody that specifically recognizes the PML1 (97.6 kDa) and PML4 (70.0 kDa) proteins (named PML1 Ab). Using a knockdown approach, we concluded that the two upper bands in Figure 32A, migrating close to 100 and 130 kDa, are PML1 protein, one of which is likely to be post-translationally modified. A very highly form of modified

PML4 is also possible but less likely as no siRNA-sensitive 70 KDa band was detected

(Figure 32B). The bands migrating below 75 kDa are non-specific signals because their

abundance was not proportionally reduced as were the 100 and 130 kDa bands, in PML knockdown samples (132).

PML1 mRNA contains a long (2811 bp) 3’-UTR that is absent in the other PML isoforms.

This suggests that PML1 may be specifically regulated by miRNAs, which primarily target the 3’-UTR. To search for miRNAs that may target PML1 mRNA, we employed a computational approach which is based on conserved sequences across multiple species of potential miRNA binding sites in the 3’-UTR of PML1 mRNA, and an analysis of miRNAs with known biological function. Through this search, we identified 2 miRNAs, miR-24 and miR-133, that potentially target PML1 3’-UTR (Figure 32C). To test whether miR-24 and miR-133 could affect PML1 protein expression, HUVECs were transfected

124

with a control miRNA mimic (miR-Ctrl), a miR-24 mimic or a miR-133 mimic. We found that cells transfected with miR-24 or miR-133 express lower levels of PML1 protein than cells transfected with miR-Ctrl (Figure 32D).

125

126

Figure 32. miR-24 and miR-133 target PML1 mRNA, a major PML mRNA isoform

in HUVECs. A, A Schematic representation of PML nuclear isoforms. All PML mRNA

isoforms contain exons 1-6 encoding a common N-terminal 552 amino acids which

harbors a functional nuclear localization sequence. Additionally, they also contain

distinct 3’-UTRs (not shown) that are derived from alternative splicing. The exon

junctions of the PML gene are indicated. PML1 mRNA contains a long 3’-UTR including

exon 9, part of which encodes distinct C-terminal amino acid sequences of PML1 protein.

B, PML1 is the major isoform in HUVECs. Whole cell lysates prepared from HUVECs

transfected with a siCtrl or two different common PML siRNA (siP-1 and siP-2) were

analyzed by immunoblotting with anti-panPML, anti-PML1 and anti-β-actin antibodies.

β-actin served as an internal loading control. Lanes 1 and 4, lanes 2 and 5 and lanes 3 and

6 are from same cell lysates. All samples were loaded in the same gel. Prior to incubating

with anti-PML antibodies, the membrane was cut and probed with indicated antibodies.

Anti-panPML peptide antibodies recognize all nuclear PML isoforms while anti-PML1

antibodies only recognize PML1 (97 KDa) and PML4 (70 KDa). The slowest migrating band (denoted as asterisks) cross react with the anti-panPML (lanes 1-3) or anti-PML1 antibodies are non-specific signals because their intensity were not proportionally decreased in PML knockdown cells. C, Putative targeted sequences of miR-24 and miR133 in human and mouse PML1 mRNA 3’-UTR. The seed regions were labeled in red, and a G-U pair labeled in blue. D, miR-24 or miR-133 decreases PML1 protein expression under normal culture conditions. Whole cell lysates prepared from HUVECs transfected with the indicated miRNAs were analyzed by immunoblotting with anti-

PML1 and anti-β-actin antibodies.

127

Expression of miR-24, but not miR-133, increases PML1 protein levels under

starvation condition or high passage HUVECs

Interestingly, as shown in Figure 33A, PML1 protein expression was increased in miR-24

transfected into high passage HUVECs (passage 10). To confirm that the effect of miR-

24 on PML1 protein expression correlated with cell passage number, we transfected miR-

24 in HUVECs that were passaged from 4 to 10. We found that miR-24 switches from having an inhibitory activity to having a stimulating activity as passage number increases

and that PML1 expression is increased in higher passage HUVECs (Figure 33B). It has been reported that some miRNAs switch from a translation repressor to an activator in cell-cycle arrested cells that are responding to serum starvation or cell post-confluence

(238,239). Interestingly, PML1 protein expression is decreased in serum-starved

HUVECs (Figure 33C). However, PML protein expression was increased in serum- starved HUVECs when cells were transfected with miR-24 but not miR-133 (Figure 33D).

We further examined whether this regulation extends to other endothelial cell types.

Indeed, PML1 protein decreased in serum-starved HAECs and HMVECs, and overexpression of miR-24 increased PML expression in each case (Figures 33E-F). To determine the effect of endogenous miR-24 on PML1 protein, we inhibited endogenous miR-24 by introducing an antagonist of miR-24 and found that overexpression of the miR-24 antagonist decreased PML1 protein expression in serum-starved HUVECs

(Figure 33G, lane 3 vs lane 4). The effect of miR-24 on Pml1 expression in starved or non-starved HUVECs was further confirmed in mouse aorta endothelial cells (Figure

33H).

128

129

Figure 33. Overexpression of miR-24 enhances PML1 protein accumulation in high

passage HUVECs or when cells are serum-starved. A-B, Transient transfection of a

miR-24 mimic increased PML1 protein levels in high passage HUVECs. Whole cell

lysates were prepared from HUVECs at passage 10 (A) or the indicated passage number

(B) of the transfection with indicated miRNA or siRNA were analyzed by

immunoblotting with anti-PML1 and anti-β-actin antibodies. C-F, Overexpression of

miR-24 mimic increased PML1 protein expression under starvation conditions. Passage 4

HUVECs (C-D), HAECs (E) and HMVECs (F) were starved (EGM-2 media with 0.1%

FBS) for the days indicated. Whole cell lysates were prepared from these cells and

followed by immunoblotting (lanes 1-4). ECs at passage 4 were transfected with the

indicated miRNAs and starved for two days. Whole cell lysates were prepared followed

by immunoblotting with anti-PML1 and anti-β-actin antibodies (lanes 5-6). G, HUVECs

were transfected with indicated miRNA mimic (lanes 1-2) or antagonist (lanes 3-4) followed by 2 days of serum starvation. Immunoblotting analysis was performed to measure expression of PMLI. H, Mouse aorta endothelial cells were isolated and transfected with the indicated miRNAs with or without starvation. Immunoblotting analysis was performed as in (G) using an anti-mouse PML antibody.

130

miR24 enhances PML mRNA translation in serum-starved HUVECs

miRNAs regulate their targets primarily by affecting mRNA translation or stability. We

examined the distribution of PML mRNA by polyribosome analysis using extracts

prepared from HUVECs transfected with a miR-Ctrl or miR-24 following serum

starvation. After sucrose gradient centrifugation, total RNA was isolated followed by

Northern blotting and qRT-PCR. As shown by the Northern blots in Figure 34A-B, miR-

24 transfected samples have a higher percentage of PML mRNA in heavy polysomes compared to miR-Crtl transfected cells. The major PML mRNA isoform migrating larger than 5 Kb, which corresponds to PML1 mRNA (5.6 Kb). The size of PML4 transcript, if present would be about 2.2 Kb. This experiment concludes that the protein signals detected in Fig 1 are PML1 and a post-translationally modified form.

Using PML1 mRNA-specific primers and primers common to all PML mRNA isoforms

(Figure 34C), we carried out RT-qPCR. We found that PML1 mRNA in miR-Ctrl- transfected cells showed a total of 42% of PML1 mRNA in polysomes, while a total of 58% of PML1 mRNA present in polysomes prepared from miR-24 transfected cells.

Collectively, these data indicate that miR-24 promotes PML1 mRNA translation upon starvation.

To confirm PML1 mRNA is a miR-24 target, full-length PML1 mRNA 3’-UTR was cloned into a luciferase reporter vector, pmiRGLO (Figure 34D). Under non-starvation conditions, miR-24 reduced the activity of the luciferase reporter gene that contains the wild-type 3’-UTR of PML1 mRNA by 62%, compared to miR-Ctrl transfected samples

(Figure 34E), but increased the reporter activity approximately 2 fold under starvation conditions. However, these fluctuations were not observed in a reporter construct that

131

132

133

Figure 34. miR-24 enhances PML1 mRNA translation in starvation condition. A-D,

HUVECs were transfected with miR-Ctrl or miR-24 and starved for 2 days. Whole cell extracts were subjected to polyribosome analysis on sucrose gradients (A). The distribution of PML1 mRNA was analyzed by Northern blotting and the intensity of the signals were quantified by Image J (B). The percentage of translating PML1 mRNA was determined by dividing the sum of the PML1 mRNA expression in monoribosomes

(fractions 7-8) and polyribosomes (fractions 9-14) over the expression of total PML mRNA isoforms (B) Northern data from (A). (C) RT-PCR. PML1-specifc primers and a primer pair common to all PML isoforms were used for RT-qPCR to quantify the levels of PML1 and total PML mRNA, respectively. D-E, The effect of miR-24 on PML1 mRNA 3’-UTR reporter activity. D, A schematic diagram of human full-length wild-type or mutant PML1 mRNA 3’-UTR reporter constructs. E, HUVECs were transfected with indicated reporter plasmids and miRNA with or without 48 hours of serum starvation.

Whole cell extracts were prepared and subjected to dual luciferase assays. Firefly luciferase activity is normalized to Renilla luciferase activity. The data shown are the mean±s.d of 3 independent experiments. Two-tail t-tests were used to determine statistical significance (* p<0.05).

134

contained a mutant miR-24 binding site with a 3-nucleotide substitution. These results

suggest that PML1 could be a direct target of miR-24 both in non-starvation and starvation conditions with opposite effects.

The effect of miR-24 on PML1 protein expression is independent of FXR1

It has been reported that activation of TNFα mRNA translation in starvation conditions is

FXR1 (Fragile X Mental Retardation Syndrome-Related Protein 1)-dependent (257). To explore whether FXR1 plays a role in miR-24-mediated enhancement of PML1 mRNA translation, we knocked down FXR1 by siRNAs and determined the ability of miR-24 to regulate PML1 mRNA translation. We found that transfection of miR-24 still increased

PML1 protein accumulation in FXR1 knockdown cells (Figure 35A).

Because both miR-24 and miR-133 target PML, we asked whether there is crosstalk between miR-24 and miR-133 in the regulation of PML mRNA translation under starvation condition. As shown in Figure 35B, transfection of miR-24 alone increases

PML1 protein expression (lane 1 vs lane 2), whereas transfection of miR-133 alone slightly down-regulates PML1 protein expression (lane 1 vs lane 3). When miR-24 and miR-133 were both transfected, we found that PML1 expression level was slightly higher than the control cell (lane 1 vs lane 4), but lower than miR-24 transfected cells (lane 2 vs lane 4). These data suggest that the effects of miR-24 and miR-133 on PML mRNA translation function independently of each other. miR-24 inhibits angiogenesis in a PML1-dependet manner

We have previously demonstrated that knockdown of PML in HUVECs increased angiogenesis in a capillary tube formation assay (198) and altered an array of genes implicated in angiogenesis (107). Our observation that overexpression of miR-24

135

Figure 35. The role of FXR1 and miR-133 in miR-24-mediated induction of PML1 protein accumulation. A, HUVECs were infected with a control (shCtrl) or FXR1 shRNA (shFXR1). After puromycin selection, cells were transfected with miR-Ctrl or

miR-24 and starved as described in Figure 33, harvested and subjected to

immunoblotting analysis with anti-PML1, anti-FXR1 and anti-β-actin antibodies. B,

HUVECs were transfected with miR-Ctrl, miR-24, miR-133 or both miR-24 and miR-

133. Two days after starvation, cells were lysed and subjected to immunoblotting analysis with anti-PML1 and anti-β-actin antibodies.

136 increases PML1 mRNA translation under starvation conditions suggests that miR-24 might inhibit EC angiogenesis, in part, by up-regulating PML protein expression.

Consistent with this notion, no significant difference in PML mRNA level between miR-

Ctrl and miR-24 transfected ECs was observed under starvation conditions. However, the angiostatic gene, Notch4 (258), was increased, while angiogenic genes CCL2, SERPIEN1, and EPAS1 (259-261), were decreased in miR-24 transfected cells (Figure 36).

Furthermore, knockdown of PML attenuated the effect of miR-24 on the expression of these PML target genes, suggesting that this regulation is PML-dependent.

To evaluate the biological significance of miR-24 in angiogenesis in serum-starved

HUVECs, we determined the effects of miR-24 on EC angiogenesis. Similar to that observed in Figure 33D, miR-24 increases PML protein expression in serum-starved

HUVECs (Figure 37A). We next performed in vitro capillary tube formation assays and found that miR-24 reduces tube formation in starved HUVECs (Figure 37B-C). To further assess whether PML plays a role in miR-24-regulated angiogenesis, we knocked down PML in miR-24 transfected HUVECs (Figure 37D). As shown in Figure 6E-F, overexpression of miR-24 significantly reduced tube formation, but this effect was largely compromised in PML knockdown, starved HUVECs.

