Slx4 Interacting Protein (Slx4ip)

SLX4 INTERACTING PROTEIN (SLX4IP): A VITAL PRIMER FOR ALTERNATIVE LENGTHENING OF TELOMERE (ALT)-LIKE PROCESSES PROMOTING REPLICATIVE IMMORTALITY IN CASTRATION-RESISTANT PROSTATE CANCER WITH ANDROGEN RECEPTOR LOSS

TAWNA L. MANGOSH

Submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

Dissertation Advisor: Dr. Derek J. Taylor, Ph.D.

Department of Pharmacology CASE WESTERN RESERVE UNIVERSITY August 2021

CASE WESTERN RESERVE UNIVERSITY SCHOOL OF GRADUATE STUDIES

We hereby approve the thesis/dissertation of

Tawna L. Mangosh

Candidate for the degree of Doctor of Philosophy*.

Marvin Nieman, Ph.D. (Committee Chair) Magdalena Grabowska, Ph.D. (Committee Member) Jason Mears, Ph.D. (Committee Member) William Schiemann, Ph.D. (Committee Member) Derek Taylor, Ph.D. (Committee Member)

Date of Defense May 25th, 2021

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

DEDICATION

For my grandparents, who always taught me to just be(e) happy.

TABLE OF CONTENTS

TABLE OF FIGURES ...... 7 ACKNOWLEDGEMENTS ...... 9 Abstract ...... 12 CHAPTER 1: INTRODUCTION AND BACKGROUND ...... 14 1.1 Telomeres and Replicative Capacity ...... 15 1.1.1 The Dynamic Nucleoprotein Structure of Telomeres ...... 16 1.1.2 The Terminal Fates of Telomeric Attrition ...... 17 1.2 Orthodox Telomere Synthesis via Telomerase ...... 21 1.2.1 Telomerase Structure and Function ...... 22 1.2.2 Structural and Functional Regulators of Telomerase ...... 22 1.2.3 Transcriptional Regulators of Telomerase ...... 23 1.3 The Alternative Lengthening of Telomeres (ALT) Pathway ...... 24 1.3.1 ALTernative Mechanism of Telomere Elongation ...... 25 1.3.2 Identification of ALT Activity Through Surrogate Measures...... 27 1.3.3 ALT-associated Promyelocytic Leukemia Bodies ...... 31 1.4 Telomere Maintenance Mechanism Plasticity ...... 34 1.4.1 Engineered Telomere Maintenance Mechanism Coexistence ...... 34 1.4.2 ALT as a Mechanism of Therapeutic Resistance ...... 36 1.4.3 ALTiness versus Bona Fide ALT ...... 38 1.4.4 The Elusive Role of SLX4IP in Telomere Maintenance ...... 39 1.5 The Complexity of Prostate Cancer Disease Progression ...... 40 1.5.1 Acquisition of Castration Resistance in Advanced Disease ...... 41 1.5.2 Molecular Mechanisms Driving AR-negative Disease ...... 44 1.5.3 Telomere Maintenance in Castration-resistant Prostate Cancer ...... 46 1.6 Statement of Purpose ...... 49 CHAPTER 2: SLX4IP PROMOTES TELOMERE MAINTENANCE IN ANDROGEN RECEPTOR–INDEPENDENT CASTRATION-RESISTANT PROSTATE CANCER THROUGH ALT-LIKE TELOMERIC PML LOCALIZATION ...... 52

2.1 Abstract ...... 53 2.2 Introduction ...... 54 2.3 Results ...... 58 2.3.1 AR-independent CRPC in vitro models exhibit an atypical ALT-like phenotype ...... 58 2.3.2 SLX4IP expression correlates with AR-independence and ALT-like hallmarks ...... 60 2.3.3 SLX4IP is essential for ALT-like hallmarks and telomere maintenance in AR-independent CRPC ...... 63 2.3.4 SLX4IP-mediated ALT-like hallmark depletion and induction of senescence is maintained in vivo ...... 66 2.3.5 Androgen-deprived conditions promote ALT-like PML localization in an SLX4IP-dependent manner ...... 67 2.4 Discussion ...... 70 2.5 Methods ...... 74 2.6 Acknowledgements ...... 81 CHAPTER 3: SLX4IP N-TERMINUS DICTATES TELOMERIC LOCALIZATION IN ALT-LIKE CASTRATION-RESISTANT PROSTATE CANCER IN VITRO ...... 101 3.1 Abstract ...... 102 3.2 Introduction ...... 103 3.3 Results ...... 107 3.3.1 The N-terminus of SLX4IP is responsible for the promotion of APB foci in CRPC in vitro models ...... 107 3.3.2 Manipulation of SLX4IP expression alters abundance of nuclear PML foci but not PML expression ...... 109 3.3.3 The N-terminus of SLX4IP is responsible for telomeric localization.. 110 3.3.4 SLX4IP N-terminus rescues the late-onset senescent phenotype induced by SLX4IP knockdown ...... 112 3.4 Discussion ...... 114

3.5 Materials and Methods ...... 117 3.6 Acknowledgements ...... 121 CHAPTER 4: DISCUSSION AND FUTURE DIRECTIONS ...... 140 4.1 ALTiness: The FinALT Telomeric Frontier ...... 141 4.2 SLX4IP-mediated ALTiness or true ALT activity ...... 142 4.2.1 Canonical ALT Mediators Do Not Translate to ALTiness ...... 144 4.2.2 ALTiness via DNA Damage Machinery Hijacking ...... 146 4.3 Regulation of SLX4IP: From Transcriptional to Post-translational ...... 147 4.3.1 Androgen Deprivation Therapy and SLX4IP ...... 148 4.3.2 Transcriptional Programs Shared by PML and SLX4IP ...... 149 4.3.3 Telomeric Recruitment via SLX4IP SUMOylation ...... 151 4.4 TMM Inhibition as a Last Resort ...... 154 4.4.1 Dual TMM Inhibition Circumvents Resistance ...... 155 4.4.2 Androgen Deprivation Therapy, ALT Inhibition, and Senolytics ...... 156 4.5 Concluding Remarks ...... 157 REFERENCES ...... 159

LIST OF FIGURES

Figure 1. 1. Telomere maintenance mechanisms instill replicative immortality in cancer...... 20 Figure 1. 2. Evaluation of ALT hallmarks identify cells primarily using ALT for telomere maintenance...... 33 Figure 1. 3. Telomere maintenance mechanism plasticity in prostate cancer disease progression...... 48

Figure 2. 1: Identification of atypical ALT phenotype in AR-independent CRPC in vitro...... 82 Figure 2. 2: Validation of TMM characterization techniques...... 84 Figure 2. 3: SLX4IP promotes ALT-like PML localization events in CRPC in vitro...... 85 Figure 2. 4: Expression of SLX4IP in mCRPC versus primary disease...... 87 Figure 2. 5: SLX4IP knockdown is accompanied by disappearance of ALT- like PML localization events and accelerated telomere shortening...... 88 Figure 2. 6: SLX4IP knockdown correlates with accelerated telomere shortening...... 90 Figure 2. 7: SLX4IP knockdown triggers accumulation of senescence- associated markers...... 91 Figure 2. 8: SLX4IP knockdown is accompanied by a blunted atypical ALT phenotype...... 93 Figure 2. 9: SLX4IP knockdown promotes telomere shortening and senescence...... 94 Figure 2. 10: SLX4IP knockdown in AR-independent CRPC leads to ALT- like hallmark loss, reduced tumor volume, and induction of senescent- associated markers in vivo...... 95 Figure 2. 11: Androgen deprivation triggers SLX4IP-dependent ALT-like PML localization events...... 97

Figure 2. 12: Androgen-independent marker expression and telomere length changes following androgen deprivation...... 99 Figure 2. 13: SLX4IP knockdown triggers accumulation of senescence- associated markers following androgen deprivation...... 100

Figure 3. 1: N-terminus of SLX4IP promotes APB structures in C4-2B cells .... 122 Figure 3. 2: SLX4IP constructs are unable to recapitulate ALT-like phenotype observed with full-length SLX4IP...... 124 Figure 3. 3: N-terminus of SLX4IP induces additional APB-positive cells in DU145 and PC-3 cells...... 125 Figure 3. 4: SLX4IP constructs are unable to exacerbate the ALT-like phenotype of AR-negative CRPC models...... 126 Figure 3. 5: SLX4IP expression directly correlates with endogenous PML expression and number of PML nuclear foci in CRPC in vitro models...... 128 Figure 3. 6: Genetic manipulation of SLX4IP expression alters number of PML nuclear foci but not PML expression...... 129 Figure 3. 7: Telomeric localization of SLX4IP at APB foci occurs in DU145 and PC-3 cells...... 131 Figure 3. 8: N-terminus of SLX4IP coordinates telomeric localization at APBs...... 133 Figure 3. 9: Loss of APBs following SLX4IP knockdown is rescued by SLX4IP N-terminus introduction...... 135 Figure 3. 10: SLX4IP N-terminus prevents late-onset accumulation of senescent markers...... 137 Figure 3. 11: N-terminus of SLX4IP is required for telomere maintenance in ALT-like CRPC in vitro...... 139

Figure 4. 1: The regulation of SLX4IP-dependent ALT-like telomere maintenance in CRPC in vitro...... 152

ACKNOWLEDGEMENTS

This work would not have been possible without my incredible support system and their continued confidence in me throughout my academic journey.

Firstly, I would like to thank Dr. Derek Taylor for being the most incredible and supportive boss and mentor. Derek has given me so much creative freedom and independence to drive my research project. Because of this, I have developed a new level of confidence and self-reliance in the lab that I will take with me as I close this chapter of my academic career. I am also extremely grateful for the opportunities that Derek allowed me to pursue (and helped me to find) outside of the lab to set me up for success in my future career, even if that meant less time for research. This level of support is what truly defines a great boss and mentor and without Derek’s mentorship, I would not be where I am today.

I am also thankful for the support and mentorship I have received from my thesis committee members, Drs. Marvin Nieman, Magdalena Grabowska, Jason

Mears, and William Schiemann. I tell others all the time about how lucky I am to have such a fantastic thesis committee. Each member has played an important role during my journey as a Ph.D. student and I greatly appreciate all that they have helped me achieve. Guidance and mentorship from Bill early on in my journey really allowed me to hit the ground running, setting me up for success. Magda has provided me a level of guidance and mentorship that makes me feel as if I am an honorary member of the Grabowska Lab. Jason has helped me grow as an educator, and the teaching opportunities he has helped me find will be invaluable for my future career. Marvin is always finding opportunities for me to gain 9

experience in the higher education space and for his efforts, I am extremely grateful. Additionally, I want to thank Drs. Ruth Keri and Amy Wilson-Delfosse for their informal mentoring. Their support and insight has been incredibly valuable throughout my journey. I also want to thank the entire pharmacology department because so many faculty members, postdocs, and graduate students have helped me along the way, both in and outside of the lab. Lastly, I would not have been able to this without my amazing lab mates. While their scientific expertise helped propel my research forward, their friendship and emotional support is what means the most in the end. I would especially like to thank Wilnelly Hernandez-Sanchez,

Mengyuan Xu, Magdalena Malgowska, Xuehuo Zeng, Nathaniel Robinson, Daniel

Leonard, and Wei Huang.

I would not be on this path if it were not for Dr. Ryan Schneider, who introduced me to pharmacology research and education in pharmacy school. My interactions with Schneider showed me how rewarding a career in academia as a

Pharm.D./Ph.D. could be and his example helped solidify my career aspirations.

Though we constantly joke about how I am Schneider’s “mini-me”, I truly hope that one day I can achieve his level of excellence as an educator, mentor, and leader.

I am so lucky to have Schneider as a life-long mentor and friend and I would not have been able to do this without his guidance.

Lastly, I would like to thank my family and friends for their encouragement and support during my seemingly endless academic career. Thank you to my parents, Kim and Bob, for their love and unwavering confidence in me. They have been there for me through every snowy college softball game, every stressful

conference presentation, and every weekend I needed help entertaining our dogs so I could write without dog toys appearing on my keyboard. I am grateful for every effort and sacrifice they have made to help me succeed. A special thank you goes to my husband, Chris, for being there for me every single step of this journey.

Thank you Chris for listening to every presentation at least twice, knowing exactly when I need a science hiking break, listening to me rant about failed experiments, and so much more. I am grateful for the sacrifices Chris has made to allow me to pursue my career aspirations and for always keeping my crazy in check during this stressful process. I would not have been able to do this without his love and support.

SLX4 Interacting Protein (SLX4IP): A Vital Primer for Alternative Lengthening of Telomere (ALT)-like Processes Promoting Replicative Immortality in Castration-resistant Prostate Cancer with Androgen Receptor Loss

Abstract

TAWNA L. MANGOSH

Telomere maintenance mechanisms (TMMs) are responsible for instilling replicative immortality, with the two TMMs being telomerase or the alternative lengthening of telomeres (ALT). TMMs were thought to be mutually exclusive however, recently ALT hallmarks were identified in traditionally telomerase- positive cancers, supporting TMM coexistence. Moreover, activation of both TMMs can occur following disease progression, like in prostate cancer.

In advanced prostate cancer, therapeutic interventions unavoidably trigger progression to Castration-resistant prostate cancer (CRPC). CRPC is acquired through resistance mechanisms that can be grouped as either AR-positive or AR- negative. Therapeutics have been developed for CRPC however, they provide little benefit in AR-negative disease.

In the context of TMMs, primary prostate cancer uses telomerase exclusively and CRPC was assumed reliant on telomerase by extension. However, recent identification of ALT hallmarks in CRPC, which were lacking in the primary

tumor, suggests TMM coexistence occurs during progression to CRPC. However, it remains unclear if TMM coexistence correlates with AR status.

Here, we show that AR-positive CRPC models demonstrated telomerase activity but lacked ALT hallmarks while AR-negative models exhibited TMM coexistence. The presence of ALT hallmarks in AR-negative CRPC directly correlated with expression of SLX4IP, a known ALT regulator. In AR-negative

CRPC, SLX4IP knockdown caused ALT hallmarks loss, telomere shortening, and senescence. In a model of AR-negative CRPC progression, AR loss was accompanied by SLX4IP-dependent ALT hallmarks. Notably, reduced SLX4IP expression prevented ALT hallmarks, accelerated telomere shortening, and induced senescence in our model of disease progression.

Efforts to understand how SLX4IP promotes TMM coexistence in AR- negative CRPC revealed the N-terminus of SLX4IP promoted ALT hallmarks similar to that of the full-length protein. Moreover, introduction of the N-terminus of

SLX4IP partially rescued the telomere-induced senescent phenotype triggered by endogenous SLX4IP knockdown. Further investigation revealed the N-terminus directs SLX4IP’s localization at the telomere, priming sites for ALT hallmark formation and downstream telomere maintenance.

These findings highlight the unique dependence of AR-negative CRPC on

SLX4IP-mediated ALT hallmarks which are crucial for maintaining telomeres and replicative immortality. Our studies provide a foundation for future investigations targeting TMM coexistence through SLX4IP to fulfill the therapeutic need of difficult to treat AR-negative CRPC.

CHAPTER 1: INTRODUCTION AND BACKGROUND

Portions of this chapter were previously published in:

Whited, T. L., and Taylor, D. J. "Expanding the chemotherapeutic potential of an established nucleoside analog with selective targeting of telomerase." Molecular

& Cellular Oncology 5.6 (2018): e1536844.

1.1 Telomeres and Replicative Capacity

Telomeres function as the protective cap of eukaryotic chromosomes.1–42

These highly dynamic and tightly regulated structures consist of a tandemly repeating DNA sequence that forms a double-stranded region followed by a single- stranded G-rich overhang.2,4 The double-stranded regions range in length from ~2 to 30 kb with a single-stranded overhang of ~150 bp.2,4 The primary purpose of this repetitive sequence is to serve as a nucleotide buffer to prevent loss of genomic information.5–7 Because of the end-replication problem associated with semiconservative DNA polymerases, linear chromosomal ends cannot be completely replicated which leads to a loss of ~50 to 200 distal nucleotides with each cellular division.5–7 Without telomeres, this net loss would be deducted from essential genetic information. Therefore, the repetitive telomeric sequence provides this nucleotide buffer to prevent genomic erosion.

These DNA repeats, specifically 5’-TTAGGG-3’ in mammals,4 is protected from recognition by DNA damage response proteins by a sheath of multimeric protein complexes coined shelterin.8 The shelterin complex consists of six telomere-specific proteins: two double-stranded DNA binding proteins TRF1 and

TRF2 (Telomere Repeat Binding Factor 1 and 2) and the single-stranded DNA binding protein POT1 (Protection of Telomere 1); all of which directly bind telomeric

DNA.9–11 RAP1 (Repressor/activator Protein 1) binds directly and modulates the function of TRF2.12 TIN2 (TRF2-interacting Nuclear Factor 2) and TPP1 (ACD gene) act as a bridge tethering the double and single-stranded telomere-binding proteins.13 Loss of but one shelterin component is sufficient to elicit aberrant DNA

damage response processes;11,14–17 as such, the primary role of shelterin is to ensure the integrity of this nucleoprotein genomic buffer via camouflage.

1.1.1 The Dynamic Nucleoprotein Structure of Telomeres

The representation of telomeres as static nucleotide buffers, only modulated by the end-replication problem, does not accurately reflect the dynamic regulation of this nucleoprotein’s structure and length. In addition to the end- replication problem, some telomeric processing can lead to the loss of telomere length. For instance, TRF1 and TRF2 are able to manipulate telomeric DNA in a length-dependent manner.18,19 If telomere length reaches a specific threshold,

TRF1 and TRF2 promote telomeric bending where the 3’ single-stranded DNA overhang invades the double-stranded segment forming new base pairing and stabilizing the formation of telomere loops or t-loop structures.18,19 These t-loops are thought to block DNA replication and telomere elongation machinery from accessing the affected regions and prevents the degradation of the 3’ overhang.18,20–22 However, if t-loops remain unresolved during DNA replication the entire t-loop region is lost.22 Telomere trimming can also occur at t-loops to eliminate extremely long telomeres and cleavage of these structures results in circular extrachromosomal telomeric repeats (ECTR).20–22 Similarly, loss of telomeric DNA can occur if G-quadruplex structures, a four-stranded DNA secondary structure formed by the 3’ G-rich overhang, are unresolved at the time of DNA replication.21 These telomeric processing steps lead to the formation of telomeric ends that more closely resemble double-stranded DNA breaks (DSB).

To ensure the 3’ overhang remains intact and of sufficient length, the 5’ telomeric ends are shortened by exonucleases.23 This exonuclease processing is somewhat counteracted by the CST (CTC1, STN1, TEN1) complex, which provides some degree of 5’ end reconstitution while still preserving the 3’ overhang.23,24 However, the net effect of these processing steps is still telomeric loss. Though telomeres function as a buffer to avoid genomic attrition, there is a finite number of cell divisions that can be buffered by telomeres because of the net loss of nucleotides via the telomeric processing steps described above.

1.1.2 The Terminal Fates of Telomeric Attrition

In somatic cells, telomeres progressively shorten with each replication cycle and, if appropriate safeguards are not in place, telomere attrition can lead to deleterious genomic consequences (Figure 1.1A).5,7,11 When telomeres are of sufficient length and protected by shelterin, chromosomal ends are unable to be recognized as DSB by DNA damage response proteins.8 However, once the minimum threshold length is reached, shelterin binding is lost and telomeres are subjected to aberrant DNA damage signaling.11 Similarly, impairment of telomeric protective factors phenocopy this consequence and inhibition of TRF2 is commonly used as a model for inducing telomeric dysfunction.11,14–17 Loss of TRF2 binding leads to rapid degradation of the 3’ telomeric overhang so telomeric ends resemble DSB thereby triggering DNA damage signaling.17

Initially these unprotected telomeric ends, generated through telomere attrition and shelterin destabilization, are marked by phosphorylated γH2AX

foci.2526 These foci contain additional DNA damage and repair factors such as

53BP1 (p53 Binding Protein 1), Rad17, Mre11 (Meiotic Recombination 11), NBS1

(Nibrin), ATM (Ataxia-telangiectasia Mutated) and many others.26 These complex dysfunctional telomeric structures coined TIFs (telomere dysfunction-induced foci) are a marker commonly used to evaluate telomere integrity in somatic and malignant cellular populations.27 Telomeres marked by these DNA damage

“sensors” can undergo chromosomal fusions via non-homologous end joining

(NHEJ), leading to significant genomic instability.10,11,17 However once a telomere reaches the critically short length threshold,28 DNA damage checkpoints trigger cell cycle arrest to prevent further propagation of genomic instability (Figure

1.1A).28–31 The consequence of this tumor suppressive mechanism is p53- dependent senescence or apoptosis (Figure 1.1A).31,32

However, if growth arrest checkpoints are bypassed through tumor suppressor disruption, like p53 loss, cellular populations undergo telomeric crisis defined by rapid telomere shortening and further chromosomal fusions.33 Because of this compounding genomic instability, the majority of the cellular population does not survive (Figure 1.1A).33 However, those cells able to escape telomeric crisis harbor incredibly unstable genomes due to tumor suppressor loss and chromosomal fusion-breakage cycles,33,34 making them well positioned for malignant transformation.35,36 As a prerequisite, telomere maintenance mechanisms (TMMs) must be activated during this catastrophic genomic event to elongate telomeres (Figure 1.1A).35–38 This ensures a fine balance between sufficient genomic instability to ensure malignant propagation without being

excessive to a point that is not conducive for cell survival. Additionally, activation of TMMs instills replicative immortality, a defining characteristic of malignant transformation.39 The two TMMs promoting unlimited replicative capacity in cancer are telomerase and the alternative lengthening of telomeres (ALT) pathway (Figure

1.1A).38,40–42

Figure 1. 1. Telomere maintenance mechanisms instill replicative immortality in cancer. (A) Schematic representation of cell fates triggered by continued telomere shortening over consecutive replicative cycles. (B) Schematic representation of the mechanism behind telomerase-mediated telomere elongation followed by CST-mediated strand fill-in. (C) Schematic representation of the mechanism behind ALT-mediated telomere elongation.

1.2 Orthodox Telomere Synthesis via Telomerase

Telomerase is an enzyme capable of catalyzing the addition of telomeric repeats to the ends of linear chromosomes to promote telomere elongation and instill additional replicative capacity.40,41 In healthy somatic cells, with the exception of stem cells, telomere length dictates the lifespan and replicative limit of the cellular population.28 Because of this restriction on replication, telomerase is not needed for telomere maintenance in somatic populations and therefore expression and activity is undetectable.37,41

In stem cells, a prolonged proliferative lifespan is necessary so telomerase is activated to ensure telomere length is maintained over additional cell divisions.43

However, telomerase activity in the majority of adult stem cells is relatively low; in rare instances, telomerase upregulation can be triggered following stem cell commitment to a lineage that undergoes rapid expansion and turnover like hematopoietic progenitor cells.43 Despite this telomerase activity, the telomeres of stem cells still shorten with replicative ageing triggering senescence, albeit at slower rates than somatic cells.43–45

However during the process of malignant transformation, telomerase is activated and upregulated in ~80 to 95% of all cancers.37 This activation leads to the elongation and maintenance of telomeres just beyond the critical length limit but not much further, resulting in relatively shorter telomere lengths.46 In cancer cells, activation of telomerase ensures replicative immortality coupled with the ideal balance between genomic instability and integrity needed for cancer cell survival.

1.2.1 Telomerase Structure and Function

Telomerase’s unique ribonucleoprotein complex minimally consists of a catalytic reverse transcriptase component, TERT (telomerase reverse transcriptase), and an RNA template complementary to the repetitive telomeric sequence, TR or TERC (telomerase RNA component).1 To promote extension, telomerase is recruited to the telomere via interactions with the shelterin component TPP1 (Figure 1.1B).47–49 Following recruitment, telomerase aligns five nucleotides of the TR template with the 3’ telomeric end (Figure 1.1B).50 The next six nucleotides of TR are used as a template by TERT to synthesize the addition of the telomeric hexametric sequence (Figure 1.1B).4,50 After reverse transcription, telomerase translocates on the telomere to realign TR with the 3’ end to restart the catalytic cycle (Figure 1.1B).50 Once the 3’ telomeric overhang is sufficiently extended, the CST complex synthesizes fill-in of the complementary strand, thereby extending the double-stranded telomeric region while preserving a portion of the telomerase-extended 3’ overhang (Figure 1.1B).23,24

1.2.2 Structural and Functional Regulators of Telomerase

The regulation of telomerase assembly and function relies on a multitude of factors beyond TERT and TR. Additionally, there are a number of telomerase- associated proteins that stably interact with telomerase forming a complete telomerase holoenzyme structure.51 These factors include dyskerin, NHP2

(Nuclear Protein Family A, Member 2), NOP10 (Nuclear Protein Family A, Member

3) and GAR1 (Nuclear Protein Family A, Member 1) for example.51 Highlighting the importance of these telomerase-associated proteins, disease-causing mutations have been identified in the genes encoding each of these factors, which result in rapid degradation of TR.52–54

Further promoting telomerase function, a number of chaperones involved in assembly and trafficking of telomerase associate transiently. For instance, TCAB1

(Telomerase Cajal Body Protein 1) can bind telomerase to shift its distribution to the nucleoplasm where telomerase-telomere interactions occur and mutant

TCAB1 causes diseases of premature telomere shortening.53,55 Together, a variety of telomerase-associated factors are required for the proper organization and trafficking of telomerase to the telomere to participate in telomeric expansion. Once telomerase is shuttled to the telomere, it is recruited via TPP1 through its TEL- patch surface.49 The TEL-patch surface provides the interaction site for TERT domains, triggering both telomerase recruitment and activation at the telomere.47–

49 Only the TPP1-TERT interaction has been shown necessary for telomerase- mediated telomere elongation. Additional regulatory effects on telomerase stem from telomeric chromatin dynamics and secondary telomeric DNA structures, but the impact of these factors is less defined.21,56

1.2.3 Transcriptional Regulators of Telomerase

Despite the complexity of telomerase holoenzymes, only TR and TERT are necessary for the establishment of telomerase activity in malignancy.1 TR is ubiquitously expressed across tissue types and is regarded as a non-limiting factor

for telomerase activity.57 However, TERT mRNA expression is tightly regulated transcriptionally and directly correlates with telomerase activity, making TERT the primary determinant for telomerase activation.37,41

Malignant transformation requires telomerase activation via upregulation of

TERT expression.37,41 This upregulation occurs through a number of genetic mechanisms, such as TERT amplifications,58 genomic rearrangements,59 splice variants, 60 and promoter mutations,61 or epigenetic mechanisms, such as promoter methylation62 and microRNAs.63 In addition, a plethora of transcription factors bind the TERT promoter region to modulate gene expression, including c-

Myc,64 ER (estrogen receptor),65,66 AR (androgen receptor),67,68 NF-κB (Nuclear

Factor- κB),69,70 STAT3 (Signal Transducer and Activator of Transcription 3),71 and

STAT5 (Signal Transducer and Activator of Transcription 5)72 to name a few.

Altogether, these efforts to define transcriptional programs dictating TERT expression have highlighted that this regulation is extremely complex involving the interplay of a number of coexisting molecular pathways commonly dysregulated in malignancy.

1.3 The Alternative Lengthening of Telomeres (ALT) Pathway

In the remaining ~10 to 15% of cancers, ALT is engaged to elongate telomeres and establish replicative immortality.42 ALT is primarily engaged in cancers of mesenchymal and neuroepithelial origin.73–76 In contrast to telomerase, relatively little is known about its exact protein requirements and the mechanism behind ALT-mediated telomere maintenance. Minimally, ALT is known to rely upon

homologous recombination (HR) to promote extension, utilizing telomere-based templates for synthesis of additional repeats.42,77,78 Because the ALT mechanism is poorly characterized, ALT activity cannot be directly measured like telomerase.

Therefore, in order to identify the presence of ALT, a number of surrogate measures of ALT activity must be identified.79 The majority of these surrogate measures are direct byproducts of HR events essential for ALT-mediated telomere elongation.79 Additionally, these telomere-directed HR events necessary for ALT repurpose and recruit a variety of DNA damage response proteins already present promoting genomic integrity.80 Multiple DNA damage response pathways converge at telomeres to elicit telomere elongation and ALT byproduct formation, but detailed mechanistic insight is lacking in the ALT field. Without this knowledge of ALT’s molecular mechanism, TMM governance of replicative immortality cannot be fully understood.

1.3.1 ALTernative Mechanism of Telomere Elongation

Telomeres are relatively challenging to replicate by DNA polymerases because of their repetitive sequence and propensity to form secondary structures, such as t-loops and G-quadruplexes.21,81,82 Therefore, telomeric regions are prone to replicative stress at baseline, which is defined as the stalling of DNA replication fork progression and DNA synthesis deficiencies.38 Moreover, when DNA damage checkpoint signaling is compromised during telomeric crisis and malignant transformation, replicative stress accumulates as cells in crisis progress through additional replication cycles in the presence DNA lesions and genomic

instability.32,33,35,36 ALT telomeres are particularly prone to replicative stress and this is suggested to be an initiating event for ALT activity.83,84

In addition to being susceptible to elevated replicative stress, the compaction of chromatin at ALT telomeres is relaxed compared to telomerase- positive environments.85,86 Firstly, this increased accessibility renders an HR- permissive state at ALT telomeres allowing for recruitment of DNA damage response proteins.85,87 Secondly, mutations in the chromatin remodelers ATRX

(Alpha Thalassemia/Mental Retardation Syndrome X-linked) and DAXX (Death

Associated Protein 6) are a common occurrence in ALT cancers.88–90 The loss of functional ATRX/DAXX leads to replicative stress through altered nucleosome assembly at telomeres.86 Moreover, introduction of functional ATRX in ALT in vitro models lead to a loss of ALT activity.89 These findings suggest that the ALT telomeric architecture significantly varies from that of telomerase-regulated telomeres and these prerequisites make ALT telomeres poised for HR.

