Identification of Telomere Length Regulators By

IDENTIFICATION OF TELOMERE LENGTH REGULATORS BY

POOLED LIBRARY SCREENING

by Steven Wang

A dissertation submitted to The Johns Hopkins University in conformity with the requirements of the degree of Doctor of Philosophy

Baltimore, Maryland January 2017

Abstract

Telomeres protect chromosome ends from damage (d'Adda di Fagagna et al.,

2003). In the presence of telomerase, telomere length is maintained at an established equilibrium (Greider, 1996). Alterations to this equilibrium cause short telomeres, which manifest as degenerative disease, or long telomeres, which can facilitate initiation and growth of cancer (Stanley and Armanios, 2015). Altering the activity of telomere length regulators has therapeutic application in short and long telomere syndromes (Stanley and

Armanios, 2015). We tested whether elongation of short telomeres, in bone marrow stem and progenitor cells, was sufficient to enhance engraftment. We also identified and characterized new regulators of telomere length using a pooled screening approach.

We transplanted short telomere or wildtype mouse hematopoietic stem and progenitor cells, with telomerase or GFP lentivirus, into lethally irradiated hosts. Short telomere mice, transduced with telomerase, had greater survival and engraftment, compared to short telomere mice transduced with GFP. We found that this effect was not due to changes in mean telomere length, but a reduction in critically short telomeres. This suggests that elongation of the shortest telomeres is sufficient to ameliorate degenerative phenotypes.

We also developed a pooled genetic screen to identify new telomere length regulators. We transduced cells with an shRNA or sgRNA library against human kinases, then isolated short telomere cells by fluorescently labeling telomeres and cell sorting.

Enrichment of shRNA or sgRNA inserts in the short telomere fraction were determined using bowtie2 (Langmead and Salzberg, 2012) and MAGeCK analysis (Li et al., 2014).

We identified BRD4 as a strong positive regulator of telomere length. We also identified CK1, and the MEK/ERK pathway as more moderate positive regulators of length. Inhibition of BRD4, with 4 different BRD4 inhibitors, was sufficient to completely block telomere elongation by telomerase over expression, in a drug dependent manner. BRD4 has pleiotropic roles and is currently being investigated in cancer and inflammatory disease (Devaiah et al., 2016b). We believe shortening telomeres, is one mechanism by which BRD4 inhibition slows cancer proliferation. Telomere length should also be considered when using BRD4 inhibitors for inflammatory disease, particularly in the context of short telomere syndromes.

Thesis Advisor: Carol W. Greider Ph.D.

Thesis Reader: Heng Zhu Ph.D.

iii

Acknowledgements

I feel very fortunate to have been a member of the Greider lab, and am thankful to many people for making my time here an enriching experience.

First, I’d like to thank Carol for her mentorship and support. Carol’s passion for science has created a lab environment where it is a real joy to discuss science, and I’ve benefitted tremendously from it. I would also like to thank her for her support of my career and professional development. She is truly someone who recognizes the importance of education, as much as research.

I’d also like to thank Carla and Margaret for all their help. Many people say they couldn’t have done it without them, but this is probably literally true in my case. Thank you Carla for entertaining all my late night conversations, and Margaret for always bringing a positive perspective when things go wrong.

I’ve had the privilege of being surrounded by many talented and fun graduate students. Thanks to Stella for being a fantastic friend/mentor/life consultant since day one of my rotation. I’d also like to thank Alex for always being down for an excursion for free beer or food. I am indebted to many alumni, including Jon Alder, Sofia Rehermann and Chris Viggiani, who welcomed me to the lab and from whom I learned a lot.

I am grateful to my committee, Heng, Steve, DJ and Jef for helpful discussion and feedback throughout the years. I’d also like to thank Hao Zhang at the Public Health Cell

Sorting Facility and John Weger at UC Riverside for many long and helpful discussions about cell sorting and sequencing.

Lastly, I’d like to thank all my friends and family from Toronto, college and

Baltimore. Thanks for all your support through the years and always making me laugh.

Table of Contents

Abstract ii

Acknowledgements iv

Table of Contents v

List of Tables ix

List of Figures x

Chapter 1 Introduction 1

1.1 Telomeres and telomerase 1

1.2 Telomere length homeostasis 2

1.3 Telomere disease 3

1.4 Role of kinases in telomere length regulation 4

1.5 Measuring telomere length 5

1.6 Screens for telomere length regulators 6

Chapter 2 Restoration of telomerase in mTR-/- mice enhances hematopoietic

stem and progenitor cell engraftment 9

2.1 Introduction 9

2.2 Results and Discussion 11

2.2.1 Bone marrow transplant to test the effect of lentiviral

telomerase restoration on hematopoietic stem and progenitor

cell engraftment 12

2.2.2 Effect of telomerase restoration on hematopoietic stem and

progenitor cell engraftment 13

2.2.3 Effect of telomerase restoration on telomere length 15

2.2.4 Generation of novel lentiviral constructs to restore telomerase

in vivo 18

2.3 Materials and Methods 20

2.3.1 Transgenic Mice 20

2.3.2 Lentivirus production and transduction 21

2.3.3 Bone marrow harvest and collecting lineage depleted cells 22

2.3.4 Bone Marrow Transplant 23

2.3.5 Peripheral blood collection and flow cytometry 24

2.3.6 Cell Sorting 25

2.3.7 Quantitative Real-Time PCR 25

2.3.8 Telomere flow-FISH 26

2.3.9 Telomere Quantitative FISH 27

2.3.10 Telomerase Repeat Amplification Protocol 28

2.3.11 Telomere Restriction Fragment Southern Blot 30

Chapter 3 Identifying novel kinase regulators of telomere length by pooled

screening approach 48

3.1 Introduction 48

3.2 Results and Discussion 49

3.2.1 Identifying cell line for pooled telomere length screening 49

3.2.2 Validating shRNA screening approach 52

3.2.3 Pooled shRNA screening 53

3.2.4 Analysis of sequencing results from pooled shRNA screen 55

3.2.5 Pooled CRISPR Screening approach 58

3.2.2 Analysis of sequencing results from pooled CRISPR screen 60

3.2.3 Summary 61

3.3 Materials and Methods 62

3.3.1 shRNA knockdown 62

3.3.2 qPCR Analysis 62

3.3.3 Telomere flow-FISH analysis 62

3.3.4 Pooled shRNA kinase library 63

3.3.5 Pooled CRISPR kinase library 63

3.3.6 Generating constitutive Cas9 expressing cell line 63

3.3.7 Western Blotting 64

3.3.8 Transduction and tissue culture of transduced cells 65

3.3.9 Telomere flow-FISH cell sorting 65

3.3.10 FACS of cells by telomere length 66

vii

3.3.11 Illumina sample preparation and sequencing 67

3.3.12 shRNA screen sequencing analysis 68

3.3.13 CRISPR screen sequencing analysis 68

Chapter 4 Chemical inhibition of BRD4, CK1 and MEK/ERK block

telomere elongation 81

4.1 Introduction 81

4.2 Results and Discussion 83

4.2.1 Telomeres are rapidly elongated by telomerase overexpression 83

4.2.2 BRD4 inhibition blocks telomere elongation 85

4.2.3 CK1 inhibition blocks telomere elongation 86

4.2.3 MEK1/2 and ERK1/2 inhibition blocks telomere elongation 87

4.2.3 Summary 89

4.3 Materials and Methods 90

4.4.1 Small molecule inhibitors 90

4.4.2 Virus Production and Titering 91

4.4.3 Inhibition of SVA mediated telomere elongation 92

4.4.4 Telomere Southern Blot 92

References 102

Curriculum Vitae 115

viii

List of Tables

Table 2.1 Primer list 46

Table 2.2 Mouse mTR and mTERT lentiviral constructs 47

Table 3.1 HeLa shRNA screen top 20 hits 77

Table 3.2 293FT shRNA screen top 20 hits 78

Table 3.3 CRISPR screen top 20 hits 79

Table 3.3 Primer List 80

List of Figures

Figure 2.1 Experimental design to test effect of mTR expression on

hematopoietic stem and progenitor cell engraftment 30

Figure 2.2 Kaplan-Meier survival curve for primary and secondary transplant 32

Figure 2.3 Donor engraftment in primary transplant 34

Figure 2.4 Donor engraftment in secondary transplant 36

Figure 2.5 mTR expression and telomere length in mTR or FUGW transduced

WT or G4 cells 38

Figure 2.6 Signal free ends and P/Q ratio of GFP+ sorted splenocytes 40

Figure 2.7 Expression of mTR and mTERT compared to mTR alone 42

Figure 2.8 LVT3b lentiviral telomerase construct elongates telomeres and

generates bright GFP 44

Figure 3.1 Selecting cell line to perform pooled telomere length screen 69

Figure 3.2 Knockdown of hTERT or POT1 shortens, or elongates telomeres

respectively, in 293FT cells 71

Figure 3.3 Optimizations for fluorescence activated cell sorting and flow-FISH 73

Figure 3.4 POT1 and hTERT shRNA inserts are recovered in the expected

fractions upon sorting long and short telomere cells 75

Figure 4.1 Inhibition of BRD4 with small molecules blocks SVA mediated

telomere elongation 94

Figure 4.2 Inhibition of BRD4 by JQ1 blocks telomere elongation in a dose

dependent manner 96

Figure 4.3 Inhibition of CK1 by D4476 inhibits telomere elongation in a dose

dependent manner 98

Figure 4.4 Chemical inhibition of ERK1/2 or MEK1/2 partially blocks

telomere elongation 100

xi CHAPTER 1. INTRODUCTION

Chapter 1. Introduction

1.1 Telomeres and telomerase

Eukaryotic cells organize their DNA as linear chromosomes. This creates two important problems. First, the cell must distinguish between sites of DNA damage and free chromosome ends. Second, the cell must prevent loss of DNA due to the end- replication problem, during DNA replication. Telomeres have evolved to solve both of these problems. Telomeres are nucleoprotein structures located at the ends of linear chromosomes that protect the chromosome from damage. They are G rich, repetitive sequences of DNA bound by specific telomere binding proteins (Blackburn and Szostak,

1984). In mammals, the DNA sequence is TTAGGG (Moyzis et al., 1988) and the protein complex is called shelterin (Smogorzewska and de Lange, 2004).

The shelterin complex distinguishes the telomere from DNA damage, and negatively regulates the binding of proteins that signal DNA damage (Denchi and de

Lange, 2007). When this complex is deregulated or abolished, telomeres are recognized as double-stranded DNA breaks (Denchi and de Lange, 2007). This leads to a DNA damage signal cascade, which can cause telomeric fusions, cellular senescence and apoptosis (d'Adda di Fagagna et al., 2003; Harley et al., 1990; IJpma and Greider, 2003;

Lundblad and Szostak, 1989).

Telomeres are elongated by telomerase (Greider and Blackburn, 1985).

Telomerase is an enzyme, whose primary components include TERT, the protein catalytic component, and TR, the RNA template component (Greider and Blackburn,

1 CHAPTER 1. INTRODUCTION

1989; Lingner et al., 1997). Telomerase solves the end replication problem by adding telomeric repeats to telomeres. In mammals, the regulation of telomerase is tightly controlled to maintain equilibrium between telomere shortening, through end-replication, and telomere lengthening through telomerase.

1.2 Telomere length homeostasis

Telomere length is maintained at an established equilibrium, in the presence of telomerase. In a given cell, there is a distribution of different telomere lengths, and the telomere length distribution is maintained through multiple cell divisions (Greider, 1996).

This equilibrium is regulated by telomerase interaction with telomere binding proteins. In mammals, the shelterin complex is comprised of RAP1, TRF1, TRF2, TPP1, TIN2 and

POT1 (Smogorzewska and de Lange, 2004). In general, these proteins are negative regulators of telomerase activity (Abreu et al., 2010; Kendellen et al., 2009). However, during late S phase, telomerase is recruited to the telomere and adds telomere repeats

(Wellinger et al., 1993). Alterations to telomerase, or shelterin can affect the recruitment or processitivity of telomerase at the telomere, and alter the equilibrium length (Abreu et al., 2010; Kendellen et al., 2009; Smogorzewska and de Lange, 2004; Wang et al., 2007).

While some elements of this regulation have been described, many of the mechanistic details remain unclear.

Short telomeres are modeled in mice by successively breeding telomerase null mice. Late generation telomerase null mice have short telomeres, and display degenerative phenotypes, particularly in high turnover tissues (Blasco et al., 1997). These

2 CHAPTER 1. INTRODUCTION phenotypes are caused by short telomeres, not telomerase status, as early generation telomerase null mice have no phenotype (Blasco et al., 1997). Furthermore, telomerase is limiting and haploinsufficient in mammalian cells (Liu et al., 2000). Successive breeding of mice that are heterozygous for mTR, yields short telomere mice with degenerative phenotypes, in later generations (Hathcock et al., 2002; Hemann et al., 2001).

1.3 Telomere disease

In humans, short telomeres cause degenerative age-related disease (Armanios and

Blackburn, 2012). In adults, telomerase is only expressed in germ-line and tissue specific stem cells (Hiyama et al., 1995). This leads to gradual telomere shortening in somatic tissues, with age (Allsopp et al., 1992; Harley et al., 1990; Hastie et al., 1990). When telomeres become critically short, they limit cellular self-renewal cellular proliferation in a particular tissue (d'Adda di Fagagna et al., 2003).

Short telomeres can also arise as a result of genetic disease. For example, autosomal dominant mutations in the dyskerin gene cause dyskeratosis congenital (Heiss et al., 1998; Mitchell et al., 1999; Vulliamy et al., 2001). Loss of dyskerin lowers TR levels. Despite maintaining an intact copy of TR, patients still develop short telomeres and degenerative phenotypes, illustrating the haploinsufficiency of TR in human disease

(Armanios et al., 2005). Families with autosomal dominant dyskeratosis 3 ongenital display genetic anticipation (Armanios et al., 2005). Children inherit short telomeres, from parents, and successive generations have shorter and shorter telomeres, due to low

TR dosage. These shorter telomeres are accompanied by more severe disease phenotypes

3 CHAPTER 1. INTRODUCTION

(Armanios and Blackburn, 2012). While short telomere disease often manifests in high turn over tissue, such as bone marrow, liver and GI, they can also affect low turnover tissue such as the lung and liver (Armanios et al., 2007; Kojima et al., 1997). Recently, short telomeres were identified as a cause of idiopathic pulmonary fibrosis (Armanios et al., 2007) and susceptibility to emphysema (Alder et al., 2011).

Telomere maintenance is a hallmark of cancer (Hanahan and Weinberg, 2011).

Cancer cells need to maintain their telomeres in order to continue dividing (Greider,

1999). They accomplish this in telomerase-dependent and independent ways. The majority of cancers maintain telomeres by reactivating expression of telomerase (Artandi and DePinho, 2010; Kim et al., 1994). Some cancers are able to maintain telomeres in the absence of telomerase, through recombination-based methods (Bechter et al., 2004;

Conomos et al., 2013). Recently, promoter mutations in TERT and some telomere binding proteins, have been associated with an increased risk of cancer (Huang et al.,

2013; Killela et al., 2013; Ramsay et al., 2013; Shi et al., 2014). Sporadic mutations in

TERT increase telomerase levels. They are thought to generate new transcription factor binding sites, and are highly recurrent in melanoma (Huang et al., 2013). Germline TERT promoter mutations have also been identified as segregating with disease (Horn et al.,

2013). Together, these findings highlight the importance of telomerase activation in cancer.

1.4 Role of kinases in telomere length regulation

4 CHAPTER 1. INTRODUCTION

Kinases play an important role in regulating telomere length. In yeast, Tel1, a kinase involved in double-stranded break repair, has been well characterized as a positive regulator of telomere length (Greenwell et al., 1995). More recently, it has been demonstrated that ATM, the mammalian homologue of Tel1, is required for de novo telomere elongation (Lee et al., 2015). Previous work has also demonstrated that cdk1 is essential for telomere elongation in yeast (Frank et al., 2006). There is mounting evidence that kinases play an important role in telomere length regulation. Many kinases have been implicated in mammalian telomere regulation, including AKT, PI3K, mTOR,

PKC, MAPK, VRK1, and PIM1 (Chang et al., 2006; Choi et al., 2012; Cottage et al.,

2012; Kang et al., 1999; Kupiec and Weisman, 2012; Ludlow et al., 2012; Sitaram et al.,

2009). Kinases play an important regulatory role because phosphorylation is a powerful way to change protein function. Moreover, kinases are an attractive therapeutic target.

Inhibition of kinases by small molecules is specific, generates robust effects, and is well tolerated by cells (Zhang et al., 2009). Many of the best cancer drugs we have, such as imatinib, are examples of kinase inhibitors with excellent therapeutic effect.

1.5 Measuring telomere length

Telomere length is typically measured by TRF analysis, qPCR, Q-FISH and flow-

FISH.

In TRF Analysis (Harley et al., 1990), genomic DNA is digested with a restriction enzyme that spares the telomere. Telomere fragments are run on an agarose gel, and

5 CHAPTER 1. INTRODUCTION visualized on a membrane by hybridization with a radioactive telomere probe. This method is often considered the “gold standard” of telomere length measurements.

In the qPCR method, telomere repeats are amplified, which generates a fluorescent signal. By comparing the increase in fluorescent signal over multiple rounds of amplification, one is able to determine the number of telomere templates in the sample, as a proxy for the telomere length (Cawthon, 2002).

Q-FISH visualizes telomeres on metaphase spreads by hybridizing a fluorescent telomere probe (Poon and Lansdorp, 2001). Individual chromosomes and telomeres can be visualized by microscopy. Telomere length is determined by quantifying fluorescent signal at each chromosome end.

Lastly, flow-FISH uses flow cytometry to measure signal of a hybridized fluorescent telomere probe. Each cell, with hybridized telomere probe, will produce a fluorescent signal, corresponding to its telomere length (Baerlocher et al., 2006).

1.6 Screens for telomere length regulators

There is significant need to identify unknown genes involved in telomere length homeostasis. Firstly, it will elucidate many of the regulatory mechanisms that control telomere length, particularly in concert with other cell processes such as checkpoint control, DNA damage response, DNA replication and cell cycle progression. Secondly, while 13 genes to date have been identified to cause short telomeres in human disease, only 50% of families with inherited short telomeres have a known mutation (Stanley and

6 CHAPTER 1. INTRODUCTION

Armanios, 2015). Thus, there is significant need to identify other telomere genes, which may serve to inform clinical decisions, or serve as therapeutic targets.

Regulators of telomere length have identified by a number of methods including yeast genetics, two hybrid screens, biochemistry, proteomics, as well as linkage analysis of families with genetic telomere disease. Genetic screens are a powerful tool to identify novel regulators of telomere length. Pooled shRNAs screens have been used to identify regulators of replicative senescence (Burrows et al., 2010). Screening telomere length in a library of yeast knockout strains identified many novel, and previously identified regulators of telomere length (Askree et al., 2004; Ungar et al., 2009). Interaction screens, in yeast and humans, have also been performed to identify proteins that interact with the telomere, telomere binding proteins and telomerase (Hardy et al., 1992; Lin et al., 2015). While many of the key players have been identified in yeast, relatively few screens have been performed in mammalian systems. Importantly, while there are significant differences in the sequence identity of telomere proteins, across taxa, the mechanisms of equilibrium homeostasis are well conserved (Greider, 1996). Although mammalian homologues of yeast telomere genes may not play a role in mammalian telomere length, it is likely that phosphorylation is a key regulator in humans, as it is in yeast.

Fewer screens have been performed in mammalian systems. An siRNA screen in human cells identified proteins that regulate telomerase activity (Cerone et al., 2011).