We further carried out aortic ring explant assays to determine the effect of miR-24 on microvessel outgrowths ex-vivo. We found more microvessels sprouting from aortic rings prepared from Pml KO mice than those isolated from the WT animals. In contrast, at days 4 and 6, less microvessel outgrowth was observed in miR-24 transfected wild-type aortic rings. Moreover, the ability of miR-24 to inhibit microvessel outgrowth was abolished in Pml knockout aortic rings (Figure 38).

137

Figure 36. miR-24 regulates the expression of genes involved in angiogenesis in a

PML-dependent manner. HUVECs were transfected with siCtrl or siPML. One day

later, the cells were transfected with miR-Ctrl or miR-24 followed by serum starvation

for another 2 days. Total RNA was extracted and subjected to RT-qPCR with the indicated gene-specific primers.

138

Figure 37. miR-24 inhibits capillary tube formation in a PML-dependent manner. A,

HUVECs were transfected with miR-Ctrl or miR-24 followed by serum starvation. An aliquot of cells was subjected to immunoblotting analysis with anti-PML1 and anti-β- actin antibodies. B-C, Another aliquot of cells was subjected to capillary tube formation assays. D-F, HUVECs were transfected with siCtrl or siPML. One day later, cells were transfected with miR-Ctrl or miR-24 followed by serum starvation for 2 days. The immunoblotting analysis and capillary tube formation assays were performed as described in (A).

139

Figure 38. PML is essential for the inhibitory effect of miR-24 on microvessel

outgrowth of mouse aortic rings. Mouse aortic rings isolated as described in Methods were embedded in Matrigel supplemented with Endothelial Cell Growth Medium-2 plus

FBS (2%) and growth factors. Six hours later, the rings were incubated with miR-Ctrl or miR-24 followed by starvation. Microvessel outgrowth was assessed every 2 days until day 6 using a Leica Wetzlar microscope with bright-field optics. Vessel quantification was performed by counting the numbers of neovessels within the corresponding regions of interest and the mean numbers of sprouts per ring±s.d. Two-tail t-tests were used to determine statistical significance (* p<0.05).

140

DISCUSSION

PML is a tumor suppressor that regulates cell cycle arrest, apoptosis, senescence, cell migration and angiogenesis. We have previously demonstrated that PML inhibits EC migration and capillary tube formation (2,198), suggesting an angiostatic activity. Here, we also show that PML inhibits angiogenesis ex vivo using an aortic ring assay. One of current focuses of this field is to restore PML function in cancer cells with low

expression. While most studies focused on transcriptional induction of PML mRNA or

stability control of PML protein, little is known regarding translational regulation of PML

mRNAs. As the PML gene expresses many spliced isoforms, most of which harbor

distinct 3’-UTR, it is possible that different PML isoforms are subjected to distinct

translational regulation through their unique 3’-UTRs. The current study demonstrates

that miR-24 and miR-133 target the PML1 mRNA 3’-UTR and regulate the expression of

PML1 protein, the major isoform in HUVECs (Figure 32B and 34A). Our results further demonstrate a novel finding that while both miR-24 and miR-133 down-regulate PML1 expression under normal culture conditions, miR-24 switches to up-regulate PML1 mRNA translation under serum starvation culture conditions (Figure 33 and 34). In addition, miR-24 inhibits in vitro and ex vivo models of angiogenesis in a PML1- dependent manner.

The majority of miRNAs function by promoting mRNA decay or decreasing mRNA

translation (237). However, it has been reported that a few miRNAs are capable of

enhancing translation of their targeted mRNAs under certain culture conditions (238-241).

For example, Steitz’s group has reported that upon starvation, miR-369 enhances TNFα

mRNA translation in cultured monocytes in a manner that is dependent on FXR1- and the

141

AU-rich element (ARE) present in the 3’UTR of TNFα mRNA (238,257). Furthermore,

Ruppert’s group reported that miR-206 is capable of promoting translation of KLF4 mRNA in immortalized cells post-confluence, though the mechanism remains largely unknown (239,262). Our data demonstrate that miR-24 enhances PML1 protein levels with little or no effect on mRNA level. Indeed, polyribosome analysis followed by

Northern or RT-qPCR data showed that miR-24 stimulates PML1 mRNA translation in same starved HUVECs. This conclusion is further supported by luciferase reporter assays.

Additionally, knockdown experiments indicated that miR-24 increased PML1 protein levels independent of FXR1 (Figure 35), suggesting a novel mechanism underlying mR-

24-mediated activation of PML1 mRNA translation. In summary, our finding that miR-24

promotes PML1 mRNA translation in HUVECs, HAECs and HMVECs upon serum

starvation and provides additional evidence that miRNAs act in a cell context-specific

manner.

Individual miRNAs usually target multiple mRNAs. miR-24 has previously been shown to target several mRNAs including E2F2, MYC, PAK4, ALK4 and DHFR (251,263-265).

However, our study shows, for the first time, that miR-24 is capable of activating

translation of its targeted mRNA upon starvation. Interestingly, miR-24 has been

proposed to promote apoptosis and inhibit angiogenesis through targeting of the

endothelial GATA2 and the p21-activated kinase PAK4 (253). Nonetheless, using aortic

rings isolated from wild-type and Pml-/- animals, we were able to show that miR-24

inhibits angiogenesis in vitro and ex vivo in part, through inducing PML protein levels

(Figure 38). Furthermore, miR-24 regulates the expression of angiogenic and angiostatic

142

genes in a PML-dependent manner (Figure 36). Taken together, these data suggest that

the miR-24-PML axis inhibits angiogenesis upon starvation.

Our finding that overexpression of a miR-24 mimic activates PML1 mRNA translation raised the question of whether endogenous miR-24 is capable of regulating PML1 protein levels. Indeed, we demonstrated that introduction of an antagonist against miR-24

decreased PML1 protein accumulation (Figure 33G). This observation indicates that the miR-24-mediated increase in PML1 protein level upon starvation is biologically relevant.

Unexpectedly, we also discovered that miR-24 increased PML1 protein levels in high passage HUVECs, even in the absence of serum starvation. High passaging of primary cells induces replicative senescence (266). Interestingly, it was recently shown that miR-

24 is up-regulated in the senescent HUVECs (267). Furthermore, PML protein levels are up-regulated in high passage HUVECs (Figure 33B) and PML promotes senescence

(197,268). These observations suggest a positive feedback of PML in senescent HUVECs and a strong physiological correlation between the induction of miR-24 and PML in senescent cells. Because the gene expression patterns are significantly different between early and high passage HUVECs (269) and between HUVECs grown in normal and serum starvation, it is possible that the constituents of the translation machinery at distinct states are different, thus resulting in distinct effects of miR-24 on PML1 mRNA translation. It will be of significant interest to elucidate the biological meaning of why miR-24 converts from a PML inhibitor to an activator when cells are serum-starved.

Furthermore, the molecular mechanism underlying this switch warrants further investigation.

143

CHAPTER 5: FUTURE DIRECTIONS

The overall goal of my thesis was to elucidate novel mechanisms underlying PML

regulation. Ultimately, we would like to test the hypothesis that PML is a potential target

for cancer therapy. Our current studies have uncovered 3 novel mechanisms by which

PML protein expression is controlled at the posttranslational level. Based on our observations, we proposed the following future directions for the regulation of PML expression and PML-NBs formation (Figure 39):

1) Our cell studies showed that the E3 ubiquitin ligase UHRF1 promotes PML protein degradation via the proteasome pathway. Moreover, knockdown of UHRF1 expression or

abolishing UHRF1 E3 ligase activity increased PML protein accumulation and inhibited

EC migration and capillary tube formation (angiogenesis), two hallmarks of

tumorigenesis. Nonetheless, future investigation using animal models will be essential to

evaluate whether inhibition of UHRF1 slows tumor growth. In summary, these

observations suggest that pharmacological inhibition of UHRF1 is a potential therapeutic

strategy for PML-related cancer.

2) Our studies also showed that SIRT1 and SIRT5 deacetylate PML protein and promote

PML-NB formation. SIRT1 and SIRT5 are indispensable for PML-mediated cell death in

response to H2O2. It would be interesting to test our model using small molecule

activators or inhibitors of SIRT1 and SIRT5 deacetylase. Furthermore, the observation

that H2O2 promotes Sirt1- and Sirt5-dependent nuclear accumulation of PML and PML

NBs in HeLa cells defined a novel mechanistic regulation of PML activity by controlling its subcellular distribution.

144

3) Our unpublished data indicate that miR-24 and miR-133 target the Pml1 3’-UTR and repress PML1 mRNA translation in HUVECs under normal culture conditions. However, it is not clear whether these two miRs behave similarly in cancer cells and in tumorigenesis animal models. Interestingly, the long non-coding RNA H19 (lincRNA

H19) has been proposed to have multiple seed sequences for binding miRs including miR-24 and miR-133 (Table 2) (270). This observation raises the possibility that H19 enhances PML protein expression by sequestering miR-24 and/or miR-133 and attenuates their function. We are currently testing this hypothesis.

Long non-coding RNA, H19, is an imprinted maternally expressed transcript and located in an imprinted region of chromosome 11 near the insulin-like growth factor 2 (IGF2) gene. My unpublished data shown that knockdown UHRF1, which is important for maintenance of DNA methylation, increases IGF2 mRNA expression in MCF7 cells. H19 is one of the most abundant transcripts but the function of this lincRNA is not well studied (271). H19 has been reported to have both tumor suppressor and oncogenic functions. Mutation in H19 or abnormal methylation status at H19 gene locus region is associated with Beckwith-Wiedemann Syndrome and Wilms tumorigenesis (272). In addition, H19, as a molecular sponge, inhibits oncogenic miR-675 and up-regulates miRNA-675 targeting mRNA, transforming growth factor β induced protein (TGFBI).

This H19-miR-675 axis acts as a suppressor in prostate cancer metastasis (273).

However, H19 respectively inhibits tumor suppressor miR-141 and let-7 in gastric cancer and ovarian cancer (274,275).

145

To test the hypothesis that whether H19 inhibits miR-24 and miR-133, and then restore

PML function, it is important to use cell types which highly express both miR-24 and miR-133. After examining 8 breast cell lines including breast cancer cell lines (MCF10A,

MCF12A, MCF7, MDA-MB-231, MDA-MB-468, T47D, 4T1 and SK-BR-3), we found

MDA-MB-231 cells express highest levels of both miR-24 and miR-133 among all cell

lines (data not shown). It would be interesting to examine: 1) the correlation between the

expression of miR-24, miR-133, H19 and PML protein; 2) whether H19 regulates PML

translation and expression and 3) the functions of H19 in MDA-MB-231cell and whether

these functions are mediated by PML.

To explore the potential mechanism of how miR-24 switch from repression to activation

of PML translation in different cell culture conditions, the purification of PML1 3’-UTR

associated mRNP complex would be one the potential approaches. The detailed methods

can be referred to Joan A. Steitz’s study (257).

146

Figure 39. Summary of three distinct mechanisms for post-transcriptional targeting

PML protein.

147

Table 2. Summary of Putative miRNA-Binding Sites on Human H19. This Table was adopted from (270). RNA hybrid program was used to predict binding sites for 18 muscle-related miRNAs. miRNA # sites miR-133b 11 miR-133a 10 miR-24 10 miR-24-3p 10 miR-145 7 miR-378 6 5 miR-127-3p miR-136 4 miR-532-3p 4 miR-210 4 miR-100 3 miR-125b-3p 3 miR-125a-3p 3 miR-26a 2 miR-206 1 miR-1 1 miR-19b 1 miR-23b 1

148

The second part of my future directions is to study the downstream targets of PML

protein and extend the mechanism, originally discovered in PML, to other proteins.

One key direction for future study will be the role of PML on epigenetics and chromatin

organization. Many histone modifying enzymes and enzymatic components of chromatin

remodeling complexes interact with PML. For example, protein acetyltransferase (p300),

deacetylase (HDAC1, HDAC2, HDAC3, HDAC7, SIRT1 and SIRT5) (3,65,69,152,153),

methyltransferases (SETDB1 and SUV39H1) (154,155), component of polycomb repressive complex (EZH2) (156), and epigenetic regulator UHRF1 physically associate with PML (119). However, little is known about whether PML regulates the activity of these factors. Understanding the epigenetic regulation by PML is a pivotal step toward elucidating the mechanism of tumor suppressor role of PML. Currently, γ-irradiation and chemical therapies IFN and IL6, have been shown to stimulate accumulation of PML protein (49,103,104,109). PML protein can also be induced by several small molecules, including MLN4924 (target KLHL20), emodin (target CK2), XMD8-92 (target BMK1),

TSA (HDAC inhibitor), as well as other stimuli such as ROS/H2O2, EGF, sulforaphane,

MG132, As2O3 and DNA damage regents (115,146,157,158). It will be informative to see

whether combinatorial treatment with these reagents enhances potency of their anti- cancer activity by synergistically increasing PML protein accumulation.

Studies from Pandolfi’s group suggested that PML also has oncogenic function in chronic myeloid leukemia due to its requirement for the maintenance of hematopoietic stem cell

(159). The same group also reported that PML is overexpressed in triple-negative breast cancer patients (160). Our lab recently demonstrated that Pml KO mice exhibited increased fatty acid oxidation in liver, which contributes to reduced incidence of Western

149

diet-induced dysplastic hepatic nodules. How PML may switch from a tumor suppressor

in one tissue to an oncoprotein in another tissue is an outstanding question and warrants further investigation.