Replicative stress leads to the slowing and even stalling of replication fork progression at ALT telomeres.21,81,82,91 This stress can be alleviated by replication fork regression, removal or repair of DNA lesions, and replication fork renewal.83

However if the replicative stress is not alleviated, replication fork collapse occurs generating a double stranded telomeric break (Figure 1.1C).92 ATM kinase activation and recruitment of the MRN (Mre11, Rad50, NBS1) complex results, ultimately leading to telomere end processing to regenerate the 3’ overhang to allow for a homology search (Figure 1.1C).83,93 At this point, the regenerated telomeric terminal can be processed through NHEJ or HR.80,86,92,94 The presence

of TRF2 at telomeres inhibits NHEJ of telomeric ends preventing chromosomal fusions and TRF2 is needed to recruit DNA damage proteins and ALT mediators to the telomere.95 For these reasons, regenerated telomere termini are preferentially processed via HR after 3’ telomeric overhang invasion occurs of a nearby double-stranded region (Figure 1.1C).83,93 The homology search and intertelomeric strand invasion triggers break-induced telomere synthesis (Figure

1.1C).96,97 This is followed by BLM (BLM RecQ Like Helicase)-mediated promotion of telomere extension by DNA polymerases followed by dissolution the HR junction

(Figure 1.1C).83,97,98 Because incredibly long telomere tracts can be detrimental, tipping the genomic instability scale from favorable for malignant progression to unsuitable for cell survival, ALT safeguards are in place to circumvent continuous extension. After triggering break-induced telomere synthesis, the resolution of recombination intermediates by SLX4 (BTBD12 or FANCP) and associated nucleases results in a telomeric crossover event so the telomeric population is diversified without a net change in telomere length.99–101 The recruitment of conserved DNA repair pathways to alleviate replicative stress at ALT telomeres while concurrently promoting telomere elongation provides a starting point for understanding the complexity of ALT’s molecular mechanism.

1.3.2 Identification of ALT Activity Through Surrogate Measures

Unlike telomerase, ALT activity cannot be directly measured and instead surrogate measures are used to evaluate the ALT status of a cellular population

(Figure 1.2).79 These surrogate measures are unique and specific for cells using

ALT as their TMM of choice. These surrogate measures include extrachromosomal telomeric repeats (ECTR),102,103 telomere sister chromatid exchange (TSCE),77 extreme telomere length and heterogeneity,42 and ALT-associated PML bodies

(APBs).104

Arguably, the most unique ALT hallmark is the abundance of telomeric DNA repeats in cells that are separate from chromosomes. These extrachromosomal

DNA structures can present as double-stranded telomeric circles or t-circles, large branched atypical telomeric DNA complexes or t-complexes, linear double- stranded telomere segments, or as partially single-stranded telomeric circles.

102,103,105 Additionally, homology searches and strand invasion necessary for ALT- mediated recombination is thought to utilize circular telomeric species in a similar fashion as nearby telomeres; however, the circular template triggers rolling circle amplification (RCA) for telomere synthesis (Figure 1.2A).105,106

Partially single-stranded telomeric circles are the most frequently evaluated hallmark of the ECTR species listed (Figure 1.2A).102,103,106 These particular telomeric circles can be further classified as either C-circles or G-circles based on whether the circle originates from the G-rich or C-rich telomeric strand (Figure

1.2A). Specifically, C-circles are correlative with ALT status, extremely responsive to changes in ALT activity, and easily quantifiable.105 Therefore, C-circle abundance is regarded as a “gold standard’ hallmark of ALT.105 Because C-circles contain both double stranded and single stranded DNA regions within their circular structures they are self-primed at the double-stranded region originating from the telomeric G-strand.105 This self-priming poises these structures for RCA by Φ29

polymerase, a highly processive DNA polymerase capable of producing RCA products of up to ~70 kb (Figure 1.2A).105 Once amplified, these C-circles can be hybridized with a telomeric probe and detected via chemiluminescent, fluorescent, or radioactive detection (Figure 1.2A).105 Generally, cells are deemed ALT-positive if their relative C-circle abundance ratio of amplified reaction versus control reaction lacking Φ29 polymerase is greater than or equal to two.105

Another byproduct of telomeric HR marking ALT cells is telomere sister chromatid exchange (TSCE) (Figure 1.2B).77 TSCE occurs more frequently in ALT cells versus telomerase-positive cells, where TSCE is nearly undetectable.77 This elevation in TSCE is not accompanied by an increase in genomic SCE.77 The molecular explanation behind TSCE in ALT remains elusive. TSCE can be detected after cellular replication in the presence of BrdU followed by hybridization of telomeric DNA with a fluorescent G- and/or C-probe.107 The newly synthesized daughter strands with BrdU incorporation are then degraded.107 Telomerase- positive cells will demonstrate a single G-probe and C-probe signal at each chromatid end.107 However, if this process reveals two G-probe and C-probe signals at either one or both chromatid ends, TSCE has occurred and ALT is likely engaged at the telomere (Figure 1.2B).107,108 The current theory is that telomeric

DNA breakages occur following DNA replication, such as with replication fork collapse.83,109 This triggers unequal TSCE and non-random segregation of sister chromatids resulting in one daughter cell with telomere extension and additional replicative capacity and one daughter cell with telomere shortening and restricted replicative capacity.110 This proposed mechanism assumes cells are inherently

programmed to ensure lengthened telomeres are segregated into a single daughter cell.110 While this proposal explains the unique presence of TSCE in ALT, it does not fit with HR-mediated model described previously and has not been experimentally proven. Though unequal TSCE and HR-mediated ALT mechanisms are inherently different, the mutual exclusivity of these two mechanisms has yet to be determined.

While it remains unclear which of the three proposed mechanisms of ALT

(HR-mediated, telomeric RCA, and unequal TSCE) are responsible for telomere elongation and if they coexist, there is a commonality amongst the three mechanisms: generation of extreme telomere length and significant length heterogeneity.42 HR-mediated telomere synthesis can generate extremely long telomeres, which is why SLX4 is inherently programmed to trigger telomere crossover events to prevent disadvantageous telomere overextension.97,108,111

RCA of telomeres using circular ECTR is thought to similarly generate extremely extended tracts of telomeric DNA further elevating the average telomere length of cells using ALT.112,113 While HR-mediated and RCA-mediated telomere synthesis creates extremely long telomeres within the telomeric population, unequal TSCE is thought to develop the significant telomere length heterogeneity demonstrated in ALT cells.110 TSCE is thought to preferentially select one daughter cell to retain long telomeres rendering the other daughter cell telomerically deficient, thereby creating a wide distribution of telomere lengths.110 Together these proposed ALT mechanisms explain the unique telomeric phenotype exhibited by ALT cells.

1.3.3 ALT-associated Promyelocytic Leukemia Bodies

ALT-associated PML or APBs differ from the other ALT hallmarks described above in that APBs are complex nuclear structures necessary for the orchestration of ALT activity (Figure 1.2C).104 These nuclear structures are large ALT-specific foci containing PML protein and telomeric DNA. APBs are identified via IF or IF-

FISH through colocalization of PML with either TRF2 or telomeric DNA hybridized with a telomeric probe (Figure 1.2C).104 While these components are minimally required to be defined as an APB, the structure and constituents of APBs are incredibly complex. APBs generally contain large clusters of telomeres, DNA damage response proteins, and telomere-binding proteins positioning the telomeres involved in the most opportune environment for ALT-mediated recombination.104,114–117 Additionally, PML is vital for the formation of APBs because depletion leads to loss of telomere clustering and ALT-mediated telomere synthesis.118 Despite the essentiality of APBs, the regulation of APB assembly and how APB-mediated proximity of ALT components triggers telomere maintenance is unknown.

One hypothesis that exists is that the enrichment of SUMOylated telomere components recruits PML to orchestrate the direction of ALT-mediators and promote telomere clustering.119 PML foci or PML bodies beyond the telomere are formed through numerous SUMO/SIM (SUMOylation/SUMO-interaction motif) interactions between PML proteins. These interactions create a structure that forms a membrane-less organelle via liquid-liquid phase separation (LLPS).119,120

LLPS refers to the process of de-mixing a single homogenous solution into two

distinct phases, one of which is a condensed liquid phase and the other a dilute liquid phase.120 Similarly, SUMO/SIM-facilitated LLPS has been suggested to drive

APB formation.121 When fusion proteins containing multiple SUMO and SIM sites were attached to telomeric sites in telomerase-positive cells nuclear condensates containing PML and telomere clusters rapidly formed, mimicking the traditional

APB presentation.121 Together, these findings support APB formation as a prerequisite for ALT by forming nuclear condensates placing ALT mediators and telomeric substrates in close proximity. However, APB formation alone is not sufficient for telomere synthesis.121 Compounding upon the knowledge void surrounding the complex ALT mechanism, key regulators of APB formation and

ALT activation at these foci must be identified to accurately describe TMMs in malignancy.

Figure 1. 2. Evaluation of ALT hallmarks identify cells primarily using ALT for telomere maintenance. (A) Schematic representation of ALT hallmark detection including telomeric C- circle abundance, (B) telomere sister chromatid exchange and (C) ALT-associated PML body formation.

1.4 Telomere Maintenance Mechanism Plasticity

Historically, telomerase and ALT were thought to be mutually exclusive.122

Additionally, ALT was thought to occur only in cancers with undetectable telomerase activity.38 Moreover, telomerase negativity was previously used as a hallmark of ALT.79 However, recent findings suggest that both TMM pathways are inherently programmed in malignancy and can coexist at varying degrees within a single cell or within a cellular population.75,123–125 This evidence for a complex TMM continuum suggests that the designation of telomerase-positive or ALT-positive status does not accurately represent malignant TMM phenotypes, specifically in the case of TMM coexistence. Moreover, cases of TMM plasticity have identified opportunities for cancer cells to engage the secondary TMM if for any reason the primary TMM is no longer favorable.107,126,127 Understanding how replicative immortality is instilled and perpetuated through cancer progression requires the elucidation of the purpose behind TMM plasticity and coexistence and key drivers of this phenomenon.

1.4.1 Engineered Telomere Maintenance Mechanism Coexistence

The coexistence of telomerase and ALT was first demonstrated through several genetically engineered in vitro approaches. For instance, ectopic expression of TERT and TR led to telomerase reconstitution in WI-38 VA13 cells, an immortalized fibroblast cell line that relies upon ALT.128 In this model of engineered TMM coexistence, telomerase activity and APBs were present for over

90 population doublings.128 Upon visualization of telomere length in this

coexistence model, extremely long telomeres characteristic of ALT were maintained and relatively short telomeres were preferentially elongated.128 The result was reduced telomere length heterogeneity with increased average telomere length over time.128 Similarly, overexpression of TERT in ALT-positive immortalized fibroblasts, GM847 cells,129 and immortalized ovarian epithelial cells,

HIO118 and HIO107 cells,130 triggered preferential elongation of short telomeres while maintaining extremely long telomeres and APBs in long-term culture. These findings support the capability of cancer cells to use telomerase and ALT concurrently to instill additional replicative capacity without any obvious detrimental effects.

Successful development of TMM coexistence models calls attention to the lack of knowledge surrounding TMM regulators. Specifically, if telomerase and ALT can promote telomere maintenance in parallel then what regulators are triggered to dictate the use of one TMM over the other in the majority of cancers.

Telomerase reconstitution studies clearly establish that telomerase is not a repressor of ALT activity because of the continued presence of APBs and extreme telomere length.129 However, the fusion of telomerase-positive cell lines with ALT- positive GM847 cells allowed for only short term TMM coexistence; but, long-term culture revealed shorter average telomere length, limited telomere length distribution, and disappearance of ALT hallmarks.129 Though telomerase is unlikely to fill this role, an unknown ALT repressor must be present in telomerase-positive cells for these effects to be demonstrated in telomerase-ALT hybrids.129 Because of the elusive mechanism of ALT, ALT reconstitution in telomerase-positive cells

has yet to be investigated in a similar fashion. However, based on the generation of these hybrids it is unlikely that ALT can be activated in a telomerase-positive environment with an intact ALT repressive mechanism.129

Therefore, the most likely model of TMM coexistence begins with the activation of telomerase during malignant transformation.35–38 This is followed by an unknown telomeric event in which ALT repressive programs are lost so both

TMMs can work concertedly to ensure sufficient telomere maintenance.32,33,35,36

Because there are very few characterized ALT repressors and promotors, it is possible that these telomerase-ALT fusion hybrids are not completely accurate depictions of TMM coexistence.129 Little is known about the exact protein requirements for ALT and fusion of a telomerase-positive cell with an ALT-positive cell may alter or dilute these ALT prerequisites, rendering ALT disadvantageous in the absence of a true ALT repressive mechanism. Identifying and understanding key regulators of TMM plasticity promoting coexistence is crucial for understanding how cancer exploits TMMs to develop the most advantageous telomeric environment.

1.4.2 ALT as a Mechanism of Therapeutic Resistance

The selective requirement of replicative immortality for malignant cells, but not somatic cells, makes TMMs an attractive and selective therapeutic target for cancer.131 Telomerase is particularly advantageous because of its relatively well- defined molecular mechanism, targetable enzymatic function as a reverse transcriptase, and upregulation in the majority of cancer types.37,50,53 Indeed,

antitelomerase therapeutic approaches have been developed but these approaches have highlighted clear disadvantages of telomerase inhibition; one such disadvantage is ALT activation as a mechanism of resistance.132

Triggering TMM plasticity to engage ALT in the absence of sufficient telomerase activity was first demonstrated following genetic modifications driving telomerase inhibition in vitro. The introduction of a dominant negative human TERT mutant in T24 bladder cancer cells led to a loss of telomerase activity, appearance of APBs, and increased telomere length and heterogeneity.126 Similarly, dominant negative TERT expression in HCT15 colon carcinoma cells triggered a similar

TMM switch but TSCE was evaluated in the place of APBs.107 The RNA template of telomerase has also been targeted resulting in a similar activation of TMM plasticity in a model of esophageal squamous carcinogenesis.133

These effects of genetic telomerase interference have not been directly shown with antitelomerase therapeutics but the above approaches highlight the concern for resistance via TMM plasticity. Presumably, TMM plasticity would be a potential concern for resistance to ALT inhibition, though true ALT inhibitors have yet to be developed because of its mysterious mechanism. ATR (Ataxia

Telangiectasia and Rad3-related Protein) inhibitors have been proposed as a therapeutic targeting ALT but this inhibition has not been rigorously evaluated.134,135 Nevertheless, TMM plasticity remains a primary concern when targeting TMMs as a cancer therapeutic. In order to understand TMM plasticity and overcome this hurdle of TMM resistance, key divers must be identified.

1.4.3 ALTiness versus Bona Fide ALT

True or bona fide ALT assumes that the malignant population in question exhibits all ALT hallmarks at robust levels.79 Traditionally, bona fide ALT cells must demonstrate undetectable telomerase activity coupled with multiple ALT hallmarks that dramatically differ from their telomerase-positive counterparts. However, the

ALT spectrum has recently been shown to be much more complex.38,79 Just as there is a TMM continuum spanning from telomerase to ALT, there exists a continuum of ALT phenotypes or ALTiness that do not fit under the archaic bona fide ALT definition. The concept of ALTiness was demonstrated in a variety of cancer types classically defined as telomerase-positive.75,136 These examples further support TMM plasticity and TMM coexistence as crucial components of malignant telomere homeostasis.

The first incomplete ALT phenotype was described, before ALT was even identified, when a group of telomerase-positive retinoblastomas demonstrated a singular ALT hallmark, being incredibly long telomeres.137 Moreover, early characterization of classically telomerase-positive cancers harbored extreme telomere length and heterogeneity.75 With the identification of additional ALT markers beyond telomere length distribution, ALTiness was identified in a number of tumor types previously defined as telomerase-positive including breast,75,124,136 colorectal,75,138 hepatic,75 melanoma,75,136 gastric,75,139 bladder75 and prostate cancer.75 Generally, this atypical ALT presentation included detectable telomerase activity to varying degrees coupled with at least one ALT hallmark, but not the complete bona fide ALT phenotype. Moreover, this coexistence can occur through

the divergence of telomerase and ALT-positive subpopulations within a single tumor or through the activation of both telomerase and ALT in a single cell.124,137

The identification of ALTiness in patients diagnosed with historically telomerase- positive cancers supports the clinical relevance of TMM coexistence and plasticity.

1.4.4 The Elusive Role of SLX4IP in Telomere Maintenance

Due to the uncharted nature of TMM coexistence, triggered through activation of TMM plasticity, switches dictating these presentations are poorly characterized. SLX4IP (SLX4 interacting protein) is one of the only TMM regulators known to elicit transitions from telomerase to ALT and vice versa.140–142 SLX4IP was originally identified as a binding partner of SLX4, an essential player in ALT- mediated telomeric recombination.143 Moreover, SLX4IP in coordination with the

SLX4-associated nuclease hub promotes intra-stand crosslink repair at genomic sites but it is unclear if this function translates to ALT-associated telomeres.144

Despite known interactions with ALT regulators, studies have yet to reveal the actual function of SLX4IP at the telomere.

Further complicating the elucidation of SLX4IP’s function, SLX4IP alterations and consequential effects on ALT seem to vary based on cancer type.

In bona fide ALT U2OS osteosarcoma cells and WI-38 VA13 immortalized fibroblasts, CRISPR-Cas9 mediated knockout of SLX4IP led to a significant increase in ALT markers including C-circles, APBs, TSCE events, and telomere length heterogeneity.140 In these true ALT models, SLX4IP is thought to promote

SLX4-medaited inhibition of telomere synthesis by BLM to ensure telomeres are

not overextended.140 Without SLX4IP, BLM activity is unchecked leading to overactive ALT and extremely long telomeres triggering cellular survival defects.140

Though this function of SLX4IP aligns with the proposed true ALT mechanisms outlined above, this regulation does not translate to models that trend toward

ALTiness and TMM coexistence.

The functional requirement of SLX4IP may vary based on whether bona fide

ALT is activated during malignant transformation or if ALT-like processes are adopted as a consequence of disease progression or therapeutic resistance.

Contradicting the above study, knockdown of SLX4IP in an in vitro model of breast cancer, that demonstrated TMM coexistence at baseline, triggered the adoption of a telomerase-positive phenotype.142 Specifically loss of SLX4IP significantly reduced C-circles and APBs and induced telomerase activity.142 Together, these two studies represent the only published efforts to elucidate the function of SLX4IP in the context of ALT. These conflicting roles of SLX4IP suggest that its function is either dependent upon cancer type or upon the time and basis for ALT activation.

To determine the role SLX4IP plays in TMM plasticity promoting ALTiness and if this SLX4IP-mediated mechanism mimics true ALT will require further investigation.

1.5 The Complexity of Prostate Cancer Disease Progression

Prostate cancer remains the second leading cause of cancer-related death in men in the United States despite the diagnosis of localized primary disease being associated with a very favorable prognosis.145 This statistic stems from those

patients presenting with advanced metastatic disease or presenting with localized disease that progresses following therapeutic intervention.146 Because prostate cancer relies upon the androgen receptor (AR) signaling axis, treatment revolves around androgen deprivation therapy (ADT) in cases of both localized and advanced prostate cancer (Figure1.3A).146 Localized disease is treated with combinations of surgery, radiation and ADT.146 One such ADT regimen is a luteinizing hormone-releasing hormone (LHRH) agonist, which stimulates the release of LH and a downstream surge of androgens. But, long-term LHRH agonist treatment triggers a negative feedback loop preventing further androgen production.147,148 LHRH agonists are used in combination with first-generation antiandrogens, like bicalutamide.148 These agents are AR antagonists preventing activation of AR by testosterone and its more potent form 5-dihydrotesterone

(DHT).149 When used with LHRH agonists, antiandrogens prevent the initial LHRH- associated androgen surge from promoting prostate cancer growth.147,148

Advanced or metastatic disease is similarly treated with radiation in combination with ADT regimens.146 However, because these therapies are not curative in advanced disease, a subset of patients will progress on their initial ADT to castration-resistant prostate cancer (CRPC) that harbors a bleak prognosis

(Figure1.3A).150,151

1.5.1 Acquisition of Castration Resistance in Advanced Disease

The clinical course of prostate cancer progression to CRPC is extremely heterogeneous due to the variety of resistance mechanisms adopted to circumvent

ADT.68,146,151–154 Adding further complexity, drivers of these resistance mechanisms are poorly understood and no single mechanism alone reliably accounts for CRPC progression in patients. The majority of CRPC investigations focus on mechanisms restoring AR signaling despite maintained application of

ADT.155,156

In CRPC, continued AR signaling in the face of ADT can be explained through a number of molecular mechanisms identified in both experimental models and patients. Firstly, CRPC growth is stimulated via the production of intracrine androgens by cancer cells using adrenal steroids and cholesterol as substrates for androgen synthesis.157 Secondly and the most widely studied mechanisms triggering AR signaling reactivation are AR alterations.68,158,159 Elevated expression of AR can be achieved through amplification of the AR gene locus and in fact, one study demonstrated 30% of CRPC tumors harbored AR gene amplifications that were not present in patient-matched primary tumors.160,161

Similarly, AR overexpression can be triggered through transcriptional or translational upregulation and this effect sensitizes CRPC cells to minimized androgen levels and overcomes antiandrogen inhibition.162,163 ADT can also select for mutations within the ligand-binding domain of AR, which occurs in 15 to 20% of CRPC patients.164–169 These AR mutants either promote androgen hypersensitivity or broaden ligand specificity to include steroidal hormones beyond androgens. With certain AR mutations, changes in the ligand-binding domain exploits the presence of antiandrogens so that they now agonize AR, driving CRPC growth.168,169 Lastly, upregulation of alternative splicing generates constitutively

active AR splice-variants (AR-Vs).158,170,171 The ligand-independent activity of AR-

Vs results from truncation of the C-terminus resulting in a loss of the ligand-binding domain.158,170,171 Collectively, these AR-dependent or AR-positive CRPC programs drive cancer growth via AR signaling regardless of ADT interventions

(Figure1.3A).

The majority of research efforts have focused on resistance mechanisms associated with continued AR expression, but the loss of AR and adoption of AR- independent programs perpetuating cancer growth can similarly drive CRPC;151–

154 however, this mechanistic avenue is not very clear. It was proposed that the application of ADT selects for prostate cancer clones able to bypass AR signaling to develop a CRPC phenotype.172–174 Several other theories have surfaced explaining the transition to an AR-independent or AR-negative CRPC environment but few efforts have been made to understand this elusive CRPC phenotype

(Figure1.3A).

Second-generation antiandrogens and hormonal therapies have been developed for CRPC patients but they provide little benefit in AR-negative CRPC because these agents remain targeted at the AR-signaling axis (Figure1.3A).175–

177 For instance, enzalutamide is a second-generation antiandrogen functioning as a competitive AR antagonist but is also able to prevent AR nuclear translocation, binding of DNA, and recruitment of coactivators.176,177 Another agent is abiraterone, which inhibits synthesis of intratumoral androgens from steroidal hormones.157,175 While these second-generation agents have positively influenced the therapeutic landscape of AR-positive CRPC, they are not efficacious in AR-

negative disease.152,178 Moreover, these agents can induce AR loss and adoption of AR-independent programs further limiting therapeutic options as AR-positive

CRPC patients advance on the continuum of prostate cancer progression

(Figure1.3A).152

1.5.2 Molecular Mechanisms Driving AR-negative Disease

AR-negative CRPC is thought to result from ADT-mediated selection of prostate cancer cells capable of bypassing AR signaling through the activation of alternative mechanisms driving malignant growth.172–174 Several programs capable of sustaining CRPC in the absence of AR have been identified but the diversity of these programs suggests AR-negative CRPC encompasses significant phenotypic heterogeneity driven by a multitude of mechanisms.

Potentially promoting AR signaling bypass in CRPC is the cell survival protein Bcl-2 (B-cell Lymphoma 2).179,180 Bcl-2 is an anti-apoptotic protein acting as a regulator of cell survival by inhibiting the pro-apoptotic proteins Bax and Bak, which trigger mitochondrial-mediated apoptotic signals.181 Expression of Bcl-2 in previously Bcl-2-negative prostate cancer xenografts promoted significant tumor growth despite the initial castration-induced growth arrest.182 Additionally, inhibition of Bcl-2 in prostate cancer xenografts delayed the onset of castration resistance.183 Despite the identification of elevated Bcl-2 expression in CRPC experimental models and patients, this elevation is not universally required for the adoption of mechanisms bypassing AR.180

AR-independence can also be achieved after the substitution of the glucocorticoid receptor for AR in CRPC.173,184,185 Interestingly, CHIP-seq revealed significant similarity between transcriptomes driven by AR versus glucocorticoid receptors in CRPC.173,185 Moreover, in vitro and in vivo resistance to enzalutamide correlated with an upregulation of glucocorticoid receptor expression and this substitute receptor was required for continued enzalutamide-resistant growth.184

While possible, it is unclear if other steroid hormone receptors can be hijacked to bypass AR signaling in CRPC.

The cancer stem cell theory has been proposed as the reason behind ADT- mediated selection of AR-negative CRPC. The cancer stem cell theory states that a rare population of cancer cells within a tumor possess stem-like properties, like tumor initiating capabilities and are incredibly therapeutically resistant.186 In prostate cancer, cancer stem cell populations have been characterized and this body of evidence suggests that these stem cells express low AR levels or are AR- negative and therefore resistant to ADT.187,188 Moreover, the gene expression profile of metastatic CRPC more closely mimics the expression profile suggested to drive this stem cell phenotype.189 Therefore, ADT may select for these AR- negative prostate cancer stem cells, which then repopulate the tumor with a CRPC cellular population that bypasses AR.

Similar to the prostate cancer stem cell theory outlined above, the subpopulation of neuroendocrine (NE) cells may be responsible for generating an

AR-negative CRPC phenotype. NE cells are rare cells found within prostate epithelium that secrete neuropeptides associated with prostate cancer survival.190

They are identified by a number of neuroendocrine differentiation (NED) markers, such as synaptophysin, neuron-specific enolase 2 (ENO2), and chromogranin A, combined with a lack of AR expression.190 While de novo neuroendocrine prostate cancer accounts for less than 2% of prostate cancer cases, ~5 to 10% of primary prostate adenocarcinomas contain focal NED.191,192 The presence of focal NED may allow for ADT-mediated selection of these particular populations.

More commonly, NED appears following progression to CRPC after treatment of primary prostate adenocarcinoma with long-term ADT.154,193–195 In the presence of second generation therapeutics, AR-positive prostate cancer can lose

AR expression and gain NED markers.154,193–195 During this progression, evidence supports a transition of prostate cancer cells from an epithelial phenotype to a neuroendocrine phenotype, instead of mere selection of an already present AR- negative population.

With the growing use of these second-generation agents and lack of efficacious therapies for AR-negative CRPC, mechanisms driving this phenotype are of particular concern. Additional efforts are needed understand molecular pathways that drive AR bypass in difficult to treat AR-negative CRPC phenotypes.

1.5.3 Telomere Maintenance in Castration-resistant Prostate Cancer

Attempts to understand the complexity of molecular pathways driving progression of CRPC have unveiled TMM plasticity as a potential switch worth further investigation. In primary prostate cancer, telomerase is the sole mechanism employed for telomere length maintenance and androgens induce the expression

of TERT in vitro (Figure1.3B).196,197 Because of the reliance of primary disease on telomerase, CRPC was assumed similarly dependent on telomerase despite little effort to evaluate ALT hallmarks (Figure1.3B). In fact, AR mutants and splice variants driving AR-positive CRPC differentially regulate TERT expression and telomerase activity.67,198 While these findings strengthen the case for telomerase in the context of AR-positive CRPC, it remains to be determined if AR-negative programs takeover the regulation of telomerase expression or induce ALT activity at the telomere (Figure1.3B).

Interestingly, the only documented occurrence of ALT in prostate cancer occurred following progression of advanced prostate cancer to CRPC.199 This case study revealed the presence of long, heterogeneous telomere lengths in a distant metastatic site following therapeutic intervention and CRPC progression.199

Notably, a wide telomere length distribution was not noted within the primary tumor, which exhibited nearly undetectable telomeric signal typical of prostate cancer.199 This metastatic site also demonstrated a loss of ATRX, which is a common occurrence in cancers using ALT, but cannot drive ALT alone.199 This correlative but not causative relationship was similarly demonstrated in CRPC when inactivation of ATRX alone in CWR-R1 cells was unable to trigger a transition to ALT.200 Though loss of ATRX may correlate with ALT acquisition in CRPC, it is not the key driver. Together, these studies represent the only published occurrence of the elusive ALT phenotype in CRPC. Also highlighted is the potential for TMM plasticity and CRPC disease progression to occur in parallel, possibly driven by a shared molecular program.