However, this screen did not measure telomere length. Short telomeres are the direct cause of senescence, apoptosis and disease, however, no mammalian screen to date has

7 CHAPTER 1. INTRODUCTION used telomere length as an output. Here we develop a novel method of screening human genes involved in telomere length regulation.

8 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Chapter 2. Restoration of telomerase in mTR-/- mice enhances

hematopoietic stem and progenitor cell engraftment

2.1 Introduction

Short telomeres cause cellular senescence and limit tissue self-renewal (Hao et al.,

2005; Harley et al., 1990). Telomeres shorten with age (Hastie et al., 1990), and short telomeres manifest as degenerative disease, particularly in high turnover tissue

(Armanios and Blackburn, 2012). Successive breeding of mTERT or mTR knockout mice eventually yields telomerase knockout mice with short telomeres, which model the degenerative phenotypes of human disease (Blasco et al., 1997; Liu et al., 2000). It has been shown that these degenerative phenotypes are dependent on telomere length, not telomerase levels (Hao et al., 2005). Moreover, both TERT and TR are haploinsufficient

(Armanios et al., 2005; Blasco et al., 1997), indicating that telomerase is limiting in cells.

Given this limiting level, it would suggest that increasing telomerase levels in vivo may increase telomere length. In dyskeratosis congenita, short telomeres cause bone marrow failure and aplastic anemia (Armanios et al., 2005; Marrone et al., 2004; Yamaguchi et al., 2005). Therapeutic options are limited, because there are no current methods of elongating telomeres. Short telomeres limit the self-renewal and engraftment ability of hematopoietic stem cells (Rossi et al., 2007). Thus, a question in the field, is whether in vivo restoration of telomerase in bone marrow could restore functional ability and ameliorate degenerative phenotypes, in hematopoietic stem and progenitor cells. We

9 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH sought to test with experiments that could serve as “proof-of-principle” for future gene therapies involving telomere length restoration.

There is strong evidence that ectopic telomerase expression can increase telomere length. Lentiviral expression of TR and TERT is able to elongate telomeres in human cells in vitro (Cristofari and Lingner, 2006). Restoration of telomerase in patient derived dyskeratosis congenita cells was also sufficient to increase telomerase levels, and increase telomere length (Westin et al., 2007; Wong and Collins, 2006). In both cases, expression of TERT and TR together, generated a greater change in telomere length and telomerase activity, than either TERT or TR alone (Cristofari and Lingner, 2006). There is also significant evidence in vivo that telomerase restoration can enhance degenerative phenotypes of short telomeres. G4 mTR-/+ mice show better survival and body weight, compared to G4-/- littermates (Hemann et al., 2001). Genetically inducing overexpression of mTERT is sufficient to restore degenerative phenotypes in the GI, testes, bone marrow and CNS (Jaskelioff et al., 2011). Recently, a group used AAV vectors to re-introduce telomerase into Tert or Trf1 knockout mice (Bar et al., 2016).

Telomerase was able to target the hematopoietic stem cell niche, increase telomere length and enhance blood counts (Bar et al., 2016). Taken together, these data strongly support in vivo activation of telomerase, as an efficient method of restoring telomere length.

There has been some controversy as to whether cellular phenotypes of telomerase loss are cell autonomous. Previous work showed that functional reconstitution, following transplant of various hematopoietic organs, was determined by the recipient genotype, rather than host (Ju et al., 2007; Song et al., 2010; Song et al., 2012). This model predicts

10 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH that loss of telomerase is not cell autonomous, and that transplant of wild type telomerase tissue, is not sufficient to rescue degenerative phenotypes in short telomere tissue. A potential explanation of these results, is that stem cells, which lack telomerase, generate a different niche, compared to wildtype stem cells. This niche is composed of the microenvironment in which the stem cells resides, and will not change upon telomerase restoration. Work in this chapter distinguishes between cell autonomous and non-cell autonomous by evaluating the affect of telomerase restoration in telomerase knockout cells.

Another open question in the field is whether telomerase preferentially elongates the shortest telomere. Telomerase is limiting, and cannot target all telomeres in a cell cycle (Zhu et al., 1996). There is considerable evidence in yeast that short telomeres are preferentially elongated over long ones(Teixeira et al., 2004; Xu et al., 2013).

Saccharomyces cerevisiae has degenerate telomeres, making it possible to distinguish telomeres replicated by DNA replication, and those synthesized by telomerase. Cloning and sequencing of these telomeres upon telomerase activation reveals that telomerase prefers to elongate short telomeres over long telomeres (Teixeira et al., 2004). Single chromosome PCR of human telomeres reveals this is also true in human cells in vitro

(Britt-Compton et al., 2009). Previous work has shown that upon crossing short telomere mice with long telomere mice, the shortest telomere, not average telomere length, is what determines degenerative phenotype (Hemann et al., 2001).

2.2 Results and Discussion

11 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

2.2.1 Bone marrow transplant to test the effect of lentiviral telomerase restoration on hematopoietic stem and progenitor cell engraftment

We wanted to evaluate the effect of telomerase restoration, on engraftment of mTR-/- hematopoietic stem and progenitor cells (HSPCs). We used the bone marrow transplant model to study this because bone marrow transplant forces self-renewal and differentiation of HSPCs during transplant engraftment. Furthermore, bone marrow transplant allows us to measure engraftment of donor cells by evaluating their competitive repopulation. This will allow us to quantify the effect of mTR restoration.

The bone marrow transplant used in this experiment is summarized in figure 2.1.

This experiment is designed to evaluate the effect of mTR restoration in a competitive repopulation experiment. We harvested Lin- cells from wildtype or G4 mTR -/- mice on the C57BL/6 background. G4 mTR-/- mice are 4th generateion mTR knockout mice that display degenerative phenotypes caused by short telomeres (Blasco et al., 1997). Lin- cells are uncommitted bone marrow cells, composed mostly of undifferentiated stem and progenitor cells (Osawa et al., 1996). Donor mice on the C57BL/6 background carry the

CD45.2 antigen, which cab ne used to distinguish donor and host. Lin- cells were transduced with either SVA mTR (mTR-GFP) or FUGW (GFP alone) lentivirus. Lin- cells were then transplanted into lethally irradiated Pep Boy hosts, which carry the

CD45.1 antigen. Engraftment in peripheral blood was evaluated every 4 weeks. At 24 weeks, primary recipients were sacrificed. Telomere length was measured by flow-FISH and mTR and mTERT expression was determined by qPCR. In half of each treatment

12 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH group, Lin- cells were harvested and transplanted into secondary lethally irradiated recipients, to further allow expansion of donor cells. In the secondary recipients, engraftment in peripheral blood was measured every 4 weeks. Telomere length, mTR and mTERT expression, signal free ends and P/Q ratios were determined at 16 weeks post transplant.

2.2.2 Effect of telomerase restoration on hematopoietic stem and progenitor cell engraftment

Engraftment of donor cells was measured by examining cells in the peripheral blood, every 4 weeks post transplant for 24 weeks in the primary transplant, and every 4 weeks for 16 weeks in the secondary transplant (figure 2.1). We determined engraftment by evaluating the percentage of peripheral blood mononuclear cells that carried the

CD45.2 antigen from the donor. Since both SVA mTR and FUGW contain GFP, we were able to measure the engraftment of lentiviral transduced cells by measuring the percentage of donor cells that were GFP+. We reasoned that if mTR restoration enhanced

G4 mTR-/- cells’ engraftment, mTR-GFP transduced cells should outcompete non- transduced cells in generating differentiated blood cells, thus a greater percentage of the donor population will be GFP+.

In addition to measuring engraftment, we also evaluated the survival of recipients transplanted with WT or G4 HSPCs with either mTR-GFP or GFP (FUGW) only lentivirus. While, we noted no difference in survival between groups in the primary transplant (figure 2.2 a), upon secondary transplantation, the G4 mTR-/- group

13 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH reconstituted with FUGW showed lower survival compared to G4 mTR-/- group reconstituted with mTR (figure 2.2 b). This suggests that the additional cell divisions of a secondary transplant may cause G4 mTR-/- HSPCs to fail at engrafting. We conclude that restoration of mTR by lentivirus is sufficient to restore the survival defect of mice transplanted with G4 mTR-/- HPSCs.

We were able to evaluate the engraftment within various cellular lineages by gating on cells with particular cell surface markers. We observed a greater percentage of

GFP+ cells in the donor compartment of the G4 mTR -/- mTR group, compared to the G4 mTR-/- FUGW group (figure 2.3 a,b). This difference was pronounced in T-cells (figure

2.3 c) and weaker in B-cells and monocytes/granulocytes (figure 2.3 d, e). However, it was very pronounced in whole peripheral blood (figure 2.3 a,b). While this difference was not statistically significant, it strongly suggests that mTR expression confers a growth and engraftment advantage, compared to FUGW alone. This difference does not manifest until 12 weeks post transplant, suggesting mTR restoration provides an advantage in long-term restoration of the hematopoietic system, rather than the initial engraftment. mTR expression in wildtype bone marrow did not affect GFP+ percentage in the donor population, suggesting mTR only has an effect in the context of short telomeres (figure 2.3 a). We observed high variation in the percentage of donor cells that were GFP+, which contributed to the high variation in engraftment at later time points.

One cause of this is that SVA mTR generated very dim GFP, which made it difficult to accurately measure the percentage of cells that were GFP+. This issue is addressed in

2.2.4. An alternative way of measuring the relative contribution of mTR or FUGW to

14 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH engraftment is comparing the ratio of GFP+ in mTR transduced to FUGW transduced groups. This analysis shows that the ratio of GFP+ in mTR GFP compared to mTR

FUGW rises significantly after 12 weeks, while it remains relatively stable in the WT mTR compared to WT FUGW (figure 2.3 f). This further underscores the conclusion that restoration of mTR may confer an advantage in engraftment.

We did not observe significant changes in percent GFP+ donor cells in the mTR transduced, compared to FUGW transduced cells upon secondary transplantation, in both

WT and G4 mTR-/- donors (figure 2.4 a). This was also true in the T-cell, B-cell and monocyte/granulocyte compartment (figure 2.4 b, c, d). We expected further transplantation stress to accentuate the differences between G4 FUGW and G4 mTR group. However, unexpectedly, %GFP+ in donor cells plateaued up to 20 weeks. This could be due to survival bias. We saw greater death, post transplant, in the G4 FUGW group, compared to the G4 mTR group. Since we were only able to assay engraftment in survivors, it’s possible that the %GFP+ donors in the G4 FUGW group was inflated, because the mice with the lowest %GFP+ donor cells had already died.

2.2.3 Effect of telomerase restoration on telomere length

Given the differences in engraftment conferred by transduction with mTR lentivirus, we asked whether mTR transduction in HSPCs was increasing telomere length, upon transplantation. At 24 weeks post-primary transplant, we sacrificed mice and sorted GFP+ splenocytes by fluorescence activated cell sorting (FACS). We determined telomere length by flow-FISH, a flow cytometry based method that measures

15 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH telomere length based on fluorescence of a hybridized PNA telomere probe (Baerlocher et al., 2006). Interestingly, despite the effect on engraftment, we observed no difference in mean telomere length in the mTR transduced or FUGW transduced cells (figure 2.5 b). qPCR analysis of mTR levels revealed that mTR and mTERT levels in mTR transduced

G4 cells were comparable to wildtype levels (figure 2.5 a).

We hypothesized that the observed difference in engraftment may be due to restoration of critically short telomeres, rather than a bulk increase in mean telomere length. There is evidence that telomerase preferentially lengthens the shortest telomeres

(Britt-Compton et al., 2009; Teixeira et al., 2004). Particularly in the context of modest mTR expression, we may not observe a change in mean telomere length, though the shortest telomeres are still being lengthened enough to prevent DNA damage. To test this, we performed Q-FISH on metaphase spreads of GFP+ splenocytes. Q-FISH is able to visualize single telomeres on chromosomes. Ends of chromosomes without telomere signal are called signal free ends, and represent critically short telomeres. We observed a significant decrease in signal free ends in G4 mTR, compared to G4 FUGW, suggesting that mTR restoration rescued these critically short telomeres (figure 2.6 b).

These data are consistent with the current model of telomere-induced senescence.

In this model, when telomeres fall below a threshold length, they are recognized as DNA damage, signal a double stranded break and lead to cellular senescence or apoptosis

(Garvik et al., 1995; Lydall and Weinert, 1995). We propose that, in these experiments, the modest levels of mTR were sufficient to keep very short telomeres above this threshold, thereby preventing senescence and allowing stem cell self-renewal. While our

16 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH data cannot exclude the fact that telomerase is equally elongating short and long telomeres, it is consistent with the hypothesis that telomerase prefers elongating the shortest telomere. We propose that in the context of modest telomerase restoration, telomerase elongates the shortest telomeres. This is important for telomerase restoration as gene therapy, because telomerase reactivation is a hallmark of cancer (Hanahan and

Weinberg, 2011). Significant over expression of telomerase may provide conditions that increase the likelihood of cancer development. If a modest increase in telomerase is sufficient to enhance engraftment, it may be a better therapeutic option.

Despite seeing a difference in engraftment in the primary transplant, we wondered why we didn’t observe a difference between mTR and FUGW transduction upon secondary transplantation. Particularly given that telomeres should shorten even more in the secondary transplant, as G4 stem cells are forced to self-renew in the absence of telomerase. Restoration of critically short telomeres can happen through telomerase, but it can also occur through telomerase independent methods. Alternative lengthening of telomeres (ALT) maintains or lengthens telomeres through recombination-based methods that are telomerase independent (Bryan et al., 1995). ALT can be measured by a skewing of the ratio of telomere lengths on the P and Q arm of telomeres. Indeed, in the secondary transplant, we noted an increase in P/Q ratio in G4 FUGW GFP+ splenocytes, compared to G4 mTR (figure 2.6 b). This suggests that the G4 FUGW group had developed the ability to maintain its telomeres in a telomerase independent fashion, via an ALT mechanism. This is also supported by the fact that the G4 FUGW group had no

17 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH significant decrease in telomere length despite the great self-renewal and differentiation load imposed by two serial transplantations (figure 2.5 b).

2.2.4 Generation of novel lentiviral constructs to restore telomerase in vivo

Restoration of mTR alone, in G4 mTR-/- bone marrow did not robustly enhance telomerase expression, or telomere length, in our bone marrow transplant model (Figure

2.6 a,b). We wished to develop a lentiviral vector that could generate significant changes in mean telomere length. This would allow us to better evaluate the role of telomere length restoration in engraftment. It has been shown that expression of TR and TERT together, significantly enhances the rate of telomere elongation, compared to TR alone

(Cristofari and Lingner, 2006). Consistent with this, we found that expression of mTR and mTERT from a single lentiviral vector, significantly enhanced telomerase expression and telomerase activity, compared to mTR alone (figure 2.7 a, b). However, the GFP expression from this mTR and mTERT virus, which we called SVA, was very low. This made it difficult to accurately distinguish GFP positive and negative cells, which is the primary metric we used to gauge engraftment. Therefore, generating a novel construct with higher mTR, mTERT and eGFP expression would allow us to have greater effects on telomere length, and more accurately determine engraftment.

We generated many novel mTR and mTERT constructs in an effort to create a construct that provide robust telomerase expression and activity, and high GFP expression (table 2.2). We tested multiple promoters, orientations and orders for an mTR mTERT virus. We found that we were only able to get mTR expression when we drove

18 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH its expression from its endogenous promoter (table 2.2). We also found that addition of a

CAAX motif to GFP, which localized it to the plasma membrane, did not increase GFP brightness on the flow cytometer (table 2.2). Lastly, we found that the identify of the promoter downstream from mTR, affected mTR’s expression. In the case of LVT5, although eF1a generated high levels of mTERT and eGFP, it inhibited expression of mTR, which was upstream from it (table 2.2).

We decided to use “LVT3b” because it generated the best compromise between bright GFP and robust telomere lengthening upon transduction (figure 2.8 b, c). This construct has the endogenous promoter driving mTR, followed by PGK driving mTERT-

2a-GFP. Moreover, in order to generate a negative control construct with the exact same transduction efficiency and GFP expression, we generate deletion variants of all constructs, which had the template region of TR deleted (figure 2.8 a). The mTR molecule transcribed from this deletion mutant will not form functional telomerase.

We have shown that co-expression of mTR and mTERT, led to more robust changes in telomere length in vitro. We predict that future experiments co-expressing both components will also lead to greater increases in telomere length in vivo, as well as enhanced engraftment. Secondly we have generated lentiviral vector that expresses higher levels of GFP, allowing us to effectively differentiate transduced and untransduced cells. Future experiments will determine whether LVT3b can create a more robust telomere length restoration and engraftment enhancement. An additional advantage brighter GFP is an increase in titer accuracy. Both SVA and LVT3b are large vectors and

19 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH generally produces poor titers. Previously, low GFP brightness made it difficult for us to distinguish GFP+ cells, which made tittering in accurate. This led to variable transduction efficiencies in Lin- cells. Recipients received populations of cells, which started with different baseline levels of GFP. While we were able to use relative changes in GFP levels to determine engraftment efficiency, we would be able to track absolute GFP levels, if each mouse started with relatively comparable % GFP+. Brighter GFP in

LVT3b will allow us to address this problem.

Thus, LVT3b is a promising construct, which will allow us to further evaluate the effect of robust telomere length change on engraftment.

2.3 Materials and Methods

2.3.1 Transgenic Mice

mTR-/- mice were bred from (Blasco et al., 1997). Generation 4 mTR -/- mice with short telomeres were generated by crossing mTR +/- mice from each generation.

Wildtype mice were on the C57BL/6J genetic background. Recipient mice were of the

B6.SJL-Ptprca Pepcb/BoyJ strain and carried CD45.1 leukocyte marker (Jackson

Laboratory).

2.3.2 Lentivirus production and transduction

SVA mTR is a lentiviral vector containing mTR driven by its endogenous promoter, followed by IRES and eGFP. To generate SVA mTR lentivirus, we first coated

15-cm polystyrene plates (BD Falcon) with 10 mL of 100 μg/ml poly-D lysine for 30

20 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH minutes. Next we plated 8 million 293FT cells in DMEM media (Gibco) with 10% fetal bovine serum (Gibco) and penstrep glutamine (Gibco). After 24 hours, we changed media to 1% FBS in DMEM and co-transfected SVA mTR (mTR lentiviral vector), pCMVΔ8.91 (containing gag and pol lentiviral genes), and VSVG (containing the env lentiviral gene). Transduction was performed with Lipofectamine 2000 (Life

Technologies) in Opti-MEM media (Gibco). Supernatant was collected after 48 hours, briefly centrifuged to remove cellular debris and filtered through a 0.45 μm CN filter

(Thermo) to eliminate remaining debris. Viral supernatant was concentrated using 100 kDa ultracentrifugal filters (Millipore) and centrifuging for 10 minutes at 4000g.

Concentrated virus was then collected, aliquot and frozen at -80.

To transduce Lin- cells with SVA mTR virus, we added virus at an MOI of 2.5 directly to isolated Lin- cells in the presence of 8 μg/mL polybrene. Cells are transduced for 24 hours, before being washed with PBS and injected into recipients.

2.3.3 Bone marrow harvest and collecting lineage depleted cells

8-12 week old Donor mice were sacrificed by exposure to isofluorane, followed by cervical dislocation. The femur, tibia, spine and hips were removed from the animal.

Tissue was removed from bones. Bone marrow was collected by crushing bones with mortar and pestle, adding PBS (Gibco) and passing cells through a 70 μm cell strainer.