150

References

1. Gurrieri, C., Capodieci, P., Bernardi, R., Scaglioni, P. P., Nafa, K., Rush, L. J., Verbel, D. A., Cordon-Cardo, C., and Pandolfi, P. P. (2004) Loss of the tumor suppressor PML in human cancers of multiple histologic origins. J Natl Cancer Inst 96, 269-279 2. Guan, D., Factor, D., Liu, Y., Wang, Z., and Kao, H. Y. (2012) The epigenetic regulator UHRF1 promotes ubiquitination-mediated degradation of the tumor-suppressor protein promyelocytic leukemia protein. Oncogene 3. Guan, D., Lim, J. H., Peng, L., Liu, Y., Lam, M., Seto, E., and Kao, H. Y. (2014) Deacetylation of the tumor suppressor protein PML regulates hydrogen peroxide- induced cell death. Cell Death Dis 5, e1340 4. de The, H., Lavau, C., Marchio, A., Chomienne, C., Degos, L., and Dejean, A. (1991) The PML-RAR alpha fusion mRNA generated by the t(15;17) translocation in acute promyelocytic leukemia encodes a functionally altered RAR. Cell 66, 675-684 5. Kakizuka, A., Miller, W. H., Jr., Umesono, K., Warrell, R. P., Jr., Frankel, S. R., Murty, V. V., Dmitrovsky, E., and Evans, R. M. (1991) Chromosomal translocation t(15;17) in human acute promyelocytic leukemia fuses RAR alpha with a novel putative transcription factor, PML. Cell 66, 663-674 6. de The, H., Le Bras, M., and Lallemand-Breitenbach, V. (2012) The cell biology of disease: Acute promyelocytic leukemia, arsenic, and PML bodies. J Cell Biol 198, 11-21 7. Bernardi, R., and Pandolfi, P. P. (2007) Structure, dynamics and functions of promyelocytic leukaemia nuclear bodies. Nat Rev Mol Cell Biol 8, 1006-1016 8. Lallemand-Breitenbach, V., and de The, H. (2010) PML nuclear bodies. Cold Spring Harb Perspect Biol 2, a000661 9. Sahin, U., Lallemand-Breitenbach, V., and de The, H. (2014) PML nuclear bodies: regulation, function and therapeutic perspectives. J Pathol 234, 289-291 10. Guo, L., Giasson, B. I., Glavis-Bloom, A., Brewer, M. D., Shorter, J., Gitler, A. D., and Yang, X. (2014) A cellular system that degrades misfolded proteins and protects against neurodegeneration. Mol Cell 55, 15-30 11. Gamell, C., Jan Paul, P., Haupt, Y., and Haupt, S. (2014) PML tumour suppression and beyond: Therapeutic implications. FEBS Lett 12. Miki, T., Xu, Z., Chen-Goodspeed, M., Liu, M., Van Oort-Jansen, A., Rea, M. A., Zhao, Z., Lee, C. C., and Chang, K. S. (2012) PML regulates PER2 nuclear localization and circadian function. EMBO J 31, 1427-1439 13. Nisole, S., Maroui, M. A., Mascle, X. H., Aubry, M., and Chelbi-Alix, M. K. (2013) Differential Roles of PML Isoforms. Front Oncol 3, 125 14. Borden, K. L. (1998) RING fingers and B-boxes: zinc-binding protein-protein interaction domains. Biochem Cell Biol 76, 351-358 15. Reymond, A., Meroni, G., Fantozzi, A., Merla, G., Cairo, S., Luzi, L., Riganelli, D., Zanaria, E., Messali, S., Cainarca, S., Guffanti, A., Minucci, S., Pelicci, P. G., and Ballabio, A. (2001) The tripartite motif family identifies cell compartments. EMBO J 20, 2140-2151 16. Shen, T. H., Lin, H. K., Scaglioni, P. P., Yung, T. M., and Pandolfi, P. P. (2006) The mechanisms of PML-nuclear body formation. Mol Cell 24, 331-339 17. Hodges, M., Tissot, C., Howe, K., Grimwade, D., and Freemont, P. S. (1998) Structure, organization, and dynamics of promyelocytic leukemia protein nuclear bodies. Am J Hum Genet 63, 297-304 18. Jensen, K., Shiels, C., and Freemont, P. S. (2001) PML protein isoforms and the RBCC/TRIM motif. Oncogene 20, 7223-7233

151

19. Salomoni, P., and Pandolfi, P. P. (2002) The role of PML in tumor suppression. Cell 108, 165-170 20. Batty, E. C., Jensen, K., and Freemont, P. S. (2012) PML nuclear bodies and other TRIM- defined subcellular compartments. Adv Exp Med Biol 770, 39-58 21. Seeler, J. S., and Dejean, A. (1999) The PML nuclear bodies: actors or extras? Curr Opin Genet Dev 9, 362-367 22. Van Damme, E., Laukens, K., Dang, T. H., and Van Ostade, X. (2010) A manually curated network of the PML nuclear body interactome reveals an important role for PML-NBs in SUMOylation dynamics. Int J Biol Sci 6, 51-67 23. Li, H., Leo, C., Zhu, J., Wu, X., O'Neil, J., Park, E. J., and Chen, J. D. (2000) Sequestration and inhibition of Daxx-mediated transcriptional repression by PML. Mol Cell Biol 20, 1784-1796 24. Fagioli, M., Alcalay, M., Tomassoni, L., Ferrucci, P. F., Mencarelli, A., Riganelli, D., Grignani, F., Pozzan, T., Nicoletti, I., and Pelicci, P. G. (1998) Cooperation between the RING + B1-B2 and coiled-coil domains of PML is necessary for its effects on cell survival. Oncogene 16, 2905-2913 25. Cao, T., Borden, K. L., Freemont, P. S., and Etkin, L. D. (1997) Involvement of the rfp tripartite motif in protein-protein interactions and subcellular distribution. J Cell Sci 110 ( Pt 14), 1563-1571 26. Peng, H., Begg, G. E., Schultz, D. C., Friedman, J. R., Jensen, D. E., Speicher, D. W., and Rauscher, F. J., 3rd. (2000) Reconstitution of the KRAB-KAP-1 repressor complex: a model system for defining the molecular anatomy of RING-B box-coiled-coil domain- mediated protein-protein interactions. J Mol Biol 295, 1139-1162 27. Gambacorta, M., Flenghi, L., Fagioli, M., Pileri, S., Leoncini, L., Bigerna, B., Pacini, R., Tanci, L. N., Pasqualucci, L., Ascani, S., Mencarelli, A., Liso, A., Pelicci, P. G., and Falini, B. (1996) Heterogeneous nuclear expression of the promyelocytic leukemia (PML) protein in normal and neoplastic human tissues. Am J Pathol 149, 2023-2035 28. Lee, H. E., Jee, C. D., Kim, M. A., Lee, H. S., Lee, Y. M., Lee, B. L., and Kim, W. H. (2007) Loss of promyelocytic leukemia protein in human gastric cancers. Cancer Lett 247, 103- 109 29. Zhang, P., Chin, W., Chow, L. T., Chan, A. S., Yim, A. P., Leung, S. F., Mok, T. S., Chang, K. S., Johnson, P. J., and Chan, J. Y. (2000) Lack of expression for the suppressor PML in human small cell lung carcinoma. Int J Cancer 85, 599-605 30. Koken, M. H., Linares-Cruz, G., Quignon, F., Viron, A., Chelbi-Alix, M. K., Sobczak-Thepot, J., Juhlin, L., Degos, L., Calvo, F., and de The, H. (1995) The PML growth-suppressor has an altered expression in human oncogenesis. Oncogene 10, 1315-1324 31. Mu, Z. M., Le, X. F., Vallian, S., Glassman, A. B., and Chang, K. S. (1997) Stable overexpression of PML alters regulation of cell cycle progression in HeLa cells. Carcinogenesis 18, 2063-2069 32. Wang, Z. G., Delva, L., Gaboli, M., Rivi, R., Giorgio, M., Cordon-Cardo, C., Grosveld, F., and Pandolfi, P. P. (1998) Role of PML in cell growth and the retinoic acid pathway. Science 279, 1547-1551 33. Rego, E. M., Wang, Z. G., Peruzzi, D., He, L. Z., Cordon-Cardo, C., and Pandolfi, P. P. (2001) Role of promyelocytic leukemia (PML) protein in tumor suppression. J Exp Med 193, 521-529 34. Mao, Y. S., Zhang, B., and Spector, D. L. (2011) Biogenesis and function of nuclear bodies. Trends Genet 27, 295-306

152

35. Reineke, E. L., and Kao, H. Y. (2009) Targeting promyelocytic leukemia protein: a means to regulating PML nuclear bodies. Int J Biol Sci 5, 366-376 36. Huang, Q., Li, F., Liu, X., Li, W., Shi, W., Liu, F. F., O'Sullivan, B., He, Z., Peng, Y., Tan, A. C., Zhou, L., Shen, J., Han, G., Wang, X. J., Thorburn, J., Thorburn, A., Jimeno, A., Raben, D., Bedford, J. S., and Li, C. Y. (2011) Caspase 3-mediated stimulation of tumor cell repopulation during cancer radiotherapy. Nat Med 17, 860-866 37. Bernardi, R., and Pandolfi, P. P. (2003) Role of PML and the PML-nuclear body in the control of programmed cell death. Oncogene 22, 9048-9057 38. Wang, Z. G., Ruggero, D., Ronchetti, S., Zhong, S., Gaboli, M., Rivi, R., and Pandolfi, P. P. (1998) PML is essential for multiple apoptotic pathways. Nat Genet 20, 266-272 39. Quignon, F., De Bels, F., Koken, M., Feunteun, J., Ameisen, J. C., and de The, H. (1998) PML induces a novel caspase-independent death process. Nat Genet 20, 259-265 40. Levine, A. J., and Oren, M. (2009) The first 30 years of p53: growing ever more complex. Nat Rev Cancer 9, 749-758 41. Bernardi, R., Scaglioni, P. P., Bergmann, S., Horn, H. F., Vousden, K. H., and Pandolfi, P. P. (2004) PML regulates p53 stability by sequestering Mdm2 to the nucleolus. Nat Cell Biol 6, 665-672 42. Kurki, S., Latonen, L., and Laiho, M. (2003) Cellular stress and DNA damage invoke temporally distinct Mdm2, p53 and PML complexes and damage-specific nuclear relocalization. J Cell Sci 116, 3917-3925 43. Louria-Hayon, I., Grossman, T., Sionov, R. V., Alsheich, O., Pandolfi, P. P., and Haupt, Y. (2003) The promyelocytic leukemia protein protects p53 from Mdm2-mediated inhibition and degradation. J Biol Chem 278, 33134-33141 44. Yang, Q., Liao, L., Deng, X., Chen, R., Gray, N. S., Yates, J. R., 3rd, and Lee, J. D. (2013) BMK1 is involved in the regulation of p53 through disrupting the PML-MDM2 interaction. Oncogene 32, 3156-3164 45. Yang, S., Kuo, C., Bisi, J. E., and Kim, M. K. (2002) PML-dependent apoptosis after DNA damage is regulated by the checkpoint kinase hCds1/Chk2. Nat Cell Biol 4, 865-870 46. D'Orazi, G., Cecchinelli, B., Bruno, T., Manni, I., Higashimoto, Y., Saito, S., Gostissa, M., Coen, S., Marchetti, A., Del Sal, G., Piaggio, G., Fanciulli, M., Appella, E., and Soddu, S. (2002) Homeodomain-interacting protein kinase-2 phosphorylates p53 at Ser 46 and mediates apoptosis. Nat Cell Biol 4, 11-19 47. Hofmann, T. G., Moller, A., Sirma, H., Zentgraf, H., Taya, Y., Droge, W., Will, H., and Schmitz, M. L. (2002) Regulation of p53 activity by its interaction with homeodomain- interacting protein kinase-2. Nat Cell Biol 4, 1-10 48. Li, Q., He, Y., Wei, L., Wu, X., Wu, D., Lin, S., Wang, Z., Ye, Z., and Lin, S. C. (2011) AXIN is an essential co-activator for the promyelocytic leukemia protein in p53 activation. Oncogene 30, 1194-1204 49. Guo, A., Salomoni, P., Luo, J., Shih, A., Zhong, S., Gu, W., and Pandolfi, P. P. (2000) The function of PML in p53-dependent apoptosis. Nat Cell Biol 2, 730-736 50. Langley, E., Pearson, M., Faretta, M., Bauer, U. M., Frye, R. A., Minucci, S., Pelicci, P. G., and Kouzarides, T. (2002) Human SIR2 deacetylates p53 and antagonizes PML/p53- induced cellular senescence. EMBO J 21, 2383-2396 51. Carnero, A., and Paramio, J. M. (2014) The PTEN/PI3K/AKT Pathway in vivo, Cancer Mouse Models. Front Oncol 4, 252 52. Trotman, L. C., Alimonti, A., Scaglioni, P. P., Koutcher, J. A., Cordon-Cardo, C., and Pandolfi, P. P. (2006) Identification of a tumour suppressor network opposing nuclear Akt function. Nature 441, 523-527

