UvA-DARE (Digital Academic Repository)

Genome integrity maintenance during spermatogonial development

Zheng, Y.

Publication date 2018 Document Version Final published version License Other Link to publication

Citation for published version (APA): Zheng, Y. (2018). Genome integrity maintenance during spermatogonial development.

General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

Download date:06 Oct 2021

Genome integrity maintenance during spermatogonial development

Yi Zheng

Genome integrity maintenance during spermatogonial development PhD Thesis, University of Amsterdam, The Netherlands

© Yi Zheng 2018, Amsterdam All rights reserved. No parts of this dissertation may be reproduced, stored in a retrieval system of any nature, or transmitted in any form or by any means without written permission from the author.

This thesis describes research performed in the Reproductive Biology Laboratory of the Center for Reproductive Medicine, Academic Medical Center, University of Amsterdam, The Netherlands.

ISBN: 978-94-6332-307-9 Cover: SMC5/6 molecule by Dideke Emma Verver Printing: GVO drukkers & vormgevers B.V. Genome integrity maintenance during spermatogonial development

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. dr. ir. K.I.J. Maex ten overstaan van een door het College voor Promoties ingestelde commissie, in het openbaar te verdedigen in de Agnietenkapel op donderdag 15 februari 2018, te 14.00 uur

door

Yi Zheng geboren te Sichuan, China Promotiecommissie

Promotor: Prof. dr. S. Repping AMC-UvA Co-promotor: Dr. G. Hamer AMC-UvA

Overige leden: Dr. ir. W.M. Baarends Erasmus Universiteit Rotterdam Prof. dr. N. Zelcer AMC-UvA Prof. dr. C.J.F. van Noorden AMC-UvA Dr. N.A.P. Franken AMC-UvA Prof. dr. D.G. de Rooij Universiteit Utrecht

Faculteit der Geneeskunde Table of contents

Chapte 1 7 General introduction and outline of the thesis

Chapter 2 21 Non-SMC element 2 (NSMCE2) of the SMC5/6 complex helps to resolve topological stress Verver DE#, Zheng Y#, Speijer D, Hoebe R, Dekker HL, Repping S, Stap J, Hamer G #equal contribution International Journal of Molecular Sciences. 2016 Oct 26;17(11). pii: E1782

Chapter 3 49 Trivial role for NSMCE2 during in vitro proliferation and differentiation of male germline stem cells Zheng Y, Jongejan A, Mulder CL, Mastenbroek S, Repping S, Wang Y, Li J, Hamer G Reproduction. 2017 Sep;154(3):81-95

Chapter 4 77 On the increasing sensitivity of differentiating spermatogonia to DNA damage Zheng Y, Jongejan A, Mulder CL, van Daalen SKM, Mastenbroek S, Hwang G, Jordan P, Repping S, Hamer G Submitted

Chapter 5 107 Spermatogonial stem cell autotransplantation and germline genomic editing: a future cure for spermatogenic failure and prevention of transmission of genomic diseases Mulder CL#, Zheng Y#, Jan SZ, Struijk RB, Repping S, Hamer G*, van Pelt AM #equal contribution, *corresponding author Human Reproduction Update. 2016 Sep;22(5):561-73

Chapter 6 139 General discussion and implications for future research

Chapter 7 155 Summary Samenvatting

Acknowledgements 160 PhD portfolio 162 About the author 163 List of publications 164

Chapter 1

General introduction and outline of the thesis 8 Chapter 1

Background

Spermatogenic failure An estimated 10-15% of couples suffer from subfertility [1, 2], defined as the inability to conceive after one year of unprotected intercourse [3, 4]. Although the most important factor that affects human fertility is female age, in about half of these couples reduced semen quality is commonly observed [3, 5]. Reduced semen quality can be characterized by low sperm counts (oligozoospermia), low sperm motility (asthenozoospermia), low number of morphologically normal sperm (teratozoospermia) or the most extreme clinical presentation- a complete absence of sperm in the semen (azoospermia) [6]. Azoospermia can be subdivided into obstructive and non-obstructive azoospermia [2]. In the case of obstructive azoospermia, the process of spermatogenesis is most often not affected, but the spermatozoa cannot reach the semen due to a physical obstruction. In the case of non- obstructive azoospermia, the lack of sperm in the semen is caused by severely decreased or absent sperm production in the testis, often referred to as spermatogenic failure. Despite the clinical importance, very little is known about the etiology of spermatogenic failure. There are only a few established causes for spermatogenic failure, including DNA damage caused by chemo- or radiotherapy [7], structural or numerical chromosomal abnormalities [5] and Y- chromosome deletions [8]. Nonetheless, the etiology of spermatogenic failure remains unknown in most cases. It is presumed that genetic mutations lie at the base of many cases of spermatogenic failure [9, 10]. Yet, no direct treatment options for spermatogenic failure are currently available to allow these men to achieve genetic parenthood. The only option to date is the use of testicular sperm extraction (TESE) in combination with intra-cytoplasmic sperm injection (ICSI). The drawback is however that the chance of finding spermatozoa upon TESE in men with non-obstructive azoospermia is roughly 50% and that ICSI implies ovarian hyperstimulation of the unaffected female partner as well as fertilization and culture of the resulting embryos in vitro. If indeed spermatogenic failure is genetic in origin, this would require a precisely patient-specific targeted therapeutic approach, or germline genome modification to restore the genome into its original ‘fertile’ state. This is currently not yet feasible.

Spermatogenesis and spermatogonial stem cells (SSCs) Spermatogenesis is an intricate developmental process ultimately leading to the continuous production of spermatozoa. The whole process comprises three consecutive developmental stages: the spermatogonial stage (mitotic proliferation and differentiation), the spermatocyte stage (meiosis) and the spermatid stage (spermiogenesis) [11]. Specifically, General introduction 9 spermatogenesis initiates from type A spermatogonia that undergo multiple mitotic divisions and then differentiate into intermediate and type B spermatogonia. Type B spermatogonia will then divide to form pre-leptotene spermatocytes that replicate their DNA and enter meiosis. The spermatocytes will subsequently undergo two consecutive meiotic divisions (meiosis I and II) to generate round spermatids which then further develop into elongating spermatids and eventually mature sperm. The type A spermatogonia can be divided into undifferentiated and differentiating spermatogonia. The undifferentiated spermatogonia proliferate freely and maintain spermatogonial density in the testis. In contrast, the differentiating spermatogonia are irreversibly committed towards meiosis and their divisions are strictly regulated. An important subset of the undifferentiated spermatogonia are the spermatogonial stem cells (SSCs). These cells can be defined by their ability to generate and maintain donor-derived spermatogenesis when transplanted into infertile recipient testes [12]. To maintain lifelong male fertility, a perfect balance between SSC self-renewal and differentiation is essential. Too much self-renewal may lead to tumor-like germ cell clusters, while excessive differentiation will lead to germ cell depletion [13]. Despite the apparent importance of this balance, knowledge regarding the molecular mechanisms underlying SSC self-renewal and differentiation remains limited [11].

The spermatogonial response to DNA damage DNA damage, for instance caused by irradiation or chemotherapy, often results in germ cell apoptosis. Many cancer patients undergoing chemo- or radiotherapy are therefore confronted with reduced fertility [14-16]. Furthermore, DNA damage in germ cells that is not correctly repaired can lead to genetic mutations or chromosomal aberrations that can be transmitted to the offspring. For this reason it is thought that germ cells hold a unique response to DNA damage. Indeed, they are generally much more prone to undergo apoptosis in response to DNA damage than somatic cells [17, 18]. Even among the different types of spermatogonia differences in radiosensitivity exist. Differentiating spermatogonia are more radiosensitive and inclined to undergo apoptosis in response to irradiation than the undifferentiated spermatogonial population [19]. Even between the undifferentiated spermatogonia differences exist, with the self-renewing SSCs being the most resistant to DNA damage [20-22]. It seems that, while differentiating spermatogonia with DNA damage are readily eliminated, preservation of SSCs, and thus long-term male fertility, to some extent prevails over the risk of mutated progeny. The mechanisms that determine these differential responses of spermatogonial subtypes to DNA lesions remain largely unknown. 10 Chapter 1

Studies aimed at investigating the role of specific genes in the DNA damage response of SSCs are hampered by the fact that many of these genes are embryonically lethal and thus conventional knockout (KO) strategies are unlikely to work. In addition, reliable conditional KO systems are not available for the earliest stages of spermatogenesis due to the lack of suitable promoters.

Chromatin dynamics and the structural maintenance of chromosomes (SMC) 5/6 complex Spermatogenesis, including spermatogonial differentiation, is associated with a continuous and drastic transformation of chromatin structure and function. Failure to maintain correct spatio-temporal organization of chromatin can lead to genomic instability, which often results in germ cell apoptosis or, when all spermatogenic checkpoints fail, transmission of the genomic abnormalities to the offspring [11, 23]. Thus, regulation of chromatin composition and function and maintenance of genome integrity are of paramount importance to the progression of spermatogenesis and safe reproduction. Genomic integrity maintenance and other chromatin-based processes, e.g. DNA replication, transcription and cellular differentiation, are for a large part orchestrated by SMC protein complexes: SMC1/3 (cohesin), SMC2/4 (condensin) and SMC5/6. Of these the SMC5/6 complex is composed of SMC5 and SMC6 and several non-SMC elements (NSMCE1-4 in mammals, Figure 1). Together, these components can form a ring-like structure able to hold two double-stranded DNA molecules together [24]. Of the NSMCEs, NSMCE2 specifically associates with SMC5, where it displays an E3 small ubiquitin-related modifier (SUMO) ligase activity that is involved in DNA damage repair [25-27].

Figure 1: The structure of the SMC5/6 protein complex. SMC5 and SMC6 proteins form a ring-like structure together with several NSMCEs. NSMCE2 specifically associates with SMC5. Image by Dideke Emma Verver [24].

General introduction 11

In the mouse and human testis, SMC5/6 has been described to be involved in several meiotic processes including chromosome segregation, homologous chromosome synapsis and meiotic sex chromosome inactivation [28, 29]. Interestingly, protein staining for SMC6 has recently been found to coincide with spermatogonial differentiation [29]. However, the specific function of SMC5/6 in differentiating spermatogonia is currently not known. It may be present in germ cells to prevent dangerous and error-prone recombination events in highly repetitive genomic sequences such as the regions that surround centromeres [28-33]. Alternatively, the SMC5/6 complex may be involved in the maintenance of replication fork stability and the prevention of replication-induced DNA damage [34-36]. A recent paper also showed NSMCE2 to be expressed in mouse male germ cells from spermatogonia to round spermatids [37]. Nevertheless, whether the SMC5/6 complex plays a role in spermatogonial response to DNA damage has not been further investigated.

SSC culture SSCs account for only 0.02-0.03% of all germ cells [38]. Given the sparsity of SSCs in the testis, SSC culture has become an established tool to expand and study this relatively rare cell population in vitro. A breakthrough in SSC culture was accomplished in 2003, when Shinohara’s group first reported a long-term culture system for mouse SSCs [39]. In this culture system, primary undifferentiated spermatogonia, termed male germline stem (GS) cells, are able to propagate in vitro for years without losing SSC properties [40]. Since then, successful long-term SSC cultures from rats [41], hamsters [42], rabbits [43] and tree shrew [44] have been achieved. SSCs from these species do not only propagate in vitro for a long time but can also be used to generate transgenic offspring [45]. Based on the culture system for rodent SSCs, we were the first to establish a culture system for adult [46] and prepubertal human SSCs [47]. However, current cultures of human SSCs are not optimal and remain to be improved. Most importantly, the developmental capacity of cultured human SSCs, i.e. the capacity to initiate and maintain spermatogenesis, still needs to be demonstrated.

SSC transplantation In 1994, Brinster and colleagues developed a method of SSC transplantation by injecting donor germ cells into the efferent duct or rete testis of a recipient testis [12, 48]. After transplantation, only SSCs are assumed to relocate to the stem cell niche at the basal membrane of seminiferous tubules, from where they can reinitiate and maintain spermatogenesis. It is well acknowledged that only SSCs have the capacity of producing 12 Chapter 1 donor-derived mature spermatozoa, whilst other more advanced germ cells will most likely degenerate or further differentiate and disappear after a certain period of time. Transplantation has therefore become the golden assay to determine the stem cell capacity of spermatogonia. Later, this technique has successfully been used in rodent species other than mice (e.g. rats), in domestic animals (e.g. boars, bulls, goats, sheep, cats, dogs) and even non-human primates [49]. In 2012, Hermann et al. [50] first reported that autologous or allogeneic transplantation of SSCs into the testis of recipient rhesus monkeys could reestablish spermatogenesis and produce donor-derived sperm that were functional and able to develop to embryos after fertilization. Human SSC transplantation is currently regarded as a future treatment option for prepubertal boys who have lost their germ cells due to the gonadotoxic side-effects of cancer treatment such as chemo- or radiotherapy [51]. In order to preserve fertility in these prepubertal boys, a testicular biopsy can be obtained before the initiation of chemo- or radiotherapy. This testicular biopsy, or cultured and expanded cells from the biopsy, can be cryopreserved. When the patient is cured from cancer and reaches adulthood, the cells can be auto-transplanted into the testis where they will then hopefully initiate and maintain spermatogenesis and restore fertility. This method is currently still under development in our laboratory and the first auto-transplantation is expected to be conducted within five years from now on. Because the efficiency of SSC transplantation highly relies on the number of transplanted SSCs [52], development of culture systems to expand human SSCs is crucial. In addition, the (epi)genomic stability of the cultured SSCs, the efficiency of the transplantation technique in humans, and the safety of the patients and offspring need to be guaranteed before clinical application of SSC transplantation can be considered [53].

Genome modification of SSCs The establishment of stable long-term culture systems to expand SSCs in vitro lays the groundwork for the possibility to genetically modify SSCs. SSCs in culture can be subjected to genetic manipulation, using similar protocols as developed for embryonic stem (ES) cells. More specifically, SSCs can be transfected with plasmid vectors by prevailing methods such as calcium phosphate precipitation, lipofection or electroporation, albeit very inefficiently [54]. To improve the efficiency, SSCs are typically transduced with viral-based vectors, such as adenovirus, adeno-associated virus (AAV), retrovirus or lentivirus. Of these, lentiviral transduction is advantageous in that transgenes can integrate into the genome and will thus be stably expressed in SSCs [55]. In this way, genetically modified animal models, including mice, rats and recently tree shrew, have been produced [44, 56, 57]. Nevertheless, as General introduction 13 retrovirus and lentivirus can integrate into the genome of the recipient, they are not suitable for clinical use in the human. Traditionally, genome editing has been achieved via homologous recombination in ES cells, which is inefficient and time-consuming. Over the last decade, novel tools using engineered nucleases to generate site-specific double-strand breaks (DSBs) have revolutionized the field of genome editing. Of these, the novel CRISPR-Cas9 technique (Figure 2) is unprecedentedly simple and efficient. The emergence of CRISPR-Cas9 greatly facilitates genetic manipulation of SSCs, and to date both targeted gene KO and gene correction have been achieved in rodent SSCs [58-60]. In the future, the combination of SSC culture and CRISPR-Cas9 is expected to enable germline modification of all mammalian species including humans.

Figure 2: A schematic illustration of the type II SpCas9 system. This most commonly used CRISPR-Cas9 system is composed of a single-guide RNA (sgRNA) which contains a specific 20-nt sequence to bind to the complementary genomic DNA (gDNA) sequence. The 20-bp gDNA sequence must precede 5’-NGG, the protospacer-adjacent motif (PAM). In conjunction with the endonuclease Cas9, a DSB can occur at ~3-bp upstream of the PAM. The triggered DSB is typically repaired by non-homologous end joining (NHEJ), which is error-prone and causes insertions/deletions (indels) flanking the DSB site, possibly resulting in a frame-shift and therefore a pre-mature stop codon. Alternatively, when a homologous DNA sequence is provided as a repair template, homology-directed repair (HDR) can occur and precise genomic editing can be achieved.

Since SSCs are capable of transmitting genetic information from one generation to the next, they are an ideal target for genetic manipulation to produce transgenic animal models for biomedical research and animal production. The use of genetically modified SSCs could also form a direct treatment option for men with spermatogenic failure in whom the disease is caused by a genetic abnormality. In addition, the potential use of genome modification in human SSCs opens up the possibility to prevent the paternal transmission of any genetic 14 Chapter 1 abnormality to offspring. Genome modification of human SSCs is in this way in essence a potential alternative for existing methods aimed at preventing the transmission of genetic abnormalities to offspring such as prenatal and preimplantation genetic diagnosis (PGD) and an alternative for the possible future application of genome editing of human embryos. The potential clinical application of germline genome editing, be it in SSCs or in embryos, is a controversial and highly debated subject [61, 62].

Aim and outline of the thesis The specific aim of this thesis was to unravel the mechanisms that determine and regulate the dynamic response to DNA damage during spermatogonial development, with a specific focus on the role of the SMC5/6 complex. The development of SSC culture, together with the development of CRISPR-Cas9 as described above, now for the first time open up the possibility to study this using spermatogonia-specific genome modification. In the current thesis, we combined an established culture system for spermatogonial proliferation and differentiation with the CRISPR-Cas9 system to knock out genes of interest in the response to DNA damage. Furthermore, we additionally addressed the broad and potentially large clinical prospects and implications of using CRISPR-Cas9 to treat spermatogenic failure or to prevent the transmission of genetic abnormalities to human offspring by transplantation of genetically modified human SSCs.

In Chapter 2 we describe the use of CRISPR-Cas9 and a human osteosarcoma cell line (U2OS) to investigate the molecular mechanisms by which the SMC5/6 complex functions in genomic integrity maintenance. In Chapter 3 we report an optimized protocol to generate genetically modified mouse male germline stem cell lines using CRISPR-Cas9. Using this protocol, we generated a male germline stem cell line devoid of NSMCE2, a subunit of the SMC5/6 complex, and then interrogated the role of NSMCE2 in spermatogonial proliferation and differentiation. In Chapter 4 we analyze the transcriptomes of irradiated and non-irradiated undifferentiated and differentiating mouse male germline stem cells to gain insights into the differential DNA damage responses of undifferentiated and differentiating spermatogonia. In Chapter 5 we review the state of the art with respect to SSC transplantation and genomic editing using CRISPR-Cas9, followed by envisioning the clinical prospects of SSC transplantation, with or without genomic editing, to restore male fertility or prevent transmission of genomic disorders. General introduction 15

In Chapter 6 we give an overview of the findings presented in this thesis and discuss the implications of our results for future research. In Chapter 7 we summarize the results presented in this thesis. 16 Chapter 1

References

1. Sharlip ID, Jarow JP, Belker AM, Lipshultz LI, Sigman M, Thomas AJ, Schlegel PN, Howards SS, Nehra A, Damewood MD, Overstreet JW, Sadovsky R. Best practice policies for male infertility. Fertil Steril 2002; 77:873-882. 2. Kumar N, Singh AK. Trends of male factor infertility, an important cause of infertility: A review of literature. J Hum Reprod Sci 2015; 8:191-196. 3. Evers JL. Female subfertility. Lancet 2002; 360:151-159. 4. Gnoth C, Godehardt E, Frank-Herrmann P, Friol K, Tigges J, Freundl G. Definition and prevalence of subfertility and infertility. Hum Reprod 2005; 20:1144-1147. 5. de Kretser DM. Male infertility. Lancet 1997; 349:787-790. 6. van der Steeg JW, Steures P, Eijkemans MJ, JD FH, Hompes PG, Kremer JA, van der Leeuw-Harmsen L, Bossuyt PM, Repping S, Silber SJ, Mol BW, van der Veen F, et al. Role of semen analysis in subfertile couples. Fertil Steril 2011; 95:1013-1019. 7. Marjault HB, Allemand I. Consequences of irradiation on adult spermatogenesis: Between infertility and hereditary risk. Mutat Res 2016; 770:340-348. 8. Noordam MJ, Repping S. The human Y chromosome: a masculine chromosome. Curr Opin Genet Dev 2006; 16:225-232. 9. Gianotten J, Westerveld GH, Leschot NJ, Tanck MW, Lilford RJ, Lombardi MP, van der Veen F. Familial clustering of impaired spermatogenesis: no evidence for a common genetic inheritance pattern. Hum Reprod 2004; 19:71-76. 10. Lilford R, Jones AM, Bishop DT, Thornton J, Mueller R. Case-control study of whether subfertility in men is familial. BMJ 1994; 309:570-573. 11. Jan SZ, Hamer G, Repping S, de Rooij DG, van Pelt AM, Vormer TL. Molecular control of rodent spermatogenesis. Biochim Biophys Acta 2012; 1822:1838-1850. 12. Brinster RL, Zimmermann JW. Spermatogenesis following male germ-cell transplantation. Proc Natl Acad Sci U S A 1994; 91:11298-11302. 13. Silber SJ. Evaluation and treatment of male infertility. Clin Obstet Gynecol 2000; 43:854-888. 14. Meistrich ML. Effects of chemotherapy and radiotherapy on spermatogenesis in humans. Fertil Steril 2013; 100:1180-1186. 15. Jeruss JS, Woodruff TK. Preservation of fertility in patients with cancer. N Engl J Med 2009; 360:902-911. 16. Stahl O, Boyd HA, Giwercman A, Lindholm M, Jensen A, Kjaer SK, Anderson H, Cavallin- Stahl E, Rylander L. Risk of birth abnormalities in the offspring of men with a history of cancer: a cohort study using Danish and Swedish national registries. J Natl Cancer Inst 2011; 103:398-406. General introduction 17

17. Paris L, Cordelli E, Eleuteri P, Grollino MG, Pasquali E, Ranaldi R, Meschini R, Pacchierotti F. Kinetics of gamma-H2AX induction and removal in bone marrow and testicular cells of mice after X-ray irradiation. Mutagenesis 2011; 26:563-572. 18. Rube CE, Zhang S, Miebach N, Fricke A, Rube C. Protecting the heritable genome: DNA damage response mechanisms in spermatogonial stem cells. DNA Repair (Amst) 2011; 10:159-168. 19. van der Meer Y, Huiskamp R, Davids JA, van der Tweel I, de Rooij DG. The sensitivity to X rays of mouse spermatogonia that are committed to differentiate and of differentiating spermatogonia. Radiat Res 1992; 130:296-302. 20. van Beek ME, Meistrich ML, de Rooij DG. Probability of self-renewing divisions of spermatogonial stem cells in colonies, formed after fission neutron irradiation. Cell Tissue Kinet 1990; 23:1-16. 21. Aloisio GM, Nakada Y, Saatcioglu HD, Pena CG, Baker MD, Tarnawa ED, Mukherjee J, Manjunath H, Bugde A, Sengupta AL, Amatruda JF, Cuevas I, et al. PAX7 expression defines germline stem cells in the adult testis. J Clin Invest 2014; 124:3929-3944. 22. Komai Y, Tanaka T, Tokuyama Y, Yanai H, Ohe S, Omachi T, Atsumi N, Yoshida N, Kumano K, Hisha H, Matsuda T, Ueno H. Bmi1 expression in long-term germ stem cells. Scientific Reports 2014; 4. 23. de Rooij DG, de Boer P. Specific arrests of spermatogenesis in genetically modified and mutant mice. Cytogenet Genome Res 2003; 103:267-276. 24. Verver DE, Hwang GH, Jordan PW, Hamer G. Resolving complex chromosome structures during meiosis: versatile deployment of Smc5/6. Chromosoma 2016; 125:15-27. 25. Andrews EA, Palecek J, Sergeant J, Taylor E, Lehmann AR, Watts FZ. Nse2, a component of the Smc5-6 complex, is a SUMO ligase required for the response to DNA damage. Mol Cell Biol 2005; 25:185-196. 26. Potts PR, Yu H. Human MMS21/NSE2 is a SUMO ligase required for DNA repair. Mol Cell Biol 2005; 25:7021-7032. 27. Zhao X, Blobel G. A SUMO ligase is part of a nuclear multiprotein complex that affects DNA repair and chromosomal organization. Proc Natl Acad Sci U S A 2005; 102:4777-4782. 28. Gomez R, Jordan PW, Viera A, Alsheimer M, Fukuda T, Jessberger R, Llano E, Pendas AM, Handel MA, Suja JA. Dynamic localization of SMC5/6 complex proteins during mammalian meiosis and mitosis suggests functions in distinct chromosome processes. J Cell Sci 2013; 126:4239-4252. 29. Verver DE, van Pelt AM, Repping S, Hamer G. Role for rodent Smc6 in pericentromeric heterochromatin domains during spermatogonial differentiation and meiosis. Cell Death Dis 2013; 4:e749. 30. Farmer S, San-Segundo PA, Aragón L. The smc5-smc6 complex is required to remove chromosome junctions in meiosis. PLoS One 2011; 6:e20948. 31. Lilienthal I, Kanno T, Sjogren C. Inhibition of the Smc5/6 Complex during Meiosis Perturbs Joint Molecule Formation and Resolution without Significantly Changing Crossover or Non-crossover Levels. PLoS Genet 2013; 9:e1003898. 18 Chapter 1

32. Torres-Rosell J, Sunjevaric I, De Piccoli G, Sacher M, Eckert-Boulet N, Reid R, Jentsch S, Rothstein R, Aragon L, Lisby M. The Smc5-Smc6 complex and SUMO modification of Rad52 regulates recombinational repair at the ribosomal gene locus. Nat Cell Biol 2007; 9:923-931. 33. Xaver M, Huang L, Chen D, Klein F. Smc5/6-mms21 prevents and eliminates inappropriate recombination intermediates in meiosis. PLoS Genet 2013; 9:e1004067. 34. Carter SD, Sjogren C. The SMC complexes, DNA and chromosome topology: right or knot? Crit Rev Biochem Mol Biol 2012; 47:1-16. 35. Jeppsson K, Kanno T, Shirahige K, Sjogren C. The maintenance of chromosome structure: positioning and functioning of SMC complexes. Nat Rev Mol Cell Biol 2012; 15:601-614. 36. Langston RE, Weinert T. Nifty Alleles, a Plethora of Interactions, and Imagination Advance Understanding of Smc5/6's Roles with Chromosomes. Mol Cell 2015; 60:832-833. 37. Jacome A, Gutierrez-Martinez P, Schiavoni F, Tenaglia E, Martinez P, Rodriguez-Acebes S, Lecona E, Murga M, Mendez J, Blasco MA, Fernandez-Capetillo O. NSMCE2 suppresses cancer and aging in mice independently of its SUMO ligase activity. Embo J 2016; 34:2604-2619. 38. Tegelenbosch RA, de Rooij DG. A quantitative study of spermatogonial multiplication and stem cell renewal in the C3H/101 F1 hybrid mouse. Mutat Res 1993; 290:193-200. 39. Kanatsu-Shinohara M, Ogonuki N, Inoue K, Miki H, Ogura A, Toyokuni S, Shinohara T. Long- term proliferation in culture and germline transmission of mouse male germline stem cells. Biol Reprod 2003; 69:612-616. 40. Kanatsu-Shinohara M, Ogonuki N, Iwano T, Lee J, Kazuki Y, Inoue K, Miki H, Takehashi M, Toyokuni S, Shinkai Y, Oshimura M, Ishino F, et al. Genetic and epigenetic properties of mouse male germline stem cells during long-term culture. Development 2005; 132:4155-4163. 41. Ryu BY, Kubota H, Avarbock MR, Brinster RL. Conservation of spermatogonial stem cell self- renewal signaling between mouse and rat. Proc Natl Acad Sci U S A 2005; 102:14302-14307. 42. Kanatsu-Shinohara M, Muneto T, Lee J, Takenaka M, Chuma S, Nakatsuji N, Horiuchi T, Shinohara T. Long-term culture of male germline stem cells from hamster testes. Biol Reprod 2008; 78:611-617. 43. Kubota H, Wu X, Goodyear SM, Avarbock MR, Brinster RL. Glial cell line-derived neurotrophic factor and endothelial cells promote self-renewal of rabbit germ cells with spermatogonial stem cell properties. FASEB J 2011; 25:2604-2614. 44. Li CH, Yan LZ, Ban WZ, Tu Q, Wu Y, Wang L, Bi R, Ji S, Ma YH, Nie WH, Lv LB, Yao YG, et al. Long-term propagation of tree shrew spermatogonial stem cells in culture and successful generation of transgenic offspring. Cell Res 2017; 27:241-252. 45. Kanatsu-Shinohara M, Shinohara T. Spermatogonial stem cell self-renewal and development. Annu Rev Cell Dev Biol 2013; 29:163-187. 46. Sadri-Ardekani H, Mizrak SC, van Daalen SK, Korver CM, Roepers-Gajadien HL, Koruji M, Hovingh S, de Reijke TM, de la Rosette JJ, van der Veen F, de Rooij DG, Repping S, et al. Propagation of human spermatogonial stem cells in vitro. JAMA 2009; 302:2127-2134. General introduction 19

47. Sadri-Ardekani H, Akhondi MA, van der Veen F, Repping S, van Pelt AM. In vitro propagation of human prepubertal spermatogonial stem cells. JAMA 2011; 305:2416-2418. 48. Brinster RL, Avarbock MR. Germline transmission of donor haplotype following spermatogonial transplantation. Proc Natl Acad Sci U S A 1994; 91:11303-11307. 49. Dores C, Alpaugh W, Dobrinski I. From in vitro culture to in vivo models to study testis development and spermatogenesis. Cell Tissue Res 2012; 349:691-702. 50. Hermann BP, Sukhwani M, Winkler F, Pascarella JN, Peters KA, Sheng Y, Valli H, Rodriguez M, Ezzelarab M, Dargo G, Peterson K, Masterson K, et al. Spermatogonial stem cell transplantation into rhesus testes regenerates spermatogenesis producing functional sperm. Cell Stem Cell 2012; 11:715-726. 51. Brinster RL. Male germline stem cells: from mice to men. Science 2007; 316:404-405. 52. Dobrinski I, Ogawa T, Avarbock MR, Brinster RL. Computer assisted image analysis to assess colonization of recipient seminiferous tubules by spermatogonial stem cells from transgenic donor mice. Mol Reprod Dev 1999; 53:142-148. 53. Struijk RB, Mulder CL, van der Veen F, van Pelt AM, Repping S. Restoring fertility in sterile childhood cancer survivors by autotransplanting spermatogonial stem cells: are we there yet? Biomed Res Int 2013; 2013:903142. 54. Kanatsu-Shinohara M, Toyokuni S, Shinohara T. Genetic selection of mouse male germline stem cells in vitro: offspring from single stem cells. Biol Reprod 2005; 72:236-240. 55. Dann CT. Transgenic modification of spermatogonial stem cells using lentiviral vectors. Methods Mol Biol 2013; 927:503-518. 56. Kanatsu-Shinohara M, Shinohara T. Germline Modification Using Mouse Spermatogonial Stem Cells. Methods in Enzymology, Vol 477: Guide to Techniques in Mouse Development, Part B: Mouse Molecular Genetics, Second Edition 2010; 477:17-36. 57. Hamra FK, Chapman KM, Nguyen DM, Williams-Stephens AA, Hammer RE, Garbers DL. Self renewal, expansion, and transfection of rat spermatogonial stem cells in culture. Proc Natl Acad Sci U S A 2005; 102:17430-17435. 58. Chapman KM, Medrano GA, Jaichander P, Chaudhary J, Waits AE, Nobrega MA, Hotaling JM, Ober C, Hamra FK. Targeted Germline Modifications in Rats Using CRISPR/Cas9 and Spermatogonial Stem Cells. Cell Rep 2015; 10:1828-1835. 59. Sato T, Sakuma T, Yokonishi T, Katagiri K, Kamimura S, Ogonuki N, Ogura A, Yamamoto T, Ogawa T. Genome Editing in Mouse Spermatogonial Stem Cell Lines Using TALEN and Double- Nicking CRISPR/Cas9. Stem Cell Reports 2015; 5:75-82. 60. Wu Y, Zhou H, Fan X, Zhang Y, Zhang M, Wang Y, Xie Z, Bai M, Yin Q, Liang D, Tang W, Liao J, et al. Correction of a genetic disease by CRISPR-Cas9-mediated gene editing in mouse spermatogonial stem cells. Cell Res 2015; 25:67-79. 61. Lanphier E, Urnov F, Haecker SE, Werner M, Smolenski J. Don't edit the human germ line. Nature 2015; 519:410-411. 20 Chapter 1

62. Porteus MH, Dann CT. Genome editing of the germline: broadening the discussion. Mol Ther 2015; 23:980-982.

Chapter 2

Non-SMC element 2 (NSMCE2) of the SMC5/6 complex helps to resolve topological stress

Dideke E. Verver# Yi Zheng# Dave Speijer Ron Hoebe Henk L. Dekker Sjoerd Repping Jan Stap Geert Hamer

#equal contribution International Journal of Molecular Sciences 2016 Oct 26;17(11). pii: E1782

22 Chapter 2

Abstract The structural maintenance of chromosomes (SMC) protein complexes shape and regulate the structure and dynamics of chromatin, thereby controlling many chromosome- based processes such as cell cycle progression, differentiation, gene transcription and DNA repair. The SMC5/6 complex is previously described to promote DNA double-strand breaks (DSBs) repair by sister chromatid recombination, and found to be essential for resolving recombination intermediates during meiotic recombination. Moreover, in budding yeast, SMC5/6 provides structural organization and topological stress relief during replication in mitotically dividing cells. Despite the essential nature of the SMC5/6 complex, the versatile mechanisms by which SMC5/6 functions and its molecular regulation in mammalian cells remain poorly understood. By using a human osteosarcoma cell line (U2OS), we show that after the CRISPR-Cas9-mediated removal of the SMC5/6 subunit NSMCE2, treatment with the topoisomerase II inhibitor etoposide triggered an increased sensitivity in cells lacking NSMCE2. In contrast, NSMCE2 appeared not essential for a proper DNA damage response or cell survival after DSB induction by ionizing irradiation (IR). Interestingly, by way of immunoprecipitations (IPs) and mass spectrometry, we found that the SMC5/6 complex physically interacts with the DNA topoisomerase II α (TOP2A). We therefore propose that the SMC5/6 complex functions in resolving TOP2A-mediated DSB-repair intermediates generated during replication.