Cells were then washed with PBS and resuspended in RBC lysis buffer (eBioscience) and incubated on ice for 5 minutes, in order to lyse all red blood cells. Cells were then spun down, washed again wish PBS and resuspended in PBS with 2% FBS and 0.1 mM EDTA

21 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH pH 8.0. Lin- stem and progenitor cells were enriched using the Mouse Hematopoietic

Progenitor Enrichment Kit (StemCell Technologies) following the EasySep protocol.

Isolated Lin- cells were then counted, and plated at 1 million cells/24-well dish in FTK medium (StemSpan) containing 100 ng/ml mSCF, 50 ng/ml mFLT-3 ligand, 10 ng/ml mTPO (all cytokines from PeproTech), and 1X gentamicin (Gibco) This media limits differentiation of stem and progenitor cells. Cells were then transduced overnight with virus as in section 2.4.2.

2.3.4 Bone Marrow Transplant

At least 2 hours prior to transplant, recipient mice with lethally irradiated with

950 rads of x-ray radiation. We prepared donor Lin- cells at a concentration of 2 x 106 cells/mL in PBS (Gibco). Recipient mice had their tail veins dilated by exposure to a heat lamp, and were transplanted with 250 µL (5 x 105 cells) of Lin- donor cells. We also included a radiation control, which was not transplanted with donor cells. This mouse typically expired 10-14 days post irradiation, and is used to ensure recipient mice were given a lethal dose of irradiation. Mice were monitored daily following transplant. Mice in a severely weakened state, evidenced by low body weight, ruffled fur or hunched back were euthanized immediately for ethical reasons as they were unlikely to survive much longer.

2.3.5 Peripheral blood collection and flow cytometry

22 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

We obtained peripheral blood from recipient mice every 4 weeks by puncturing the facial vein with a sharp lancet and collecting 150-250 μL of peripheral blood. Blood was washed with PBS, and resuspended in RBC lysis buffer (eBioscience) for 5 minutes on ice to lyse red blood cells. In addition to transplant recipients, we also collected peripheral blood from C57BL6/J and a B6.SJL-Ptprca Pepcb/BoyJ mice to serve as staining controls. Cells were then washed with PBS and resuspended in 200 µL of PBS with 2% FBS. Antibody was used at a 1:500 concentration, and incubated for 30 minutes at 4°C. Antibodies used were PE mouse anti-mouse CD45.1 (donor cells), PE-Cy™7 mouse anti-mouse CD45.2 (recipient cells), APC hamster anti-mouse CD3e (t-cells),

APC rat anti-mouse CD45R/B220 (b-cells), and APC rat anti-mouse CD11b (monocytes and granulocytes) (all BD Biosciences). We also used isotype controls to allow offline compensation of fluorophore bleed through. This included all isotype controls to determiner total background, as well as isotype controls in the context of a single antibody to determine the bleed through of that particular antibody in the context of background. Isotype controls used were PE rat IgG2 κ, PE-Cy™7 rat IgG2 κ, and APC rat IgG1 κ (all BD Biosciences). Data was collected on a FACSCalibur flow cytometer

(BD Biosciences) and analyzed using FloJo software (FloJo).

2.3.6 Cell Sorting

At terminal transplant time points, we sacrificed mice by isofluorane inhalation and cervical dislocation, then collected whole spleens. Whole spleens were mashed through a 70 μm cell strainer (BD Falcon) to obtain a single sell suspension, then washed

23 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH with PBS (Gibco). We sorted 5 x 106 GFP+ cells by FACS using a MoFlo cell sorter (BD

Biosciences) at the JHSPH Cell Sorting Facility. Sorted cells were collected in PBS on ice, washed with PBS and cultured in DMEM with 10% FBS.

2.3.7 Quantitative Real-Time PCR

We used quantitative real-time PCR (qPCR) to determine the levels of mTR in sorted GFP+ splenocytes. First, we isolated RNA using the RNAeasy RNA isolation kit

(Qiagen). Next, we generated cDNA from RNA using random hexamer primers with the

SuperScriptIII First Strand Synthesis Kit (Thermo). We assumed that there was the same amount of cDNA after the reverse transcription as there was RNA added. qPCR was performed 96-well. qPCR was performed in a CFX96 Thermocycler (Bio-Rad). Each reaction contained 1X SYBR Green Supermix (Bio-Rad), containing buffer, enzyme and

SYBR Green, 5 µM primers and 5 ng cDNA. Samples were run in triplicate. We normalized mTR expression to HPRT expression. We generated a standard curve for by using known quantities of HPRT-TOPO and mTR-TOPO, which are TOPO vectors where HPRT or mTR have been cloned in. Primers are listed in table 2.1. Because mTR does not contained introns that are spliced out, we also included a “no reverse- transcriptase” control for every sample to determine amplification of contaminating gDNA. Cycling conditions for mTR were: 5 minutes at 95C; 15 seconds at 95C, 30 seconds at 68C, 45 seconds at 72C, 10 seconds at 82C (35 cycles); 3 minutes at 72C.

Cycling conditions for HPRT were 5 minutes at 95C; 15 seconds at 95C, 30 seconds at

24 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

58C, 30 seconds at 72C (35 cycles), 3 minutes at 72C. Relative mTR levels were determined by the ΔΔCt method.

2.3.8 Telomere flow-FISH

Flow-FISH measures telomere length by hybridizing a fluorescent telomere probe to fixed cells, then analyzing fluorescence of those cells by flow cytometry. We used a protocol adapted from the one previously described (Baerlocher et al., 2006). Briefly, splenocytes are passed through a 70 µm cell strainer (BD Falcon) to generate a single cell suspension. We then lyse red blood cells by incubating in RBS lysis buffer (eBioscience) for 5 minutes on ice. Cells are then washed with PBS and fixed for 30 minutes at 4C in

70% Ethanol. Next, they are washed in PBS and resuspended in the hybridization mix, which is comprised of 70% formamide (Fisher Scientific), 0.5% blocking reagent

(Roche), 0.01 M Tris, and 0.4 mg/mL FITC-labeled PNA probe (Applied Biosystems).

Hybridization is performed by incubating for 15 minutes at 87C. Cells are kept at room temperature overnight, washed, then resuspended in PI staining solution for 30 minutes at room temperature. PI staining solution is comprised of 0.1% Triton X-100 (Sigma), 20

μg/mL RNAse A (Sigma), and 20 μg/mL propidium iodide (Sigma). Telomere length is determined by flow cytometry on a FACSCalibur flow cytometer (BD Biosciences) by gating on G1 cells with PI, and determining mean FITC signal on a linear scale.

2.3.9 Telomere Quantitative FISH

25 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

We measured signal free ends (SFE) and P/Q ratios by Quantitative FISH (Q-

FISH). We generated metaphase spreads from splenocytes as previously described

(Hemann et al., 2001). We hybridized Cy3 PNA telomere probes (Applied Biosystems) to metaphase spreads, and imaged with Zeiss Axioskop microscope (Zeiss). Telomere length for P/Q ratio was determined using TFL-TELO software as described (Poon and

Lansdorp, 2001). SFEs were defined as telomere ends with no detectable Cy3 signal.

2.4.9 Telomerase Repeat Amplification Protocol

We measured telomerase activity using the telomerase repeat amplification protocol TRAP (Kim et al., 1994). First, cell lysates are generated by resuspending pellets in 1X CHAPS lysis buffer (10 mM Tris-HCl pH 7.5, 1 mM MgCl2, 1 mM EGTA pH 8.0, 0.1 mM benzamidine, 5 mM -mercaptoethanol (BME), 0.5% CHAPS, 10% glycerol) for 30 minutes on ice. The cells are then spun down and the supernatant is collected. Half of the sample is treated with RNase A to control for non-telomerase dependent products. Next, telomerase substrates are exposed to the lysate by incubating lysates in TRAP reaction buffer (200 mM Tris-HCl pH 8.3,15 mM MgCl2, 630 mM KCl,

0.5% Tween-20, 10 mM EGTA, pH 8.0), 0.1 mM dNTPs, and 34 μM TS primer for 1 hour at 30°C. Next, these products are amplified by transferring 2 µL of the products from the first step to a 48 µL mastermix containing 1X Taq buffer (Denville), 0.1 mM dNTPs (TaKaRa), 2 U Taq polymerase (Denville), 20 μM end-labeled TS primer (4 μCi

γ-32P-ATP, 1X PNK forward buffer (Invitrogen), 20 μM TS primer, 1 U T4 PNK

(Invitrogen), and primer mix (8.5 μM reverse primer, 17 μM internal control reverse

26 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH primer K1, 10-12 μM internal control template TSK1). The products are amplified by incubating at 94C x 10 min; 30 cycles of 94C x 30s, 59C x 30 sec. These products are run for 20 minutes at 20W on a

10% non-denaturing polyacrylamide gel (1X TBE, 10% acrylamide/bis solution

(19:1) (BioRad), 2% glycerol, 0.144% ammonium persulfate (Sigma), 0.04% N,N,N′,N′-

Tetramethylethylenediamine (Sigma)). The gel is then dried, exposed on a phosphorimager (Fuji) for 1 hour, and scanned on a STORM 860 imager (GE

Healthcare).

2.3.10 Plasmid Construction

LVT3b was generated from a previously constructed plasmid called LV-TEL 3.0

(Vidal-Cardenas 2011). LV-TEL 3.0 contained mTR driven by its endogenous promoter, followed by PGK, mTERT-2a-eGFP. This resembled the desired final product but contained an indel in the 2a peptide sequence. We generated LVT 3b by cloning in a synthesized 2a fragment between mTERT and eGFP, by Gibson assembly. The general strategy was to first make 2 cuts with restriction fragments upstream and downstream of the 2a peptide, to remove the previous fragment. BlpI cut upstream in PGK and XbaI cut downstream in Tert. We then generated 2 PCR fragments, one upstream and one downstream of 2a. These fragments were generated with primers that overlapped the adjoining sequence, so they could be ligated together during Gibson assembly. The forward primer on the downstream component (LVT3b_3) and the reverse primer on the upstream component (LVT3b_2) contained overlap sequences to a synthesized 2a

27 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH sequence with the correct sequence. The upstream fragment, downstream fragment, synthesized 2a fragment, and restriction enzyme generated vector were then ligated together using Gibson assembly.

The same primers were used on LV-TEL 3.0 del, to generate a corresponding mTR deletion mutant.

2.3.11 Telomere Restriction Fragment Southern Blot

We measured telomere length by telomere restriction fragment Southern blot. Genomic

DNA was extracted using Puregene Core Kit A (Qiagen). 2 μg of gDNA was digested with MseI (NEB) and loaded onto a 0.7 % TAE Agarose gel. The gel was run at 37 V for

16 hours, then denatured for 30 minutes in 0.5 M NaOH and 1.5 M NaCl, then neutralized for 30 minutes in 1.5 M NaCl and 0.5 M Tris-HCl pH 7.4. The DNA was transferred from the gel to a nylon membrane (Amersham Hybond N+) by the weighted method overnight. Next, DNA is crosslinked to the membrane with a UV Stratalinker

(Stratagene). The membrane is then pre-incubated for 2 hours at 65C with Church’s

Buffer. Telomere probe and 2-log ladder (Life Technologies) is then end labeled with

Klenow fragment polymerase (NEB) 33 μM of dATP, dTTP, dGTP and 50 μCi of α-32P dCTP (3000 Ci/mmol). Unincorporated nucleotides are removed by running on a G50 column (GE Healthcare). Labeled probe is counted, denatured at 100C, and 106 counts/ mL telomere probe and 2 x 104 counts/mL of 2-log ladder are added. The membrane is probed overnight at 65C. The next day, the membrane is washed twice with 2X SSC

28 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

0.1% SDS, then twice with 0.5X SSC 0.1 SDS. The membrane is exposed on a Fuji phosphorimager (Fuji) and imaged on a STORM scanner (GE Healthcare).

29 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.1 Experimental design to test effect of mTR expression on hematopoietic stem and progenitor cell engraftment

Lin- progenitor cells are harvested from G4 mTR-/- or wildtype C57BL/6 mice. C57BL/6 mice carry the CD45.2 antigen. mTR-GFP (SVA mTR) or GFP alone (FUGW) lentivirus was transduced overnight into donor Lin- cells and transplanted into lethally irradiated

Pepboy hosts, containing the CD45.1 antigen. Donor engraftment in peripheral blood was measured every 4 weeks by flow cytometry. At 24 weeks post transplant, half of each group was transplanted into secondary recipients to force further hematopoiesis. The other half of each group was sacrificed, and GFP+ splenocytes were isolated by cell sorting. Telomere length in sorted splenocytes was determined by flow-FISH. mTR and mTERT expression was determined by qPCR. Engraftment in peripheral blood of secondary transplant was determined every 4 weeks. At 16 weeks post secondary transplant, GFP+ splenocytes were isolated by cell sorting, and telomere length, mTR and mTERT expression were determined, as in the primary transplant.

30 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.1

Measure mTR-GFP or GFP engra ment alone len virus Every 4 weeks

Secondary Transplant into Transplant lethally irradiated host (CD45.1) Sort GFP+ splenocytes

Sort GFP+ splenocytes Evaluate telomere length, mTR & G4 mTR-/- or WT Lin- Cells Measure engra ment mTERT expression (CD45.2) every 4 weeks

31 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.2 Kaplan-Meier survival curve for primary and secondary transplant

A) Recipient survival in primary transplant up to 168 days (n=10 per group). Radiation control was lethally irradiated but not transplanted with bone marrow.

B) Lin- cells were collected from primary recipients and transplanted into secondary recipients (n=10 per group). Survival was measured up to 140 days post transplant. p=0.0242 for G4 mTR compared to G4 FUGW

32 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.2

A BMT 5.2 Primary Transplant Survival

100 WT FUGW

l 80 WT mTR

v i G4 FUGW

v r u 60 G4 mTR

n Radiation Control e 40

P 20

0 0 50 100 150 Days

B Secondary Transplant Survival

100

l 80

u 60

t n WT FUGW e 40 c

e WT mTR P 20 G4 FUGW

G4 mTR 0 0 50 100 150 Days

33 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.3 Donor engraftment in primary transplant

Percentage of donor cells that were GFP+, was used to measure the ability of transduced cells in each group to contribute to engraftment. Since not all donor cells were transduced, an increase in percent GFP+ corresponds to enhanced engraftment ability of the transduced cells, relative to the untransduced. GFP was measured by flow cytometry on mononuclear cells from peripheral blood.

A) Overall percent GFP+ donor cells from WT donors, transduced with FUGW or mTR

B) Overall percent GFP+ donor cells from G4 mTR -/- donors, transduced with FUGW or mTR

C) Percent GFP+ donor cells from CD3 T cells

D) Percent GFP+ donor cells from B220 B cells

E) Percent GFP+ donor cells from CD11b monocytes and granulocytes

F) Ratio of percent GFP+ donor cells in mTR transduced to percent GFP+ donor cells in

FUGW transduced. This ratio is used as a relative measure of the effect of the virus in enhancing hematopoietic ability of transduced cells.

34 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.3

A Wildtype Donor GFP B G4 mTR -/- Donor GFP

100 100 WT FUGW G4 FUGW WT mTR

80 80 G4 mTR )

) % % 60

( 60

(

P F F 40 40

G G

20 20

0 0 0 10 20 30 0 10 20 30 Time (weeks) Time (weeks)

C D Donor Derived T-cells Donor Derived B-cells

100 100 WT FUGW WT FUGW 80 WT mTR 80 WT mTR G4 FUGW G4 FUGW

+ +

P 60 G4 mTR P 60 G4 mTR

F F

G G

% 40 % 20 20 0 0 0 4 8 2 6 0 4 0 4 8 2 6 0 4 1 1 2 2 1 1 2 2 Weeks Weeks

E F Primary Donor GFP

Donor Derived Monocytes and Granulocytes 2.5 WT mTR/WT FUGW 100 WT FUGW 2.0 G4 mTR/G4 FUGW 80 WT mTR

o 1.5 i

G4 FUGW t a

P 60 G4 mTR R

F 1.0

G 40 % 0.5 20 0.0 0 10 20 30 0 Time (weeks) 0 4 8 2 6 0 4 1 1 2 2 Weeks

35 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.4 Donor engraftment in secondary transplant

Lin- cells were harvest from primary recipients and transplanted into secondary recipients. Percent GFP+ donor cells was used to measure the competitive advantage in hematopoiesis, conferred by transduction of the indicated virus.

A) Percent GFP+ donor cells in all mononuclear cells in peripheral blood

B) Percent GFP+ donor cells in CD3 T cells

C) Percent GFP+ donor cells in B220 B cells

D) Percent GFP+ donor cells in CD11b monocytes and granulocytes

36 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.4

A B Total Donor GFP T-Cell Donor GFP 100 100 WT FUGW WT mTR 80 80

P G4 FUGW

P F

F G

G t G4 mTR

t WT FUGW n

n e

c r c WT mTR 40

r 40 e

e P

P G4 FUGW 20 20 G4 mTR 0 0 0 5 10 15 20 25 0 5 10 15 20 25 Weeks Weeks

B-Cell Donor GFP Monocyte and Granulocyte Donor GFP C D 100 100 WT FUGW 80 WT FUGW 80 WT mTR

P WT mTR P

F F

G G4 FUGW 60 G 60

G4 FUGW

t t

n n G4 mTR

e G4 mTR e

c c

r 40 r 40

e e

P P 20 20

0 0 0 5 10 15 20 25 0 5 10 15 20 25 Weeks Weeks

37 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.5 mTR expression and telomere length in mTR or FUGW transduced WT or G4 cells

GFP+ splenocytes were isolated by FACS 24, weeks post primary transplant, to determine the effect of mTR or FUGW transduction.

A) qPCR for mTR levels in GFP+ splenocytes

B) Flow-FISH to determine telomere length in GFP+ Splenocytes. WT and G4 are untransplanted controls

38 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.5

A mTR levels in Sorted GFP+ Splenocytes

v 3

a l e 1 R

W R W R G T G T U m U m F T F 4 T W 4 G W G

39 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.6 Signal free ends and P/Q ratio of GFP+ sorted splenocytes

A) Signal free ends in telomere Q-FISH represent critically short telomeres. Transduction of mTR, compared to FUGW, eliminated signal free ends in GFP+ splenocytes

B) The ratio of telomere lengths on the P/Q arm of chromosomes is a measure of the alternative telomere lengthening, since alternative lengthening increases the variance in

P/Q ratio. Shown are representative P/Q ratios, after 20 weeks secondary transplant, of a

G4 FUGW sample and G4 mTR sample.

40 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.6

Signal Free ends in GFP+ Splenocytes A 0.5

h 0.4

a t

M / 0.3

e 0.2

a n 0.1

g i

0.0 W R G T U m F 4 4 G G

Representative P/Q ratio B G4 FUGW

Representative P/Q ratio G4 mTR

41 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.7 Expression of mTR and mTERT compared to mTR alone

A) Expression of mTR, measure by qPCR, of G4 mTR-/- fibroblasts transduced with mTR and mTERT (SVA) or mTR alone (SVA mTR). WT is untransduced wildtype

C57BL/6 fibroblast.

B) Telomerase repeat activity protocol (TRAP) to measure telomerase activity in G4 mTR-/- fibroblasts treated with FUGW (GFP only), mTR and mTERT (SVA) or mTR alone (SVA mTR). RNAse is added to samples as a negative control to ensure activity is telomerase specific, since telomerase contains an essential RNA component.