153

53. Culjkovic, B., Tan, K., Orolicki, S., Amri, A., Meloche, S., and Borden, K. L. (2008) The eIF4E RNA regulon promotes the Akt signaling pathway. J Cell Biol 181, 51-63 54. Trotman, L. C., Wang, X., Alimonti, A., Chen, Z., Teruya-Feldstein, J., Yang, H., Pavletich, N. P., Carver, B. S., Cordon-Cardo, C., Erdjument-Bromage, H., Tempst, P., Chi, S. G., Kim, H. J., Misteli, T., Jiang, X., and Pandolfi, P. P. (2007) Ubiquitination regulates PTEN nuclear import and tumor suppression. Cell 128, 141-156 55. Song, M. S., Salmena, L., Carracedo, A., Egia, A., Lo-Coco, F., Teruya-Feldstein, J., and Pandolfi, P. P. (2008) The deubiquitinylation and localization of PTEN are regulated by a HAUSP-PML network. Nature 455, 813-817 56. Giorgi, C., Ito, K., Lin, H. K., Santangelo, C., Wieckowski, M. R., Lebiedzinska, M., Bononi, A., Bonora, M., Duszynski, J., Bernardi, R., Rizzuto, R., Tacchetti, C., Pinton, P., and Pandolfi, P. P. (2010) PML regulates apoptosis at endoplasmic reticulum by modulating calcium release. Science 330, 1247-1251 57. Regad, T., Bellodi, C., Nicotera, P., and Salomoni, P. (2009) The tumor suppressor Pml regulates cell fate in the developing neocortex. Nat Neurosci 12, 132-140 58. Vernier, M., Bourdeau, V., Gaumont-Leclerc, M. F., Moiseeva, O., Begin, V., Saad, F., Mes-Masson, A. M., and Ferbeyre, G. (2011) Regulation of E2Fs and senescence by PML nuclear bodies. Genes Dev 25, 41-50 59. Ferbeyre, G., de Stanchina, E., Querido, E., Baptiste, N., Prives, C., and Lowe, S. W. (2000) PML is induced by oncogenic ras and promotes premature senescence. Genes Dev 14, 2015-2027 60. Wu, W. S., Xu, Z. X., Hittelman, W. N., Salomoni, P., Pandolfi, P. P., and Chang, K. S. (2003) Promyelocytic leukemia protein sensitizes tumor necrosis factor alpha-induced apoptosis by inhibiting the NF-kappaB survival pathway. J Biol Chem 278, 12294-12304 61. Wu, W. S., Xu, Z. X., and Chang, K. S. (2002) The promyelocytic leukemia protein represses A20-mediated transcription. J Biol Chem 277, 31734-31739 62. Wu, W. S., Xu, Z. X., Ran, R., Meng, F., and Chang, K. S. (2002) Promyelocytic leukemia protein PML inhibits Nur77-mediated transcription through specific functional interactions. Oncogene 21, 3925-3933 63. Vallian, S., Chin, K. V., and Chang, K. S. (1998) The promyelocytic leukemia protein interacts with Sp1 and inhibits its transactivation of the epidermal growth factor receptor promoter. Mol Cell Biol 18, 7147-7156 64. Kawasaki, A., Matsumura, I., Kataoka, Y., Takigawa, E., Nakajima, K., and Kanakura, Y. (2003) Opposing effects of PML and PML/RAR alpha on STAT3 activity. Blood 101, 3668- 3673 65. Khan, M. M., Nomura, T., Kim, H., Kaul, S. C., Wadhwa, R., Shinagawa, T., Ichikawa-Iwata, E., Zhong, S., Pandolfi, P. P., and Ishii, S. (2001) Role of PML and PML-RARalpha in Mad- mediated transcriptional repression. Mol Cell 7, 1233-1243 66. Lin, D. Y., Lai, M. Z., Ann, D. K., and Shih, H. M. (2003) Promyelocytic leukemia protein (PML) functions as a glucocorticoid receptor co-activator by sequestering Daxx to the PML oncogenic domains (PODs) to enhance its transactivation potential. J Biol Chem 278, 15958-15965 67. Lin, D. Y., Huang, Y. S., Jeng, J. C., Kuo, H. Y., Chang, C. C., Chao, T. T., Ho, C. C., Chen, Y. C., Lin, T. P., Fang, H. I., Hung, C. C., Suen, C. S., Hwang, M. J., Chang, K. S., Maul, G. G., and Shih, H. M. (2006) Role of SUMO-interacting motif in Daxx SUMO modification, subnuclear localization, and repression of sumoylated transcription factors. Mol Cell 24, 341-354

154

68. Chang, C. C., Naik, M. T., Huang, Y. S., Jeng, J. C., Liao, P. H., Kuo, H. Y., Ho, C. C., Hsieh, Y. L., Lin, C. H., Huang, N. J., Naik, N. M., Kung, C. C., Lin, S. Y., Chen, R. H., Chang, K. S., Huang, T. H., and Shih, H. M. (2011) Structural and functional roles of Daxx SIM phosphorylation in SUMO paralog-selective binding and apoptosis modulation. Mol Cell 42, 62-74 69. Gao, C., Cheng, X., Lam, M., Liu, Y., Liu, Q., Chang, K. S., and Kao, H. Y. (2008) Signal- dependent regulation of transcription by histone deacetylase 7 involves recruitment to promyelocytic leukemia protein nuclear bodies. Mol Biol Cell 19, 3020-3027 70. Ulbricht, T., Alzrigat, M., Horch, A., Reuter, N., von Mikecz, A., Steimle, V., Schmitt, E., Kramer, O. H., Stamminger, T., and Hemmerich, P. (2012) PML promotes MHC class II gene expression by stabilizing the class II transactivator. J Cell Biol 199, 49-63 71. Chen, Y., Wright, J., Meng, X., and Leppard, K. N. (2015) Promyelocytic Leukemia Protein Isoform II Promotes Transcription Factor Recruitment To Activate Interferon Beta and Interferon-Responsive Gene Expression. Mol Cell Biol 35, 1660-1672 72. Shtutman, M., Zhurinsky, J., Oren, M., Levina, E., and Ben-Ze'ev, A. (2002) PML is a target gene of beta-catenin and plakoglobin, and coactivates beta-catenin-mediated transcription. Cancer Res 62, 5947-5954 73. Vallian, S., Gaken, J. A., Gingold, E. B., Kouzarides, T., Chang, K. S., and Farzaneh, F. (1998) Modulation of Fos-mediated AP-1 transcription by the promyelocytic leukemia protein. Oncogene 16, 2843-2853 74. Salomoni, P., Bernardi, R., Bergmann, S., Changou, A., Tuttle, S., and Pandolfi, P. P. (2005) The promyelocytic leukemia protein PML regulates c-Jun function in response to DNA damage. Blood 105, 3686-3690 75. Wu, J., Zhou, L. Q., Yu, W., Zhao, Z. G., Xie, X. M., Wang, W. T., Xiong, J., Li, M., Xue, Z., Wang, X., Zhang, P., Mao, B. B., Hao, D. L., Lv, X., and Liu, D. P. (2014) PML4 facilitates erythroid differentiation by enhancing the transcriptional activity of GATA-1. Blood 123, 261-270 76. Tsuzuki, S., Towatari, M., Saito, H., and Enver, T. (2000) Potentiation of GATA-2 activity through interactions with the promyelocytic leukemia protein (PML) and the t(15;17)- generated PML-retinoic acid receptor alpha oncoprotein. Mol Cell Biol 20, 6276-6286 77. Dellaire, G., and Bazett-Jones, D. P. (2004) PML nuclear bodies: dynamic sensors of DNA damage and cellular stress. Bioessays 26, 963-977 78. Yeager, T. R., Neumann, A. A., Englezou, A., Huschtscha, L. I., Noble, J. R., and Reddel, R. R. (1999) Telomerase-negative immortalized human cells contain a novel type of promyelocytic leukemia (PML) body. Cancer Res 59, 4175-4179 79. Grobelny, J. V., Godwin, A. K., and Broccoli, D. (2000) ALT-associated PML bodies are present in viable cells and are enriched in cells in the G(2)/M phase of the cell cycle. J Cell Sci 113 Pt 24, 4577-4585 80. Chung, I., Osterwald, S., Deeg, K. I., and Rippe, K. (2012) PML body meets telomere: the beginning of an ALTernate ending? Nucleus 3, 263-275 81. Flynn, R. L., Cox, K. E., Jeitany, M., Wakimoto, H., Bryll, A. R., Ganem, N. J., Bersani, F., Pineda, J. R., Suva, M. L., Benes, C. H., Haber, D. A., Boussin, F. D., and Zou, L. (2015) Alternative lengthening of telomeres renders cancer cells hypersensitive to ATR inhibitors. Science 347, 273-277 82. Xu, Z. X., Timanova-Atanasova, A., Zhao, R. X., and Chang, K. S. (2003) PML colocalizes with and stabilizes the DNA damage response protein TopBP1. Mol Cell Biol 23, 4247- 4256

155

83. Dellaire, G., Ching, R. W., Ahmed, K., Jalali, F., Tse, K. C., Bristow, R. G., and Bazett-Jones, D. P. (2006) Promyelocytic leukemia nuclear bodies behave as DNA damage sensors whose response to DNA double-strand breaks is regulated by NBS1 and the kinases ATM, Chk2, and ATR. J Cell Biol 175, 55-66 84. Zhong, S., Hu, P., Ye, T. Z., Stan, R., Ellis, N. A., and Pandolfi, P. P. (1999) A role for PML and the nuclear body in genomic stability. Oncogene 18, 7941-7947 85. Bischof, O., Kim, S. H., Irving, J., Beresten, S., Ellis, N. A., and Campisi, J. (2001) Regulation and localization of the Bloom syndrome protein in response to DNA damage. J Cell Biol 153, 367-380 86. Le, S., Moore, J. K., Haber, J. E., and Greider, C. W. (1999) RAD50 and RAD51 define two pathways that collaborate to maintain telomeres in the absence of telomerase. Genetics 152, 143-152 87. Carbone, R., Pearson, M., Minucci, S., and Pelicci, P. G. (2002) PML NBs associate with the hMre11 complex and p53 at sites of irradiation induced DNA damage. Oncogene 21, 1633-1640 88. Wu, G., Lee, W. H., and Chen, P. L. (2000) NBS1 and TRF1 colocalize at promyelocytic leukemia bodies during late S/G2 phases in immortalized telomerase-negative cells. Implication of NBS1 in alternative lengthening of telomeres. J Biol Chem 275, 30618- 30622 89. Zheng, P., Guo, Y., Niu, Q., Levy, D. E., Dyck, J. A., Lu, S., Sheiman, L. A., and Liu, Y. (1998) Proto-oncogene PML controls genes devoted to MHC class I antigen presentation. Nature 396, 373-376 90. Lane, A. A., and Ley, T. J. (2003) Neutrophil elastase cleaves PML-RARalpha and is important for the development of acute promyelocytic leukemia in mice. Cell 115, 305- 318 91. Gurrieri, C., Nafa, K., Merghoub, T., Bernardi, R., Capodieci, P., Biondi, A., Nimer, S., Douer, D., Cordon-Cardo, C., Gallagher, R., and Pandolfi, P. P. (2004) Mutations of the PML tumor suppressor gene in acute promyelocytic leukemia. Blood 103, 2358-2362 92. Gao, Y. M., Zhong, L., Zhang, X., Hu, X. X., and Liu, B. Z. (2013) PML(NLS(-)) inhibits cell apoptosis and promotes proliferation in HL-60 cells. Int J Med Sci 10, 498-507 93. Bellodi, C., Kindle, K., Bernassola, F., Dinsdale, D., Cossarizza, A., Melino, G., Heery, D., and Salomoni, P. (2006) Cytoplasmic function of mutant promyelocytic leukemia (PML) and PML-retinoic acid receptor-alpha. J Biol Chem 281, 14465-14473 94. Bellodi, C., Kindle, K., Bernassola, F., Cossarizza, A., Dinsdale, D., Melino, G., Heery, D., and Salomoni, P. (2006) A cytoplasmic PML mutant inhibits p53 function. Cell Cycle 5, 2688-2692 95. Terris, B., Baldin, V., Dubois, S., Degott, C., Flejou, J. F., Henin, D., and Dejean, A. (1995) PML nuclear bodies are general targets for inflammation and cell proliferation. Cancer Res 55, 1590-1597 96. Chan, J. Y., Chin, W., Liew, C. T., Chang, K. S., and Johnson, P. J. (1998) Altered expression of the growth and transformation suppressor PML gene in human hepatocellular carcinomas and in hepatitis tissues. Eur J Cancer 34, 1015-1022 97. Chen, Z., Lu, X., Wang, Z., Jin, G., Wang, Q., Chen, D., Chen, T., Li, J., Fan, J., Cong, W., Gao, Q., and He, X. (2015) Co-expression of PKM2 and TRIM35 predicts survival and recurrence in hepatocellular carcinoma. Oncotarget 6, 2538-2548 98. Shimada, N., Shinagawa, T., and Ishii, S. (2008) Modulation of M2-type pyruvate kinase activity by the cytoplasmic PML tumor suppressor protein. Genes Cells 13, 245-254