Figure 1. 3. Telomere maintenance mechanism plasticity in prostate cancer disease progression. (A) Schematic representation of prostate cancer disease progression and acquisition of Castration-resistant prostate cancer (CRPC) phenotypes. (B) Schematic representation of defined and predicted TMMs in primary prostate cancer, AR-positive CRPC and AR-negative CRPC. Tx = treatment.

1.6 Statement of Purpose

There exists a significant knowledge void with regard to key drivers of ALT activity. This knowledge deficit is even more pronounced in the context of traditionally telomerase-positive cancer types in which ALTiness, through TMM plasticity, can be adopted. Without efforts to understand ALT acquisition, characterize the ALTiness spectrum, and identify ALT drivers, malignant TMM regulation will remain a mystery. There is a body of evidence, albeit limited, supporting the acquisition of ALTiness following prostate cancer progression to

CRPC.199 Whether these ALT hallmarks correlate with AR status remains to be determined. If an association exists between ALTiness and AR-negative CRPC, then it is possible a common molecular mechanism governs both phenotypic transitions. Unfortunately, drivers of AR loss and AR-independent programs are not well defined; similar to the knowledge void surrounding ALT. Investigations of

ALT in the context of AR-negative CRPC may unveil a molecular mechanism governing both phenotypic transitions that can aid in deciphering the role TMM plasticity plays in malignant progression. Therefore, the studies highlighted herein were designed to assess the role of the ALT regulator, SLX4IP, in promoting TMM plasticity in the context of prostate cancer progression. The purpose of these studies was to address the following; (i) is SLX4IP promoting ALTiness following progression to AR-negative CRPC and (ii) what is the mechanism by which

SLX4IP promotes ALT hallmarks in a previously telomerase-positive environment.

First, we determined that primary prostate cancer and AR-positive CRPC in vitro models exhibit telomerase activity in the absence of ALT hallmarks. However,

AR-negative CRPC demonstrated ALTiness characterized by low telomerase activity with elevated C-circles and APB abundance. Interestingly, SLX4IP expression directly correlated with APBs in AR-negative CRPC. Removal of

SLX4IP from AR-negative CRPC in vitro models triggered loss of APBs, telomere shortening and senescent phenotype, which was partially rescued following telomerase reconstitution. Modelling of AR-positive CRPC progression to AR- negative disease revealed AR loss was accompanied by an induction of SLX4IP- dependent APB formation. Without sufficient SLX4IP expression, dramatic telomere shortening led to senescence in this model of CRPC progression. In conclusion, SLX4IP-dependent APB promotion is necessary for telomere maintenance in CRPC with AR loss.

Next, we identified the region of SLX4IP capable of promoting APB formation and downstream telomere maintenance in CRPC. Here we show that the N-terminus of SLX4IP, representing the first 140 amino acids of the protein, is sufficient to promote APBs in a variety of CRPC in vitro models. Additionally, introduction of the N-terminus of SLX4IP is able to rescue the loss of APBs, telomere shortening, and induction of senescence triggered by endogenous

SLX4IP knockdown in AR-negative CRPC. Furthermore, the N-terminus is responsible for directing SLX4IP to the telomere to prime telomeric sites for APB formation and downstream ALT activity. These data identify the N-terminus of

SLX4IP responsible for telomeric localization to promote APBs necessary for telomere maintenance.

Taken together, these data highlight the critical role of SLX4IP in promoting

ALTiness in the context of disease progression to AR-negative CRPC. Additionally, these studies shed light on regulators of TMM plasticity and molecular drivers governing the replicative immortality of highly resistant AR-negative CRPC phenotypes.

CHAPTER 2: SLX4IP PROMOTES TELOMERE MAINTENANCE IN ANDROGEN RECEPTOR–INDEPENDENT CASTRATION-RESISTANT PROSTATE CANCER THROUGH ALT-LIKE TELOMERIC PML LOCALIZATION

A version of this chapter was reproduced from:

Mangosh, T. L., Awadallah, W. N., Grabowska, M. M., and Taylor, D. J. "SLX4IP

promotes telomere maintenance in androgen receptor–independent castration- resistant prostate cancer through ALT-like telomeric PML localization." Molecular

Cancer Research 19.2 (2021): 301-316.

2.1 Abstract

In advanced prostate cancer, resistance to androgen deprivation therapy is achieved through numerous mechanisms, including loss of the androgen receptor

(AR) allowing for AR-independent growth. Therapeutic options are limited for AR- independent castration-resistant prostate cancer (CRPC), and defining mechanisms critical for survival is of utmost importance for targeting this lethal disease. Our studies focus on identifying telomere maintenance mechanism

(TMM) hallmarks adopted by CRPC to promote survival. TMMs are responsible for telomere elongation to instill replicative immortality and prevent senescence, with the two TMM pathways available being telomerase and alternative lengthening of telomeres (ALT). Here, we show that AR-independent CRPC demonstrates an atypical ALT-like phenotype with variable telomerase expression and activity, whereas AR-dependent models lack discernible ALT hallmarks. Additionally, AR- independent CRPC cells exhibited elevated levels of SLX4IP, a protein implicated in promoting ALT. SLX4IP overexpression in AR-dependent C4-2B cells promoted an ALT-like phenotype and telomere maintenance. SLX4IP knockdown in AR- independent DU145 and PC-3 cells led to ALT-like hallmark reduction, telomere shortening, and induction of senescence. In PC-3 xenografts, this effect translated to reduced tumor volume. Using an in vitro model of AR-independent progression, loss of AR in AR-dependent C4-2B cells promoted an atypical ALT-like phenotype in an SLX4IP-dependent manner. Insufficient SLX4IP expression diminished ALT- like hallmarks and resulted in accelerated telomere loss and senescence. This study demonstrates a unique reliance of AR-independent CRPC on SLX4IP-

mediated ALT-like hallmarks and loss of these hallmarks induce telomere shortening and senescence, thereby impairing replicative immortality.

2.2 Introduction

Prostate cancer relies on androgen receptor (AR) signaling, and advanced disease is treated with androgen deprivation therapy despite the nearly unavoidable progression to castration-resistant prostate cancer (CRPC).150,151,201

Castration-resistance is acquired through a number of AR-dependent mechanisms such as AR mutations, amplifications, and splice variants.68 Resistance is also attained following AR loss and activation of AR-independent pathways.152,153,190,202,203 Accordingly, second-generation antiandrogens have been developed for CRPC,175,204 however, they provide little benefit in CRPC patients with AR loss.152,178 Moreover, AR-dependent CRPC treatment with second-generation agents induces AR loss.152 Because of the lack of efficacious therapeutics and poor prognosis of AR-independent CRPC, it is crucial to identify regulatory mechanisms required for AR-independent CRPC survival.

Our studies have focused on defining telomere maintenance mechanism

(TMM) phenotypes in AR-dependent and independent CRPC. Cancer cells acquire replicative immortality by activating TMMs, which allows for maintained telomere length over successive cell divisions.38,40–42 As replicative polymerases are unable to fully replicate the ends of linear DNA,7 telomeric DNA of somatic cells progressively shortens with each replication cycle until a minimum threshold limit is reached provoking senescence to prevent deleterious genomic consequences.28

Conversely, malignancies employ TMMs, through activation of telomerase or the alternative lengthening of telomeres (ALT) pathway, to evade this terminal fate and instill replicative immortality.38,40–42 Telomerase is a reverse transcriptase capable of catalyzing the addition of telomeric repeats to chromosomal ends40 and is the primary contributor to telomere elongation in 85-90% of cancers.37 ALT activation accounts for the remaining 10-15% of cancers38 and involves telomere-directed homologous recombination (HR) to promote elongation.42 The ALT mechanism remains challenging to characterize due to the inability to measure ALT activity directly, unlike the telomerase pathway.37,79 As such, the identification of ALT relies upon surrogate measures including ALT-associated Promyelocytic leukemia protein (PML) bodies (APBs) found at sites of telomeric HR,104 extra-telomeric circular C-rich DNA byproducts of ALT called C-circles,102 and significant telomere length heterogeneity.42 Historically, ALT was predominantly associated with cancers with undetectable telomerase activity38 but emerging data indicates TMM identity is not universally binary; instead, both telomerase and ALT are inherently programmed in malignancy and coexist to varying degrees, further confounding

TMM characterization.75,123 Despite the complicated mechanism and identification of ALT, recent evidence has begun to illuminate the underestimation of ALT’s contribution to telomere maintenance, including prostate cancer.

Both in vitro and in vivo, telomere elongation in primary prostate cancer has been wholly attributed to telomerase.196,197 This is not surprising as androgens in primary disease models promote expression of the telomerase reverse transcriptase component, TERT.197 Historically, CRPC regardless of AR

expression, was assumed to rely upon telomerase for telomere elongation,196 which is unsurprising due to the challenges surrounding ALT identification, presence of detectable telomerase activity, and underestimation of TMM coexistence. However, a case study comparing primary and castration-resistant disease identified ALT hallmarks in a distant metastatic site, which were lacking in the primary tumor.199 This finding represents one of the few documented occurrences of the understudied ALT phenotype in prostate cancer and, more importantly, highlights a link between TMM coexistence and disease progression.

Thus, understanding the acquisition of ALT hallmarks in prostate cancer progression is critical for understanding the purpose of TMM coexistence.

Additional examples of TMM coexistence have emerged in cases beyond prostate cancer progression underscoring the complexity of malignant TMM programming.76,124,139 Moreover, cancer cells initiate a shift from telomerase to ALT as a mechanism of resistance to telomerase inhibition highlighting the necessity of both TMM programs, though both mechanisms may not be engaged equally or simultaneously.107,126,127,133 Further supporting the potential of TMM coexistence, telomerase reconstitution or fusion of telomerase- and ALT-positive cells led to a blended TMM phenotypic presentation.128,129 Despite the identification of TMM coexistence, key regulators have yet to be well characterized. Recently, SLX4IP was implicated as a regulator of TMM plasticity. SLX4IP interacts with SLX4, a nuclease scaffold essential for ALT-mediated telomere maintenance.143,144 SLX4 shuttles its associated nucleases and SLX4IP to participate in the resolution of HR junctions at the telomere following its recruitment by TRF2, an essential telomere

end-binding protein.25,143 Knowledge regarding the function of SLX4IP beyond this interaction is limited but emerging data demonstrates SLX4IP differentially regulates TMMs based on cancer type. Specifically, SLX4IP knockout promotes

ALT hallmarks in osteosarcoma models140 but in breast cancer models SLX4IP knockdown leads to the loss of ALT hallmarks and induction of telomerase.142

Despite this disease-dependent difference, it is evident that SLX4IP is involved in

TMM plasticity though its involvement in TMM coexistence linked to prostate cancer disease progression remains unexplored.

In order to understand the acquisition of ALT hallmarks and TMM coexistence in CRPC we have characterized the TMM phenotype of five diverse

CRPC in vitro models. Surprisingly, highly variable telomerase expression and activity was noted amongst CRPC models. CRPC cell lines utilizing AR- independent modes of resistance exhibit an atypical ALT phenotype coupled with elevated SLX4IP expression, while AR-dependent models demonstrate a lack of

ALT hallmarks. Additionally, we show that overexpression of SLX4IP promotes unique ALT-like telomeric PML localization events closely mimicking traditional

APBs. In contrast, we show SLX4IP knockdown leads to loss of ALT-like PML localization events in AR-independent CRPC cell lines. Consequently, these cells undergo a notable reduction in telomere length coupled with the induction of senescence-associated markers, which was partially rescued following telomerase overexpression. The observed SLX4IP-mediated promotion of ALT-like PML localization events and telomere maintenance was further confirmed after inducing

AR loss in AR-dependent CRPC cells via androgen deprivation. Herein, we identify

a unique relationship between SLX4IP and ALT-like hallmarks that are necessary for the maintenance of telomere length and therefore replicative immortality of AR- independent CRPC in vitro.

2.3 Results

2.3.1 AR-independent CRPC in vitro models exhibit an atypical ALT-like phenotype

Due to the understudied nature of ALT’s contribution to telomere maintenance in CRPC, we sought to determine if TMM phenotype correlated with

AR dependence. Five in vitro CRPC models were interrogated for telomerase expression, activity, ALT hallmarks, and telomere length distribution. C4-2B, CWR-

R1 and 22Rv1 cell lines retain AR expression (AR-dependent)205,206 while DU145 and PC-3 cells exhibit AR loss (AR-independent).207,208 Despite being castration- sensitive, LAPC4 cells represent a characterized telomerase-positive AR- dependent model capable of acquiring ALT hallmarks200 and was included as a control. Relative mRNA expression of TERT and telomerase activity across cell lines was analyzed alongside telomerase-positive HCT116 and ALT-positive

U2OS controls (Figure 2.1A and B, Figure 2.2D). Interestingly, significant variability was noted amongst CRPC cell lines regardless of AR status.

With the exception of C4-2B cells, CRPC cell lines exhibited limited telomerase activity. Based on this result, we reasoned ALT-like processes might be engaged to assist with telomere maintenance in CRPC cells with limited telomerase activity. As ALT activity cannot be directly measured, we evaluated

surrogate measures of ALT activity, including APB abundance, C-circles, and telomere length heterogeneity.42,102,104 APBs were visualized via IF-FISH for ALT- associated PML protein localizing at the telomere, which is highly indicative of ALT- mediated HR events.104 Traditionally, APBs are instances of nearly complete overlap of PML foci with ultra-bright telomeric foci illustrated in the ALT-positive

U2OS control (Figure 2.1C and D).104 However, we identified atypical telomeric

PML localization events exhibiting only partial overlap, representing an ALT-like hallmark (Figure 2.1C and D). AR-independent DU145 and PC-3 cell lines had a modest proportion of APB-positive cells at ~7% and ~9%, respectively, compared to ALT-positive control with ~39%; however, AR-dependent cell lines had significantly fewer APB-positive cells ranging from ~0.3 to 1.8% (Figure 2.1C and

D). In addition, DU145 and PC-3 cells had a significantly greater proportion of cells with ALT-like PML localization events at ~22% and ~34%, respectively, whereas

AR-dependent cell lines exhibited these events infrequently ranging from ~1.2 to

8.7%. To complement this finding, telomeric C-circles that are byproducts of ALT- related events were evaluated via C-circle Amplification (CCA).105 A signal ratio greater than or equal to two when comparing the CCA reaction to control is highly indicative of ALT105 and only AR-independent cell lines reached this threshold, albeit at a significantly lower level than U2OS control (Figure 2.1E). Lastly, telomere length was evaluated to determine if CRPC cell lines contain long, heterogeneous telomeres, a known characteristic of traditional ALT-positive cells.42 Interestingly, DU145 and PC-3 telomeres do not mimic the telomere length distribution present in the U2OS control and only exhibit an average telomere

length of ~4.9kb and 6.0 kb, respectively (Figure 2.1F). Moreover, they do not differ dramatically from the telomere length distribution patterns of their AR-dependent counterparts.

Taken together these data indicate CRPC cell lines exhibit variable telomerase expression and activity. Those that are AR-dependent do not contain an appreciable amount of ALT-like hallmarks supporting telomerase as the primary

TMM. Conversely, in DU145 and PC-3 cells lacking AR expression, relatively low telomerase activity is accompanied by an atypical ALT-like phenotype including

ALT-like telomeric PML localization events and C-circles but lacking long, heterogeneous telomeres.

2.3.2 SLX4IP expression correlates with AR-independence and ALT-like hallmarks

ALT hallmarks, including those in CRPC,199 often correlate with loss of chromatin remodeler ATRX. While ATRX deletions are potential ALT initiating events, these genomic alterations are not universally causative of ALT hallmarks across in vitro models.88,200 Mutational analysis of ATRX209–211 and protein expression across CRPC cell lines revealed DU145 and PC-3 cells, which display an atypical ALT-like phenotype, retain wild-type ATRX expression (Figure 2.3A).

Thus, ATRX loss may be responsible for ALT initiation in some cases, but is noncontributory to the ALT-like hallmarks we observed. As such, additional ALT regulators were investigated.

The relatively uncharacterized protein, SLX4IP, was identified as a promoter of ALT-mediated telomere maintenance in breast cancer models;142 however, its role in CRPC remains ambiguous. Inspection of SLX4IP expression data from the Prostate Cancer Transcriptome Atlas (PCTA)212 revealed significantly higher expression in metastatic CRPC (mCRPC) patients, which includes both AR-dependent and independent cases, when compared to those with primary disease (Figure 2.4). Interestingly, relative SLX4IP protein expression in DU145 and PC-3 cells, which exhibit ALT-like hallmarks and AR-independence, is significantly higher than in AR-dependent cell lines that lack ALT-like hallmarks

(Figure 2.3A and B). The correlation between elevated SLX4IP expression, ALT- like hallmarks, and known role of SLX4IP in telomere maintenance supports

SLX4IP as a potential regulator of the observed atypical ALT phenotype in AR- independent CRPC in vitro.142–144

To investigate the ability of SLX4IP to promote ALT-like hallmarks in CRPC cell lines, a 3X FLAG-tagged SLX4IP construct was stably overexpressed via retroviral transduction in C4-2B (C4-2B.SLX4IP) and PC-3 cells (PC-3.SLX4IP)

(Figure 2.3C). In C4-2B.SLX4IP cells, a significant reduction in TERT expression and telomerase activity was concomitant with SLX4IP overexpression (Figure 2.3D and E, Figure 2.4). This effect was diminished in PC-3 cells that have limited telomerase expression and activity at baseline (Figure 2.3D and E, Figure 2.4). As predicted, SLX4IP overexpression coincided with a significant increase in the percent of cells with ALT-like PML localization events in both C4-2B.SLX4IP and

PC-3.SLX4IP cell populations (Figure 2.3F and G). Though modest increases in

the percent of APB-positive cells were noted with SLX4IP overexpression, the changes were deemed not statistically significant. Lastly, SLX4IP overexpression resulted in a significant elevation in C-circle signal ratio in C4-2B cells (Figure

2.3H). A slight, but statistically insignificant, elevation was noted in C-circle abundance in PC-3.SLX4IP cells. Together, these data indicate SLX4IP overexpression is capable of promoting ALT-like hallmarks to varying degrees but is most consistently correlated with elevations in ALT-like PML localization events.

These data support the role of SLX4IP in promoting ALT-like hallmarks, but it is unclear whether SLX4IP expression correlates with telomere length changes.

After defining PD time (Figure 2.4), TRF analysis revealed control C4-2B and PC-

3 cells exhibited minimal telomere length changes over 45 days (Figure 2.3I and

J). If SLX4IP-mediated ALT-like hallmarks are not promoting telomere maintenance it would be expected that, at least in the case of C4-2B.SLX4IP cells, telomeres would shorten due to the significant reduction in telomerase expression and activity observed (Figure 2.3D and E). However, slight telomere lengthening was noted in both C4-2B.SLX4IP and PC-3.SLX4IP cells, supporting the role of

SLX4IP-mediated ALT-like PML localization events in promoting telomere maintenance (Figure 2.3I and J, Figure 2.4). Despite this observation, changes in telomere length distribution were not present, consistent with the atypical ALT-like presentation described above.

Here we establish elevated SLX4IP expression promotes ALT-like PML localization events and C-circles in C4-2B cells and elevates ALT-like hallmarks from baseline in PC-3 cells. Moreover, these phenotypic changes are coupled to

telomere maintenance, suggesting SLX4IP–mediated ALT-like PML localization promotes functional telomeric elongation events.

2.3.3 SLX4IP is essential for ALT-like hallmarks and telomere maintenance in AR-independent CRPC

To identify the effects of SLX4IP loss on the atypical ALT phenotype of AR- independent CRPC in vitro, stable cell lines with SLX4IP knockdown (KD.1, KD.2) and non-targeting control (NS) derived from PC-3 cells were generated using lentiviral transduction (Figure 2.5A and B). A ~50% knockdown of SLX4IP in PC-3 cells led to a reduction in both percent of APB-positive cells and percent of cells with ALT-like telomeric PML localization events (Figure 2.5C and D), while having no significant effect on C-circle abundance (Figure 2.5E). With this specific loss of

ALT-like PML localization events, we reasoned SLX4IP knockdown may trigger a compensatory induction of telomerase to provide adequate telomere maintenance.213 However, TERT expression and telomerase activity was not induced, but marginally diminished, in PC-3.KD.1 and KD.2 cells (Figure 2.5F and

G, Figure 2.6). To solidify the role of SLX4IP-mediated ALT-like PML localization events in telomere maintenance, telomere length changes were investigated following SLX4IP knockdown.

If the observed SLX4IP-mediated atypical ALT phenotype is involved in telomere maintenance then a reduction in SLX4IP expression without compensatory telomerase induction should lead to accelerated telomere shortening.126 To test this hypothesis, telomere length changes were evaluated

over 45 days coupled with PD time determination. Calculation of PD time revealed a dramatic reduction in cell number over time in PC-3.KD.1 and KD.2 cells suggesting an impairment in proliferative capacity, a phenotype observed with impaired telomere maintenance (Figure 2.6).28,31 In addition, PC-3.KD.1 and KD.2 cells demonstrated an average telomere loss of ~700 to 800 bp over 24 PDs, whereas control lines exhibited minimal telomere length changes in the equivalent number of PDs (Figure 2.5H, Figure 2.6). This translates to more than triple the rate of telomere loss in control lines (Figure 2.5I, Figure 2.6). Because cells unable to sufficiently maintain telomere length are programmed to senesce to prevent loss of genomic information as they approach a minimum threshold length,28,31 we reasoned that accelerated telomeric shortening in PC-3.KD.1 and KD.2 would lead to premature senescence. As expected, a considerable increase in the proportion of senescence-associated β-galactosidase (SA β-gal) staining214 was observed in late passage cells with SLX4IP knockdown compared to late passage controls

(Figure 2.7A and B). Elevations in relative p21 protein expression, a senescence- associated marker,215 further corroborated the senescent phenotype (Figure 2.7C and D). Moreover, SLX4IP-mediated senescence is most likely due to inadequate telomere length maintenance, as no significant differences in β-gal positive cells

(Figure 2.7A and B) or p21 protein expression (Figure 2.7C and D) were observed in early passage PC-3.NS.1, NS.2, KD.1, and KD.2 cells.

In contrast to PC-3 cells, SLX4IP knockdown in DU145 cells revealed a compensatory increase in telomerase expression and activity coupled with a reduction in cells with ALT-like telomeric PML localization (Figure 2.8). Though

these cells have the ability to induce telomerase expression and promote functional assembly of the enzyme, a significant impairment in PD time and modest telomere shortening was observed (Figure 2.9). Moreover, SLX4IP knockdown led to a prominent increase in both β-gal positively stained cells and relative p21 protein expression in late passage DU145-derived cell lines that was not present in early passage cells (Figure 2.9). Altogether, these data demonstrate diminished

SLX4IP expression is associated with the induction of senescence in late passage

PC-3 and DU145 cells.

The inability of telomerase induction to fully rescue DU145.KD.1 and KD.2 cell lines from telomere shortening and senescent-associated markers raises the question of whether stable telomerase overexpression in PC-3.KD.1 and KD.2 cells, that do not induce telomerase, can promote sufficient telomere maintenance and prevent senescence. To answer this question, PC-3.KD.1 and KD.2 cells with stable telomerase overexpression (PC-3.KD.1+TELO, PC-3.KD.2+TELO) were generated via retroviral transduction and followed over 45 days. Following confirmation of telomerase overexpression (Figure 2.7E and F, Figure 2.9) and PD determination (Figure 2.9), TRF analysis revealed minimal telomere length changes in PC-3.KD.1+TELO cells over 32 PD versus ~600 bp loss in PC-3.KD.1 cells over 24 PDs (Fig 4G). Telomere length reductions of ~500 bp in PC-

3.KD.2+TELO cells over 32 PDs versus ~800 bp in only 30 PDs in PC-3.KD.2 cells were also noted (Fig 4G). These data translate to a deceleration in the rate of telomere loss to only ~4 and ~14 bp per PD in PC-3.KD.1+TELO and PC-

3.KD.2+TELO cells, respectively (Figure 2.7H). Telomere length preservation via

telomerase was further reflected by a significant reduction in the proportion of SA

β-gal positive cells and p21 expression in PC-3.KD.1+TELO cells and p21 expression in PC-3.KD.2+TELO cells when compared to cells with SLX4IP knockdown alone (Figure 2.7I, J, K, and L). Together these data support the senescent phenotype observed following SLX4IP knockdown is very likely through a telomere-based mechanism.

Loss of SLX4IP-mediated ALT-like PML localization events revealed senescence-associated markers in both PC-3 and DU145 cells demonstrating AR- independence. Remarkably, the compensatory induction of telomerase observed in DU145 cells was unable to completely prevent telomere shortening and senescence. This effect was recapitulated following telomerase reconstitution in

PC-3 cells with SLX4IP knockdown. These data support ALT-like PML localization is necessary for adequate telomere maintenance in AR-independent CRPC in vitro models. Additionally, these data identify SLX4IP as an essential regulator of the atypical ALT phenotype in this AR-independent environment.

2.3.4 SLX4IP-mediated ALT-like hallmark depletion and induction of senescence is maintained in vivo

To ascertain if loss of ALT-like PML localization events via SLX4IP knockdown is maintained in vivo, male nude mice were inoculated subcutaneously with PC-3.NS.1 or PC-3.KD.1 cells and tumor volumes subsequently monitored over 30 days. Immunohistochemistry revealed SLX4IP knockdown led to a significant increase in the percent of p21-positively stained cells (Figure 2.10A and

B). Consequently, this effect translated to a significant reduction in average tumor volume at time of sacrifice (Figure 2.10C). As no considerable difference in Ki-67 or active caspase-3 staining was observed between groups, reductions in tumor volume are likely due to the senescent phenotype associated with SLX4IP knockdown (Figure 2.10A). Indeed, SLX4IP knockdown resulted in a reduction of

ALT-like PML localization events (Figure 2.10D and E). Analogous to our in vitro studies, evaluation of telomerase activity among xenografts did not reveal a compensatory increase in telomerase activity following SLX4IP knockdown (Figure

2.10F and G). Together, these in vivo results further support SLX4IP as a mediator of ALT-like PML localization in AR-independent CRPC.

2.3.5 Androgen-deprived conditions promote ALT-like PML localization in an SLX4IP-dependent manner

Androgen deprivation therapy triggers AR-independence in patients.151,193,216,217 Similarly, AR-dependent cell lines undergo AR loss when placed in growth media lacking androgens.218–220 Because we identified a unique relationship between ALT-like hallmarks and AR-independent CRPC, we reasoned

AR-dependent CRPC cells lacking ALT-like hallmarks grown in androgen-deprived conditions may limit telomerase and gain ALT-like hallmarks as they transition to

AR-independence. Moreover, this promotion of ALT-like hallmarks, specifically

ALT-like PML localization, is expected to be reliant on SLX4IP. Therefore, C4-2B cells with stable knockdown of SLX4IP (KD.1, KD.2) and non-targeting control (NS)

were generated (Figure 2.11A and B) and placed in charcoal-stripped media to mimic androgen deprivation and induce AR loss.218–221

Analysis of SLX4IP expression revealed a modest yet significant increase in SLX4IP expression when comparing C4-2B.NS cells cultured in charcoal- stripped media (+CSS), hypothesized to adopt an ALT-like phenotype, to their normal growth media (fetal bovine serum, +FBS) counterparts (Figure 2.11A and

B). Reductions in AR expression and upregulation of ENO2 expression, markers of AR-independence, across groups confirmed the transition was initiated under androgen-deprived conditions (Figure 2.11A, Figure 2.12).221,222 Notably, a difference in AR and ENO2 expression was not observed between cell lines grown in +CSS indicating SLX4IP itself is not responsible for inducing AR-independence.

As expected, relatively high telomerase expression and activity was noted in cells grown in +FBS, however both were significantly reduced with +CSS (Figure

2.11C and D, Figure 2.12). Moreover, few APB-positive cells and ALT-like PML localization events were identified in C4-2B.NS, KD.1, and KD.2 cells grown in

+FBS. Following +CSS growth, C4-2B.NS cells exhibited a significant increase in the percent of cells with ALT-like PML localization events but not APBs (Figure

2.11E and F). This significant increase was diminished with SLX4IP knockdown.

Together these data support C4-2B cells shift from high telomerase expression and activity to low telomerase expression and activity with ALT-like PML localization events following androgen deprivation. Without sufficient SLX4IP expression, this phenotypic shift is blunted.