42 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.7

B SVA SVA mTR Lentivirus: FUGW

RNase: - + - + - +

10000 1000

s 100

p 10 x

R 1

m 0.1 0.01 W T R T G R T W U E m F T A m V + R S O T + K m R O T A K V R m S T + m O K R T m

43 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.8 LVT3b lentiviral telomerase construct elongates telomeres and generates bright GFP

A) Graphical depiction of novel LVT3b and LVT3b del construct, expressing mTR and mTERT from a single construct. LVT3b del contain mutant mTR with a deleted template region

B) Flow cytometry of mouse fibroblasts transduced with LVT3b or SVA lentivirus. X- axis: GFP expression, Y-Axis: RFP expression. Cells in bottom right section are GFP+

C) Telomere Southern blot of G4 mTR-/- fibroblasts transduced with FUGW (GFP only),

SVA (old mTR mTERT virus), LVT3b (new mTR mTERT virus) or WT untransduced fibroblasts.

44 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 2.8

A LVT3b

LVT3b del

B SVA LVT3b

SVALVT3b WT FUGW

45 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Table 2.1 Primer list

To clone LVT3b and LVT3b del Primer Sequence LVT3b_1 GATGGACAGCACCGCTGAGCAATGGAAGCG GTA G LVT3b_2 CATCCATTCCCAGCGTAATC LVT3b_3 CGCGCTCCTCGTTGCCCC LVT3b_4 CTGTTCACCTGCAAGTCTAGAAATAGACCG TGACACTTCAACCGCAAG eGFP_2a_mTERT_oligo GATTACGCTGGGAATGGATGAGCTGTACAA GGGATCCGGAGAAGGCCGCGGATCCTTGTT AACATGTGGTGACGTCGAAGAAAACCCGGG GCCCATGACCCGCGCTCCTCGTTGCCCCGC qPCR for mTR, mTERT, HPRT Primer Sequence HPRT_F TGATCAGTCAACGGGGGACA HPRT_R TTCGAGAGGTCCTTTTCACCA mTERT_F GGATTGCCACTGGCTCCG mTERT_R TGCCTGACCTCCTCTTGTGAC mTR_F TGTGGGTTCTGGTCTTTTGTTCTCCG mTR_R GTTTTTGAGGCTCGGGAACGCG

TRAP K1 TRAP ATCGCTTCTCGGCCTTTT TSK1 TRAP AATCCGTCGAGCAGCCGAGAAGCGAT TRAP TS AATCCGTCGAGCAGAGTT TRAP Reverse CCCTTACCCTTACCCTTACCCTAA

46 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Table 2.2 Mouse mTR and mTERT lentiviral constructs

Name Contents mTR mTERT TRAP GFP Telomere Elongation FUGW hUBC, eGFP none none none +++ none SVA mTR endo prom, mTR, +++ +++ +++ + +++ PGK prom, mTERT- IRES-eGFP LVT3 mTR endo prom, mTR, ++ ++ ++ + - PGK prom, eGFP-2a- mTERT (missense) LVT3b mTR endo prom, mTR, ++ ++ ++ ++ ++ PGK prom, eGFP-2a- mTERT LVT3b mTR endo prom, mTR none ++ none ++ none del del, PGK prom, eGFP-2a- mTERT LVT3c mTR endo prom, mTR, - - - + - PGK prom, eGFP-CAAX motif-2a-mTERT LVT4 mTR endo prom, mTR, - - ++ + - hUBC prom, eGFP-2a- mTERT LVT5 mTR endo prom, mTR, + +++ +++ ++ + EF1a prom, eGFP-2a- mTERT LVT7 U6 prom, mTR, EF1a - - none - - prom, eGFP-2a-mTERT LVT7G U6 prom, G, mTR, EF1a - - none - - prom, eGFP-2a-mTERT LVT10 mTR endo prom, mTR, - - - + - PGK prom, mTERT- IRES-eGFP-CAXX Motif Note: “-” denotes no data available

47 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Chapter 3. Identifying novel kinase regulators of telomere

length by pooled screening approach

3.1 Introduction

Telomere length is maintained at an established equilibrium, in the presence of telomerase (Greider, 1996). Many of the details, of telomere length regulation, are known in yeast. In humans, many of the players are known, the molecular details, of how length is regulated, are not fully understood. Telomere length is regulated at many levels, including transcription (Kimura et al., 2004; Wu et al., 1999), cell cycle regulation (Frank et al., 2006; Zhu et al., 1996), RNA processing (Melek et al., 1996; Mochizuki et al.,

2004), and post translational modifications (Greenwell et al., 1995; Hang et al., 2011;

Mallory and Petes, 2000).

Post-translational modification is a robust method of regulating telomere length.

Phosphorylation (Greenwell et al., 1995), ubiquitination (Chang et al., 2003), sumoylation (Hang et al., 2011), and chromatin modification (Greider, 1992) have all be implicated in telomere length regulation. Understanding these processes is critical for understanding telomere biology, but also provides valuable targets for therapeutic development. Presently, there are few inhibitors targeting telomerase, or known telomere binding proteins. While there have been some attempts at developing telomerase inhibitors (Roth et al., 2010), direct inhibition of telomere binding proteins is problematic, because they regulate both telomere length and end protection. Loss of

48 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH telomere binding proteins leads to deprotected telomeres, which are prone to signaling

DNA damage or fusing with other telomeres (Denchi and de Lange, 2007). Kinases are attractive therapeutic targets, given the ease of developing bioavailable, chemical inhibitors against them (Zhang et al., 2009). Previous work has shown that the kinases

ATM (Lee et al., 2015) and CDK1 (Frank et al., 2006) are required for telomere elongation. Thus, we sought to develop a genetic screen to identify novel kinase regulators of telomere length.

Recent work has demonstrated the power of genetic screens using pooled libraries of shRNAs (Strezoska et al., 2012) or sgRNAs in the presence of Cas9 (Wang et al.,

2014). Telomere length can be measured with flow-FISH, which uses hybridization of a fluorescent telomere probe and measuring fluorescence on a flow cytometer (Baerlocher et al., 2006). We adapted this method to separate cells based on telomere length, using fluorescence activated cell sorting (FACS). While flow-FISH provides single cell accuracy in selecting cells with short or long telomeres, it requires fixing cells. Therefore, we are unable to use multiple rounds of selection to strengthen selection for long or short telomere cells. We combined the robustness of pooled library screening, and the specific selective power flow-FISH to create a powerful method for identifying novel regulators of telomere length.

Results and Discussion

3.2.1 Identifying cell line for pooled telomere length screening

49 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

We wished to identify a cell line, in which to perform a pooled screen for novel regulators of telomere length. We first set out to identify a cell line with moderate telomere length, in which we could measure dynamic changes in length in either direction, and that was amenable to FACS. In addition to telomere length we also measured transduction efficiency of multiple cell lines by transducing of a GFP lentivirus and measuring GFP (Figure 3.1a and b). We evaluated KG-1, HL60, L-1210, Daudi,

X16C8.5, Raji, 32D U-266 and K-562 cells. We identified K-562 as a candidate cell line due to its high transduction efficiency and moderate telomere length (figure 3.1 a, b).

However, K-562s were not amenable to efficient propidium iodide (PI) staining (Figure

3.1c). Flow-FISH requires treating cells with harsh denaturing conditions to hybridize the telomere probe. This had a differential effect on PI staining in different cell lines. PI staining is required for flow-FISH, so we can identify cells in the G1 phase of the cell cycle. An accurate cell cycle stain is required to gate on G1 cells because S phase cells and G2 cells will have more telomeres and thus will artificially appear, in this single cell assay, to have longer telomeres. The transduction efficiency for the cells lines with medium telomere length was quite low so we looked for other lines. Two other cell lines, with high transduction efficiency, 293FT and HeLa, generated more accurate cell cycle profiles (Figure 3.1c). They had a medium range of telomere lengths, which were comparable to K-562 (Figure 3.1d). Since these cell lines had the telomere length, transduction efficiency and cell cycle profile we were looking for, we decided to use these cell lines in our screens to identify pathways that, when knocked down, change telomere length.

50 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

To test whether the screen might work ,we carried out a pilot experiment. We knocked down POT1 and hTERT, by shRNA, in separate populations of 293FT cells and cultured for 7 weeks. Previous work has shown that 7 weeks is required to detect changes in telomere length, upon POT1 or hTERT knockdown (Counter et al., 1992; Loayza and De

Lange, 2003). POT1 knockdown was expected to lengthen telomeres (Loayza and De

Lange, 2003), and hTERT knockdown was expected to shorten them (Harley et al.,

1990). qPCR indicated that POT1 levels were reduced by approximately 70% and hTERT levels were reduced by approximately 50% (Figure 3.2a). hTERT levels are limiting in the cell (Hathcock et al., 2002), so even modest knockdown was expected to affect telomere length. We used flow-FISH to measure telomere length. Consistent with previous reports, hTERT knockdown significantly shortened telomeres, while POT1 knockdown significantly lengthened them (Figure 3.2b). Importantly, loss of these known telomere regulators generated a robust change in telomere length. The larger the change in telomere lengths that is generated, the stronger the selective power of the screen will be, indicating our selection strategy could go forward.

We analyzed the telomere signals of POT1, hTERT and non-silencing shRNAs on the cell sorter and noted that loss of POT1 generated long telomeres that did not overlap with the lengths in non-silencing, or hTERT shRNA treated cells (figure 3.3c). Moreover, while hTERT knockdown shifted the telomere signal to the left, compared to non- silencing, the effect was not as dramatic as the long telomere cells (figure 3.3a), presumably because there is a hard limit to how short a telomere can become, before a

51 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH cell undergoes senescence or apoptosis. These data demonstrated that the telomere lengths generated by loss of known regulators could by distinguished and sorted by

FACS.

3.3.2 Validating shRNA screening approach

We tested the validity of our screening approach by performing a pilot experiment with known telomere length regulators. We reasoned that if the screen is working, it should accurately distinguish known positive and negative regulators of telomere length.

In this pilot experiment, we transduced a single pool of cells with equal amounts of hTERT, Non-silencing and POT1 shRNA lentivirus. Loss is hTERT is known to shorten telomeres (Harley et al., 1990) and loss of POT1 is known to lengthen them (Loayza and

De Lange, 2003). Each of these shRNAs contains a different stem region, but constant regions flanking the stem. After 7 weeks of culture to allow sufficient length change, we sorted the 7% longest and shortest telomeres by flow-FISH. We collected gDNA from the two sorted populations, as well as the unsorted population as a control. From gDNA, we amplified the integrated shRNA from the flanking constant regions. We then sequenced the amplicons by Illumina sequencing, to determine the relative frequency of cells with hTERT, POT1 or non-silencing shRNA insert in the short, unsorted or long fraction. We found that POT1 shRNA was enriched in the long telomere fraction, and hTERT shRNA was enriched in the short telomere fraction consistent with the known biology of these genes (Figure 3.4a). These data demonstrate that genes with known effects on telomere length, are recovered and enriched in the correct population upon

52 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH flow-FISH sorting and sequencing genomic inserts. These data may underestimate the degree of enrichment, because each shRNA was transduced into 1/3 of cells in the population. Since the population is growing competitively, an shRNA that is found in 1/3 of cells, to start with, can only be enriched by 3 fold as a maximum. However, an shRNA that begins as 1/4675th of a population is capable of enriching significantly more, since it comprises a smaller proportion of the initial fraction.

Given these findings, we felt confident that our screening method would allow us to identify genes that altered telomere length when knocked down.

3.3.3 Pooled shRNA screening

We use both 293FT and HeLa cells, independently, to perform a pooled screen for kinase regulators of telomere length.

In this screen, we used a pooled library of lentivirus, containing shRNAs against human protein kinases and some kinase related genes (GE Dharmacon). This library has been optimized for pooled screening in a viability screen (Strezoska et al., 2012); it contains 4675 shRNAs directed against 706 kinase and kinase related genes. We transduced HeLa and 293FT cells with 500-fold representation of virus at an MOI of 0.1.

This means that on average 500 cells that were transduced with each shRNA. A high representation is required because our screen uses a relatively weak selection. It also provides additional precision in evaluating more subtle changes in shRNA representation.

In this case, only a single round of selection was possible. We cannot use successive rounds of enrichment, for long or short telomeres, because the cells are fixed for flow-

53 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

FISH, and no longer viable. We used a low MOI, so that each cell is transduced with 1 or fewer viruses to avoid complications with more than one shRNA interacting in a cell.

Transductions were performed in triplicate in both cell lines. We cultured these cells for 7 weeks to allow sufficient changes in telomere length to accumulate in the population.

At 7 weeks, we performed flow-FISH and FACS to sort out cells with the 7% shortest and 7% longest telomeres. Unsorted cells were collected as a control. We identified many optimizations to the traditional flow-FISH protocol that enhanced our ability to accurately sort cells at a high speed (figure 3.3). Flow-FISH requires formamide treatment of whole cells to allow hybridization of PNA telomere probes. This treatment causes cells to aggregate, even upon cell straining, and resist DNA staining to saturation.

We found that fixing cells with paraformaldehyde, followed by dehydration in methanol produced a far superior cell cycle profile, compared to traditional 70% ethanol fixation

(figure 3.3a, b). Since we required a 500 fold representation of ~5000 shRNAs, we needed to sort approximately 2.5 million cells in the long and short fractions, which required sorting between 200 and 400 million cells total. Thus we required faster cell sorting to facilitate this. We found that addition of 0.1% SDS to PI staining solution significantly decreased cell aggregation, allowing for a much faster flow rate (figure 3.3b, c).

Since shRNA inserts were integrated into the genome by a lentivirus, we first isolated genomic DNA from sorted populations. Due to flow-FISH treatment, traditional

DNA extraction kits (Qiagen), recommended by Illumina were unable to recover sufficient yields of genomic DNA. We found that phenol-chloroform extraction provided

54 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH much higher genomic DNA yield. To amplify shRNA inserts, we used primers in a flanking common region of the virus to amplify the variable stem region that uniquely identifies the shRNAs. These amplified fragments were then purified and sequenced using Illumina deep sequencing, to determine the relative frequency of each shRNA insert in the long telomere, short telomere and unsorted populations. We aligned Illumina reads to reference shRNA sequences by bowtie2 (Langmead and Salzberg, 2012), and determined enriched genes with MAGeCK analysis (Li et al., 2014).

3.3.4 Analysis of sequencing results from pooled shRNA screen

We aligned Illumina sequencing data to the reference shRNA sequence library using bowtie2, as described (Strezoska et al., 2012). In this workflow, we first determined the number of counts of each shRNA by aligning to the provided reference library (GE

Dharmacon). Next, we normalized these counts, based on the total number of reads in a particular sample. This gave us a normalized list of reads for each shRNA in the short fraction, long fraction and unsorted fraction, in triplicate, for both HeLa and 293FT screens. We next used MAGeCK analysis (Li et al., 2014) to rank genes associated with the short and long fractions. MAGeCK analysis was originally designed for pooled

CRISPR/Cas9 screens, but since it can generate gene rankings from counts tables, we adapted it for our shRNA screen. We reasoned that a gene that is involved in telomere length would create a larger enrichment in the long or short fraction, and have multiple shRNAs that target that gene enriched in that particular fraction. MAGeCK uses the p-

55 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH value, from 3 replicates, and mean enrichment of individual shRNAs to rank genes likely to be involved in telomere length.

In the HeLa screen, we obtained 37 genes enriched in the short fraction and 30 genes in enriched the long fraction that were statistically significantly (p<0.05). In

293FT, we recovered 41 genes enriched in the short fraction and 31 genes enriched in the long fraction that were statistically significantly (p<0.05). The top 20 genes in long and short fractions for HeLa (Table 3.1) and 293FT (Table 3.2) are listed below.

There are relatively few “gold standard” kinases with well-established roles in mammalian telomere length regulation that we could use as positive controls for enrichment. Some positive kinase regulators have been identified but virtually no negative ones have been described. Therefore, we were unable to refer to a known kinase regulator of telomere length, as a benchmark of enrichment for bona fide telomere length regulators. We did however note that the top hit in the 293FT short fraction, MAPK15, was identified, in a previous screen, to lower telomerase activity when it was knocked down (Cerone et al., 2011),. This demonstrates that we were able to recover at least one known kinase involved in telomere length regulation.

Upon manual inspection, some genes had significant variation between biological replicates. This was more pronounced in 293FT, than HeLa. This may be explained by the fact that each control is the original unsorted sample, from which the long or short fraction was sorted. This control sample controls for varying growth rates and jackpot events in that population, but this could be very different in the other biological replicates. Therefore, each replicate of control, controls best for the experimental

56 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH replicate that was derived from it. Thus, we used a paired version of MAGeCK analysis to account for this. In this version, we performed MAGeCK on each individual replicate and it’s corresponding control, and then averaged the ranks of each gene to come up with a final ranking. This analysis yielded significantly different data for the 293FT cells, with nearly half of the top 20 changing, while it did not significantly affect the HeLa cells

(Table 3.1, Table 3.2). We referred to the default MAGeCK analysis as “unpaired”, and the modified paired version as “paired”. We also found a significant discordance between genes identified by HeLa and 293FT, by both paired and unpaired analysis methods. This could be due to differences in cell lines, such as mutations altering particular kinase pathways. Another potential source of error is the accuracy of the cell cycle profile. We use PI staining to differentiate G1 cells, so we only select telomere length within cells with the same DNA content. 293FT cells were significantly more variable than HeLa in the cell cycle staining. Suggesting that some of the variability in 293FT may be explained by contaminating S and G2/M cells. Because the HeLa cell line showed more consistency between replicates, we decided to favor it in selecting candidates to further investigate.

We determined the list of candidates to further investigate by comparing their ranking using the paired and unpaired method, exclusivity to one fraction, broad tissue expression and known biological function. We used exclusivity to one fraction as a criterion because we noted many top hits came up in both the long and short fraction.

Loss of telomere regulators is expected to shift mean telomere length, but not alter the breadth of different lengths. Therefore we expected true telomere regulators to be enriched in only the long or short fractions, but not both. Since there were a many of

57 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH genes enriched in both fractions, we concluded they were likely due to PCR bias and did not represent those genes that broadened telomere length distribution. We instead focused on genes that were enriched exclusively in the long or short fraction, but not both. Tissue expression of candidate kinases was determined through the Human Protein Atlas. Broad tissue expression was used as a criterion because telomerase is active in the stem cells of many tissue types (Hiyama et al., 1995). Therefore, we reasoned that genes, which were important in regulating telomere length, would be expressed in a broad array of tissues.

For example, a gene, which is expressed exclusively in neurons, is unlikely to have a global role in telomere length regulation. To determine biological function, we manually inspected candidates and evaluated their function through published literature. We favored genes with nuclear localization and association with known telomere pathways such as DNA damage, cell cycle regulation, checkpoint regulation, DNA replication and chromatin modification.

3.3.5 Pooled CRISPR Screening approach

Having established the best methods to carry out the flow-FISH screen, we wanted to use these methods to test the next generation of genome editing libraries. shRNA libraries were very successful in past screens. CRISPR libraries recently became available and might offer a complementary approach to identify genes involved in telomere length regulation. We performed a pooled CRISPR screen to complement the pooled shRNA screening approach. CRISPR has the advantage that it will completely knock out a gene by disrupting the genomic locus. shRNA, on the other hand, can have a

58 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH variable level of knock down, depending on the shRNA. This can be advantageous if the gene is essential, because a partial loss will produce a viable phenotype. Alternatively,

CRISPR produces gene knockouts, and enrichment in a population is may due to complete loss, rather than partial loss of function. Pooled CRISPR screens have been used successfully to demonstrate the feasibility this approach (Shalem et al., 2014; Wang et al., 2014). In addition, libraries have been developed that are enriched for sgRNAs against particular groups of genes, including kinases. We used a kinase CRISPR sub pool containing 5070 sgRNAs against 507 human kinases (Wang et al., 2014).