156

99. Ikushima, H., and Miyazono, K. (2010) TGFbeta signalling: a complex web in cancer progression. Nat Rev Cancer 10, 415-424 100. Lin, H. K., Bergmann, S., and Pandolfi, P. P. (2004) Cytoplasmic PML function in TGF-beta signalling. Nature 431, 205-211 101. Seo, S. R., Ferrand, N., Faresse, N., Prunier, C., Abecassis, L., Pessah, M., Bourgeade, M. F., and Atfi, A. (2006) Nuclear retention of the tumor suppressor cPML by the homeodomain protein TGIF restricts TGF-beta signaling. Mol Cell 23, 547-559 102. Salomoni, P., Dvorkina, M., and Michod, D. (2012) Role of the promyelocytic leukaemia protein in cell death regulation. Cell Death Dis 3, e247 103. Stadler, M., Chelbi-Alix, M. K., Koken, M. H., Venturini, L., Lee, C., Saib, A., Quignon, F., Pelicano, L., Guillemin, M. C., Schindler, C., and et al. (1995) Transcriptional induction of the PML growth suppressor gene by interferons is mediated through an ISRE and a GAS element. Oncogene 11, 2565-2573 104. Kim, T. K., Lee, J. S., Oh, S. Y., Jin, X., Choi, Y. J., Lee, T. H., Lee, E., Choi, Y. K., You, S., Chung, Y. G., Lee, J. B., DePinho, R. A., Chin, L., and Kim, H. (2007) Direct transcriptional activation of promyelocytic leukemia protein by IFN regulatory factor 3 induces the p53- dependent growth inhibition of cancer cells. Cancer Res 67, 11133-11140 105. Sasaki, M., Ikeda, H., Sato, Y., and Nakanuma, Y. (2008) Proinflammatory cytokine- induced cellular senescence of biliary epithelial cells is mediated via oxidative stress and activation of ATM pathway: a culture study. Free Radic Res 42, 625-632 106. Chelbi-Alix, M. K., Pelicano, L., Quignon, F., Koken, M. H., Venturini, L., Stadler, M., Pavlovic, J., Degos, L., and de The, H. (1995) Induction of the PML protein by interferons in normal and APL cells. Leukemia 9, 2027-2033 107. Cheng, X., and Kao, H. Y. (2012) Microarray analysis revealing common and distinct functions of promyelocytic leukemia protein (PML) and tumor necrosis factor alpha (TNFalpha) signaling in endothelial cells. BMC Genomics 13, 453 108. Cheng, X., Liu, Y., Chu, H., and Kao, H. Y. (2012) Promyelocytic leukemia protein (PML) regulates endothelial cell network formation and migration in response to tumor necrosis factor alpha (TNFalpha) and interferon alpha (IFNalpha). J Biol Chem 287, 23356-23367 109. Hubackova, S., Krejcikova, K., Bartek, J., and Hodny, Z. (2012) Interleukin 6 signaling regulates promyelocytic leukemia protein gene expression in human normal and cancer cells. J Biol Chem 287, 26702-26714 110. de Stanchina, E., Querido, E., Narita, M., Davuluri, R. V., Pandolfi, P. P., Ferbeyre, G., and Lowe, S. W. (2004) PML is a direct p53 target that modulates p53 effector functions. Mol Cell 13, 523-535 111. Lapi, E., Di Agostino, S., Donzelli, S., Gal, H., Domany, E., Rechavi, G., Pandolfi, P. P., Givol, D., Strano, S., Lu, X., and Blandino, G. (2008) PML, YAP, and p73 are components of a proapoptotic autoregulatory feedback loop. Mol Cell 32, 803-814 112. Scaglioni, P. P., Rabellino, A., Yung, T. M., Bernardi, R., Choi, S., Konstantinidou, G., Nardella, C., Cheng, K., and Pandolfi, P. P. (2012) Translation-dependent mechanisms lead to PML upregulation and mediate oncogenic K-RAS-induced cellular senescence. EMBO Mol Med 4, 594-602 113. Hsu, K.-S., Guan, B.-J., Cheng, X., Guan, D., Lam, M., Hatzoglou, M., and Kao, H.-Y. Translational control of PML contributes to TNFa-induced apoptosis of MCF7 breast cancer cells and decreased angiogenesis in HUVECs. Cell Death Differ accepted 114. Reineke, E. L., and Kao, H. Y. (2009) PML: An emerging tumor suppressor and a target with therapeutic potential. Cancer Ther 7, 219-226

157

115. Cheng, X., and Kao, H. Y. (2012) Post-translational modifications of PML: consequences and implications. Front Oncol 2, 210 116. Chen, R. H., Lee, Y. R., and Yuan, W. C. (2012) The role of PML ubiquitination in human malignancies. J Biomed Sci 19, 81 117. Lallemand-Breitenbach, V., Jeanne, M., Benhenda, S., Nasr, R., Lei, M., Peres, L., Zhou, J., Zhu, J., Raught, B., and de The, H. (2008) Arsenic degrades PML or PML-RARalpha through a SUMO-triggered RNF4/ubiquitin-mediated pathway. Nat Cell Biol 10, 547-555 118. Tatham, M. H., Geoffroy, M. C., Shen, L., Plechanovova, A., Hattersley, N., Jaffray, E. G., Palvimo, J. J., and Hay, R. T. (2008) RNF4 is a poly-SUMO-specific E3 ubiquitin ligase required for arsenic-induced PML degradation. Nat Cell Biol 10, 538-546 119. Guan, D., Factor, D., Liu, Y., Wang, Z., and Kao, H. Y. (2013) The epigenetic regulator UHRF1 promotes ubiquitination-mediated degradation of the tumor-suppressor protein promyelocytic leukemia protein. Oncogene 32, 3819-3828 120. Louria-Hayon, I., Alsheich-Bartok, O., Levav-Cohen, Y., Silberman, I., Berger, M., Grossman, T., Matentzoglu, K., Jiang, Y. H., Muller, S., Scheffner, M., Haupt, S., and Haupt, Y. (2009) E6AP promotes the degradation of the PML tumor suppressor. Cell Death Differ 16, 1156-1166 121. Yuan, W. C., Lee, Y. R., Huang, S. F., Lin, Y. M., Chen, T. Y., Chung, H. C., Tsai, C. H., Chen, H. Y., Chiang, C. T., Lai, C. K., Lu, L. T., Chen, C. H., Gu, D. L., Pu, Y. S., Jou, Y. S., Lu, K. P., Hsiao, P. W., Shih, H. M., and Chen, R. H. (2011) A Cullin3-KLHL20 Ubiquitin ligase- dependent pathway targets PML to potentiate HIF-1 signaling and prostate cancer progression. Cancer Cell 20, 214-228 122. Fanelli, M., Fantozzi, A., De Luca, P., Caprodossi, S., Matsuzawa, S., Lazar, M. A., Pelicci, P. G., and Minucci, S. (2004) The coiled-coil domain is the structural determinant for mammalian homologues of Drosophila Sina-mediated degradation of promyelocytic leukemia protein and other tripartite motif proteins by the proteasome. J Biol Chem 279, 5374-5379 123. Qin, Y., Wang, J., Gong, W., Zhang, M., Tang, Z., Zhang, J., and Quan, Z. (2014) UHRF1 depletion suppresses growth of gallbladder cancer cells through induction of apoptosis and cell cycle arrest. Oncol Rep 31, 2635-2643 124. Chen, H. Y., Hu, J. Y., Chen, T. H., Lin, Y. C., Liu, X., Lin, M. Y., Lang, Y. D., Yen, Y., and Chen, R. H. (2015) KLHL39 suppresses colon cancer metastasis by blocking KLHL20- mediated PML and DAPK ubiquitination. Oncogene 125. Wei, S., Kozono, S., Kats, L., Nechama, M., Li, W., Guarnerio, J., Luo, M., You, M. H., Yao, Y., Kondo, A., Hu, H., Bozkurt, G., Moerke, N. J., Cao, S., Reschke, M., Chen, C. H., Rego, E. M., Lo-Coco, F., Cantley, L. C., Lee, T. H., Wu, H., Zhang, Y., Pandolfi, P. P., Zhou, X. Z., and Lu, K. P. (2015) Active Pin1 is a key target of all-trans retinoic acid in acute promyelocytic leukemia and breast cancer. Nat Med 21, 457-466 126. Lin, Y. C., Lu, L. T., Chen, H. Y., Duan, X., Lin, X., Feng, X. H., Tang, M. J., and Chen, R. H. (2014) SCP phosphatases suppress renal cell carcinoma by stabilizing PML and inhibiting mTOR/HIF signaling. Cancer Res 74, 6935-6946 127. Wu, H. C., Lin, Y. C., Liu, C. H., Chung, H. C., Wang, Y. T., Lin, Y. W., Ma, H. I., Tu, P. H., Lawler, S. E., and Chen, R. H. (2014) USP11 regulates PML stability to control Notch- induced malignancy in brain tumours. Nat Commun 5, 3214 128. Fu, C., Ahmed, K., Ding, H., Ding, X., Lan, J., Yang, Z., Miao, Y., Zhu, Y., Shi, Y., Zhu, J., Huang, H., and Yao, X. (2005) Stabilization of PML nuclear localization by conjugation and oligomerization of SUMO-3. Oncogene 24, 5401-5413

158

129. Campagna, M., Herranz, D., Garcia, M. A., Marcos-Villar, L., Gonzalez-Santamaria, J., Gallego, P., Gutierrez, S., Collado, M., Serrano, M., Esteban, M., and Rivas, C. (2011) SIRT1 stabilizes PML promoting its sumoylation. Cell Death Differ 18, 72-79 130. Peche, L. Y., Scolz, M., Ladelfa, M. F., Monte, M., and Schneider, C. (2012) MageA2 restrains cellular senescence by targeting the function of PMLIV/p53 axis at the PML- NBs. Cell Death Differ 19, 926-936 131. Rabellino, A., Carter, B., Konstantinidou, G., Wu, S. Y., Rimessi, A., Byers, L. A., Heymach, J. V., Girard, L., Chiang, C. M., Teruya-Feldstein, J., and Scaglioni, P. P. (2012) The SUMO E3-ligase PIAS1 regulates the tumor suppressor PML and its oncogenic counterpart PML- RARA. Cancer Res 72, 2275-2284 132. Gao, C., Ho, C. C., Reineke, E., Lam, M., Cheng, X., Stanya, K. J., Liu, Y., Chakraborty, S., Shih, H. M., and Kao, H. Y. (2008) Histone deacetylase 7 promotes PML sumoylation and is essential for PML nuclear body formation. Mol Cell Biol 28, 5658-5667 133. Tatham, M. H., Kim, S., Jaffray, E., Song, J., Chen, Y., and Hay, R. T. (2005) Unique binding interactions among Ubc9, SUMO and RanBP2 reveal a mechanism for SUMO paralog selection. Nat Struct Mol Biol 12, 67-74 134. Huang, M. E., Ye, Y. C., Chen, S. R., Chai, J. R., Lu, J. X., Zhoa, L., Gu, L. J., and Wang, Z. Y. (1988) Use of all-trans retinoic acid in the treatment of acute promyelocytic leukemia. Blood 72, 567-572 135. Chen, G. Q., Zhu, J., Shi, X. G., Ni, J. H., Zhong, H. J., Si, G. Y., Jin, X. L., Tang, W., Li, X. S., Xong, S. M., Shen, Z. X., Sun, G. L., Ma, J., Zhang, P., Zhang, T. D., Gazin, C., Naoe, T., Chen, S. J., Wang, Z. Y., and Chen, Z. (1996) In vitro studies on cellular and molecular mechanisms of arsenic trioxide (As2O3) in the treatment of acute promyelocytic leukemia: As2O3 induces NB4 cell apoptosis with downregulation of Bcl-2 expression and modulation of PML-RAR alpha/PML proteins. Blood 88, 1052-1061 136. Zhang, S. Y., Zhu, J., Chen, G. Q., Du, X. X., Lu, L. J., Zhang, Z., Zhong, H. J., Chen, H. R., Wang, Z. Y., Berger, R., Lanotte, M., Waxman, S., Chen, Z., and Chen, S. J. (1996) Establishment of a human acute promyelocytic leukemia-ascites model in SCID mice. Blood 87, 3404-3409 137. de The, H., and Chen, Z. (2010) Acute promyelocytic leukaemia: novel insights into the mechanisms of cure. Nat Rev Cancer 10, 775-783 138. Zhang, X. W., Yan, X. J., Zhou, Z. R., Yang, F. F., Wu, Z. Y., Sun, H. B., Liang, W. X., Song, A. X., Lallemand-Breitenbach, V., Jeanne, M., Zhang, Q. Y., Yang, H. Y., Huang, Q. H., Zhou, G. B., Tong, J. H., Zhang, Y., Wu, J. H., Hu, H. Y., de The, H., Chen, S. J., and Chen, Z. (2010) Arsenic trioxide controls the fate of the PML-RARalpha oncoprotein by directly binding PML. Science 328, 240-243 139. Reineke, E. L., Liu, Y., and Kao, H. Y. (2010) Promyelocytic leukemia protein controls cell migration in response to hydrogen peroxide and insulin-like growth factor-1. J Biol Chem 285, 9485-9492 140. Reineke, E. L., Lam, M., Liu, Q., Liu, Y., Stanya, K. J., Chang, K. S., Means, A. R., and Kao, H. Y. (2008) Degradation of the tumor suppressor PML by Pin1 contributes to the cancer phenotype of breast cancer MDA-MB-231 cells. Mol Cell Biol 28, 997-1006 141. Lim, J. H., Liu, Y., Reineke, E., and Kao, H. Y. (2011) Mitogen-activated protein kinase extracellular signal-regulated kinase 2 phosphorylates and promotes Pin1 protein- dependent promyelocytic leukemia protein turnover. J Biol Chem 286, 44403-44411 142. Scaglioni, P. P., Yung, T. M., Cai, L. F., Erdjument-Bromage, H., Kaufman, A. J., Singh, B., Teruya-Feldstein, J., Tempst, P., and Pandolfi, P. P. (2006) A CK2-dependent mechanism for degradation of the PML tumor suppressor. Cell 126, 269-283