Keywords Structural Maintenance of Chromosomes 5/6 complex (SMC5/6); Non-SMC Element 2 (NSMCE2); Topoisomerase II α (TOP2A); DNA double-strand breaks (DSBs); Ionizing Radiation (IR); CRISPR-Cas9 NSMCE2 helps to resolve topological stress 23

Introduction The structural maintenance of chromosome (SMC) protein complexes shape and determine chromatin structure and function and are therefore implicated in many, if not all, fundamental chromosome-based processes. The three SMC protein complexes, cohesin, condensin and SMC5/6, all consist of two SMC subunits and several non-SMC elements, the NSMCEs. The resulting ring-like complexes possess the capacity to hold two DNA double- strands together, and are therefore able to physically shape the DNA in specific chromatin structures [1, 2]. By doing so, SMC complexes control chromosome segregation, DNA repair, transcription and replication, among other processes [1, 3-5]. Of the three SMC complexes, the SMC5/6 complex has been most directly and exclusively described to be involved in DNA damage repair and genomic integrity maintenance [6-8]. In mammals, SMC6 is highly expressed in the testis [9, 10] and we have recently found SMC6 to be involved in crucial processes during mouse and human spermatogenesis, including spermatogonial differentiation and meiosis [9, 10]. In various organisms, ranging from yeast to humans, SMC5/6 is involved in numerous meiotic processes [11] such as chromosome segregation [9, 10, 12, 13], homologous chromosome synapsis [9, 12-16] and meiotic sex chromosome inactivation [10, 13]. Co-localization studies have suggested that SMC5/6 prevents dangerous and error-prone homologous recombination (HR) in highly repetitive, densely packed DNA regions such as the rDNA and pericentromeric heterochromatin [10, 13, 15-18]. SMC5/6 seems to be involved in double-strand break (DSB) repair as it is enriched at DSB sites in budding yeast and C. elegans [12, 14, 16]; it localizes side by side with RAD51 in budding yeast and humans [9, 12, 16] and its deletion results in an increase in RAD51 foci and chromosome fragmentation in C. elegans [14]. Furthermore, Smc5/6 has been found to play a role in the resolution of meiotic recombination intermediates and mutations of Smc5, Smc6 or the SUMO ligase domain of Nse2 lead to the accumulation of toxic joint molecules in yeast and C. elegans [12, 15, 16, 19-22]. In budding and fission yeast the Smc5/6 complex is essential for the maintenance of replication fork stability, the prevention of joint molecules and the resolution of such joint molecules that would otherwise lead to mitotic failure (reviewed in [23-25]). In mice, ablation of SMC6 results in embryonic lethality, whilst a mutation in its ATP hydrolysis motif only generates a mild phenotype [26]. NSMCE2 has also been shown to be essential for mouse development and it can suppress cancer and aging by limiting recombination and facilitating chromosome segregation [27]. In line with these studies, a recent paper describes that 24 Chapter 2 depletion of SMC5 in mouse embryonic stem cells led to accumulation of cells in G2 and subsequent mitotic failure and apoptosis [28]. From this increasing amount of data, it has become overwhelmingly clear that SMC5/6 is essential for maintaining genomic integrity by a variety of means. However, the exact roles of the SMC5/6 complex in mammalian especially human cells remain poorly understood. By using a commonly used human osteosarcoma cell line (U2OS), we extended our knowledge regarding the roles of SMC5/6 in human genome integrity maintenance.

Materials and methods Cells and culture

U2OS cells were cultured at 37 °C and 10% CO2 in Dulbecco’s modified Eagle’s medium (DMEM; (high glucose, pyruvate, L-glutamine); Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% Fetal Calf Serum (FCS), penicillin (100 U/mL) and streptomycin (100 U/mL).

Design of single-guide RNAs (sgRNAs) and construction of the CRISPR-Cas9 plasmids An online CRISPR-Cas9 Design Tool provided by Zhang’s lab (http://crispr.mit.edu) was exploited to identify the targeting sequences. The 20-nt targeting sequences preceding 5′-NGG (the protospacer-adjacent motif, PAM), locating in early and conserved coding exons among transcript variants, and with least predicted off-target sites in human genome were selected. To target SMC6 and NSMCE2 in U2OS cells, two sgRNAs were designed for each gene: 5′-GGTGACGAAGACGAATGTAA-3′ (in exon 3) and 5′-ATGCTTGGACCTTTTAAGTT- 3′ (in exon 4) for SMC6, 5′-TTCCAAGCCTGTATCAACTC -3′ and 5′- AGCCTGTATCAACTCTGGTA-3′ (both in exon 3) for NSMCE2. The corresponding sense and antisense strands of oligos were purchased from Sigma-Aldrich (St. Louis, MO, USA). The CRISPR plasmids pSpCas9(BB)-2A-GFP (pX458) were obtained from Addgene (Addgene plasmid 48138). The pX458 plasmids were digested with FastDigest BbsI (Fermentas, Waltham, MA, USA). The oligos were then annealed and cloned into digested pX458 according to the protocol described by Ran et al. [29].

Transfection of U2OS cells with CRISPR plasmids The constructed pX458 plasmids were transfected into U2OS cells with the 4D- Nucleofector System (Lonza, Basel, Switzerland). For each nucleofection reaction, approximately 500 ng plasmids were transfected into 100,000 U2OS cells using program CM-104 and the SE Cell Line 4D-Nucleofector X Kit S (Lonza), according to the NSMCE2 helps to resolve topological stress 25 manufacturer’s instructions. One day after nucleofection, the top 5%-10% GFP+ U2OS cells were separated by FACS with a BD FACSAria cell sorter. In order to derive a stable NSMCE2-deficient cell line from single U2OS cells, FACS was conducted to deposit single GFP+ cells into 96-well plates (one cell per well).

Surveyor assay for verification of genome editing Genomic DNA of transfected U2OS cells was extracted with QIAamp DNA Mini Kit (Qiagen, Hilden, Germany), according to the protocol provided by the manufacturer. The genomic region (417 bp) containing both mutation sites in exon 3 of NSMCE2 was amplified by PCR with Herculase II fusion polymerase (Agilent Technologies, Santa Clara, CA, USA). The forward and reverse primers used for PCR were 5′-AATTTCAAGATGCCAGGACGT-3′ and 5′-GGATCTTCAAATCTTTGCCCAT-3′, respectively. PCR products were purified by QIAquick PCR Purification Kit (Qiagen). Genomic modifications in the amplified region were then detected with Surveyor Mutation Detection Kit (Integrated DNA Technologies, Coralville, IA, USA), according to the manufacturer’s instructions. After Surveyor nuclease digestion, the PCR products were run on a 2% agarose gel with ethidium bromide (EB) for visualization. The insertion/deletion (indel) occurrence was estimated with the formula described by Ran et al. [29].

Sequencing analysis of the targeting site in NSMCE2 Genomic DNA was extracted from each single cell-derived colonies, and the region flanking the targeting site was amplified by PCR with Easy-A High-Fidelity PCR Cloning Enzyme (Agilent Technologies). The primers used were identical to those for the Surveyor assay. After purification, the PCR products were cloned into TOPO TA cloning vectors (Thermo Fisher Scientific). The ligated vectors were transformed into One Shot TOP10 Chemically Competent E. coli (Thermo Fisher Scientific). For each reaction, forty colonies were randomly picked and subjected to Sanger-sequencing to analyze mutations from all alleles.

Off-target analysis in the NSMCE2-devoid cell line Genomic DNA was extracted from WT and NSMCE2 null U2OS cells, respectively. To gain an overview of off-target effects in the established NSMCE2 null cell line, the 10 top- ranking potential off-target sites provided by the online CRISPR Design Tool (http://crispr.mit.edu) were analyzed. The selected sites included those preceding 5′-NAG, the alternative PAM. The genomic regions flanking each potential off-target sites were amplified by PCR with Herculase II fusion polymerase (Agilent Technologies). The selected 26 Chapter 2 off-target sites and the corresponding PCR primers are depicted in Table S1. Purified PCR products were Sanger-sequenced for analysis of off-target effects.

Live cell microscopy Cells were plated in multi-chambered cover-glass slides (Labtek II, Nunc) in a density of 1000 cells/cm2 and cultured overnight before starting imaging. Both U2OS WT and NSMCE2 null cells were imaged at the same time, using a IRBE inverted phase contrast microscope and a N Plan Apo L 40×/0.55 Ph2 objective. Images were captured every 10 min for a total of 170 h, at 37 °C in an atmosphere containing 10% CO2. Medium was refreshed under the microscope. Cell cycle duration was determined by calculating the time between two cell divisions. Generations were aligned, in which we chose the third generation as the one in which the medium was refreshed, based on the WT cells.

Distribution of cells over the cell cycle phase To determine the distribution of cells over the different cell cycle phases, colcemid (KaryoMAX Colcemid Solution; Thermo Fisher Scientific) was added to the culture medium to a final concentration of 0.1 µg/mL and DNA histograms were made after incubation. Cells were detached by 0.05% trypsin/EDTA (Thermo Fisher Scientific), pelleted in serum- containing medium and washed in PBS (Phosphate Buffered Saline). Cells were fixed and stored in 100% EtOH at 4 °C. On the day of FACS analysis, cells were pelleted by centrifugation and all EtOH was removed. Cells were resuspended in PBS and RNAse A (final 1 mg/mL; Roche, Basel, Switzerland) was added. After vortexing, propidium iodide (PI; final 25 µg/mL; Sigma-Aldrich) was added and the cell suspension was thoroughly vortexed again. Cells were incubated for 15 min at 37 °C, after which the cell suspension was transferred through a 21 G needle twice, in order to disrupt cell aggregates. DNA content was analyzed on a LSR Fortessa FACS analyzer (BD Biosciences, San Jose, CA, USA) using DivaTM acquisition and analysis software. Figures were constructed using FlowJo X software.

Ionizing irradiation (IR) Exponentially growing cells were exposed to IR emitted by a 137Cs source (95% β- emission). For immunocytochemistry, cells received 1 Gy of IR. For clonogenic assays, cells received 0-8 Gy of IR.

NSMCE2 helps to resolve topological stress 27

Immunocytochemistry (ICC) For ICC, cells were plated on glass coverslides overnight, after which they were fixed in 4% formaldehyde/PBS for 10 min at room temperature (RT). In the case of IR treatment, IR- treated cells (and their non-IR counterparts) were fixed at varying time points, ranging from 0 min to 6 h post IR. In the case of etoposide treatment, cells were incubated with 3 µM etoposide for 1 h at 37 °C/10% CO2. Cells were fixed at varying time points after removal of etoposide, ranging from 0 min to 6 h post etoposide treatment. After fixation, cells were permeabilized for 10 min at RT in PBST (0.25% Tween-20/PBS). Next, non-specific adhesion sites were blocked for 45 min at RT in PBST containing 1% bovine serum albumin (BSA). To visualize SMC6 and γH2AX, slides were incubated for 2 h at RT in primary antibodies guinea pig (GP) anti-SMC6 (1/200; custom made, peptide: KRPRQEELEDFDKDGDEDE) and mouse anti-γH2AX (1/10,000; 05-636, Merck Millipore, Darmstadt, Germany), diluted in 1% BSA/PBST. After incubation in corresponding secondary antibodies (Goat-anti-GP Alexa488, Goat-anti-Mouse Alexa555, respectively; all diluted 1/1000 in 1% BSA/PBST), slides were washed and counterstained with DAPI and mounted in Prolong Gold. Between all steps, except blocking and primary antibody incubation, 3 × 5 min washes in PBS were performed. Widefield fluorescence microscopy images were acquired at RT using a Plan Fluotar 100×/1.30 oil objective on a Leica DM5000B widefield microscope equipped with a Leica DFC365 FX CCD camera. Images were analyzed using Leica Application Suite Advanced Fluorescence software. Figures were constructed using Adobe Photoshop CS5 version 12.0. Confocal images for subsequent quantification were acquired at RT using a Leica TCS SP8 SMD confocal microscope equipped with a 63×/1.40 HC Plan Apo oil CS2 objective (Leica, Wetzlar, Germany). For excitation of DAPI, a 405 nm UV Diode was used and for excitation of other fluochromes, the VIS Argon 470–670 nm White Light Laser (WLL) was used. Fluorescent signal was detected by PMTs and a HyD detector, and acquisition of the image (stacks) was performed using LAS AF X software.

Quantitative imaging In order to enable identical staining and acquisition conditions for all samples, four time points were chosen for quantitative imaging, 0 min, 30 min, 3 h and 6 h post IR or etoposide treatment. Confocal image stacks were acquired using the following settings: resonant scan = on; galvo flow and bidirectional X = off; line average = 4; acquisition = between lines; field of view = 792 × 792 (zoom = 5.0); Z-step size = 0.20 µm; stack size = 8 µm total (42 steps). One pixel = 47 × 47 × 200 nm (XYZ). 28 Chapter 2

Images were deconvolved using Huygens Essential software, with a maximum number of iterations of 40, and a SNR setting of 12 (green channel) or 10 (red channel). By visual inspection, cells with at least two clear SMC6 foci were identified for further analysis with MatLab software. Using MatLab software, we isolated the γH2AX foci with a minimal size of 50 pixels (0.022 µm3) present in the nucleus. Next, the separate SMC6 foci were isolated (minimal size 10 pixels), and the amount of γH2AX foci overlapping with a SMC6 focus was determined. Statistical significance was determined by applying the Student’s t-test (two- tailed, unpaired).

Clonogenic assay Clonogenic assays were performed as described previously [30]. Four hours after plating, cells were exposed to 0-8 Gy of IR or incubated for 1 h at 37 °C in 0-30 μM etoposide. In each experiment, each dose was administered to 2 different cell densities. Experiments were repeated at least 3 times. Survival capacity was calculated relative to the non-treated control condition. Statistical significance was determined by applying the Student’s t-test (one-tailed, paired).

Protein isolation Cells were detached, washed in PBS and pelleted by centrifugation. These cell pellets were either snapfrozen in liquid nitrogen and stored or lysed directly. Cells were lysed in RIPA buffer containing 1× PIC and 1× PhosSTOP (Roche) for 1 h on ice. The lysate was centrifuged (16,000 rcf, 10 min, 4 °C) to clarify the extract.

Western blot analysis Western blot analysis of cell lysates was performed as previously described [10], using the primary antibodies: SMC6 GP (1/200; custom made), SMC5 (1/1000; A300-236A, Bethyl Laboratories, Montgomery, TX, USA), NSMCE2 (1/500; NBP1-76263, Novus Biologicals, Littleton, CO, USA), TOP2A (1/1,000; TG2011-1, TopoGEN, Buena Vista, CO, USA) and β- actin (1/5000; A1978, Sigma-Aldrich).

Immunoprecipitation (IP) IP was performed on lysed cells, using Dynabeads Protein A (Thermo Fisher Scientific). Per IP, 1 × 106 cells and 50 µL dynabeads were used. Dynabeads were resuspended in AB Binding & Washing buffer containing 2 µL anti-SMC6 GP (custom made) or anti-SMC5 (A300-236A, Bethyl Laboratories) antibody and incubated for 30 min with rotation at RT. NSMCE2 helps to resolve topological stress 29

Using a magnet, the supernatant was removed, and the beads were washed by resuspension in AB Binding & Washing buffer. After the removal of the buffer, the beads were incubated in the cell lysate for 30 min with rotation at RT, after which the supernatant was collected and the beads were washed. Elution of the precipitated antigen was achieved after resuspension of the beads in RIPA buffer containing 1× PIC and 1× PhosSTOP, addition of LDS Sample Buffer and Sample Reducing Agent and heating of the sample for 10 min at 70 °C. In preparation for Western blot analysis, the supernatant was also supplemented with LDS Sample Buffer and Sample Reducing Agent and heated for 10 min at 70 °C.

Mass spectrometry Abcam rabbit anti-SMC6 (ab18039, Abcam, Cambridge, UK) was used for IP followed by mass spectrometry of the several bands detected by the antibody in U2OS cells, following the protocol described above. A total of approximately 7 × 106 cells, 100 µL dynabeads and 5 µL SMC6 Abcam antibody were used. All immunoprecipitated material was loaded on a single lane of a 4%-12% bis-tris gradient gel (Thermo Fisher Scientific). After running, the gel was washed 3 times 10 min in H2O to remove SDS, and subsequently stained in a colloidal coomassie solution (PageBlue Protein Staining Solution; 24620, Thermo Fisher Scientific) for

1 h at RT, after which the excess of staining was washed away with H2O. Gels were stored in

1% acetic acid/H2O at 4 °C. Protein bands of interest were excised, alkylated and subjected to tryptic digestion according to standard protocols. Further mass spectrometry analysis was performed as described previously [31].

Results CRISPR-Cas9-mediated targeting of the SMC5/6 complex In order to investigate the role of the SMC5/6 complex during different cellular processes such as DNA repair, we used the novel CRISPR-Cas9 system to generate cells lacking a fully functional SMC5/6 complex. U2OS cells were transfected with constructed CRISPR plasmids (pX458) to target SMC6 or NSMCE2. One day after transfection, GFP+ cells harboring CRISPR plasmids were sorted by fluorescence-activated cell sorting (FACS) and then cultured for five days (Figure 1A). To verify the occurrence of genome mutations in the sorted cell fractions, a Surveyor assay was performed based on PCR amplicons of the genomic DNA region around the targeting sites. Targeting of NSMCE2, but not SMC6, yielded fragments with the expected sizes after Surveyor nuclease digestion, indicating successful genome editing. The insertion/deletion (indel) occurrence brought by two 30 Chapter 2 individual single-guide RNAs (sgRNAs) targeting NSMCE2 was 17.2% and 16.6%, respectively (Figure 1B). To derive a monoclonal knockout cell line, FACS was conducted to deposit single GFP+ cells into 96-well plates. Single cells were then expanded for one to two months. Consistent with the results of Surveyor assay, all single cell-derived colonies appeared wild type for SMC6 after Sanger sequencing. In addition, for NSMCE2, we did not achieve complete NSMCE2-knockout after propagation. However, one monoclonal cell population (NSMCE2-1B) showed only one remaining wild type NSMCE2 allele, which was effectively mutated after a second round of transfection and single cell sorting using the NSMCE2-1B clone, resulting in the generation of a complete NSMCE2 null cell line (NSMCE2-1R, Figure 1C). Both Sanger sequencing and Western blot analysis showed the full null mutations in NSMCE2-1R cells (Figure 1C, D). Subsequently, by sequencing the 10 top-ranking potential off-target sites in the established NSMCE2 null cell line (Table S1), no off-target alterations were detected.

Figure 1: CRISPR-Cas9-mediated targeting of NSMCE2. (A) Left panel, transfected U2OS cells (GFP+) under bright field and fluorescence; right panel, FACS enrichment of GFP+ cells. Bar = 50 µm. (B) Expected cleavage bands (approximately 304 and 113 bp, arrowheads) generated by Surveyor nuclease digestion. For negative control (−) transfection of the pX458 plasmids without NSMCE2 sgRNA was performed. (C) Sequencing analysis for characterization of the CRISPR-Cas9-induced frameshift mutations. Red letters represent the 20-nt targeting sequences, while blue letters refer to the protospacer-adjacent motif (PAM). (D) Western blot analysis of the NSMCE2 protein in the final NSMCE2 null and WT cells. β-Actin was used as a loading control. NSMCE2 helps to resolve topological stress 31

Characterization of NSMCE2 null cells Morphologically, NSMCE2 null cells generally resemble WT cells, although NSMCE2 null cells clearly show more vacuoles, indicating increased cellular stress in the absence of NSMCE2 (Figure 2A). In addition, time-lapse imaging revealed a significant 1.37-fold increase in the cell cycle duration of NSMCE2 null cells (Figure 2B). When investigating the distribution of cells among different cell cycle phases, the DNA histogram of NSMCE2 null cells showed a recurring increase of approximately 10% in G0-1 phase compared to WT (Figure 2C). To investigate whether all of the NSMCE2 null cells participate in the cell cycle, we treated WT and NSMCE2 null cells with the M-phase blocking agent colcemid [32].

Although both WT and NSMCE2 null cells showed a rapid depletion of G0-1 cells after colcemid treatment (Figure 2D, E), which is in accordance with the rapid cycling nature of

U2OS cells, there were always ~10% more NSMCE2 null cells remaining in G0-1, and even after 96 h, a clear subpopulation of 16% remained (Figure 2D, E), indicating that these cells do not participate in the cell cycle. Protein levels of SMC5 and SMC6 were not evidently affected by the absence of NSMCE2 (Figure 2F).

32 Chapter 2

Figure 2: Analysis of NSMCE2 null cell growth characteristics. (A) Phase contrast images of WT and NSMCE2 null cells. The latter shows a large number of vacuoles in the cytoplasm (arrow). Bar = 20 µm. (B) Average cell cycle duration of WT and NSMCE2 null cells over multiple generations observed by live cell imaging. Data are presented as mean ± standard error of mean (SEM), n = 3. (C) Cell cycle phase distribution analysis of WT and NSMCE2 null cells by DNA histograms shows a 10% increase NSMCE2 null cells in G0-1. (D) G0-1 phase depletion by colcemid treatment. WT and NSMCE2 null cells were treated with colcemid for 0–96 h. In NSMCE2 null cells, a fraction of cells remained in

G0-1 even after 96 h. (E) Quantification for depletion of G0-1 cells with the time of colcemid treatment. Data are presented as mean ± SEM, n = 3. (F) Western blot analysis of SMC5 and SMC6 proteins in WT and NSMCE2 null cells. β-Actin was used as a loading control.

Irradiation-induced SMC6 foci formation occurs independent of NSMCE2 Because NSMCE2 is reported to be essential for the SMC5/6 function in the repair of DNA DSBs [33-37], we performed immunocytochemical stainings (ICC) for SMC6 on WT and NSMCE2 null cells at different time points after exposure to 1 Gy of ionizing irradiation (IR) (Figure 3A). Indeed, both in WT and NSMCE2 null cells, IR induced SMC6 foci that co- localize with DSBs (marked by γH2AX foci) and that gradually decrease post irradiation. This expression pattern of SMC6 was similar for both WT and NSMCE2 null cells. After we generated and applied a MatLab image analysis script that objectively isolates and quantifies the number of SMC6 and γH2AX foci in each cell, we found no difference in average number of γH2AX foci per nucleus at the chosen time points, indicating that both cell lines process DSBs in a similar fashion (Figure 3B). Importantly, not all γH2AX foci were represented by a SMC6 focus (Figure 3A). We therefore also determined the percentage of γH2AX foci positive for SMC6. We found that roughly 50% of the DSB sites were positive for SMC6 in both cell lines, indicating that the accumulation of SMC6 to sites of DSB damage is equally efficient in WT and NSMCE2 null cells (Figure 3C).

More etoposide-induced double-strand break (DSB) formation without NSMCE2 Next we interrogated whether the absence of NSMCE2 would influence the repair of etoposide-induced DNA damage. The cells were exposed to etoposide, a cytotoxic agent that acts by forming a complex with the DNA and the topoisomerase II enzyme [38, 39]. In normal conditions, type II topoisomerase releases superhelical stress and untangles chromosomes by creating transient DSBs, through which an unbroken DNA helix is transferred before ligation of the break, thereby averting incidents such as stalled replication forks or replication-induced joint molecules caused by DNA supercoiling [40]. Because NSMCE2 helps to resolve topological stress 33 etoposide prevents re-ligation of the DNA strands after transient DSBs, etoposide treatment will eventually lead to an increase of permanent DSBs [38, 39]. Indeed, when exposing the WT and NSMCE2 null cells to etoposide, DSBs (marked by γH2AX) became readily discernible (Figure 4A). However, in contrast to IR, the number of etoposide-induced DSBs increased significantly in the absence of NSMCE2 (p < 0.02 after 30 min and p < 0.03 after 3 h, Figure 4B). Similar to IR, not all etoposide-induced DSB foci contained SMC6, and the percentage of DSB foci containing SMC6 showed no significant difference between WT and NSMCE2 null cells after etoposide treatment (Figure 4C).

Figure 3: Ionizing irradiation (IR)-induced double-strand break (DSB) foci formation in the absence of NSMCE2. (A) WT and NSMCE2 null cells were subjected to 1 Gy of IR, fixed at different time points post IR (0 min: immediately after IR) and stained for γH2AX (a marker for DNA damage, red) and SMC6 (green). SMC6 localized to sites of DNA damage in both WT and NSMCE2 null cells. Bar = 5 μm. (B) Quantification of the average number of γH2AX foci in each cell. Data are presented as mean ± SEM, n = 3. (C) Quantification of the average number of γH2AX foci that overlap with a SMC6 focus. Data are presented as mean ± SEM, n = 3. 34 Chapter 2

Figure 4: Etoposide-induced DSB foci formation in the absence of NSMCE2. (A) WT and NSMCE2 null cells were subjected to 3 µM etoposide for 1 h, fixed at different time points post treatment (0 min: immediately after etoposide treatment) and stained for γH2AX (a marker for DNA damage, red) and SMC6 (green). SMC6 localized to sites of DNA damage in both WT and NSMCE2 null cells. Bar = 5 μm. (B) Quantification of the average number of γH2AX foci in each cell. Significantly more γH2AX foci were formed in NSMCE2 null cells. Data are presented as mean ± SEM, n = 3. (C) Quantification of the average number of γH2AX foci that overlap with a SMC6 focus. Data are presented as mean ± SEM, n = 3.

Absence of NSMCE2 affects survival upon etoposide-induced DSBs To measure the role of NSMCE2 in survival upon IR- or etoposide-induced DSBs, we subjected both NSMCE2 null and WT cells to a clonogenic assay [30]. Firstly, the plating efficiency, i.e., the percentage of plated single cells that develop into a cell colony of at least 50 cells determined in control conditions, was over three-fold lower in NSMCE2 null than in NSMCE2 helps to resolve topological stress 35

WT cells (average of 25% compared to 75%, respectively) (Figure 5A). Interestingly, the relative survival when cells were exposed to increasing doses of IR (0-8 Gy) only showed a small difference between the WT and NSMCE2 null cells, with the latter being slightly more sensitive (not statistically different though, Figure 5B). However, when exposed to 1 h of increasing doses of etoposide, NSMCE2 null cells showed a clearly reduced viability compared to WT cells (p < 0.05 in the case of 30 µM, Figure 5C).

Figure 5: Absence of NSMCE2 affects survival upon etoposide-induced DSBs. (A) Plating efficiency of WT and NSMCE2 null cells during clonogenic assays. The survival capacity of plated cells under non-challenged conditions was reduced in NSMCE2 null cells compared to WT. Data are presented as mean ± SEM, n = 8. (B) Clonogenic assay after increasing doses of IR. NSMCE2 null cells seemed to be slightly more sensitive to IR than WT cells (p > 0.05). Data are presented as mean ± SEM, n = 3. (C) Clonogenic assay after increasing doses of etoposide. NSMCE2 null cells were significantly more sensitive to etoposide than WT cells. Data are presented as mean ± SEM, n = 3.

36 Chapter 2

The SMC5/6 complex physically interacts with topoisomerase II α (TOP2A) To validate that the SMC5/6 complex is indeed involved in topoisomerase II-mediated relief of topological stress, we performed immunoprecipitations (IPs) using anti-SMC5 and SMC6 antibodies. Both SMC5 and SMC6 clearly co-immunoprecipitated with each other, and TOP2A clearly co-immunoprecipitated with SMC5 (Figure 6A), suggesting that the three proteins are physically linked. To unequivocally establish the physical interaction among these proteins, we conducted an additional IP with another antibody Abcam rabbit anti-SMC6, followed by mass spectrometry of the bands that could represent SMC5/6 and TOP2A (Figure 6B). We found that the band we presumed to represent full length SMC6 indeed contained the SMC6 protein. Not unexpectedly, because SMC5 and SMC6 are physically linked in the SMC5/6 complex, and SMC5 and SMC6 have equal sizes, we additionally identified the SMC5 protein to be present at the same height (Figure 6B). Interestingly, the lower running band (Figure 6B), which has been suggested to represent SMC6 lacking posttranslational modifications [9, 10, 41], was convincingly identified as SEC23IP, a protein previously shown to be involved in spermiogenesis [42]. Finally, we investigated the largest band (approximately 175 kDa) that could potentially represent TOP2A and that was also pulled down in this SMC6 IP (Figure 6B). Of the 23 identified peptides, 3 are homologous between TOP2A and TOP2B, and 20 are unique to TOP2A. Because no specific peptides for TOP2B were identified we conclude that this band indeed represented the protein TOP2A (Data S1-S3).

Figure 6: The SMC5/6 complex physically interacts with TOP2A. (A) Immunoprecipitations (IPs) and Western blot analysis of SMC5 and SMC6. Both SMC5 and SMC6 co-immunoprecipitated with each other. Additionally, TOP2A co-immunoprecipitated with SMC5. (B) For IP followed by mass spectrometry, Abcam rabbit anti-SMC6 was used. SMC6-IP material was loaded on a 4%-12% bis-tris gradient gel and stained with coomassie blue. Arrows indicate the bands that were isolated for mass spectrometry analysis (green: potential SMC5/6 proteins; orange: potential TOP2A protein).