Reductions in telomerase with a blunted atypical ALT phenotype makes telomere shortening an expected fate. Following PD time determination (Figure

2.12), cell lines grown in each condition were followed for 45 days to characterize telomere length changes. As expected, C4-2B cells grown in +FBS regardless of

SLX4IP expression exhibited minimal telomere length changes consistent with their telomerase-positive status (Figure 2.12). C4-2B-derived cell lines grown in

+CSS demonstrated telomere shortening; however, this phenotype was markedly amplified in the case of SLX4IP knockdown (Fig 6G, Figure 2.12). Specifically, C4-

2B.KD.1 and KD.2 cells exhibited a dramatic loss of ~1 800 and ~2 500 bp in 45

PDs compared to the ~1 400 bp loss in 54 PDs in C4-2B.NS cells. Insufficient telomere maintenance is unsurprising because C4-2B-derived cell lines demonstrate limited telomerase expression and activity following growth in +CSS, regardless of SLX4IP expression. However, the accelerated loss of telomere length in C4-2B.KD.1 and KD.2 cells supports the role of ALT-like PML localization events in promoting some degree of telomere maintenance in the C4-2B.NS cells following androgen deprivation, albeit not to the same degree as telomerase in

+FBS.

After identification of telomere shortening, senescent-associated markers were evaluated. Early passage comparisons revealed SLX4IP knockdown and growth in +CSS independently triggers immediate β-gal positivity (Fig 7A and B) and elevations in relative p21 protein expression (Fig 7C and D) to some degree.

This pattern remained consistent from early to late passage cells with the exception of C4-2B.KD.1 and KD.2 cells grown in +CSS that exhibit dramatic

telomere shortening. β-gal staining of these late passage C4-2B.KD.1 and KD.2 cells revealed a significant increase in β-gal positive cells compared to C4-2B.NS cells grown in +CSS and compared to C4-2B.KD.1 and KD.2 cells grown in +FBS

(Figure 2.13A and B). Consequently, these late passage β-gal positive populations can likely be attributed to the impaired telomere maintenance highlighted above.

Corroborating this pattern, relative p21 protein expression mimicked our previous finding and was significantly elevated in late passage SLX4IP knockdown cells cultured in +CSS (Figure 2.13C and D). Therefore, with androgen deprivation, AR- dependent CRPC limits telomerase expression and activity and gains ALT-like

PML localization events in an SLX4IP-dependent manner. Without sufficient

SLX4IP expression and promotion of these ALT-like hallmarks, telomere shortening and senescence is induced in this in vitro model of disease progression following androgen deprivation.

2.4 Discussion

Presently, there exists a lack of targeted therapeutic options for patients with AR-independent CRPC, which encompasses a group of heterogeneous diseases including therapeutically-acquired neuroendocrine prostate cancer152 and double-negative prostate cancer.203 Here, we report that SLX4IP-dependent

ALT-like PML localization events coexist with telomerase to promote telomere maintenance and perpetuate replication in in vitro models of this highly resistant

CRPC cohort. Whether these ALT-like hallmarks are universally required for telomere maintenance across all models of AR-independence and whether this

phenotype emerges as a consequence of androgen deprivation in patient tissues remains to be evaluated. Several in vitro models undergo AR-independent reprogramming following acquisition of resistance to second-generation antiandrogens indicated for CRPC treatment.223 Investigation of SLX4IP-mediated telomere maintenance in these models of therapy-acquired AR-independence would elucidate if this is consistent across AR-independent CRPC heterogeneity, complementing our findings.

The inability of AR-independent CRPC to overcome even a ~50% loss of

SLX4IP expression, and subsequent senescence, provides an elegant therapeutic target, but the knowledge void around SLX4IP-mediated ALT-like hallmarks must first be addressed. Our studies demonstrated a strong relationship between

SLX4IP, ALT-like PML localization events and telomere maintenance, yet it is unclear as to why other ALT hallmarks remain unchanged. The incomplete ALT phenotype observed in our studies may be due to the retention of functional

ATRX.88 ATRX alterations have been identified in a small subset of metastatic

CRPC cases, including the CRPC case demonstrating ALT hallmarks, but these mutations are rare.165,199,209,210 Elevations in SLX4IP and ATRX loss may be required for CRPC to recapitulate a traditional ALT phenotype. Coincidentally, several studies have demonstrated that ATRX loss is unable to initiate the traditional ALT phenotype; instead, an incomplete atypical ALT phenotypic change has been observed, similar to our findings.112,140,224 For example, ATRX knockout in mouse embryonic cells in vivo did not cause an increase in telomere length or heterogeneity224 and ATRX knockout in telomerase-positive glioma cell lines did

not increase telomeric C-circle abundance.112 Contrary to this hypothesis, investigation of osteosarcoma tumors revealed limited telomerase activity and significant C-circle abundance despite retention of ATRX expression,140 supporting the possibility of an ALT-like phenotype on a background of functional ATRX.

Though nontraditional, the ALT-like phenotype described in our studies linked to

SLX4IP expression warrants further characterization to determine if ALT-like processes are driving this unique presentation and if elevations in SLX4IP expression are stepping-stones to bona fide ALT.

The coexistence of an atypical ALT-like phenotype on a background of variable telomerase activity described in our studies challenges the traditional dichotomous relationship of ALT and telomerase. Several groups have demonstrated androgens transcriptionally regulate TERT in primary prostate cancer.67,197,198 Following progression to CRPC, AR mutants and splice variants are capable of differentially regulating telomerase expression.67 Furthermore, these AR mutants and splice variants coexist in vitro and in vivo making telomerase expression the collective output of AR perturbations present.68 Despite this complexity, AR-mediated regulation of telomerase supports our data demonstrating the reliance of AR-independent CRPC in vitro models on SLX4IP- mediated ALT-like hallmarks. It is possible that either AR suppresses SLX4IP transcriptionally or TERT, shown to have non-canonical extra-telomeric functions, interferes with SLX4IP.225 Following progression to AR-independent CRPC, we have shown SLX4IP is free to promote TMM coexistence and ALT-like hallmarks to compensate for limited telomerase activity. Interestingly, late passage AR-

independent DU145 cells demonstrated rebound telomerase expression and activity following ALT-like hallmark reduction despite maintained AR loss whereas

PC-3 cells did not. This compensatory effect was described in other malignant investigations,142 but why this varies between two AR-independent CRPC cell lines remains undetermined. Nonetheless, the observed induction of telomerase was incapable of completely filling the TMM deficiency in the timeframe studied and both cell lines underwent telomere shortening and senescence. Because of this attempted resistance to SLX4IP-mediated telomere shortening, defining mechanisms regulating both SLX4IP and compensatory telomerase in AR- independent CRPC will be crucial for describing the purpose of TMM coexistence associated with disease progression.

In summary, AR-independent CRPC exhibits a unique ALT-like phenotype with variable telomerase activity while CRPC retaining AR lacks ALT hallmarks, relying on telomerase in vitro. ALT-like hallmarks correlate with elevated SLX4IP expression and introduction of exogenous SLX4IP promotes ALT-like PML localization events and telomere maintenance. Additionally, loss of SLX4IP in AR- independent CRPC renders an environment lacking ALT-like PML localization events triggering telomere shortening and senescence. This phenotype was recapitulated in vivo resulting in reduced tumor volume. Triggering AR- independent progression, AR-dependent CRPC cells grown in androgen-deprived conditions limit telomerase and exhibit ALT-like PML localization events in an

SLX4IP-dependent manner. These ALT-like hallmarks promoted a greater degree of telomere length preservation compared to those cell lines lacking sufficient

SLX4IP expression. Together, these data reveal SLX4IP-mediated telomere maintenance as a unique mechanism necessary for the replicative immortality of

AR-independent CRPC in vitro.

2.5 Methods

Cell Culture: Cell lines were obtained from ATCC and maintained at 37°C in 5%

CO2. U2OS, HCT116, WI-38 VA13, and HEK293T cells were cultured in DMEM

(Gibco) with 10% (v/v) FBS (Sigma), 1% of Penicillin 10 000 units/mL,

Streptomycin 10 000 μg/mL, and 25 μg/mL Amphotericin B mixture (Antibiotic-

Antimycotic, Gibco). C4-2B, CWR-R1, 22Rv1, DU145, and PC-3 cells were cultured in RPMI1640 (Gibco) with 10% FBS and 1% Antibiotic-Antimycotic.

LAPC4 cells were cultured in IMDM (Sigma) with 10% (v/v) FBS (Sigma) and 1 nM dihydrotestosterone (Sigma). For androgen-deprived conditions, C4-2B cells were cultured in RPMI1640, No Phenol Red (Gibco) with 10% Charcoal Stripped FBS

(Sigma) and 1% Antibiotic-Antimycotic. Parental and genetically modified cell lines were subjected to Mycoplasma testing via MycoAlert™ PLUS Detection Kit (July

2018, Lonza) and STR validation (April 2019, CWRU Genomics Core). To determine PD times, cells were plated and trypsinized at 72 and 120 hours. Using trypan blue exclusion, viable cell number was determined using the Countess II

(ThermoFisher). All experiments including designated early passage time points were completed with cells from population doubling (PD) 2 through 8. Late passage time points were completed with cells from the latest PD included in each telomere length experiment or after.

Stable Cell Line Generation: To generate cell lines stably overexpressing 3X

FLAG-tagged-SLX4IP, a gBlock® was designed to incorporate a 3X FLAG tag.

The 3X FLAG-SLX4IP DNA sequence (NCBI Nucleotide, NM_001009608.3) was flanked with a SalI endonuclease restriction site on the C-terminus and a BamHI restriction site on the N-terminus. The SLX4IP open reading frame was engineered for bacterial codon optimization using the gBlock® Gene Fragment design tool

(IDT). The gBlock® was cloned into pBABE-puro, a gift from Hartmut Land, Jay

Morgenstern, and Bob Weinberg (Addgene#1764).226 Retrovirus was produced using Lipofectamine™ 2000 (ThermoFisher) to co-transfect pBABE-puro or pBABE-puro.SLX4IP.3XFLAG with pCMV-VSV-G (at a ratio of 6:1, Clontech) into

GP2-293T cells (Clontech). Virus was harvested and cleared via centrifugation and/or sterile filtration at 48 hours post-transfection and infections were carried out in the presence of 10 μg/mL of polybrene (Sigma) followed by selection in 1 μg/mL

(C4-2B) or 1.5 μg/mL (DU145, PC-3) of puromycin (Sigma). To generate cell lines stably overexpressing telomerase, an identical strategy was used as outlined above using pBABE-puro.UThTERT+U3-hTR-500, a gift from Kathleen Collins

(Addgene#27665).227

To generate cell lines with stable knockdown of SLX4IP, two short hairpin

RNAs (shRNA) targeting SLX4IP in the pLKO.1-puro lentiviral expression plasmid were purchased from Sigma (Clone ID NM_001009608.1-426s1c1 and

NM_001009608.1-247s1c1). Lentivirus was produced using Lipofectamine™ 2000

(ThermoFisher) to co-transfect HEK293T cells with pLKO.1-puro Non-Mammalian

shRNA Control (Sigma#SCH002), pLKO.1-puro.shSLX4IP.1, or pLKO.1- puro.shSLX4IP.2 with pMD2.G, pRRE, and pRSV-Rev, which were gifts from

Didier Trono (at a ratio of 4:1:1:1, Addgene#12259, #12251, #12253).228 Viral infections and selection were carried out as described above.

Mouse Xenograft Studies and Immunohistochemistry: In vivo experiments were performed with approval from the Institutional Animal Care and Use

Committee at Case Western Reserve University, which is certified by the American

Association of Accreditation for Laboratory Animal Care. Six-week-old male nude mice were inoculated with a 1:1 mix of Matrigel (Gibco 354234) and 1x106 cells subcutaneously. Mouse weights and tumor volumes were determined three times weekly. Tumor volumes were calculated using the hemiellipsoid volume formula following Vernier caliper measurements of tumor diameters in three dimensions.

At 30 days post-inoculation, tumors were excised and processed for immunohistochemistry and telomerase activity. Immunohistochemistry and hematoxylin and eosin staining was carried out as published previously.229 Primary antibody dilutions: SLX4IP (1:200, Sigma, HPA046372), p21 (1:500, Cell

Signaling, 2947), Ki-67 (1:1 000, Sigma, HPA001164), and active caspase-3 (1:1

000, Millipore, AB3623). Three fields of view for each slide were imaged using an

Olympus BX43 Upright Microscope and cells were manually counted and scored using ImageJ Software.230

RNA Isolation and Analysis: RNA was isolated using TRIzol reagent (Invitrogen), reverse transcribed using the High-Capacity RNA-to-cDNA Kit (ThermoFisher), and quantified via Quantitative real time PCR (RT-qPCR) on an Applied

Biosystems™ StepOnePlus™ real time PCR system as described previously.231

Established primer pairs for TERT and for GAPDH were used.231

Western Blotting Analysis: Western blotting analysis was completed as previously described.231 Briefly, 50 μg of lysate was resolved on a 4-20% Mini-

PROTEAN® TGX™ Gel (BioRad) and transferred to Nitrocellulose membrane

(Millipore) using the Trans-Blot® Turbo™ (BioRad). Blots were blocked in 5% nonfat milk diluted in TBST and incubated overnight at 4°C with primary antibody.

Blots were incubated for one hour at room temperature with either HRP-conjugated

(Santa Cruz), IRDye 800CW or IRDye 680RD (LI-COR) secondary antibodies diluted to 1:10 000. Development with ProtoGlow ECL reagent (National

Diagnostics) was used when necessary. Blots were imaged on an Odyssey® Fc

Imaging System (LI-COR), quantified using Image Studio™ 5.2 Software, and normalized to GAPDH. Primary antibody dilutions: 1:20 000 GAPDH (Cell

Signaling, 2118), 1:5 000 SLX4IP (Sigma, HPA046372), 1:5 000 FLAG (Sigma,

F1804), 1:10 000 AR (Santa Cruz, sc-816), 1:3 000 ENO2 (Cell Signaling, 9536),

1:5 000 p21 (Cell Signaling, 2947), 1:2 000 ATRX (Cell Signaling, 14820).

Telomere Repeat Amplification Protocol: The real-time Q-TRAP assay was performed as previously published.232 Tumor lysates were prepared following

homogenization with Kontes® PELLET PESTLE® Grinders (Kimble) under similar lysis conditions. Reactions were performed on an Applied Biosystems™

StepOnePlus™ real time PCR system. Collected Ct values were then converted to relative telomerase activity units using the telomerase-positive standard curve generated as previously described. Technique validation experiments were carried out with telomerase-positive HEK293T and HCT116 cells and telomerase-negative

U2OS and WI-38 VA13 cells with a standard curve generated using HCT116 cells

(Figure 2.2A and B). Additional negative controls were generated by incubating telomerase-positive samples with 1 μg of RNase at 37°C for 20 minutes.

Immunofluorescence-Fluorescence in situ Hybridization Analysis: A previously published immunofluorescence-fluorescence in situ hybridization (IF-

FISH) protocol was carried out on cells seeded on sterile glass coverslips and tumor sections.231 Coverslips were incubated with PML antibody (Santa Cruz, sc-

5621) diluted 1:200 for one hour at room temperature. Alexa-488® conjugated anti- rabbit secondary antibody (Jackson) was diluted 1:100 and added to coverslips for overnight incubation at 4°C. Hybridization with 333ng/mL of a TelC-Cy5-labeled peptide nucleic acid (PNA) oligonucleotide telomere probe (N-

CCTAACCTAACCTAA-C, PNA BIO) in PNA buffer (10mM Tris pH 7.5, 70% formamide) was carried out at 95°C for 5 min followed by overnight incubation at room temperature. Coverslips were stained with 2 μg/mL of DAPI for 10 min and mounted with Fluoromount-G (ThermoFisher). Stained slides were imaged following blinding using the Leica HyVolution SP8 gated STED Microscope and

LAS X Software. Cell count and regions of interest (ROIs) representing potential colocalization events were identified in ImageJ.230 These ROIs were analyzed via

Coloc 2 using the Manders’ Correlation algorithm,233 where a correlation coefficient of >0.25 was further analyzed. This coefficient represents the total PML signal relative to the intersection of the PML and telomeric signal in a given ROI.

Following this analysis, ROIs were categorized as either true APBs (coefficients of

>0.85) versus ALT-like PML localization events (coefficients of 0.25-0.85). Each cell was categorized based on highest ROI correlation coefficient found within the cell. Colocalization events within U2OS control cells were included for comparison.

C-circle Isolation and Analysis: A previously published C-circle dot blot assay was performed with minor modifications.105 Genomic DNA was harvested using the Genelute™ Mammalian Genomic DNA Miniprep Kit (Sigma) according to manufacturer’s instructions. 160 ng of DNA was incubated under conditions described in the previously published protocol. Reactions were blotted onto

Hybond® membranes (ThermoFisher) and UV cross-linked (Spectrolinker™ XL-

1000, Spectronics). Membranes were hybridized with 30 μg/mL of Digoxigenin

(DIG) conjugated telomere probe (CCCTAACCCTAACCCTAACCCTAA-DIG, IDT) in DIG Easy Hyb buffer (Roche) overnight at 37°C. 0.5X saline-sodium citrate

(SSC) buffer with 0.1% SDS was used to remove excess probe. Membranes were blocked in DIG Blocking Buffer (Roche) for 30 min followed by incubation with 1:3

000 dilution of Digoxigenin-AP, Fab fragments (Roche) for 3 hours at room temperature. DIG Wash buffer (Roche) was used to remove excess antibody and

amplified C-circles were detected with CDP-Star® kit (Roche). Membranes were imaged with the Odyssey® Fc Imager (LI-COR) and signal intensity ratios calculated as previously described using Image Studio 5.2 Software. Technique validation experiments were carried out with 160 ng of DNA from telomerase- positive HEK293T, HCT116, and A549 cells and a standard curve of DNA from

ALT-positive U2OS and WI-38 VA13 cells (Figure 2.2C).

Telomere Restriction Fragment Analysis: Genomic DNA was harvested using the GenElute™ Mammalian Genomic DNA Miniprep Kit (Sigma) according to manufacturer’s instructions and telomere restriction fragment (TRF) analysis was carried out as previously published.231 Membranes were hybridized as described in the above C-circle analysis protocol at 42°C. 2x saline-sodium citrate (SSC) buffer with 0.1% SDS was used to remove excess probe. Membranes were blocked in DIG Blocking Buffer (Roche) for 30 min followed by incubation with 1:10

000 dilution of Digoxigenin-AP, Fab fragments (Roche) for 3 hours at room temperature. Additional membrane processing and imaging steps were carried out as described above in the C-circle analysis protocol.

Senescence-associated β-galactosidase Activity Staining: Cells were seeded on sterile glass coverslips and grown to 80% confluency. Senescence β- galactosidase staining kit was used following the manufacturer’s protocol (Cell

Signaling Technologies). Slides were imaged following blinding using an Olympus

BX43 Upright Microscope and cells were manually counted and scored using

ImageJ Software.230

Statistical Analysis: Statistics were performed using two-tailed Student’s t test

(in vitro and in vivo data comparing two groups) or one-way ANOVA with multiple comparisons (in vitro data comparing at least three groups) in GraphPad Prism 8 software. Patient expression data was analyzed through the PCTA based on disease course using Rank sums-test between mCRPC and primary subsets

(www.thepcta.org).212 p values less than 0.05 were considered statistically significant. All in vitro data are represented as mean values from three independent experiments performed in triplicate unless otherwise noted.

2.6 Acknowledgements

The authors would like to thank all members of the Taylor and Grabowska labs for helpful comments and suggestions related to this investigation. The

Grabowska lab is supported by start-up funds (MMG) provided by the Case

Research Institute, a joint venture between University Hospitals and Case Western

Reserve University. The Taylor Lab is supported by the NIH (R01 GM133841 and

R01 CA240993 to DJT). We would also like to acknowledge the Molecular

Therapeutics Training Program (T32 GM008056 to TLM).

Figure 2. 1: Identification of atypical ALT phenotype in AR-independent CRPC in vitro. (A) Relative TERT mRNA expression and (B) telomerase activity across CRPC cell lines. C4-2B, CWR-R1 and 22Rv1 retain AR expression (red) and DU145 and PC-3 exhibit AR loss (blue). ALT-positive U2OS (purple), telomerase-positive HCT116 (green), and telomerase-positive LAPC4 cells retaining AR expression (grey) were included controls. All groups compared to U2OS control. (C) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) at telomeres (red) with zoom insets provided. Scale bar: 5 μm. (D) Quantification of (C) for percent positive cells with at least one APB. Cells lacking APBs were quantified for ALT-like PML localization events. Representative foci are included defining these events. All groups compared to HCT116 control. (E) Representative dot blot demonstrating the presence of telomeric C-circles. Φ+ lane indicates reactions with Φ29 polymerase and Φ- lane indicates control reaction. Quantification of C-circle abundance defined as the signal ratio of Φ+ to Φ- reaction is shown below. Dotted line at y=2 indicates threshold suggesting ALT activity. All groups compared to HCT116 control. (F) TRF analysis demonstrating average telomere length across cell lines. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 2: Validation of TMM characterization techniques. (A) Standard curve of telomerase-positive HCT116 cells using Q-TRAP protocol. ALT-positive U2OS and WI-38 VA-13 cells included as negative controls. (B) Relative telomerase activity of telomerase-positive HEK293T and HCT116 cells and ALT-positive U2OS and WI-38 VA-13 cells calculated from (A). (C) Representative dot blots demonstrating the presence of telomeric C-circles. Φ+ lane indicates reactions with Φ29 polymerase and Φ- lane indicates control reaction. Standard curve of ALT-positive U2OS and WI-38 VA-13 DNA versus signal ratio of Φ+ to Φ- reaction is shown below. Telomerase-positive HEK293T, HCT116, and A549 cells included as negative controls (DNA at 160ng). (D) Representative standard curve of telomerase-positive HCT116 cells using Q- TRAP protocol for Figure 2.1B. ALT-positive U2OS and RNase-treated HCT116 sample included as negative controls. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 3: SLX4IP promotes ALT-like PML localization events in CRPC in vitro. (A) ATRX and SLX4IP protein expression across CRPC cell lines. (B) Quantification of SLX4IP expression in (A). (C) Confirmation of stable SLX4IP overexpression in C4-2B and PC-3 cells. EV: Empty Vector. (D) Relative TERT mRNA expression and (E) telomerase activity following stable SLX4IP overexpression in C4-2B and PC-3 cells. (F) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) at telomeres (red) with zoom insets provided. Scale bar: 5 μm. (G) Quantification of (F) for percent positive cells with at least one APB. Cells lacking typical APBs were quantified for ALT-like PML localization events. Representative foci are included defining these events. (H) Representative dot blot demonstrating the presence of telomeric C-circles in C4-2B and PC-3 cells with SLX4IP overexpression. Φ+ lane indicates reactions with Φ29 polymerase and Φ- lane indicates control reaction. Telomerase-positive HCT116 (−) and ALT-positive WI-38 VA-13 (+) cells included as controls. Quantification of C-circle abundance defined as the signal ratio of Φ+ to Φ- reaction is shown below. Dotted line at y=2 indicates threshold signal ratio suggesting ALT activity. (I) TRF analysis demonstrating telomere length changes over 45 days in C4-2B and (J) PC-3 cells overexpressing SLX4IP versus control. PD: Calculated population doublings. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 4: Expression of SLX4IP in mCRPC versus primary disease. (A) SLX4IP expression data from the PCTA arranged based upon patient diagnosis of benign disease, primary disease and Gleason score, or mCRPC. Ranksums-test was used to compare primary (GS<7,=7,>7) versus mCRPC cohorts. (B) Representative standard curve of telomerase-positive HCT116 cells using Q-TRAP protocol for Figure 2.3E. ALT-positive U2OS sample included as negative control. (C) Using trypan blue exclusion, relative cell number was followed over 120 hours for C4-2B and (D) PC-3 cells with SLX4IP overexpression. (E) Telomere length changes over time measured via PDs from Figure 2.3I and J. Simple linear regression models are depicted by dotted lines and slopes reported as bp/PD. Hexagon data points represent 45 day time points. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 5: SLX4IP knockdown is accompanied by disappearance of ALT- like PML localization events and accelerated telomere shortening. (A) Confirmation of stable SLX4IP knockdown using two shRNAs (KD.1, KD.2) in PC-3 cells with scrambled shRNA control (NS) at the protein level. (B) Quantification of (A). (C) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) at telomeres (red) with zoom insets provided. Scale bar: 5 μm. (D) Quantification of (C) for percent positive cells with at least one APB. Cells lacking typical APBs were quantified for ALT-like PML localization events. Representative foci are included defining these events. (E) Representative dot blot demonstrating the presence of telomeric C-circles. Φ+ lane indicates reactions with Φ29 polymerase and Φ- lane indicates control reaction. Telomerase-positive HCT116 (−) and ALT-positive U2OS (+) cells included as controls. Quantification of C-circle abundance defined as the signal ratio of Φ+ to Φ- reaction is shown below. Dotted line at y=2 indicates threshold signal ratio suggesting ALT activity. (F) Relative TERT mRNA expression and (G) telomerase activity following stable SLX4IP knockdown in PC-3 cells. (H) TRF analysis demonstrating telomere length changes over 45 days in PC-3 cells with SLX4IP knockdown (KD.1) versus control (NS.1). PD: Calculated population doublings. (I) Telomere length changes over time measured via PDs from (H). Simple linear regression models are depicted by dotted lines and slopes reported as bp/PD. Hexagon data points represent 45 day time points. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 6: SLX4IP knockdown correlates with accelerated telomere shortening. (A) Representative standard curve of telomerase-positive HCT116 cells using Q- TRAP protocol for Figure 2.5G. ALT-positive U2OS sample included as negative control. (B) Using trypan blue exclusion, relative cell number was followed over 120 hours in PC-3.NS.1 versus KD.1 or (C) NS.2 versus KD.2 cells. (D) Telomere restriction fragment analysis demonstrating telomere length changes over 45 days in PC-3 cells with SLX4IP knockdown (KD.2) versus control (NS.2). PD: Calculated population doublings. (E) Telomere length changes over time measured via PDs from (D). Simple linear regression models are depicted by dotted lines and slopes reported as bp/PD. Hexagon data points represent 45 day time points. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 7: SLX4IP knockdown triggers accumulation of senescence- associated markers. (A) Representative bright field images demonstrating SA β-gal staining (arrowheads) for senescence in early and late-passage PC-3 cells with stable knockdown of SLX4IP. Scale bar: 20 μm. (B) Quantification of relative proportion of SA β-gal positive cells in (A). (C) p21 expression following knockdown of SLX4IP in early and late passage PC-3 cells. (D) Quantification of relative p21 expression in (C). (E) Confirmation of stable overexpression of telomerase in PC-3.KD.1 and KD.2 with SLX4IP knockdown via relative TERT mRNA expression and (F) telomerase activity. (G) TRF analysis demonstrating telomere length changes over 45 days in PC-3 cells with SLX4IP knockdown (KD.1, KD.2) and telomerase overexpression (+TELO). PD: Calculated population doublings. (H) Telomere length changes over time measured via PDs from (G). Simple linear regression models are depicted by dotted lines and slopes reported as bp/PD. Hexagon data points represent 45 day time points. (I) Representative bright field images demonstrating SA β-gal staining (arrowheads) for senescence in late-passage PC- 3.KD.1 and KD.2 cells with telomerase overexpression. Scale bar: 20 μm. (J) Quantification of relative proportion of SA β-gal positive cells in (I) compared to controls. (K) p21 expression in late passage PC-3.KD.1 and KD.2 cells with telomerase overexpression. (L) Quantification of relative p21 expression in (K) compared to controls. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 8: SLX4IP knockdown is accompanied by a blunted atypical ALT phenotype. (A) Confirmation of stable SLX4IP knockdown using two shRNAs (KD.1, KD.2) in DU145 cells with scrambled shRNA control (NS) at the protein level. (B) Quantification of (A). (C) Relative TERT mRNA expression and (D) telomerase activity following SLX4IP knockdown in DU145 cells. (E) Representative standard curve of telomerase-positive HCT116 cells using Q-TRAP protocol for (D). ALT- positive U2OS sample included as negative control. (F) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) at telomeres (red) with zoom insets provided. Scale bar: 5 μm. (G) Quantification of (F) for percent positive cells with at least one APB. Cells lacking typical APBs were quantified for ALT-like PML localization events. Representative foci are included defining these events. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 9: SLX4IP knockdown promotes telomere shortening and senescence. (A) Using trypan blue exclusion, relative cell number was followed over 120 hours in DU145.NS cells versus DU145.KD.1 and KD.2 cells. (B) TRF analysis demonstrating telomere length changes over 45 days in DU145 cells with SLX4IP knockdown (KD.1, KD.2) versus control (NS). PD: Calculated population doublings. (C) Telomere length changes over time measured via PDs from B. Simple linear regression models are depicted by dotted lines and slopes reported as bp/PD. Hexagon data points represent 45 day time points. (D) Representative bright field images demonstrating SA β-gal staining (arrowheads) for senescence in early and late-passage DU145 cells with stable knockdown of SLX4IP. Scale bar: 20 μm. (E) Quantification of relative proportion of SA β-gal positive cells in (D). (F) p21 expression in early and late passage DU145 cells with stable knockdown of SLX4IP. (G) Quantification of relative p21 expression in (F). (H) Representative standard curve of telomerase-positive HCT116 cells using Q-TRAP protocol for Figure 2.7F. ALT-positive U2OS sample included as negative control. (I) Using trypan blue exclusion, relative cell number was followed over 120 hours in PC- 3.KD.1 and KD.2 cells with telomerase overexpression (+TELO). Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 10: SLX4IP knockdown in AR-independent CRPC leads to ALT-like hallmark loss, reduced tumor volume, and induction of senescent- associated markers in vivo. (A) Representative immunohistochemistry 40X images demonstrating SLX4IP, p21, Ki-67, active caspase-3, and H&E staining from PC-3.NS.1 and KD.1 xenograft sections. Scale bar: 60 μm. (B) Quantification of p21 staining in (A) for percent p21-positive cells. (C) Average tumor volume of PC-3.NS.1 (n=12) and KD.1 (n=10) on day 30. (D) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) at telomeres (red) with zoom insets provided. Scale bar: 5 μm. (E) Quantification of (D) for percent positive cells with at least one APB. Cells lacking typical APBs were quantified for ALT-like PML localization events. Representative foci are included defining these events. (F) Relative telomerase activity following SLX4IP knockdown in PC-3 xenografts. (G) Representative standard curve of telomerase-positive C4-2B xenograft using Q- TRAP protocol for (F). RNase-treated C4-2B sample included as negative control. Data represented as mean+SD; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 11: Androgen deprivation triggers SLX4IP-dependent ALT-like PML localization events. (A) Confirmation of stable SLX4IP knockdown using two shRNAs (KD.1, KD.2) in C4-2B cells with scrambled shRNA control (NS) and confirmation of AR- independent transition via AR and ENO2 expression following growth in androgen deprived conditions (+CSS) versus normal growth media (+FBS). (B) Quantification of SLX4IP expression in (A). (C) Relative TERT mRNA expression and (D) telomerase activity following SLX4IP knockdown in C4-2B cells grown in +FBS or +CSS. (E) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) at telomeres (red) with zoom insets provided. Scale bar: 5 μm. (F) Quantification of (E) for percent positive cells with at least one APB. Cells lacking typical APBs were quantified for ALT-like PML localization events. Representative foci are included defining these events. (G) TRF analysis demonstrating telomere length changes in C42B cells with SLX4IP knockdown (KD.1, KD.2) grown in +CSS versus C4-2B.NS grown in +CSS or +FBS. PD: Calculated population doublings. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 12: Androgen-independent marker expression and telomere length changes following androgen deprivation. (A) Quantification of AR and (B) ENO2 expression in Figure 2.11A. (C) Representative standard curve of telomerase-positive HCT116 cells using Q- TRAP protocol for Figure 2.11D. ALT-positive U2OS sample included as negative control. (D) Using trypan blue exclusion, relative cell number was followed over 120 hours in C4-2B.NS cells versus C4-2B.KD.1 and KD.2 cells. (E) TRF analysis demonstrating telomere length changes in C42B cells with SLX4IP knockdown (KD.1, KD.2) versus C4-2B.NS grown in +FBS media. PD: Calculated population doublings. (F) Telomere length changes over time measured via PDs from (E) and Figure 2.11G. Simple linear regression models are depicted by dotted lines and slopes reported as bp/PD. Hexagon data points represent 45 day time points. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 2. 13: SLX4IP knockdown triggers accumulation of senescence- associated markers following androgen deprivation. (A) Representative bright field images demonstrating SA β-gal staining (arrowheads) for senescence in early and late-passage C4-2B cells with stable knockdown of SLX4IP grown in +CSS and +FBS. Scale bar: 20 μm. (B) Quantification of relative proportion of SA β-gal positive cells in (A). (C) Relative p21 expression in early and late-passage C4-2B cells with stable knockdown of SLX4IP grown in +CSS and +FBS. (D) Quantification of relative p21 expression in (C). Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

100

CHAPTER 3: SLX4IP N-TERMINUS DICTATES TELOMERIC LOCALIZATION IN ALT-LIKE CASTRATION-RESISTANT PROSTATE CANCER IN VITRO

A version of this chapter has been submitted for publication:

Mangosh, T. L., Grabowska, M. M., and Taylor, D. J. "SLX4IP N-terminus

dictates telomeric localization in ALT-like castration-resistant prostate cancer in

vitro.” 2021.