159

143. Yang, Q., Deng, X., Lu, B., Cameron, M., Fearns, C., Patricelli, M. P., Yates, J. R., 3rd, Gray, N. S., and Lee, J. D. (2010) Pharmacological inhibition of BMK1 suppresses tumor growth through promyelocytic leukemia protein. Cancer Cell 18, 258-267 144. Gresko, E., Ritterhoff, S., Sevilla-Perez, J., Roscic, A., Frobius, K., Kotevic, I., Vichalkovski, A., Hess, D., Hemmings, B. A., and Schmitz, M. L. (2009) PML tumor suppressor is regulated by HIPK2-mediated phosphorylation in response to DNA damage. Oncogene 28, 698-708 145. Hayakawa, F., Abe, A., Kitabayashi, I., Pandolfi, P. P., and Naoe, T. (2008) Acetylation of PML is involved in histone deacetylase inhibitor-mediated apoptosis. J Biol Chem 283, 24420-24425 146. Guo, S., Cheng, X., Lim, J. H., Liu, Y., and Kao, H. Y. (2014) Control of antioxidative response by the tumor suppressor protein PML through regulating Nrf2 activity. Mol Biol Cell 147. Henderson, B. R., and Eleftheriou, A. (2000) A comparison of the activity, sequence specificity, and CRM1-dependence of different nuclear export signals. Exp Cell Res 256, 213-224 148. Negrette-Guzman, M., Huerta-Yepez, S., Tapia, E., and Pedraza-Chaverri, J. (2013) Modulation of mitochondrial functions by the indirect antioxidant sulforaphane: a seemingly contradictory dual role and an integrative hypothesis. Free Radic Biol Med 65, 1078-1089 149. Zhang, Y., Talalay, P., Cho, C. G., and Posner, G. H. (1992) A major inducer of anticarcinogenic protective enzymes from broccoli: isolation and elucidation of structure. Proc Natl Acad Sci U S A 89, 2399-2403 150. Juge, N., Mithen, R. F., and Traka, M. (2007) Molecular basis for chemoprevention by sulforaphane: a comprehensive review. Cell Mol Life Sci 64, 1105-1127 151. Dinkova-Kostova, A. T., and Kostov, R. V. (2012) Glucosinolates and isothiocyanates in health and disease. Trends Mol Med 18, 337-347 152. Morey, L., Brenner, C., Fazi, F., Villa, R., Gutierrez, A., Buschbeck, M., Nervi, C., Minucci, S., Fuks, F., and Di Croce, L. (2008) MBD3, a component of the NuRD complex, facilitates chromatin alteration and deposition of epigenetic marks. Mol Cell Biol 28, 5912-5923 153. Wu, W. S., Vallian, S., Seto, E., Yang, W. M., Edmondson, D., Roth, S., and Chang, K. S. (2001) The growth suppressor PML represses transcription by functionally and physically interacting with histone deacetylases. Mol Cell Biol 21, 2259-2268 154. Cho, S., Park, J. S., and Kang, Y. K. (2011) Dual functions of histone-lysine N- methyltransferase Setdb1 protein at promyelocytic leukemia-nuclear body (PML-NB): maintaining PML-NB structure and regulating the expression of its associated genes. J Biol Chem 286, 41115-41124 155. Carbone, R., Botrugno, O. A., Ronzoni, S., Insinga, A., Di Croce, L., Pelicci, P. G., and Minucci, S. (2006) Recruitment of the histone methyltransferase SUV39H1 and its role in the oncogenic properties of the leukemia-associated PML-retinoic acid receptor fusion protein. Mol Cell Biol 26, 1288-1296 156. Villa, R., Pasini, D., Gutierrez, A., Morey, L., Occhionorelli, M., Vire, E., Nomdedeu, J. F., Jenuwein, T., Pelicci, P. G., Minucci, S., Fuks, F., Helin, K., and Di Croce, L. (2007) Role of the polycomb repressive complex 2 in acute promyelocytic leukemia. Cancer Cell 11, 513-525 157. Wolyniec, K., Carney, D. A., Haupt, S., and Haupt, Y. (2013) New Strategies to Direct Therapeutic Targeting of PML to Treat Cancers. Front Oncol 3, 124

160

158. Sahin, U., Ferhi, O., Jeanne, M., Benhenda, S., Berthier, C., Jollivet, F., Niwa-Kawakita, M., Faklaris, O., Setterblad, N., de The, H., and Lallemand-Breitenbach, V. (2014) Oxidative stress-induced assembly of PML nuclear bodies controls sumoylation of partner proteins. J Cell Biol 204, 931-945 159. Ito, K., Bernardi, R., Morotti, A., Matsuoka, S., Saglio, G., Ikeda, Y., Rosenblatt, J., Avigan, D. E., Teruya-Feldstein, J., and Pandolfi, P. P. (2008) PML targeting eradicates quiescent leukaemia-initiating cells. Nature 453, 1072-1078 160. Carracedo, A., Weiss, D., Leliaert, A. K., Bhasin, M., de Boer, V. C., Laurent, G., Adams, A. C., Sundvall, M., Song, S. J., Ito, K., Finley, L. S., Egia, A., Libermann, T., Gerhart-Hines, Z., Puigserver, P., Haigis, M. C., Maratos-Flier, E., Richardson, A. L., Schafer, Z. T., and Pandolfi, P. P. (2012) A metabolic prosurvival role for PML in breast cancer. J Clin Invest 122, 3088-3100 161. Cheng, X., Guo, S., Liu, Y., Chu, H., Hakimi, P., Berger, N. A., Hanson, R. W., and Kao, H. Y. (2013) Ablation of promyelocytic leukemia protein (PML) re-patterns energy balance and protects mice from obesity induced by a western diet. J Biol Chem 162. Arita, K., Ariyoshi, M., Tochio, H., Nakamura, Y., and Shirakawa, M. (2008) Recognition of hemi-methylated DNA by the SRA protein UHRF1 by a base-flipping mechanism. Nature 455, 818-821 163. Avvakumov, G. V., Walker, J. R., Xue, S., Li, Y., Duan, S., Bronner, C., Arrowsmith, C. H., and Dhe-Paganon, S. (2008) Structural basis for recognition of hemi-methylated DNA by the SRA domain of human UHRF1. Nature 455, 822-825 164. Hashimoto, H., Horton, J. R., Zhang, X., Bostick, M., Jacobsen, S. E., and Cheng, X. (2008) The SRA domain of UHRF1 flips 5-methylcytosine out of the DNA helix. Nature 455, 826- 829 165. Bostick, M., Kim, J. K., Esteve, P. O., Clark, A., Pradhan, S., and Jacobsen, S. E. (2007) UHRF1 plays a role in maintaining DNA methylation in mammalian cells. Science 317, 1760-1764 166. Unoki, M., Nishidate, T., and Nakamura, Y. (2004) ICBP90, an E2F-1 target, recruits HDAC1 and binds to methyl-CpG through its SRA domain. Oncogene 23, 7601-7610 167. Hu, L., Li, Z., Wang, P., Lin, Y., and Xu, Y. (2011) Crystal structure of PHD domain of UHRF1 and insights into recognition of unmodified histone H3 arginine residue 2. Cell Res 168. Xie, S., Jakoncic, J., and Qian, C. (2011) UHRF1 Double Tudor Domain and the Adjacent PHD Finger Act Together to Recognize K9me3-Containing Histone H3 Tail. J Mol Biol 169. Citterio, E., Papait, R., Nicassio, F., Vecchi, M., Gomiero, P., Mantovani, R., Di Fiore, P. P., and Bonapace, I. M. (2004) Np95 is a histone-binding protein endowed with ubiquitin ligase activity. Mol Cell Biol 24, 2526-2535 170. Yan, F., Tan, X. Y., Geng, Y., Ju, H. X., Gao, Y. F., and Zhu, M. C. (2011) Inhibition Effect of siRNA-Downregulated UHRF1 on Breast Cancer Growth. Cancer Biother Radiopharm 26, 183-189 171. Li, X., Meng, Q., Rosen, E. M., and Fan, S. (2011) UHRF1 confers radioresistance to human breast cancer cells. Int J Radiat Biol 87, 263-273 172. Unoki, M., Daigo, Y., Koinuma, J., Tsuchiya, E., Hamamoto, R., and Nakamura, Y. (2010) UHRF1 is a novel diagnostic marker of lung cancer. Br J Cancer 103, 217-222 173. Daskalos, A., Oleksiewicz, U., Filia, A., Nikolaidis, G., Xinarianos, G., Gosney, J. R., Malliri, A., Field, J. K., and Liloglou, T. (2010) UHRF1-mediated tumor suppressor gene inactivation in nonsmall cell lung cancer. Cancer

161

174. Wang, F., Yang, Y. Z., Shi, C. Z., Zhang, P., Moyer, M. P., Zhang, H. Z., Zou, Y., and Qin, H. L. (2012) UHRF1 Promotes Cell Growth and Metastasis Through Repression of p16(ink4a) in Colorectal Cancer. Ann Surg Oncol 175. Du, Z., Song, J., Wang, Y., Zhao, Y., Guda, K., Yang, S., Kao, H. Y., Xu, Y., Willis, J., Markowitz, S. D., Sedwick, D., Ewing, R. M., and Wang, Z. (2010) DNMT1 stability is regulated by proteins coordinating deubiquitination and acetylation-driven ubiquitination. Sci Signal 3, ra80 176. Sabatino, L., Fucci, A., Pancione, M., Carafa, V., Nebbioso, A., Pistore, C., Babbio, F., Votino, C., Laudanna, C., Ceccarelli, M., Altucci, L., Bonapace, I. M., and Colantuoni, V. (2012) UHRF1 coordinates peroxisome proliferator activated receptor gamma (PPARG) epigenetic silencing and mediates colorectal cancer progression. Oncogene 177. Jenkins, Y., Markovtsov, V., Lang, W., Sharma, P., Pearsall, D., Warner, J., Franci, C., Huang, B., Huang, J., Yam, G. C., Vistan, J. P., Pali, E., Vialard, J., Janicot, M., Lorens, J. B., Payan, D. G., and Hitoshi, Y. (2005) Critical role of the ubiquitin ligase activity of UHRF1, a nuclear RING finger protein, in tumor cell growth. Mol Biol Cell 16, 5621-5629 178. Babbio, F., Pistore, C., Curti, L., Castiglioni, I., Kunderfranco, P., Brino, L., Oudet, P., Seiler, R., Thalman, G. N., Roggero, E., Sarti, M., Pinton, S., Mello-Grand, M., Chiorino, G., Catapano, C. V., Carbone, G. M., and Bonapace, I. M. (2012) The SRA protein UHRF1 promotes epigenetic crosstalks and is involved in prostate cancer progression. Oncogene 179. Unoki, M., Kelly, J. D., Neal, D. E., Ponder, B. A., Nakamura, Y., and Hamamoto, R. (2009) UHRF1 is a novel molecular marker for diagnosis and the prognosis of bladder cancer. Br J Cancer 101, 98-105 180. Yang, G. L., Zhang, L. H., Bo, J. J., Chen, H. G., Cao, M., Liu, D. M., and Huang, Y. R. (2011) UHRF1 is associated with tumor recurrence in non-muscle-invasive bladder cancer. Med Oncol 181. Un, F., Qi, C., Prosser, M., Wang, N., Zhou, B., Bronner, C., and Yen, Y. (2006) Modulating ICBP90 to suppress human ribonucleotide reductase M2 induction restores sensitivity to hydroxyurea cytotoxicity. Anticancer Res 26, 2761-2767 182. Jin, W., Chen, L., Chen, Y., Xu, S. G., Di, G. H., Yin, W. J., Wu, J., and Shao, Z. M. (2010) UHRF1 is associated with epigenetic silencing of BRCA1 in sporadic breast cancer. Breast Cancer Res Treat 123, 359-373 183. Alhosin, M., Sharif, T., Mousli, M., Etienne-Selloum, N., Fuhrmann, G., Schni-Kerth, V. B., and Bronner, C. (2011) Down-regulation of UHRF1, associated with re-expression of tumor suppressor genes, is a common feature of natural compounds exhibiting anti- cancer properties. J Exp Clin Cancer Res 30, 41 184. Jeanblanc, M., Mousli, M., Hopfner, R., Bathami, K., Martinet, N., Abbady, A. Q., Siffert, J. C., Mathieu, E., Muller, C. D., and Bronner, C. (2005) The retinoblastoma gene and its product are targeted by ICBP90: a key mechanism in the G1/S transition during the cell cycle. Oncogene 24, 7337-7345 185. Kim, J. K., Esteve, P. O., Jacobsen, S. E., and Pradhan, S. (2009) UHRF1 binds G9a and participates in p21 transcriptional regulation in mammalian cells. Nucleic Acids Res 37, 493-505 186. de The, H., Chomienne, C., Lanotte, M., Degos, L., and Dejean, A. (1990) The t(15;17) translocation of acute promyelocytic leukaemia fuses the retinoic acid receptor alpha gene to a novel transcribed locus. Nature 347, 558-561