Discussion To study the SMC5/6 complex in an experimental setup, we used the novel CRISPR- Cas9 system to target the SMC6 and NSMCE2 genes that encode the SMC6 and NSMCE2 subunits of the SMC5/6 complex, respectively. For NSMCE2, we did not get a complete NSMCE2 helps to resolve topological stress 37 knockout after the first round of transfection and monoclonal isolation. The results were not unexpected, given that U2OS is a cell line with chromosome counts in the hypertriploid range, and that generating a true knockout is technically challenging since it requires disruption of all functional copies of the gene. Consequently, a second round of gene targeting and single cell expansion was performed. Eventually we established a bona fide cell line devoid of NSMCE2. In the case of SMC6, however, we were unable to generate any CRISPR-modified cells. Previous papers show that ablation of SMC6 resulted in embryonic lethality in mice [26] and that conditional knockout of SMC5 in mouse embryonic stem cells induced apoptosis [28]. Therefore, the failure to generate SMC6-deprived clones in our studies most likely reflects that SMC6 is also essential for cell survival in humans. To minimize the CRISPR-Cas9-induced off-target effects, we selected sgRNAs with the highest specificity to coding exons of SMC6 and NSMCE2. Later, we analyzed the 10 top- ranking potential off-target sites in the established NSMCE2 null cell line by sequencing, and detected no off-target mutations. The results are in line with recent papers showing low frequency of CRISPR-Cas9-induced off-target alterations in human cells [43, 44]. Hence, although the possibility of off-target effects in modified cells could not be thoroughly excluded, the differential phenotypes between NSMCE2 null and WT cells are thought to authentically mirror gene functions. We found clear differences in growth characteristics of the NSMCE2 null cells compared to WT cells. The mutated cells have a prolonged cell cycle, and a clear larger portion of the cells arrest in the G0-1 phase. Because this latter fraction is still equally present after an extensive time in culture and multiple passages, it must be continuously supplemented by cells exiting the cell cycle. Considering the differences in phenotypes, i.e., the presence of the G0-1-arrested cells, the overall slower growth rate and the reduced plating efficiency of NSMCE2 null cells, it is plausible that the absence of NSMCE2 is not immediately lethal to the majority of the cells, but poses growth challenges in normal culture conditions that will ultimately arrest the cells. These data are in line with a recent study performed in human breast cancer cells (MCF-7), in which the depletion of NSMCE2 by RNA interference (RNAi) caused a slower cell growth and increased percentage of G1 phase cells (70% vs. 55%-59% in control) [45]. Interestingly, depletion of NSMCE2 led to reduced levels of E2F1 protein and its downstream target genes that are required for G1-S transition. This decreased growth rate was rescued by ectopic expression of Flag-NSMCE2 but not its SUMO ligase-inactive mutant, suggesting that the SUMO ligase activity of NSMCE2 ensures proper G1-S transition in these human cancer cells [45]. 38 Chapter 2

Although NSMCE2 is frequently linked to the DNA repair function of the SMC5/6 complex [33-37, 46, 47], the survival capacity of NSMCE2 null cells is only slightly affected after increasing doses of IR. Likewise, IR-induced DSBs marked by γH2AX appear with similar kinetic properties in WT and NSMCE2 null cells. We therefore conclude that NSMCE2 is not crucial for the repair of IR-induced DSBs. Nevertheless, the recruitment of SMC6 to DSBs early post irradiation suggests that the SMC5/6 complex is involved in the early repair of DSBs. In addition, NSMCE2 null cells did not display a differential survival rate in response to cisplatin, a cytostatic agent causing DNA adducts and crosslinks that are generally repaired by nucleotide excision repair (NER, Figure S1). In contrast to the effects of IR or cisplatin, exposure to increasing doses of the topoisomerase inhibitor etoposide does impair the survival capacity of NSMCE2 null cells. In normal conditions, type II topoisomerase releases superhelical stress and untangles chromosomes by creating transient DSBs, through which an unbroken DNA helix is transferred before ligation of the break, in order to unwind the DNA double helix to prevent supercoiling [40, 48]. Etoposide acts as a cytotoxic agent by forming a complex with the DNA and the topoisomerase II enzyme, thereby preventing re-ligation of the DNA strands after transient DSBs. Thus, etoposide treatment will ultimately lead to an increase of permanent DSBs [38, 39]. The increased sensitivity to etoposide of NSMCE2 null cells is intriguing, especially in the light of the interaction between SMC5/6 and TOP2A found in this study. Interplay between TOP2A and the SMC5/6 complex has been suggested by several studies in mouse and yeast [13, 49-52], but physical interaction between them is not confirmed. Recently, it has been reported that Smc5/6 immunoprecipitated with the type II topoisomerase in budding yeast [53]. Here, we for the first time unequivocally demonstrate that TOP2A is indeed associated with the SMC5/6 complex in human cells. The absence of a detectable band for TOP2A in the SMC6 guinea pig (GP) IP is most likely due to the lower efficiency of the SMC6 GP antibody, which is supported by the observation that the SMC5 IP generated more SMC6 protein than the SMC6 GP IP. Topological tension is conceived when DNA molecules become supercoiled, for instance preceding an advancing replication fork during chromosome duplication. This positive supercoiling has to be removed in order for the replication fork to proceed. One mechanism to avoid accumulation of supercoiled DNA ahead of the fork is to allow the replication fork to proceed in a rotating manner following the turn of the helix. This will indeed prevent the accumulation of supercoils ahead of the fork, but will simultaneously induce the formation of sister chromatid intertwinings (SCI) behind the fork. Another way to release the supercoiling is through topoisomerases, the enzymes creating single-strand nicks and NSMCE2 helps to resolve topological stress 39 double-strand breaks. In this case TOP1 and TOP2 nick the DNA double helix ahead of the fork, thereby allowing the unwinding of the supercoiled helix and the release of topological tension. Because topological tension increases with the length of the chromosome, topoisomerases are supposed to be more important to replication of longer chromosomes. Indeed, budding yeast cells lacking functional topoisomerase I show a length-dependent delay in replication [54]. In line with the observed interaction between SMC5/6 and TOP2A, the association of budding yeast Smc5/6 with chromosomes during S-phase is also linearly correlated with chromosome length, indicating that Smc5/6 somehow measures chromosome length, probably by sensing topological tension [54, 55]. Moreover, Smc5/6 seems to also play a role in topological strain release, since budding yeast cells lacking functional Smc6 or Nse2 show a delay in replication similar to Top2 mutants [54, 55]. Our own data, showing that inhibition of topoisomerase activity has a more profound effect on cells harboring an impaired SMC5/6 complex, together with physical interaction between the SMC5/6 complex and TOP2A, further corroborate the presumed co-operation between TOP2A and SMC5/6 at replication forks. Considering the two mechanisms of tension release, SMC5/6 could function both before and after the replication fork. Ahead of the fork, SMC5/6 could be responsible for the correct repair of the TOP2A-induced DSBs. When the ligase function of TOP2A is inhibited by etoposide, re-initiation of replication might rely more on SMC5/6, which will be challenged when NSMCE2 deletion impairs SMC5/6 function. On the other side of the fork, SMC5/6 might be required to stabilize the SCIs, as proposed previously [54]. In this model, SMC5/6 associates to SCIs, thereby fixating them and allowing fork rotation, and reducing topological tension. In parallel, budding yeast Smc5/6 is also involved in the actual resolving recombination intermediates in order to prevent toxic chromosome structures [56]. In addition, budding yeast Top2 is also essential for the removal of SCIs that would otherwise lead to segregation errors during the subsequent M-phase [57-59]. Since both induction and removal of SCIs involves transferring one DNA double helix through another via transient formation and repair of a DSB [48], it is likely that SMC5/6 is also working together with TOP2A at the level of SCIs. Furthermore, in budding yeast, Top2 activity relies on sumoylation and failure to sumoylate Top2 disrupts the ability of Top2 to separate replicated chromosomes [60]. Of note, human TOP2 is also found conjugated to SUMO [61]. However, RANBP2 seems to be the major SUMO E3 ligase for TOP2A in mice [62], and a mutation compromising NSMCE2 sumoylation activity does not affect murine lifespan [27]. Nevertheless, we cannot rule out that SMC5/6 function at the replication fork might involve NSMCE2-mediated sumoylation of TOP2A, which could explain the interaction between SMC5/6 and TOP2A and the effects seen in NSMCE2 null cells. 40 Chapter 2

Our findings, demonstrating a physical interaction between SMC5/6 and TOP2A and an increased sensitivity of NSMCE2 null cells to etoposide, suggest that the SMC5/6 complex helps to resolve topological stress. Since this physical interaction is even present in cells that are not challenged by IR or cytotoxic agents, SMC5/6 and TOP2A seem already to function together during S-phase under normal non-challenged circumstances. In this respect, it is plausible that the fraction of NSMCE2 null cells arresting in G0-1 phase is actually a representation of cells with stalled or collapsed replication forks very early in the replication process, which occurs naturally, yet cannot be resolved properly due to the lack of NSMCE2. While in WT cells these replication forks are normally repaired and restarted by SMC5/6, the absence of NSMCE2 first induces a delay in early replication progression, explaining the increased cell cycle duration of NSMCE2 null cells. Subsequently, residual repair defects may trigger cells to eventually arrest in G0. In this respect, several studies in yeast have suggested that Smc5/6-mutated cells will undergo cell division despite the presence of chromosomal abnormalities caused by defective DNA repair mechanisms [33, 49, 52, 63, 64]. However, over time, the amount of chromosomal abnormalities within a cell will accumulate, eventually leading to cell cycle arrest. In conclusion, in the light of current literature and the data we present here, we propose that the SMC5/6 complex functions in resolving TOP2A-mediated recombination intermediates endogenously generated early during DNA replication in human cells.

Acknowledgments We thank Berend Hooibrink and Daisy Picavet of the core facility Cellular Imaging of the Academic Medical Center (AMC) for assistance with and use of their equipment for FACS analysis and confocal microscopy. Furthermore, we thank Klaas Franken, Hans Rodermond and Bregje van Oorschot for assistance with and use of their 137Cs source for IR. Finally, we thank Philip W. Jordan for generously providing the TOP2A antibody and for fruitful discussions. This study has been supported by an AMC Fellowship, the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (CIG 293765) to Geert Hamer and the China Scholarship Counsel (CSC) number 201306300081 to Yi Zheng.

Author contributions Dideke E. Verver, Yi Zheng, Jan Stap and Geert Hamer conceived and designed the experiments. Dideke E. Verver, Yi Zheng, Dave Speijer, Ron Hoebe and Henk L. Dekker performed the experiments. Dideke E. Verver, Yi Zheng, Jan Stap and Geert Hamer NSMCE2 helps to resolve topological stress 41 analyzed the data. Dideke E. Verver, Yi Zheng, Sjoerd Repping and Geert Hamer wrote the manuscript.

Conflicts of interest The authors declare no conflict of interest.

Abbreviations SMC Structural Maintenance of Chromosomes NSMCE Non-SMC Element DSB DNA double-strand break IR Ionizing Radiation TOP Topoisomerase SCI Sister chromatid intertwining

42 Chapter 2

References

1. Nasmyth K, Haering CH. The structure and function of SMC and kleisin complexes. Annu Rev Biochem 2005; 74:595-648. 2. Peters JM, Tedeschi A, Schmitz J. The cohesin complex and its roles in chromosome biology. Genes Dev 2008; 22:3089-3114. 3. Hirano T. At the heart of the chromosome: SMC proteins in action. Nat Rev Mol Cell Biol 2006; 7:311-322. 4. Terret ME, Sherwood R, Rahman S, Qin J, Jallepalli PV. Cohesin acetylation speeds the replication fork. Nature 2009; 462:231-234. 5. Wendt KS, Peters JM. How cohesin and CTCF cooperate in regulating gene expression. Chromosome Res 2009; 17:201-214. 6. De Piccoli G, Cortes-Ledesma F, Ira G, Torres-Rosell J, Uhle S, Farmer S, Hwang JY, Machin F, Ceschia A, McAleenan A, Cordon-Preciado V, Clemente-Blanco A, et al. Smc5-Smc6 mediate DNA double-strand-break repair by promoting sister-chromatid recombination. Nat Cell Biol 2006; 8:1032- 1034. 7. Potts PR. The Yin and Yang of the MMS21-SMC5/6 SUMO ligase complex in homologous recombination. DNA Repair (Amst) 2009; 8:499-506. 8. Wu N, Yu H. The Smc complexes in DNA damage response. Cell Biosci 2012; 2:5. 9. Verver DE, Langedijk NS, Jordan PW, Repping S, Hamer G. The SMC5/6 complex is involved in crucial processes during human spermatogenesis. Biol Reprod 2014; 91:22. 10. Verver DE, van Pelt AM, Repping S, Hamer G. Role for rodent Smc6 in pericentromeric heterochromatin domains during spermatogonial differentiation and meiosis. Cell Death Dis 2013; 4:e749. 11. Verver DE, Hwang GH, Jordan PW, Hamer G. Resolving complex chromosome structures during meiosis: versatile deployment of Smc5/6. Chromosoma 2016; 125:15-27. 12. Copsey A, Tang S, Jordan PW, Blitzblau HG, Newcombe S, Chan AC, Newnham L, Li Z, Gray S, Herbert AD, Arumugam P, Hochwagen A, et al. Smc5/6 coordinates formation and resolution of joint molecules with chromosome morphology to ensure meiotic divisions. PLoS Genet 2013; 9:e1004071. 13. Gomez R, Jordan PW, Viera A, Alsheimer M, Fukuda T, Jessberger R, Llano E, Pendas AM, Handel MA, Suja JA. Dynamic localization of SMC5/6 complex proteins during mammalian meiosis and mitosis suggests functions in distinct chromosome processes. J Cell Sci 2013; 126:4239-4252. 14. Bickel JS, Chen L, Hayward J, Yeap SL, Alkers AE, Chan RC. Structural maintenance of chromosomes (SMC) proteins promote homolog-independent recombination repair in meiosis crucial for germ cell genomic stability. PLoS Genet 2010; 6:e1001028. 15. Lilienthal I, Kanno T, Sjogren C. Inhibition of the Smc5/6 Complex during Meiosis Perturbs Joint Molecule Formation and Resolution without Significantly Changing Crossover or Non-crossover Levels. PLoS Genet 2013; 9:e1003898. NSMCE2 helps to resolve topological stress 43

16. Xaver M, Huang L, Chen D, Klein F. Smc5/6-mms21 prevents and eliminates inappropriate recombination intermediates in meiosis. PLoS Genet 2013; 9:e1004067. 17. Farmer S, San-Segundo PA, Aragón L. The smc5-smc6 complex is required to remove chromosome junctions in meiosis. PLoS One 2011; 6:e20948. 18. Torres-Rosell J, Sunjevaric I, De Piccoli G, Sacher M, Eckert-Boulet N, Reid R, Jentsch S, Rothstein R, Aragon L, Lisby M. The Smc5-Smc6 complex and SUMO modification of Rad52 regulates recombinational repair at the ribosomal gene locus. Nat Cell Biol 2007; 9:923-931. 19. Agostinho A, Meier B, Sonneville R, Jagut M, Woglar A, Blow J, Jantsch V, Gartner A. Combinatorial regulation of meiotic holliday junction resolution in C. elegans by HIM-6 (BLM) helicase, SLX-4, and the SLX-1, MUS-81 and XPF-1 nucleases. PLoS Genet 2013; 9:e1003591. 20. Hong Y, Sonneville R, Agostinho A, Meier B, Wang B, Blow JJ, Gartner A. The SMC-5/6 Complex and the HIM-6 (BLM) Helicase Synergistically Promote Meiotic Recombination Intermediate Processing and Chromosome Maturation during Caenorhabditis elegans Meiosis. PLoS Genet 2016; 12:e1005872. 21. O'Neil NJ, Martin JS, Youds JL, Ward JD, Petalcorin MI, Rose AM, Boulton SJ. Joint molecule resolution requires the redundant activities of MUS-81 and XPF-1 during Caenorhabditis elegans meiosis. PLoS Genet 2013; 9:e1003582. 22. Wehrkamp-Richter S, Hyppa RW, Prudden J, Smith GR, Boddy MN. Meiotic DNA joint molecule resolution depends on Nse5-Nse6 of the Smc5-Smc6 holocomplex. Nucleic Acids Res 2012; 40:9633-9646. 23. Carter SD, Sjogren C. The SMC complexes, DNA and chromosome topology: right or knot? Crit Rev Biochem Mol Biol 2012; 47:1-16. 24. Jeppsson K, Kanno T, Shirahige K, Sjogren C. The maintenance of chromosome structure: positioning and functioning of SMC complexes. Nat Rev Mol Cell Biol 2012; 15:601-614. 25. Langston RE, Weinert T. Nifty Alleles, a Plethora of Interactions, and Imagination Advance Understanding of Smc5/6's Roles with Chromosomes. Mol Cell 2015; 60:832-833. 26. Ju L, Wing J, Taylor E, Brandt R, Slijepcevic P, Horsch M, Rathkolb B, Racz I, Becker L, Hans W, Adler T, Beckers J, et al. SMC6 is an essential gene in mice, but a hypomorphic mutant in the ATPase domain has a mild phenotype with a range of subtle abnormalities. DNA Repair (Amst) 2013; 12:356-366. 27. Jacome A, Gutierrez-Martinez P, Schiavoni F, Tenaglia E, Martinez P, Rodriguez-Acebes S, Lecona E, Murga M, Mendez J, Blasco MA, Fernandez-Capetillo O. NSMCE2 suppresses cancer and aging in mice independently of its SUMO ligase activity. Embo J 2016; 34:2604-2619. 28. Pryzhkova MV, Jordan PW. Conditional mutation of Smc5 in mouse embryonic stem cells perturbs condensin localization and mitotic progression. J Cell Sci 2016; 129:1619-1634. 29. Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, Zhang F. Genome engineering using the CRISPR-Cas9 system. Nat Protoc 2013; 8:2281-2308. 30. Franken NA, Rodermond HM, Stap J, Haveman J, van Bree C. Clonogenic assay of cells in vitro. Nat Protoc 2006; 1:2315-2319. 44 Chapter 2

31. Stax MJ, Mouser EE, van Montfort T, Sanders RW, de Vries HJ, Dekker HL, Herrera C, Speijer D, Pollakis G, Paxton WA. Colorectal mucus binds DC-SIGN and inhibits HIV-1 trans-infection of CD4+ T-lymphocytes. PLoS One 2015; 10:e0122020. 32. Sluder G. Role of spindle microtubules in the control of cell cycle timing. J Cell Biol 1979; 80:674-691. 33. Andrews EA, Palecek J, Sergeant J, Taylor E, Lehmann AR, Watts FZ. Nse2, a component of the Smc5-6 complex, is a SUMO ligase required for the response to DNA damage. Mol Cell Biol 2005; 25:185-196. 34. Kliszczak M, Stephan AK, Flanagan AM, Morrison CG. SUMO ligase activity of vertebrate Mms21/Nse2 is required for efficient DNA repair but not for Smc5/6 complex stability. DNA Repair (Amst) 2012; 11:799-810. 35. McDonald WH, Pavlova Y, Yates JR, 3rd, Boddy MN. Novel essential DNA repair proteins Nse1 and Nse2 are subunits of the fission yeast Smc5-Smc6 complex. J Biol Chem 2003; 278:45460- 45467. 36. Rai R, Varma SP, Shinde N, Ghosh S, Kumaran SP, Skariah G, Laloraya S. Small ubiquitin- related modifier ligase activity of Mms21 is required for maintenance of chromosome integrity during the unperturbed mitotic cell division cycle in Saccharomyces cerevisiae. J Biol Chem 2011; 286:14516-14530. 37. Zhao X, Blobel G. A SUMO ligase is part of a nuclear multiprotein complex that affects DNA repair and chromosomal organization. Proc Natl Acad Sci U S A 2005; 102:4777-4782. 38. Chen GL, Yang L, Rowe TC, Halligan BD, Tewey KM, Liu LF. Nonintercalative antitumor drugs interfere with the breakage-reunion reaction of mammalian DNA topoisomerase II. J Biol Chem 1984; 259:13560-13566. 39. Nitiss JL. DNA topoisomerase II and its growing repertoire of biological functions. Nat Rev Cancer 2009; 9:327-337. 40. Wang JC. DNA topoisomerases. Annu Rev Biochem 1996; 65:635-692. 41. Taylor EM, Moghraby JS, Lees JH, Smit B, Moens PB, Lehmann AR. Characterization of a novel human SMC heterodimer homologous to the Schizosaccharomyces pombe Rad18/Spr18 complex. Mol Biol Cell 2001; 12:1583-1594. 42. Arimitsu N, Kogure T, Baba T, Nakao K, Hamamoto H, Sekimizu K, Yamamoto A, Nakanishi H, Taguchi R, Tagaya M, Tani K. p125/Sec23-interacting protein (Sec23ip) is required for spermiogenesis. FEBS Lett 2011; 585:2171-2176. 43. Smith C, Gore A, Yan W, Abalde-Atristain L, Li Z, He C, Wang Y, Brodsky RA, Zhang K, Cheng L, Ye Z. Whole-genome sequencing analysis reveals high specificity of CRISPR/Cas9 and TALEN-based genome editing in human iPSCs. Cell Stem Cell 2014; 15:12-13. 44. Veres A, Gosis BS, Ding Q, Collins R, Ragavendran A, Brand H, Erdin S, Cowan CA, Talkowski ME, Musunuru K. Low incidence of off-target mutations in individual CRISPR-Cas9 and TALEN targeted human stem cell clones detected by whole-genome sequencing. Cell Stem Cell 2014; 15:27-30. NSMCE2 helps to resolve topological stress 45

45. Ni HJ, Chang YN, Kao PH, Chai SP, Hsieh YH, Wang DH, Fong JC. Depletion of SUMO ligase hMMS21 impairs G1 to S transition in MCF-7 breast cancer cells. Biochim Biophys Acta 2012; 1820:1893-1900. 46. Pebernard S, McDonald WH, Pavlova Y, Yates JR, Boddy MN. Nse1, Nse2, and a novel subunit of the Smc5-Smc6 complex, Nse3, play a crucial role in meiosis. Mol Biol Cell 2004; 15:4866- 4876. 47. Raschle M, Smeenk G, Hansen RK, Temu T, Oka Y, Hein MY, Nagaraj N, Long DT, Walter JC, Hofmann K, Storchova Z, Cox J, et al. DNA repair. Proteomics reveals dynamic assembly of repair complexes during bypass of DNA cross-links. Science 2015; 348:1253671. 48. Wang JC. Cellular roles of DNA topoisomerases: a molecular perspective. Nat Rev Mol Cell Biol 2002; 3:430-440. 49. Harvey SH, Sheedy DM, Cuddihy AR, O'Connell MJ. Coordination of DNA damage responses via the Smc5/Smc6 complex. Mol Cell Biol 2004; 24:662-674. 50. Takahashi Y, Dulev S, Liu X, Hiller NJ, Zhao X, Strunnikov A. Cooperation of sumoylated chromosomal proteins in rDNA maintenance. PLoS Genet 2008; 4:e1000215. 51. Uemura T, Yanagida M. Isolation of type I and II DNA topoisomerase mutants from fission yeast: single and double mutants show different phenotypes in cell growth and chromatin organization. Embo J 1984; 3:1737-1744. 52. Verkade HM, Bugg SJ, Lindsay HD, Carr AM, O'Connell MJ. Rad18 is required for DNA repair and checkpoint responses in fission yeast. Mol Biol Cell 1999; 10:2905-2918. 53. Kanno T, Berta DG, Sjogren C. The Smc5/6 Complex Is an ATP-Dependent Intermolecular DNA Linker. Cell Rep 2015; 12:1471-1482. 54. Kegel A, Betts-Lindroos H, Kanno T, Jeppsson K, Strom L, Katou Y, Itoh T, Shirahige K, Sjogren C. Chromosome length influences replication-induced topological stress. Nature 2011; 471:392-396. 55. Lindroos HB, Strom L, Itoh T, Katou Y, Shirahige K, Sjogren C. Chromosomal association of the Smc5/6 complex reveals that it functions in differently regulated pathways. Mol Cell 2006; 22:755- 767. 56. Menolfi D, Delamarre A, Lengronne A, Pasero P, Branzei D. Essential Roles of the Smc5/6 Complex in Replication through Natural Pausing Sites and Endogenous DNA Damage Tolerance. Mol Cell 2015; 60:835-846. 57. Bermejo R, Doksani Y, Capra T, Katou YM, Tanaka H, Shirahige K, Foiani M. Top1- and Top2-mediated topological transitions at replication forks ensure fork progression and stability and prevent DNA damage checkpoint activation. Genes Dev 2007; 21:1921-1936. 58. Kim RA, Wang JC. Function of DNA topoisomerases as replication swivels in Saccharomyces cerevisiae. J Mol Biol 1989; 208:257-267. 59. Spell RM, Holm C. Nature and distribution of chromosomal intertwinings in Saccharomyces cerevisiae. Mol Cell Biol 1994; 14:1465-1476. 46 Chapter 2

60. Bachant J, Alcasabas A, Blat Y, Kleckner N, Elledge SJ. The SUMO-1 isopeptidase Smt4 is linked to centromeric cohesion through SUMO-1 modification of DNA topoisomerase II. Mol Cell 2002; 9:1169-1182. 61. Mao Y, Desai SD, Liu LF. SUMO-1 conjugation to human DNA topoisomerase II isozymes. J Biol Chem 2000; 275:26066-26073. 62. Dawlaty MM, Malureanu L, Jeganathan KB, Kao E, Sustmann C, Tahk S, Shuai K, Grosschedl R, van Deursen JM. Resolution of sister centromeres requires RanBP2-mediated SUMOylation of topoisomerase IIalpha. Cell 2008; 133:103-115. 63. Ampatzidou E, Irmisch A, O'Connell MJ, Murray JM. Smc5/6 is required for repair at collapsed replication forks. Mol Cell Biol 2006; 26:9387-9401. 64. Miyabe I, Morishita T, Hishida T, Yonei S, Shinagawa H. Rhp51-dependent recombination intermediates that do not generate checkpoint signal are accumulated in Schizosaccharomyces pombe rad60 and smc5/6 mutants after release from replication arrest. Mol Cell Biol 2006; 26:343-353. NSMCE2 helps to resolve topological stress 47

Supplementary materials

Table S1: The selected off-target sites and the corresponding PCR primers. Potential off-target Genomic loci Mis PCR primers sites (GRCh38/hg38) matches ACCCAAATCTGTATC 62065416-62065438, F: GTCCCTCCATCTTGGTGCCT 1, 2, 7, 8 AACTC AGG chromosome 11 R: GGGTCTTTGCTGCCTGTGA CTCTTATCCTGTATC 43274942-43274964, F: CACTGACAACAGGCATGAAAT 1, 4, 5, 7 AACTC AAG chromosome 2 R: CTGGAGACTGAGGCAGGAGA TTTCTATTCTGTATC 153197613-153197635, F: ATTGGGCCTTATGAACTGATTC 3, 5, 7, 8 AACTC AAG chromosome 7 R: TGGTCTACGCAGGGTAAGGATA TGCCAAGGCTGTAT 110987269-110987291, F: ACTGCTTGGAACAGTGAACATG 2, 8, 15 GAACTC AGG chromosome 2 (in gene) R: GCTGAGACTGATGAGCGATAAA TCCAAAGTCTGTATC 2809500-2809522, 2, 4, 8, F: CAGCCAAACCATATCATTCTGT AACTT TAG chromosome 8 20 R: TGTTTTCATGTTTGTGGCAGTG TTCCAAGACTATATC 114757228-114757250, F: GTATTGCAGCAAGCCATTACC 8, 11, 20 AACTA AAG chromosome 11 R: AAGAATCTGCTCTGGAGGGAG TTCCTGGCCTGTATC 52436381-52436403, F: ACGGAACCAGGTGAAGGAA 5, 6, 19 AACAC AAG chromosome 12 R: TTGGCACTTGGAGCGGTAG TTTGAAGTCTGTATC 104356169-104356191, 3, 4, 8, F: CACTCTTACTTTGTTCCCCACA ATCTC AAG chromosome 5 17 R: CCTCACTTGCCTTGCCTATT TACTAAACATGTATC 31343808-31343830, F: GCTGGCGGGTGCAATTAGT 2, 4, 7, 9 AACTC AGG chromosome X R: AGAGCAAGACCCTGACCCTAA GCCCAAGCCAGTAT 128369750-128369772, 1, 2, 10, F: CGGAAAGTGGGAGTAAGAAATC CACCTC AAG chromosome 5 17 R: GCTCTAATCACTGGCTATGCTAT 48 Chapter 2

Figure S1: Clonogenic assay after increasing doses of cisplatin. No clear difference in sensitivity to cisplatin could be detected between WT and NSMCE2 null cells.

Data S1-S3 can be found at www.mdpi.com/1422-0067/17/11/1782/s1.

Chapter 3

Trivial role for NSMCE2 during in vitro proliferation and differentiation of male germline stem cells

Yi Zheng Aldo Jongejan Callista L. Mulder Sebastiaan Mastenbroek Sjoerd Repping Yinghua Wang Jinsong Li Geert Hamer

Reproduction 2017 Sep;154(3):81-95 50 Chapter 3

Abstract Spermatogenesis, starting with spermatogonial differentiation, is characterized by ongoing and dramatic alterations in composition and function of chromatin. Failure to maintain proper chromatin dynamics during spermatogenesis may lead to mutations, chromosomal aberrations or aneuploidies. When transmitted to the offspring, these can cause infertility or congenital malformations. The structural maintenance of chromosomes (SMC) 5/6 protein complex has recently been described to function in chromatin modeling and genomic integrity maintenance during spermatogonial differentiation and meiosis. Among the subunits of the SMC5/6 complex, non-SMC element 2 (NSMCE2) is an important small ubiquitin-related modifier (SUMO) ligase. NSMCE2 has been reported to be essential for mouse development, prevention of cancer and aging in adult mice and topological stress relief in human somatic cells. By using in vitro cultured primary mouse spermatogonial stem cells (SSCs), referred to as male germline stem (GS) cells, we investigated the function of NSMCE2 during spermatogonial proliferation and differentiation. We first optimized a protocol to generate genetically modified GS cell lines using CRISPR-Cas9 and generated an Nsmce2-/- GS cell line. Using this Nsmce2-/- GS cell line, we found that NSMCE2 was dispensable for proliferation, differentiation and topological stress relief in mouse GS cells. Moreover, RNA sequencing analysis demonstrated that the transcriptome was only minimally affected by the absence of NSMCE2. Only differential expression of Sgsm1 appeared highly significant, but with SGSM1 protein levels being unaffected without NSMCE2. Hence, despite the essential roles of NSMCE2 in somatic cells, chromatin integrity maintenance seems differentially regulated in the germline. NSMCE2 in male germline stem cells 51

Introduction The process of spermatogenesis is characterized by ongoing and dramatic alterations in composition and function of chromatin. Chromatin is the supra-molecular complex, consisting of DNA and proteins, which packages, shapes and orchestrates the genome and safeguards genomic stability. Incorrect spatio-temporal organization of chromatin can initiate germ cell apoptosis, leading to spermatogenic arrest and male infertility [1]. If all spermatogenic arrest mechanisms fail, incorrect chromatin architecture can even cause chromosomal aberrations or aneuploidies, leading to congenital abnormalities in the offspring [2]. Numerous chromatin-based processes, such as cell cycle progression, cellular differentiation and genomic integrity maintenance, are to a large extent modulated by the structural maintenance of chromosomes (SMC) protein complexes: SMC1/3 (cohesin), SMC2/4 (condensin) and SMC5/6. Besides SMC5 and SMC6, the SMC5/6 complex contains several non-SMC elements (NSMCEs), including NSMCE2. Together with these NSMCEs, the SMC5/6 complex has a ring-like structure large enough to hold two double-stranded DNA molecules together. This property of the SMC5/6 complex is pivotal for recombination- mediated DNA damage repair [3-5] and resolving replication-induced topological stress in yeast [6-8]. Of the four NSMCEs in mammals, NSMCE2 specifically associates with SMC5, where it exhibits a C-terminal SP-RING domain with an E3 small ubiquitin-related modifier (SUMO) ligase activity. This SUMO ligase activity of NSMCE2 is required for the DNA damage repair activity of the SMC5/6 complex [9-11]. Using various model organisms, the function of the SMC5/6 complex during meiosis has been extensively investigated and reviewed [12]. Meiotic processes involving the SMC5/6 complex include chromosome segregation [13-17], homologous chromosome synapsis [13-15, 18-20] and meiotic sex chromosome inactivation [14, 16]. Very likely, the SMC5/6 complex prevents dangerous and error-prone recombination events in highly repetitive, densely packed genomic regions such as the rDNA and pericentromeric heterochromatin [14, 16, 19-22]. Aberrant recombination between such repetitive sequences would otherwise interfere with the tightly regulated process of meiotic recombination, for instance, by causing intra-chromosomal recombination events. Indeed, in yeast and C. elegans, mutations in Smc5, Smc6 or the SUMO ligase domain of Nsmce2 lead to the accumulation of toxic joint molecules caused by failure to resolve meiotic recombination intermediates [13, 19, 20, 23-26]. In mitotically dividing yeast, the SMC5/6 complex is essential for the maintenance of replication fork stability and the prevention of replication-induced topological stress [6-8]. Consistently, depletion of Smc5 in mouse embryonic stem (ES) cells leads to accumulation 52 Chapter 3

of cells in G2 and subsequent mitotic failure and apoptosis [27]. Using a human osteosarcoma cell line (U2OS), we have recently shown that SMC6 interacts with DNA topoisomerase II α (TOP2A) [28]. TOP2A is a topoisomerase that prevents supercoiling of replicating DNA [29, 30]. Moreover, we have shown that the CRISPR-Cas9-mediated removal of NSMCE2 in these cells led to increased sensitivity to the topoisomerase inhibitor etoposide [28]. In the mouse and human, the SMC6 protein is most highly expressed in the testis where it appears to be involved in spermatogonial differentiation [15, 16] and meiosis [14-16]. Likewise, also NSMCE2 has been identified to be expressed in developing mouse male germ cells, from spermatogonia to round spermatids [31]. Recently, a study using a conditional knock-out (KO) mouse model showed essential roles for SMC5/6 during meiotic chromosome segregation [17]. However, suitable KO models to study the SMC5/6 complex in mitotically dividing spermatogonia are currently not available. In mice, complete ablation of SMC6 or NSMCE2 results in embryonic lethality [31, 32]. Hence, the role of SMC5/6 or NSMCE2 in spermatogonia remains largely unknown. Male germline stem (GS) cells, initially termed by Shinohara’s group, refer to the cultured spermatogonial stem cells (SSCs) able to propagate in vitro for over 2 years without losing SSC properties [33]. In the current study, we first optimized a protocol to generate genetically modified mouse GS cell lines using CRISPR-Cas9. By applying this optimized protocol, we generated an Nsmce2-/- GS cell line. Using this GS cell line, we studied the role of Nsmce2 during in vitro spermatogonial proliferation and differentiation, gene expression and the spermatogonial response to topological stress.

Materials and methods Animal use and care Neonatal (4-5 d.p.p) DBA/2J (Charles River) male mice were used for GS cell isolation. To acquire neonatal testis materials, donor mice were first anesthetized by 4% isoflurane total body anesthesia followed by killing by decapitation and inactivation of the brain. Testes were collected and the tunica albuginea was removed. Testicular tissues were cryopreserved in supplemented MEM (Gibco, Thermo Fisher Scientific) containing 20% fetal bovine serum (FBS) and 8% DMSO in a Coolcell freezing device and stored in liquid nitrogen (-196°C) for future GS cell isolation. All animal procedures were in accordance with and approved by the animal ethical committee of the Academic Medical Center, University of Amsterdam.

NSMCE2 in male germline stem cells 53

Design of Nsmce2- single-guide (sgRNA) and construction of CRISPR-Cas9 plasmids The online Optimized CRISPR Design Platform (http://crispr.mit.edu/) was utilized to design Nsmce2-sgRNA. 5’-ACCCGTTACATATCCTTCAG-3’, followed by the protospacer- adjacent motif (PAM) TGG (in exon 2) was selected as the target site. The corresponding forward and reverse strand oligonucleotide was synthesized by Sigma-Aldrich, and then annealed and cloned into the commercial linearized vector GeneArt CRISPR Nuclease Vector with OFP Reporter (Thermo Fisher Scientific), following the protocol provided by the manufacturer. The correct double-strand oligonucleotide insertion was confirmed by Sanger sequencing after transformation and plasmid extraction.

GS cell culture and differentiation A mouse GS cell line was established following a previously published protocol [34]. Briefly, germ cells were isolated from the cryopreserved testes of 4-5 d.p.p DBA/2J male mice by a two-step enzymatic dissociation. After overnight incubation on gelatin-coated wells, the floating and loosely attached cells were collected and cultured in the complete GS cell medium composed of StemPro-34 SFM medium (Thermo Fisher Scientific), StemPro-34 Supplement (Thermo Fisher Scientific), 1% FBS, 10 ng/ml recombinant human GDNF (Peprotech), 10 ng/ml recombinant human bFGF (Peprotech), as well as other 17 components as previously reported [34, 35]. The cells were refreshed every 2-3 days and passaged every 5-7 days at a ratio of 1: 4-6. From the third passage, the cells were transferred to inactivated primary mouse embryonic fibroblast (MEF) feeder cells that had been treated with 10 µg/ml mitomycin-C (Sigma-Aldrich) for 2-3 hours at 37 °C. The cells were maintained at 37°C in an atmosphere of 5% CO2 in air. After ~1 month, the growth of GS cells became stable. GS cells were cultured on MEFs unless otherwise stated. For feeder-free culture, GS cells were seeded on wells pre-coated with laminin (20 µg/ml, Sigma- Aldrich). For retinoic acid (RA)-induced differentiation of GS cells, the feeder-free culture was adopted, and GS cells were exposed to medium containing 2µM all-trans-RA (Sigma-Aldrich) for 3 days. In control groups, vehicle (0.1% ethanol) was added to the medium.