We generated a cell line of 293FT cells, which constitutively expressed Cas9. We chose 293FT because this line had the most dynamic range of telomere lengths, based on previous experiments (Figure 3.2b). We generated this cell line by transducing 293FT cells with lentiCRISPR V2 eGFP, a lentiviral vector encoding Cas9-2a-eGFP. We then sorted cells with high eGFP expression, and isolated clones by serial dilution. We used western blotting of 6 clones with high GFP expression, to select the one with the highest

Cas9 expression.

The CRISPR screen was performed in a similar manner to the shRNA screen. It was carried out in triplicate in the Cas9-eGFP 293FT cell line. We performed this screen at 100-fold representation of sgRNAs, because we expected fewer sgRNAs to be enriched, since essential genes would be expected drop out entirely if they were homozygously deleted. 100 and even 20 fold representation of pooled screens have been shown to produce the same results as 500 fold representation screens (Strezoska et al.,

2012). The lower representation allowed us to sort the shortest and longest 6% of cells

59 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH instead of 7%, which provided more stringency in selection. We used primers specific to flanking regions of the sgRNA to amplify sgRNA inserts from genomic DNA. PCR products were sequenced by Illumina sequencing and enrichment was determined using

MAGeCK analysis (Li et al., 2014) as described below.

3.3.6 Analysis of sequencing results from pooled CRISPR screen

We analyzed CRISPR screen results in similar manner to our shRNA screen.

MAGeCK contains alignment algorithms as well, so we were able to directly analyze

FASTQ files from Illumina sequencing, rather than first align with bowtie2. We performed both paired and unpaired analysis, as with the shRNA. The top 20 hits from the short and long, MAGeCK analysis are listed in table 3.3.

We saw less concordance between biological replicates, compared to the shRNA screen. This may be explained by the lower 100-fold representation that we chose, compared to 500-fold in the shRNA. Furthermore, CRISPR relies on NHEJ of the genomic locus to knockout all alleles of a particular gene. Since 293FT cells have variable ploidy, it is possible that some cells did not have homozygous deletions. This would lead to an effective lower representation, because only a fraction of transduced cells would knockout the gene.

The CRISPR screen also had poor concordance with the 293FT and HeLa shRNA screen. Only one short fraction hit, out of the top 20, was shared with HeLa short, and 1 long fraction hit, out of the top 20, was shared with 293FT long.

60 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

We decided on candidates to further validate in the same manner as in the shRNA screen

(3.3.4), by evaluating ranking in paired and unpaired analysis, exclusive enrichment in one fraction, broad tissue expression and known biological function. We gave genes that were shared between the shRNA and CRISPR screen the highest priority. We then selected candidates, which had not been tested from the shRNA screen, based on availability of drug inhibitors, known function and broad tissue distribution.

3.3.8 Summary

Here we have developed a versatile method of identify novel genes that regulate telomere length. We treated cells will pooled lentiviral libraries of either shRNA or sgRNAs, and then sorted cells based on telomere length by telomere flow-FISH. We amplified shRNA or sgRNA inserts in these cells and sequenced them by Illumina sequencing. By comparing frequency of shRNA or sgRNA reads in the long or short fraction, compared to the unsorted fraction, we were able to define putative negative and positive regulators of telomere length.

While there are limitations to this assay, particularly the relatively week selective power of flow-FISH, it is highly versatile. We were unable to perform multiple rounds of culture and flow-FISH for long or short telomere cells, because flow-FISH requires fixing cells. Here we focused on knocking out or knocking down kinases. However, this method can be used for other kinds of libraries, or even the whole genome. This provides

61 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH a powerful method to discover more novel telomere length regulators in future experiments.

3.3 Methods

3.3.1 shRNA knockdown

We knocked down expression of hTERT (V3LHS_340163) and POT1 (RHS4430-

101129426) using pGIPZ lentiviral shRNA vectors (GE Dharmacon). A non-silencing shRNA (GE Dharmacon, RHS4346) was used as a negative control. We generated lentivirus, and transduced cells as previously described (2.3.2). Cells were selected with 2

μg/mL puromycin for 72 hours, to eliminate untransduced cells.

3.3.2 qPCR Analysis

Gene expression of hTERT and POT1 were evaluated 7 weeks post transduction. 7 weeks is required for sufficient changes in telomere length. qPCR was performed as previously described (2.3.7), using primers against hTERT and POT1. ARF3 was use as a housekeeping gene to normalize for cell number.

3.3.3 Telomere flow-FISH analysis

Telomere length was determined by flow-FISH, as previously described (2.3.8), with some modifications. Rather than fixation with ethanol, cells were fixed at room temperature in 1.5% paraformaldehyde for 10 minutes, followed by overnight incubation in 100% methanol.

62 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

3.3.4 Pooled shRNA kinase library

We used the Decode Pooled Human GIPZ Kinase Library to transduce cells (GE

Dharmacon). This is a library of lentiviral particles containing shRNAs, on the pGIPZ vector, against human kinases and kinase related genes. The library contains 4675 shRNAs against 709 genes. This library can be used in pooled screens, with either positive or negative selection (Strezoska et al., 2012).

3.3.5 Pooled CRISPR kinase library

Pooled CRISPR kinase library was a gift from Eric Lander and David Sabbatini

(Addgene #51044). This library is a sub-pool, enriched for kinases, of a larger whole genome CRISPR library (Wang et al., 2014). It contains 5070 sgRNAs against 507 genes, on the pLX-sgRNA backbone. This backbone does not contain Cas9, which was expressed constitutively from the cell line.

3.3.6 Generating constitutive Cas9 expressing cell line

We generated a constitutive Cas9 expressing cell line, in which we performed the pooled

CRISPR screen. The CRISPR sub-pool backbone contained blasticidin resistance, therefore we needed a cell line that was blasticidin sensitive. We transduced 293FT cells with lentiCRISPR eGFP V2. This plasmid contained Cas9-FLAG-2a-eGFP in its expression cassette. This plasmid was a gift from Valina Dawson. We sorted transduced cells by FACS (Becton Dickinson) and isolated a population of cells with medium high

63 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

GFP. Extremely high GFP expression is toxic to cells so we selected a population with moderate-high expression. These cells were counted, by haemocytometer, and plated at limiting dilutions in 96-well plates. Individual colonies were isolated, and Cas9 protein level was determined by western blot. The cell line with the highest Cas9 expression level was selected for use in the screen.

3.3.7 Western Blotting

Protein was isolated by lysis with RIPA buffer (Cell Signaling, #98016S). Protein concentration was determined by BCA analysis (Thermo #23227). 30 μg of protein was run on a Novex 4-12% Bis-Tris gel (Thermo Scientific) at 200 V for 1 hour, then transferred to a nitrocellulose membrane for 1 hour at 30 V in an XCell II Blot Module

(Life Technologies). The membrane was blocked in Odyssey blocking buffer (LI-COR,

#927-40000) for 1 hour at RT. The membrane was then incubated in with the following primary antibodies at 4°C: goat anti-actin (Santa Cruz sc-1616), mouse anti-FLAG

(Sigma F3165). The membrane was washed 5 times for 10 minutes with 1X TBST, then cut in half to separate larger proteins, where Cas9-FLAG was predicted to run, and smaller proteins, where actin was predicted to run. The membranes were then incubated for 2 hours at room temperature in Odyssey blocking Buffer (LI-COR) with the following conjugated secondary antibodies: IRDye® goat anti-mouse 800 for the larger protein membrane, IRDye® donkey anti-goat 680 for the smaller protein membrane (LI-

COR). The membranes were then washed 5 times for 10 minutes with 1X TBST, then 2

64 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH times for 5 minutes in TBS. Blots were scanned using Odyssey near-infrared scanner (LI-

COR).

3.3.8 Transduction and tissue culture of transduced cells

Transduction of the both shRNA and CRISPR libraries was carried out at MOI=0.1, as previously described (2.3.2). Assuming a Poisson distribution of transduction events, an

MOI=0.1 will generate >98% of cells with 1 or fewer lentiviral integrations. In the shRNA screen, transduction was performed at 500 fold representation, while in the

CRISPR screen, transduction was performed at 100 fold representation. Transduced cells were selected with either 2 μg/mL of puromycin for 3 days, in the case of the shRNA screen, or 10 μg/mL of blasticidin for 7 days, in the case of the CRISPR screen. To ensure fold representation was maintained, cells were cultured such that the population never dropped below the number of originally transduced cells. Cells were cultured, in

DMEM media (Gibco) with 10% fetal bovine serum (Gibco) and penstrep glutamine

(Gibco), for 7 weeks in to allow sufficient telomere length change.

3.3.9 Telomere flow-FISH cell sorting

We adapted flow-FISH, use for analyzing telomere length (Baerlocher et al., 2006), to sort cells based on telomere length. A large number of cells were required, in order to collect enough genomic DNA to maintain appropriate representation of each integrated shRNA or sgRNA. A number of optimizations were required to enhance flow cytometry efficiency and sorting accuracy, in order to get enough cells. First, 4 x107 were

65 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH trypsizinied with 0.05% Trypsin (Gibco) for 5 minutes at 37°C, then washed in PBS

(Gibco). Cells were passed through a 70 μM cell strainer (Corning) to eliminate aggregated cells. Cells were pelleted, resuspended in PBS (Gibco) and added dropwise to

1.5 % paraformaldeyhyde to a final concentration of 500,000 cells/mL. Cells were incubated with shaking for 10 minutes are room temperature, passed through the cell strainer again, then resuspended in ice cold methanol at a concentration of 500,000 cells/mL. Cells were incubated for 1-3 weeks at 4°C in methanol.

To hybridize probe, cells were first passed through a cell strainer, then washed in PBS with 1% BSA (Sigma). Cells were then resuspended in hybridization solution containing

FITC telomere probe, at a concentration of 20 million cells/mL, as described (Baerlocher et al., 2006). Probe was hybridized by incubation at 87°C for 15 minutes, then room temperature overnight.

Cells were washed as described in Baerlocher et al., 2006, and resuspended in modified

PI staining solution, at a concentration of 5 million cells/mL. Modified PI staining solution contains PBS with 0.1% Triton X-100, 200 μg/mL RNase A (Sigma), 20 μg/mL propidium iodide (Sigma) and 0.1% SDS. Cells were incubated for 30 minutes, protected from light, before cell sorting

3.3.10 FACS of cells by telomere length

66 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Cell sorting was performed on a MoFlo cell sorter as described (Becton Dickenson,

2.3.6). First, we gated on single cells by selecting singlets by pulse width, then gated on

G1 based on PI staining, then finally sorting out cells with the longest and shortest 7% of telomere lengths, based on FITC telomere signal. A minimum of 1 million cells were collected in each fraction to collect genomic DNA. A population of unsorted cells were also collected for genomic DNA.

3.3.11 Illumina sample preparation and sequencing

Genomic DNA was harvested from sorted cells by phenol-chloroform extraction

(Sambrook et al., 1989). Phenol chloroform extraction yielded far higher genomic DNA concentration than Puregene DNA extraction (Qiagen), likely due to the harsh formamide treatment during probe hybridization. For shRNA and CRISPR library treated cells, the variable region, of the genomic insert, was amplified with primers annealing to flanking common regions, as described (Strezoska et al., 2012; Wang et al., 2014). Primers contained indices to allow multiplexing of samples in a single Illumina lane. Amplicons were visualized by gel electrophoresis on a 1% TAE agarose gel, and purified with

Agencourt AMPure XP Beads (Beckman Coulter)

PCR products were sequenced on an Illumina HiSeq 2000 to obtain single end 50 base pair reads. Between 3 and 9 samples were multiplexed on a single lane, depending on quantity of PCR product.

67 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

3.3.12 shRNA screen sequencing analysis

Illumina reads from the shRNA screen were analyzed as described (Strezoska et al.,

2012). Reads are first parsed by barcode to separate multiplexed samples. Next, they are aligned to a reference genome, of kinase shRNAs, by bowtie2, as described (Langmead and Salzberg, 2012). These alignments generate count tables with numbers of reads of each shRNA in each sample. We then use MAGeCK analysis to rank gene enrichment, based on the enrichment of individual shRNAs (Li et al., 2014). We performed MAGeCK analysis in a paired and unpaired manner. In the paired method, we performed MAGeCK on each individual sample, then determined the mean rank of each gene by averaging the rank of all 3 replicates. In the unpaired method, we determined gene rankings by comparing the mean number of reads in 3 experimental samples, compared to the mean number of reads in 3 control samples. From these rankings we manually selected genes in the long and short telomere fraction to further characterize.

3.3.13 CRISPR screen sequencing analysis

Illumina reads from the CRISPR screen were analyzed as described (Wang et al., 2016).

The method was very similar to the shRNA screen analysis, except we used MAGeCK software to align sgRNA reads to a custom sgRNA reference genome. From this,

MAGeCK determined read numbers for each sgRNA, then ranked genes based on enrichment, p-value and number of sgRNAs against that gene that were enriched. Like the shRNA analysis, this was also performed in a paired and unpaired manner.

Primers

68 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 3.1 Selecting cell line to perform pooled telomere length screen

A) Transduction efficiency of indicated cell lines, measured by percent GFP+ after 72 post transduction with FUGW at MOI=1.

B) Telomere length of indicated cell lines, measured by telomere flow-FISH. Bovine thymocyte had a known telomere length and served as a control for telomere length.

C) Flow cytometry plot of cell cycle profile of K-562, 293FT and Hela, following staining with PI

D) Telomere length of K-562, 293FT and HeLa, measured by flow-FISH

69 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 3.1

A 40 35 30 25 20 15

10 Percent Percent GFP+ 5 0

B 900 800 700 600 500 400 300 200 100

0 Telomere Telomere Signal (a.f.u.)

C 150 K-562 D Flow-FISH Telomere Length

s 50

l l

e 100 C

l 40

a n

50 g

i S

43 e

r e 0

m 20

0 200 400 600 800 1000 o l

FL2-H: PI e T 10 200 250 0 2 a T 6 L F 200 5 e 3 150 - H 9

293FT HeLa K 2

s s

l l l l

e e 150

C C Cell Line

# 100 #

100

50 50 58.3 65.1

0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 FL2-H: PI FL2-H: PI

70 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 3.2 Knockdown of hTERT or POT1 shortens, or elongates telomeres respectively, in 293FT cells

A) Q-PCR of POT1 or hTERT levels, following shRNA knockdown, in 293FT cells

B) Telomere length of 293FT cells following hTERT or POT1 knockdown, measured by flow-FISH

C) Overlay of flow-FISH telomere signals of hTERT, non-silencing and POT1 shRNA.

71 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 3.2

A 293T qRT-PCR 293T qRT-PCR

1.0 n 1.0

e p

T 0.5

0.5

T O

h P

0.0 0.0 g n A g A ci N in N n R c R e h n h il s le s -s T i n -S 1 o R n T E o O N T N P h B 6 Week 293T Flow-FISH

l 100

n g

e r

e 50

m o

e T

0 g A A in c N N n R R e h h il s s -s T 1 n R T o E O N T P

72 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 3.3 Optimizations for fluorescence activated cell sorting and flow-FISH

Adaptations to the traditional telomere flow-FISH protocol were made to enhance DNA staining and cell cycle profile (A) and decrease aggregation to increase cell sorting speed and accuracy.

A) PI staining of HeLa cells fixed with 70% ethanol

B) PI staining of HeLa cells fixed with paraformaldehyde and methanol

C) Flow cytometry of PI signal for HeLa cells stained with traditional PI staining solution

(Pulse width vs Area, circled population represents singlets)

D) Flow cytometry of PI signal for HeLa cells stained with 0.1% SDS PI staining solution (Pulse width vs Area, circled population represents singlets)

73 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 3.3

A B

C D

74 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 3.4 POT1 and hTERT shRNA inserts are recovered in the expected fractions upon sorting long and short telomere cells

A) The percentage of the short fraction, unsorted fraction or long fraction that contained hTERT, non-silencing, or POT1 inserts. hTERT reads are enriched in the short fraction because loss of hTERT causes short telomeres, which enriches the short fraction for cells with hTERT reads. POT1 reads are enriched in the long fraction because loss of POT1 causes long telomeres, which enriches the long fraction for cells with POT1 reads.

75 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Figure 3.4

A shRNA Enrichment in Sorted Fractions

1.0 hTERT shRNA

0.8 Non-silencing

n POT1 shRNA

t 0.6

r 0.4

P 0.2

0.0 rt k g o l n h u o S B L Sorted Fractions

76 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Table 3.1 HeLa shRNA screen top 20 hits

A – Short Fraction B – Long Fraction

Rank Unpaired Paired Rank Unpaired Paired 1 GUCY2D GUCY2D 1 STK16 ERBB3 2 TEC PIK3R4 2 PRKCG STK16 3 CDC42BPG TEC 3 TEC IKBKE 4 ANKK1 CDC42BPG 4 ANKK1 MAPK14 5 EIF2AK1 PIK3R1 5 RET TEC 6 PCK1 PRKCG 6 PRAGMIN PRKCG 7 PIK3R4 CDK6 7 PI4KB PRAGMIN 8 TRIO EIF2AK1 8 ITK LMTK2 9 ILK ILK 9 KCNH2 ANKK1 10 ITPKA ANKK1 10 IKBKE ROR1 11 STK16 CSK 11 MAPK14 ITK 12 CSK PCTK1 12 DGKB CDC42BPG 13 CLK1 CLK1 13 CDC42BPG STK17A 14 PKN3 MAP3K11 14 PRKACV PRKACA 15 PIK3R1 MASTL 15 ROR1 PI4KB 16 MASTL STK16 16 MAP3K15 PIK3R1 17 CAM2N1 CAMK2N1 17 TAOK1 MAP3K11 18 WNK2 BRD4 18 NEK5 STK32A 19 LATS2 NEK5 19 PIK3R1 ADPGK 20 BRD4 TRIO 20 PFTK1 AK7

77 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Table 3.2 293FT shRNA screen top 20 hits

A – Short Fraction B – Long Fraction

Rank Unpaired Paired Rank Unpaired Paired 1 MAPK15 STK10 1 FES LIMK1 2 TAOK2 SRMS 2 DAPK2 ADCK1 3 CDK3 PKMYT1 3 NUP62 PI4KA 4 STK10 CDKN2DD 4 ADCK1 CAMK2A 5 PNCK MAP3K10 5 KALRN KALRN 6 HK2 PDGFRL 6 RPS6KA5 CDC2L1 7 PDGFRL RPS6KA4 7 MAPK15 ACVR1B 8 EPHB2 AURKB 8 PDK4 DYRK1B 9 MAP3K11 PKN3 9 BCR VRK3 10 MAPK12 CALM3 10 DYRK1B TYK2 11 CDKN2D CSK 11 PI4KA INSRR 12 NME3 TK2 12 LIMK1 LCK 13 STK16 PANK4 13 CHEK1 MAPK15 14 MAPK11 WNK2 14 IRAK2 PANK1 15 CSNK1A1 PIM3 15 ACVR1B MAP3K12 16 RELA STK3 16 LCK GRK6 17 RPS6KA2 STK16 17 TK2 CDK4 18 INSRR STK35 18 DDR1 SPHK2 19 CALM3 CDK4 19 MYO3B TSSK6 20 CDKN1C TSSK6 20 CSNK2A2 EIF2AK2

78 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Table 3.3 CRISPR screen top 20 hits

A – Short Fraction B – Long Fraction

Ran Unpaired Paired Rank Unpaired Paired 1 HIPK2 MAPK10 1 NEK4 NEK9 2k PIK3R4 HIPK2 2 FLT4 CSNK1A1L 3 TBK1 PIK3R4 3 CSN1A1L PRKCE 4 CAMK2B FGFR4 4 GUCY2C IGF1R 5 GRK7 GRK7 5 PIM2 FLT4 6 MAPK8 MAPK8 6 GRK1 NEK3