162

187. Boutell, C., Canning, M., Orr, A., and Everett, R. D. (2005) Reciprocal activities between herpes simplex virus type 1 regulatory protein ICP0, a ubiquitin E3 ligase, and ubiquitin- specific protease USP7. J Virol 79, 12342-12354 188. Khurana, S., Chakraborty, S., Cheng, X., Su, Y. T., and Kao, H. Y. (2011) The actin-binding protein, actinin alpha 4 (ACTN4), is a nuclear receptor coactivator that promotes proliferation of MCF-7 breast cancer cells. J Biol Chem 286, 1850-1859 189. Leng, R. P., Lin, Y., Ma, W., Wu, H., Lemmers, B., Chung, S., Parant, J. M., Lozano, G., Hakem, R., and Benchimol, S. (2003) Pirh2, a p53-induced ubiquitin-protein ligase, promotes p53 degradation. Cell 112, 779-791 190. Felle, M., Joppien, S., Nemeth, A., Diermeier, S., Thalhammer, V., Dobner, T., Kremmer, E., Kappler, R., and Langst, G. (2011) The USP7/Dnmt1 complex stimulates the DNA methylation activity of Dnmt1 and regulates the stability of UHRF1. Nucleic Acids Res 191. Ma, H., Chen, H., Guo, X., Wang, Z., Sowa, M. E., Zheng, L., Hu, S., Zeng, P., Guo, R., Diao, J., Lan, F., Harper, J. W., Shi, Y. G., Xu, Y., and Shi, Y. (2012) M phase phosphorylation of the epigenetic regulator UHRF1 regulates its physical association with the deubiquitylase USP7 and stability. Proc Natl Acad Sci U S A 109, 4828-4833 192. van Wijk, S. J., de Vries, S. J., Kemmeren, P., Huang, A., Boelens, R., Bonvin, A. M., and Timmers, H. T. (2009) A comprehensive framework of E2-RING E3 interactions of the human ubiquitin-proteasome system. Mol Syst Biol 5, 295 193. Rajakumara, E., Wang, Z., Ma, H., Hu, L., Chen, H., Lin, Y., Guo, R., Wu, F., Li, H., Lan, F., Shi, Y. G., Xu, Y., Patel, D. J., and Shi, Y. (2011) PHD Finger Recognition of Unmodified Histone H3R2 Links UHRF1 to Regulation of Euchromatic Gene Expression. Mol Cell 43, 275-284 194. Nady, N., Lemak, A., Walker, J. R., Avvakumov, G. V., Kareta, M. S., Achour, M., Xue, S., Duan, S., Allali-Hassani, A., Zuo, X., Wang, Y. X., Bronner, C., Chedin, F., Arrowsmith, C. H., and Dhe-Paganon, S. (2011) Recognition of multivalent histone states associated with heterochromatin by UHRF1. J Biol Chem 195. Wang, D. Y., Done, S. J., McCready, D. R., Boerner, S., Kulkarni, S., and Leong, W. L. (2011) A new gene expression signature, the ClinicoMolecular Triad Classification, may improve prediction and prognostication of breast cancer at the time of diagnosis. Breast Cancer Res 13, R92 196. Unoki, M., Brunet, J., and Mousli, M. (2009) Drug discovery targeting epigenetic codes: the great potential of UHRF1, which links DNA methylation and histone modifications, as a drug target in cancers and toxoplasmosis. Biochem Pharmacol 78, 1279-1288 197. Pearson, M., Carbone, R., Sebastiani, C., Cioce, M., Fagioli, M., Saito, S., Higashimoto, Y., Appella, E., Minucci, S., Pandolfi, P. P., and Pelicci, P. G. (2000) PML regulates p53 acetylation and premature senescence induced by oncogenic Ras. Nature 406, 207-210 198. Cheng, X., Liu, Y., Chu, H., and Kao, H. Y. (2012) Promyelocytic leukemia protein (PML) regulates endothelial cell network formation and migration in response to tumor necrosis factor alpha (TNFalpha) and interferon alpha (IFNalpha). J Biol Chem 199. Sarkari, F., Wang, X., Nguyen, T., and Frappier, L. (2011) The herpesvirus associated ubiquitin specific protease, USP7, is a negative regulator of PML proteins and PML nuclear bodies. PLoS One 6, e16598 200. Canning, M., Boutell, C., Parkinson, J., and Everett, R. D. (2004) A RING finger ubiquitin ligase is protected from autocatalyzed ubiquitination and degradation by binding to ubiquitin-specific protease USP7. J Biol Chem 279, 38160-38168 201. Ascoli, C. A., and Maul, G. G. (1991) Identification of a novel nuclear domain. J Cell Biol 112, 785-795

163

202. Yang, X. J., and Seto, E. (2008) Lysine acetylation: codified crosstalk with other posttranslational modifications. Mol Cell 31, 449-461 203. Choudhary, C., Kumar, C., Gnad, F., Nielsen, M. L., Rehman, M., Walther, T. C., Olsen, J. V., and Mann, M. (2009) Lysine acetylation targets protein complexes and co-regulates major cellular functions. Science 325, 834-840 204. Saunders, L. R., and Verdin, E. (2007) Sirtuins: critical regulators at the crossroads between cancer and aging. Oncogene 26, 5489-5504 205. Finkel, T., Deng, C. X., and Mostoslavsky, R. (2009) Recent progress in the biology and physiology of sirtuins. Nature 460, 587-591 206. Guarente, L. (2011) Franklin H. Epstein Lecture: Sirtuins, aging, and medicine. N Engl J Med 364, 2235-2244 207. Herranz, D., and Serrano, M. (2010) SIRT1: recent lessons from mouse models. Nat Rev Cancer 10, 819-823 208. Song, N. Y., and Surh, Y. J. (2012) Janus-faced role of SIRT1 in tumorigenesis. Ann N Y Acad Sci 1271, 10-19 209. Deng, C. X. (2009) SIRT1, is it a tumor promoter or tumor suppressor? Int J Biol Sci 5, 147-152 210. Stunkel, W., Peh, B. K., Tan, Y. C., Nayagam, V. M., Wang, X., Salto-Tellez, M., Ni, B., Entzeroth, M., and Wood, J. (2007) Function of the SIRT1 protein deacetylase in cancer. Biotechnol J 2, 1360-1368 211. Byles, V., Chmilewski, L. K., Wang, J., Zhu, L., Forman, L. W., Faller, D. V., and Dai, Y. (2010) Aberrant cytoplasm localization and protein stability of SIRT1 is regulated by PI3K/IGF-1R signaling in human cancer cells. Int J Biol Sci 6, 599-612 212. Lim, J. H., Lee, Y. M., Chun, Y. S., Chen, J., Kim, J. E., and Park, J. W. (2010) Sirtuin 1 modulates cellular responses to hypoxia by deacetylating hypoxia-inducible factor 1alpha. Mol Cell 38, 864-878 213. Yeung, F., Hoberg, J. E., Ramsey, C. S., Keller, M. D., Jones, D. R., Frye, R. A., and Mayo, M. W. (2004) Modulation of NF-kappaB-dependent transcription and cell survival by the SIRT1 deacetylase. EMBO J 23, 2369-2380 214. Nie, Y., Erion, D. M., Yuan, Z., Dietrich, M., Shulman, G. I., Horvath, T. L., and Gao, Q. (2009) STAT3 inhibition of gluconeogenesis is downregulated by SirT1. Nat Cell Biol 11, 492-500 215. Yuan, Z., Zhang, X., Sengupta, N., Lane, W. S., and Seto, E. (2007) SIRT1 regulates the function of the Nijmegen breakage syndrome protein. Mol Cell 27, 149-162 216. Rajamohan, S. B., Pillai, V. B., Gupta, M., Sundaresan, N. R., Birukov, K. G., Samant, S., Hottiger, M. O., and Gupta, M. P. (2009) SIRT1 promotes cell survival under stress by deacetylation-dependent deactivation of poly(ADP-ribose) polymerase 1. Mol Cell Biol 29, 4116-4129 217. Ming, M., Shea, C. R., Guo, X., Li, X., Soltani, K., Han, W., and He, Y. Y. (2010) Regulation of global genome nucleotide excision repair by SIRT1 through xeroderma pigmentosum C. Proc Natl Acad Sci U S A 107, 22623-22628 218. Sundaresan, N. R., Pillai, V. B., Wolfgeher, D., Samant, S., Vasudevan, P., Parekh, V., Raghuraman, H., Cunningham, J. M., Gupta, M., and Gupta, M. P. (2011) The deacetylase SIRT1 promotes membrane localization and activation of Akt and PDK1 during tumorigenesis and cardiac hypertrophy. Sci Signal 4, ra46 219. Menssen, A., Hydbring, P., Kapelle, K., Vervoorts, J., Diebold, J., Luscher, B., Larsson, L. G., and Hermeking, H. (2012) The c-MYC oncoprotein, the NAMPT enzyme, the SIRT1-

164

inhibitor DBC1, and the SIRT1 deacetylase form a positive feedback loop. Proc Natl Acad Sci U S A 109, E187-196 220. Wong, S., and Weber, J. D. (2007) Deacetylation of the retinoblastoma tumour suppressor protein by SIRT1. Biochem J 407, 451-460 221. Pickard, A., Wong, P. P., and McCance, D. J. (2010) Acetylation of Rb by PCAF is required for nuclear localization and keratinocyte differentiation. J Cell Sci 123, 3718-3726 222. Ikenoue, T., Inoki, K., Zhao, B., and Guan, K. L. (2008) PTEN acetylation modulates its interaction with PDZ domain. Cancer Res 68, 6908-6912 223. Nakagawa, T., Lomb, D. J., Haigis, M. C., and Guarente, L. (2009) SIRT5 Deacetylates carbamoyl phosphate synthetase 1 and regulates the urea cycle. Cell 137, 560-570 224. Lai, C. C., Lin, P. M., Lin, S. F., Hsu, C. H., Lin, H. C., Hu, M. L., Hsu, C. M., and Yang, M. Y. (2013) Altered expression of SIRT gene family in head and neck squamous cell carcinoma. Tumour Biol 34, 1847-1854 225. Peng, L., Ling, H., Yuan, Z., Fang, B., Bloom, G., Fukasawa, K., Koomen, J., Chen, J., Lane, W. S., and Seto, E. (2012) SIRT1 negatively regulates the activities, functions, and protein levels of hMOF and TIP60. Mol Cell Biol 32, 2823-2836 226. O'Hagan, H. M., Wang, W., Sen, S., Destefano Shields, C., Lee, S. S., Zhang, Y. W., Clements, E. G., Cai, Y., Van Neste, L., Easwaran, H., Casero, R. A., Sears, C. L., and Baylin, S. B. (2011) Oxidative damage targets complexes containing DNA methyltransferases, SIRT1, and polycomb members to promoter CpG Islands. Cancer Cell 20, 606-619 227. Rhee, S. G. (2006) Cell signaling. H2O2, a necessary evil for cell signaling. Science 312, 1882-1883 228. Ren, Q., and Gorovsky, M. A. (2001) Histone H2A.Z acetylation modulates an essential charge patch. Mol Cell 7, 1329-1335 229. Megee, P. C., Morgan, B. A., Mittman, B. A., and Smith, M. M. (1990) Genetic analysis of histone H4: essential role of lysines subject to reversible acetylation. Science 247, 841- 845 230. Le, X. F., Yang, P., and Chang, K. S. (1996) Analysis of the growth and transformation suppressor domains of promyelocytic leukemia gene, PML. J Biol Chem 271, 130-135 231. Li, T., Diner, B. A., Chen, J., and Cristea, I. M. (2012) Acetylation modulates cellular distribution and DNA sensing ability of interferon-inducible protein IFI16. Proc Natl Acad Sci U S A 109, 10558-10563 232. de la Vega, L., Grishina, I., Moreno, R., Kruger, M., Braun, T., and Schmitz, M. L. (2012) A redox-regulated SUMO/acetylation switch of HIPK2 controls the survival threshold to oxidative stress. Mol Cell 46, 472-483 233. Nathan, D., Ingvarsdottir, K., Sterner, D. E., Bylebyl, G. R., Dokmanovic, M., Dorsey, J. A., Whelan, K. A., Krsmanovic, M., Lane, W. S., Meluh, P. B., Johnson, E. S., and Berger, S. L. (2006) Histone sumoylation is a negative regulator in Saccharomyces cerevisiae and shows dynamic interplay with positive-acting histone modifications. Genes Dev 20, 966- 976 234. Kotaja, N., Karvonen, U., Janne, O. A., and Palvimo, J. J. (2002) The nuclear receptor interaction domain of GRIP1 is modulated by covalent attachment of SUMO-1. J Biol Chem 277, 30283-30288 235. Rardin, M. J., He, W., Nishida, Y., Newman, J. C., Carrico, C., Danielson, S. R., Guo, A., Gut, P., Sahu, A. K., Li, B., Uppala, R., Fitch, M., Riiff, T., Zhu, L., Zhou, J., Mulhern, D., Stevens, R. D., Ilkayeva, O. R., Newgard, C. B., Jacobson, M. P., Hellerstein, M., Goetzman, E. S., Gibson, B. W., and Verdin, E. (2013) SIRT5 Regulates the Mitochondrial Lysine Succinylome and Metabolic Networks. Cell Metab 18, 920-933