GS cell electroporation The constructed CRISPR plasmids targeting Nsmce2 were delivered to low-passage GS cells (˂P10) by Neon electroporator (Thermo Fisher Scientific), following the manufacturer’s guidance. The program used for electroporation was voltage 1100, width 20ms and pulse 2. Two days after electroporation, OFP+ cells were sorted by fluorescence- activated cell sorter (FACS, BD Biosciences) and cultured on MEFs for recovery. 54 Chapter 3

Surveyor assay One week after FACS sorting, the genomic DNA of GS cells was extracted, and the genomic region around the target site was amplified by PCR with the forward primer 5’- GATGATGGCACAGTGCTTGG-3’ and the reverse primer 5’- GGCAGTTCTGAGTGGAGGATTA C-3’. Herculase II fusion polymerase (Agilent Technologies) was used for high-fidelity PCR amplification. PCR products were purified, denatured and re-annealed to generate DNA heteroduplexes, followed by the Surveyor nuclease (Integrated DNA Technologies) digestion, according to the protocol provided by the manufacturer. The Surveyor nuclease digestion products were run and visualized on agarose gels. The incidence of insertion/deletion (indel) was calculated using a previously described formula [36].

GS cell clonal isolation and expansion One week after FACS sorting, the recovered GS cells were dissociated by accutase (Thermo Fisher Scientific) and filtered through a 50µm mesh to remove cell aggregates. The single GS cells were plated on 6-well plates pre-coated with laminin, at a density of 2,000- 4,000 cells/well. One week after plating, single GS cell-derived patches were detached with 0.5mM EDTA, manually picked under the microscope, and each cell patch was transferred to one well of a 96-well plate pre-coated with MEFs. After ~1 month, the expanded colonies were dissociated by accutase and transferred to individual wells of a 48-well plate with MEFs and to larger wells thereafter. Since the clonal expansion proceeded to 6-well plates, the cell culture was carried out routinely.

Genotyping monoclonal GS cell lines Before genotyping, a subpopulation of monoclonal GS cell lines was cultured on laminin for several passages to thoroughly eliminate the mixed MEFs. Then, the genomic DNA was extracted from each monoclonal GS cell line, and the region around the target site was amplified by PCR with the uniform primers for the Surveyor assay, but with a different high-fidelity polymerase (Easy-A high-fidelity PCR cloning enzyme, Agilent Technologies). The purified PCR products were cloned into T-Vector pMD19 (TaKaRa) following the manufacturer’s guidance. After transformation and overnight incubation, twenty colonies for each reaction were picked at random and sequenced.

NSMCE2 in male germline stem cells 55

Off-target analysis The 10 top-ranking potential off-target loci, provided by the online Optimized CRISPR Design Platform (http://crispr.mit.edu/), were analyzed. In brief, the genomic DNA was first extracted from monoclonal GS cell lines, and the regions flanking each potential off-target site were PCR-amplified with the Herculase II fusion polymerase (Agilent Technologies) and the primers shown in Table 1. The purified PCR products were then sequenced for off-target analyses.

Table 1: The selected potential off-target sites and the corresponding PCR primers for sequencing. Potential off-target sites Genomic loci Mis- PCR primers (GRCm38/mm10) matches ACCTGTTACATATCCTTCGG 60636826-60636848, F: AAGTTAGTTTAGGGCACAAAGG 4, 19 AGG chromosome 12, - R: AGTCCACCAGGTTAGAAAAGC TCCTGTTACGTATCCTTCAG 4770918-4770940, F: GCTCCCTGGCTTTCTCATT 1, 4, 10 CAG chromosome 1, + R: CTTGCCCGTGTCCTCTACTA ACACTGTACATATCCTTCAG 22477599-22477621, F: TGGTCCAAAATTCGCTGTAA 3, 5, 6 CAG chromosome 18, - R: GTGGCATCAGGGCAAACA AAGCTTTCCATATCCTTCAG 107576559-107576581, F: CCTGGAACTAACTCTAGGGATG 2, 3, 5, 8 TGG chromosome 6, + R: CTGGATGTTTCTTAATGGGACT AGGCGTCTCATATCCTTCAG 148326544-148326566, F: ATTTCCAGCAGAGTCCCACT 2, 3, 7, 8 TGG chromosome 2, - R: GACCCCAAGGCCATTATTC TTCAGTTAGATATCCTTCAG 91837662-91837684, F: CCTAAGCTGCTGCCTAAAAG 1, 2, 4, 9 TGG chromosome 11, - R: CACATGACATTCTGATCTTGCA GCCAGTTCCACATCCTTCAG 34570031-34570053, F: TGCTCAAGGAGGAGGAAACT 1, 4, 8, 11 CAG chromosome 2, + R: GGAAGGCTGGAAGTGGTGT ACATGTTTCAGATCCTTCAG 75040008-75040030, F: GTGGCAAATCTGGGTGGA 3, 4, 8, 11 AAG chromosome 3, - R: TCTGAGGGTAGGCTGTGAGG ACTTCTTACAAATCCTTCAG 16346304-16346326, F: TGGCAGTGATGAGAAAACGA 3, 4, 5, 11 TGG chromosome 14, + R: GCTCTGAGGATGGAATGGGT GCCAGTTCCATTTCCTTCAG 123031365-123031387, F: AGCTGGCAGTCTATGAGTCAAT 1, 4, 8, 12 TAG chromosome 10, + R: TGCCCAGTGCGATTTGAT

Western blot analysis Before Western blot analysis, a subpopulation of Nsmce2+/+ and Nsmce2-/- GS cells were cultured on laminin for several passages to thoroughly eliminate the mixed MEFs. Then, the protein was isolated and Western blot analysis was conducted as previously reported [16, 28], with the LI-COR Odyssey imaging system (Biosciences). The primary antibodies used were rabbit anti-NSMCE2 (1: 200; provided by Oscar Fernandez-Capetillo), rabbit anti-SMC5 56 Chapter 3

(1: 500; A300-236A, Bethyl Laboratories), guinea pig anti-SMC6 (1: 200; custom made, peptide: KRPRQEELEDFDKDGDEDE), mouse anti-PLZF (1: 100; D-9, Santa Cruz Biotechnology), mouse anti-OCT4 (1: 100; C-10, Santa Cruz Biotechnology), rabbit anti- STRA8 (1: 1,000; ab49602, Abcam), rabbit anti-SGSM1 (1: 1,000; ab171943, Abcam), mouse anti-β-actin (1: 5,000; A1978, Sigma-Aldrich) and rabbit anti-GAPDH (1: 400; FL-335, Santa Cruz Biotechnology).

Cell cycle analysis Cell cycle analysis based on DNA content was performed as previously described [28]. DNA content was analyzed with the FACS analyzer (BD Biosciences) and the figures were constructed via the FlowJo software. Data were presented as the mean ± standard error of mean (S.E.M.). Differences between groups were assessed using the Student’s t-test. A difference was considered significant when p˂0.05.

EdU assay The cell proliferation assay was performed using a Click-iT EdU Alexa Fluor 488 imaging kit (Thermo Fisher Scientific), following the protocol provided by the manufacturer. In brief, GS cells were grown on laminin-coated glass coverslips in 24-well plates. On the day of the treatment, cells were incubated with 10µM EdU diluted in complete medium for 2 h at 37 °C, followed by fixation in 4% paraformaldehyde (PFA) for 10 min. After permeabilization, cells were incubated with the reaction cocktail for 30 min at room temperature (RT), and then counterstained with DAPI for 5 min. Cells were mounted on glass slides with the Prolong Gold anti-fade mountant (Thermo Fisher Scientific) and later subjected to visualization under the microscope. For quantification of EdU+ cells, at least 300 cells were analyzed in each group. Data were presented as the mean ± S.E.M. Differences between groups were assessed using the Student’s t-test. A difference was considered significant when p˂0.05.

Immunocytochemistry GS cells were grown on laminin-coated glass coverslips in 24-well plates for all immunocytochemical experiments. In case of etoposide treatment, GS cells were incubated with 10µM etoposide for 3 h at 37 °C, then fixed at different time points, i.e. 0 h (immediately post treatment), 1, 3 and 5 h after treatment respectively, in 4% PFA for 10 min. Cells were permeabilized in phosphate-buffered saline (PBS) with 0.1% triton-X for 15 min, followed by 1 h of blocking in PBS with 1% bovine serum albumin (BSA) and 0.25% Tween20. Primary NSMCE2 in male germline stem cells 57 antibodies were applied to cells at 4°C overnight. The primary antibodies used were mouse anti-PLZF (1: 50; D-9, Santa Cruz Biotechnology), mouse anti-OCT4 (1: 50; C-10, Santa Cruz Biotechnology), rabbit anti-LIN28A (1: 1,000; ab46020, Abcam), rabbit anti-ID4 (1: 100; M106, CalBioreagents), mouse anti-ɣ-H2AX (1: 20,000; 05-636, Merck Millipore) and guinea pig anti-SMC6 (1: 400; custom made). Omission of the primary antibodies and replacement with mouse, rabbit and guinea pig isotype IgGs were used as negative controls. After washing with PBS on the next day, the cells were incubated with the corresponding secondary antibodies (Alexa Fluor 488 or 555, 1: 1,000; Thermo Fisher Scientific) for 1h at RT. After counterstaining with DAPI for 5 min at RT, the cells were mounted on glass slides with the Prolong Gold anti-fade mountant (Thermo Fisher Scientific) and later subjected to visualization under the microscope. For quantification of ɣ-H2AX+ cells (˃5 ɣ-H2AX foci/cell), at least 50 cells were analyzed in each group. Data were presented as the mean ± S.E.M. Differences between groups were assessed using the Student’s t-test. A difference was considered significant when p˂0.05.

Microscopy Fluorescence microscopy images were acquired at RT using a Leica DM5000B microscope equipped with a Leica DFC365 FX CCD camera. Images were analyzed using Leica Application Suite Advanced Fluorescence (LAS AF) software. The presented figures were constructed using Adobe Photoshop CS6.

RNA sequencing (RNA-seq) Total RNA was extracted from Nsmce2-/- and Nsmce2+/+ GS cells, respectively, using PureLink RNA Micro Kit (Thermo Fisher Scientific). Biological triplicates from different passages were prepared. Total RNA was sent to BGI Tech Solutions (Hong Kong, China) for library construction and sequencing (Illumina HiSeq 4000, paired end 100bp). Reads were subjected to quality control and aligned to the UCSC mm10 (GRCm38.p4) genome using HISAT2 (v2.0.4) [37]. Counts were obtained using HTSeq (v0.6.1) [38] using the UCSC mm10 GTF. Statistical analyses were performed using the edgeR [39] and limma [40] R (v3.2.2)/Bioconductor (v3.0) packages. All genes with no counts in any of the samples were removed (15,633 genes), whilst genes with more than 1 count-per-million reads (CPM) in at least 2 of the samples were kept (31,435 genes). Count data were transformed to log2-counts per million (logCPM), normalized by applying the trimmed mean of M-values method [39] and precision weighted using voom [41]. Differential expression was assessed using an empirical Bayes moderated t-test within limma’s linear model framework including the precision 58 Chapter 3 weights estimated by voom. Resulting P values were corrected for multiple testing using the Benjamini-Hochberg false discovery rate. Corrected P values of ˂0.05 were considered as statistically significant. Genes were re-annotated using biomaRt (v2.26.1) using the Ensembl genome databases (v85). The resulting DEGs (adj.P˂0.05) and entire RNA-seq data are shown in Table 2 and Supplementary Table 1 (see section on supplementary data given at the end of this article) respectively. Gene Set Enrichment Analysis (GSEA) software [42, 43] was used to analyze differentially expressed gene sets, and the GSEA results are shown in Supplementary Table 2.

Table 2: The DEGs obtained by analysis of RNA-seq data from Nsmce2-/- and Nsmce2+/+ GS cells. Gene symbol Description logFC adj.P.Val (˂0.05) Mx1 MX dynamin-like GTPase 1 2.647288067 0.046661415 Skor1 SKI family transcriptional corepressor 1 0.715634301 0.046661415 Myog myogenin 4.244660863 0.044188003 Gbgt1 globoside alpha-1,3-N- 1.234137519 0.046661415 acetylgalactosaminyltransferase 1 Sgsm1 small G protein signaling modulator 1 -1.795957187 0.003750701 Zfp358 zinc finger protein 358 -1.292106048 0.046661415 Smim22 small integral membrane protein 22 2.450729769 0.046661415 Gm9 predicted gene 9 6.259824968 0.046661415 Gm30332 predicted gene 30332 -6.010552517 0.003750701

Accession numbers All sequence data have been submitted to NCBI (SRA) and will be available under the accession number: ID PRJNA379902.

Results Construction of CRISPR-Cas9 plasmids and gene targeting of Nsmce2 in GS cells To perform loss of function study for Nsmce2, we first designed a sgRNA to target exon 2 of the mouse Nsmce2 locus (Figure 1A). The target site is located in an early and conserved coding sequence of the two Nsmce2 transcript variants. Frameshift mutations at this site will eliminate NSMCE2 function, including its only known activity, the E3 SUMO ligase through its C-terminal SP-RING domain [9-11]. To enable recognition and fluorescent sorting of transfected cells, we then constructed CRISPR-Cas9 plasmids containing this sgRNA and an orange fluorescent protein (OFP) reporter (Figure 1B). To establish a GS cell line, we isolated germ cells from the testes of 4-5 d.p.p DBA/2J male mice and cultured the isolated germ cells following an established protocol published by NSMCE2 in male germline stem cells 59

Shinohara’s group [34]. After 3-4 weeks, the GS cells were transferred to inactivated MEFs for subculture and they formed the distinctive grape-like colonies (Figure 1C), consistent with former reports [34, 44]. GS cells have been shown to be extremely refractory to commonly used transfection methods such as calcium phosphate precipitation and lipofection [45]. Nevertheless, novel electroporation devices can be harnessed to transfect spermatogonia with moderate efficiency [46-50]. To this end, we utilized a Neon electroporator to deliver CRISPR vectors into early passage GS cells. Two days after electroporation, OFP+ cells (Figure 1D) were enriched by FACS (Figure 1E) and transferred to MEF feeder cells for recovery. One week after FACS sorting, we conducted a Surveyor assay [36] based on the sorted cells. The target site-specific PCR amplicons were digested by Surveyor nucleases, yielding fragments with expected sizes indicative of indel mutations, demonstrating the occurrence of gene editing (Figure 1F).

Figure 1: CRISPR-Cas9-mediated targeting of Nsmce2 in GS cells. (A) The design of a sgRNA targeting exon 2 of the mouse Nsmce2 locus. Blue letters represent the 20-bp target, whereas red letters refer to the PAM NGG. (B) A schematic overview of the constructed CRISPR-Cas9 plasmids targeting Nsmce2. (C) The grape-like GS cell colonies when cultured on MEFs. Bar = 100µm. (D) GS cells on laminin 1 day after electroporation. Bar = 100µm. (E) FACS enrichment of OFP+ cells harboring CRISPR-Cas9 plasmids. (F) Surveyor assay-induced cleavage bands (approximately 300 and 218bp respectively, arrowheads). Negative control (-) represents the group without transfection.

Generation of Nsmce2-/- GS cell lines Next, we derived monoclonal Nsmce2-/- GS cell lines. Initially, we exploited FACS to deposit single GS cells into individual wells of one 96-well plate (1 cell/well) pre-coated with MEF feeder cells. However, after 2 weeks of culture, we did not observe any GS cell clones. 60 Chapter 3

To overcome this hurdle, we determined to pick single cell-derived patches. Specifically, one week after FACS sorting, the sorted and recovered GS cells were plated on 6-well plates pre-coated with laminin, at a low density (2,000-4,000 cells/well) (Figure 2A). The wells were pre-coated with laminin instead of MEFs because GS cells can attach rapidly to laminin, thereby circumventing the problem of polyclonal formation when single GS cells are plated at a low density. By contrast, we observed that GS cells attached slowly to MEFs and that many single cells readily aggregated before attachment. Previous reports have shown that MEFs as feeders are dispensable and GS cells can also be cultured on laminin for a long time [51-53]. One week after plating, approximately 1-10% of single GS cells formed cell patches comprising 4-6 cells (Figure 2B). They were then detached with EDTA, manually picked under the microscope and each cell patch was transferred to one well of a 96-well plate pre-coated with MEFs for further clonal expansion. After ~1 month, the expanded colonies (Figure 2C) were dissociated and transferred to individual wells of a 48-well plate coated with MEFs and to larger wells thereafter. Eventually, the clonal expansion proceeded to 6-well plates. The whole process of the clonal expansion took ~2.5 months. In total, we derived 7 GS cell lines from 48 picked single-cell patches. PCR amplification of the genomic DNA region around the target site followed by TA cloning and Sanger sequencing revealed that 2 of them carried gene modifications at the Nsmce2 locus. Unfortunately, one Nsmce2-/- GS cell line failed to expand, probably due to premature passage when cells remained at a small number. Fortunately, the other Nsmce2-/- GS cell line, harboring bi-allelic frameshift mutations (Figure 2D), was successfully expanded. Western blot analysis, using a transfected and single cell-derived Nsmce2+/+ GS cell line as a positive control, confirmed the eradication of the corresponding NSMCE2 protein (Figure 2E). Protein levels of SMC5 and SMC6 were not influenced by removal of NSMCE2 (Figure 2E). We validated these results using lysates from Nsmce2+/+ and Nsmce2-/- MEFs (described in [31]) (Figure 2E). Finally, we sequenced the 10 top-ranking potential off-target sites (Table 1) in the established Nsmce2-/- cell line and detected no off-target mutations. This Nsmce2-/- GS cell line, together with the control Nsmce2+/+ GS cell line used in Figure 2E, were used for all further experiments.

Removal of NSMCE2 does not influence the spermatogonial cell cycle or proliferation Morphologically, Nsmce2-/- GS cells were indistinguishable from their Nsmce2+/+ counterparts, and both formed the characteristic grape-like colonies on MEFs (Figure 3A), distinct from the ES-like multipotent GS (mGS) cell colonies. Because we have previously demonstrated a prolonged cell cycle of Nsmce2-null U2OS cells [28], we investigated NSMCE2 in male germline stem cells 61 whether the removal of NSMCE2 also alters the proliferation/cell cycle time of GS cells. The two single cell-derived GS cell lines, Nsmce2+/+ and Nsmce2-/- respectively, did not have significantly different cell doubling time (Figure 3B). Consistently, cell cycle analysis by DNA histogram showed no difference between the two cell populations with respect to the ratios of cells in different cell cycle phases (Figure 3C). To further investigate the role of NSMCE2 in controlling GS cell proliferation, we performed an EdU incorporation assay. The proportion of cells that incorporate the thymidine analog EdU (a measurement for DNA synthesis) did not differ between Nsmce2+/+ and Nsmce2-/- GS cells (Figure 3D). Furthermore, both cell lines could normally propagate in vitro for more than 35 passages, without significant changes in cellular morphology or cell doubling time. Overall, the results suggest that the removal of NSMCE2 does not influence in vitro spermatogonial proliferation.

Figure 2: Generation of Nsmce2-/- GS cell lines. (A) Single GS cells immediately after plating on laminin. Bar = 100µm. (B) Single cell-derived patches 1 week after seeding. Bar = 100µm. (C) A single GS cell-derived colony after 1 month of culture in a well of 96-well plate coated with MEFs. Bar = 100µm. (D) Sanger-sequencing analysis of the Nsmce2-/- GS cell line. (E) Immunoblotting of NSMCE2, SMC5 and SMC6 in the established Nsmce2+/+ and Nsmce2-/- GS cell lines, and in the control Nsmce2+/+ and Nsmce2-/- MEFs. β-Actin is used as a loading control.

62 Chapter 3

Figure 3: The deprivation of NSMCE2 does not influence the self-renewal of GS cells. (A) Representative images of Nsmce2+/+ and Nsmce2-/- GS cells on MEFs. Bar = 100µm. (B) The doubling time of Nsmce2+/+ and Nsmce2-/- GS cells. Data are presented as the mean ± S.E.M., n=3. (C) Cell cycle analysis of Nsmce2+/+ and Nsmce2-/- GS cells showing the percentages of cells in different cell cycle phases (G0-1, S and G2-M). Data are presented as the mean ± S.E.M., n=3. (D) Representative images (the upper part) and quantification (the lower part) of Nsmce2+/+ and Nsmce2-/- GS cells that incorporate EdU. Cells were counterstained with DAPI to visualize nuclei. Bar = 20µm. Data are presented as the mean ± S.E.M., n=3.

Nsmce2-/- GS cells express typical markers of undifferentiated spermatogonia and can be induced to differentiation To validate their undifferentiated spermatogonial identity, we performed immunocytochemistry on Nsmce2-/- and Nsmce2+/+ GS cells. Both cell populations showed staining for SSC/progenitor cell markers PLZF, LIN28A, OCT4 and ID4 (Figure 4A), indicating their undifferentiated state. Negative controls, i.e. omission of primary antibodies or NSMCE2 in male germline stem cells 63 replacement with isotype IgGs, did not yield any staining (Figure 4A). Because our previous studies have shown that SMC6 marks spermatogonial differentiation [16], we next investigated whether NSMCE2 plays a role during spermatogonial differentiation. To this end, we induced spermatogonial differentiation by adding RA to the culture medium according to previous papers [54, 55]. In line with previous papers [51, 54], when cultured on laminin, both Nsmce2+/+ and Nsmce2-/- GS cells started to exhibit the typical rhomboid morphology with long pseudopod-like extensions that is characteristic for undifferentiated spermatogonia. Treatment with RA for 3 days made them, as expected for differentiating spermatogonia [54], gradually lose this phenotype and become more round (Figure 4B), indicative of their differentiation. Moreover, Western blot analysis showed that exposure to RA considerably reduced the protein levels of PLZF and OCT4, whereas that of differentiation marker STRA8 [56, 57] was markedly increased in both cell lines (Figure 4C). Overall, these results demonstrate that Nsmce2-/- GS cells have normal undifferentiated spermatogonial characteristics and can be induced to differentiation normally.

64 Chapter 3

Figure 4: Nsmce2-/- GS cells express typical markers of undifferentiated spermatogonia and can be induced to differentiation. (A) The staining of SSC/progenitor cell markers in GS cells on laminin. Bar = 10µm. (B) Representative images of GS cells (on laminin) with/without RA treatment. Asterisks (*) indicate the long pseudopod-like extension. Bar = 25µm. (C) Western blot analysis of PLZF, OCT4 and STRA8 proteins in GS cells with/without RA-induced differentiation. GAPDH and β- actin are used as loading controls.

Etoposide-induced DNA damage repair occurs independently of NSMCE2 in GS cells Because we have recently shown that SMC6 physically interacts with TOP2A and that NSMCE2 is implicated in the response to etoposide-induced topological stress and in subsequent DNA damage repair in U2OS cells [28], we investigated whether NSMCE2 functions similarly in GS cells. To this end, we exposed Nsmce2+/+ and Nsmce2-/- GS cells to etoposide and quantified DNA damage formation and repair marked by ɣ-H2AX (Figure 5A and B). Albeit at a low number, ɣ-H2AX staining could be discerned in a small fraction of cells prior to etoposide treatment. Exposure to etoposide for 3 h triggered an increase of cells displaying ɣ-H2AX staining, indicative of increased DNA damage. However, the absence of NSMCE2 did not lead to more ɣ-H2AX+ cells after etoposide treatment (Figure 5A and B). Also the decrease of ɣ-H2AX+ cells, indicative of DNA repair, followed similar dynamics between Nsmce2+/+ and Nsmce2-/- GS cells. Moreover, in contrast to U2OS cells [28], ɣ- H2AX did not co-localize with SMC6 in GS cells, regardless of etoposide treatment (Figure 5C). Negative controls, i.e. omission of primary antibodies or replacement with isotype IgGs, did not yield any staining (Figure 5A and C). The above data suggest that NSMCE2 is not involved in the response to etoposide-induced topological stress in GS cells.

Deprivation of NSMCE2 results in significant downregulation of Sgsm1 at the RNA but not protein level To gain a broader perspective on the overall molecular effects of NSMCE2 removal, we conducted a RNA-seq experiment to compare the transcriptomes of Nsmce2-/- and Nsmce2+/+ GS cells. According to the RNA-seq data of biological triplicates from different passages (Figure 6A), inactivation of NSMCE2 generated only 9 differentially expressed genes (DEGs, adj.P˂0.05). Of these 9 DEGs, 6 genes showed upregulation and 3 genes showed downregulation (Table 2 and Supplementary Table 1). These RNA-seq data thus suggest that removal of NSMCE2 has only very minimal effects on the spermatogonial transcriptome. Notably, one known gene, Sgsm1, showed a very significant downregulation without NSMCE2 (Table 2) and appeared to be highly expressed in human spermatogonia NSMCE2 in male germline stem cells 65

(Human Protein Atlas, v15 [58]). Nevertheless, Western blot analysis showed comparable protein expression of SGSM1 in Nsmce2-/- and Nsmce2+/+ GS cells (Figure 6B). Subsequent GSEA indicated alterations of gene sets such as H3K27Me3 or H3K4Me2 related to histone methylation and cancer development (Supplementary Table 2). Nonetheless, none of the genes within these gene sets were differentially expressed (adj.P˂0.05).

Figure 5: Etoposide-induced DNA damage formation. (A) Representative images of ɣ-H2AX staining in Nsmce2+/+ and Nsmce2-/- GS cells after etoposide treatment. GS cells were seeded on laminin prior to etoposide treatment. Control, without etoposide treatment; 0h, immediately post treatment. Bar = 10µm. (B) Percentage of ɣ-H2AX+ cells after etoposide treatment. Control, without etoposide treatment; 0h, immediately post treatment. Data are presented as the mean ± S.E.M., n=3. (C) No co-localization of ɣ-H2AX with SMC6. Bar = 5µm.

66 Chapter 3

Figure 6: The overall effects of NSMCE2 removal on the transcriptome of GS cells. (A) The multidimensional scaling (MDS) plot (left panel) and heat map (right panel) illustrating the minimal transcriptome variation between the WT and KO cells. KO, Nsmce2-/- GS cells; WT, Nsmce2+/+ GS cells. (B) Western blot analysis of SGSM1 in Nsmce2+/+ and Nsmce2-/- GS cells. β-Actin is used as a loading control.

Discussion SSCs, a subpopulation of undifferentiated spermatogonia, are best characterized by their capability of self-renewal to maintain sufficient numbers as well as differentiation to mature spermatozoa, thereby maintaining life-long male fertility. SSCs hold great value in reproductive medicine. They can reestablish spermatogenesis following transplantation into recipient testes, and thus they can be harnessed to restore fertility of, for instance, childhood cancer survivors who lose their germ cells due to chemotherapy or radiotherapy [59]. Moreover, in combination with genomic modification, SSCs could theoretically be employed to cure spermatogenic failure with known genetic causes or prevent inheritance of genomic diseases [59]. For rodent SSCs, CRISPR-Cas9 has been applied successfully for precise genomic modification and even to correct a genetic disease in the offspring [48-50]. Apart from clinical applications, SSCs are also of substantial utility in biomedical research, e.g. manipulation of SSCs provides an advantageous avenue to generate various animal models with specific genotypes and phenotypes [60]. In spite of this, the establishment of genetically modified SSC lines has been inefficient so far, primarily due to low transfection efficiency, difficulty in monoclonal isolation and expansion of SSCs in vitro. These technical difficulties impede the generation of transgenic animal models and the research on genes underlying spermatogenesis. Similar to other primary stem cells, SSCs are very refractory to prevailing transfection approaches such as calcium phosphate precipitation and lipofection [45]. Previous studies have also indicated that adeno-associated virus (AAV) and integration-deficient lentivirus are ineffective to fulfill gene editing in SSCs [47]. Currently, novel electroporation devices are increasingly used to NSMCE2 in male germline stem cells 67 transfect SSCs [46-50]. In our study, we employed a Neon electroporator to deliver the large CRISPR-Cas9 vectors into GS cells. The cells harboring CRISPR-Cas9 vectors were then enriched and subjected to clonal isolation. To achieve single cell-derived GS cell clones, we first cultured single GS cells in a full 96-well plate, but failed to get any cell clones. Because the initial cell density has been reported to have a big influence on subsequent GS cell culture [45], we then attempted to pick single cell-derived patches. We found that GS cells attached slowly to MEFs and that many single cells readily aggregated before attachment. We therefore used laminin instead of MEF feeder cells. GS cells rapidly attach to laminin, thereby greatly facilitating monoclonal isolation when single GS cells are plated at a low density. Collectively, the entire optimized protocol (elaborated in ‘Materials and methods’ section) is less laborious and more efficient compared to recent reports [48-50]. Using this optimized protocol, we successfully generated an Nsmce2-/- GS cell line lacking the NSMCE2 subunit of the SMC5/6 complex. The SMC5/6 complex has been shown to play pivotal roles in many important biological processes, in particular, genomic integrity maintenance and DNA damage repair [3-5]. In yeast, all subunits of the SMC5/6 complex are indispensable [11] and also in mice KO of SMC6 leads to embryonic lethality [32]. Consistently, deprivation of SMC5 in mouse

ES cells leads to abnormal mitotic progression, accumulation in G2 of the cell cycle and apoptosis [27]. In our preliminary experiments, we attempted to use CRISPR-Cas9 to knock- out Smc5 and Smc6 in GS cells. However, we observed cell death of more than 90% of the transfected cells, which may indicate that the deletion of one of these genes is lethal to GS cells. In contrast, Nsmce2 KO did not influence the survival of GS cells, and we eventually generated an Nsmce2-/- GS cell line. NSMCE2 depletion can destabilize the SMC5/6 complex, characterized by the reduced expression of SMC5 and/or SMC6 [61, 62]. Using Western blot analysis, we showed that the protein levels of SMC5 and SMC6 were not affected by NSMCE2 depletion in GS cells. Indeed, the stable presence of SMC5 and SMC6 does not mean that the whole SMC5/6 complex remains stable. However, combined with the fact that we did not detect any other significant phenotype, it does suggest that NSMCE2 is not essential for the SMC5/6 stability and function in GS cells. We have recently found that the removal of NSMCE2 in U2OS cells generated significant differences in phenotypes, including slower cell growth, accumulation of cells in

G0-1 of the cell cycle, as well as a decreased plating efficiency [28]. In line with these, an earlier study in human MCF-7 breast cancer cells showed that knockdown of Nsmce2 resulted in slower cell growth and impaired G1-S transition. Ectopic expression of the full- length NSMCE2, but not its SUMO ligase-inactive mutant, rescued the reduced cell growth, 68 Chapter 3 implying that the normal growth of these breast cancer cells requires the NSMCE2 SUMO ligase function [63]. However, consistent with a study conducted in chicken DT40 cells [61], we found that the viability and growth of GS cells was not influenced by the removal of NSMCE2. Nsmce2-/- GS cells showed similar doubling time, EdU incorporation and cell cycle progression as their Nsmce2+/+ counterparts. Furthermore, GS cells showed normal cellular morphology without nuclear abnormalities, such as micronuclei or nucleoplasmic bridges that are observed in mouse and human Nsmce2-deficient fibroblasts [31, 62]. These disparities imply cell type-specific roles of NSMCE2. We have recently found that SMC6 protein expression specifically marks differentiating spermatogonia [16], the spermatogonial subpopulation irreversibly committed toward meiosis. However, the exact role of NSMCE2 during spermatogonial differentiation has not been elucidated. We therefore investigated whether absence of NSMCE2 would influence the capacity of GS cells to differentiate. GS cells can be induced to differentiation by adding RA to the culture medium [54, 55]. RA-induced differentiation will lead to downregulation of SSC/progenitor cell markers (e.g. PLZF and OCT4), upregulation of differentiation markers (e.g. STRA8), reduced self-renewal and increased apoptosis and eventually to a decline in the total cell number [54, 55, 64]. Before RA-induced differentiation, Nsmce2-/- GS cells exhibited routine expression profiles characteristic for undifferentiated spermatogonia. Here, we found that Nsmce2-/- GS cells could be normally induced to differentiate, exhibiting similar morphology and characteristics as their Nsmce2+/+ counterparts. Hence, NSMCE2 does not seem to be involved in in vitro spermatogonial proliferation or differentiation. NSMCE2 is extensively reported to be crucial for the response to DNA damage, mostly in yeast [9-11, 61, 65, 66]. However, recent studies performed in multiple mouse and human cell types have indicated that NSMCE2 is redundant for the repair of ionizing radiation (IR)- induced DNA damage in mammalian cells [28, 31, 67]. Indeed, we have recently shown that also in U2OS cells, NSMCE2 is redundant for the repair of double-strand breaks (DSBs) induced by IR [28]. Nonetheless, in the same paper, we showed that CRISPR-Cas9- mediated removal of NSMCE2 led to increased sensitivity to etoposide. Etoposide is a cytotoxic agent that, by forming a complex with DNA and topoisomerase II, causes replication-induced topological stress and DSBs at replication forks [68]. Surprisingly, here we found that removal of NSMCE2 in GS cells did not cause increased sensitivity to etoposide. Moreover, etoposide-induced DSBs marked by ɣ-H2AX were repaired efficiently in both Nsmce2+/+ and Nsmce2-/- GS cells. Also in contrast to U2OS cells [28], etoposide- induced ɣ-H2AX in GS cells did not co-localize with SMC6. It is thus plausible that GS cells hold different mechanisms to resolve topological stress than somatic cells, which might not rely on NSMCE2 or even the SMC5/6 complex. NSMCE2 in male germline stem cells 69

To gain a broader perspective on the overall molecular effects of NSMCE2 removal, we finally conducted a whole transcriptome RNA-seq analysis for Nsmce2+/+ and Nsmce2-/- GS cells. We previously hypothesized that, by altering chromatin structure, the SMC5/6 complex would influence spermatogonial gene transcription [15, 16]. GSEA did indicate the downregulation of gene sets involved in chromatin regulation, such as H3K27Me3 or H3K4Me2 related to histone methylation and cancer development, suggesting that chromatin architecture or function is somehow affected by the absence of NSMCE2. However, removal of NSMCE2 only led to 9 DEGs, of which none were present in these gene sets. Of the 9 DEGs, Sgsm1 showed a very significant downregulation without NSMCE2. SGSM1, as well as its two paralogs SGSM2 and SGSM3, all consist of RUN and TBC motifs and have been reported to orchestrate small G protein-mediated signaling transduction. Unlike SGSM2 and SGSM3, which show ubiquitous expression in a wide range of tissues, SGSM1 is primarily expressed in human brain, heart and testes [69]. Moreover, in human testes, SGSM1 is highly expressed in spermatogonia (Human Protein Atlas, v15 [58]). In our present study, we also demonstrated the high protein level of SGSM1 in GS cells. However, while removal of NSMCE2 significantly downregulated Sgsm1 transcripts, its protein level was not affected. Perhaps the SGSM1 protein level was somehow stabilized in the absence of NSMCE2. Alternatively, assuming that SGSM1 protein stability is not affected by NSMCE2, the detected lower Sgsm1 mRNA level was still sufficient to ensure the wildtype protein level. Nevertheless, the detected Sgsm1 downregulation in our study and localization in human spermatogonia indicate an unknown role of this protein in GS cells. To acquire more knowledge about this, loss-of-function studies for SGSM1 could be performed in the future. The absence of evident phenotypes and the minimal transcriptome variation caused by NSMCE2 removal raise the question what the exact role of NSMCE2 in GS cells is. Recently, a case report described 2 female patients with heterozygous frameshift mutations in Nsmce2 that result in decreased expression of NSMCE2. These patients exhibited serious phenotypes like primordial dwarfism, extreme insulin resistance and, notably, primary ovarian failure [62]. Given that NSMCE2 is required for yeast meiosis [70], the possibility remains that further downstream steps beyond spermatogonial proliferation and differentiation require the function of NSMCE2. On the other hand, the specific expression of SMC6 in differentiating spermatogonia [16], the stable presence of SMC5 and SMC6 without NSMCE2 and the observed lethality of GS cells after transfection with CRISPR-Cas9 plasmids targeting Smc5/6 all suggest that the SMC5/6 complex plays important roles in spermatogonia, but that this spermatogonial function of SMC5/6 is not affected by NSMCE2 removal. 70 Chapter 3

Male germline stem cells are responsible for the lifelong daily production of millions of sperm and the transmission of genetic information to the offspring. Decades of studies have well demonstrated that this unique adult stem cell population is largely distinct from somatic cells in terms of cellular activities, cell fate commitment, developmental plasticity, chromatin architecture and remodeling as well as (epi)genetic features [71-73]. They also hold unique mechanisms to maintain genomic stability [74], which can partially explain the divergent roles of NSMCE2 and SMC5/6 in GS cells. It has been known that the SMC5/6 complex, including NSMCE2, is essential for genome integrity maintenance in somatic cells, demonstrated by, e.g. the embryonic lethality of SMC6 or NSMCE2 KO [31, 32] and the role of NSMCE2 in the prevention of cancer and aging in adult mice [31]. Nonetheless, the SMC5/6 complex in male germline stem cells seems to function normally without NSMCE2. Hence, how germline stem cells safeguard genomic integrity, and the role of SMC5/6 herein, remain to be further investigated, especially given the current technical development such as CRISPR-Cas9 and the potential clinical application of these cells, for instance, in fertility preservation, curing spermatogenic failure or preventing transmission of genetic diseases [59].