101

3.1 Abstract

To ensure replicative immortality in cancer, telomeres are maintained through activation of telomere maintenance mechanisms (TMMs), being telomerase or the alternative lengthening of telomeres (ALT) pathway. TMMs were thought to be mutually exclusive. However, ALT hallmarks were identified in cancers traditionally defined as telomerase-positive, supporting TMM coexistence.

For instance, telomerase was thought to be the primary TMM in castration- resistant prostate cancer (CRPC) but CRPC in vitro models with androgen receptor

(AR) loss demonstrated ALT hallmarks and required ALT-associated PML bodies

(APBs) for sustained telomere maintenance. This relationship was reliant on the uncharacterized ALT regulator SLX4IP. In order to understand how SLX4IP is responsible for promoting TMM coexistence in AR-negative CRPC models, regions of SLX4IP were interrogated for their ability to promote APB formation. The

N-terminus of SLX4IP was the only region capable of promoting APBs to a similar degree as full-length SLX4IP across CRPC cell lines. Moreover, APB promotion by the N-terminus of SLX4IP was able to partially rescue telomere shortening and senescent induction triggered by SLX4IP knockdown in AR-negative CRPC cells.

Additionally, the N-terminus prompted SLX4IP recruitment to telomeres and APBs.

Therefore, in order to elicit APB formation and telomere maintenance, SLX4IP must localize to the telomere via its N-terminal region, thus priming telomeres for

APB assembly. Without the N-terminal region, APB formation is minimized, telomeres shorten, and senescence is induced. These findings identify a role of

102

the uncharacterized ALT regulator SLX4IP in the promotion of TMM coexistence to perpetuate replicative immortality in CRPC in vitro.

3.2 Introduction

Telomeres compose the ends of linear chromosomes and consist of repeating DNA sequences camouflaged by a shelterin protein sheath 2–4,8.

Telomeres act as a safeguard against the end replication problem, thereby preventing genomic attrition.7 In non-malignant environments, telomeric DNA shortens with successive cell divisions until reaching a certain length threshold at which point senescence is triggered to prevent genomic loss.28 Conversely, malignant cells activate telomere maintenance mechanisms (TMMs) to overcome telomere attrition, acquire additional replicative capacity, and avoid senescence.38,40–42 Telomerase is activated in ~90% of cancers37 and, by catalyzing the addition of telomeric repeats to chromosomal ends, supports replicative immortality.40 Accounting for the remaining ~10% of cancers,38 the alternative lengthening of telomeres (ALT) pathway requires the redirection of DNA damage response proteins to the telomere to engage in homologous recombination (HR), thus extending the protective telomeric cap.42 Unlike the telomerase pathway, techniques to directly measure ALT activity have yet to be developed due to the poorly defined nature of its molecular mechanism.37,79

Instead, to be defined as ALT-positive, a collection of ALT-related surrogate measures must be evaluated. These markers include limited telomerase expression and activity42, telomere length heterogeneity42, telomeric C-circles102,

103

and ALT-associated Promyelocytic leukemia protein (PML) bodies (APBs).104

Through a combination of telomerase activity measures and ALT phenotypic signature identification, it is now possible to more accurately describe malignant

TMM programming and, more importantly, define the critical drivers instilling replicative immortality.

TMMs were originally thought to be mutually exclusive events.38,42,79

However, telomerase and ALT have been found to be foundational mechanisms programmed in malignancy regardless of the primary TMM employed.75,107,123,126,127,133 Moreover, these TMMs can be engaged simultaneously and coexist to varying degrees107,126–129,133 refuting the previous mutually exclusive TMM hypothesis. As such, emerging evidence support a complex TMM continuum in which telomerase and ALT coexist. Further highlighting the need for more complete TMM characterization in malignancy, ALT characteristics have been reported in a number of cancer types historically defined as telomerase-positive75,199; one such example is castration-resistant prostate cancer (CRPC).199

CRPC emerges following androgen deprivation therapy of primary or castration-sensitive prostate cancer.146,150,151 Mechanisms of castration-resistance can be divided into two types being AR-positive, which includes AR mutations, amplifications, and splice variants,68 or AR-negative, which includes AR loss and gain of neuroendocrine features.152,153,190,202 Though therapeutics for CRPC are available,175,204 they are minimally efficacious in AR-negative CRPC patients because they target the AR signaling axis.152,178 Additionally, treatment of AR-

104

positive CRPC with second-generation therapeutics induces an AR-negative phenotype.152 The emergence of AR-negative CRPC is particularly interesting as

AR status changes following progression correlate with TMM pathways in vitro.141

Primary prostate cancer, which is almost solely AR-positive, relies exclusively upon telomerase for telomere maintenance.196,197 Traditionally, CRPC was assumed to be similarly reliant on telomerase.196 However, a case study identified ALT-like telomeric characteristics in a metastatic site of a CRPC patient.199 Notably, this ALT-like presentation was not present in the primary tumor.199 Interestingly, analysis of AR-positive CRPC cell lines revealed a lack of

ALT characteristics and conserved telomerase activity, whereas AR-negative cell lines displayed ALT-like hallmarks including limited telomerase expression and activity, increased C-circle formation, and increased atypical APB telomeric structures.141 Interestingly, these latter models lacked telomere length heterogeneity and retained some telomerase activity, highlighting TMM coexistence in AR-negative CRPC cell lines.141 For these reasons and within the context of the current investigation, these AR-negative and AR-positive models will be referred to as ALT-like and telomerase-positive, respectively.

Despite considerable research efforts to elucidate ALT mechanisms, promoters of TMM coexistence remain poorly understood. SLX4 Interacting

Protein (SLX4IP) has emerged as a key player in regulating ALT and TMM plasticity.140–142 SLX4IP interacts with SLX4 as part of a nuclease scaffold hub that participates in ALT-related HR.143,144 Specifically, SLX4, SLX4IP and associated nucleases are recruited to the telomere via shelterin component TRF2 to resolve

105

telomeric HR junctions.25,143 SLX4IP also helps to differentially regulate primary

TMM, which is dependent on cancer type. For example, in osteosarcoma models,

SLX4IP knockout promotes an ALT phenotype in vivo,140 while SLX4IP knockdown in breast cancer models results in the loss of ALT-related characteristics in vitro.142

In the context of CRPC progression, elevated SLX4IP expression promotes an atypical ALT-like presentation that is characterized by an increase in atypical APB formation.141 Moreover, SLX4IP knockdown led to the specific loss of APB foci in

ALT-like AR-negative CRPC in vitro followed by consequential telomere shortening and induction of senescence.141 The reliance of ALT-like CRPC on

SLX4IP-dependent APB foci for telomere maintenance provides a starting point for unraveling the complexity behind TMM coexistence and the role it plays in malignant progression.

Because TMM coexistence occurs following CRPC progression and relies upon SLX4IP-dependent APB foci for telomere preservation, we chose to further characterize SLX4IP’s function in the context of CRPC in vitro models. First, we introduced SLX4IP constructs into CRPC cell lines to identify which SLX4IP regions are responsible for promoting the ALT-like presentation described previously.141 Whereas full-length SLX4IP was the only construct capable of recapitulating the complete ALT-like phenotype, the SLX4IP N-terminal domain

(SLX4IP.1-140) was sufficient to induce APB formation. Moreover, the promotion of APBs by SLX4IP.1-140 partially rescued the late-onset senescent phenotype triggered by endogenous SLX4IP knockdown and consequential telomere shortening. Additionally, the N-terminus of SLX4IP promoted telomeric localization

106

triggering APB formation and downstream telomere maintenance. Altogether, we propose SLX4IP localizes to the telomere, through interactions coordinated by its

N-terminal domain, to prime telomeric sites for the formation of APBs and, thus telomere maintenance in ALT-like CRPC models. Without the N-terminal region,

SLX4IP is unable to localize to the telomere, leading to a loss of APBs and late accumulation of senescent markers. Together these results provide foundational knowledge for understanding SLX4IP’s role in TMM coexistence as it relates to

CRPC disease progression models. More broadly, our findings help to define the function of SLX4IP in propagating replicative immortality in malignancy.

3.3 Results

3.3.1 The N-terminus of SLX4IP is responsible for the promotion of APB foci in CRPC in vitro models

In CRPC cell lines, SLX4IP overexpression promotes ALT-like characteristics including reduction of telomerase expression and activity, promotion of C-circles and APBs, and telomere length maintenance.141 Despite the characterization of this relationship between SLX4IP and ALT features, it is unclear which regions of SLX4IP are necessary for this phenotypic transition. To identify these crucial regions, five 3xFLAG-tagged SLX4IP constructs were designed

(Figure 3.1a) and stably overexpressed in C4-2B cells alongside full-length SLX4IP

(1-408) and empty vector (EV) (Figure 3.1b). Surprisingly, these constructs were unable to recapitulate the significant reduction in TERT mRNA expression (Figure

3.2a), reduction in telomerase activity (Figure 3.2b), and elevation in C-circle

107

abundance (Figure 3.2c) observed with overexpression of SLX4IP.1-408 in C4-2B cells. However, SLX4IP.1-140 overexpression, representing the first 140 amino acids of the SLX4IP protein, was capable of recapitulating the significant elevation in APB-positive cells observed with SLX4IP.1-408 introduction (Figure 3.1c and d).

Notably, the remaining truncations were unable to induce APBs to a similar degree as SLX4IP.1-408 suggesting SLX4IP.1-140 is minimally sufficient to promote this increase in APB-positive cells.

To confirm the N-terminal region is capable of promoting APB foci in additional CRPC cell lines, SLX4IP constructs were stably overexpressed in

DU145 and PC-3 cells (Figure 3.3a). Both cell types exhibit ALT-like markers at baseline, which includes telomeric C-circles and APB foci.141 We therefore predicted that the introduction of SLX4IP.1-140 would exacerbate the already present ALT-like phenotype, while the remaining constructs would have minimal effect. Just as before, a significant reduction in TERT mRNA expression (Figure

3.4a), reduction in telomerase activity (Figure 3.4b), and elevation in C-circle abundance (Figure 3.4c) was not observed with SLX4IP construct introduction.

Similar to the pattern observed with C4-2B cells, overexpression of SLX4IP.1-408 and 1-140 significantly increased the proportion of APB-positive cells to a similar degree (Figure 3.3b and c). Together, these data suggest full-length SLX4IP is necessary to support the complete ALT-like phenotype in CRPC in vitro models, but the N-terminus of SLX4IP, specifically the first 140 amino acids, is sufficient for the promotion of APB foci at telomeres.

108

3.3.2 Manipulation of SLX4IP expression alters abundance of nuclear PML foci but not PML expression

We have previously shown that SLX4IP-dependent APB foci are necessary for telomere length preservation in ALT-like CRPC in vitro models.141 With the identification of the SLX4IP N-terminus as the region responsible for APB foci promotion in CRPC cells, we sought to further elucidate the general mechanism behind this ALT-like presentation. We first explored whether SLX4IP increased the proportion of APB-positive cells indirectly by promoting elevated PML expression.

To do so, SLX4IP and PML protein expression was analyzed across six prostate cancer in vitro models. This study included one primary prostate cancer cell line

(LAPC4)200,234 and five CRPC models. Of the CRPC models, C4-2B, CWR-R1, and 22Rv1 cells are telomerase-positive and retain androgen receptor (AR) expression141,205,206 while DU145 and PC-3 cells are ALT-like and AR- negative.141,207,208 Interestingly, SLX4IP expression directly correlated with PML expression across all cell lines tested (Figure 3.5a and b). Furthermore, a direct correlation between SLX4IP expression and number of nuclear PML foci per cell was identified (Figure 3.5c and d).

To further define the observed relationship between SLX4IP and PML expression, SLX4IP.1-408 was overexpressed in C4-2B, DU145 and PC-3 cells with EV control (Figure 3.6a). Upon evaluation of PML expression, no significant differences were found between EV and SLX4IP.1-408 cell line pairs (Figure 3.6a).

However, SLX4IP overexpression did lead to a significant elevation in the number of nuclear PML foci in each cell line pair tested (Figure 3.6b). Complementary to

109

these findings, SLX4IP knockdown in DU145 and PC-3 cells did not significantly alter PML expression (Figure 3.6c) but did lead to a significant loss of nuclear PML foci when compared to non-targeting control (NS) (Figure 3.6d). Since the genetic manipulation of SLX4IP does not alter PML protein expression, these data suggest that SLX4IP does not promote APB foci through transcriptional, translational, or post-translational regulation. However, our results highlight a direct relationship between SLX4IP expression and nuclear PML foci formation, thereby suggesting a role for SLX4IP in triggering telomeric PML recruitment and assembly of APB structures.

3.3.3 The N-terminus of SLX4IP is responsible for telomeric localization

Upon defining the relationship between SLX4IP expression and nuclear

PML foci recruitment, we hypothesized SLX4IP may function as a primer at the telomere for APB structure assembly and downstream telomere maintenance. To test this, we coimmunoprecipitated endogenous SLX4IP from telomerase-positive

C4-2B and 22Rv1 cell lines and ALT-like DU145 and PC-3 cell lines and then probed for the telomere end-binding protein TRF2. We reasoned that if SLX4IP localization at the telomere to trigger APB foci formation, then a greater proportion of SLX4IP would be expected to reside at the telomere in ALT-like cell lines.

Congruent with this hypothesis, TRF2-containing SLX4IP complexes are elevated in ALT-like cell lines when compared to their telomerase-positive counterparts

(Figure 3.7a). Additionally, analysis of telomeric foci via IF-FISH revealed an average of ~37% of DU145 and PC-3 cells had at least one SLX4IP-positive

110

telomere versus only ~10% and ~12% in C4-2B and 22Rv1 cells, respectively

(Figure 3.7b and c). Moreover, analysis of APBs showed an average of ~20% of

DU145 and ~18% of PC-3 cells had at least one SLX4IP-positive APB foci (Figure

3.7b and c). These events were significantly less frequent amongst C4-2B and

22Rv1 cells with only ~4% and ~1% of cells, respectively. We attribute this observation to be due, at least in part, to ALT-like CRPC cell lines having a significantly higher proportion of APB-positive cells as compared to those that are telomerase-positive (Figure 3.7b and c). But in support of our proposed role of

SLX4IP in mediating telomere recruitment of PML to the telomere, there was no statistically significant difference in the percent of APBs bound by SLX4IP between cell lines tested (Figure 3.7b and c). Taken together, these findings show endogenous SLX4IP can localize to the telomere and can be found at a significant proportion of APBs in CRPC in vitro models.

Moving forward, we tested if the N-terminal region, which is responsible for promoting APBs, is doing so through telomeric localization. To test this, 3xFLAG- tagged SLX4IP constructs were coimmunoprecipitated from C4-2B and PC-3 cells overexpressing SLX4IP.1-408, 1-140, 141-408, or control and complexes were probed for TRF2. In both cell lines, TRF2 enrichment was greater in complexes containing SLX4IP.1-408 and 1-140 when compared to cells with SLX4IP.141-408

(Figure 3.8a). This result was further confirmed following analysis of telomeric and

APB foci via IF-FISH. In C4-2B cells, SLX4IP.1-408 and 1-140 construct overexpression triggered a similar percent of cells with SLX4IP-positive telomeres and SLX4IP-positive APB foci; however, a significantly lower percent of cells

111

exhibited these events with SLX4IP.141-408 introduction (Figure 3.8b and c). A nearly identical pattern was observed in PC-3-derived cell lines (Figure 3.8b and c). These results identify the first 140 amino acids of SLX4IP as the primary region responsible for directing SLX4IP to the telomere.

Here we show that SLX4IP is capable of interacting with telomeres and can be found at sites of APB formation. Moreover, the N-terminal region of SLX4IP that is responsible for promoting APB structures is also responsible for the telomeric localization of SLX4IP. These findings suggest SLX4IP’s N-terminus directs

SLX4IP to the telomere to act as a primer for APB formation in CRPC in vitro models.

3.3.4 SLX4IP N-terminus rescues the late-onset senescent phenotype induced by SLX4IP knockdown

Telomere maintenance in ALT-like CRPC in vitro models requires SLX4IP- dependent promotion of APBs.141 Consequently, loss of SLX4IP expression leads to telomere shortening and induction of senescent markers. It is unclear if the promotion of APB foci by the N-terminal region of SLX4IP is sufficient to promote telomere maintenance similar to full-length SLX4IP. To test this possibility,

SLX4IP.1-408, 1-140, and 141-408 were overexpressed in PC-3 cells with a background of endogenous SLX4IP knockdown (KD.1 and KD.2) (Figure 3.9a). As expected, introduction of SLX4IP.1-408 and 1-140, but not 141-408, led to a significant increase in the percent of APB-positive cells when compared to PC-

3.KD.1 and KD.2 cells (Figure 3.9b). After establishing the proportion of APB-

112

positive cells in each cell line, senescent markers were evaluated at early and late passages.

Since the N-terminus of SLX4IP is responsible for APB formation, we reasoned that expression of SLX4IP constructs containing this N-terminal domain

(SLX4IP.1-408 and 1-140) would similarly blunt the previously characterized telomere-mediated senescent phenotype that occurs with SLX4IP knockdown.141

After identification of APB rescue, senescent-associated markers were evaluated.

β-galactosidase (β-gal) staining214 and p21 protein expression215 of early and late passage cells from each PC-3-derived line were analyzed. When comparing early passages, minimal differences in β-gal positivity and induction of p21 protein expression was noted (Figure 3.10b, c and d). However, this pattern of senescent markers was marginally elevated when comparing early and late time points for

PC-3.KD.1+1-408 and KD.1+1-140 cells (Figure 3.10b and c). This pattern was also true for PC-3.KD.2+1-408 and KD.2+1-140 cells. In contrast, when comparing late and early passages, a significant elevation in β-gal-positive cells and p21 protein expression was present in cells with SLX4IP.141-408 introduction (Figure

8b, c and d). This pattern of senescent marker induction was not significantly different from the pattern observed with SLX4IP knockdown alone. These findings suggest the SLX4IP-dependent APB rescue identified minimizes the induction of senescence in cells with SLX4IP.1-408 and 1-140 introduction. Altogether, we conclude that the N-terminal region of SLX4IP localizes to the telomere priming it for APB foci formation to maintain telomere length in ALT-like CRPC in vitro models (Figure 3.11). Without the N-terminus, SLX4IP is no longer directed to the

113

telomere, APB foci formation is limited, and telomeres shorten inducing a senescent phenotype (Figure 3.11)

3.4 Discussion

Presently, a knowledge void surrounds the molecular mechanism behind

SLX4IP’s role in ALT-like processes and TMM coexistence. Moreover, regulatory and functional regions of SLX4IP remain poorly characterized. Here we show that the N-terminal domain is required for telomere localization and maintenance.

Previous studies have demonstrated that SLX4IP’s N-terminus bridges the interaction between SLX4 and SLX4-asssociated nucleases to participate in intra- strand crosslink repair144; however, it is unclear if these interactions remain intact in the context of ALT. Additionally, SLX4IP’s localization at the telomere is reported to be SLX4-dependent in bona fide ALT in vitro models.140 While indirect SLX4IP- telomeric interactions coordinated by SLX4 is possible, the ability of the SLX4 complex to function as an E3 ligase235 and known importance of SUMOylation in telomeric and ALT regulator recruitment confounds this hypothesis.115 Therefore, it will be important to determine if SLX4 bridges SLX4IP’s interaction with telomeres of ALT-positive cells and if SUMOylation of SLX4IP regulates its ALT- associated telomeric interactions.

The direct correlation between SLX4IP and PML protein expression described in our study highlights another deficit with regard to SLX4IP regulation, specifically transcription. In CRPC progression models, the transition from an AR- positive to AR-negative phenotype is accompanied by ALT-like APB formation and

114

induction of SLX4IP expression, suggesting the molecular programs driving AR loss may also promote SLX4IP expression.141 Likewise, transcriptional regulators altering SLX4IP expression may similarly alter PML expression. Interestingly, the pro-inflammatory cytokine IL-6 that is regulated through the JAK/STAT3 pathway induces PML expression in both somatic and malignant environments.236 In the context of prostate cancer, significantly higher expression of IL-6 is associated with

AR loss in CRPC models 237 and regulates neuroendocrine differentiation, a molecular program capable of driving CRPC growth in the absence of AR.238,239

This connection between SLX4IP and PML expression, ALT associated phenotypes, IL-6 regulation, and AR-negative CRPC identifies an avenue that may uncover the transcriptional program governing SLX4IP-dependent telomere maintenance.

Unfortunately, our dissection of the SLX4IP protein was unsuccessful in identifying regions that are responsible for several ALT characteristics triggered by full-length SLX4IP overexpression in CRPC cell lines, including enhanced C-circle formation and inhibition of telomerase expression and activity.141 This finding suggests there are multiple regions of SLX4IP that cooperate to induce these phenotypic changes. For example, the lack of a specific region of SLX4IP might prevent conformational changes of a distant region that is necessary for the complete phenotypic transition. Alternatively, the C-terminal domain of SLX4IP may be necessary for regulating protein-protein interactions with unidentified proteins necessary for these functions. Nevertheless, it is important to determine how each characteristic is promoted by SLX4IP because these investigations will

115

provide clues as to whether a true ALT mechanism is engaged at the telomere in

CRPC.

Although ALT-like CRPC models do exhibit certain ALT hallmarks, they lack telomere length heterogeneity, a canonical ALT marker.141 Furthermore, these models have significantly fewer C-circles and retain some telomerase activity compared to bona fide ALT cell lines.141 Our findings also suggest that telomere maintenance in ALT-like CRPC does not completely reflect a true ALT mechanism; but instead, might engage already present telomeric machinery in an ALT-like fashion to escape senescence. For instance, the telomere maintenance we observe is reliant upon SLX4IP-dependent APBs only, and not the entire ALT-like phenotype observed.

While ALT-like CRPC may not fully recapitulate the traditional ALT phenotype, the underlying mechanisms driving telomere maintenance cannot be substituted for telomerase in this environment. Importantly, telomerase alone is unable to sufficiently maintain telomeres in ALT-like CRPC cells, as cells with impaired ALT capabilities and telomerase reconstitution well above endogenous levels were unable to maintain their telomere length and prevent senescence.141

Moving forward, it will be important to determine how each ALT characteristic is independently promoted by SLX4IP because these investigations will provide clues as to whether a true ALT mechanism is engaged at the telomere in CRPC.

While investigating a traditional ALT mechanistic framework provides a valid starting point in malignancy with TMM coexistence, it is possible that the ALT-like

116

processes engaged in ALT-like CRPC progression may diverge mechanistically from traditional ALT.

Because of the unique reliance of ALT-like CRPC on SLX4IP-dependent

APBs for telomere maintenance, investigation of SLX4IP regions important for telomere maintenance was completed in CRPC in vitro models. Interestingly, only full-length SLX4IP promoted the complete ALT-like phenotype suggesting the entire protein is likely required to elicit this transition. However, the N-terminus, comprising the first 140 amino acids, was responsible for inducing APB formation in C4-2B, DU145, and PC-3 cell lines. Moreover, the N-terminus directed SLX4IP to the telomere and was found at sites of APB formation. To test if the N-terminus was capable of preventing late-onset senescence, consistent with a telomere- based mechanism, this region was overexpressed in ALT-like PC-3 cells with

SLX4IP knockdown. SLX4IP knockdown caused dramatic senescence induction at late passages but this was partially rescued by the N-terminal region of SLX4IP via APB formation. Taken together these results highlight the importance of

SLX4IP’s N-terminus in promoting ALT-like telomere maintenance in a CRPC environment and provide a foundation to guide future TMM investigations in newly identified ALT-like malignant environments.

3.5 Materials and Methods

Cell Culture and Stable Cell Line Generation: All cell lines were maintained at

37°C in 5% CO2 following acquisition from ATCC. LAPC4 cells were grown in

IMDM (Sigma) supplemented with 10% (v/v) FBS (Sigma) and 1 nM

117

dihydrotestosterone (Sigma). C4-2B, CWR-R1, 22Rv1, DU145, and PC-3 cells were grown in RPMI1640 (Gibco) supplemented with 10% FBS and 1% of

Penicillin 10 000 units/mL, Streptomycin 10 000 μg/mL, and 25 μg/mL

Amphotericin B mixture (Antibiotic-Antimycotic, Gibco). HEK293T, HCT116, and

U2OS cells were grown in DMEM (Gibco) supplemented with 10% (v/v) FBS

(Sigma) and 1% Antibiotic-Antimycotic (Gibco). All cell lines were tested for

Mycoplasma using the MycoAlert™ PLUS Detection Kit (Lonza) and STR validated

(CWRU Genomics Core) prior to experimental testing and following each genetic modification.

Generation and selection of stable cell lines overexpressing SLX4IP and

SLX4IP truncation constructs utilized an identical strategy as described previously.141 To generate the 3XFLAG-tagged SLX4IP truncation constructs depicted in Figure 3.1a, full-length 3XFLAG-tagged SLX4IP was sub-cloned by

TOP Gene Technologies, Inc. Generation and selection of stable cell lines with

SLX4IP knockdown utilized an identical strategy as described previously.141

RNA and Protein Isolation and Analysis: RNA was isolated, quantified, and analyzed as described previously.141 Protein isolation, western blotting analysis and quantification was completed as published previously.141 Primary antibody dilutions are as follows: 1:10,000 GAPDH (Cell Signaling, 2118), 1:3,000 SLX4IP

(Sigma, HPA046372), 1:3,000 FLAG (Sigma, F1804), 1:5,000 TRF2 (Novus

Biologicals, NB110-57130), 1:1,000 PML (Cell Signaling, E5R8T and Santa Cruz

Biotechnology, sc-5621), 1:3,000 p21 (Cell Signaling, 2947).