7 DCLK1 TSSK3 7 MAPK15 TSSK6 8 PRKD2 PKMYT1 8 PRKCE ZAK 9 TAOK2 CDC42BPB 9 LRRK2 PRKG2 10 STK3 GRK4 10 PRKG2 MAP4K5 11 AATK SMG1 11 EIF2AK3 GRK7 12 NEK6 MAST2 12 NEK9 PIK3R4 13 RIOK1 CDK13 13 TYRO3 NEK10 14 PRKCB RIOK1 14 OBSCN TBK1 15 FLT4 AATK 15 TSSK4 TGFBR2

16 SMG1 CSNK1E 16 GRK7 NEK11 17 MAPK10 KIT 17 MAP3K9 PIM2 18 CDC42BP PRKD2 18 NEK10 MINK1 19 PKMYT1 DMPK 19 MKNK2 TEK B 20 BRAF CDKL1 20 ADRBK2 ROCK1

79 CHAPTER 3. IDENTIFYING NOVEL KINASE REGULATORS OF TELOMERE LENGTH BY POOLED SCREENING APPROACH

Table 3.4 Primer List

Pooled shRNA Screen Primer Sequence Decode Forward AATGATACGGCGACCACCGAGATCTACACCGGTGCCTGAG PCR Primer TTTGTTTGAA Decode Reverse CAAGCAGAAGACGGCATACGAGATGGCATTAAAGCAGCGT PCR Primer ATCCAC Decode Illumina GAAGGCTCGAGAAGGTATATTGCTGTTGACAGTGAGCG Sequencing Primer

Pooled CRISPR Screen Primer Sequence Sequencing Primer CGGTGCCACTTTTTCAAGTTGATAACGGACTAGCCTTATTTT AACTTGCTATTTCTAGCTCTAAAAC Indexing Primer TTTCAAGTTACGGTAAGCATATGATAGTCCATTTTAAAACA TAATTTTAAAACTGCAAACTACCCAAGAAA Reverse Primer AATGATACGGCGACCACCGAGATCTACACCGACTCGGTGC CACTTTT Forward Primer1 CAAGCAGAAGACGGCATACGAGATCATCACGTTTCTTGGG TAGTTTGCAGTTTT Forward Primer2 CAAGCAGAAGACGGCATACGAGATCCGATGTTTTCTTGGGT AGTTTGCAGTTTT Forward Primer3 CAAGCAGAAGACGGCATACGAGATCTTAGGCTTTCTTGGGT AGTTTGCAGTTTT Forward Primer4 CAAGCAGAAGACGGCATACGAGATCTGACCATTTCTTGGG TAGTTTGCAGTTTT Forward Primer5 CAAGCAGAAGACGGCATACGAGATCACAGTGTTTCTTGGG TAGTTTGCAGTTTT Forward Primer6 CAAGCAGAAGACGGCATACGAGATCGCCAATTTTCTTGGG TAGTTTGCAGTTTT Forward Primer7 CAAGCAGAAGACGGCATACGAGATCCAGATCTTTCTTGGG TAGTTTGCAGTTTT Forward Primer8 CAAGCAGAAGACGGCATACGAGATCACTTGATTTCTTGGGT AGTTTGCAGTTTT Forward Primer9 CAAGCAGAAGACGGCATACGAGATCGATCAGTTTCTTGGG TAGTTTGCAGTTTT Forward Primer10 CAAGCAGAAGACGGCATACGAGATCTAGCTTTTTCTTGGGT AGTTTGCAGTTTT Forward Primer11 CAAGCAGAAGACGGCATACGAGATCGGCTACTTTCTTGGG TAGTTTGCAGTTTT Forward Primer12 CAAGCAGAAGACGGCATACGAGATCCTTGTATTTCTTGGGT AGTTTGCAGTTTT

80 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Chapter 4. Chemical inhibition of BRD4, CK1 and MEK/ERK

block telomere elongation

4.1 Introduction

Replicative immortality is a hallmark of cancer (Hanahan and Weinberg, 2011).

Most adult mammalian cells do not express telomerase, and will eventually undergo replicative senescence as telomeres shorten (Harley et al., 1990). One of the most common ways cancer cells overcome this is by activating telomerase expression (de

Lange, 1994; Heidenreich et al., 2014). Telomerase is able to maintain telomeres such that they are not recognized as DNA damage, and the cell is able to divide indefinitely.

Thus, developing telomerase inhibitors, or inhibitors of telomere elongation, are an area of significant investigation. Here we identify BRD4, CK1 and the MEK/ERK pathway as positive regulators of telomere length. Identification of these pathways generates new potential targets for blocking telomere elongation in cancer.

BRD4 is a BET family protein. BET family proteins are characterized by bromodomains, which bind to acetylated lysines on histones. BRD4 has both histone acetyl transferase activity (Devaiah et al., 2016a), and kinase activity (Devaiah et al.,

2012). BRD4 has previously been found to be required for the growth of acute myeloid leukemia (Zuber et al., 2011) and other cancers (French et al., 2003). Many BRD4 inhibitors have entered various stages of clinical trials (Boi et al., 2015); NCT02259114;

NCT01587703; NCT02308761; NCT01949883; NCT02157636. Despite it’s therapeutic

81 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION consideration, relatively little is known about BRD4’s mechanism of action. BRD4 inhibition blocks MYC transcription (Mertz et al., 2011), but this is unlikely to account for all of its effect, since it is more efficacious than traditional MYC inhibitors.

BRD4 has pleiotropic roles associated with its role as a scaffold, kinase and histone acetyl transferase (Devaiah et al., 2016b). BRD4 acts as a mitotic “bookmark”, by acting as a scaffold for transcription factors on promoters of genes to be expressed after mitosis (Dey et al., 2009; Zhao et al., 2011). It also directly regulates transcription by recruiting pTEFb (Yang et al., 2005) and phosphorylating the C-terminal domain of RNA

Polymerase II (Devaiah et al., 2012). BRD4 uses its histone acetyl transferase activity to evict nucleosomes from chromatin, through acetylation of H3 and (Devaiah et al., 2016a).

It has also been found to affect cell cycle progression, by regulating Aurora Kinase B

(You et al., 2009).

In addition to BRD4, we also found the inhibition of CK1 and MEK/ERK, blocked telomere elongation. The CK1 family of kinases has diverse cellular roles including DNA repair, proliferation, cytoskeleton regulation, trafficking, apoptosis, and differentiation (Knippschild et al., 2014). It’s broadly expressed in mammalian tissues, but has multiple isoforms with distinct regulatory roles (Knippschild et al., 2005). It has been shown to activate p53 during stress or DNA damage (Dumaz et al., 1999). CK1 also regulates diverse pathways including Wnt, Hippo and Hedgehog (Knippschild et al.,

2014). It is an active target of research in cancer, partially due to its role in regulating p53, but also its role in the aforementioned pathways. Identifying isoform specific roles and developing isoform specific drugs, also remains an active area of investigation.

82 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

The MAPK kinase pathway is a well-studied and diverse signaling pathway, which communicates extracellular information to the cell. Ligand binding to a cell surface receptor activates Ras, which activates Raf, which phosphorylates MEK, which phosphorylates ERK (Chin et al., 1999). This activates a variety downstream effects related to proliferation, differentiation and inflammation. Constitutive activation of the

MAPK is a common source of unregulated proliferation in cancer (Osborne et al., 2012).

Many inhibitors targeting this pathway are in various stages of clinical trial, or already being used to treat cancer. The MAPK pathway has also been found to regulate telomerase transcription (Ge et al., 2006) and expression levels of telomere binding proteins (Picco et al., 2016).

Here we identify BRD4, CK1 and MEK/ERK as candidate positive regulators of telomere length.

4.2 Results and Discussion

4.2.1 Telomeres are rapidly elongated by telomerase overexpression

We set out to validate shRNA and CRISPR screen results with an independent assay. We initially considered knocking out candidate genes by CRISPR/Cas9, and evaluating telomere length. Telomere length changes typically take greater than 6 weeks to detect by Southern blot. Thus, this method would limit the number of genes we could evaluate, particularly as a first pass to validate screen results. We decided to evaluate candidates based upon their ability to block telomere elongation, upon chemical

83 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION inhibition. This assay has been previously used to demonstrate the requirement of ATM for telomere elongation (Lee et al., 2015). In this assay, mouse fibroblasts are transduced with a SVA, a telomerase lentivirus, which overexpresses telomerase. In the presence of vehicle alone, SVA causes robust telomere elongation after just 6 days. However, upon chemical inhibition of a protein such as ATM, which is required for telomere elongation, no telomere elongation is observed.

While this assay limited us to evaluating candidates enriched in the short fraction of our screen, it is robust, fast and generates a clear answer. Furthermore, we were able to detect partial elongation blockage, by observing telomeres with partial elongation. An additional benefit of this assay is that telomerase is overexpressed from a randomly integrated lentivirus, so its transcription will be unaffected by factors that regulate endogenous telomerase transcription. Many proteins have been previously found to affect

TERT transcription (Kang et al., 1999; Wu et al., 1999). This assay will exclusively detect proteins that regulate telomere length at the post-transcriptional level.

We have previously observed that this assay is best suited to testing chemical inhibitors, rather than siRNA knockdown. This may be due to the significant telomerase overexpression, which may overcome knockdown of a positive regulator by mass action.

Chemical inhibitors, however, can be present at much higher concentrations in the cell, and provide much greater inhibition of target proteins, compared to siRNA. Since our screen focused on kinases, many of our candidates had well-established chemical inhibitors, which we used to validate the candidate telomere length regulators..

84 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

4.2.2 BRD4 inhibition blocks telomere elongation

BRD4 was identified as a potential positive regulator of telomere length in our shRNA screen. It is a bromo-domain containing protein with histone acetyl transferase activity, as well as kinase activity (Devaiah et al., 2012). It is known to regulate gene transcription, by modifying chromatin at promoters, or acting as a scaffold for additional transcription factors (Rahman et al., 2011). It has also been found to regulate phosphorylation of the C-terminal domain of RNA Polymerase II (Devaiah et al., 2012).

Inhibition of BRD4 with 0.1 μM JQ1, a well-established BRD4 inhibitor

(Filippakopoulos et al., 2010), blocked telomere elongation by SVA (figure 4.1 a).

Telomere length 6 days after transduction with SVA was similar to day 2, indicating that

BRD4 inhibition blocked telomere elongation in this context. The magnitude of this blockage was similar to the result seen with 10 μM KU-55933, a potent ATM inhibitor

(figure 4.1 a). We next examined additional BRD4 inhibitors; including OTX015, I-

BET151 and MS436 and found these also blocked SVA induced telomere elongation

(figure 4.1 b, c). Since inhibition was seen with four independent BRD4 inhibitors, it is likely that the blocking of telomere elongation is in fact due to BRD4 inhibition rather than off target effect.

To examine whether JQ1 inhibition of telomere elongation was dose dependent, we tested several concentration of this drug. As previously observed, 100 nM JQ1 blocked nearly all SVA mediated elongation (figure 4.2). However, 11 nM JQ1 had virtually no effect on telomere elongation, while 33 nM JQ1 produced an intermediate telomere length. This demonstrates that JQ1’s effect, on blocking telomere elongation,

85 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION titrates with its concentration. This is further underscores the conclusion that inhibiting

BRD4 is what is responsible for blocking telomere elongation.

4.2.3 CK1 inhibition blocks telomere elongation

CK1 was identified as potential positive regulators of telomere length in our pooled CRISPR screen. We tested the effect, of chemically inhibiting this kinase, on telomere elongation by SVA. We used three different CK1 inhibitors: D4476, 5-

Iodotubercidin and IC261. We also used KU-55933, an ATM inhibitor, as a positive control for blocking telomere elongation.

Inhibition of CK1 by 10 μM D4476 blocked telomere elongation (figure 4.3 a), to a comparable degree as 10 μM KU-55933, an ATM inhibitor, which is known to inhibit telomere elongation (Lee et al., 2015). This indicated that loss of CK1 strongly inhibited telomere elongation. However, 0.5 μM 5-Iodotubercidin or 0.5 μM IC261, which are also

CK1 inhibitors, did not affect telomere elongation (figure 4.3 a). Therefore, either the effect we saw with D4476 was an off target effect, or 5-Iodotubercidin and IC261 did not block CK1 as strongly as D4476 did. We observed that D4476’s effect was concentration dependent, as lower concentrations of D4476 caused less blockage of telomere elongation

(figure 4.3 b). Notably, even 0.37 μM D4476 was able to block some telomere elongation, compared to DMSO control. These data indicate to us that CK1 may be a potential positive regulator of telomere length.

86 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Because only 1 out of 3 CK1 inhibitors we tested blocked telomere elongation, we cannot definitvely conclude that CK1 is required for telomere elongation. However, we have shown that CK1 inhibition it is sufficient to block telomere elongation in this assay. It may be that the difference observed between different inhibitors, is due to variable degree of CK1 inhibition. Firstly, 5-Iodotubercidin is not a specific CK1 inhibitor. It was originally identified as an adenosine kinase inhibitor (Phillis and Smith-Barbour, 1993).

It was only later identified to affect CK1 (Massillon et al., 1994), however, the off target effects were broad, and its inhibition of CK1 has not been characterized or validated.

D4476 is a more potent inhibitor of CK1 than IC261, with an in vitro IC50 of 200 nM, compared to an IC50 of 16 μM for IC261 (Selleckchem). IC261 was also more toxic to mouse fibroblasts than D4476, as 2.5 μM IC261 caused significant cell death, while

D4476 was tolerated at nearly 50 μM (data not shown). Given the higher in vitro IC50 and lower tolerated in vivo concentration, it is may be that that IC261 caused less CK1 inhibition than D4476 in this assay. D4476 has been shown to be a superior CK1 inhibitor in vivo using a variety of downstream effects (Bryja et al., 2007; Rachidi et al.,

2014; Rena et al., 2004). D4476 also inhibits all CK1 isoforms while IC261 only targets some (Rena et al., 2004), which may point to telomere specific isoforms of CK1. Given these data, it may be that the greater telomere elongation blockage of D4476 is due to its greater specificity for Ck1 inhibition.

4.2.4 MEK1/2 and ERK1/2 inhibition blocks telomere elongation

87 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

We identified a number of components of the MAPK pathway that affected telomere elongation, in our pooled shRNA and CRISPR screens. MEK1/2 and ERK1/2 are two critical, and well studied, components of the Ras-Raf-MEK-ERK pathway

(McCubrey et al., 2007). We sought to determine the effect of inhibiting MEK1/2 or

ERK1/2 on telomere elongation by SVA.

Specific inhibition of ERK1/2 by 10 μM GDC-0994 blocked telomere elongation

(figure 4.4 a). However, treatment with 10 μM FR-180204, another ERK1/2 specific inhibitor, had minimal effect on telomere elongation compared to DMSO control (figure

4.4 a). Treatment with 10 nM Trametinib, a MEK1/2 specific inhibitor, partially blocked telomere elongation, as did treatment with 5 μM selumetinib, a MEK1/2 and ERK1/2 inhibitor.

These data indicate that the MEK/ERK pathway is a positive regulator of telomere elongation. However, the specific mechanism of action is still unclear. The effect on telomere elongation is not due to changes in telomerase transcription, since telomerase was expressed from a lentivirus, driven by a constitutively active promoter.

Inhibition of ERK1/2 alone, by GDC-0994, caused the most significant blockage of telomere elongation , although FR-180204, another ERK1/2 inhibitor, had minimal effect on elongation (figure 4.4 a). This effect was slightly weaker than that observed from

BRD4, CK1 or ATM inhibition. Inhibition of MEK1/2 alone is sufficient to moderately block elongation (figure 4.4 b). Inhibition of MEK1/2 and ERK1/2 together also moderately blocked telomere elongation (figure 4.4 b). There are two potential explanations for why we observe different effects on telomere elongation, with drugs

88 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION targeting the same pathway: some drugs may have off target effects, which are causing or preventing the telomere elongation phenotype, or some drugs are not effectively inhibiting the target. We have not assessed the degree to which MEK or ERK was inhibited by the drug. Therefore we cannot rule out variable kinase inhibition by each drug. However, in this case all drugs used have been demonstrated as potent and specific inhibitors of their intended target in vitro (Selleckchem). An alternative explanation would be that each drug has a different effect on downstream effectors, which are the direct regulators of telomere elongation. For example, RSK and the NFκB pathways are downstream of MEK/ERK (Carter and Hunninghake, 2000; Roux et al., 2007). We identified elements of both pathways, which may be downstream effectors of the

MEK/ERK pathway that are directly regulating telomere length.

4.2.5 Summary

We have evidence that chemical inhibition of BRD4, CK1 and MEK/ERK block telomere elongation in a telomerase overexpression assay. We identified 4 independent

BRD4 inhibitors that all blocked telomere elongation. In addition we found that 1/3 CK1 inhibitors blocked telomere elongation. Those that were not effective could be due to poor CK1 inhibition. Finally, inhibition of ERK1/2, MEK1/2 or both block telomere elongation, although the effect was weaker than BRD4 or CK1. These results may be due to either variable inhibition by different drugs, or variable inhibition of a yet-to-be identified downstream effector in the pathway.

89 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Further work is required to determine if these drugs can block telomere elongation by endogenous telomerase,. Moreover, identifying the mechanism of these targets may allow us to design more potent and specific inhibitors.

These data suggest that BRD4, CK1 and the MEK/ERK pathways may be therapeutic targets for telomerase inactivation in cancer. Some BRD4 inhibitors are already in clinical trials as cancer drugs, and many ERK/MEK pathway inhibitors are approved for clinical use. Elucidating the mechanism by which these targets block telomere elongation, will allow us to better target cancers or other indications, where inhibiting aberrant telomere elongation has therapeutic benefit.

4.3 Materials and Methods

4.3.1 Small molecule inhibitors

Small molecule inhibitors were dissolved in DMSO, then added to cells at indicated concentrations. Small molecules are listed below:

Drug Vendor Catalogue Number Target KU-55933 R&D Systems 3544 ATM

JQ1 Selleckchem S7110 BRD4

OTX015 Selleckchem S7360 BRD4

I-BET151 Selleckchem S2780 BRD4

MS436 Selleckchem S7305 BRD4

D4476 Selleckchem S7642 CK1

90 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

5-Iodotubercidin Selleckchem S8314 CK1

IC261 Selleckchem S8237 CK1

GDC-0994 Selleckchem S7554 ERK1/2

FR-180204 Selleckchem S7524 ERK1/2

Trametinib Selleckchem S2673 MEK1/2

Selumetinib Selleckchem S8314 MEK1/2, ERK1/2

4.3.2. Virus Production and Titering

SVA is a lentiviral vector containing mTR, driven by its endogenous promoter, and mTERT-IRES-eGFP, driven by the PGK promoter. To generate SVA lentivirus, we first coated 15-cm polystyrene plates (BD Falcon) with 10 mL of 100 μg/ml poly-D lysine for

30 minutes. Next we plated 8 million 293FT cells in DMEM media (Gibco) with 10% fetal bovine serum (Gibco) and penicillin-streptomycin-glutamine (Gibco). After 24 hours, we changed media to 1% FBS in DMEM and co-transfected SVA mTR (mTR lentiviral vector), pCMVΔ8.91 (containing gag and pol lentiviral genes), and VSVG

(containing the env lentiviral gene). Transduction was performed with Lipofectamine

2000 (Life Technologies) in Opti-MEM media (Gibco). Supernatant was collected after

48 hours, briefly centrifuged to remove cellular debris and filtered through a 0.45 μm CN filter (Thermo) to eliminate remaining debris. Viral supernatant was concentrated using

100 kDa ultracentrifugal filters (Millipore) and centrifuging for 10 minutes at 4000g.