Supplementary data This is linked to the online version of the paper at http://dx.doi.org/10.1530/REP-17- 0173.

Declaration of interest The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding This study has been supported by an AMC Fellowship, the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (CIG 293765) to Geert Hamer and the China Scholarship Counsel (CSC) number 201306300081 to Yi Zheng.

Acknowledgements The authors thank Ieva Masliukaite and Dr Kaijun Liu, for assistance with FACS analysis and sorting, respectively, and Saskia K M van Daalen for assistance with picking single cell-derived patches. In addition, they thank Oscar Fernandez-Capetillo for kindly providing the NSMCE2 antibody and MEF extracts. NSMCE2 in male germline stem cells 71

References

1. de Rooij DG, de Boer P. Specific arrests of spermatogenesis in genetically modified and mutant mice. Cytogenet Genome Res 2003; 103:267-276. 2. Jan SZ, Hamer G, Repping S, de Rooij DG, van Pelt AM, Vormer TL. Molecular control of rodent spermatogenesis. Biochim Biophys Acta 2012; 1822:1838-1850. 3. De Piccoli G, Cortes-Ledesma F, Ira G, Torres-Rosell J, Uhle S, Farmer S, Hwang JY, Machin F, Ceschia A, McAleenan A, Cordon-Preciado V, Clemente-Blanco A, et al. Smc5-Smc6 mediate DNA double-strand-break repair by promoting sister-chromatid recombination. Nat Cell Biol 2006; 8:1032- 1034. 4. Potts PR. The Yin and Yang of the MMS21-SMC5/6 SUMO ligase complex in homologous recombination. DNA Repair (Amst) 2009; 8:499-506. 5. Wu N, Yu H. The Smc complexes in DNA damage response. Cell Biosci 2012; 2:5. 6. Carter SD, Sjogren C. The SMC complexes, DNA and chromosome topology: right or knot? Crit Rev Biochem Mol Biol 2012; 47:1-16. 7. Jeppsson K, Kanno T, Shirahige K, Sjogren C. The maintenance of chromosome structure: positioning and functioning of SMC complexes. Nat Rev Mol Cell Biol 2012; 15:601-614. 8. Langston RE, Weinert T. Nifty Alleles, a Plethora of Interactions, and Imagination Advance Understanding of Smc5/6's Roles with Chromosomes. Mol Cell 2015; 60:832-833. 9. Andrews EA, Palecek J, Sergeant J, Taylor E, Lehmann AR, Watts FZ. Nse2, a component of the Smc5-6 complex, is a SUMO ligase required for the response to DNA damage. Mol Cell Biol 2005; 25:185-196. 10. Potts PR, Yu H. Human MMS21/NSE2 is a SUMO ligase required for DNA repair. Mol Cell Biol 2005; 25:7021-7032. 11. Zhao X, Blobel G. A SUMO ligase is part of a nuclear multiprotein complex that affects DNA repair and chromosomal organization. Proc Natl Acad Sci U S A 2005; 102:4777-4782. 12. Verver DE, Hwang GH, Jordan PW, Hamer G. Resolving complex chromosome structures during meiosis: versatile deployment of Smc5/6. Chromosoma 2016; 125:15-27. 13. Copsey A, Tang S, Jordan PW, Blitzblau HG, Newcombe S, Chan AC, Newnham L, Li Z, Gray S, Herbert AD, Arumugam P, Hochwagen A, et al. Smc5/6 coordinates formation and resolution of joint molecules with chromosome morphology to ensure meiotic divisions. PLoS Genet 2013; 9:e1004071. 14. Gomez R, Jordan PW, Viera A, Alsheimer M, Fukuda T, Jessberger R, Llano E, Pendas AM, Handel MA, Suja JA. Dynamic localization of SMC5/6 complex proteins during mammalian meiosis and mitosis suggests functions in distinct chromosome processes. J Cell Sci 2013; 126:4239-4252. 15. Verver DE, Langedijk NS, Jordan PW, Repping S, Hamer G. The SMC5/6 complex is involved in crucial processes during human spermatogenesis. Biol Reprod 2014; 91:22. 16. Verver DE, van Pelt AM, Repping S, Hamer G. Role for rodent Smc6 in pericentromeric heterochromatin domains during spermatogonial differentiation and meiosis. Cell Death Dis 2013; 4:e749. 72 Chapter 3

17. Hwang G, Sun F, O'Brien M, Eppig JJ, Handel MA, Jordan PW. SMC5/6 is required for the formation of segregation-competent bivalent chromosomes during meiosis I in mouse oocytes. Development 2017. 18. Bickel JS, Chen L, Hayward J, Yeap SL, Alkers AE, Chan RC. Structural maintenance of chromosomes (SMC) proteins promote homolog-independent recombination repair in meiosis crucial for germ cell genomic stability. PLoS Genet 2010; 6:e1001028. 19. Lilienthal I, Kanno T, Sjogren C. Inhibition of the Smc5/6 Complex during Meiosis Perturbs Joint Molecule Formation and Resolution without Significantly Changing Crossover or Non-crossover Levels. PLoS Genet 2013; 9:e1003898. 20. Xaver M, Huang L, Chen D, Klein F. Smc5/6-mms21 prevents and eliminates inappropriate recombination intermediates in meiosis. PLoS Genet 2013; 9:e1004067. 21. Farmer S, San-Segundo PA, Aragón L. The smc5-smc6 complex is required to remove chromosome junctions in meiosis. PLoS One 2011; 6:e20948. 22. Torres-Rosell J, Sunjevaric I, De Piccoli G, Sacher M, Eckert-Boulet N, Reid R, Jentsch S, Rothstein R, Aragon L, Lisby M. The Smc5-Smc6 complex and SUMO modification of Rad52 regulates recombinational repair at the ribosomal gene locus. Nat Cell Biol 2007; 9:923-931. 23. Agostinho A, Meier B, Sonneville R, Jagut M, Woglar A, Blow J, Jantsch V, Gartner A. Combinatorial regulation of meiotic holliday junction resolution in C. elegans by HIM-6 (BLM) helicase, SLX-4, and the SLX-1, MUS-81 and XPF-1 nucleases. PLoS Genet 2013; 9:e1003591. 24. Hong Y, Sonneville R, Agostinho A, Meier B, Wang B, Blow JJ, Gartner A. The SMC-5/6 Complex and the HIM-6 (BLM) Helicase Synergistically Promote Meiotic Recombination Intermediate Processing and Chromosome Maturation during Caenorhabditis elegans Meiosis. PLoS Genet 2016; 12:e1005872. 25. O'Neil NJ, Martin JS, Youds JL, Ward JD, Petalcorin MI, Rose AM, Boulton SJ. Joint molecule resolution requires the redundant activities of MUS-81 and XPF-1 during Caenorhabditis elegans meiosis. PLoS Genet 2013; 9:e1003582. 26. Wehrkamp-Richter S, Hyppa RW, Prudden J, Smith GR, Boddy MN. Meiotic DNA joint molecule resolution depends on Nse5-Nse6 of the Smc5-Smc6 holocomplex. Nucleic Acids Res 2012; 40:9633-9646. 27. Pryzhkova MV, Jordan PW. Conditional mutation of Smc5 in mouse embryonic stem cells perturbs condensin localization and mitotic progression. J Cell Sci 2016; 129:1619-1634. 28. Verver DE, Zheng Y, Speijer D, Hoebe R, Dekker HL, Repping S, Stap J, Hamer G. Non-SMC Element 2 (NSMCE2) of the SMC5/6 Complex Helps to Resolve Topological Stress. Int J Mol Sci 2016; 17. 29. Wang JC. DNA topoisomerases. Annu Rev Biochem 1996; 65:635-692. 30. Wang JC. Cellular roles of DNA topoisomerases: a molecular perspective. Nat Rev Mol Cell Biol 2002; 3:430-440. NSMCE2 in male germline stem cells 73

31. Jacome A, Gutierrez-Martinez P, Schiavoni F, Tenaglia E, Martinez P, Rodriguez-Acebes S, Lecona E, Murga M, Mendez J, Blasco MA, Fernandez-Capetillo O. NSMCE2 suppresses cancer and aging in mice independently of its SUMO ligase activity. Embo J 2016; 34:2604-2619. 32. Ju L, Wing J, Taylor E, Brandt R, Slijepcevic P, Horsch M, Rathkolb B, Racz I, Becker L, Hans W, Adler T, Beckers J, et al. SMC6 is an essential gene in mice, but a hypomorphic mutant in the ATPase domain has a mild phenotype with a range of subtle abnormalities. DNA Repair (Amst) 2013; 12:356-366. 33. Kanatsu-Shinohara M, Ogonuki N, Iwano T, Lee J, Kazuki Y, Inoue K, Miki H, Takehashi M, Toyokuni S, Shinkai Y, Oshimura M, Ishino F, et al. Genetic and epigenetic properties of mouse male germline stem cells during long-term culture. Development 2005; 132:4155-4163. 34. Kanatsu-Shinohara M, Ogonuki N, Inoue K, Miki H, Ogura A, Toyokuni S, Shinohara T. Long- term proliferation in culture and germline transmission of mouse male germline stem cells. Biol Reprod 2003; 69:612-616. 35. Kanatsu-Shinohara M, Shinohara T. Germline Modification Using Mouse Spermatogonial Stem Cells. Methods in Enzymology, Vol 477: Guide to Techniques in Mouse Development, Part B: Mouse Molecular Genetics, Second Edition 2010; 477:17-36. 36. Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, Zhang F. Genome engineering using the CRISPR-Cas9 system. Nat Protoc 2013; 8:2281-2308. 37. Kim D, Langmead B, Salzberg SL. HISAT: a fast spliced aligner with low memory requirements. Nat Methods 2015; 12:357-360. 38. Anders S, Pyl PT, Huber W. HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics 2015; 31:166-169. 39. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 2010; 26:139-140. 40. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 2015; 43:e47. 41. Law CW, Chen Y, Shi W, Smyth GK. voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol 2014; 15:R29. 42. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Mesirov JP. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 2005; 102:15545-15550. 43. Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, Lehar J, Puigserver P, Carlsson E, Ridderstrale M, Laurila E, Houstis N, Daly MJ, et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 2003; 34:267-273. 44. Kubota H, Avarbock MR, Brinster RL. Growth factors essential for self-renewal and expansion of mouse spermatogonial stem cells. Proc Natl Acad Sci U S A 2004; 101:16489-16494. 74 Chapter 3

45. Kanatsu-Shinohara M, Toyokuni S, Shinohara T. Genetic selection of mouse male germline stem cells in vitro: offspring from single stem cells. Biol Reprod 2005; 72:236-240. 46. Zeng W, Tang L, Bondareva A, Luo J, Megee SO, Modelski M, Blash S, Melican DT, Destrempes MM, Overton SA, Gavin WG, Ayres S, et al. Non-viral transfection of goat germline stem cells by nucleofection results in production of transgenic sperm after germ cell transplantation. Mol Reprod Dev 2012; 79:255-261. 47. Fanslow DA, Wirt SE, Barker JC, Connelly JP, Porteus MH, Dann CT. Genome editing in mouse spermatogonial stem/progenitor cells using engineered nucleases. PLoS One 2014; 9:e112652. 48. Chapman KM, Medrano GA, Jaichander P, Chaudhary J, Waits AE, Nobrega MA, Hotaling JM, Ober C, Hamra FK. Targeted Germline Modifications in Rats Using CRISPR/Cas9 and Spermatogonial Stem Cells. Cell Rep 2015; 10:1828-1835. 49. Wu Y, Zhou H, Fan X, Zhang Y, Zhang M, Wang Y, Xie Z, Bai M, Yin Q, Liang D, Tang W, Liao J, et al. Correction of a genetic disease by CRISPR-Cas9-mediated gene editing in mouse spermatogonial stem cells. Cell Res 2015; 25:67-79. 50. Sato T, Sakuma T, Yokonishi T, Katagiri K, Kamimura S, Ogonuki N, Ogura A, Yamamoto T, Ogawa T. Genome Editing in Mouse Spermatogonial Stem Cell Lines Using TALEN and Double- Nicking CRISPR/Cas9. Stem Cell Reports 2015; 5:75-82. 51. Kanatsu-Shinohara M, Miki H, Inoue K, Ogonuki N, Toyokuni S, Ogura A, Shinohara T. Long- term culture of mouse male germline stem cells under serum-or feeder-free conditions. Biol Reprod 2005; 72:985-991. 52. Kanatsu-Shinohara M, Inoue K, Ogonuki N, Morimoto H, Ogura A, Shinohara T. Serum- and feeder-free culture of mouse germline stem cells. Biol Reprod 2011; 84:97-105. 53. Kanatsu-Shinohara M, Ogonuki N, Matoba S, Morimoto H, Ogura A, Shinohara T. Improved serum- and feeder-free culture of mouse germline stem cells. Biol Reprod 2014; 91:88. 54. Dann CT, Alvarado AL, Molyneux LA, Denard BS, Garbers DL, Porteus MH. Spermatogonial stem cell self-renewal requires OCT4, a factor downregulated during retinoic acid-induced differentiation. Stem Cells 2008; 26:2928-2937. 55. Wang S, Wang X, Ma L, Lin X, Zhang D, Li Z, Wu Y, Zheng C, Feng X, Liao S, Feng Y, Chen J, et al. Retinoic Acid Is Sufficient for the In Vitro Induction of Mouse Spermatocytes. Stem Cell Reports 2016; 7:80-94. 56. Zhou Q, Li Y, Nie R, Friel P, Mitchell D, Evanoff RM, Pouchnik D, Banasik B, McCarrey JR, Small C, Griswold MD. Expression of stimulated by retinoic acid gene 8 (Stra8) and maturation of murine gonocytes and spermatogonia induced by retinoic acid in vitro. Biol Reprod 2008; 78:537-545. 57. Zhou Q, Nie R, Li Y, Friel P, Mitchell D, Hess RA, Small C, Griswold MD. Expression of stimulated by retinoic acid gene 8 (Stra8) in spermatogenic cells induced by retinoic acid: an in vivo study in vitamin A-sufficient postnatal murine testes. Biol Reprod 2008; 79:35-42. NSMCE2 in male germline stem cells 75

58. Uhlen M, Fagerberg L, Hallstrom BM, Lindskog C, Oksvold P, Mardinoglu A, Sivertsson A, Kampf C, Sjostedt E, Asplund A, Olsson I, Edlund K, et al. Proteomics. Tissue-based map of the human proteome. Science 2015; 347:1260419. 59. Mulder CL, Zheng Y, Jan SZ, Struijk RB, Repping S, Hamer G, van Pelt AM. Spermatogonial stem cell autotransplantation and germline genomic editing: a future cure for spermatogenic failure and prevention of transmission of genomic diseases. Hum Reprod Update 2016; 22:561-573. 60. Zheng Y, Zhang Y, Qu R, He Y, Tian X, Zeng W. Spermatogonial stem cells from domestic animals: progress and prospects. Reproduction 2014; 147:R65-74. 61. Kliszczak M, Stephan AK, Flanagan AM, Morrison CG. SUMO ligase activity of vertebrate Mms21/Nse2 is required for efficient DNA repair but not for Smc5/6 complex stability. DNA Repair (Amst) 2012; 11:799-810. 62. Payne F, Colnaghi R, Rocha N, Seth A, Harris J, Carpenter G, Bottomley WE, Wheeler E, Wong S, Saudek V, Savage D, O'Rahilly S, et al. Hypomorphism in human NSMCE2 linked to primordial dwarfism and insulin resistance. J Clin Invest 2014; 124:4028-4038. 63. Ni HJ, Chang YN, Kao PH, Chai SP, Hsieh YH, Wang DH, Fong JC. Depletion of SUMO ligase hMMS21 impairs G1 to S transition in MCF-7 breast cancer cells. Biochim Biophys Acta 2012; 1820:1893-1900. 64. Chen J, Cai T, Zheng C, Lin X, Wang G, Liao S, Wang X, Gan H, Zhang D, Hu X, Wang S, Li Z, et al. MicroRNA-202 maintains spermatogonial stem cells by inhibiting cell cycle regulators and RNA binding proteins. Nucleic Acids Res 2016. 65. McDonald WH, Pavlova Y, Yates JR, 3rd, Boddy MN. Novel essential DNA repair proteins Nse1 and Nse2 are subunits of the fission yeast Smc5-Smc6 complex. J Biol Chem 2003; 278:45460- 45467. 66. Rai R, Varma SP, Shinde N, Ghosh S, Kumaran SP, Skariah G, Laloraya S. Small ubiquitin- related modifier ligase activity of Mms21 is required for maintenance of chromosome integrity during the unperturbed mitotic cell division cycle in Saccharomyces cerevisiae. J Biol Chem 2011; 286:14516-14530. 67. Fernandez-Capetillo O. The (elusive) role of the SMC5/6 complex. Cell Cycle 2016; 15:775- 776. 68. Hande KR. Etoposide: four decades of development of a topoisomerase II inhibitor. Eur J Cancer 1998; 34:1514-1521. 69. Yang H, Sasaki T, Minoshima S, Shimizu N. Identification of three novel proteins (SGSM1, 2, 3) which modulate small G protein (RAP and RAB)-mediated signaling pathway. Genomics 2007; 90:249-260. 70. Pebernard S, McDonald WH, Pavlova Y, Yates JR, Boddy MN. Nse1, Nse2, and a novel subunit of the Smc5-Smc6 complex, Nse3, play a crucial role in meiosis. Mol Biol Cell 2004; 15:4866- 4876. 71. Guo Y, Hai Y, Gong Y, Li Z, He Z. Characterization, isolation, and culture of mouse and human spermatogonial stem cells. J Cell Physiol 2014; 229:407-413. 76 Chapter 3

72. Manku G, Culty M. Mammalian gonocyte and spermatogonia differentiation: recent advances and remaining challenges. Reproduction 2015; 149:R139-157. 73. Tseng YT, Liao HF, Yu CY, Mo CF, Lin SP. Epigenetic factors in the regulation of prospermatogonia and spermatogonial stem cells. Reproduction 2015; 150:R77-91. 74. Marjault HB, Allemand I. Consequences of irradiation on adult spermatogenesis: Between infertility and hereditary risk. Mutat Res 2016; 770:340-348.

Chapter 4

On the increasing sensitivity of differentiating spermatogonia to DNA damage

Yi Zheng Aldo Jongejan Callista L. Mulder Saskia K.M. van Daalen Sebastiaan Mastenbroek Grace Hwang Philip W. Jordan Sjoerd Repping Geert Hamer

Submitted 78 Chapter 4

Abstract Lifelong mammalian male fertility is maintained through an intricate balance between spermatogonial proliferation and differentiation. DNA damage in spermatogonia, for instance caused by chemo- or radiotherapy, can induce cell cycle arrest or germ cell apoptosis, possibly resulting in male infertility. Because genetic aberrations in spermatogonia can be transmitted to future generations, these cells are generally more radiosensitive, and hence more prone to undergo apoptosis, than somatic cells. Among spermatogonial subtypes the response to DNA damage is differentially modulated; undifferentiated spermatogonia are relatively radio-resistant, whereas differentiating spermatogonia are very radiosensitive. To investigate the molecular mechanisms underlying this difference, we used an in vitro system consisting of primary cultured undifferentiated mouse spermatogonia that can be induced to differentiate. Using RNA-sequencing analysis, we analyzed the response of undifferentiated and differentiating spermatogonia to ionizing radiation (IR). At the RNA level, both undifferentiated and differentiating spermatogonia showed a very similar response to IR. However, at the protein level, undifferentiated spermatogonia showed a much stronger upregulation of p53 in response to IR than differentiating spermatogonia. Our results suggest that the difference in radiosensitivity between undifferentiated and differentiating spermatogonia is largely caused by properties, like chromatin architecture, proliferation activity, protein content or post-translational modifications that are already induced upon differentiation, rather than an alternative gene expression pattern in response to irradiation in both cell types.

Keywords Spermatogonia; Differentiation; Spermatogonial markers; DNA damage response; Transcriptome Transcriptome of irradiated spermatogonia 79

Introduction Spermatogenesis is an intricate process that takes place in the seminiferous tubules within the testis. In mammals, the entire process of spermatogenesis is comprised of three consecutive phases: a mitotic phase (spermatogonial proliferation and differentiation), a meiotic phase (spermatocyte meiotic divisions to generate haploid spermatids) and spermiogenesis (elongation and maturation of spermatids) [1]. For continuous spermatogenesis spermatogonial stem cells (SSCs) are essential. SSCs can be defined as a subpopulation of undifferentiated spermatogonia able to generate and maintain donor- derived spermatogenesis when transplanted into infertile recipient testes [2, 3]. Continuous spermatogenesis requires a constant balance between SSC self-renewal, proliferation and differentiation [1]. Within the seminiferous tubules, spermatogenesis occurs in an orchestrated spatio-temporal fashion in which specific germ cell types are grouped in specific stages of the seminiferous epithelium. The undifferentiated spermatogonia may divide freely during all of these epithelial stages. In contrast, differentiating spermatogonia are irreversibly committed towards meiosis and their subsequent divisions are strictly dictated by the epithelial stage in which they are present [4]. To guarantee the continuous generation of healthy and functional sperm, DNA integrity is constantly being monitored during spermatogenesis [1]. DNA damage, for instance caused by gonadotoxic chemicals or ionizing radiation (IR), can result in gene mutations or chromosomal aberrations and often leads to spermatogenic arrest and subsequent male infertility. Indeed, in the human, treatment with chemo- or radiotherapy in adult males often results in impaired fertility [5]. Because genetic aberrations can potentially be transmitted to future generations when DNA damage checkpoints mechanisms fail during spermatogenesis, spermatogenic cells are thought to be generally more radiosensitive than somatic cells [6]. Because genetic aberrations in SSCs have the potential to result in lifelong generation of mutated sperm, one might expect that the SSCs, when compared to all other spermatogonia, are most prone to undergo apoptosis in response to IR. However, this appears not to be the case. It turns out that differentiating spermatogonia are actually much more radiosensitive and show a stronger apoptotic response [7]. Among the undifferentiated spermatogonia, the self-renewing SSCs are most resistant to DNA damage induced by either the alkylating agent busulfan or IR [8-10]. Evidently, while damaged differentiating spermatogonia are more easily sacrificed, preservation of SSCs, and thus long-term fertility, seems to outweigh a certain risk of mutated offspring. What determines the differences in radiosensitivity among spermatogonial subtypes is currently unknown. One process that characterizes spermatogonial differentiation is the 80 Chapter 4 condensation of repetitive sequences surrounding the centromeres in pericentromeric heterochromatin domains. From spermatogonial differentiation and onwards through meiosis, these heterochromatic regions are marked by the presence of SMC6, a major subunit of the SMC5/6 complex known to be involved in the repair of DNA double-strand breaks (DSBs) [11, 12]. Within these regions the SMC5/6 complex has been postulated to inhibit aberrant homology-driven recombinational repair of DSBs that would otherwise easily occur between repetitive sequences [12]. Hence, this changed chromatin architecture in differentiating spermatogonia may profoundly influence their response to DNA damage. Apart from chromatin architecture, also several DNA damage response proteins are differentially regulated during spermatogonial differentiation. For instance, phosphorylated histone H2AX (ɣ-H2AX), usually marking DSBs, has been described to increase with spermatogonial differentiation [13, 14] and is highly expressed in intermediate and B spermatogonia [15]. The DNA damage response protein p53 has been found to be induced in all spermatogonia by irradiation, but knockout of p53 seems to predominantly affect the apoptotic response of the undifferentiated spermatogonia [16-18]. Nevertheless, transplantation assays of mutated SSCs revealed that deficiency in a specific p53 pathway (Trp53-Trp53inp1-Tnfrsf10b) actually increased survival of SSCs after irradiation [19]. The same study also reported that the apoptosis-inducing protein BBC3 was specifically active in differentiating spermatogonia after irradiation [19]. To investigate the relation between the IR-induced DNA damage response and spermatogonial differentiation, we used an established culture system for undifferentiated mouse spermatogonia [20, 21]. In this culture system, primary isolated mouse SSCs, then referred to as male germline stem (GS) cells, can propagate in vitro for years without losing SSC properties [21]. GS cells can also be induced to differentiate by adding retinoic acid (RA) to the culture medium [22, 23]. Moreover, by way of RNA-sequencing (RNA-seq), the transcriptome of RA-induced differentiating GS cells was reported recently [23]. To gain insights into the differential DNA damage responses of undifferentiated and differentiating spermatogonia, we investigated the transcriptomes of irradiated and non-irradiated GS cells with or without RA treatment.

Materials and methods Animals Neonatal (4-5 d.p.p) DBA/2J male mice were used for GS cell isolation, and adult (~8 weeks) C57BL/6J male mice were used for irradiation and immuno-histochemical analysis. For histological analysis on neonatal testis sections, 8 d.p.p old C57BL/6J male mice were Transcriptome of irradiated spermatogonia 81 used. All animal procedures were in accordance with and approved by the animal ethical committee of the Academic Medical Center, University of Amsterdam or in accordance with the National Institutes of Health and US Department of Agriculture criteria approved by the Institutional Animal Care and Use Committees of Johns Hopkins University.

GS cell culture A mouse GS cell line was established as previously reported [20, 24]. Briefly, testes were harvested from neonatal DBA/2J male mice, and after removing the tunica albuginea, testicular tissues were mechanically dissociated and subjected to a collagenase-trypsin dissociation to obtain a single-cell suspension. Germ cells were enriched by an overnight differential plating and cultured in a medium mainly composed of StemPro-34 SFM medium (Thermo Fisher Scientific), StemPro-34 Supplement (Thermo Fisher Scientific), 1% fetal bovine serum (FBS), recombinant human GDNF (15 ng/ml, Peprotech), recombinant human bFGF (10 ng/ml, Peprotech), as well as other components as previously reported [20]. The cells were cultured on mitotically inactivated mouse embryonic fibroblasts (MEFs) since the third passage, and were refreshed every 2-3 days and passaged every 5-7 days at a ratio of

1:4-6. The cells were maintained at 37°C in an atmosphere of 5% CO2 in air.

RA treatment Before RA treatment, GS cells cultured on MEFs were transferred to laminin (20 µg/ml, Sigma-Aldrich)-coated wells. On the next day, GS cells were treated with 2µM all-trans-RA (Sigma-Aldrich) in culture medium for 48-72 hours. In control groups, vehicle (0.1% ethanol in medium) was applied to the cells.

Ionizing irradiation (IR) Before IR treatment, GS cells cultured on MEFs were transferred to laminin (20 µg/ml, Sigma-Aldrich)-coated wells. On the next day, GS cells were subjected to 1 Gy of IR emitted by a 137Cs source (95% β-emission). To prepare irradiated mice, adult C57BL/6J male mice were exposed to a whole-body IR (1 Gy).

Quantitative-real time PCR (Q-PCR) Total RNA was extracted from GS cells which had been subjected to RA treatment for 2 days or IR 3 hours before, using ISOLATE II RNA Mini Kit (Bioline) and following the protocol provided by the manufacturer. After treatment with DNase (Qiagen) and tests for 82 Chapter 4 genomic DNA free, RNA samples were reversely transcribed, using SensiFAST cDNA Synthesis Kit (Bioline). The synthesized cDNA was then used for Q-PCR reactions, using the Roche LightCycler 480 platform (the 384-well plate format). The Q-PCR reaction was performed in a 10µl volume system including 2× LightCycler 480 SYBR Green I Master (Roche). Ppt2 and Mtg1 were used as reference genes, and the data were analyzed using the -ΔΔCt method. Data were presented as the mean ± standard error of mean (SEM) of 3 independent experiments (n=3). Differences between groups were assessed using the Student’s t-test. P˂0.05 was considered statistically significant and P˂0.01 was considered extremely significant. The primers for Q-PCR analysis are listed in Table 1.

Table 1: Primer sequences for Q-PCR analysis. Gene Forward primer Reverse primer Product size (bp) Ppt2 CCTGCTGGACTATATCAATGAGAC TCTCGGAACCCTTGTACCTG 114 Mtg1 CACGATGTAGCACGCTGGTT GGGTTTCGACCTGAAAATGGG 128 Plzf TCTCGGAACCCTTGTACCTG ACCGAAAGAGGTGGAGACTGA 132 Oct4 CACGAGTGGAAAGCAACTCA CTTCTGCAGGGCTTTCATGT 125 Stra8 GGAAGGCAGTTTACTCCCAGTC GATTCCCATCTTGCAGGTTGA 144 Clu CAGTTCCCAGACGTGGATT GGGCAGGATTGTTGGTTG 157 Ntrk3 TTTGGGGTGTCCATAGCAG AGCCACAGGACCCTTCATT 120 Wnt16 CTTCCCATCAGAAACACCACA GCGGCAGTCCACAGACATTA 114 Agtr2 GAAGAACAGAATTACCCGTGAC AGGGAAGCCAGCAAATGA 80 Bmp2 ATCTGTACCGCAGGCACTC ACGGCTTCTTCGTGATGG 112 Sarm1 CAAGGAGATTGTGACTGCTTTA GGTACTCATGGGACCATTTGA 143 Insm1 GGTGTTCCCCTGCAAGTACT CTATTCTCAGACGGGTGGC 90 Slit2 ATGGAGAACAGAATCAGCACC TCGCAGTCCCGAGAAACA 124 Adgrg1 AGCCAAGTCCTGGGTGAGA TTGATGCCGGGTCTTCAA 154 Myog GTCCCAACCCAGGAGATCA AACAGACATATCCTCCACCGT 107 Pax6 AACAGACATATCCTCCACCGT TATCATAACTCCGCCCATTC 135 Vsx2 CACTACCCAGATGTCTACGCC CACTTCTCCCTCTTCCTCCAC 116 Adora1 AACCCAGCATCCTCATCTACA GTGGTCGTTCCAAATCTTCA 125 Sfrp2 GTGGTCGTTCCAAATCTTCA GCTCTTTGTCTCCAGGATGAT 130 Irf1 AATGCGGATGAGACCCTG ATGTCCCAGCCGTGCTTA 129 Pdx1 AATCCACCAAAGCTCACGC CGGGTCCTCTTGTTTTCCTC 82 Bbc3 GAGCGGCGGAGACAAGAAGA ATCCCTGGGTAAGGGGAGGA 96 Plk2 ATGTGGAACCCCAAATTATCTC GGTCTTCCTAGCAGCATCGTAT 114 Trp53inp1 GACACCAGTGATTCCTGCTTC GGACTTGTTTCCACCTTGATAG 122 Transcriptome of irradiated spermatogonia 83

Ddias TGTCCTTGAAAGTGGCAGAA GTGTAAACCAGTGGCCGTAA 98 Ccng1 ATAATGGCCTCAGAATGACTGC CCAAGATGCTTCGCCTGTAC 160 Klhl42 CAATCCCATCACCAACGAG TAGAAACAGCCTGCCCACC 116 Psrc1 TGCCCACCGTGAGTTCTT GTGGGTGATTCCTTCTTTATGC 139 Eda2r ACTTGTGCTGTCATCAATCGG CGTGTCTTTCGGTAGAACCTG 98 Pdrg1 GGAAGGAGCCAAGGTGAAGT CCTCGGCCAACTCCTCTAC 117 Sesn2 AACTACATCCACTGCGTCTTTG CATCCTACGGGTCGTCTTCT 135 Slc2a10 TACTTGTTCCTGAAACCAAAGG TCCAGGCGATGGTACTGAA 121 Far2 TTATTGGAACACCGTCAGCC CAGCATTCTGGGTTTCCTTC 85

Western blot Proteins were extracted from the cells and quantified with Qubit Protein Assay Kit (Thermo Fisher Scientific). Then Western blot analysis was performed as reported previously [12, 24, 25], using the LI-COR Odyssey imaging system (LI-COR Biosciences). The primary antibodies used were mouse anti-PLZF (1:100; D-9, Santa Cruz Biotechnology), mouse anti- OCT4 (1:200; C-10, Santa Cruz Biotechnology), rabbit anti-STRA8 (1:500; ab49602, Abcam), rabbit anti-p53 (1:100; FL-393, Santa Cruz Biotechnology), rabbit anti-GAPDH (1:400; FL- 335, Santa Cruz Biotechnology) and mouse anti-β-actin (1:5,000; A1978, Sigma-Aldrich). For quantification of the relative p53 expression, p53 band intensity is divided by that of GAPDH. Data were presented as the mean ± SEM of 4 independent experiments (n=4). Differences among groups were assessed using the one-way ANOVA followed by LSD test. P˂0.05 was considered statistically significant and P˂0.01 was considered extremely significant.