118

Coimmunoprecipitation experiments were carried out using the

Dynabeads™ Co-Immunoprecipitation Kit according to manufacturer instructions

(ThermoFisher Scientific). The provided immunoprecipitation (IP) buffer included was modified with an additional 50 mM NaCl and 0.05% Triton X-100. Cells were grown in 15 cm dishes and scraped using the modified IP buffer, treated with 5 μL of recombinant DNaseI (Sigma), rotated at 4° C for one hour, and cleared by centrifugation. Following protein quantification, 2 mg of protein was incubated with

1.5 mg of Dynabeads coupled with 10 µL of FLAG antibody (Sigma, F1804) or 20

µL of SLX4IP antibody (Sigma, HPA046372) per protocol. After completion, whole cell lysates and coimmunoprecipitation samples were analyzed by western blotting as described above.

Telomere Repeat Amplification Protocol and C-circle Isolation and Analysis:

A previously published Telomere Repeat Amplification Protocol (TRAP) and C- circle dot blot assay was performed and analyzed without modifications.105,141,232

A standard curve of telomerase-positive C4-2B cells alongside ALT-positive U2OS control were included for each TRAP assay and included in supplemental figures.

Controls were included in each C-circle analysis using HEK293T or HCT116 as a telomerase-positive control and U2OS cells as an ALT-positive control.

Immunofluorescence-Fluorescence in situ Hybridization Analysis: Analysis of nuclear PML foci, ALT-like APBs, and SLX4IP-positive ALT-like APBs was completed using a previously published immunofluorescence-fluorescence in situ

119

hybridization (IF-FISH) protocol with minor modifications.141,233 Coverslips were incubated with PML antibody (diluted 1:200, Santa Cruz Biotechnology, sc-5621 or sc-377390) either alone, with SLX4IP antibody (diluted 1:200, Sigma,

HPA046372) or with FLAG antibody (diluted 1:200, Sigma, F1804) for one hour at room temperature. Alexa-488® conjugated anti-rabbit and/or Alexa-594® conjugated anti-mouse secondary antibody (Jackson ImmunoResearch) was diluted 1:100 and added to coverslips for overnight incubation at 4°C. Hybridization with TelC-Cy5-labeled telomere probe and DAPI staining was carried out just as described. With masking, imaging of stained slides was carried out using a Leica

HyVolution SP8 gated STED Microscope and LAS X Software. Cell count, nuclear foci, and potential colocalization events were identified and quantified in ImageJ.230

Telomere Restriction Fragment Analysis and Senescence-associated β- galactosidase Staining: Telomere restriction fragment (TRF) analysis was completed as described previously but using a total of 8 μg of genomic DNA for digestion.141,231 Senescence-associated (SA) β-galactosidase (β-gal) activity staining was completed as published.141

Statistical Analysis: Statistics were performed in GraphPad Prism 8 software using a two-tailed Student’s t test when comparing two groups or a one-way

ANOVA with multiple comparisons when comparing at least three groups. All in vitro data with the exception of IF and IF-FISH data are represented as mean values from three independent experiments performed in triplicate unless

120

otherwise noted. IF and IF-FISH are represented as violin plots with medians and quartiles and include quantification from each cell from three independent experiments (nuclear PML foci) or each field of view from three independent experiments (ALT-like APBs and SLX4IP-positive ALT-like APBs). p values less than 0.05 were considered statistically significant.

3.6 Acknowledgements

We thank members of the Taylor and Grabowska labs for helpful manuscript comments and experimental design discussions. The Grabowska lab is supported by start-up funds (MMG) provided by the Case Research Institute. The Taylor Lab is supported by the NIH (R01 GM133841 and R01 CA240993 to DJT). The CWRU

SOM Light Microscopy Core Facility is supported by NIH (S10 OD016164).

121

Figure 3. 1: N-terminus of SLX4IP promotes APB structures in C4-2B cells. (a) Schematic representation of 3xFLAG-tagged SLX4IP constructs. (b) Confirmation of stable overexpression of SLX4IP constructs alongside full-length SLX4IP (1-408) and empty vector (EV) in C4-2B cells. (c) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) at telomeres (red) in C4-2B cells with SLX4IP overexpression. White outline represents nuclear boundary. Scale bar: 5 μm. (d) Quantification of (c) for percent positive cells with at least one APB. Data represented as mean+SD; n≥3; All constructs compared to 1-408 positive control; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

122

123

Figure 3. 2: SLX4IP constructs are unable to recapitulate ALT-like phenotype observed with full-length SLX4IP. (a) Relative TERT mRNA expression and (b) telomerase activity following stable SLX4IP overexpression in C4-2B cells. Standard curve of telomerase-positive C4- 2B cells using TRAP protocol is shown to the right. ALT-positive U2OS and C4-2B cells treated with RNaseI included as negative controls. (c) Representative dot blot demonstrating the presence of telomeric C-circles in C4-2B cells with SLX4IP overexpression. +Φ lane indicates reaction with Φ29 polymerase and −Φ lane indicates control reaction. Telomerase-positive HCT116 (TELO) and ALT-positive U2OS (ALT) cells included as controls. Quantification of C-circle abundance defined as the signal ratio of +Φ to −Φ reaction is shown below. Dotted line at y=2 indicates threshold signal ratio suggesting ALT activity. Data represented as mean+SD; n≥3; All constructs compared to 1-408 positive control; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

124

Figure 3. 3: N-terminus of SLX4IP induces additional APB-positive cells in DU145 and PC-3 cells. (a) Confirmation of stable overexpression of SLX4IP constructs alongside full- length (1-408) SLX4IP and empty vector (EV) in DU145 and PC-3 cells. (b) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) at telomeres (red) in DU145 and PC-3 cells with SLX4IP overexpression. White outline represents nuclear boundary. Scale bar: 5 μm. (c) Quantification of (b) for percent positive cells with at least one APB. Data represented as mean+SD; n≥3; All constructs compared to 1-408 positive control; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

125

Figure 3. 4: SLX4IP constructs are unable to exacerbate the ALT-like phenotype of AR-negative CRPC models. (a) Relative TERT mRNA expression and (b) telomerase activity following stable SLX4IP overexpression in DU145 and PC-3 cells. Standard curve of telomerase- positive C4-2B cells using TRAP protocol is shown below. C4-2B cells treated with RNaseI included as negative controls. (c) Representative dot blot demonstrating the presence of telomeric C-circles in DU145 and PC-3 cells with SLX4IP overexpression. +Φ lane indicates reaction with Φ29 polymerase and −Φ lane indicates control reaction. Telomerase-positive HCT116 (TELO) and ALT-positive U2OS (ALT) cells included as controls. Quantification of C-circle abundance defined as the signal ratio of +Φ to −Φ reaction is shown below. Dotted line at y=2 indicates threshold signal ratio suggesting ALT activity. Data represented as mean+SD; n≥3; All constructs compared to 1-408 positive control; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

126

127

Figure 3. 5: SLX4IP expression directly correlates with endogenous PML expression and number of PML nuclear foci in CRPC in vitro. (a) PML and SLX4IP protein expression across CRPC cell lines. Quantification of PML and SLX4IP expression in (a) is shown to the right. (b) Relative PML expression plotted versus relative SLX4IP expression in CRPC cell lines. Simple linear regression model depicted by dotted line with R2 value reported. (c) Representative IF images demonstrating the presence of nuclear PML foci across CRPC cell lines. White outline represents nuclear boundary. Scale bar: 5 μm. (d) Quantification of (c) for number of nuclear PML foci per cell. Data represented as mean+SD; n≥3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

128

Figure 3. 6: Genetic manipulation of SLX4IP expression alters number of PML nuclear foci but not PML expression. (a) PML expression in C4-2B, DU145 and PC-3 cells with full-length SLX4IP (1- 408) overexpression versus empty vector (EV). Quantification of PML expression in (a) is shown to the right. (b) Representative IF images demonstrating the presence of nuclear PML foci in C4-2B, DU145 and PC-3 cells with full-length SLX4IP overexpression versus EV. White outline represents nuclear boundary. Scale bar: 5 μm. Quantification of (b) for number of nuclear PML foci per cell is shown below. (c) PML expression in DU145 and PC-3 cells with SLX4IP knockdown (KD.1, KD.2) versus non-targeting control (NS). Quantification of PML expression in (c) is shown to the right. (d) Representative IF images demonstrating the presence of nuclear PML foci in DU145 and PC-3 cells with SLX4IP knockdown versus non-targeting control. White outline represents nuclear boundary. Scale bar: 5 μm. Quantification of (d) for number of nuclear PML foci per cell is shown to the right. Data represented as mean+SD; n≥3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

129

130

Figure 3. 7: Telomeric localization of SLX4IP at APB foci occurs in DU145 and PC-3 cells. (a) Representative immunoblots of coIP and whole cell lysate (input) samples from CRPC cells. CoIP was completed with SLX4IP antibody and then complexes were probed for SLX4IP, TRF2, SLX4, and GAPDH. (b) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) and SLX4IP (blue) at telomeres (red) in CRPC cell lines. White outline represents nuclear boundary. Scale bar: 5 μm. (c) Quantification of (b) for percent positive cells with at least one SLX4IP-positive telomere (magenta), percent positive cells with at least one SLX4IP-positive APB (white), percent APB-positive cells (yellow), and percent SLX4IP-positive APBs (white). Data represented as mean+SD; n≥3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

131

132

Figure 3. 8: N-terminus of SLX4IP coordinates telomeric localization at APBs. (a) Representative immunoblots of coIP and whole cell lysate (input) samples from C4-2B and PC-3 cells with stable overexpression of SLX4IP constructs. CoIP was completed with FLAG antibody and then complexes were probed for FLAG, TRF2, SLX4 and GAPDH. (b) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) and FLAG (blue) at telomeres (red) in C4-2B and PC-3 cells with stable overexpression of SLX4IP constructs. White outline represents nuclear boundary. Scale bar: 5 μm. (c) Quantification of (b) for percent positive cells with at least one FLAG-positive telomere (magenta) and percent positive cells with at least one FLAG-positive APB. (white). Data represented as mean+SD; n≥3; All constructs compared to 1-408 positive control; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

133

134

Figure 3. 9: Loss of APBs following SLX4IP knockdown is rescued by SLX4IP N-terminus introduction. (a) Confirmation of stable overexpression of SLX4IP constructs alongside full- length (1-408) SLX4IP in PC-3 cells with SLX4IP knockdown (KD.1 and KD.2). PC- 3.NS represents non-targeting control. (b) Representative IF-FISH images demonstrating the presence (arrowheads) or absence of PML (green) at telomeres (red) in PC-3 cells with SLX4IP construct overexpression on a background of SLX4IP knockdown. White outline represents nuclear boundary. Scale bar: 5 μm. (c) Quantification of (b) for percent APB-positive cells. Data represented as mean+SD; All constructs compared to 1-408 positive control; n≥3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

135

136

Figure 3. 10: SLX4IP N-terminus prevents late-onset accumulation of senescent markers. (a) Representative bright field images demonstrating β-gal staining (arrowheads) for senescence in early and late passage PC-3-derived cells. Scale bar: 20 μm. (b) Quantification of percent β-gal positive cells in (a). (c) p21 expression following knockdown of SLX4IP in early and late passage PC-3-derived cells. (d) Quantification of relative p21 expression in (c). Data represented as mean+SD; All constructs compared to KD.1 or KD.2 positive controls; n≥3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

137

138

Figure 3. 11: N-terminus of SLX4IP is required for telomere maintenance in ALT-like CRPC in vitro. Schematic representation of the N-terminal region of SLX4IP localizing to the telomere, priming it for APB formation to maintain telomere length in ALT-like CRPC. Without this region, SLX4IP is unable to localize at the telomere, APB foci formation is hindered, and senescence is induced due to insufficient telomere maintenance.

139

CHAPTER 4: DISCUSSION AND FUTURE DIRECTIONS

Portions of this chapter were previously published in:

Whited, T. L., and Taylor, D. J. "Expanding the chemotherapeutic potential of an established nucleoside analog with selective targeting of telomerase." Molecular

& Cellular Oncology 5.6 (2018): e1536844.

140

4.1 ALTiness: The FinALT Telomeric Frontier

The poorly characterized AR-negative mechanisms driving CRPC progression is a source of significant mortality among CRPC patients.151

Identification of phenotypes differentiating AR-positive from AR-negative disease may uncover targetable molecular mechanisms essential for CRPC progression and AR loss. Here we show AR-negative CRPC demonstrates a unique reliance on SLX4IP-dependent ALTiness that was not required by their AR-positive counterparts.141 ALTiness in this particular case was defined by limited telomerase activity, abundance of C-circles, and presence of APBs.141 Interestingly, telomere maintenance in AR-negative CRPC was reliant only on SLX4IP-dependent APB formation.141 The atypical ALT presentation exhibited by AR-negative CRPC in vitro models is mimicked across a variety of cancer types not typically associated with ALT activity.75,136 To add further complexity, representative models of

ALTiness do not contain a traditional ALT phenotype and ALT hallmarks are generally not robustly demonstrated.75,136,141 Therefore, the concept of ALTiness is in limbo between telomerase and ALT on the TMM spectrum but the role this hybrid TMM phenotype plays in malignancy is a mystery.

It is possible that there is a balancing act between TMM programs and other malignant programs to identify the most advantageous molecular make-up to ensure disease progression. In the example of primary prostate cancer progression to AR-positive CRPC, first-generation antiandrogens disrupt both the

AR-signaling axis and androgen-driven telomerase activity but activation of AR mutants or splice variants reestablish AR signaling, driving both disease

141

progression and TMMs.149,197,198 Similarly, introduction of second-generation antiandrogens in AR-positive CRPC disrupts AR signaling driven by wild-type and mutant AR.177 AR-positive CRPC has already exploited AR-dependent resistance mechanisms potentially limiting the molecular avenues for circumventing

ADT.158,159,162,169 As a result, AR loss and adoption of AR-independent programs is a likely fate.151–154,194,240 However, this transition to an AR-negative phenotype may prevent telomerase expression thereby hindering telomere maintenance.197,198 To circumvent this limitation, regulators of DNA damage and telomere maintenance may be recruited to the telomere to participate in atypical

ALT, perpetuating telomere maintenance. This proposed role of ALTiness in disease progression permits TMM flexibility, allowing malignant adaptation in the presence of therapeutics without sacrificing telomere maintenance. Therefore,

ALTiness may serve as a rescue TMM when malignant programs must be adopted that impair telomerase’s capabilities.

4.2 SLX4IP-mediated ALTiness or true ALT activity

The detailed molecular mechanisms driving bona fide ALT and its exact protein requirements remain to be fully explored but a body of evidence supports

HR events as foundational for ALT activity.78,79,123 In contrast, TMM coexistence driving ALTy presentations was only recently identified in patients with traditionally telomerase-positive cancers; therefore, mechanisms governing this phenotype remain unknown.75,124,136,138,139 Importantly, SLX4IP has been evaluated in both

ALTiness and bone fide ALT models but these studies demonstrated contradicting

142

roles of SLX4IP.140–142 Osteosarcoma models represent true ALT activity and

SLX4IP knockout led to an induction of ALT hallmarks and extreme telomere elongation.140 However, knockdown of SLX4IP in ALTy cell lines triggered loss of

ALT hallmarks.141,142 This differential regulation by SLX4IP coupled with distinct differences in ALTy versus true ALT phenotypes indicates the programs promoting

ALT may not be equally engaged across the ALT spectrum.

Activation of pathways driving ALT or ALTiness may be dictated by telomeric chromatin status.56,85,86 ALT cells harbor relatively relaxed chromatin structures compared to telomerase-positive cells.86 The resulting elevated telomere accessibility is permissive for additional DNA secondary structure formation further reducing telomeric chromatin compaction.84,241 This accessibility also renders a telomeric environment permissive for ALT-mediated HR, RCA and

TSCE.84,241–243

In contrast, ALTiness generally retains some telomerase activity and clonal divergence of TMMs supports the acquisition of ALTiness in a cell previously reliant on telomerase.124,129 In our model of disease progression, telomerase- positive AR-positive CRPC cells lose a majority of their original telomerase activity as they transition to an ALTy AR-negative phenotype.141 Because cells demonstrating ALTiness likely originate from cells dependent on telomerase, it is possible that telomeric chromatin compaction more closely resembles that of a telomerase-positive environment.86 This restricted accessibility of telomeric chromatin may prevent the localization of true ALT regulators at the telomere, providing an explanation for the differences in regulation and presentation.

143

ATRX and DAXX status may also dictate telomeric chromatin compaction permissive for ALT.88–90,243,244 Loss of ATRX/DAXX is highly correlative with ALT activity in cancer.86,245 ATRX and DAXX cooperate to promote histone variant H3.3 enrichment at the telomere.246 Genomic enrichment of H3.3 generally marks sites of active, accessible chromatin; however, telomeric enrichment of H3.3 indirectly prevents telomeric chromatin relaxation by preventing secondary structure formation.86,241,243 If ATRX or DAXX is lost and H3.3 is not loaded at the telomere, then telomeric chromatin is rendered more accessible and permissive for ALT activity.86 In our study, ALT-like AR-negative CRPC retains ATRX expression despite the presence of ALT hallmarks. If ATRX and DAXX remain intact, then it is likely that the telomeric chromatin structure of ALTy cancers significantly differs in accessibility from true ALT.86 This further limits the recruitment of true ALT mediators to the telomere resulting in an incomplete ALT phenotype, like that identified in AR-negative CRPC.

4.2.1 Canonical ALT Mediators Do Not Translate to ALTiness

Loss of ATRX is highly correlative with ALT activity but is not consistently lost in models of ALTiness.86,245 The implications of this specific loss of chromatin remodelers in true ALT are not limited to telomeric chromatin compaction.

Enrichment of H3.3 via ATRX within the genome marks transcriptionally active sites.243,247 Therefore, ATRX status in ALT versus ALTiness likely dictates divergent transcriptomes in addition to telomeric chromatin compaction. Loss of

ATRX may prevent accessibility of the TERT promotor or regulatory elements

144

driving transcription of ALT repressors, thereby generating a telomeric environment permissive for ALT activity. In contrast, ALTiness can retain functional

ATRX and telomerase activity despite ALT activation, supporting a divergent transcriptional signature.141

The elevated genomic instability characteristic of ALT, due to accumulation of replicative stress, may similarly contribute to a differential gene expression profile from that of an originally telomerase-positive cell turned ALTy.21,81,82,91,246

Generation of a bona fide ALT phenotype requires the expression of a number of canonical ALT mediators to work collaboratively with already present DNA damage response proteins, generating stereotypical ALT hallmarks at robust levels.79,80 In the case of ALTiness, telomerase activity is maintained and the ALT phenotype is incomplete and/or blunted.75,124,136,138,139,141,142,199 One potential explanation for this is a lack of canonical ALT regulator expression due to a genomic chromatin structure more closely reflecting that of a telomerase-reliant environment, where true ALT regulators are not necessary.86 Unfortunately, the lack of defined ALT regulators make understanding how genomic chromatin structure influences ALT- associated gene expression profiles difficult.

Further highlighting the discordance between ALT and ALTiness is the role of SLX4IP in regulating each TMM phenotype.140–142,144 SLX4IP’s role in ALT is not well defined but SLX4IP interacts with and regulates several canonical ALT regulators.140,143,144 In true ALT models, SLX4IP antagonizes BLM-mediated telomere synthesis by promoting SLX4-dependent cross over events.140

Specifically, knockout of SLX4IP in true ALT models led to extreme telomere

145

elongation and robust induction of ALT hallmarks.140 Conversely, in two different models of ALTiness, SLX4IP knockdown led to a loss of ALT hallmarks.141,142

Interestingly in the context of ALTiness, we show that SLX4IP still interacts with

TRF2, SLX4, and APB structures, all of which are true ALT mediators, but elicits the opposite effect.141 While some canonical ALT mediators are still engaged in

ALTiness there is a clear mechanistic difference. This difference is likley regulated by divergent protein requirements and these protein requirements dictate SLX4IP function. Dissimilarities in ATRX status, telomeric chromatin compaction and

SLX4IP regulation suggest that molecular mechanisms driving ALT may not translate directly to ALTiness; but, instead demonstrate some degree of overlap.86,88,140–142 Characterizing these difference will begin to unravel the complex TMM continuum available for malignancy.

4.2.2 ALTiness via DNA Damage Machinery Hijacking

Acquisition of TMM coexistence following disease progression and/or therapeutic resistance categorizes ALTiness as a mechanism of resistance.107,126,132,133,141,199 Here we show telomerase as the primary TMM is not sustainable in AR-positive CRPC cells grown in androgen-deprived conditions, likely due to the inhibition of AR-dependent TERT expression.67,141,198 In this disease progression model, AR loss is accompanied by significant APB formation that aids in telomere maintenance.141 When telomerase is disadvantageous, ALT- like mechanisms are engaged to ensure sufficient telomere length. It is clear from the deficient bona fide ALT phenotype that the mechanism behind ALTiness may

146

be lacking in protein components through either limited or lack of expression. In line with this hypothesis, ALT is thought to repurpose already present DNA damage response proteins and telomere-associated proteins, instilling dual functionality.80 However, baseline expression of these dual-function proteins may not be sufficient for eliciting robust ALT activity in the absence of canonical ALT regulators. The result is a subdued ALT-like presentation recapitulating some, but not all aspects of true ALT. True ALT and DNA damage response programs provide a starting point for investigations targeted at ALTiness but protein requirement variance is expected based on these phenotypic differences.

4.3 Regulation of SLX4IP: From Transcriptional to Post-translational

A commonality amongst ALT and ALTy models is the lack of knowledge regarding all aspects of SLX4IP regulation. SLX4IP expression is induced following

AR loss and adoption of AR-negative programs in CRPC, suggesting the programs driving this transition may similarly drive SLX4IP and ALTiness.141 Unfortunately, regulatory programs bypassing AR in CRPC remain poorly defined and inconsistent across models, providing an unideal starting point for SLX4IP investigations.154,173,183,187,188,193,195 Similarly mysterious are the molecular pathways driving both ALT and ALTiness. Without efforts to define drivers of these phenotypes, the multilevel regulation of SLX4IP in these contexts will remain difficult to characterize.

147

4.3.1 Androgen Deprivation Therapy and SLX4IP

Induction of SLX4IP following long-term androgen deprivation in a model of

CRPC progression provides clues for unraveling SLX4IP transcriptional regulation in this context.141 Firstly, the AR signaling axis promotes telomerase expression but its role as a transcriptional repressor of SLX4IP has yet to be investigated.67,198

In the absence of ADT, AR may function as a repressor of SLX4IP driving AR- positive CRPC’s reliance on telomerase for telomere maintenance. However, ADT triggers a loss of telomerase expression and induction of SLX4IP, promoting ALT hallmarks as the transition to AR-negative CRPC occurs.141 Importantly, in our model of disease progression, AR expression is never lost, but only reduced, during the time course of our study. This suggests SLX4IP induction and ALTy telomere maintenance may preclude the transition to an AR-negative phenotype.141 In support of this relationship, transient SLX4IP overexpression, in which ALT hallmarks are short lived, is unable to reduce AR expression and induce

NED in AR-positive CRPC (Figure 4.1A). Additionally, SLX4IP knockdown in AR- negative CRPC does not return AR expression or reduce NED markers (Figure

4.1B and C). However at later passages, stable overexpression of SLX4IP led to

~50% reduction in AR and FOXA1 (Forkhead Box Protein A1), a pioneering transcription factor of AR known to suppress NED (Figure 4.1D).165,248 This reduction was accompanied by an elevation in the NED marker ENO2 (Figure

4.1D). This transition coincided with ALT hallmark induction.141 Taken together these data support a stepwise mechanism of ADT-triggered SLX4IP induction,

148

activation of ALTiness, followed by a transition to AR independence; however, the transcriptional brake placed on SLX4IP prior to ADT remains elusive.

Previous studies have suggested telomerase may function as a transcriptional repressor of true ALT activity but this was later disproven after continued ALT following telomerase reconstitution.129 Keeping in mind that

ALTiness and ALT differ mechanistically,86,88,140–142 telomerase as an ALT repressor was evaluated in only bona fide ALT models;129 therefore, until tested, telomerase cannot be discounted as a repressor of ALTiness via SLX4IP. AR- negative CRPC cell lines exhibit significantly higher SLX4IP expression compared to AR-positive models.141 Notably, telomerase reconstitution via TERT overexpression in AR-negative CRPC did not significantly alter SLX4IP expression

(Figure 4.1J, K, and L). These finding clearly demonstrate SLX4IP expression is not repressed by telomerase. Therefore, transcriptional programs regulating

SLX4IP-dependent ALTiness in CRPC are linked to changes initiated under androgen deprivation but independent of AR-regulated telomerase.

4.3.2 Transcriptional Programs Shared by PML and SLX4IP

In our second investigation, we identified a correlation between PML and

SLX4IP expression in CRPC models. Specifically, AR-negative cell lines demonstrated a significant elevation in the expression of both proteins.

Importantly, SLX4IP does not transcriptionally regulate PML. The correlation between SLX4IP and PML expression and their necessary roles in ALT, suggest

PML transcriptional programs may regulate SLX4IP. In parallel, the ability of long-

149

term androgen deprivation to induce SLX4IP expression as cells transition to AR- negative phenotypes suggest programs activated to bypass AR may similarly promote SLX4IP expression.141 Providing a valid starting point for investigation, these relationships narrow down the potential transcriptional regulators of SLX4IP to programs shared by PML and AR-negative progression.

PML is a direct transcriptional target of p53 and inactivating p53 mutations in prostate cancer are generally infrequent.249–252 However, p53 mutations have been associated with cases of advanced disease and disease progression.251,253,254 Additionally, our studies focus on the AR-negative DU145 and PC-3 cells lines, which demonstrate elevated SLX4IP and PML expression despite loss of functional p53.234 Therefore, p53 does not govern SLX4IP regulation in this context.

Cytokines like interferon (IFN)-γ and interleukin (IL)-6 can regulate both

PML and AR independent programs.236,237,239,255–257 Interferon release leads to the rapid induction of PML expression and the promoter of PML contains an IFN-γ activation site.256,257 Additionally, IFN-γ induces NED, the most characterized AR- negative CRPC phenotype, in prostate epithelial cells though this has not been directly tested in CRPC cells.255 Similarly, IL-6 upregulates PML expression across somatic and malignant populations through the JAK/STAT pathway.236 In CRPC,

IL-6 expression is associated with AR loss and NED.237–239 Investigation of cytokine-driven transcriptional programs may uncover those responsible for

SLX4IP expression. The regulatory programs described above provide investigational avenues for elucidating key regulators of SLX4IP.

150

4.3.3 Telomeric Recruitment via SLX4IP SUMOylation

Canonical ALT regulators, telomere-binding proteins and regulators of

ALTiness identified in our studies require SUMOylation in order to be recruited to the telomere to participate in ALT-mediated telomere maintenance.118–121 For instance, orchestration of APB formation required for ALT is initiated by recruitment of SUMOylated PML to the telomere.121 In our second study, SLX4IP was recruited to the telomere and APBs in a similar fashion but the SUMOylation status of

SLX4IP has not been definitively shown. Interestingly, the SLX4 complex functions as a SUMO E3 ligase and may therefore SUMOylate SLX4IP.235 Following

SUMOylation, SLX4IP may localize at the telomere and promote PML recruitment and APB formation through further SUMO/SIM.120,121 Interestingly, SLX4IP contains five predicted SUMOylation sites, one of which falls within the N-terminal region shown in our studies to be required for APB formation and telomere maintenance in ALT-like CRPC (Figure 4.1M).258,259 Moreover, three predicted

SIMs are found within the N-terminal region (Figure 4.1M).258 Mutational analysis of these sites will determine if SUMOylation post-translationally regulates

SLX4IP’s function at the telomere.

151

Figure 4. 1: The regulation of SLX4IP-dependent ALT-like telomere maintenance in CRPC in vitro. (A) Changes in AR, FOXA1, and ENO2 protein expression in C4-2B transiently overexpressing SLX4IP. Quantification of (A) is shown to the right. (B) Changes in AR, FOXA1 and ENO2 protein expression in PC-3 cells with stable knockdown of SLX4IP. Quantification of (B) is shown to the right. (C) Changes in AR, FOXA1 and ENO2 protein expression in DU145 cells with stable knockdown of SLX4IP. Quantification of (C) is shown to the right. (D) Changes in AR, FOXA1, and ENO2 protein expression in C4-2B and PC-3 cells stably overexpressing SLX4IP at late passages. Quantification of (D) in C4-2B cells and PC-3 cells is shown to the right. (J) Relative telomerase activity following stable telomerase reconstitution (TELO) in PC-3 cells. (K) Relative TERT and SLX4IP mRNA expression following stable telomerase reconstitution (TELO) in PC-3 cells. (L) SLX4IP protein expression following stable telomerase reconstitution (TELO) in PC-3 cells. (M) Schematic representation of predicted SUMOylation sites (SUMO) and SUMOylation- interaction motifs (SIM) in the SLX4IP sequence. Data represented as mean+SD; n=3; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

152

153

4.4 TMM Inhibition as a Last Resort

Investigations of TMMs, like those described herein, are not only geared toward understating the continuum of TMMs but to identify therapeutically advantageous targets to fill the void in highly resistant cancer types, like AR- negative CRPC.151,178 When targeted therapeutic options are exhausted cytotoxic chemotherapeutics are a last resort and generally do not dramatically improve prognoses.152,260–262 In these cases of therapeutic resistance, additional targeted regimens to preclude traditional chemotherapeutics is needed.