Concentrated virus was then collected, aliquoted and frozen at -80C.

91 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

To titer virus, 105 293FT cells were seeded and cultured over night. We assumed a doubling time of 24 hours, therefore a population of 2x105 cells at the time of transduction. Cells were treated with 8 μg/ml of polybrene (Sigma) and infected with 1,

2, 4 and 8 μL of virus. Media was changed 24 hours post transduction and cells were collected at 48 hours post transduction. The percentage of GFP-positive cells was determined by flow cytometry (Becton Dickinson). From these data, we converted percent GFP-positive to initial transduction units, by multiplying percent GFP by 2x105, the initial number of cells at time of transduction. We then plotted transduction units, versus volume added, and generated a line of best fit across the linear range of data points. The slope of this line represented the titer of the virus.

4.3.3 Inhibition of SVA mediated telomere elongation

First, we plated 2x105 SL13 cells onto a 6-well dish in DMEM (Gibco) with 10% FBS

(Sigma) and 1% Penicillin-Streptomycin-Glutamine (Gibco). SL13 cells are a mouse fibroblast line that was derived as described (Lee et al., 2015). After 24 hours, the drug or vehicle was added at the indicated concentration for the duration of the experiment. 24 hours later, SVA virus was added at an MOI of 0.5. We estimated 2 population doublings during the initial 48 hours; therefore 400,000 transduction units of SVA were added to

800,000 cells. Media was changed at 24 hours post transduction, and cell pellets were collected at 2 and 6 days post transduction.

4.3.4 Telomere Southern Blot

92 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

We measured telomere length by telomere restriction fragment Southern blot. Genomic

DNA was extracted using Puregene Core Kit A (Qiagen). 1-3 μg of gDNA was digested with MseI (NEB) and loaded onto a 0.7 % TAE Agarose gel. The gel was run at 37 V for

(Stratagene). The membrane is then pre-incubated for 2 hours at 65C with Church’s

Buffer. Telomere probe and 2-log ladder (Life Technologies) is then end labeled with

0.1% SDS, then twice with 0.5X SSC 0.1 SDS. The membrane is exposed on a phosphorimager (GE Healthcare) and imaged on a STORM scanner (GE Healthcare).

93 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Figure 4.1 Inhibition of BRD4 with small molecules blocks SVA mediated telomere elongation

Telomeres were rapidly elongated in 6 days, after transduction with SVA, a lentivirus that expresses telomerase. Cells were treated chemical inhibitors, to determine the effect of the inhibitor on telomere elongation.

A) Southern blot of telomere length at days 2 and 6 post SVA transduction, in the presence of DMSO, 10 μM KU-55933 (an ATM inhibitor), or 0.1 μM JQ1 (a BRD4 inhibitor

B) Southern blot of telomere length at days 2 and 6 post SVA transduction, in the presence of DMSO or 0.4 μM OTX015 (a BRD4 inhibitor)

C) Southern blot of telomere length at days 2 and 6 post SVA transduction, in the presence of DMSO, 1 μM I-BET151 or 5 μM MS436 (both BRD4 inhibitors).

94 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Figure 4.1 A B DMSO KU-55933 JQ1 DMSO OTX015 Day: 2 6 2 6 2 6 Day: 2 6 2 6

C DMSO IBET151 MS436

Day: 2 6 2 6 2 6

95 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Figure 4.2 Inhibition of BRD4 by JQ1 blocks telomere elongation in a dose dependent manner

Cells were transduced with SVA and either DMSO, 10 μM KU-55933, 100 nM JQ1, 33 nM JQ1, or 11 nM JQ1. Telomere length was measured at days 2 and 6 post-transduction by telomere Southern blot.

96 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Figure 4.2

DMSO KU-55933 JQ1 Day: 2 6 2 6 2 6 2 6 2 6

97 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Figure 4.3 Inhibition of CK1 by D4476 inhibits telomere elongation in a dose dependent manner

Telomeres were rapidly elongated in 6 days, after transduction with SVA, a lentivirus that expresses telomerase. Cells were treated with chemical inhibitors, to determine the effect of the inhibitor on telomere elongation.

A) Southern blot of telomere length at days 2 and 6 post SVA transduction, in the presence of DMSO, 10 μM D4476 (a CK1 inhibitor), 0.5 μM 5-Iodotubercidin (a CK1 inhibitor), or 0.5 μM IC261 (a CK1 inhibitor)

B) Southern blot of telomere length at days 2 and 6 post SVA transduction, in the presence of DMSO or 3.33 μM, 1.11 μM or 0.37 μM D4476

98 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Figure 4.3 5-Iodotub- A DMSO KU-55933 D4476 DMSO ercidin IC261 Day: 2 6 2 6 2 6 2 6 2 6 2 6

B D4476 DMSO Day: 2 6 2 6 2 6 2 6

99 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Figure 4.4 Chemical inhibition of ERK1/2 or MEK1/2 partially blocks telomere elongation

Telomeres were rapidly elongated in 6 days, following transduction with SVA, a lentivirus that expresses telomerase. Cells were treated with chemical inhibitors, to determine the effect of the inhibitor on telomere elongation.

A) Southern blot of telomere length at days 2 and 6 post SVA transduction, in the presence of DMSO, 10 μM GDC-0994 (an ERK 1/2 inhibitor), or 10 FR-180204 (an

ERK1/2 inhibitor).

B) Southern blot of telomere length at days 2 and 6 post SVA transduction, in the presence of DMSO, 10 nM trametinib (a MEK1/2), or 5 μM selumetinib (a MEK1/2 and

ERK1/2 inhibitor)

100 CHAPTER 4. CHEMICAL INHIBITION OF BRD4, CK1 AND MEK/ERK BLOCK TELOMERE ELONGATION

Figure 4.4 GDC- FR- A DMSO 0994 180204

Day: 2 6 2 6 2 6

B DMSO Trametinib DMSO Selumetinib Day: 2 6 2 6 2 6 2 6

101

References

Abreu, E., Aritonovska, E., Reichenbach, P., Cristofari, G., Culp, B., Terns, R.M., Lingner, J., and Terns, M.P. (2010). TIN2-tethered TPP1 recruits human telomerase to telomeres in vivo. Mol Cell Biol 30, 2971-2982.

Alder, J.K., Guo, N., Kembou, F., Parry, E.M., Anderson, C.J., Gorgy, A.I., Walsh, M.F., Sussan, T., Biswal, S., Mitzner, W., et al. (2011). Telomere Length is a Determinant of Emphysema Susceptibility. Am J Respir Crit Care Med.

Allsopp, R.C., Vaziri, H., Patterson, C., Goldstein, S., Younglai, E.V., Futcher, A.B., Greider, C.W., and Harley, C.B. (1992). Telomere length predicts replicative capacity of human fibroblasts. Proc Natl Acad Sci USA 89, 10114-10118.

Armanios, M., and Blackburn, E.H. (2012). The telomere syndromes. Nat Rev Genet 13, 693-704.

Armanios, M., Chen, J.L., Chang, Y.P., Brodsky, R.A., Hawkins, A., Griffin, C.A., Eshleman, J.R., Cohen, A.R., Chakravarti, A., Hamosh, A., et al. (2005). Haploinsufficiency of telomerase reverse transcriptase leads to anticipation in autosomal dominant dyskeratosis congenita. Proc Natl Acad Sci U S A 102, 15960-15964.

Armanios, M.Y., Chen, J.J., Cogan, J.D., Alder, J.K., Ingersoll, R.G., Markin, C., Lawson, W.E., Xie, M., Vulto, I., Phillips, J.A., 3rd, et al. (2007). Telomerase mutations in families with idiopathic pulmonary fibrosis. N Engl J Med 356, 1317-1326.

Artandi, S.E., and DePinho, R.A. (2010). Telomeres and telomerase in cancer. Carcinogenesis 31, 9-18.

Askree, S.H., Yehuda, T., Smolikov, S., Gurevich, R., Hawk, J., Coker, C., Krauskopf, A., Kupiec, M., and McEachern, M.J. (2004). A genome-wide screen for Saccharomyces cerevisiae deletion mutants that affect telomere length. Proc Natl Acad Sci U S A 101, 8658-8663.

Baerlocher, G.M., Vulto, I., de Jong, G., and Lansdorp, P.M. (2006). Flow cytometry and FISH to measure the average length of telomeres (flow FISH). Nat Protoc 1, 2365-2376.

102

Bar, C., Povedano, J.M., Serrano, R., Benitez-Buelga, C., Popkes, M., Formentini, I., Bobadilla, M., Bosch, F., and Blasco, M.A. (2016). Telomerase gene therapy rescues telomere length, bone marrow aplasia, and survival in mice with aplastic anemia. Blood 127, 1770-1779.

Bechter, O.E., Zou, Y., Walker, W., Wright, W.E., and Shay, J.W. (2004). Telomeric recombination in mismatch repair deficient human colon cancer cells after telomerase inhibition. Cancer Res 64, 3444-3451.

Blackburn, E.H., and Szostak, J.W. (1984). The molecular structure of centromeres and telomeres. Annu Rev Biochem 53, 163-194.

Blasco, M.A., Lee, H.-W., Hande, P.M., Samper, E., Lansdorp, P.M., DePinho, R.A., and Greider, C.W. (1997). Telomere shortening and tumor formation by mouse cells lacking telomerase RNA. Cell 91, 25-34.

Boi, M., Gaudio, E., Bonetti, P., Kwee, I., Bernasconi, E., Tarantelli, C., Rinaldi, A., Testoni, M., Cascione, L., Ponzoni, M., et al. (2015). The BET Bromodomain Inhibitor OTX015 Affects Pathogenetic Pathways in Preclinical B-cell Tumor Models and Synergizes with Targeted Drugs. Clin Cancer Res 21, 1628-1638.

Britt-Compton, B., Capper, R., Rowson, J., and Baird, D.M. (2009). Short telomeres are preferentially elongated by telomerase in human cells. FEBS Lett 583, 3076-3080.

Bryan, T.M., Englezou, A., Gupta, J., Bacchetti, S., and Reddel, R.R. (1995). Telomere elongation in immortal human cells without detectable telomerase activity. EMBO J 14, 4240-4248.

Bryja, V., Schulte, G., Rawal, N., Grahn, A., and Arenas, E. (2007). Wnt-5a induces Dishevelled phosphorylation and dopaminergic differentiation via a CK1-dependent mechanism. J Cell Sci 120, 586-595.

Burrows, A.E., Smogorzewska, A., and Elledge, S.J. (2010). Polybromo-associated BRG1-associated factor components BRD7 and BAF180 are critical regulators of p53 required for induction of replicative senescence. Proc Natl Acad Sci U S A 107, 14280- 14285.

Carter, A.B., and Hunninghake, G.W. (2000). A constitutive active MEK --> ERK pathway negatively regulates NF-kappa B-dependent gene expression by modulating TATA-binding protein phosphorylation. J Biol Chem 275, 27858-27864.

103

Cawthon, R.M. (2002). Telomere measurement by quantitative PCR. Nucleic acids research 30, e47.

Cerone, M.A., Burgess, D.J., Naceur-Lombardelli, C., Lord, C.J., and Ashworth, A. (2011). High-throughput RNAi screening reveals novel regulators of telomerase. Cancer Res 71, 3328-3340.

Chang, J.T., Lu, Y.C., Chen, Y.J., Tseng, C.P., Chen, Y.L., Fang, C.W., and Cheng, A.J. (2006). hTERT phosphorylation by PKC is essential for telomerase holoprotein integrity and enzyme activity in head neck cancer cells. Br J Cancer 94, 870-878.

Chang, W., Dynek, J.N., and Smith, S. (2003). TRF1 is degraded by ubiquitin-mediated proteolysis after release from telomeres. Genes Dev 17, 1328-1333.

Chin, L., Tam, A., Pomerantz, J., Wong, M., Holash, J., Bardeesy, N., Shen, Q., O'Hagan, R., Pantginis, J., Zhou, H., et al. (1999). Essential role for oncogenic Ras in tumour maintenance. Nature 400, 468-472.

Choi, Y.H., Lim, J.K., Jeong, M.W., and Kim, K.T. (2012). HnRNP A1 phosphorylated by VRK1 stimulates telomerase and its binding to telomeric DNA sequence. Nucleic acids research 40, 8499-8518.

Conomos, D., Pickett, H.A., and Reddel, R.R. (2013). Alternative lengthening of telomeres: remodeling the telomere architecture. Front Oncol 3, 27.

Cottage, C.T., Neidig, L., Sundararaman, B., Din, S., Joyo, A.Y., Bailey, B., Gude, N., Hariharan, N., and Sussman, M.A. (2012). Increased mitotic rate coincident with transient telomere lengthening resulting from pim-1 overexpression in cardiac progenitor cells. Stem Cells 30, 2512-2522.

Counter, C.M., Avilion, A.A., LeFeuvre, C.E., Stewart, N.G., Greider, C.W., Harley, C.B., and Bacchetti, S. (1992). Telomere shortening associated with chromosome instability is arrested in immortal cells which express telomerase activity. Embo J 11, 1921-1929.

Cristofari, G., and Lingner, J. (2006). Telomere length homeostasis requires that telomerase levels are limiting. Embo J 25, 565-574.

104

d'Adda di Fagagna, F., Reaper, P.M., Clay-Farrace, L., Fiegler, H., Carr, P., Von Zglinicki, T., Saretzki, G., Carter, N.P., and Jackson, S.P. (2003). A DNA damage checkpoint response in telomere-initiated senescence. Nature 426, 194-198. de Lange, T. (1994). Activation of telomerase in a human tumor. Proc Natl Acad Sci 91, 2882-2885.

Denchi, E.L., and de Lange, T. (2007). Protection of telomeres through independent control of ATM and ATR by TRF2 and POT1. Nature 448, 1068-1071.

Devaiah, B.N., Case-Borden, C., Gegonne, A., Hsu, C.H., Chen, Q., Meerzaman, D., Dey, A., Ozato, K., and Singer, D.S. (2016a). BRD4 is a histone acetyltransferase that evicts nucleosomes from chromatin. Nat Struct Mol Biol 23, 540-548.

Devaiah, B.N., Gegonne, A., and Singer, D.S. (2016b). Bromodomain 4: a cellular Swiss army knife. J Leukoc Biol 100, 679-686.

Devaiah, B.N., Lewis, B.A., Cherman, N., Hewitt, M.C., Albrecht, B.K., Robey, P.G., Ozato, K., Sims, R.J., 3rd, and Singer, D.S. (2012). BRD4 is an atypical kinase that phosphorylates serine2 of the RNA polymerase II carboxy-terminal domain. Proc Natl Acad Sci U S A 109, 6927-6932.

Dey, A., Nishiyama, A., Karpova, T., McNally, J., and Ozato, K. (2009). Brd4 marks select genes on mitotic chromatin and directs postmitotic transcription. Mol Biol Cell 20, 4899-4909.

Dumaz, N., Milne, D.M., and Meek, D.W. (1999). Protein kinase CK1 is a p53-threonine 18 kinase which requires prior phosphorylation of serine 15. FEBS Lett 463, 312-316.

Filippakopoulos, P., Qi, J., Picaud, S., Shen, Y., Smith, W.B., Fedorov, O., Morse, E.M., Keates, T., Hickman, T.T., Felletar, I., et al. (2010). Selective inhibition of BET bromodomains. Nature 468, 1067-1073.

Frank, C.J., Hyde, M., and Greider, C.W. (2006). Regulation of Telomere Elongation by the Cyclin-Dependent Kinase CDK1. Mol Cell 24, 423-432.

French, C.A., Miyoshi, I., Kubonishi, I., Grier, H.E., Perez-Atayde, A.R., and Fletcher, J.A. (2003). BRD4-NUT fusion oncogene: a novel mechanism in aggressive carcinoma. Cancer Res 63, 304-307.

105

Garvik, B., Carson, M., and Hartwell, L. (1995). Single-stranded DNA arising at telomeres in cdc13 mutants may constitute a specific signal for the RAD9 checkpoint. Mol Cell Biol 15, 6128-6138.

Ge, Z., Liu, C., Bjorkholm, M., Gruber, A., and Xu, D. (2006). Mitogen-activated protein kinase cascade-mediated histone H3 phosphorylation is critical for telomerase reverse transcriptase expression/telomerase activation induced by proliferation. Mol Cell Biol 26, 230-237.

Greenwell, P.W., Kronmal, S.L., Porter, S.E., Gassenhuber, J., Obermaier, B., and Petes, T.D. (1995). TEL1, a gene involved in controlling telomere length in S. cerevisiae, is homologous to the human ataxia telangiectasia gene. Cell 82, 823-829.

Greider, C.W. (1992). Telomere chromatin and gene expression. Curr Biol 2, 62-64.

Greider, C.W. (1996). Telomere length regulation. Annu Rev Biochem 65, 337-365.

Greider, C.W. (1999). Telomerase activation: one step on the road to cancer? Trends Genet 15, 109-112.

Greider, C.W., and Blackburn, E.H. (1985). Identification of a specific telomere terminal transferase activity in Tetrahymena extracts. Cell 43, 405-413.

Greider, C.W., and Blackburn, E.H. (1989). A telomeric sequence in the RNA of Tetrahymena telomerase required for telomere repeat synthesis. Nature 337, 331-337.

Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next generation. Cell 144, 646-674.

Hang, L.E., Liu, X., Cheung, I., Yang, Y., and Zhao, X. (2011). SUMOylation regulates telomere length homeostasis by targeting Cdc13. Nat Struct Mol Biol 18, 920-926.

Hao, L.Y., Armanios, M., Strong, M.A., Karim, B., Feldser, D.M., Huso, D., and Greider, C.W. (2005). Short telomeres, even in the presence of telomerase, limit tissue renewal capacity. Cell 123, 1121-1131.

Hardy, C.F., Sussel, L., and Shore, D. (1992). A RAP1-interacting protein involved in transcriptional silencing and telomere length regulation. Genes Dev 6, 801-814.

106

Harley, C.B., Futcher, A.B., and Greider, C.W. (1990). Telomeres shorten during ageing of human fibroblasts. Nature 345, 458-460.

Hastie, N.D., Dempster, M., Dunlop, M.G., Thompson, A.M., Green, D.K., and Allshire, R.C. (1990). Telomere reduction in human colorectal carcinoma and with ageing. Nature 346, 866-868.

Hathcock, K.S., Hemann, M.T., Opperman, K.K., Strong, M.A., Greider, C.W., and Hodes, R.J. (2002). Haploinsufficiency of mTR results in defects in telomere elongation. Proc Natl Acad Sci U S A 99, 3591-3596.

Heidenreich, B., Rachakonda, P.S., Hemminki, K., and Kumar, R. (2014). TERT promoter mutations in cancer development. Curr Opin Genet Dev 24, 30-37.

Heiss, N.S., Knight, S.W., Vulliamy, T.J., Klauck, S.M., Wiemann, S., Mason, P.J., Poustka, A., and Dokal, I. (1998). X-linked dyskeratosis congenita is caused by mutations in a highly conserved gene with putative nucleolar functions. Nat Genet 19, 32-38.

Hemann, M.T., Strong, M., Hao, L.-Y., and Greider, C.W. (2001). The shortest telomere, not average telomere length, is critical for cell viability and chromosome stability. Cell 107, 67-77.

Hiyama, K., Hirai, Y., Kyoizumi, S., Akiyama, M., Hiyama, E., Piatyszek, M.A., Shay, Jw, Ishioka, S., and Yamakido, M. (1995). Activation of telomerase in human lymphocytes and hematopoietic progenitor cells. J Immunol 155, 3711-3715.

Horn, S., Figl, A., Rachakonda, P.S., Fischer, C., Sucker, A., Gast, A., Kadel, S., Moll, I., Nagore, E., Hemminki, K., et al. (2013). TERT promoter mutations in familial and sporadic melanoma. Science 339, 959-961.