Immunocytochemistry (ICC) and immunohistochemistry (IHC) For ICC, GS cells were grown on laminin (20 µg/ml, Sigma-Aldrich)-coated glass coverslips in 24-well plates. In case of IR treatment, GS cells were fixed in 4% paraformaldehyde (PFA) at 3 hours post IR. Cells were permeabilized and blocked as previously described [24, 25], followed by 4°C overnight incubation with the following primary antibodies: mouse anti-PLZF (1:50; D-9, Santa Cruz Biotechnology), mouse anti-OCT4 (1:50; C-10, Santa Cruz Biotechnology), rabbit anti-LIN28A (1:1,000; ab46020, Abcam), rabbit anti- ID4 (1:100; M106, CalBioreagents), mouse anti-ɣ-H2AX (1:20,000; 05-636, Merck Millipore) and guinea pig anti-SMC6 (1:400; custom made, peptide: KRPRQEELEDFDKDGDEDE [24, 25]). Replacing primary antibodies with phosphate buffered saline (PBS) containing 0.5% bovine serum albumin (BSA) was used as negative controls. On the next day, the cells were washed and incubated with the corresponding host-specific secondary antibodies (Alexa 84 Chapter 4

Fluor 488 or 555, 1:1,000; Thermo Fisher Scientific), and counterstained with DAPI. The cells were mounted on glass slides with the Prolong Gold anti-fade mountant (Thermo Fisher Scientific) and later visualized under the microscope. Microscopy was performed as previously described [24]. For IHC, testes were collected from normal adult and neonatal mice, or the experimental and control mice at 3 hours post IR, fixed in diluted Bouin’s solution or 4% PFA, and embedded in paraffin. Testis sections were sliced at a thickness of 5µm. After deparaffinization and rehydration, testis sections were alternatively subjected to microwave- mediated antigen retrieval in sodium citrate buffer (0.01M, pH 6.0), followed by blocking in Super Block (ScyTek Laboratories) or PBS with 1% BSA and 0.1% Tween-20, for 1 hour at room temperature (RT). Then the sections were incubated with primary antibodies diluted in Normal antibody Diluent (ImmunoLogic) or PBS containing 0.5% BSA and 0.1% Tween-20, at 4°C overnight. The primary antibodies used were rabbit anti-WNT16 (1:100; H-96, Santa Cruz Biotechnology), rabbit anti-SLIT2 (1:80; ab7665, Abcam), rabbit anti-PDX1 (1:1,000; ab47267, Abcam), rabbit anti-ADORA1 (1:600; ab82477, Abcam), rabbit anti-p53 (1:100; FL- 393, Santa Cruz Biotechnology), rabbit anti-BBC3 (1:250; ab9643, Abcam), rabbit anti-PLK2 (1:100; H-90, Santa Cruz Biotechnology), rabbit anti-PDRG1 (1:100; 16968-1-AP, Proteintech) and rabbit anti-SESN2 (1:100; 10795-1-AP, Proteintech). The isotype rabbit IgG was used in negative controls. On the next day, the sections were washed and incubated with Powervision Poly-HRP-Anti-mouse/rabbit/rat secondary antibody (ImmunoLogic) for 1 hour at RT. After washing, the sections were stained with diaminobenzidine (DAB) and counterstained with hematoxylin. Then the slides were dehydrated, embedded in Entellan (Merck Millipore), and visualized under the microscope. Microscopy was performed as previously described [12].

RNA-sequencing (RNA-seq) analysis Total RNA was extracted from GS cells which had been subjected to RA treatment for 2 days and/or IR 3 hours before, using ISOLATE II RNA Mini Kit (Bioline) and following the protocol provided by the manufacturer. After treatment with DNase (Qiagen) and tests for genomic DNA-free, RNA samples from 3 independent experiments (biological triplicates) were sent to BGI Tech Solutions (HongKong) where libraries were constructed (TruSeq, 160bp) and paired-end sequencing was performed (101bp, Illumina HiSeq 4000). RNA-seq analysis was conducted as described previously [24]. In brief, clean reads were subjected to quality control, and then aligned to UCSC mm10 GRCm38.p4 GTF using HISAT2 (v2.0.4) [26]. Counts were obtained using HTSeq (v0.6.1) [27]. Count tables were made and all genes without counts in any of the samples were removed, whilst genes with more than 1 Transcriptome of irradiated spermatogonia 85 count-per-million reads (CPM) in 2 of the samples were kept. Genes were re-annotated using biomaRt and mm10 of Ensembl. Count data were transformed to log2-counts per million (logCPM) using voom, estimating the mean-variance relationship. Differential expression was assessed using a moderated t-test and the linear model framework from the limma package. Benjamini-Hochberg false discovery rate was used to correct for multiple testing of the resulting p-values. The entire analysis was performed using R v3.2.2 and Bioconductor v3.0 [28, 29]. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were conducted using DAVID bioinformatics resources 6.8 [30, 31], and Gene Set Enrichment Analysis (GSEA) [32, 33] was used to analyze differentially expressed gene sets.

Accession numbers All sequencing data have been submitted to NCBI (SRA) and will be available under the accession number: SUB2838515 (BioProject ID PRJNA392875).

Results Culture, differentiation and irradiation of GS cells First, we established a GS cell line according to a previously published and well- demonstrated protocol [20]. GS cells can be maintained on either a feeder layer of mouse embryonic fibroblasts (MEFs) or in laminin-coated wells [34]. Like previously described [34], GS cells cultured on MEFs formed typical grape/morula-like colonies, whereas they started to form chain-like structures when seeded in laminin-coated wells (Figure 1A). The cultured GS cells were positive for putative SSC/progenitor markers PLZF, LIN28A, OCT4 and ID4 (Figure 1B), indicative of their undifferentiated spermatogonial properties. To induce GS cell differentiation, we transferred them to laminin-coated wells and added RA to the culture medium. Multiple studies have shown that RA can drive spermatogonial differentiation both in vivo and in vitro [22, 23, 35-37]. Consistent with these reports, exposure to RA for 3 days made the GS cells gradually lose the structure characteristic for undifferentiated spermatogonia and display an enlarged cell size and round morphology with clear cellular boundaries (Figure 1C), indicative of spermatogonial differentiation. Q-PCR and Western blot analyses showed substantial downregulation of the SSC self-renewal genes Plzf and Oct4, while RNA and protein levels of the differentiation marker Stra8 markedly increased after 3 days of RA exposure (Figure 1D, E). These results demonstrate that in vitro spermatogonial differentiation was successfully induced in our culture system. 86 Chapter 4

To investigate their response to induced DSBs, GS cells on laminin were subjected to 1Gy of IR, a dose that causes substantial DNA damage but does not necessarily kill these cells [7]. To visualize the DSBs, we stained the cells with ɣ-H2AX, a widely used DSB marker. Before IR, the cells displayed a weak ɣ-H2AX staining. Three hours post IR, clear nuclear ɣ- H2AX foci were observed (Figure 1F), demonstrating the induction of DSBs. In line with our recent findings showing that, in contrast to somatic cells [25], SMC6 does not co-localize with ɣ-H2AX at etoposide-induced DSBs in spermatogonia [24], spermatogonial SMC6 did not co- localize with IR-induced ɣ-H2AX foci (Figure 1F, right panel).

Transcriptome of irradiated spermatogonia 87

Figure 1: In vitro culture, differentiation and irradiation of GS cells. (A) Representative images of GS cells cultured on MEFs or laminin. Bar = 100µm. (B) PLZF, LIN28A, OCT4 and ID4 staining of GS cells on laminin. NC: omitting primary antibodies. Bar = 10µm. (C) Representative images of GS cells (on laminin) with/without exposure to RA. Asterisks (*) refer to the long extensions typical for undifferentiated spermatogonia. Bar = 25µm. (D) Q-PCR analysis of Plzf, Oct4 and Stra8 expression in GS cells with/without exposure to RA. Data are presented as the mean ± standard error of mean (SEM), n=3. *: P˂0.05; **: P˂0.01. (E) Western blot analysis of PLZF, OCT4 and STRA8 expression in GS cells with/without exposure to RA. β-actin or GAPDH is used as the loading control. (F) ɣ-H2AX and SMC6 staining in GS cells with/without IR. NC: omitting primary antibodies. Left panel: bar = 10µm; right panel: bar = 5µm.

Spermatogonial differentiation induces a high transcriptomic change Differentiating spermatogonia are known to be much more radiosensitive than undifferentiated spermatogonia [7]. To study the underlying molecular mechanism causing this increased radiosensitivity, we performed a whole transcriptomic RNA-seq analysis for GS cells with/without RA and/or IR treatment (Figure 2A). A multidimensional scaling (MDS) plot and heat map showed that RA treatment triggered a much larger transcriptomic change than IR (Figure 2B). 1,748 upregulated and 966 downregulated differentially expressed genes (DEGs, adj.P<0.05 and fold change>2) resulted from RA treatment (Table S1). Upregulated DEGs included markers for differentiating spermatogonia and (pre-)meiotic germ cells such as c-Kit, Stra8, Rec8, Prdm9. Likewise, well-known marker genes for SSCs/progenitors, e.g. Plzf, Oct4, Gfra1, c-Ret, Nanos2, Nanos3, Lin28A, Id4 and Pax7, showed a significant downregulation (Table S1). Gene Ontology (GO) analysis demonstrated that RA-induced upregulated genes were involved in cell adhesion, differentiation and response to retinoic acid. Downregulated genes were mainly related to transcription, proliferation, regulation of cell death and apoptosis (Figure 2C, Table S5). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis revealed that differentiation-upregulated genes were enriched with pathways mediating extracellular matrix (ECM)-receptor interaction, PI3K-Akt and Rap1 signaling, whereas downregulated genes were primarily related to stem cell pluripotency regulation and cancer pathways (Figure 2D, Table S6). We further conducted a Q-PCR analysis for 16 RA-induced DEGs representative of the aforementioned cellular processes (Figure 2E). For most genes the Q-PCR results were in line with the RNA-seq data, confirming the validity of the RNA-seq analysis.

88 Chapter 4

Figure 2: Transcriptomic profiles of GS cells in response to RA and IR treatment. (A) A schematic overview of the experimental design. (B) The MDS plot (left panel) and heat map (right panel) showing the expression profiles of 4 experimental groups of cells (-RA-IR, +RA-IR, -RA+IR, +RA+IR). (C) GO term enrichment analysis of RA-induced up- and downregulated genes. (D) The representative enriched pathways shown by KEGG analysis. (E) Q-PCR analysis of a panel of RA- induced DEGs. Data are presented as the mean ± SEM, n=3. *: P˂0.05; **: P˂0.01. (F) Q-PCR analysis of a panel of IR-induced DEGs. Data are presented as the mean ± SEM, n=3. *: P˂0.05.

Transcriptome of irradiated spermatogonia 89

Undifferentiated and differentiating spermatogonia display similar transcriptomic changes in response to IR Also IR yielded transcriptomic variations in both undifferentiated and differentiating spermatogonia (Figure 2B). Between the -RA-IR and -RA+IR cell populations, 31 and 8 DEGs (adj.P<0.05 and fold change>2) were up- and downregulated, respectively (Table S2). KEGG analysis disclosed that IR-upregulated genes, including Bbc3, Gtse1, Ccng1, Cdkn1a and Sesn2, were predominantly related to the p53 signaling pathway. To validate the RNA- seq data, we used Q-PCR to quantify the expression of 12 IR-induced DEGs that are implicated in the p53 signaling pathway, cell cycle arrest or apoptosis. The Q-PCR results were in line with the RNA-seq data for most genes (Figure 2F). Next, we investigated IR-induced DEGs in differentiating spermatogonia. Between the +RA-IR and +RA+IR cell populations, there were 20 DEGs (adj.P<0.05 and fold change>2) that all showed upregulation (Table S3). When comparing these 20 DEGs from the +RA-IR vs +RA+IR comparison to the 39 DEGs from the -RA-IR vs -RA+IR comparison, we found that 15 DEGs, including well-known IR response genes such as Bbc3, Gtse1 and Ccng1, were upregulated by IR in both undifferentiated and differentiating spermatogonia. There were 24 DEGs only belonging to the -RA-IR vs -RA+IR comparison, and 5 DEGs only belonging to the +RA-IR vs +RA+IR comparison. One of the reasons that these 24 and 5 specific DEGs were found when simply comparing the (-RA-IR vs -RA+IR) to the (+RA-IR vs +RA+IR) group is the threshold used in our study (adj.P<0.05 and fold change>2). Genes can be found differentially expressed only in one group while being just below the threshold in the other group. Moreover, expression of these DEGs can be influenced by RA-induced differentiation independent of IR. To overcome these problems, we also directly compared the -RA+IR and +RA+IR cell populations after correction for their respective baseline levels of gene expression without exposure to IR (i.e. -RA-IR and +RA-IR, respectively). In this way, the gene expression affected by differentiation independent of IR is separated from that in response to IR. Apart from the genes affected by RA treatment independent of IR, no significant DEGs (adj.P<0.05 and fold change>2) were detected between the -RA+IR and +RA+IR cell populations (Table S4). Hence, undifferentiated and differentiating spermatogonia do not show significantly differential gene expression in response to IR. In order to find significant up- or downregulation of specific pathways or gene sets, we performed a Gene Set Enrichment Analysis (GSEA). In contrast to GO or KEGG analysis, GSEA is not based on observed DEGs but instead takes into account all minor variations at 90 Chapter 4 the whole transcriptome level. GSEA showed that mesenchymal transition pathways went up in +RA+IR cell populations compared to their -RA+IR counterparts (Table S7).

Undifferentiated spermatogonia display a more robust p53 induction in response to IR Because the IR-upregulated genes were predominantly related to the p53 signaling pathway and p53 is known to be upregulated in spermatogonia in response to IR [17], we performed a Western blot analysis to investigate IR-induced p53 in both undifferentiated and differentiating spermatogonia. Interestingly, we found that undifferentiated spermatogonia displayed a much stronger induction of p53 than differentiating spermatogonia at 3 hours post IR (Figure 3).

Figure 3: Western blot analysis of p53 protein levels in -IR-RA, -IR+RA, +IR-RA, +IR+RA cell groups. GAPDH is used as a loading control. p53 band intensity is divided by that of GAPDH. Data are presented as the mean ± SEM, n=4. *: P˂0.05; **: P˂0.01.

Protein localization of differentiation-induced genes in the testis To study the protein localization of differentiation-induced DEGs, we performed immunohistochemistry (IHC) on testis sections. We examined the protein localization patterns of 4 RA-induced DEGs (Wnt16, Slit2, Pdx1, Adora1) for which working antibodies were available in adult mouse testes (Figure 4A-D). In accordance with the RNA-seq and Q-PCR data showing differentiation-induced upregulation of Wnt16 gene expression, WNT16 staining was observed in the cytoplasm of pachytene spermatocytes, round and elongating spermatids (Figure 4A). Again, in line with the RNA-seq and Q-PCR data showing differentiation-induced upregulation of Slit2 gene Transcriptome of irradiated spermatogonia 91 expression, we observed SLIT2 staining in the nuclei of round spermatids. In addition, some staining was observed in chromatid bodies in pachytene spermatocytes (Figure 4B).

92 Chapter 4

Figure 4: The protein localization of several RA-induced DEGs in the adult testis. Shown are localization of (A) WNT16, (B) SLIT2, (C) PDX1 and (D) ADORA1 in adult testis sections. Examples of several cell types are marked: A, type A spermatogonia; Int, intermediate spermatogonia; B, type B spermatogonia; pL, pre-leptotene spermatocytes; L, leptotene spermatocytes; Z, zygotene spermatocytes; P, pachytene spermatocytes; D, diplotene spermatocytes; R, round spermatids; E, elongating spermatids; Ser, Sertoli cells; Ley, Leydig cells. Stages of the seminiferous epithelium are indicated with Roman numerals. Bar = 20µm. Negative controls, using the isotype rabbit IgG, do not show any staining apart from interstitial cells (Figure S1).

RNA-seq and Q-PCR analyses showed that Pdx1 was significantly downregulated by RA-induced differentiation. In the testis, PDX1 staining was exclusively found in the nuclei of a subpopulation of type A spermatogonia (Figure 4C). Spermatogonia with clear heterochromatin patches at the rim of their nuclei, indicative of spermatogonial differentiation, and intermediate and type B spermatogonia did not stain for PDX1. To pinpoint whether it is the undifferentiated spermatogonia that are positive for PDX1 staining, we performed IHC on testis sections from neonatal (8 d.p.p) mice. Indeed, PDX1 staining was observed in undifferentiated spermatogonia but not in Sertoli cells (Figure 5A), consistent with its staining pattern in adult testis sections. While RA-induced differentiation significantly lowered the number of Adora1 transcripts, ADORA1 protein localization in the testis appeared more dynamic. Like for PDX1, ADORA1 staining was observed in the nuclei of some but not all type A spermatogonia. Also pachytene spermatocytes and round spermatids up to stage VII of the seminiferous epithelium were stained (Figure 4D). To investigate whether ADORA1 positive spermatogonia represent the undifferentiated spermatogonial population, we again performed IHC on neonatal testis sections. However, unlike for PDX1, all neonatal testicular cells, consisting of undifferentiated spermatogonia and Sertoli cells, were negative for ADORA1 staining (Figure 5B).

Transcriptome of irradiated spermatogonia 93

Figure 5: The protein localization of PDX1 (A) and ADORA1 (B) in the neonatal testis. Asterisks (*) indicate the positive undifferentiated spermatogonia. Bar = 20µm. Negative controls, using the isotype rabbit IgG, do not show any staining apart from interstitial cells (Figure S1).

Protein localization of irradiation-induced genes in the testis To study the protein localization of irradiation-induced genes in vivo, adult mice were subjected to a whole-body IR (1Gy), after which the protein localization patterns of 4 IR- induced DEGs (Bbc3, Plk2, Pdrg1, Sesn2) for which working antibodies were available were studied. IHC was performed on testis sections from experimental (3 hours post IR) and control (no IR treatment) mice. p53 was used as a positive control for the testicular radiation response [17]. As shown in Figure 6A, nuclear p53 staining was clearly induced in spermatogonia after IR. In line with our RNA-seq and Q-PCR data and a previous report [19], IR induced a strong increase of BBC3 staining in all testicular cells (Figure 6B). PLK2, on the other hand, although significantly induced at the RNA level in GS cells, showed no significant difference in response to IR with regard to its localization in vivo. PLK2 staining was observed in pachytene spermatocytes and all subsequent germ cells but not in spermatogonia, (pre-)leptotene or zygotene spermatocytes (Figure 6C). Like PLK2, also PDRG1 staining in the testis was not influenced by IR treatment. Although all germ cells displayed a vague cytoplasmic staining, clear nuclear PDRG1 staining was observed in spermatocytes and round spermatids (Figure 6D). SESN2 staining was present in Sertoli cells before IR. After IR, SESN2 staining appeared more intense and was clearly induced in spermatogonia and, albeit less intense and consistent, in spermatocytes (Figure 6E).

Discussion By perfectly balancing self-renewal and differentiation, SSCs are capable of maintaining lifelong male fertility. For rodents, long-term culture systems that retain SSC capabilities have been well established [20, 46, 47]. In these culture systems, undifferentiated spermatogonia, then termed GS cells, can proliferate exponentially in vitro and, even after years of culture, still hold the capacity to initiate and maintain spermatogenesis upon transplantation into recipient testes [21]. Moreover, not only self- renewal but also spermatogonial differentiation can be induced in this culture system, providing an advantageous in vitro system to interrogate this process. Several factors, such as bone morphogenetic protein 4 (BMP4), activin A and RA, have been reported to be involved in the regulation of spermatogonial differentiation [48-50]. Of these, 94 Chapter 4

RA, the active metabolite of vitamin A, plays pivotal roles in the transition of undifferentiated spermatogonia into differentiating spermatogonia, as well as in the initiation and progression of meiosis [50]. Several studies have demonstrated that RA can be used to induce spermatogonial differentiation in culture [22, 23]. Specifically, RA exposure significantly downregulates transcription factors essential for SSC self-renewal (e.g. PLZF, OCT4), while it upregulates spermatogonial differentiation markers such as c-KIT and STRA8 [22]. Further demonstrating spermatogonial differentiation, RA-treated GS cells exhibit a significantly reduced colonization ability after transplantation [22]. Moreover, a recent article reported that RA alone was able to induce GS cell differentiation into zygotene spermatocytes [23]. Intriguingly, differentiation increases the spermatogonial sensitivity to DNA damage in vivo, for instance DNA damage induced by irradiation [7]. To study the increasing radiosensitivity of differentiating spermatogonia, we first used RA to induce GS cell differentiation as described previously [22, 23] and then compared the transcriptomes of undifferentiated and differentiating spermatogonia. Our differentiation protocol induced the transcriptomic alteration characteristic for spermatogonial differentiation. Markers for differentiating spermatogonia and (pre-)meiotic germ cells were significantly upregulated, whereas well-known markers for SSCs/progenitors or genes essential for SSC self-renewal showed a substantial downregulation. GO and KEGG analyses further uncovered that, apart from the expected response to RA, differentiation orchestrated the expression of genes involved in diverse biological processes such as cell adhesion, cell proliferation and differentiation, regulation of cell death and apoptosis, and gene transcription. Collectively, in terms of transcriptomic variation, RA-induced differentiation of GS cells for a large part mimics in vivo spermatogonial differentiation, supporting the use of this in vitro model for our study. We further examined the protein localization patterns of several RA-induced DEGs in vivo. Wnt16 and Slit2 were upregulated by RA treatment, as shown by our RNA-seq and Q- PCR data. Previous papers have reported that WNT signaling regulates the growth and differentiation of stem cells including SSCs [38, 39]. As a member of the WNT family, WNT16 has been found to be associated with estrogen withdrawal and bone loss during aging [51], and it regulates periosteal bone formation via activation of the canonical WNT signaling pathway [52]. We observed WNT16 staining mainly in the cytoplasm of pachytene spermatocytes and spermatids. SLIT2, a member of the SLIT family, has been reported to inhibit endothelial cell proliferation and migration [40]. Similar to WNT16, SLIT2 staining was observed in advanced germ cells, in this case the nuclei of round spermatids. Both protein Transcriptome of irradiated spermatogonia 95 localization patterns illustrate the idea that genes not required until later stages of spermatogenesis can already be expressed upon spermatogonial differentiation.

96 Chapter 4

Figure 6: The protein localization of several IR-induced DEGs in the adult testis. Shown are localization of (A) p53, (B) BBC3, (C) PLK2, (D) PDRG1 and (E) SESN2 in adult testis sections with/without IR treatment. Examples of several cell types are marked: A, type A spermatogonia; Int, intermediate spermatogonia; B, type B spermatogonia; pL, pre-leptotene spermatocytes; L, leptotene spermatocytes; Z, zygotene spermatocytes; P, pachytene spermatocytes; D, diplotene spermatocytes; M, metaphase I spermatocytes; R, round spermatids; E, elongating spermatids; Ser, Sertoli cells; Ley, Leydig cells. Stages of the seminiferous epithelium are indicated with Roman numerals. Bar = 20µm. Negative controls, using the isotype rabbit IgG, do not show any staining apart from interstitial cells (Figure S1).

Pdx1 and Adora1 showed a significant downregulation upon RA treatment. PDX1 is a transcription factor that acts as a regulator of pancreatic development and β cell function [41]. Intriguingly, we found staining of PDX1 exclusively in a subpopulation of type A spermatogonia. Further IHC analysis on neonatal testis sections demonstrated that the spermatogonia positive for PDX1 staining were indeed undifferentiated spermatogonia. PDX1 may thus be a suitable novel marker for these cells. Because Pdx1 is downregulated upon spermatogonial differentiation, it may be involved in SSC self-renewal or proliferation. ADORA1 belongs to the G protein-coupled receptor 1 family and plays a role in modulating respiration and metabolism [53, 54]. Like PDX1, ADORA1 staining was observed in some but not all type A spermatogonia in the adult testis. However, ADORA1 staining was absent from spermatogonia in the neonatal testis. Also pachytene spermatocytes and a fraction of round spermatids were stained, suggestive of a more dynamic role for ADORA1 during spermatogenesis. Interestingly, ADORA1 staining disappeared from round spermatids after stage VII of the seminiferous epithelium. This is exactly the stage at which spermatogonial differentiation and meiotic initiation occur in response to RA signaling [55-57]. After stage VII, the spermatid nuclei start to elongate while the cytoplasm and a more pronounced flagellum move to the luminal side of the seminiferous tubule [55, 56]. It may thus be the case that, at stage VII of the seminiferous epithelium, also spermiogenesis is regulated by RA signaling. To unravel why differentiating spermatogonia are more radiosensitive than their undifferentiated counterparts, we compared the molecular response of both undifferentiated and differentiating spermatogonia to IR by way of RNA-seq. We found that, at the transcriptome level, undifferentiated and differentiating spermatogonia displayed a similar response to IR. When comparing undifferentiated and differentiating spermatogonia, there were no IR-induced DEGs that could be attributed to spermatogonial differentiation. Thus, it is likely that the increase in radiosensitivity of differentiating spermatogonia is mainly caused by properties, like chromatin architecture, proliferation activity, protein content or post- Transcriptome of irradiated spermatogonia 97 translational modifications that are already induced upon differentiation, rather than the additional expression of genes that is induced by irradiation in both cell types. Our RNA-seq data uncovered RA-induced upregulation of apoptosis-associated genes (e.g. apoptotic peptidase activating factor 1 (Apaf1)) and downregulation of genes suppressing apoptosis (e.g. Bcl2l1, Table S1). The increased radiosensitivity of differentiating spermatogonia may thus be primed by the presence of genes that favor apoptosis over cell cycle arrest and DNA damage repair. Notably, while RNA-seq did not reveal significant DEGs between -RA+IR and +RA+IR spermatogonia after correction for their respective baselines, GSEA, which does not depend on observed DEGs but instead takes all the small contributions of the different genes and looks at them in a pathway setting, did reveal upregulation of the epithelial-mesenchymal transition (EMT) pathway in irradiated differentiating spermatogonia. EMT is a dynamic developmental process by which epithelial cells that normally interact with basal membranes convert to migratory cells with fibroblast-like morphology and mesenchymal secretory characteristics [58]. SSCs are typically regarded as an intermediate cell category between epithelial and mesenchymal cells, since they express markers for both epithelial cells (e.g. CDH1 [59]) and mesenchymal cells (e.g. THY1 [60]). Hence, supported by our RNA-seq data showing the downregulation of CDH1 and upregulation of THY1 by RA treatment (Table S1), spermatogonial differentiation, characterized by movement from the basal membrane towards the lumen, can be associated with enhanced EMT. GS cells can, albeit rarely, spontaneously reprogram to pluripotency, becoming multipotent GS (mGS) cells. Interestingly, EMT has recently been found to block this reprogramming of GS cells to pluripotency [58]. Moreover, mGS cells have been shown to be more resistant to IR than GS cells [19]. The other way around, it could thus make sense that differentiating spermatogonia, which are associated with increased EMT and decreased level of pluripotency, are less resistant to IR. In addition to upregulation of the EMT pathway revealed by GSEA, differentiating spermatogonia displayed a less robust increase of p53 protein levels in response to IR. p53 is a well-known tumor repressing factor. Activated by post-translational modifications, it regulates the transcription of genes involved in cell cycle arrest and apoptosis [61, 62]. p53 can initiate apoptosis to eliminate damaged cells or, alternatively, arrest the cell cycle. The latter will give the cells an opportunity to repair the damaged DNA before resuming the cell cycle [61, 62]. Yet, how p53 signaling triggers one of these cascades remains unknown for most cell types. Our Western blot data showed a more robust p53 induction in undifferentiated spermatogonia in response to IR. As undifferentiated spermatogonia are relatively radio-resistant, it seems plausible that spermatogonial p53 preferentially induces 98 Chapter 4 cell cycle arrest. Differentiating spermatogonia, which induce less p53 in response to IR, would then be less likely to undergo cell cycle arrest and thus have fewer opportunities to repair the inflicted DNA damage. This would then render differentiating spermatogonia more radiosensitive. We also examined the protein localization patterns of several IR-induced DEGs in vivo. Our RNA-seq and Q-PCR data demonstrated a clear upregulation of Bbc3, Plk2, Pdrg1 and Sesn2 in irradiated spermatogonia. As a positive control for the IR-induced DNA damage response, we stained testis sections from irradiated and non-irradiated mice with an anti-p53 antibody. Consistent with a previous literature [17], p53 staining was clearly discerned in spermatogonia after IR treatment. BBC3 is known to be a key determinant of the intrinsic IR- induced p53-dependent apoptotic pathway [42]. A previous study showed that spermatogonial BBC3 was specifically upregulated in differentiating spermatogonia by IR, whereas its knockdown attenuated apoptosis and increased spermatogonial survival [19]. We found an increase of BBC3 staining in all testicular cells after IR treatment, suggesting a general role in the testicular response to DNA damage. PLK2 is a serum-inducible kinase that functions in cell proliferation [43]. Its expression initiates at the G1 phase of the cell cycle, and it was recently found to modulate mitotic spindle orientation as well as mammary gland development [63]. Our protein staining showed that IR treatment did not change the expression pattern of PLK2 in the testis. Interestingly, PLK2 staining was not observed in spermatogonia, (pre-)leptotene or zygotene spermatocytes but in more advanced germ cells, in line with a previous article [64] and suggesting a role in meiosis and spermiogenesis. PDRG1, modulated by p53, is known to be induced by ultraviolet (UV) irradiation [44] and was recently identified as a novel tumor marker potentially functioning in cancer development [65]. In addition, Pdrg1 transcripts have been shown to be highly expressed in human testes [44]. Like that for PLK2, we found PDRG1 staining, irrespective of IR treatment, in pachytene spermatocytes and round spermatids. Also SESN2 is a tumor suppressor implicated in the p53 signaling pathway, modulating the cellular response to IR [66]. Before IR, we observed SESN2 staining in the Sertoli cells. In the IR group, this staining appeared more intense. Furthermore, after IR, also the spermatogonia were clearly stained, suggesting a role for SESN2 in the spermatogonial response to IR. Male cancer patients subjected to chemo- or radiotherapy are often confronted with sub-fertility caused by spermatogonial apoptosis. Nevertheless, the specific DNA damage- induced gene expression profiles of undifferentiated and differentiating spermatogonia have not been studied so far. We took advantage of a well-established in vitro model of spermatogonial differentiation to study the increasing radiosensitivity of differentiating spermatogonia, and found no DEGs that were induced specifically in undifferentiated or Transcriptome of irradiated spermatogonia 99 differentiating spermatogonia in response to IR. Nevertheless, at the protein level, undifferentiated spermatogonia showed a stronger upregulation of p53 in response to IR than differentiating spermatogonia. We therefore propose that the increased sensitivity of differentiating spermatogonia to IR-induced DNA damage is largely determined by properties, like chromatin architecture, proliferation activity, protein content or post-translational modifications that are already induced upon differentiation. The difference in radiosensitivity may reflect the divergent roles of undifferentiated and differentiating spermatogonia in maintaining genome integrity and male fertility. When damaged, differentiating spermatogonia can easily be sacrificed, thereby preventing the transmission of mutations to the offspring. When the differentiating spermatogonia are lost, the SSCs are still there to reinitiate spermatogenesis and preserve long-term fertility. In contrast, elimination of undifferentiated spermatogonia could potentially remove the SSC pool and lead to permanent infertility. No longer being able to reproduce is an evolutionary dead end. The undifferentiated spermatogonia may have therefore developed a relatively high resistance to DNA damage. From an evolutionary point of view, mutated offspring is better than no offspring.