Targeting TMMs has always been advantageous in theory because of the necessity of TMMs for replicative immortality in cancer and a lack of TMM engagement in somatic populations.37,38,40–42 The malignant specificity that comes with targeting TMMs would theoretically avoid the majority of off-target effects associated with traditional chemotherapy. There are however several concerns regarding TMM inhibition, a number of which have come to light with the advent of telomerase inhibitors.132 If TMM inhibitors truly function through impaired telomere maintenance, and not off-target effects, then patients would require chronic treatment with these agents to allow for telomere shortening to the senescent- trigger point.5,7,11,28 Moreover, the telomere-based mechanism behind these inhibitors would only prevent further tumor growth.5,28 Similar to how there are bactericidal antibiotics, that cause bacterial cell death, versus bacteriostatic antibiotics, that prevent bacterial growth,263 TMM inhibitors would function as a cancer-static agent.5,28 Several other concerns related to telomerase- and ALT- specific inhibitors regarding unintended effects have been presented, but in the

154

context of highly resistant cancers devoid of targeted therapeutic options these tradeoffs may be worthwhile.131 Thorough characterization of TMM programming in malignancy is needed to evaluate the therapeutic implications of this cancer hallmark.

4.4.1 Dual TMM Inhibition Circumvents Resistance

There is a body of evidence supporting the inherent programming of both telomerase and ALT pathways in cancer, regardless of which TMM is actively engaged in telomere maintenance.75,123–125 Additionally, telomerase inhibition has been shown to induce ALT as a mechanism of resistance in a number of in vitro models.107,126,127,133 Though this has not been directly shown with antitelomerase agents, the risk of ALT activation as a means of resistance is a primary concern.264

Dual inhibition of both telomerase and ALT would address this concern of

TMM plasticity as a resistance mechanism.107,126,127,133 However, targeting ALT in this regimen would require studies ensuring that the candidate ALT target is equally critical across the spectrum of ALT phenotypes, including ALTiness.

Unfortunately, with the limited number of defined ALT regulators this will be challenging. ATR inhibitors have been tested for their ability to selectively target

ALT cells; however, this has yielded contradicting results.134,135 Notably, SLX4IP has been shown to be a key ALT regulator in both ALT and ALTiness despite eliciting opposite effects.140–142 Knockout of SLX4IP in ALT cells led to unregulated telomere elongation that was incompatible with cell survival.140 Knockdown of

SLX4IP in studies of ALTiness prevented ALT-like telomere maintenance

155

triggering senescence.141,142 For these reasons, SLX4IP may be advantageous for targeting the ALT spectrum but its only documented function as a scaffolding protein provides a therapeutic challenge. Additional studies are needed to determine if SLX4IP has other, targetable functions or if regulators of SLX4IP can be targeted upstream addressing the need for dual TMM inhibition regimens.

4.4.2 Androgen Deprivation Therapy, ALT Inhibition, and Senolytics

In AR-positive CRPC, androgen deprivation leads to indirect telomerase inhibition followed by the activation of SLX4IP-depednent ALTiness as a mechanism of resistance.141 SLX4IP knockdown led to the inhibition of ALT-like telomere maintenance and induction of senescence.141 If SLX4IP-targetted therapeutics can be developed, then dual TMM inhibition can be achieved to circumvent the adoption of ALT in AR-negative CRPC. The result is a TMM deficient malignant environment eventually succumbing to senescence.

As discussed previously a primary concern of TMM inhibition is that telomere-targeted therapies will not cause cancer cell death, but only prevent further cancer growth once the senescent-trigger point is reached.5,7,11,28 The application of senolytics would address this concern by promoting senescent cell clearance from the tumor, resulting in regression.265 Senolytics are a class of agents that selectively target anti-apoptotic pathways necessary for senescent cell survival, thereby triggering cell death and clearance within tumors.266,267 The application of senolytics in cancer provides an additional benefit beyond senescent cell clearance.265,268 One defining characteristic of senescent cells is a

156

senescence-associated secretory phenotype where a number of pro-malignant cytokines and factors are secreted to act on nearby malignant cells promoting cancer progression.269–272 Firstly, treatment with senolytics will prevent pro- malignant input from senescent cells propagating the highly-resistant cancer, like

AR-negative CRPC.269–272 Secondly, the addition of senolytics will address a primary concern of dual TMM inhibition by clearing senescent cells causing tumor regression.5,7,11,28 Together, this tripartite telomere-based regimen may provide a therapeutic option that precludes the use of traditional chemotherapy in highly resistant cancer types.

4.5 Concluding Remarks

The previously uncharacterized protein SLX4IP has proven to be a critical regulator of ALT-mediated telomere maintenance in a variety of ALT phenotypic presentations.140–142 This protein is necessary for APB formation which is required for ALT-mediated telomere maintenance in models of disease progression and

TMM resistance.141 The key role of SLX4IP across the ALT spectrum makes it an advantageous therapeutic target as an ALT inhibitor;140–142 however, its only function as a scaffold protein makes this approach challenging.100,143,144 Therefore, additional efforts are needed to further characterize the functions of SLX4IP and upstream SLX4IP regulators to uncover targetable options within this SLX4IP- dependent mechanism. Uncovering a novel ALT inhibitor will remedy a multitude of concerns related to TMM inhibition as a therapeutic avenue in cancer.

Addressing these concerns and developing regimens geared toward TMM

157

inhibition may provide treatment options for difficult to treat cancer types where all targeted therapeutic opportunities have been exhausted, like in AR-negative

CRPC described herein.

158

REFERENCES 1. Wyatt, H. D. M., West, S. C. & Beattie, T. L. InTERTpreting telomerase

structure and function. Nucleic Acids Res. 38, (2010).

2. Blackburn, E. H. Structure and function of telomeres. Nature (1991)

doi:10.1038/350569a0.

3. Blackburn, E. H. Telomeres: No end in sight. Cell vol. 77 621–623 (1994).

4. Moyzis, R. K. et al. A highly conserved repetitive DNA sequence,

(TTAGGG)(n), present at the telomeres of human chromosomes. Proc. Natl.

Acad. Sci. U. S. A. (1988) doi:10.1073/pnas.85.18.6622.

5. Harley, C. B., Futcher, A. B. & Greider, C. W. Telomeres shorten during

ageing of human fibroblasts. Nature (1990) doi:10.1038/345458a0.

6. Olovnikov, A. M. A theory of marginotomy. The incomplete copying of

template margin in enzymic synthesis of polynucleotides and biological

significance of the phenomenon. J. Theor. Biol. (1973) doi:10.1016/0022-

5193(73)90198-7.

7. De Lange, T. How telomeres solve the end-protection problem. Science

(2009) doi:10.1126/science.1170633.

8. De Lange, T. Shelterin: The protein complex that shapes and safeguards

human telomeres. Genes and Development (2005)

doi:10.1101/gad.1346005.

9. Denchi, E. L. & De Lange, T. Protection of telomeres through independent

control of ATM and ATR by TRF2 and POT1. Nature 448, (2007).

10. Palm, W. & De Lange, T. How shelterin protects mammalian telomeres.

159

Annual Review of Genetics vol. 42 (2008).

11. Sfeir, A. & De Lange, T. Removal of shelterin reveals the telomere end-

protection problem. Science (80-. ). 336, (2012).

12. Li, B., Oestreich, S. & De Lange, T. Identification of human Rap1:

Implications for telomere evolution. Cell 101, (2000).

13. Takai, K. K., Kibe, T., Donigian, J. R., Frescas, D. & de Lange, T. Telomere

Protection by TPP1/POT1 Requires Tethering to TIN2. Mol. Cell 44, (2011).

14. Blackburn, E. H. Telomere states and cell fates. Nature vol. 408 (2000).

15. Blackburn, E. H. Switching and signaling at the telomere. Cell vol. 106

(2001).

16. Cesare, A. J. et al. Spontaneous occurrence of telomeric DNA damage

response in the absence of chromosome fusions. Nat. Struct. Mol. Biol.

(2009) doi:10.1038/nsmb.1725.

17. Van Steensel, B., Smogorzewska, A. & De Lange, T. TRF2 protects human

telomeres from end-to-end fusions. Cell 92, (1998).

18. Griffith, J. D. et al. Mammalian telomeres end in a large duplex loop. Cell 97,

(1999).

19. Bianchi, A. et al. TRF1 binds a bipartite telomeric site with extreme spatial

flexibility. EMBO J. 18, (1999).

20. Uringa, E. J. et al. RTEL1 contributes to DNA replication and repair and

telomere maintenance. Mol. Biol. Cell 23, (2012).

21. Vannier, J. B., Pavicic-Kaltenbrunner, V., Petalcorin, M. I. R., Ding, H. &

Boulton, S. J. RTEL1 dismantles T loops and counteracts telomeric G4-DNA

160

to maintain telomere integrity. Cell 149, (2012).

22. Pickett, H. A., Cesare, A. J., Johnston, R. L., Neumann, A. A. & Reddel, R.

R. Control of telomere length by a trimming mechanism that involves

generation of t-circles. EMBO J. 28, (2009).

23. Wu, P., Takai, H. & De Lange, T. Telomeric 3′ overhangs derive from

resection by Exo1 and apollo and fill-in by POT1b-associated CST. Cell 150,

(2012).

24. Surovtseva, Y. V. et al. Conserved Telomere Maintenance Component 1

Interacts with STN1 and Maintains Chromosome Ends in Higher Eukaryotes.

Mol. Cell 36, (2009).

25. Wilson, J. S. J. et al. Localization-dependent and -independent roles of SLX4

in regulating telomeres. Cell Rep. (2013) doi:10.1016/j.celrep.2013.07.033.

26. Takai, H., Smogorzewska, A. & De Lange, T. DNA damage foci at

dysfunctional telomeres. Curr. Biol. 13, (2003).

27. Cesare, A. J., Heaphy, C. M. & O’Sullivan, R. J. Visualization of telomere

integrity and function in vitro and in vivo using immunofluorescence

techniques. Curr. Protoc. Cytom. 2015, (2015).

28. Hayflick, L. The illusion of cell immortality. British Journal of Cancer (2000)

doi:10.1054/bjoc.2000.1296.

29. MacNeil, D. E., Bensoussan, H. J. & Autexier, C. Telomerase regulation from

beginning to the end. Genes vol. 7 (2016).

30. Cesare, A. J., Hayashi, M. T., Crabbe, L. & Karlseder, J. The Telomere

deprotection response is functionally distinct from the Genomic DNA

161

damage response. Mol. Cell 51, (2013).

31. D’Adda Di Fagagna, F. et al. A DNA damage checkpoint response in

telomere-initiated senescence. Nature (2003) doi:10.1038/nature02118.

32. Chin, L. et al. p53 deficiency rescues the adverse effects of telomere loss

and cooperates with telomere dysfunction to accelerate carcinogenesis. Cell

97, (1999).

33. Hayashi, M. T., Cesare, A. J., Rivera, T. & Karlseder, J. Cell death during

crisis is mediated by mitotic telomere deprotection. Nature 522, (2015).

34. Chang, S. Chromosome ends teach unexpected lessons on DNA damage

signalling. EMBO J. 31, (2012).

35. Maciejowski, J., Li, Y., Bosco, N., Campbell, P. J. & De Lange, T.

Chromothripsis and Kataegis Induced by Telomere Crisis. Cell 163, (2015).

36. Von Morgen, P. & Maciejowski, J. The ins and outs of telomere crisis in

cancer. Genome Medicine vol. 10 (2018).

37. Shay, J. W. & Bacchetti, S. A survey of telomerase activity in human cancer.

Eur. J. Cancer Part A (1997) doi:10.1016/S0959-8049(97)00062-2.

38. Dilley, R. L. & Greenberg, R. A. ALTernative Telomere Maintenance and

Cancer. Trends in Cancer (2015) doi:10.1016/j.trecan.2015.07.007.

39. Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: The next generation.

Cell vol. 144 (2011).

40. Greider, C. W. & Blackburn, E. H. Identification of a specific telomere

terminal transferase activity in tetrahymena extracts. Cell (1985)

doi:10.1016/0092-8674(85)90170-9.

162

41. Kim, N. W. et al. Specific association of human telomerase activity with

immortal cells and cancer. Science (80-. ). (1994)

doi:10.1126/science.7605428.

42. Bryan, T. M., Englezou, A., Gupta, J., Bacchetti, S. & Reddel, R. R. Telomere

elongation in immortal human cells without detectable telomerase activity.

EMBO J. (1995) doi:10.1002/j.1460-2075.1995.tb00098.x.

43. Hiyama, E. & Hiyama, K. Telomere and telomerase in stem cells. British

Journal of Cancer vol. 96 (2007).

44. Hiyama, K. et al. Activation of telomerase in human lymphocytes and

hematopoietic progenitor cells. J. Immunol. 155, (1995).

45. Tauchi, T. et al. Telomerase Inhibition with a Novel G-Quadruplex-

Interactive Agent, Telomestatin: In Vitro and In Vivo Studies in Acute

Leukemia. Blood 104, (2004).

46. Britt-Compton, B., Capper, R., Rowson, J. & Baird, D. M. Short telomeres

are preferentially elongated by telomerase in human cells. FEBS Lett. 583,

(2009).

47. Abreu, E. et al. TIN2-Tethered TPP1 Recruits Human Telomerase to

Telomeres In Vivo. Mol. Cell. Biol. 30, (2010).

48. Zhong, F. L. et al. TPP1 OB-fold domain controls telomere maintenance by

recruiting telomerase to chromosome ends. Cell 150, (2012).

49. Nandakumar, J. et al. The TEL patch of telomere protein TPP1 mediates

telomerase recruitment and processivity. Nature 492, (2012).

50. Parks, J. W. & Stone, M. D. Coordinated DNA dynamics during the human

163

telomerase catalytic cycle. Nat. Commun. 5, (2014).

51. Cohen, S. B. et al. Protein composition of catalytically active human

telomerase from immortal cells. Science (80-. ). 315, (2007).

52. Vulliamy, T. J. et al. Mutations in dyskeratosis congenita: Their impact on

telomere length and the diversity of clinical presentation. Blood 107, (2006).

53. Robart, A. R. & Collins, K. Investigation of human telomerase holoenzyme

assembly, activity, and processivity using disease-linked subunit variants. J.

Biol. Chem. 285, (2010).

54. Trahan, C., Martel, C. & Dragon, F. Effects of dyskeratosis congenita

mutations in dyskerin, NHP2 and NOP10 on assembly of H/ACA pre-RNPs.

Hum. Mol. Genet. 19, (2010).

55. Zhong, F. et al. Disruption of telomerase trafficking by TCAB1 mutation

causes dyskeratosis congenita. Genes Dev. 25, (2011).

56. Tardat, M. & Déjardin, J. Telomere chromatin establishment and its

maintenance during mammalian development. Chromosoma vol. 127

(2018).

57. Feng, J. et al. The RNA component of human telomerase. Science (80-. ).

269, (1995).

58. Zhang, A. et al. Frequent amplification of the telomerase reverse

transcriptase gene in human tumors. Cancer Res. 60, (2000).

59. Zhao, Y., Wang, S., Popova, E. Y., Grigoryev, S. A. & Zhu, J.

Rearrangement of upstream sequences of the hTERT gene during cellular

immortalization. Genes Chromosom. Cancer 48, (2009).

164

60. Wong, M. S., Wright, W. E. & Shay, J. W. Alternative splicing regulation of

telomerase: A new paradigm? Trends in Genetics vol. 30 (2014).

61. Liu, T., Yuan, X. & Xu, D. Cancer-specific telomerase reverse transcriptase

(Tert) promoter mutations: Biological and clinical implications. Genes vol. 7

(2016).

62. Lee, D. D. et al. DNA hypermethylation within TERT promoter upregulates

TERT expression in cancer. J. Clin. Invest. 129, (2019).

63. Farooqi, A. A., Mansoor, Q., Alaaeddine, N. & Xu, B. MicroRNA regulation

of telomerase reverse transcriptase (TERT): Micro machines pull strings of

papier-mâché puppets. International Journal of Molecular Sciences vol. 19

(2018).

64. Wu, K. J. et al. Direct activation of TERT transcription by c-MYC. Nat. Genet.

21, (1999).

65. Boggess, J. F., Zhou, C., Bae-Jump, V. L., Gehrig, P. A. & Whang, Y. E.

Estrogen-receptor-dependent regulation of telomerase activity in human

endometrial cancer cell lines. Gynecol. Oncol. 103, (2006).

66. Kyo, S. et al. Estrogen activates telomerase. Cancer Res. 59, (1999).

67. Moehren, U. et al. Wild-type but not mutant androgen receptor inhibits

expression of the hTERT telomerase subunit: A novel role of AR mutation

for prostate cancer development. FASEB J. (2008) doi:10.1096/fj.07-

9360com.

68. Ho, Y. & Dehm, S. M. Androgen receptor rearrangement and splicing

variants in resistance to endocrine therapies in prostate cancer.

165

Endocrinology (2017) doi:10.1210/en.2017-00109.

69. Aravindan, N., Veeraraghavan, J., Madhusoodhanan, R., Herman, T. S. &

Natarajan, M. Curcumin regulates low-linear energy transfer γ-radiation-

induced NFκB-dependent telomerase activity in human neuroblastoma cells.

Int. J. Radiat. Oncol. Biol. Phys. 79, (2011).

70. Gizard, F. et al. Telomerase activation in atherosclerosis and induction of

telomerase reverse transcriptase expression by inflammatory stimuli in

macrophages. Arterioscler. Thromb. Vasc. Biol. 31, (2011).

71. Chung, S. S., Aroh, C. & Vadgama, J. V. Constitutive activation of STAT3

signaling regulates hTERT and promotes stem cell-like traits in human

breast cancer cells. PLoS One 8, (2013).

72. Yamada, O. et al. Activation of STAT5 confers imatinib resistance on

leukemic cells through the transcription of TERT and MDR1. Cell. Signal. 23,

(2011).

73. Ulaner, G. A. et al. Absence of a telomere maintenance mechanism as a

favorable prognostic factor in patients with osteosarcoma. Cancer Res. 63,

(2003).

74. Montgomery, E., Argani, P., Hicks, J. L., DeMarzo, A. M. & Meeker, A. K.

Telomere Lengths of Translocation-Associated and Nontranslocation-

Associated Sarcomas Differ Dramatically. Am. J. Pathol. 164, (2004).

75. Heaphy, C. M. et al. Prevalence of the alternative lengthening of telomeres

telomere maintenance mechanism in human cancer subtypes. Am. J.

Pathol. (2011) doi:10.1016/j.ajpath.2011.06.018.

166

76. Hakin-Smith, V. et al. Alternative lengthening of telomeres and survival in

patients with glioblastoma multiforme. Lancet (2003) doi:10.1016/S0140-

6736(03)12681-5.

77. Londoño-Vallejo, J. A., Der-Sarkissian, H., Cazes, L., Bacchetti, S. &

Reddel, R. R. Alternative Lengthening of Telomeres Is Characterized by

High Rates of Telomeric Exchange. Cancer Res. (2004) doi:10.1158/0008-

5472.CAN-03-4035.

78. Dunham, M. A., Neumann, A. A., Fasching, C. L. & Reddel, R. R. Telomere

maintenance by recombination in human cells. Nat. Genet. (2000)

doi:10.1038/82586.

79. Cesare, A. J. & Reddel, R. R. Alternative lengthening of telomeres: Models,

mechanisms and implications. Nature Reviews Genetics (2010)

doi:10.1038/nrg2763.

80. Sobinoff, A. P. & Pickett, H. A. Alternative Lengthening of Telomeres: DNA

Repair Pathways Converge. Trends in Genetics vol. 33 (2017).

81. Bosco, N. & De Lange, T. A TRF1-controlled common fragile site containing

interstitial telomeric sequences. Chromosoma 121, (2012).

82. Sfeir, A. et al. Mammalian Telomeres Resemble Fragile Sites and Require

TRF1 for Efficient Replication. Cell 138, (2009).

83. Pan, X. et al. FANCM, BRCA1, and BLM cooperatively resolve the

replication stress at the ALT telomeres. Proc. Natl. Acad. Sci. U. S. A. 114,

(2017).

84. Min, J., Wright, W. E. & Shay, J. W. Alternative Lengthening of Telomeres

167

Mediated by Mitotic DNA Synthesis Engages Break-Induced Replication

Processes. Mol. Cell. Biol. 37, (2017).

85. O’Sullivan, R. J. & Almouzni, G. Assembly of telomeric chromatin to create

ALTernative endings. Trends in Cell Biology vol. 24 (2014).

86. Episkopou, H. et al. Alternative Lengthening of Telomeres is characterized

by reduced compaction of telomeric chromatin. Nucleic Acids Res. 42,

(2014).

87. Aymard, F. et al. Transcriptionally active chromatin recruits homologous

recombination at DNA double-strand breaks. Nat. Struct. Mol. Biol. 21,

(2014).

88. Amorim, J. P., Santos, G., Vinagre, J. & Soares, P. The role of ATRX in the

alternative lengthening of telomeres (ALT) phenotype. Genes (2016)

doi:10.3390/genes7090066.

89. Napier, C. E. et al. ATRX represses alternative lengthening of telomeres.

Oncotarget (2015) doi:10.18632/oncotarget.3846.

90. De Wilde, R. F. et al. Loss of ATRX or DAXX expression and concomitant

acquisition of the alternative lengthening of telomeres phenotype are late

events in a small subset of MEN-1 syndrome pancreatic neuroendocrine

tumors. Mod. Pathol. 25, (2012).

91. Gaillard, H., García-Muse, T. & Aguilera, A. Replication stress and cancer.

Nature Reviews Cancer vol. 15 (2015).

92. Daley, J. M., Niu, H., Miller, A. S. & Sung, P. Biochemical mechanism of DSB

end resection and its regulation. DNA Repair (Amst). 32, (2015).

168

93. Nimonkar, A. V. et al. BLM-DNA2-RPA-MRN and EXO1-BLM-RPA-MRN

constitute two DNA end resection machineries for human DNA break repair.

Genes Dev. 25, (2011).

94. Davis, A. J. & Chen, D. J. DNA double strand break repair via non-

homologous end-joining. Translational Cancer Research vol. 2 (2013).

95. Bae, N. S. & Baumann, P. A RAP1/TRF2 Complex Inhibits Nonhomologous

End-Joining at Human Telomeric DNA Ends. Mol. Cell 26, (2007).

96. Cho, N. W., Dilley, R. L., Lampson, M. A. & Greenberg, R. A.

Interchromosomal homology searches drive directional ALT telomere

movement and synapsis. Cell 159, (2014).

97. Dilley, R. L. et al. Break-induced telomere synthesis underlies alternative

telomere maintenance. Nature (2016) doi:10.1038/nature20099.

98. Roumelioti, F. et al. Alternative lengthening of human telomeres is a

conservative DNA replication process with features of break‐ induced

replication . EMBO Rep. 17, (2016).

99. Wu, L. & Hickson, I. O. The Bloom’s syndrome helicase suppresses crossing

over during homologous recombination. Nature 426, (2003).

100. Svendsen, J. M. et al. Mammalian BTBD12/SLX4 Assembles A Holliday

Junction Resolvase and Is Required for DNA Repair. Cell (2009)

doi:10.1016/j.cell.2009.06.030.

101. Sarbajna, S., Davies, D. & West, S. C. Roles of SLX1-SLX4, MUS81-EME1,

and GEN1 in avoiding genome instability and mitotic catastrophe. Genes

Dev. 28, (2014).

169

102. Cesare, A. J. & Griffith, J. D. Telomeric DNA in ALT Cells Is Characterized

by Free Telomeric Circles and Heterogeneous t-Loops. Mol. Cell. Biol.

(2004) doi:10.1128/mcb.24.22.9948-9957.2004.

103. Tokutake, Y. et al. Extra-chromosomal telomere repeat DNA in telomerase-

negative immortalized cell lines. Biochem. Biophys. Res. Commun. 247,

(1998).

104. Yeager, T. R. et al. Telomerase-negative immortalized human cells contain

a novel type of promyelocytic leukemia (PML) body. Cancer Res. (1999).

105. Henson, J. D. et al. The C-Circle Assay for alternative-lengthening-of-

telomeres activity. Methods (2017) doi:10.1016/j.ymeth.2016.08.016.

106. Tomaska, L., Nosek, J., Kramara, J. & Griffith, J. D. Telomeric circles:

Universal players in telomere maintenance. Nature Structural and Molecular

Biology vol. 16 (2009).

107. Bechter, O. E., Zou, Y., Walker, W., Wright, W. E. & Shay, J. W. Telomeric

recombination in mismatch repair deficient human colon cancer cells after

telomerase inhibition. Cancer Res. (2004) doi:10.1158/0008-5472.CAN-04-

0323.

108. Sobinoff, A. P. et al. BLM and SLX4 play opposing roles in recombination‐

dependent replication at human telomeres. EMBO J. 36, (2017).

109. Wilson, D. M. & Thompson, L. H. Molecular mechanisms of sister-chromatid

exchange. Mutation Research - Fundamental and Molecular Mechanisms of

Mutagenesis vol. 616 (2007).

110. Blagoev, K. B. & Goodwin, E. H. Telomere exchange and asymmetric

170

segregation of chromosomes can account for the unlimited proliferative

potential of ALT cell populations. DNA Repair (Amst). 7, (2008).

111. Costantino, L. et al. Break-induced replication repair of damaged forks

induces genomic duplications in human cells. Science (80-. ). 343, (2014).

112. Brosnan-Cashman, J. A. et al. ATRX loss induces multiple hallmarks of the

alternative lengthening of telomeres (ALT) phenotype in human glioma cell

lines in a cell line-specific manner. PLoS One (2018)

doi:10.1371/journal.pone.0204159.

113. Henson, J. D. et al. DNA C-circles are specific and quantifiable markers of

alternative- lengthening-of-telomeres activity. Nat. Biotechnol. 27, (2009).

114. Draskovic, I. et al. Probing PML body function in ALT cells reveals

spatiotemporal requirements for telomere recombination. Proc. Natl. Acad.

Sci. U. S. A. (2009) doi:10.1073/pnas.0907689106.

115. Chung, I., Osterwald, S., Deeg, K. I. & Rippe, K. PML body meets telomere:

The beginning of an ALTernate ending? Nucleus (United States) (2012)

doi:10.4161/nucl.20326.

116. Wu, G., Lee, W. H. & Chen, P. L. NBS1 and TRF1 colocalize at

promyelocytic leukemia bodies during late S/G2 phases in immortalized

telomerase-negative cells. Implication of NBS1 in alternative lengthening of

telomeres. J. Biol. Chem. 275, (2000).

117. Nabetani, A., Yokoyama, O. & Ishikawa, F. Localization of hRad9, hHus1,

hRad1, and hRad17 and caffeine-sensitive DNA replication at the alternative

lengthening of telomeres-associated promyelocytic leukemia body. J. Biol.

171

Chem. 279, (2004).

118. Osterwald, S. et al. PML induces compaction, TRF2 depletion and DNA

damage signaling at telomeres and promotes their alternative lengthening.

J. Cell Sci. 128, (2015).

119. Shen, T. H., Lin, H. K., Scaglioni, P. P., Yung, T. M. & Pandolfi, P. P. The

Mechanisms of PML-Nuclear Body Formation. Mol. Cell 24, (2006).

120. Banani, S. F. et al. Compositional Control of Phase-Separated Cellular

Bodies. Cell 166, (2016).

121. Min, J., Wright, W. E. & Shay, J. W. Clustered telomeres in phase-separated

nuclear condensates engage mitotic DNA synthesis through BLM and

RAD52. Genes Dev. 33, (2019).

122. Brault, M. E. & Autexier, C. Telomeric recombination induced by

dysfunctional telomeres. Mol. Biol. Cell (2011) doi:10.1091/mbc.E10-02-

0173.

123. De Vitis, M., Berardinelli, F. & Sgura, A. Telomere length maintenance in

cancer: At the crossroad between telomerase and alternative lengthening of

telomeres (ALT). International Journal of Molecular Sciences (2018)

doi:10.3390/ijms19020606.

124. Xu, B., Peng, M. & Song, Q. The co-expression of telomerase and ALT

pathway in human breast cancer tissues. Tumor Biol. (2014)

doi:10.1007/s13277-013-1534-0.

125. Gocha, A. R. S., Nuovo, G., Iwenofu, O. H. & Groden, J. Human sarcomas

are mosaic for telomerase-dependent and telomerase- independent

172

telomere maintenance mechanisms: Implications for telomere-based

therapies. Am. J. Pathol. 182, (2013).

126. Xue, Y. et al. Twisted epithelial-to-mesenchymal transition promotes

progression of surviving bladder cancer T24 cells with hTERT-dysfunction.