165

236. Geng, Y. Q., Li, T. T., Liu, X. Y., Li, Z. H., and Fu, Y. C. (2011) SIRT1 and SIRT5 activity expression and behavioral responses to calorie restriction. J Cell Biochem 112, 3755- 3761 237. Liu, N., and Olson, E. N. (2010) MicroRNA regulatory networks in cardiovascular development. Dev Cell 18, 510-525 238. Vasudevan, S., Tong, Y., and Steitz, J. A. (2007) Switching from repression to activation: microRNAs can up-regulate translation. Science 318, 1931-1934 239. Lin, C. C., Liu, L. Z., Addison, J. B., Wonderlin, W. F., Ivanov, A. V., and Ruppert, J. M. (2011) A KLF4-miRNA-206 autoregulatory feedback loop can promote or inhibit protein translation depending upon cell context. Mol Cell Biol 31, 2513-2527 240. Cordes, K. R., Sheehy, N. T., White, M. P., Berry, E. C., Morton, S. U., Muth, A. N., Lee, T. H., Miano, J. M., Ivey, K. N., and Srivastava, D. (2009) miR-145 and miR-143 regulate smooth muscle cell fate and plasticity. Nature 460, 705-710 241. Lu, H., Buchan, R. J., and Cook, S. A. (2010) MicroRNA-223 regulates Glut4 expression and cardiomyocyte glucose metabolism. Cardiovasc Res 86, 410-420 242. Truesdell, S. S., Mortensen, R. D., Seo, M., Schroeder, J. C., Lee, J. H., LeTonqueze, O., and Vasudevan, S. (2012) MicroRNA-mediated mRNA translation activation in quiescent cells and oocytes involves recruitment of a nuclear microRNP. Sci Rep 2, 842 243. Tsai, N. P., Lin, Y. L., and Wei, L. N. (2009) MicroRNA mir-346 targets the 5'-untranslated region of receptor-interacting protein 140 (RIP140) mRNA and up-regulates its protein expression. Biochem J 424, 411-418 244. Orom, U. A., Nielsen, F. C., and Lund, A. H. (2008) MicroRNA-10a binds the 5'UTR of ribosomal protein mRNAs and enhances their translation. Mol Cell 30, 460-471 245. Zhang, X., Zuo, X., Yang, B., Li, Z., Xue, Y., Zhou, Y., Huang, J., Zhao, X., Zhou, J., Yan, Y., Zhang, H., Guo, P., Sun, H., Guo, L., Zhang, Y., and Fu, X. D. (2014) MicroRNA directly enhances mitochondrial translation during muscle differentiation. Cell 158, 607-619 246. Qian, L., Van Laake, L. W., Huang, Y., Liu, S., Wendland, M. F., and Srivastava, D. (2011) miR-24 inhibits apoptosis and represses Bim in mouse cardiomyocytes. J Exp Med 208, 549-560 247. Hatziapostolou, M., Polytarchou, C., Aggelidou, E., Drakaki, A., Poultsides, G. A., Jaeger, S. A., Ogata, H., Karin, M., Struhl, K., Hadzopoulou-Cladaras, M., and Iliopoulos, D. (2011) An HNF4alpha-miRNA inflammatory feedback circuit regulates hepatocellular oncogenesis. Cell 147, 1233-1247 248. Xie, Y., Tobin, L. A., Camps, J., Wangsa, D., Yang, J., Rao, M., Witasp, E., Awad, K. S., Yoo, N., Ried, T., and Kwong, K. F. (2013) MicroRNA-24 regulates XIAP to reduce the apoptosis threshold in cancer cells. Oncogene 32, 2442-2451 249. Liu, X., Wang, A., Heidbreder, C. E., Jiang, L., Yu, J., Kolokythas, A., Huang, L., Dai, Y., and Zhou, X. (2010) MicroRNA-24 targeting RNA-binding protein DND1 in tongue squamous cell carcinoma. FEBS Lett 584, 4115-4120 250. Du, W. W., Fang, L., Li, M., Yang, X., Liang, Y., Peng, C., Qian, W., O'Malley, Y. Q., Askeland, R. W., Sugg, S. L., Qian, J., Lin, J., Jiang, Z., Yee, A. J., Sefton, M., Deng, Z., Shan, S. W., Wang, C. H., and Yang, B. B. (2013) MicroRNA miR-24 enhances tumor invasion and metastasis by targeting PTPN9 and PTPRF to promote EGF signaling. J Cell Sci 126, 1440-1453 251. Lal, A., Navarro, F., Maher, C. A., Maliszewski, L. E., Yan, N., O'Day, E., Chowdhury, D., Dykxhoorn, D. M., Tsai, P., Hofmann, O., Becker, K. G., Gorospe, M., Hide, W., and Lieberman, J. (2009) miR-24 Inhibits cell proliferation by targeting E2F2, MYC, and other

166

cell-cycle genes via binding to "seedless" 3'UTR microRNA recognition elements. Mol Cell 35, 610-625 252. Kasza, Z., Fredlund Fuchs, P., Tamm, C., Eriksson, A. S., O'Callaghan, P., Heindryckx, F., Spillmann, D., Larsson, E., Le Jan, S., Eriksson, I., Gerwins, P., Kjellen, L., and Kreuger, J. (2013) MicroRNA-24 suppression of N-deacetylase/N-sulfotransferase-1 (NDST1) reduces endothelial cell responsiveness to vascular endothelial growth factor A (VEGFA). J Biol Chem 288, 25956-25963 253. Fiedler, J., Jazbutyte, V., Kirchmaier, B. C., Gupta, S. K., Lorenzen, J., Hartmann, D., Galuppo, P., Kneitz, S., Pena, J. T., Sohn-Lee, C., Loyer, X., Soutschek, J., Brand, T., Tuschl, T., Heineke, J., Martin, U., Schulte-Merker, S., Ertl, G., Engelhardt, S., Bauersachs, J., and Thum, T. (2011) MicroRNA-24 regulates vascularity after myocardial infarction. Circulation 124, 720-730 254. Latronico, M. V., and Condorelli, G. (2009) MicroRNAs and cardiac pathology. Nat Rev Cardiol 6, 419-429 255. Wang, X., and McManus, M. (2009) Lentivirus production. J Vis Exp 256. Han, Y., Yang, K., Proweller, A., Zhou, G., Jain, M. K., and Ramirez-Bergeron, D. L. (2012) Inhibition of ARNT severely compromises endothelial cell viability and function in response to moderate hypoxia. Angiogenesis 15, 409-420 257. Vasudevan, S., and Steitz, J. A. (2007) AU-rich-element-mediated upregulation of translation by FXR1 and Argonaute 2. Cell 128, 1105-1118 258. Leong, K. G., Hu, X., Li, L., Noseda, M., Larrivee, B., Hull, C., Hood, L., Wong, F., and Karsan, A. (2002) Activated Notch4 inhibits angiogenesis: role of beta 1-integrin activation. Mol Cell Biol 22, 2830-2841 259. Stamatovic, S. M., Keep, R. F., Mostarica-Stojkovic, M., and Andjelkovic, A. V. (2006) CCL2 regulates angiogenesis via activation of Ets-1 transcription factor. J Immunol 177, 2651-2661 260. Geis, T., Doring, C., Popp, R., Grossmann, N., Fleming, I., Hansmann, M. L., Dehne, N., and Brune, B. (2015) HIF-2alpha-dependent PAI-1 induction contributes to angiogenesis in hepatocellular carcinoma. Exp Cell Res 331, 46-57 261. Takeda, N., Maemura, K., Imai, Y., Harada, T., Kawanami, D., Nojiri, T., Manabe, I., and Nagai, R. (2004) Endothelial PAS domain protein 1 gene promotes angiogenesis through the transactivation of both vascular endothelial growth factor and its receptor, Flt-1. Circ Res 95, 146-153 262. Sharma, S. B., Lin, C. C., Farrugia, M. K., McLaughlin, S. L., Ellis, E. J., Brundage, K. M., Salkeni, M. A., and Ruppert, J. M. (2014) MicroRNAs 206 and 21 cooperate to promote RAS-extracellular signal-regulated kinase signaling by suppressing the translation of RASA1 and SPRED1. Mol Cell Biol 34, 4143-4164 263. Wang, Q., Huang, Z., Xue, H., Jin, C., Ju, X. L., Han, J. D., and Chen, Y. G. (2008) MicroRNA miR-24 inhibits erythropoiesis by targeting activin type I receptor ALK4. Blood 111, 588- 595 264. Amelio, I., Lena, A. M., Viticchie, G., Shalom-Feuerstein, R., Terrinoni, A., Dinsdale, D., Russo, G., Fortunato, C., Bonanno, E., Spagnoli, L. G., Aberdam, D., Knight, R. A., Candi, E., and Melino, G. (2012) miR-24 triggers epidermal differentiation by controlling actin adhesion and cell migration. J Cell Biol 199, 347-363 265. Mishra, P. J., Humeniuk, R., Longo-Sorbello, G. S., Banerjee, D., and Bertino, J. R. (2007) A miR-24 microRNA binding-site polymorphism in dihydrofolate reductase gene leads to methotrexate resistance. Proc Natl Acad Sci U S A 104, 13513-13518

167

266. Sherr, C. J., and DePinho, R. A. (2000) Cellular senescence: mitotic clock or culture shock? Cell 102, 407-410 267. Dellago, H., Preschitz-Kammerhofer, B., Terlecki-Zaniewicz, L., Schreiner, C., Fortschegger, K., Chang, M. W., Hackl, M., Monteforte, R., Kuhnel, H., Schosserer, M., Gruber, F., Tschachler, E., Scheideler, M., Grillari-Voglauer, R., Grillari, J., and Wieser, M. (2013) High levels of oncomiR-21 contribute to the senescence-induced growth arrest in normal human cells and its knock-down increases the replicative lifespan. Aging Cell 12, 446-458 268. Pearson, M., and Pelicci, P. G. (2001) PML interaction with p53 and its role in apoptosis and replicative senescence. Oncogene 20, 7250-7256 269. Shelton, D. N., Chang, E., Whittier, P. S., Choi, D., and Funk, W. D. (1999) Microarray analysis of replicative senescence. Curr Biol 9, 939-945 270. Kallen, A. N., Zhou, X. B., Xu, J., Qiao, C., Ma, J., Yan, L., Lu, L., Liu, C., Yi, J. S., Zhang, H., Min, W., Bennett, A. M., Gregory, R. I., Ding, Y., and Huang, Y. (2013) The imprinted H19 lncRNA antagonizes let-7 microRNAs. Mol Cell 52, 101-112 271. Keniry, A., Oxley, D., Monnier, P., Kyba, M., Dandolo, L., Smits, G., and Reik, W. (2012) The H19 lincRNA is a developmental reservoir of miR-675 that suppresses growth and Igf1r. Nat Cell Biol 14, 659-665 272. Catchpoole, D., Lam, W. W., Valler, D., Temple, I. K., Joyce, J. A., Reik, W., Schofield, P. N., and Maher, E. R. (1997) Epigenetic modification and uniparental inheritance of H19 in Beckwith-Wiedemann syndrome. J Med Genet 34, 353-359 273. Zhu, M., Chen, Q., Liu, X., Sun, Q., Zhao, X., Deng, R., Wang, Y., Huang, J., Xu, M., Yan, J., and Yu, J. (2014) lncRNA H19/miR-675 axis represses prostate cancer metastasis by targeting TGFBI. FEBS J 281, 3766-3775 274. Yan, L., Zhou, J., Gao, Y., Ghazal, S., Lu, L., Bellone, S., Yang, Y., Liu, N., Zhao, X., Santin, A. D., Taylor, H., and Huang, Y. (2015) Regulation of tumor cell migration and invasion by the H19/let-7 axis is antagonized by metformin-induced DNA methylation. Oncogene 34, 3076-3084 275. Zhou, X., Ye, F., Yin, C., Zhuang, Y., Yue, G., and Zhang, G. (2015) The Interaction Between MiR-141 and lncRNA-H19 in Regulating Cell Proliferation and Migration in Gastric Cancer. Cell Physiol Biochem 36, 1440-1452

168