Author contribution Yi Zheng and Geert Hamer conceived and designed the experiments. Yi Zheng, Callista L. Mulder, Saskia K.M. van Daalen and Grace Hwang performed the experiments. Yi Zheng, Aldo Jongejan and Geert Hamer analyzed the data. Yi Zheng and Geert Hamer wrote the manuscript. Sebastiaan Mastenbroek, Philip W. Jordan, Sjoerd Repping and Geert Hamer read and revised the manuscript.

Conflict of interest There is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding This study has been supported by an AMC Fellowship, the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (CIG 293765) to G.H., the China Scholarship Counsel (CSC) number 201306300081 to Y.Z., National Institutes of Health (NIH) K99/R00 HD069458 and NIH R01 GM117155 to P.W.J., and NIH training grant CA009110 fellowship to G.Hw.

100 Chapter 4

Acknowledgment We thank Klaas Franken and the Laboratory for Experimental Oncology and Radiobiology, AMC Amsterdam, for assistance with and use of their 137Cs source for IR. Transcriptome of irradiated spermatogonia 101

References

1. Jan SZ, Hamer G, Repping S, de Rooij DG, van Pelt AM, Vormer TL. Molecular control of rodent spermatogenesis. Biochim Biophys Acta 2012; 1822:1838-1850. 2. Brinster RL, Avarbock MR. Germline transmission of donor haplotype following spermatogonial transplantation. Proc Natl Acad Sci U S A 1994; 91:11303-11307. 3. Brinster RL, Zimmermann JW. Spermatogenesis following male germ-cell transplantation. Proc Natl Acad Sci U S A 1994; 91:11298-11302. 4. de Rooij DG, Russell LD. All you wanted to know about spermatogonia but were afraid to ask. J Androl 2000; 21:776-798. 5. Meistrich ML. Effects of chemotherapy and radiotherapy on spermatogenesis in humans. Fertil Steril 2013; 100:1180-1186. 6. Marjault HB, Allemand I. Consequences of irradiation on adult spermatogenesis: Between infertility and hereditary risk. Mutat Res 2016; 770:340-348. 7. van der Meer Y, Huiskamp R, Davids JA, van der Tweel I, de Rooij DG. The sensitivity to X rays of mouse spermatogonia that are committed to differentiate and of differentiating spermatogonia. Radiat Res 1992; 130:296-302. 8. van Beek ME, Meistrich ML, de Rooij DG. Probability of self-renewing divisions of spermatogonial stem cells in colonies, formed after fission neutron irradiation. Cell Tissue Kinet 1990; 23:1-16. 9. Aloisio GM, Nakada Y, Saatcioglu HD, Pena CG, Baker MD, Tarnawa ED, Mukherjee J, Manjunath H, Bugde A, Sengupta AL, Amatruda JF, Cuevas I, et al. PAX7 expression defines germline stem cells in the adult testis. J Clin Invest 2014; 124:3929-3944. 10. Komai Y, Tanaka T, Tokuyama Y, Yanai H, Ohe S, Omachi T, Atsumi N, Yoshida N, Kumano K, Hisha H, Matsuda T, Ueno H. Bmi1 expression in long-term germ stem cells. Scientific Reports 2014; 4. 11. Gomez R, Jordan PW, Viera A, Alsheimer M, Fukuda T, Jessberger R, Llano E, Pendas AM, Handel MA, Suja JA. Dynamic localization of SMC5/6 complex proteins during mammalian meiosis and mitosis suggests functions in distinct chromosome processes. J Cell Sci 2013; 126:4239-4252. 12. Verver DE, van Pelt AM, Repping S, Hamer G. Role for rodent Smc6 in pericentromeric heterochromatin domains during spermatogonial differentiation and meiosis. Cell Death Dis 2013; 4:e749. 13. Blanco-Rodriguez J. gammaH2AX marks the main events of the spermatogenic process. Microsc Res Tech 2009; 72:823-832. 14. Rube CE, Zhang S, Miebach N, Fricke A, Rube C. Protecting the heritable genome: DNA damage response mechanisms in spermatogonial stem cells. DNA Repair (Amst) 2011; 10:159-168. 15. Hamer G, Roepers-Gajadien HL, van Duyn-Goedhart A, Gademan IS, Kal HB, van Buul PP, de Rooij DG. DNA double-strand breaks and gamma-H2AX signaling in the testis. Biol Reprod 2003; 68:628-634. 102 Chapter 4

16. Hendry JH, Adeeko A, Potten CS, Morris ID. P53 deficiency produces fewer regenerating spermatogenic tubules after irradiation. Int J Radiat Biol 1996; 70:677-682. 17. Beumer TL, Roepers-Gajadien HL, Gademan IS, van Buul PP, Gil-Gomez G, Rutgers DH, de Rooij DG. The role of the tumor suppressor p53 in spermatogenesis. Cell Death Differ 1998; 5:669- 677. 18. Hasegawa M, Zhang Y, Niibe H, Terry NH, Meistrich ML. Resistance of differentiating spermatogonia to radiation-induced apoptosis and loss in p53-deficient mice. Radiat Res 1998; 149:263-270. 19. Ishii K, Ishiai M, Morimoto H, Kanatsu-Shinohara M, Niwa O, Takata M, Shinohara T. The Trp53-Trp53inp1-Tnfrsf10b pathway regulates the radiation response of mouse spermatogonial stem cells. Stem Cell Reports 2014; 3:676-689. 20. Kanatsu-Shinohara M, Ogonuki N, Inoue K, Miki H, Ogura A, Toyokuni S, Shinohara T. Long- term proliferation in culture and germline transmission of mouse male germline stem cells. Biol Reprod 2003; 69:612-616. 21. Kanatsu-Shinohara M, Ogonuki N, Iwano T, Lee J, Kazuki Y, Inoue K, Miki H, Takehashi M, Toyokuni S, Shinkai Y, Oshimura M, Ishino F, et al. Genetic and epigenetic properties of mouse male germline stem cells during long-term culture. Development 2005; 132:4155-4163. 22. Dann CT, Alvarado AL, Molyneux LA, Denard BS, Garbers DL, Porteus MH. Spermatogonial stem cell self-renewal requires OCT4, a factor downregulated during retinoic acid-induced differentiation. Stem Cells 2008; 26:2928-2937. 23. Wang S, Wang X, Ma L, Lin X, Zhang D, Li Z, Wu Y, Zheng C, Feng X, Liao S, Feng Y, Chen J, et al. Retinoic Acid Is Sufficient for the In Vitro Induction of Mouse Spermatocytes. Stem Cell Reports 2016; 7:80-94. 24. Zheng Y, Jongejan A, Mulder CL, Mastenbroek S, Repping S, Wang Y, Li J, Hamer G. Trivial role for NSMCE2 during in vitro proliferation and differentiation of GS cells. Reproduction 2017. 25. Verver DE, Zheng Y, Speijer D, Hoebe R, Dekker HL, Repping S, Stap J, Hamer G. Non-SMC Element 2 (NSMCE2) of the SMC5/6 Complex Helps to Resolve Topological Stress. Int J Mol Sci 2016; 17. 26. Kim D, Langmead B, Salzberg SL. HISAT: a fast spliced aligner with low memory requirements. Nat Methods 2015; 12:357-360. 27. Anders S, Pyl PT, Huber W. HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics 2015; 31:166-169. 28. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 2010; 26:139-140. 29. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 2015; 43:e47. 30. Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 2009; 4:44-57. Transcriptome of irradiated spermatogonia 103

31. Huang da W, Sherman BT, Lempicki RA. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res 2009; 37:1-13. 32. Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, Lehar J, Puigserver P, Carlsson E, Ridderstrale M, Laurila E, Houstis N, Daly MJ, et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 2003; 34:267-273. 33. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Mesirov JP. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 2005; 102:15545-15550. 34. Kanatsu-Shinohara M, Miki H, Inoue K, Ogonuki N, Toyokuni S, Ogura A, Shinohara T. Long- term culture of mouse male germline stem cells under serum-or feeder-free conditions. Biol Reprod 2005; 72:985-991. 35. Zhou Q, Li Y, Nie R, Friel P, Mitchell D, Evanoff RM, Pouchnik D, Banasik B, McCarrey JR, Small C, Griswold MD. Expression of stimulated by retinoic acid gene 8 (Stra8) and maturation of murine gonocytes and spermatogonia induced by retinoic acid in vitro. Biol Reprod 2008; 78:537-545. 36. Zhou Q, Nie R, Li Y, Friel P, Mitchell D, Hess RA, Small C, Griswold MD. Expression of stimulated by retinoic acid gene 8 (Stra8) in spermatogenic cells induced by retinoic acid: an in vivo study in vitamin A-sufficient postnatal murine testes. Biol Reprod 2008; 79:35-42. 37. van Pelt AM, de Rooij DG. Retinoic acid is able to reinitiate spermatogenesis in vitamin A- deficient rats and high replicate doses support the full development of spermatogenic cells. Endocrinology 1991; 128:697-704. 38. Clevers H, Loh KM, Nusse R. Stem cell signaling. An integral program for tissue renewal and regeneration: Wnt signaling and stem cell control. Science 2014; 346:1248012. 39. Golestaneh N, Beauchamp E, Fallen S, Kokkinaki M, Uren A, Dym M. Wnt signaling promotes proliferation and stemness regulation of spermatogonial stem/progenitor cells. Reproduction 2009; 138:151-162. 40. Li GJ, Yang Y, Yang GK, Wan J, Cui DL, Ma ZH, Du LJ, Zhang GM. Slit2 suppresses endothelial cell proliferation and migration by inhibiting the VEGF-Notch signaling pathway. Mol Med Rep 2017; 15:1981-1988. 41. Gao T, McKenna B, Li C, Reichert M, Nguyen J, Singh T, Yang C, Pannikar A, Doliba N, Zhang T, Stoffers DA, Edlund H, et al. Pdx1 maintains beta cell identity and function by repressing an alpha cell program. Cell Metab 2014; 19:259-271. 42. Haupt S, Berger M, Goldberg Z, Haupt Y. Apoptosis - the p53 network. J Cell Sci 2003; 116:4077-4085. 43. Ma S, Charron J, Erikson RL. Role of Plk2 (Snk) in mouse development and cell proliferation. Mol Cell Biol 2003; 23:6936-6943. 104 Chapter 4

44. Luo X, Huang Y, Sheikh MS. Cloning and characterization of a novel gene PDRG that is differentially regulated by p53 and ultraviolet radiation. Oncogene 2003; 22:7247-7257. 45. Li H, Liu S, Yuan H, Niu Y, Fu L. Sestrin 2 induces autophagy and attenuates insulin resistance by regulating AMPK signaling in C2C12 myotubes. Exp Cell Res 2017; 354:18-24. 46. Kubota H, Avarbock MR, Brinster RL. Growth factors essential for self-renewal and expansion of mouse spermatogonial stem cells. Proc Natl Acad Sci U S A 2004; 101:16489-16494. 47. Ryu BY, Kubota H, Avarbock MR, Brinster RL. Conservation of spermatogonial stem cell self- renewal signaling between mouse and rat. Proc Natl Acad Sci U S A 2005; 102:14302-14307. 48. Carlomagno G, van Bragt MP, Korver CM, Repping S, de Rooij DG, van Pelt AM. BMP4- induced differentiation of a rat spermatogonial stem cell line causes changes in its cell adhesion properties. Biol Reprod 2010; 83:742-749. 49. Yang Y, Feng Y, Feng X, Liao S, Wang X, Gan H, Wang L, Lin X, Han C. BMP4 Cooperates with Retinoic Acid to Induce the Expression of Differentiation Markers in Cultured Mouse Spermatogonia. Stem Cells Int 2016; 2016:9536192. 50. Busada JT, Geyer CB. The Role of Retinoic Acid (RA) in Spermatogonial Differentiation. Biol Reprod 2016; 94:10. 51. Todd H, Galea GL, Meakin LB, Delisser PJ, Lanyon LE, Windahl SH, Price JS. Wnt16 Is Associated with Age-Related Bone Loss and Estrogen Withdrawal in Murine Bone. PLoS One 2015; 10:e0140260. 52. Wergedal JE, Kesavan C, Brommage R, Das S, Mohan S. Role of WNT16 in the regulation of periosteal bone formation in female mice. Endocrinology 2015; 156:1023-1032. 53. Heitzmann D, Buehler P, Schweda F, Georgieff M, Warth R, Thomas J. The in vivo respiratory phenotype of the adenosine A1 receptor knockout mouse. Respir Physiol Neurobiol 2016; 222:16-28. 54. Yang T, Gao X, Sandberg M, Zollbrecht C, Zhang XM, Hezel M, Liu M, Peleli M, Lai EY, Harris RA, Persson AE, Fredholm BB, et al. Abrogation of adenosine A1 receptor signalling improves metabolic regulation in mice by modulating oxidative stress and inflammatory responses. Diabetologia 2015; 58:1610-1620. 55. Oakberg EF. Duration of spermatogenesis in the mouse and timing of stages of the cycle of the seminiferous epithelium. Am J Anat 1956; 99:507-516. 56. Clermont Y. Quantitative analysis of spermatogenesis of the rat: a revised model for the renewal of spermatogonia. Am J Anat 1962; 111:111-129. 57. Anderson EL, Baltus AE, Roepers-Gajadien HL, Hassold TJ, de Rooij DG, van Pelt AM, Page DC. Stra8 and its inducer, retinoic acid, regulate meiotic initiation in both spermatogenesis and oogenesis in mice. Proc Natl Acad Sci U S A 2008; 105:14976-14980. 58. An J, Zheng Y, Dann CT. Mesenchymal to Epithelial Transition Mediated by CDH1 Promotes Spontaneous Reprogramming of Male Germline Stem Cells to Pluripotency. Stem Cell Reports 2017; 8:446-459. Transcriptome of irradiated spermatogonia 105

59. Tokuda M, Kadokawa Y, Kurahashi H, Marunouchi T. CDH1 is a specific marker for undifferentiated spermatogonia in mouse testes. Biol Reprod 2007; 76:130-141. 60. Kubota H, Avarbock MR, Brinster RL. Spermatogonial stem cells share some, but not all, phenotypic and functional characteristics with other stem cells. Proc Natl Acad Sci U S A 2003; 100:6487-6492. 61. Beckerman R, Prives C. Transcriptional regulation by p53. Cold Spring Harb Perspect Biol 2010; 2:a000935. 62. Kruiswijk F, Labuschagne CF, Vousden KH. p53 in survival, death and metabolic health: a lifeguard with a licence to kill. Nat Rev Mol Cell Biol 2015; 16:393-405. 63. Villegas E, Kabotyanski EB, Shore AN, Creighton CJ, Westbrook TF, Rosen JM. Plk2 regulates mitotic spindle orientation and mammary gland development. Development 2014; 141:1562- 1571. 64. Jordan PW, Karppinen J, Handel MA. Polo-like kinase is required for synaptonemal complex disassembly and phosphorylation in mouse spermatocytes. Journal Of Cell Science 2012; 125:5061- 5072. 65. Jiang L, Luo X, Shi J, Sun H, Sun Q, Sheikh MS, Huang Y. PDRG1, a novel tumor marker for multiple malignancies that is selectively regulated by genotoxic stress. Cancer Biol Ther 2011; 11:567- 573. 66. Sanli T, Linher-Melville K, Tsakiridis T, Singh G. Sestrin2 modulates AMPK subunit expression and its response to ionizing radiation in breast cancer cells. PLoS One 2012; 7:e32035. 106 Chapter 4

Supplementary data

Figure S1: Negative controls for IHC.

Tables S1-7 will be available on line upon acceptance of the paper.

Chapter 5

Spermatogonial stem cell autotransplantation and germline genomic editing: a future cure for spermatogenic failure and prevention of transmission of genomic diseases

Callista L. Mulder# Yi Zheng# Sabrina Z. Jan Robert B. Struijk Sjoerd Repping Geert Hamer Ans M.M. van Pelt

#equal contribution Human Reproduction Update 2016 Sep;22(5):561-73 108 Chapter 5

Table of Contents

Introduction Methods Clinical prospects of SSCT to restore spermatogenesis SSCT to restore fertility in adult cancer survivors SSCT to enhance fertility in oligozoospermic or azoospermic men Clinical prospects of SSCT and germline genomic editing CRISPR-Cas9 to genomically modify SSCs Curing spermatogenic failure by transplantation of genetically modified SSCs Preventing diseases in offspring by transplantation of genetically modified SSCs Epigenetic editing of SSCs Clinical and technical hurdles and drawbacks Ethical issues Concluding remarks Prospects of spermatogonial stem cell transplantation 109

Abstract

Background: Subfertility affects approximately 15% of all couples, and a severe male factor is identified in 17% of these couples. Whilst the etiology of a severe male factor remains largely unknown, prior gonadotoxic treatment and genomic aberrations have been associated with this type of subfertility. Couples with a severe male factor can resort to ICSI, with either ejaculated spermatozoa (in case of oligozoospermia) or surgically retrieved testicular spermatozoa (in case of azoospermia) to generate their own biological children. Currently there is no direct treatment for azoospermia or oligozoospermia. Spermatogonial stem cell (SSC) autotransplantation (SSCT) is a promising novel clinical application currently under development to restore fertility in sterile childhood cancer survivors. Meanwhile, recent advances in genomic editing, especially the clustered regulatory interspaced short palindromic repeats-associated protein 9 (CRISPR-Cas9) system, are likely to enable genomic rectification of human SSCs in the near future.

Objective and rationale: The objective of this review is to provide insights into the prospects of the potential clinical application of SSCT with or without genomic editing to cure spermatogenic failure and to prevent transmission of genetic diseases.

Search methods: We performed a narrative review using the literature available on PubMed not restricted to any publishing year on topics of subfertility, fertility treatments, (molecular regulation of) spermatogenesis and SSCT, inherited (genetic) disorders, prenatal screening methods, genomic editing and germline editing. For germline editing, we focussed on the novel CRISPR-Cas9 system. We included papers written in English only.

Outcomes: Current techniques allow propagation of human SSCs in vitro, which is indispensable to successful transplantation. This technique is currently being developed in a preclinical setting for childhood cancer survivors who have stored a testis biopsy prior to cancer treatment. Similarly, SSCT could be used to restore fertility in sterile adult cancer survivors. In vitro propagation of SSCs might also be employed to enhance spermatogenesis in oligozoospermic men and in azoospermic men who still have functional SSCs albeit in insufficient numbers. The combination of SSCT with genomic editing techniques could potentially rectify defects in spermatogenesis caused by genomic mutations or, more broadly, prevent transmission of genomic diseases to the offspring. In spite of the promising 110 Chapter 5 prospects, SSCT and germline genomic editing are not yet clinically applicable and both techniques require optimization at various levels.

Wider implications: SSCT with or without genomic editing could potentially be used to restore fertility in cancer survivors to treat couples with a severe male factor and to prevent the paternal transmission of diseases. This will potentially allow these couples to have their own biological children. Technical development is progressing rapidly, and ethical reflection and societal debate on the use of SSCT with or without genomic editing is pressing.

Keywords: spermatogonial stem cell autotransplantation / male infertility / male reproductive disorders / germline editing / CRISPR-Cas9 Prospects of spermatogonial stem cell transplantation 111

Introduction Subfertility, defined as failure to achieve a clinical pregnancy after at least 12 months of regular unprotected coitus [1], affects ~15% of all couples. In ~17% of these couples, a severe male factor, defined as a total motile sperm count below 3x106, is present [2]. A severe male factor may present as azoospermia (complete absence of spermatozoa in the ejaculate) or oligozoospermia (low number of spermatozoa in the ejaculate). In case of a severe male factor, a patient’s own biological children can be generated by ICSI with either ejaculated spermatozoa (in case of oligozoospermia) or surgically retrieved testicular spermatozoa by means of testicular sperm extraction (TESE) (in case of azoospermia). Currently, no direct clinical treatment exists for a severe male factor. A severe male factor is typically caused by a disturbance during spermatogenesis. Spermatogenesis occurs in the seminiferous tubules inside the testis. Essential to this process are spermatogonial stem cells (SSCs) that maintain a perfect balance between self- renewal and differentiation into mature sperm, thereby sustaining fertility throughout a man’s life. Both oligozoospermia and azoospermia can be due to a reduction in SSC numbers throughout the testis as seen in testicular biopsies taken from these men [3-5]. In some of these men, the testicular biopsies display focal spermatogenesis, where sperm is only produced in a subset of seminiferous tubules. Although the production of spermatozoa is limited due to the low number of SSCs, the existing SSCs are still functional and capable of continuous self-renewal and differentiation [6]. Whilst the etiology of a severe male factor remains largely unknown, a few genetic defects [7-9], including structural and numerical chromosomal abnormalities [10, 11] and Y‐chromosome deletions [12, 13], have been identified to associate with a severe male factor-induced subfertility. In addition, male subfertility can be attributed to previous gonadotoxic treatment [14]. Both chemotherapy and radiotherapy are known to destroy the SSC pool, resulting in oligozoospermia or azoospermia in a large proportion of cancer survivors [15]. A possible future application that could directly treat a severe male factor might be autotransplantation of SSCs. Transplantation of SSCs that are stored prior to cancer treatment is proposed as a means to restore fertility in childhood cancer survivors. This application involves transplantation of SSCs into the seminiferous tubules via the efferent duct or rete testis [16, 17]. Upon SSC transplantation (SSCT), SSCs migrate to the basement membrane of recipient seminiferous tubules, colonize the epithelium and undergo self- renewal and differentiation so that permanent spermatogenesis is established. Therefore, 112 Chapter 5

SSCT should allow natural conception without further fertility treatment, making SSCT a direct treatment for a severe male factor. In case the severe male factor is due to a genomic mutation, SSCT can only be successful if it is combined with correction of the mutation. Recent advances in genomic editing, especially those with the clustered regulatory interspaced short palindromic repeats- associated protein 9 (CRIPSR-Cas9) system, can allow for rapid, easy and highly efficient genetic alterations of a wide array of cell types and organisms including human cells [18]. Thus, if genomic editing is combined with SSCT, it would in principle allow those suffering from spermatogenic failure to have their own biological children. Furthermore, SSCT with genomic editing may be used to prevent paternal transmission of genomic diseases. In this review, the clinical prospects of SSCT with or without genomic editing for male adult cancer survivors, for men with oligozoospermia and azoospermia and for carriers of genomic diseases are discussed (Figure 1).

Methods We performed a narrative review using literature available on PubMed not restricted to any publishing year on topics of subfertility, fertility treatments, (molecular regulation of) spermatogenesis and SSCT, inherited (genetic) disorders, prenatal screening methods, genomic editing and germline editing. For germline editing, we focussed on the novel clustered regulatory interspaced short palindromic repeats-associated protein 9 (CRISPR- Cas9) system. We included papers written in English only.

Clinical prospects of SSCT to restore spermatogenesis The first successful SSCT was reported in 1994 and resulted in fertility restoration in sterile recipient mice upon transplantation of donor-derived SSCs [19]. Subsequent studies substantiated the findings by showing restoration of spermatogenesis after SSCT in mice [20-24] and other animal models [25-29], including non-human primates [30, 31]. In mice, SSCs can be transplanted into the recipient seminiferous tubules by injecting the efferent ducts, the rete testis or directly injecting seminiferous tubules [32]. Injection via the efferent ducts is mostly employed in rodents [33]. Yet, due to the anatomic and size difference in testes, ultrasound-guided transplantation via the intra-rete testis has been proved to be the least invasive and most efficient and successful protocol for non-rodent species such as pigs [26], bulls [27], primates [30, 31, 34] and ex vivo human testes [31, 35]. Crucial to this procedure is the injection of cells via the rete testis, and optimization is needed in this Prospects of spermatogonial stem cell transplantation 113 respect to augment success rates [36]. Inaddition, the quality and quantity of SSC niches in the recipient testis makes a difference to the success of autologous transplantation [37]. To date, multiple papers describe the generation of SSCT-induced healthy offspring in rodents [21-24, 38, 39] and non-rodent large animal models such as goats and sheep [40-42]. A breakthrough was recently achieved by Hermann et al. [30], who successfully performed autologous and allogeneic transplantation of SSCs from rhesus monkey, leading to the generation of donor-derived sperm in both cases. Subsequently, ICSI was conducted to fertilize oocytes, and embryos with donor paternal origin were finally produced. This demonstration in primates provides prospects for future clinical translation of SSCT. For humans, SSCT has been proposed as a future clinical application for those men with the risk of complete germ cell depletion that have no option to cryopreserve sperm, in particular, prepubertal cancer patients [17, 43-45] and, theoretically, even in case of focal spermatogenesis [46]. Prepubertal cancer patients especially rely on SSCT because spermatogenesis is not initiated yet, which means that semen cryopreservation prior to treatment is not an option. By opting for storage of a testicular biopsy before cancer therapy, SSCT can be applied later in life when the patient is cured and expresses the wish to have children [45, 47]. In this case, cryopreservation of biopsied testicular tissues constitutes an important part of fertility preservation [48]. Currently the most popular avenue to preserve pieces of testicular tissues is through controlled slow freezing [49-54], with the addition of cryoprotective agents such as dimethyl sulphoxide or ethylene glycol with or without sucrose [47, 49, 52, 53, 55-57]. Vitrification instead of controlled slow freezing has also been tested with positive outcomes [58-60]. Nevertheless, protocols for tissue preservation require further optimisation to maximize the viability of post-thawed human testicular tissues [47]. Following successful cryopreservation, conventional SSCT can be performed at a later stage, which involves the transplantation of the patient’s SSCs into one’s own testis (i.e. autotransplantation). After autotransplantation, SSCs migrate to the basement membrane of the recipient seminiferous epithelium, from where they reinitiate spermatogenesis (Figure 1). In this way, the patient can (re)gain his fertility and generate their own biological children by natural conception, as shown in animal experiments [21, 40]. The efficiency of SSCT is highly dependent on the number of transplanted SSCs [61, 62]. Previous reports suggest that SSCT is unlikely to become clinically applicable without an approach to successfully expanding human SSCs in vitro [37, 63]. In vitro expansion of SSCs has been successfully demonstrated in studies using rodent SSCs [20, 64-67]. Even after 2 years of culture, the SSCs retained the capacity to colonize the basal membrane of 114 Chapter 5 seminiferous tubules and could further develop into healthy and functional sperm [64]. In human trials, SSCs can first be isolated from biopsies and then subjected to primary culture to increase their number for the sake of SSCT. Several culture systems have been established for human adult SSCs [68-72], as well as for human prepubertal SSCs [73]. Presently (xeno)transplantation is the well-acknowledged and only available assay for functionality of human SSCs. It is shown that in vitro propagated human SSCs are capable of migrating to the niche at the basal membrane of the seminiferous tubules upon xenotransplantation into the testis of immunodeficient mice [70, 73], indicating that these cultured testicular cells still have SSC capabilities. Nevertheless, human spermatogenesis is not initiated in the mouse model. This is not unexpected given the large phylogenetic distance between mice and humans, and because of this, the murine niche cannot support full development of human SSCs into sperm [74]. Despite that, by comparing the numbers of migrated SSCs in the recipient testis after xenotransplantation of early and late passages from primary human testicular cultures, it has been established that adult as well as prepubertal SSCs can indeed proliferate in culture [70, 73]. Thus, the successful expansion of human SSCs in culture paves the way for clinical applications of SSCT. Once SSCT is clinically implemented for childhood cancer survivors, other patient groups that have the risk of becoming subfertile or that suffer from subfertility might also benefit from this treatment. In a recent study in azoospermic men, it was shown that these men hold a positive attitude toward SSCT, which was persistent even after acknowledging that a new experimental technique might have some risks for themselves or their offspring [75].

SSCT to restore fertility in adult cancer survivors Due to the high sensitivity of spermatogonia to DNA damage [14, 76], spermatogonial apoptosis and subsequent subfertility is a major side-effect of most cancer treatments. The chance of becoming azoospermic temporarily or permanently after cancer therapy is highly dependent on the type and dosage of the treatment [15]. A recent study has shown that 25% of patients who underwent chemotherapy were azoospermic after 7-218 months (median: 40 months), with the highest chance of azoospermia in Hodgkin disease survivors (63%) [77]. Cryopreservation of semen prior to treatment is historically an effective and inexpensive way to preserve fertility in adult male cancer patients. The cryopreserved semen can be used to achieve a pregnancy by cervical or intrauterine insemination (IUI) or, in case the quality of the semen is too low, by IVF or ICSI treatment. However, cancer patients often show decreased fertility at the time of cancer diagnosis. In fact, Ragni et al. [78] report that 12% of patients who wish to store their semen are azoospermic at the time of cancer Prospects of spermatogonial stem cell transplantation 115 diagnosis. Moreover, semen cryopreservation severely reduces sperm motility [79, 80], sperm count [79] and DNA integrity [81]. Hampered sperm motility significantly decreases the chance of live birth after IUI [82], and therefore the majority of patients who make use of cryopreserved spermatozoa have to resort to IVF/ICSI. However, IVF/ICSI is costly and burdensome, and requires ovarian hyperstimulation of the healthy female to retrieve oocytes, which is not risk-free and leads to onset of the ovarian hyperstimulation syndrome in 1-8% of stimulated women [83, 84]. In addition, it is well known that IVF/ICSI is associated with adverse short-term outcomes including preterm birth, lower birthweight and a higher prevalence of birth defects [85, 86]. Although SSCT does require a testis biopsy and treatment of the affected male, it would avoid IVF/ICSI treatment of the healthy female partner since natural conception might be feasible after SSCT. In the future, SSCT may therefore become an appealing alternative for fertility preservation in adult cancer patients in a similar way as described for prepubertal cancer patients by cryopreserving a testis biopsy before onset of cancer treatment.

SSCT to enhance fertility in oligozoospermic and azoospermic men Oligozoospermic and azoospermic men are capable of fertilization if they produce morphologically normal sperm in the testis. In couples where the male is oligozoospermic or azoospermic, ICSI is used to achieve fertilization with ejaculated or surgically retrieved spermatozoa, respectively. However, since oligozoospermia and azoospermia may be caused by a reduction in functional SSCs [3-5], a simple increase in SSCs may restore spermatogenesis and fertility in these men. This is especially relevant to the patients who display focal spermatogenesis at the histological level, in which some tubules show normal spermatogenesis, while others display Sertoli cell-only syndrome. The tubules with normal spermatogenesis harbour functional SSCs, and the hypothesis is that if these spermatogonia are propagated in vitro and transplanted back into the testis, they will repopulate the empty tubules and initiate spermatogenesis in these tubules. A recent article describes the characteristics of cultured SSCs deriving from patients who suffer from focal spermatogenesis due to a deletion of the azoospermia factor c (AZFc) region on the Y chromosome [46]. In vitro propagated SSCs from these men with focal spermatogenesis behaved similarly during culture and showed comparable gene expression of key spermatogonial markers when compared to SSCs originating from healthy counterparts with normal spermatogenesis. These results suggest that patients with oligozoospermia or azoospermia as a result of focal spermatogenesis might also benefit from propagation and transplantation of their own SSCs. Yet, this hypothesis needs to be demonstrated in clinical trials. However, one drawback that should be accounted for is that if a 116 Chapter 5 mutation is present on the Y chromosome, as in the case of men with AZFc deletions, male offspring of these men will harbour the same mutation and are likely to be oligozoospermic or azoospermic too [87]. It makes sense that this also holds true for mutations on other chromosomes. However, this problem also arises with other contemporary fertility treatment, such as IVF/ICSI with or without TESE. In order to prevent transmission of these genetic aberrations, additional measures have to be taken.

Prospects of spermatogonial stem cell transplantation 117

Figure 1: A schematic depiction of the proposed SSCT therapy. (A) A testicular biopsy is taken from the patient and cryopreserved. From the biopsy, SSCs are propagated in vitro, during which endogenous genomic defects may be repaired. Propagated SSCs are subsequently autotransplanted to the testis and then colonize the testis and restore spermatogenesis, enabling the patient to father a child without additional therapy. (B) The testicular histology of men with a severe male factor in different patient groups. The histology may show various phenotypes throughout the testis. For male (childhood) cancer survivors, a biopsy is cryopreserved prior to cancer therapy. Hence, thawing of the cryopreserved biopsy is indispensable to the treatment. In vitro propagation is needed for all patient groups, while genomic modification is only needed for those with a maturation arrest or carriers of diseases. In male carriers of diseases with full spermatogenesis, all germ cells including spermatids express the mutated genes, and local irradiation of the testis is required prior to transplantation to remove the mutated endogenous spermatids. After (genomically modified) SSCT, testis histology should, in theory, restore to full spermatogenesis.

Clinical prospects of SSCT and germline genomic editing While the use of SSCT for adult cancer patients or oligozoospermic and azoospermic patients that display focal spermatogenesis seems rather straightforward, azoospermic patients suffering from a maturation arrest in spermatogenesis cannot directly benefit from SSCT because transplantation of the patient’s SSCs would result in the same arrested phenotype and not cure their spermatogenic failure (Figure 1). However, in some cases, the maturation arrest may be attributed to genetic mutations or arises from epigenetic disturbances. Repair of these disorders in SSCs before SSCT would theoretically restore spermatogenesis and subsequent fertility in these patients and in addition prevent the transmission of the mutation to the offspring. The fact that human SSCs can propagate in culture for extended periods of time enables (epi)genetic editing prior to transplantation. With recent advances in the field of genomic editing, in particular the use of novel techniques such as the CRISPR-Cas9 system, SSCT with genetically modified SSCs has become feasible [88-91]. In addition, recent work has shown that epigenetic editing is also possible with CRIPSR-Cas9 [92].

CRISPR-Cas9 to genomically modify SSCs Traditional genome editing mainly relies on homologous recombination in embryonic stem cells (ESCs). Over the last decade, novel genome editing platforms such as zinc-finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs) and most recently the CRISPR-Cas9 system have been developed. These techniques are based on 118 Chapter 5 engineered nucleases that can cause a double-strand break of DNA, and are much less laborious and time-consuming compared to traditional strategies. With the emergence of these novel techniques (Table 1), the majority of cell types, including SSCs, can now be targeted.