PLoS One (2011) doi:10.1371/journal.pone.0027748.

127. Hu, J. et al. Antitelomerase therapy provokes ALT and mitochondrial

adaptive mechanisms in cancer. Cell (2012) doi:10.1016/j.cell.2011.12.028.

128. Cerone, M. A. Telomere maintenance by telomerase and by recombination

can coexist in human cells. Hum. Mol. Genet. (2001)

doi:10.1093/hmg/10.18.1945.

129. Perrem, K., Colgin, L. M., Neumann, A. A., Yeager, T. R. & Reddel, R. R.

Coexistence of Alternative Lengthening of Telomeres and Telomerase in

hTERT-Transfected GM847 Cells. Mol. Cell. Biol. (2001)

doi:10.1128/mcb.21.12.3862-3875.2001.

130. Grobelny, J. V. Effects of reconstitution of telomerase activity on telomere

maintenance by the alternative lengthening of telomeres (ALT) pathway.

Hum. Mol. Genet. (2001) doi:10.1093/hmg/10.18.1953.

131. Sugarman, E. T., Zhang, G. & Shay, J. W. In perspective: An update on

telomere targeting in cancer. Mol. Carcinog. 58, (2019).

132. Shay, J. W. & Wright, W. E. Telomerase therapeutics for cancer: Challenges

and new directions. Nature Reviews Drug Discovery vol. 5 (2006).

133. Queisser, A., Heeg, S., Thaler, M., von Werder, A. & Opitz, O. G. Inhibition

of telomerase induces alternative lengthening of telomeres during human

173

esophageal carcinogenesis. Cancer Genet. (2013)

doi:10.1016/j.cancergen.2013.10.001.

134. Flynn, R. L. et al. Alternative lengthening of telomeres renders cancer cells

hypersensitive to ATR inhibitors. Science (80-. ). 347, (2015).

135. Deeg, K. I., Chung, I., Bauer, C. & Rippe, K. Cancer cells with alternative

lengthening of telomeres do not display a general hypersensitivity to ATR

inhibition. Front. Oncol. 6, (2016).

136. Bryan, T. M., Englezou, A., Dalla-Pozza, L., Dunham, M. A. & Reddel, R. R.

Evidence for an alternative mechanism for maintaining telomere length in

human tumors and tumor-derived cell lines. Nat. Med. (1997)

doi:10.1038/nm1197-1271.

137. Pezzolo, A. et al. Intratumoral diversity of telomere length in individual

neuroblastoma tumors. Oncotarget 6, (2015).

138. Boardman, L. A. et al. Correlation of chromosomal instability, telomere

length and telomere maintenance in microsatellite stable rectal cancer: A

molecular subclass of rectal cancer. PLoS One 8, (2013).

139. Omori, Y. et al. Alternative lengthening of telomeres frequently occurs in

mismatch repair system-deficient gastric carcinoma. Cancer Sci. (2009)

doi:10.1111/j.1349-7006.2008.01063.x.

140. Panier, S. et al. SLX4IP Antagonizes Promiscuous BLM Activity during ALT

Maintenance. Mol. Cell (2019) doi:10.1016/j.molcel.2019.07.010.

141. Mangosh, T. L., Awadallah, W. N., Grabowska, M. M. & Taylor, D. J. SLX4IP

promotes telomere maintenance in androgen receptor–independent

174

castration-resistant prostate cancer through ALT-like telomeric PML

localization. Mol. Cancer Res. (2021) doi:10.1158/1541-7786.MCR-20-

0314.

142. Robinson, N. J. et al. SLX4IP and telomere dynamics dictate breast cancer

metastasis and therapeutic responsiveness. Life Sci. Alliance (2020)

doi:10.26508/LSA.201900427.

143. Wan, B. et al. SLX4 assembles a telomere maintenance toolkit by bridging

multiple endonucleases with telomeres. Cell Rep. (2013)

doi:10.1016/j.celrep.2013.08.017.

144. Zhang, H. et al. SLX4IP acts with SLX4 and XPF-ERCC1 to promote

interstrand crosslink repair. Nucleic Acids Res. (2019)

doi:10.1093/nar/gkz769.

145. Siegel, R. L., Miller, K. D. & Jemal, A. Cancer statistics, 2020. CA. Cancer

J. Clin. 70, (2020).

146. Mohler, J. L. et al. Prostate cancer, version 2.2019. JNCCN J. Natl. Compr.

Cancer Netw. 17, 479–505 (2019).

147. Mongiat-Artus, P. & Teillac, P. Role of luteinising hormone releasing

hormone (LHRH) agonists and hormonal treatment in the management of

prostate cancer. Eur. Urol. Suppl. 4, (2005).

148. Rove, K. O. & Crawford, E. D. Traditional androgen ablation approaches to

advanced prostate cancer: New insights. Can. J. Urol. 21, (2014).

149. Perlmutter, M. A. & Lepor, H. Androgen deprivation therapy in the treatment

of advanced prostate cancer. Rev. Urol. 9 Suppl 1, (2007).

175

150. Salonen, A. J. et al. The finnprostate study VII: Intermittent versus

continuous androgen deprivation in patients with advanced prostate cancer.

J. Urol. (2012) doi:10.1016/j.juro.2012.01.122.

151. Beltran, H. et al. Aggressive variants of castration-resistant prostate cancer.

Clinical Cancer Research (2014) doi:10.1158/1078-0432.CCR-13-3309.

152. Aggarwal, R. et al. Clinical and genomic characterization of treatment-

emergent small-cell neuroendocrine prostate cancer: A multi-institutional

prospective study. J. Clin. Oncol. (2018) doi:10.1200/JCO.2017.77.6880.

153. Beltran, H. et al. Divergent clonal evolution of castration-resistant

neuroendocrine prostate cancer. Nat. Med. (2016) doi:10.1038/nm.4045.

154. Abrahamsson, A. Neuroendocrine differentiation in prostatic carcinoma.

Prostate (1999) doi:10.1002/(SICI)1097-0045(19990501)39:2<135::AID-

PROS9>3.0.CO;2-S.

155. Isaacs, J. T. & Isaacs, W. B. Androgen receptor outwits prostate cancer

drugs. Nature Medicine vol. 10 (2004).

156. Scher, H. I. & Sawyers, C. L. Biology of progressive, castration-resistant

prostate cancer: Directed therapies targeting the androgen-receptor

signaling axis. Journal of Clinical Oncology vol. 23 (2005).

157. Cai, C. & Balk, S. P. Intratumoral androgen biosynthesis in prostate cancer

pathogenesis and response to therapy. Endocrine-Related Cancer vol. 18

(2011).

158. Sun, S. et al. Castration resistance in human prostate cancer is conferred by

a frequently occurring androgen receptor splice variant. J. Clin. Invest.

176

(2010) doi:10.1172/JCI41824.

159. Zhang, X. et al. Androgen receptor variants occur frequently in castration

resistant prostate cancer metastases. PLoS One (2011)

doi:10.1371/journal.pone.0027970.

160. Visakorpi, T. et al. In vivo amplification of the androgen receptor gene and

progression of human prostate cancer. Nat. Genet. 9, (1995).

161. Koivisto, P. et al. Androgen receptor gene amplification: A possible

molecular mechanism for androgen deprivation therapy failure in prostate

cancer. Cancer Res. 57, (1997).

162. Kawata, H. et al. Prolonged treatment with bicalutamide induces androgen

receptor overexpression and androgen hypersensitivity. Prostate 70, (2010).

163. Waltering, K. K. et al. Increased expression of androgen receptor sensitizes

prostate cancer cells to low levels of androgens. Cancer Res. 69, (2009).

164. Koochekpour, S. Androgen receptor signaling and mutations in prostate

cancer. Asian Journal of Andrology vol. 12 (2010).

165. Grasso, C. S. et al. The mutational landscape of lethal castration-resistant

prostate cancer. Nature (2012) doi:10.1038/nature11125.

166. Taylor, B. S. et al. Integrative Genomic Profiling of Human Prostate Cancer.

Cancer Cell 18, (2010).

167. Robinson, D. et al. Integrative clinical genomics of advanced prostate

cancer. Cell (2015) doi:10.1016/j.cell.2015.05.001.

168. Hara, T. et al. Novel mutations of androgen receptor: A possible mechanism

of bicalutamide withdrawal syndrome. Cancer Res. 63, (2003).

177

169. Sun, C. et al. Androgen receptor mutation (T877A) promotes prostate cancer

cell growth and cell survival. Oncogene 25, (2006).

170. Dehm, S. M., Schmidt, L. J., Heemers, H. V., Vessella, R. L. & Tindall, D. J.

Splicing of a novel androgen receptor exon generates a constitutively active

androgen receptor that mediates prostate cancer therapy resistance. Cancer

Res. 68, (2008).

171. Hu, R. et al. Ligand-independent androgen receptor variants derived from

splicing of cryptic exons signify hormone-refractory prostate cancer. Cancer

Res. 69, (2009).

172. Whitworth, H. et al. Identification of kinases regulating prostate cancer cell

growth using an RNAi phenotypic screen. PLoS One 7, (2012).

173. Arora, V. K. et al. XGlucocorticoid receptor confers resistance to

antiandrogens by bypassing androgen receptor blockade. Cell 155, (2013).

174. Saraon, P., Jarvi, K. & Diamandis, E. P. Molecular alterations during

progression of prostate cancer to androgen independence. Clinical

Chemistry vol. 57 (2011).

175. Fizazi, K. et al. Abiraterone acetate for treatment of metastatic castration-

resistant prostate cancer: Final overall survival analysis of the COU-AA-301

randomised, double-blind, placebo-controlled phase 3 study. Lancet Oncol.

(2012) doi:10.1016/S1470-2045(12)70379-0.

176. Scher, H. I. et al. Antitumour activity of MDV3100 in castration-resistant

prostate cancer: A phase 1-2 study. Lancet (2010) doi:10.1016/S0140-

6736(10)60172-9.

178

177. Tran, C. et al. Development of a second-generation antiandrogen for

treatment of advanced prostate cancer. Science (80-. ). 324, (2009).

178. Abida, W. et al. Genomic correlates of clinical outcome in advanced prostate

cancer. Proc. Natl. Acad. Sci. U. S. A. (2019)

doi:10.1073/pnas.1902651116.

179. Colombel, M. et al. Detection of the apoptosis-suppressing oncoprotein bcl-

2 in hormone- refractory human prostate cancers. Am. J. Pathol. 143,

(1993).

180. Furuya, Y., Krajewski, S., Epstein, J. I., Reed, J. C. & Isaacs, J. T.

Expression of bcl-2 and the progression of human and rodent prostatic

cancers. Clin. Cancer Res. 2, (1996).

181. Um, H. D. Bcl-2 family proteins as regulators of cancer cell invasion and

metastasis: A review focusing on mitochondrial respiration and reactive

oxygen species. Oncotarget 7, (2016).

182. Liu, A. Y., Corey, E., Bladou, F., Lange, P. H. & Vessella, R. L. Prostatic cell

lineage markers: Emergence of BCL2+ cells of human prostate cancer

xenograft LuCaP 23 following castration. Int. J. Cancer 65, (1996).

183. Gleave, M. et al. Progression to androgen independence is delayed by

adjuvant treatment with antisense Bcl-2 oligodeoxynucleotides after

castration in the LNCaP prostate tumor model. Clin. Cancer Res. 5, (1999).

184. Isikbay, M. et al. Glucocorticoid Receptor Activity Contributes to Resistance

to Androgen-Targeted Therapy in Prostate Cancer. Horm. Cancer 5, (2014).

185. Sahu, B. et al. FoxA1 specifies unique androgen and glucocorticoid receptor

179

binding events in prostate cancer cells. Cancer Res. 73, (2013).

186. Dean, M., Fojo, T. & Bates, S. Tumour stem cells and drug resistance.

Nature Reviews Cancer vol. 5 (2005).

187. Patrawala, L. et al. Highly purified CD44+ prostate cancer cells from

xenograft human tumors are enriched in tumorigenic and metastatic

progenitor cells. Oncogene 25, (2006).

188. Collins, A. T., Berry, P. A., Hyde, C., Stower, M. J. & Maitland, N. J.

Prospective identification of tumorigenic prostate cancer stem cells. Cancer

Res. 65, (2005).

189. Zhang, D. et al. Stem cell and neurogenic gene-expression profiles link

prostate basal cells to aggressive prostate cancer. Nat. Commun. 7, (2016).

190. Anthony di Sant’agnese, P. & de Mesy Jensen, K. L. Neuroendocrine

differentiation in prostatic carcinoma. Hum. Pathol. (1987)

doi:10.1016/S0046-8177(87)80060-6.

191. Humphrey, P. A. Histological variants of prostatic carcinoma and their

significance. Histopathology vol. 60 (2012).

192. Schwartz, L. H. et al. Small cell and anaplastic prostate cancer: Correlation

between CT findings and prostate-specific antigen level. Radiology 208,

(1998).

193. Lipianskaya, J. et al. Androgen-deprivation therapy-induced aggressive

prostate cancer with neuroendocrine differentiation. Asian Journal of

Andrology (2014) doi:10.4103/1008-682X.123669.

194. Terry, S. & Beltran, H. The many faces of neuroendocrine differentiation in

180

prostate cancer progression. Frontiers in Oncology (2014)

doi:10.3389/fonc.2014.00060.

195. Hirano, D., Okada, Y., Minei, S., Takimoto, Y. & Nemoto, N. Neuroendocrine

Differentiation in Hormone Refractory Prostate Cancer Following Androgen

Deprivation Therapy. Eur. Urol. (2004) doi:10.1016/j.eururo.2003.11.032.

196. Sommerfeld, H. J. et al. Telomerase activity: A prevalent marker of malignant

human prostate tissue. Cancer Res. (1996).

197. Guo, C., Armbruster, B. N., Price, D. T. & Counter, C. M. In vivo regulation

of hTERT expression and telomerase activity by androgen. J. Urol. (2003)

doi:10.1097/01.ju.0000074653.22766.c8.

198. Liu, S. et al. Telomerase as an important target of androgen signaling

blockade for prostate cancer treatment. Mol. Cancer Ther. (2010)

doi:10.1158/1535-7163.MCT-09-0924.

199. Haffner, M. C. et al. Tracking the clonal origin of lethal prostate cancer. J.

Clin. Invest. (2013) doi:10.1172/JCI70354.

200. Graham, M. K. et al. Functional loss of ATRX and TERC activates Alternative

Lengthening of Telomeres (ALT) in LAPC4 prostate cancer cells. Mol.

Cancer Res. (2019) doi:10.1158/1541-7786.MCR-19-0654.

201. NCCN. NCCN v1.2019 - March 6, 2019. NCCN (2019)

doi:10.1016/j.mpmed.2015.10.001.

202. Mosquera, J. M. et al. Concurrent AURKA and MYCN gene amplifications

are harbingers of lethal treatment- related neuroendocrine prostate cancer.

Neoplasia (United States) (2013) doi:10.1593/neo.121550.

181

203. Bluemn, E. G. et al. Androgen Receptor Pathway-Independent Prostate

Cancer Is Sustained through FGF Signaling. Cancer Cell (2017)

doi:10.1016/j.ccell.2017.09.003.

204. Scher, H. I. et al. Increased survival with enzalutamide in prostate cancer

after chemotherapy. N. Engl. J. Med. (2012) doi:10.1056/NEJMoa1207506.

205. Thalmann, G. N. et al. Androgen-independent Cancer Progression and Bone

Metastasis in the LNCaP Model of Human Prostate Cancer. Cancer Res.

(1994).

206. Sramkoski, R. M. et al. A new human prostate carcinoma cell line, 22Rv1.

Vitr. Cell. Dev. Biol. - Anim. (1999) doi:10.1007/s11626-999-0115-4.

207. Stone, K. R., Mickey, D. D., Wunderli, H., Mickey, G. H. & Paulson, D. F.

Isolation of a human prostate carcinoma cell line (DU 145). Int. J. Cancer

(1978) doi:10.1002/ijc.2910210305.

208. Kaighn, M. E., Narayan, K. S., Ohnuki, Y., Lechner, J. F. & Jones, L. W.

Establishment and characterization of a human prostatic carcinoma cell line

(PC-3). Invest. Urol. (1979).

209. Cerami, E. et al. The cBio Cancer Genomics Portal: An open platform for

exploring multidimensional cancer genomics data. Cancer Discov. (2012)

doi:10.1158/2159-8290.CD-12-0095.

210. Gao, J. et al. Integrative analysis of complex cancer genomics and clinical

profiles using the cBioPortal. Sci. Signal. (2013)

doi:10.1126/scisignal.2004088.

211. Ghandi, M. et al. Next-generation characterization of the Cancer Cell Line

182

Encyclopedia. Nature (2019) doi:10.1038/s41586-019-1186-3.

212. You, S. et al. Integrated classification of prostate cancer reveals a novel

luminal subtype with poor outcome. Cancer Res. (2016) doi:10.1158/0008-

5472.CAN-16-0902.

213. Plantinga, M. J. et al. Telomerase suppresses formation of ALT-associated

single-stranded telomeric C-circles. Mol. Cancer Res. (2013)

doi:10.1158/1541-7786.MCR-13-0013.

214. Debacq-Chainiaux, F., Erusalimsky, J. D., Campisi, J. & Toussaint, O.

Protocols to detect senescence-associated beta-galactosidase (SA-βgal)

activity, a biomarker of senescent cells in culture and in vivo. Nat. Protoc.

(2009) doi:10.1038/nprot.2009.191.

215. Brown, J. P., Wei, W. & Sedivy, J. M. Bypass of senescenoe after disruption

of p21(CIP1)/(WAF1) gene in normal diploid human fibroblasts. Science (80-

. ). (1997) doi:10.1126/science.277.5327.831.

216. Beltran, H. et al. Challenges in recognizing treatment-related

neuroendocrine prostate cancer. J. Clin. Oncol. (2012)

doi:10.1200/JCO.2011.41.5166.

217. KOMIYA, A. et al. The prognostic significance of loss of the androgen

receptor and neuroendocrine differentiation in prostate biopsy specimens

among castration-resistant prostate cancer patients. Mol. Clin. Oncol. (2013)

doi:10.3892/mco.2013.69.

218. Shen, R. et al. Transdifferentiation of cultured human prostate cancer cells

to a neuroendocrine cell phenotype in a hormone-depleted medium. Urol.

183

Oncol. (1997) doi:10.1016/S1078-1439(97)00039-2.

219. Burchardt, T. et al. Transdifferentiation of prostate cancer cells to a

neuroendocrine cell phenotype in vitro and in vivo. J. Urol. (1999)

doi:10.1016/S0022-5347(05)68241-9.

220. Pernicová, Z. et al. The role of high cell density in the promotion of

neuroendocrine transdifferentiation of prostate cancer cells. Mol. Cancer

(2014) doi:10.1186/1476-4598-13-113.

221. Fiandalo, M. V. et al. Serum-free complete medium, an alternative medium

to mimic androgen deprivation in human prostate cancer cell line models.

Prostate (2018) doi:10.1002/pros.23459.

222. Ather, M. H., Abbas, F., Faruqui, N., Israr, M. & Pervez, S. Correlation of

three immunohistochemically detected markers of neuroendocrine

differentiation with clinical predictors of disease progression in prostate

cancer. BMC Urol. (2008) doi:10.1186/1471-2490-8-21.

223. Kregel, S. et al. Acquired resistance to the second-generation androgen

receptor antagonist enzalutamide in castration-resistant prostate cancer.

Oncotarget (2016) doi:10.18632/oncotarget.8456.

224. Clynes, D. et al. ATRX dysfunction induces replication defects in primary

mouse cells. PLoS One (2014) doi:10.1371/journal.pone.0092915.

225. Li, Y. & Tergaonkar, V. Noncanonical functions of telomerase: Implications

in telomerase-targeted cancer therapies. Cancer Research (2014)

doi:10.1158/0008-5472.CAN-13-3568.

226. Morgenstern, J. P. & Land, H. Advanced mammalian gene transfer: High

184

titre retroviral vectors with multiple drug selection markers and a

complementary helper-free packaging cell line. Nucleic Acids Res. (1990)

doi:10.1093/nar/18.12.3587.

227. Wong, J. M. Y. & Collins, K. Telomerase RNA level limits telomere

maintenance in X-linked dyskeratosis congenita. Genes Dev. (2006)

doi:10.1101/gad.1476206.

228. Dull, T. et al. A Third-Generation Lentivirus Vector with a Conditional

Packaging System. J. Virol. (1998) doi:10.1128/jvi.72.11.8463-8471.1998.

229. Popovics, P. et al. Prostatic osteopontin expression is associated with

symptomatic benign prostatic hyperplasia. bioRxiv (2019)

doi:10.1101/2019.12.23.887612.

230. Schindelin, J. et al. Fiji: An open-source platform for biological-image

analysis. Nature Methods (2012) doi:10.1038/nmeth.2019.

231. Zeng, X. et al. Administration of a Nucleoside Analog Promotes Cancer Cell

Death in a Telomerase-Dependent Manner. Cell Rep. (2018)

doi:10.1016/j.celrep.2018.05.020.

232. Herbert, B. S., Hochreiter, A. E., Wright, W. E. & Shay, J. W. Nonradioactive

detection of telomerase activity using the telomeric repeat amplification

protocol. Nat. Protoc. (2006) doi:10.1038/nprot.2006.239.

233. MANDERS, E. M. M., VERBEEK, F. J. & ATEN, J. A. Measurement of co‐

localization of objects in dual‐ colour confocal images. J. Microsc. 169, 375–

382 (1993).

234. Van Bokhoven, A. et al. Molecular Characterization of Human Prostate

185

Carcinoma Cell Lines. Prostate (2003) doi:10.1002/pros.10290.

235. Guervilly, J. H. et al. The SLX4 complex is a SUMO E3 ligase that impacts

on replication stress outcome and genome stability. Mol. Cell (2015)

doi:10.1016/j.molcel.2014.11.014.

236. Hubackova, S., Krejcikova, K., Bartek, J. & Hodny, Z. Interleukin 6 signaling

regulates promyelocytic leukemia protein gene expression in human normal

and cancer cells. J. Biol. Chem. (2012) doi:10.1074/jbc.M111.316869.

237. Twillie, D. A. et al. Interleukin-6: A candidate mediator of human prostate

cancer morbidity. Urology (1995) doi:10.1016/S0090-4295(99)80034-X.

238. Qiu, Y., Robinson, D., Pretlow, T. G. & Kung, H. J. Etk/Bmx, a tyrosine kinase

with a pleckstrin-homology domain, is an effector of phosphatidylinositol 3′-

kinase and is involved in interleukin 6-induced neuroendocrine differentiation

of prostate cancer cells. Proc. Natl. Acad. Sci. U. S. A. (1998)

doi:10.1073/pnas.95.7.3644.

239. Spiotto, M. T. & Chung, T. D. K. STAT3 mediates IL-6-induced

neuroendocrine differentiation in prostate cancer cells. Prostate (2000)

doi:10.1002/(SICI)1097-0045(20000215)42:3<186::AID-PROS4>3.0.CO;2-

240. Dang, Q. et al. Anti-androgen enzalutamide enhances prostate cancer

neuroendocrine (NE) differentiation via altering the infiltrated mast cells →

androgen receptor (AR) → miRNA32 signals. Mol. Oncol. (2015)

doi:10.1016/j.molonc.2015.02.010.

241. Graf, M. et al. Telomere Length Determines TERRA and R-Loop Regulation

186

through the Cell Cycle. Cell 170, (2017).

242. O’Sullivan, R. J. et al. Rapid induction of alternative lengthening of telomeres

by depletion of the histone chaperone ASF1. Nat. Struct. Mol. Biol. 21,

(2014).

243. Clynes, D. et al. Suppression of the alternative lengthening of telomere

pathway by the chromatin remodelling factor ATRX. Nat. Commun. 6,

(2015).

244. Hu, Y. et al. Switch telomerase to ALT mechanism by inducing telomeric

DNA damages and dysfunction of ATRX and DAXX. Sci. Rep. (2016)

doi:10.1038/srep32280.

245. Heaphy, C. M. et al. Altered telomeres in tumors with ATRX and DAXX

mutations. Science vol. 333 (2011).

246. Lovejoy, C. A. et al. Loss of ATRX, genome instability, and an altered DNA

damage response are hallmarks of the alternative lengthening of Telomeres

pathway. PLoS Genet. 8, (2012).

247. Fang, H. T. et al. Global H3.3 dynamic deposition defines its bimodal role in

cell fate transition. Nat. Commun. 9, (2018).

248. Kim, J. et al. FOXA1 inhibits prostate cancer neuroendocrine differentiation.

Oncogene (2017) doi:10.1038/onc.2017.50.

249. De Stanchina, E. et al. PML is a direct p53 target that modulates p53 effector

functions. Mol. Cell 13, (2004).

250. Uchida, T., Wada, C., Shitara, T., Egawa, S. & Koshibal, K. Infrequent

involvement of p53 gene mutations in the tumourigenesis of Japanese

187

prostate cancer. Br. J. Cancer 68, (1993).

251. Dinjens, W. N. M., Van Der Weiden, M. M., Schroeder, F. H., Bosman, F. T.

& Trapman, J. Frequency and characterization of p53 mutations in primary

and metastatic human prostate cancer. Int. J. Cancer 56, (1994).

252. van Veldhuizen, P. J., Sadasivan, R., Garcia, F., Austenfeld, M. S. &

Stephens, R. L. Mutant p53 expression in prostate carcinoma. Prostate 22,

(1993).

253. Bookstein, R., Mac Grogan, D., Hilsenbeck, S. G., Sharkey, F. & Craig

Allred, D. p53 Is Mutated in a Subset of Advanced-Stage Prostate Cancers.

Cancer Res. 53, (1993).

254. Navone, N. M. et al. P53 protein accumulation and gene mutation in the

progression of human prostate carcinoma. J. Natl. Cancer Inst. 85, (1993).

255. Untergasser, G., Plas, E., Pfister, G., Heinrich, E. & Berger, P. Interferon-γ

induces neuroendocrine-like differentiation of human prostate basal-

epithelial cells. Prostate 64, (2005).

256. Lavau, C. et al. The acute promyelocytic leukaemia-associated PML gene is

induced by interferon. Oncogene 11, (1995).

257. Vlasáková, J. et al. Histone deacetylase inhibitors suppress IFNα-induced

up-regulation of promyelocytic leukemia protein. Blood 109, (2007).

258. Beauclair, G., Bridier-Nahmias, A., Zagury, J. F., Säb, A. & Zamborlini, A.

JASSA: A comprehensive tool for prediction of SUMOylation sites and SIMs.

Bioinformatics 31, (2015).

259. Zhao, Q. et al. GPS-SUMO: A tool for the prediction of sumoylation sites and

188

SUMO-interaction motifs. Nucleic Acids Res. 42, (2014).

260. Conteduca, V. et al. Clinical features of neuroendocrine prostate cancer.

Eur. J. Cancer 121, (2019).

261. Nader, R., El Amm, J. & Aragon-Ching, J. Role of chemotherapy in prostate

cancer. Asian Journal of Andrology vol. 20 (2018).

262. Mohler, J., Armstrong, A., Bahnson, R. & D’Amico, A. NCCN clinical practice

guidelines in oncology: prostate cancer. J Natl Compr Canc Netw (2016)

doi:10.1016/B978-0-12-800077-9.00013-X.

263. Nemeth, J., Oesch, G. & Kuster, S. P. Bacteriostatic versus bactericidal

antibiotics for patients with serious bacterial infections: Systematic review

and meta-analysis. J. Antimicrob. Chemother. 70, (2015).

264. Martínez, P. & Blasco, M. A. Telomere-driven diseases and telomere-

targeting therapies. Journal of Cell Biology vol. 216 (2017).

265. Wissler Gerdes, E. O., Zhu, Y., Tchkonia, T. & Kirkland, J. L. Discovery,

development, and future application of senolytics: theories and predictions.

FEBS J. 287, (2020).

266. Zhu, Y. et al. New agents that target senescent cells: The flavone, fisetin,

and the BCL-XL inhibitors, A1331852 and A1155463. Aging (Albany. NY).

9, (2017).

267. Zhu, Y. et al. The achilles’ heel of senescent cells: From transcriptome to

senolytic drugs. Aging Cell 14, (2015).

268. Kirkland, J. L. & Tchkonia, T. Senolytic drugs: from discovery to translation.

Journal of Internal Medicine vol. 288 (2020).

189

269. Xu, M. et al. Transplanted Senescent Cells Induce an Osteoarthritis-Like

Condition in Mice. J. Gerontol. A. Biol. Sci. Med. Sci. 72, (2017).

270. Coppé, J. P. et al. Senescence-associated secretory phenotypes reveal cell-

nonautonomous functions of oncogenic RAS and the p53 tumor suppressor.

PLoS Biol. 6, (2008).

271. Kuilman, T. & Peeper, D. S. Senescence-messaging secretome: SMS-ing

cellular stress. Nature Reviews Cancer vol. 9 (2009).

272. Coppé, J. P., Desprez, P. Y., Krtolica, A. & Campisi, J. The senescence-

associated secretory phenotype: The dark side of tumor suppression.

Annual Review of Pathology: Mechanisms of Disease vol. 5 (2010).

190