Huang, F.W., Hodis, E., Xu, M.J., Kryukov, G.V., Chin, L., and Garraway, L.A. (2013). Highly recurrent TERT promoter mutations in human melanoma. Science 339, 957-959.

IJpma, A., and Greider, C.W. (2003). Short telomeres induce a DNA damage response in Saccharomyces cerevisiae. Molec Biol Cell 14, 987-1001.

Jaskelioff, M., Muller, F.L., Paik, J.H., Thomas, E., Jiang, S., Adams, A.C., Sahin, E., Kost-Alimova, M., Protopopov, A., Cadinanos, J., et al. (2011). Telomerase reactivation reverses tissue degeneration in aged telomerase-deficient mice. Nature 469, 102-106.

107

Ju, Z., Jiang, H., Jaworski, M., Rathinam, C., Gompf, A., Klein, C., Trumpp, A., and Rudolph, K.L. (2007). Telomere dysfunction induces environmental alterations limiting hematopoietic stem cell function and engraftment. Nat Med 13, 742-747.

Kang, S.S., Kwon, T., Kwon, D.Y., and Do, S.I. (1999). Akt protein kinase enhances human telomerase activity through phosphorylation of telomerase reverse transcriptase subunit. J Biol Chem 274, 13085-13090.

Kendellen, M.F., Barrientos, K.S., and Counter, C.M. (2009). POT1 association with TRF2 regulates telomere length. Mol Cell Biol 29, 5611-5619.

Killela, P.J., Reitman, Z.J., Jiao, Y., Bettegowda, C., Agrawal, N., Diaz, L.A., Jr., Friedman, A.H., Friedman, H., Gallia, G.L., Giovanella, B.C., et al. (2013). TERT promoter mutations occur frequently in gliomas and a subset of tumors derived from cells with low rates of self-renewal. Proceedings of the National Academy of Sciences of the United States of America 110, 6021-6026.

Kim, N.W., Piatyszek, M.A., Prowse, K.R., Harley, C.B., West, M.D., Ho, P.L.C., Coviello, G.M., Wright, W.E., Weinrich, S.L., and Shay, J.W. (1994). Specific association of human telomerase activity with immortal cells and cancer. Science 266, 2011-2014.

Kimura, A., Ohmichi, M., Kawagoe, J., Kyo, S., Mabuchi, S., Takahashi, T., Ohshima, C., Arimoto-Ishida, E., Nishio, Y., Inoue, M., et al. (2004). Induction of hTERT expression and phosphorylation by estrogen via Akt cascade in human ovarian cancer cell lines. Oncogene 23, 4505-4515.

Knippschild, U., Gocht, A., Wolff, S., Huber, N., Lohler, J., and Stoter, M. (2005). The casein kinase 1 family: participation in multiple cellular processes in eukaryotes. Cell Signal 17, 675-689.

Knippschild, U., Kruger, M., Richter, J., Xu, P., Garcia-Reyes, B., Peifer, C., Halekotte, J., Bakulev, V., and Bischof, J. (2014). The CK1 Family: Contribution to Cellular Stress Response and Its Role in Carcinogenesis. Front Oncol 4, 96.

Kojima, H., Yokosuka, O., Imazeki, F., Saisho, H., and Omata, M. (1997). Telomerase activity and telomere length in hepatocellular carcinoma and chronic liver disease. Gastroenterology 112, 493-500.

108

Kupiec, M., and Weisman, R. (2012). TOR links starvation responses to telomere length maintenance. Cell Cycle 11, 2268-2271.

Langmead, B., and Salzberg, S.L. (2012). Fast gapped-read alignment with Bowtie 2. Nat Methods 9, 357-359.

Lee, S.S., Bohrson, C., Pike, A.M., Wheelan, S.J., and Greider, C.W. (2015). ATM Kinase Is Required for Telomere Elongation in Mouse and Human Cells. Cell reports 13, 1623-1632.

Li, W., Xu, H., Xiao, T., Cong, L., Love, M.I., Zhang, F., Irizarry, R.A., Liu, J.S., Brown, M., and Liu, X.S. (2014). MAGeCK enables robust identification of essential genes from genome-scale CRISPR/Cas9 knockout screens. Genome Biol 15, 554.

Lin, K.W., McDonald, K.R., Guise, A.J., Chan, A., Cristea, I.M., and Zakian, V.A. (2015). Proteomics of yeast telomerase identified Cdc48-Npl4-Ufd1 and Ufd4 as regulators of Est1 and telomere length. Nat Commun 6, 8290.

Lingner, J., Hughes, T.R., Shevchenko, A., Mann, M., Lundblad, V., and Cech, T.R. (1997). Reverse transcriptase motifs in the catalytic subunit of telomerase. Science 276, 561-567.

Liu, Y., Snow, B.E., Hande, M.P., Yeung, D., Erdmann, N.J., Wakeham, A., Itie, A., Siderovski, D.P., Lansdorp, P.M., Robinson, M.O., et al. (2000). The telomerase reverse transcriptase is limiting and necessary for telomerase function in vivo. Curr Biol 10, 1459-1462.

Loayza, D., and De Lange, T. (2003). POT1 as a terminal transducer of TRF1 telomere length control. Nature 423, 1013-1018.

Ludlow, A.T., Lima, L.C., Wang, J., Hanson, E.D., Guth, L.M., Spangenburg, E.E., and Roth, S.M. (2012). Exercise alters mRNA expression of telomere-repeat binding factor 1 in skeletal muscle via p38 MAPK. J Appl Physiol.

Lundblad, V., and Szostak, J.W. (1989). A mutant with a defect in telomere elongation leads to senescence in yeast. Cell 57, 633-643.

Lydall, D., and Weinert, T. (1995). Yeast checkpoint genes in DNA damage processing: implications for repair and arrest. Science 270, 1488-1491.

109

Mallory, J.C., and Petes, T.D. (2000). Protein kinase activity of Tel1p and Mec1p, two Saccharomyces cerevisiae proteins related to the human ATM protein kinase. Proc Natl Acad Sci U S A 97, 13749-13754.

Marrone, A., Stevens, D., Vulliamy, T., Dokal, I., and Mason, P.J. (2004). Heterozygous telomerase RNA mutations found in dyskeratosis congenita and aplastic anemia reduce telomerase activity via haploinsufficiency. Blood 104, 3936-3942.

Massillon, D., Stalmans, W., van de Werve, G., and Bollen, M. (1994). Identification of the glycogenic compound 5-iodotubercidin as a general protein kinase inhibitor. Biochem J 299 ( Pt 1), 123-128.

McCubrey, J.A., Steelman, L.S., Chappell, W.H., Abrams, S.L., Wong, E.W., Chang, F., Lehmann, B., Terrian, D.M., Milella, M., Tafuri, A., et al. (2007). Roles of the Raf/MEK/ERK pathway in cell growth, malignant transformation and drug resistance. Biochim Biophys Acta 1773, 1263-1284.

Melek, M., Greene, E.C., and Shippen, D.E. (1996). Processing of nontelomeric 3' ends by telomerase: default template alignment and endonucleolytic cleavage. Mol Cell Biol 16, 3437-3445.

Mertz, J.A., Conery, A.R., Bryant, B.M., Sandy, P., Balasubramanian, S., Mele, D.A., Bergeron, L., and Sims, R.J., 3rd (2011). Targeting MYC dependence in cancer by inhibiting BET bromodomains. Proc Natl Acad Sci U S A 108, 16669-16674.

Mitchell, J.R., Wood, E., and Collins, K. (1999). A telomerase component is defective in the human disease dyskeratosis congenita. Nature 402, 551-555.

Mochizuki, Y., He, J., Kulkarni, S., Bessler, M., and Mason, P.J. (2004). Mouse dyskerin mutations affect accumulation of telomerase RNA and small nucleolar RNA, telomerase activity, and ribosomal RNA processing. Proceedings of the National Academy of Sciences of the United States of America 101, 10756-10761.

Moyzis, R.K., Buckingham, J.M., Cram, L.S., Dani, M., Deaven, L.L., Jones, M.D., Meyne, J., Ratliff, R.L., and Wu, J.-R. (1988). A highly conserved repetitive DNA sequence, (TTAGGG)n, present at the telomeres of human chromosomes. Proc Natl Acad Sci USA 85, 6622-6626.

110

Osawa, M., Hanada, K., Hamada, H., and Nakauchi, H. (1996). Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell. Science 273, 242-245.

Osborne, J.K., Zaganjor, E., and Cobb, M.H. (2012). Signal control through Raf: in sickness and in health. Cell Res 22, 14-22.

Phillis, J.W., and Smith-Barbour, M. (1993). The adenosine kinase inhibitor, 5- iodotubercidin, is not protective against cerebral ischemic injury in the gerbil. Life Sci 53, 497-502.

Picco, V., Coste, I., Giraud-Panis, M.J., Renno, T., Gilson, E., and Pages, G. (2016). ERK1/2/MAPK pathway-dependent regulation of the telomeric factor TRF2. Oncotarget.

Poon, S.S., and Lansdorp, P.M. (2001). Quantitative fluorescence in situ hybridization (Q-FISH). Curr Protoc Cell Biol Chapter 18, Unit 18 14.

Rachidi, N., Taly, J.F., Durieu, E., Leclercq, O., Aulner, N., Prina, E., Pescher, P., Notredame, C., Meijer, L., and Spath, G.F. (2014). Pharmacological assessment defines Leishmania donovani casein kinase 1 as a drug target and reveals important functions in parasite viability and intracellular infection. Antimicrob Agents Chemother 58, 1501- 1515.

Rahman, S., Sowa, M.E., Ottinger, M., Smith, J.A., Shi, Y., Harper, J.W., and Howley, P.M. (2011). The Brd4 extraterminal domain confers transcription activation independent of pTEFb by recruiting multiple proteins, including NSD3. Mol Cell Biol 31, 2641-2652.

Ramsay, A.J., Quesada, V., Foronda, M., Conde, L., Martinez-Trillos, A., Villamor, N., Rodriguez, D., Kwarciak, A., Garabaya, C., Gallardo, M., et al. (2013). POT1 mutations cause telomere dysfunction in chronic lymphocytic leukemia. Nature genetics 45, 526- 530.

Rena, G., Bain, J., Elliott, M., and Cohen, P. (2004). D4476, a cell-permeant inhibitor of CK1, suppresses the site-specific phosphorylation and nuclear exclusion of FOXO1a. EMBO reports 5, 60-65.

Rossi, D.J., Bryder, D., Seita, J., Nussenzweig, A., Hoeijmakers, J., and Weissman, I.L. (2007). Deficiencies in DNA damage repair limit the function of haematopoietic stem cells with age. Nature 447, 725-729.

111

Roth, A., Harley, C.B., and Baerlocher, G.M. (2010). Imetelstat (GRN163L)--telomerase- based cancer therapy. Recent Results Cancer Res 184, 221-234.

Roux, P.P., Shahbazian, D., Vu, H., Holz, M.K., Cohen, M.S., Taunton, J., Sonenberg, N., and Blenis, J. (2007). RAS/ERK signaling promotes site-specific ribosomal protein S6 phosphorylation via RSK and stimulates cap-dependent translation. J Biol Chem 282, 14056-14064.

Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, second edn (Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press).

Shalem, O., Sanjana, N.E., Hartenian, E., Shi, X., Scott, D.A., Mikkelsen, T.S., Heckl, D., Ebert, B.L., Root, D.E., Doench, J.G., et al. (2014). Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 343, 84-87.

Shi, J., Yang, X.R., Ballew, B., Rotunno, M., Calista, D., Fargnoli, M.C., Ghiorzo, P., Bressac-de Paillerets, B., Nagore, E., Avril, M.F., et al. (2014). Rare missense variants in POT1 predispose to familial cutaneous malignant melanoma. Nature genetics 46, 482- 486.

Sitaram, R.T., Cairney, C.J., Grabowski, P., Keith, W.N., Hallberg, B., Ljungberg, B., and Roos, G. (2009). The PTEN regulator DJ-1 is associated with hTERT expression in clear cell renal cell carcinoma. Int J Cancer 125, 783-790.

Smogorzewska, A., and de Lange, T. (2004). Regulation of telomerase by telomeric proteins. Annu Rev Biochem 73, 177-208.

Song, Z., Wang, J., Guachalla, L.M., Terszowski, G., Rodewald, H.R., Ju, Z., and Rudolph, K.L. (2010). Alterations of the systemic environment are the primary cause of impaired B and T lymphopoiesis in telomere-dysfunctional mice. Blood 115, 1481-1489.

Song, Z., Zhang, J., Ju, Z., and Rudolph, K.L. (2012). Telomere dysfunctional environment induces loss of quiescence and inherent impairments of hematopoietic stem cell function. Aging Cell.

Stanley, S.E., and Armanios, M. (2015). The short and long telomere syndromes: paired paradigms for molecular medicine. Curr Opin Genet Dev 33, 1-9.

112

Strezoska, Z., Licon, A., Haimes, J., Spayd, K.J., Patel, K.M., Sullivan, K., Jastrzebski, K., Simpson, K.J., Leake, D., van Brabant Smith, A., et al. (2012). Optimized PCR conditions and increased shRNA fold representation improve reproducibility of pooled shRNA screens. PLoS One 7, e42341.

Teixeira, M.T., Arneric, M., Sperisen, P., and Lingner, J. (2004). Telomere length homeostasis is achieved via a switch between telomerase- extendible and -nonextendible states. Cell 117, 323-335.

Ungar, L., Yosef, N., Sela, Y., Sharan, R., Ruppin, E., and Kupiec, M. (2009). A genome-wide screen for essential yeast genes that affect telomere length maintenance. Nucleic Acids Res 37, 3840-3849.

Vulliamy, T., Marrone, A., Goldman, F., Dearlove, A., Bessler, M., Mason, P.J., and Dokal, I. (2001). The RNA component of telomerase is mutated in autosomal dominant dyskeratosis congenita. Nature 413, 432-435.

Wang, F., Podell, E.R., Zaug, A.J., Yang, Y., Baciu, P., Cech, T.R., and Lei, M. (2007). The POT1-TPP1 telomere complex is a telomerase processivity factor. Nature 445, 506- 510.

Wang, T., Lander, E.S., and Sabatini, D.M. (2016). Large-Scale Single Guide RNA Library Construction and Use for CRISPR-Cas9-Based Genetic Screens. Cold Spring Harb Protoc 2016, pdb top086892.

Wang, T., Wei, J.J., Sabatini, D.M., and Lander, E.S. (2014). Genetic screens in human cells using the CRISPR-Cas9 system. Science 343, 80-84.

Wellinger, R.J., Wolf, A.J., and Zakian, V.A. (1993). Saccharomyces telomeres acquire single-strand TG1-3 tails late in S phase. Cell 72, 51-60.

Westin, E.R., Chavez, E., Lee, K.M., Gourronc, F.A., Riley, S., Lansdorp, P.M., Goldman, F.D., and Klingelhutz, A.J. (2007). Telomere restoration and extension of proliferative lifespan in dyskeratosis congenita fibroblasts. Aging Cell 6, 383-394.

Wong, J.M., and Collins, K. (2006). Telomerase RNA level limits telomere maintenance in X-linked dyskeratosis congenita. Genes & development 20, 2848-2858.

113

Wu, K.J., Grandori, C., Amacker, M., Simon-Vermot, N., Polack, A., Lingner, J., and Dalla-Favera, R. (1999). Direct activation of TERT transcription by c-MYC. Nat Genet 21, 220-224.

Xu, Z., Duc, K.D., Holcman, D., and Teixeira, M.T. (2013). The length of the shortest telomere as the major determinant of the onset of replicative senescence. Genetics 194, 847-857.

Yamaguchi, H., Calado, R.T., Ly, H., Kajigaya, S., Baerlocher, G.M., Chanock, S.J., Lansdorp, P.M., and Young, N.S. (2005). Mutations in TERT, the gene for telomerase reverse transcriptase, in aplastic anemia. N Engl J Med 352, 1413-1424.

Yang, Z., Yik, J.H., Chen, R., He, N., Jang, M.K., Ozato, K., and Zhou, Q. (2005). Recruitment of P-TEFb for stimulation of transcriptional elongation by the bromodomain protein Brd4. Mol Cell 19, 535-545.

You, J., Li, Q., Wu, C., Kim, J., Ottinger, M., and Howley, P.M. (2009). Regulation of aurora B expression by the bromodomain protein Brd4. Mol Cell Biol 29, 5094-5103.

Zhang, J., Yang, P.L., and Gray, N.S. (2009). Targeting cancer with small molecule kinase inhibitors. Nat Rev Cancer 9, 28-39.

Zhao, R., Nakamura, T., Fu, Y., Lazar, Z., and Spector, D.L. (2011). Gene bookmarking accelerates the kinetics of post-mitotic transcriptional re-activation. Nat Cell Biol 13, 1295-1304.

Zhu, X., Kumar, R., Mandal, M., Sharma, N., Sharma, H.W., Dhingra, U., Sokoloski, J.A., Hsiao, R., and Narayanan, R. (1996). Cell cycle-dependent modulation of telomerase activity in tumor cells. Proc Natl Acad Sci U S A 93, 6091-6095.

Zuber, J., Shi, J., Wang, E., Rappaport, A.R., Herrmann, H., Sison, E.A., Magoon, D., Qi, J., Blatt, K., Wunderlich, M., et al. (2011). RNAi screen identifies Brd4 as a therapeutic target in acute myeloid leukaemia. Nature 478, 524-528.

114

CURRICULUM VITAE The Johns Hopkins University School of Medicine

Name: Steven Wang Re vised: January 3, 2017

Personal Information 725 N. Wolfe St., PCTB Rm. 605, Baltimore, MD 21205 Work: (410)-614-1469 Mobile: (215)-284-7617 Email: [email protected] Born: Qingdao, China. July 12, 1988

Educational History Ph.D. expected 2017 Molecular Biology Johns Hopkins University School of Medicine Mentor: Carol W. Greider Ph.D.

B.A. 2010 Biochemistry University of Pennsylvania

Professional Experience 2010-present Graduate Laboratory of Carol W. Greider, Johns Hopkins Researcher University School of Medicine

2015 Entrepreneur- BioHealth Innovation in-Residence Intern

2008-2010 Undergraduate Laboratory of John D. Gearhart, University of Researcher Pennsylvania School of Medicine

2007-2008 Undergraduate Laboratory of Timothy R. Brazelton, Children’s Researcher Hospital of Philadelphia

Honors and Awards 2015 Travel Award Graduate Student Assembly, Johns Hopkins University School of Medicine

2006-2009 Benjamin Franklin University of Pennsylvania Scholar

2009 Summer Undergraduate Department of Biology, Research Fellowship University of Pennsylvania

2008-2009 Vagelos Undergraduate University of Pennsylvania Research Award

2008 Dean’s Endowed University of Pennsylvania Research Award

2008 Benjamin Franklin University of Pennsylvania Scholar Research Grant

115

Publications Wang, S., Pike, A. M., Lee, S. L., Connelly, C. J., and Greider, C. W. BRD4 is a positive regulator of telomere length. In preparation.

Abstracts 2015 A pooled shRNA screen to Cold Spring Harbor Meeting on identify novel regulators of Telomeres and Telomerase telomere length. (Poster) Cold Spring Harbor Meeting on 2013 Restoration of mTR in G4 Telomeres and Telomerase mTR-/- bone marrow rescues critically short telomeres and enhances stem cell function in bone marrow transplant models. (Poster)

Patents 2016 Chemical inhibitors against kinases to block telomere elongation in cancer (provisional application, USTPO No. 62435191

116