Table 1: An overview of different genome editing techniques. Genome editing Homologous Conventional Novel Novel systems recombination engineered engineered engineered without nucleases (ZFNs nucleases nucleases engineered / TALENs) (CRISPR-Cas9) (GeCKOa) nucleases Target cells Mostly ESCs Most cell types Most cell types Most cell types Approaches to Non-viral Non-viral Non-viral Lentiviral delivering transfection / viral transfection / viral transfection / viral transduction targeting vectors transduction/ transduction / transduction / microinjection microinjection microinjection Technical High High Low Intermediate difficulty Targeting Low Variable Generally high High efficiency Low Variable Generally low Variable Off-target effects Possible to target No No No Yes a large scale of genes in parallel? Suitable for the No, due to low Not optimal Yes Currently not due clinic? efficiency and the to lentiviral typical transduction requirement of ESCs aGeCKO, genome-scale CRISPR knockout; ESC, embryonic stem cell; ZFN, zinc-finger nuclease; TALEN, transcription activator-like effector nuclease; CRISPR-Cas9, clustered regulatory interspaced short palindromic repeats-associated protein 9.

Prior to clinical application of genomically modified SSCs, the safety of the patient needs to be guaranteed. In a clinical setting, off-target effects are not acceptable, and technical simplicity would be desirable. ZFNs and TALENs were utilized to successfully manipulate mouse SSCs [90]. However, the required design and engineering of nucleases Prospects of spermatogonial stem cell transplantation 119 necessary for both ZFNs and TALENs is strenuous and of high technical difficulty. A better alternative to genetically modify SSCs seems to be the unprecedentedly simple CRISPR- Cas9 system (Figure 2), and articles describing successful manipulation of rodent SSCs by way of CRISPR are now available [88, 89, 91]. The CRISPR-Cas9 system not only bypasses the engineering of nucleases but also generates far less off-target effects compared with ZFNs [93]. According to a recent report, with genome-wide screens, no obvious off-target genetic or epigenetic changes could be detected in a large SSCT experiment involving CRISPR-Cas9-mediated gene targeting and transplantation of modified mouse SSCs [88].

Figure 2: The CRISPR-Cas9 system. The Type II Streptococcus pyogenes clustered regulatory interspaced short palindromic repeats-associated protein 9 (CRISPR-Cas9) (SpCas9) system, which is the simplest and most extensively used CRISPR-Cas9 technology, is based on a guide-RNA (gRNA) containing a specific 20 bp sequence to guide the DNA endonuclease Cas9 to a complementary target DNA sequence in the genome where it induces a DNA double-strand break (DSB). The 20-bp target genomic DNA must be upstream of a specific sequence (5’- NGG, where N represents a random nucleotide). The Cas9-induced DSB occurs ~3-bp upstream of the 5’-NGG, and can in theory be induced in any 20-bp genomic DNA sequence flanking 5’-NGG. The Cas9-induced DSB will then be repaired by either homology-directed repair (HDR), which can occur with the presence of DNA repair templates, or by non-homologous end joining (NHEJ). The error-prone NHEJ creates insertions/deletions (indels) around the DSB point. Indels, especially when occurring in early coding exons, can cause loss of gene function (gene knockout) by causing a frame shift that can lead to formation of a pre-mature stop codon. In contrast, HDR uses a template sequence for very precise repair of the DSB. Exogenous DNA repair templates (with the required sequences placed between homology arms) can be provided to the cells together with other components of the CRISPR-Cas9 system to create specific indels or modifications at target genomic loci. Thus, the CRISPR-Cas9 system can be used to insert sequences or correct disease-causing mutations in a very accurate way.

120 Chapter 5

Curing spermatogenic failure by transplantation of genetically modified SSCs A proportion of infertile men have non-obstructive azoospermia as a result of spermatogenic arrest. However, in 41% of these patients a few germ cells escape the arrest and form elongated spermatids, which can be extracted from testicular tissue by means of TESE and used for fertilisation in an ICSI procedure [94]. When no sperm is found during this procedure, no treatment options are currently available. Fortunately, Lim et al. [68] have shown that spermatogonia from patients suffering from non-obstructive azoospermia due to a maturation arrest are able to proliferate in their long-term culture system in a similar manner as men with obstructive azoospermia. Therefore, SSCs might be used in future SSCT as a valid option to treat these men. Very recently, Yuan et al. [95] employed TALENs to successfully rectify a point mutation in the mouse c-kit gene that blocks spermatogonial differentiation, and after correction spermatogenesis could be rescued, for the first time demonstrating that spermatogenic failure-related genetic defects can be corrected by genome editing platforms. Thus, it makes sense that after CRISPR-Cas9-mediated correction of the genetic defects responsible for spermatogenic failure, SSCs from azoospermic men can subsequently be used for SSCT, offering men with spermatogenic failure a patient-tailored treatment option. Besides, in small cohorts of azoospermic men, various single nucleotide polymorphisms (SNPs) associated with arrests during spermatogenesis have been identified [96-100], forming additional candidate targets for genetic modification of SSCs. A recent paper reports the use of CRISPR-Cas9 to interrogate male infertility-related SNPs in mice [101]. In addition, the recently developed genome-scale CRISPR knockout (GeCKO) enables the targeting of a variety of genes in parallel [102, 103]. Because infertility is often not believed to be a monogenic disorder but is rather thought to be caused by a spectrum of genes [104], the GeCKO system could serve as a prospective platform for gene therapy of such patients. Hence, when the infertility-causing genetic mutations are known, SSCs could first be isolated from testicular biopsies and propagated in vitro to increase their number. Subsequently, the propagated SSCs could be co-transfected with CRISPR-related vectors (to cut DNA) and exogenous DNA repair templates specific for the mutations (to induce homology-directed repair, Figure 2) in a patient-specific manner. Finally, after selection and whole genome (epi)genetic off-target analysis, the modified SSCs can be auto-transplanted into the testis to initiate spermatogenesis and produce corrected sperm.

Prospects of spermatogonial stem cell transplantation 121

Preventing diseases in offspring by transplantation of genetically modified SSCs Carriers of inherited genetic diseases, albeit often fertile, have to make important decisions when it comes to reproduction. Couples can opt to remain childless to prevent diseases in their children, opt for adoption or resort to a germ cell donor. Alternatively, these couples can attempt to have their own biological children and detect whether their prospective children are carriers of the disease via prenatal testing during pregnancy. In case the fetus is a carrier of the disease, the parents have to make the emotionally laden decision whether to terminate their pregnancy. Couples can also opt for PGD as a preventive measure for the birth of a child with a genetic defect. PGD is well established for monogenic diseases such as cystic fibrosis [105], beta thalassemia [106] and Huntington’s disease [107]. In PGD, couples undergo IVF treatment in which a single cell is aspirated from each embryo at the 6-8 cell stage to perform subsequent genetic testing for high-risk disease alleles. Only unaffected embryos are transferred to prevent the birth of children with these severe genetic diseases. Even though PGD provides a solution for couples at risk for transmitting a genetic disease, IVF treatment of the women, including ovarian hyperstimulation, is indispensable in the process of PGD. Moreover, many embryos are created, while only a few will be used to induce pregnancy. Currently there is no way to prevent genetic diseases in the offspring without creating affected embryos. Nevertheless, if the prospective father is the carrier of a disease allele, the disease-causing mutation can theoretically be corrected in isolated SSCs during in vitro culture. Subsequently, transplantation of the modified SSCs would result in genetically normal sperm and therefore prevent transfer of the disease allele to the next generations. Additionally, it would enable male carriers at risk for transmitting genetic diseases to naturally conceive a healthy child without IVF or prenatal genetic testing. Inspiringly, CRISPR-Cas9 has been shown to successfully repair mutations in disease- causing genes in different species and cell types. In mice, mutations in the Crygc gene (which causes cataracts) [108], dystrophin gene (which causes Duchenne muscular dystrophy, DMD) [109] and a Fah mutation in hepatocytes [110] have been repaired by the CRISPR-Cas9 system. In human trials, the CRISPR-Cas9 system has been used to precisely correct the hemoglobin beta and dystrophin gene, in β-thalassemia [111] and DMD patient-induced pluripotent stem cells [112], respectively. Another report describes successful repair of the cystic fibrosis transmembrane conductor receptor locus in cultured intestinal stem cells from patients with cystic fibrosis [113]. These reports raise the possibility that CRISPR-Cas9 could be used to repair inheritable mutations through the germline. 122 Chapter 5

Interestingly, a recent article reports successful genome editing of mouse SSCs with the CRISPR-Cas9 system [88]. Transplantation of the genetically modified SSCs led to fertile offspring, in which a Crygc mutation causing cataracts was corrected. To our knowledge, this is the first report that describes CRISPR-Cas9-mediated genome editing in SSCs in combination with SSCT, thereby preventing diseases in the offspring. Furthermore, as the transplanted SSCs were the cell lines derived from isolated and corrected single cells, this method can generate healthy descendants at 100% efficiency, thereby averting the problem of mosaicism. In addition to this pioneering work, two other recent articles also provide the proof of concept by showing CRISPR-Cas9 and SSCT-induced germline transmission in rodents [89, 91]. Hence, SSCT and the CRISPR-Cas9 system can be well combined in the future to prevent the transmission of inheritable diseases to the offspring.

Epigenetic editing of SSCs Mechanisms that underlie cellular functioning are orchestrated by different layers of transcriptional regulation. Apart from genetic factors, epigenetic regulation is key for the proper functioning of a cell. Epigenetic traits are defined [114] as ‘heritable phenotypes resulting from changes in a chromosome without alterations in the DNA sequence’, mediated by several factors such as DNA methylation, histone modifications and higher order chromatin structuring. Disruption of epigenetic regulation has been shown in an array of complex diseases, including cancer, diabetes and cardiovascular diseases (reviewed in [115- 117]). Thus, in theory, epigenetic editing may be a valid alternative for genetic repair to modify aberrant gene expression in SSCs from those at risk of transmitting diseases to their own biological children. The core of currently described epigenetic editing approaches is the fusion of an epigenetic modulatory enzyme to a protein with a DNA binding domain in order to affect gene expression and modulate local parameters, such as cytosine methylation and demethylation or histone modification once the enzymatic construct is in place. This powerful approach has been used to silence and/or activate specific DNA sequences by altering epigenetic parameters of target genes (reviewed in [118], [119]). For example, a recent study describes the guidance of the Ten-Eleven Translocation 2 DNA demethylation enzyme to the promoter region of the intercellular adhesion molecule 1 gene, causing local demethylation and reactivation of the gene where it was normally silenced [120]. Another group has reported the development of a programmable molecular construct consisting of a nuclease-deactivated Cas9 (dCas9) protein fused to the catalytic core of the acetyltransferase p300, which can be used to modulate histone acetylation of any Cas9-targetable genomic location [92]. One Prospects of spermatogonial stem cell transplantation 123 might imagine that epigenetic editing techniques might allow correction of the transcriptional regulation of pivotal genes in various biological processes, including germ cell development. Epigenetics have been shown to play a key role in normal germ cell development [121-123], and allele-specific DNA methylation was altered in semen from men that suffer from spermatogenic failure [124-128]. Local correction of abnormal DNA methylation or histone modification of target genes in infertile men might improve the spermatogenic potential. In theory, SSCT of epigenetically modified SSCs may also be applicable for inherited epigenetic diseases. Epidemiological data point to human transgenerational epigenetic inheritance, including the Dutch Famine Birth Cohort Study [129-132] and the Swedish Overkalix population [133]. Additionally, a few (case) studies describe inheritance of a disease-associated epimutation of a specific locus, such as the SNURF-SNRPN locus in Prader-Willi and Angelman syndrome [134] and the cancer predisposing gene MLH1 [135]. However, one must realize that the epigenome is reset in an extensive way during early embryonic development (reviewed in [136] and [137]). Even though some inherited epigenetic marks seem to escape epigenetic reprogramming, it remains unclear whether epigenetic germline editing combined with SSCT may benefit patients in the future. Therefore, more research is needed before SSCT can be applied to cure heritable epigenetic diseases.

Clinical and technical hurdles and drawbacks The field of SSCT is uprising, and a variety of patient groups may benefit from this therapy in the future. However, some hurdles still need to be overcome prior to its clinical implementation. For one thing, safety of the patient and his offspring is of major concern [45]. Human SSCs may change (epi)genetically when exposed to an in vitro environment. Yet, there is evidence of genetic stability of cultured human SSCs [138]. DNA methylation of maternal and paternal imprinted genes in uncultured murine SSCs did not alter after transplantation [21, 23], whereas cultured human SSCs showed changes in DNA methylation in some selected regions of maternal and paternal imprinted genes [138]. Conclusively, although the published data suggest that SSCT may be safe for the clinic, more (pre)clinical studies are needed in this field to ensure safety for patients, as well as for their offspring. Another challenge in cancer patients is that primary testicular cultures may be contaminated with lingering cancer cells from leukemic or metastasized patients. While some researchers were unable to successfully sort out cancer cells [139], others succeeded in removing leukemic cells from testicular cultures [140] or from cell suspensions [141]. 124 Chapter 5

In comparison, the clinical implementation of genomically modified SSCT is even more challenging. Aside from the potential risks accompanying conventional SSCT, we currently do not know whether genomic manipulation of SSCs harbours any extra risks for the patient or his offspring. In addition, a major clinical drawback of germline therapy in fertile carriers of diseases remains that the patients have to undergo local irradiation to deplete the testis of endogenous mutated spermatogonia before SSCT. Otherwise, the testis of the recipient father would produce two populations of sperm cells: those that arise from endogenous SSCs carrying the disease-causing mutation as well as those from the corrected SSCs introduced by transplantation. As a consequence, the semen of the father would contain a mixed population of spermatozoa and children conceived by natural conception could be derived from either a corrected or endogenous sperm cell. Although local irradiation has been demonstrated to be an effective measure to deplete the testis of germ cells in animal studies [41, 142], it can have a deleterious influence on outgrowth of seminiferous tubules, especially in prepubertal testes [36]. Besides, it may cause damages to surrounding organs and cells. Moreover, some endogenous spermatogonia might survive the irradiation and are still capable of developing into spermatids, thereby risking the transfer of the genomic aberrations. In this sense, the development of alternatives to exclude endogenous SSCs is needed to better strike the balance between the benefits of SSCT and the potential risks of the required total depletion of endogenous spermatogenesis. In terms of genomic manipulation, some technical problems remain to be addressed. First, the CRISPR-Cas9 components need to be delivered into cells. The way in which the Cas9 nuclease is introduced (viral or non-viral) has significant clinical implications (Table 1). In case of viral transduction, CRISPR-sequences, and possibly even residual viral sequences, integrate into the genome of the patient. This raises serious safety concerns and is not suitable for clinical application. However, non-viral delivery methods often fall short, as they can be inefficient in gene delivery (e.g. liposome-mediated transfection), are laborious (e.g. microinjection) or lead to significant cell mortality (e.g. electroporation). While SSCs have been demonstrated to be refractory to most non-viral transfection approaches [143], novel electroporation devices that are currently being used in some laboratories may be the option to transfect SSCs with adequate efficiency [88-91, 144]. Alternatively, transfection of the mRNA instead of the corresponding DNA vectors has been shown to be more efficient for genome editing [90]. Also the recently developed novel method regarding direct intracellular delivery of proteins might serve as another option for gene targeting [145]. Prospects of spermatogonial stem cell transplantation 125

In spite of the developments, one needs to realize that SSCT in combination with genomic editing to cure or prevent diseases is only feasible when the genetic mutation is known. Unfortunately, the genetic etiology remains elusive in many cases. More research on the identification of genomic mutations that cause subfertility is necessary in this respect [7- 9]. A recent article describes that CRISPR-Cas9 was exploited to successfully eradicate porcine endogenous retroviruses with the copy number as high as 62 in a porcine cell line [146], showing the robustness and versatility of CRISPR. However, various genetic diseases, including subfertility, are believed to be caused by an array of genes [104], which renders genetic correction substantially more difficult as it would require simultaneous targeting of various loci. A previous report shows that a maximum of five genes could be simultaneously disrupted by microinjection of CRISPR components into mouse ESCs [147]. At present, the novel GeCKO system seems to be the only possible approach to targeting a wide array of genes in parallel. Nevertheless, GeCKO, which requires lentiviral transduction and subsequent integration of CRISPR components into the genome, is considered unsafe for clinical application at the moment. We still need to await further development in this regard before we can target a variety of genes in parallel safely for clinical purposes. Another important issue is the potential off-target alterations induced by the CRISPR- Cas9 system. Recent studies have revealed that the 20-bp gRNA-DNA hybrid (Figure 2) has the potential to tolerate 1-3 or even more sequence mismatches [18, 148, 149]. As a consequence, normal genes containing high homology to the target sequence might also be targeted. Reassuringly, whole genome sequencing of the CRISPR-Cas9-modified SSCs showed no apparent off-target mutations [88]. Moreover, novel versions of CRISPR with enhanced specificity but without the sacrifice of on-target activity have been developed recently [150, 151], which will further facilitate its broad applications in the clinic. Also, the targeting spectrum of the CRISPR-Cas9 system needs to be expanded. The requirement of a specific sequence (e.g. 5’-NGG for type II SpCas9, Figure 2) following the target is a principal constraint. While 5’-NGG occurs quite frequently in the human genome, further extensions of the targeting range by development of other types of CRISPR that recognize distinct sequences downstream of the targeting site [152, 153] would give broader options for genome editing.

Ethical issues As genomic modification of SSCs leads to germline transmission of the modified trait to the next generation, elaborate ethical reflection and an intensive societal debate on the 126 Chapter 5 acceptability of a clinical application of germline gene editing should precede the actual clinical application of modified SSCs. Recently two groups employed the CRISPR-Cas9 system to achieve genome editing in human tripronuclear zygotes [154, 155]. Although the zygotes used were unable to develop into viable embryos, these reports still initiated major debates. Some people propose a complete ban on germline genomic editing [156], while others request a moratorium on the clinical application of germline editing but suggest permission for research in this field [157]. To date, multiple articles have been published to broaden the discussion [158-162]. We strongly support the establishment of a societal platform including molecular and (stem) cell biologists, medical professionals, ethicists, politicians, citizens and most importantly patients, to discuss under which circumstances and to what extent germline modification should be allowed.

Concluding remarks In this review, we explore different patient groups that may benefit from SSCT with or without genomic editing. We conclude that the clinical implementation of SSCT can potentially reach far beyond fertility preservation in childhood cancer patients. Conventional SSCT could help adult cancer patients preserve their fertility, while fertility may also be enhanced in oligozoospermic or azoospermic patients using this technology. Successful genomic modification of SSCs by the novel CRISPR-Cas9 system in culture could repair detrimental mutations, thereby treating patients with non-obstructive azoospermia and carriers of diseases in the future. The safety of the patient and the following generations is of paramount importance. From our perspective, by no means should these techniques be employed in the clinic until safety and efficiency has been demonstrated empirically. Therefore, more fundamental research remains to be conducted. In addition, a societal and ethical debate should precede the use of modified SSCs in a future clinical application of SSCT. Nonetheless, SSCT, with or without genome editing, is a potential powerful platform that potentially can be employed to cure infertility or even inheritable mutations in various patient groups. Research in this field is thriving, and a revolution might be visible at the horizon.

Authors’ roles C.L.M., Y.Z., G.H. and A.M.M.P. designed the outline of the review. C.L.M. and Y.Z. drafted the original manuscript. S.Z.J. and R.B.S. gave substantial contribution to the manuscript. C.L.M., Y.Z. and S.Z.J. designed and created the figures and table in this Prospects of spermatogonial stem cell transplantation 127 manuscript. S.R., A.M.M.P. and G.H. critically reviewed and revised the manuscript and approved the final version.

Funding ZonMW (TAS116003002), the China Scholarship Counsel (CSC) (201306300081), an AMC Fellowship and the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (CIG 293765).

Conflict of interest The author(s) report no financial or other conflict of interest relevant to the subject of this article. 128 Chapter 5

References

1. Zegers-Hochschild F, Adamson GD, de Mouzon J, Ishihara O, Mansour R, Nygren K, Sullivan E, Vanderpoel S, International Committee for Monitoring Assisted Reproductive T, World Health O. International Committee for Monitoring Assisted Reproductive Technology (ICMART) and the World Health Organization (WHO) revised glossary of ART terminology, 2009. Fertil Steril 2009; 92:1520- 1524. 2. van der Steeg JW, Steures P, Eijkemans MJ, Habbema JD, Hompes PG, Broekmans FJ, van Dessel HJ, Bossuyt PM, van der Veen F, Mol BW, group Cs. Pregnancy is predictable: a large-scale prospective external validation of the prediction of spontaneous pregnancy in subfertile couples. Hum Reprod 2007; 22:536-542. 3. Hentrich A, Wolter M, Szardening-Kirchner C, Lüers G, Bergmann M, Kliesch S, Konrad L. Reduced numbers of Sertoli, germ, and spermatogonial stem cells in impaired spermatogenesis. Mod Pathol. 2011; 24:1380-1389. 4. Yakirevich E, Sabo E, Dirnfeld M, Sova Y, Spagnoli G, Resnick M. Morphometrical quantification of spermatogonial germ cells with the 57B anti-MAGE-A4 antibody in the evaluation of testicular biopsies for azoospermia. Appl Immunohistochem Mol Morphol. 2003; 11:37-44. 5. Takagi S, Itoh N, Kimura M, Sasao T, Tsukamoto T. Spermatogonial proliferation and apoptosis in hypospermatogenesis associated with nonobstructive azoospermia. Fertil Steril. 2001; 76:901-907. 6. Silber SJ. Evaluation and treatment of male infertility. Clin Obstet Gynecol 2000; 43:854-888. 7. Visser L, Repping S. Unravelling the genetics of spermatogenic failure. Reproduction 2010; 139:303-307. 8. Tuttelmann F, Simoni M, Kliesch S, Ledig S, Dworniczak B, Wieacker P, Ropke A. Copy number variants in patients with severe oligozoospermia and Sertoli-cell-only syndrome. PLoS One 2011; 6:e19426. 9. Krausz C, Chianese C. Genetic testing and counselling for male infertility. Curr Opin Endocrinol Diabetes Obes 2014; 21:244-250. 10. de Kretser DM. Male infertility. Lancet 1997; 349:787-790. 11. Oates RD. The genetic basis of male reproductive failure. Urol Clin North Am 2008; 35:257- 270, ix. 12. Kuroda-Kawaguchi T, Skaletsky H, Brown LG, Minx PJ, Cordum HS, Waterston RH, Wilson RK, Silber S, Oates R, Rozen S, Page DC. The AZFc region of the Y chromosome features massive palindromes and uniform recurrent deletions in infertile men. Nat Genet 2001; 29:279-286. 13. Noordam MJ, Repping S. The human Y chromosome: a masculine chromosome. Current Opinion in Genetics & Development 2006; 16:225-232. 14. Meistrich M. Effects of chemotherapy and radiotherapy on spermatogenesis in humans. Fertil Steril. 2013; 100:1180-1186. Prospects of spermatogonial stem cell transplantation 129

15. Howell S, Shalet S. Spermatogenesis After Cancer Treatment: Damage and Recovery Journal of the National Cancer Institute Monographs 2005; 34:12-17. 16. Dores C, Alpaugh W, Dobrinski I. From in vitro culture to in vivo models to study testis development and spermatogenesis. Cell Tissue Res 2012; 349:691-702. 17. Brinster RL. Male germline stem cells: from mice to men. Science 2007; 316:404-405. 18. Sander JD, Joung JK. CRISPR-Cas systems for editing, regulating and targeting genomes. Nat Biotechnol 2014; 32:347-355. 19. Brinster RL, Zimmermann JW. Spermatogenesis following male germ-cell transplantation. Proc Natl Acad Sci U S A. 1994; 91:11298-11302. 20. Kanatsu-Shinohara M, Ogonuki N, Inoue K, Miki H, Ogura A, Toyokuni S, Shinohara T. Long- term proliferation in culture and germline transmission of mouse male germline stem cells. Biol Reprod 2003; 69:612-616. 21. Goossens E, De Rycke M, Haentjens P, Tournaye H. DNA methylation patterns of spermatozoa and two generations of offspring obtained after murine spermatogonial stem cell transplantation. Hum Reprod 2009; 24:2255-2263. 22. Kubota H, Avarbock MR, Schmidt JA, Brinster RL. Spermatogonial stem cells derived from infertile Wv/Wv mice self-renew in vitro and generate progeny following transplantation. Biol Reprod 2009; 81:293-301. 23. Wu X, Goodyear SM, Abramowitz LK, Bartolomei MS, Tobias JW, Avarbock MR, Brinster RL. Fertile offspring derived from mouse spermatogonial stem cells cryopreserved for more than 14 years. Hum Reprod 2012; 27:1249-1259. 24. Yuan Z, Hou R, Wu J. Generation of mice by transplantation of an adult spermatogonial cell line after cryopreservation. Cell Prolif 2009; 42:123-131. 25. Ryu BY, Orwig KE, Oatley JM, Lin CC, Chang LJ, Avarbock MR, Brinster RL. Efficient generation of transgenic rats through the male germline using lentiviral transduction and transplantation of spermatogonial stem cells. J Androl 2007; 28:353-360. 26. Honaramooz A, Megee SO, Dobrinski I. Germ cell transplantation in pigs. Biol Reprod 2002; 66:21-28. 27. Izadyar F, Den Ouden K, Stout TA, Stout J, Coret J, Lankveld DP, Spoormakers TJ, Colenbrander B, Oldenbroek JK, Van der Ploeg KD, Woelders H, Kal HB, et al. Autologous and homologous transplantation of bovine spermatogonial stem cells. Reproduction 2003; 126:765-774. 28. Kawasaki T, Saito K, Sakai C, Shinya M, Sakai N. Production of zebrafish offspring from cultured spermatogonial stem cells. Genes Cells 2012; 17:316-325. 29. Nobrega RH, Greebe CD, van de Kant H, Bogerd J, de Franca LR, Schulz RW. Spermatogonial stem cell niche and spermatogonial stem cell transplantation in zebrafish. PLoS One 2010; 5. 30. Hermann BP, Sukhwani M, Winkler F, Pascarella JN, Peters KA, Sheng Y, Valli H, Rodriguez M, Ezzelarab M, Dargo G, Peterson K, Masterson K, et al. Spermatogonial stem cell transplantation into rhesus testes regenerates spermatogenesis producing functional sperm. Cell Stem Cell 2012; 11:715-726. 130 Chapter 5

31. Schlatt S, Rosiepen G, Weinbauer GF, Rolf C, Brook PF, Nieschlag E. Germ cell transfer into rat, bovine, monkey and human testes. Hum Reprod 1999; 14:144-150. 32. Ogawa T, Arechaga JM, Avarbock MR, Brinster RL. Transplantation of testis germinal cells into mouse seminiferous tubules. Int J Dev Biol 1997; 41:111-122. 33. Gonzalez R, Dobrinski I. Beyond the mouse monopoly: studying the male germ line in domestic animal models. ILAR J 2015; 56:83-98. 34. Schlatt S. Spermatogonial stem cell preservation and transplantation. Mol Cell Endocrinol 2002; 187:107-111. 35. Ning L, Meng J, Goossens E, Lahoutte T, Marichal M, Tournaye H. In search of an efficient injection technique for future clinical application of spermatogonial stem cell transplantation: infusion of contrast dyes in isolated cadaveric human testes. Fertil Steril. 2012; 98:1443-1448. 36. Jahnukainen K, Ehmcke J, Quader MA, Saiful Huq M, Epperly MW, Hergenrother S, Nurmio M, Schlatt S. Testicular recovery after irradiation differs in prepubertal and pubertal non-human primates, and can be enhanced by autologous germ cell transplantation. Hum Reprod 2011; 26:1945- 1954. 37. Jahnukainen K, Ehmcke J, Soder O, Schlatt S. Clinical potential and putative risks of fertility preservation in children utilizing gonadal tissue or germline stem cells. Pediatr Res 2006; 59:40R-47R. 38. Goossens E, de Vos P, Tournaye H. Array comparative genomic hybridization analysis does not show genetic alterations in spermatozoa and offspring generated after spermatogonial stem cell transplantation in the mouse. Hum Reprod 2010; 25:1836-1842. 39. Lee J, Kanatsu-Shinohara M, Ogonuki N, Miki H, Inoue K, Morimoto T, Morimoto H, Ogura A, Shinohara T. Heritable imprinting defect caused by epigenetic abnormalities in mouse spermatogonial stem cells. Biol Reprod 2009; 80:518-527. 40. Honaramooz A, Behboodi E, Megee SO, Overton SA, Galantino-Homer H, Echelard Y, Dobrinski I. Fertility and germline transmission of donor haplotype following germ cell transplantation in immunocompetent goats. Biology of Reproduction 2003; 69:1260-1264. 41. Herrid M, Olejnik J, Jackson M, Suchowerska N, Stockwell S, Davey R, Hutton K, Hope S, Hill JR. Irradiation enhances the efficiency of testicular germ cell transplantation in sheep. Biol Reprod 2009; 81:898-905. 42. Zheng Y, Zhang Y, Qu R, He Y, Tian X, Zeng W. Spermatogonial stem cells from domestic animals: progress and prospects. Reproduction 2014; 147:65-74. 43. Goossens E, Van Saen D, Tournaye H. Spermatogonial stem cell preservation and transplantation: from research to clinic. Hum Reprod. 2013 Apr;28(4):897-907. 2013; 28:897-907. 44. Sadri-Ardekani H, Atala A. Testicular tissue cryopreservation and spermatogonial stem cell transplantation to restore fertility: from bench to bedside. Stem Cell Res Ther. 2014; 5:68. 45. Struijk R, Mulder C, Van der Veen F, van Pelt A, S. R. Restoring fertility in sterile childhood cancer survivors by autotransplanting spermatogonial stem cells: are we there yet? Biomed Res Int. 2013;2013:903142. 2013; 903142. Prospects of spermatogonial stem cell transplantation 131

46. Nickkholgh B, Korver C, van Daalen S, van Pelt A, Repping S. AZFc deletions do not affect the function of human spermatogonia in vitro Molecular Human Reproduction 2015. 47. Picton HM, Wyns C, Anderson RA, Goossens E, Jahnukainen K, Kliesch S, Mitchell RT, Pennings G, Rives N, Tournaye H, van Pelt AM, Eichenlaub-Ritter U, et al. A European perspective on testicular tissue cryopreservation for fertility preservation in prepubertal and adolescent boys. Hum Reprod 2015; 30:2463-2475. 48. Schlatt S, Ehmcke J, Jahnukainen K. Testicular stem cells for fertility preservation: preclinical studies on male germ cell transplantation and testicular grafting. Pediatr Blood Cancer 2009; 53:274- 280. 49. Keros V, Hultenby K, Borgstrom B, Fridstrom M, Jahnukainen K, Hovatta O. Methods of cryopreservation of testicular tissue with viable spermatogonia in pre-pubertal boys undergoing gonadotoxic cancer treatment. Hum Reprod 2007; 22:1384-1395. 50. Ginsberg JP, Carlson CA, Lin K, Hobbie WL, Wigo E, Wu X, Brinster RL, Kolon TF. An experimental protocol for fertility preservation in prepubertal boys recently diagnosed with cancer: a report of acceptability and safety. Hum Reprod 2010; 25:37-41. 51. Wyns C, Curaba M, Petit S, Vanabelle B, Laurent P, Wese JF, Donnez J. Management of fertility preservation in prepubertal patients: 5 years' experience at the Catholic University of Louvain. Hum Reprod 2011; 26:737-747. 52. Wyns C, Curaba M, Martinez-Madrid B, Van Langendonckt A, Francois-Xavier W, Donnez J. Spermatogonial survival after cryopreservation and short-term orthotopic immature human cryptorchid testicular tissue grafting to immunodeficient mice. Hum Reprod 2007; 22:1603-1611. 53. Keros V, Rosenlund B, Hultenby K, Aghajanova L, Levkov L, Hovatta O. Optimizing cryopreservation of human testicular tissue: comparison of protocols with glycerol, propanediol and dimethylsulphoxide as cryoprotectants. Hum Reprod. 2005; 20:1676-1687. 54. Gassei K, Orwig KE. Experimental methods to preserve male fertility and treat male factor infertility. Fertil Steril 2016; 105:256-266. 55. Kvist K, Thorup J, Byskov AG, Hoyer PE, Mollgard K, Yding Andersen C. Cryopreservation of intact testicular tissue from boys with cryptorchidism. Hum Reprod 2006; 21:484-491. 56. Wyns C, Van Langendonckt A, Wese FX, Donnez J, Curaba M. Long-term spermatogonial survival in cryopreserved and xenografted immature human testicular tissue. Hum Reprod 2008; 23:2402-2414. 57. Poels J, Abou-Ghannam G, Herman S, Van Langendonckt A, Wese FX, Wyns C. In Search of Better Spermatogonial Preservation by Supplementation of Cryopreserved Human Immature Testicular Tissue Xenografts with N-acetylcysteine and Testosterone. Front Surg 2014; 1:47. 58. Curaba M, Poels J, van Langendonckt A, Donnez J, Wyns C. Can prepubertal human testicular tissue be cryopreserved by vitrification? Fertil Steril 2011; 95:2123 e2129-2112. 59. Baert Y, Van Saen D, Haentjens P, In't Veld P, Tournaye H, Goossens E. What is the best cryopreservation protocol for human testicular tissue banking? Hum Reprod 2013; 28:1816-1826. 132 Chapter 5

60. Poels J, Van Langendonckt A, Many MC, Wese FX, Wyns C. Vitrification preserves proliferation capacity in human spermatogonia. Hum Reprod 2013; 28:578-589. 61. Nagano M. Homing Efficiency and Proliferation Kinetics of Male Germ Line Stem Cells Following Transplantation in Mice. Biol Reprod 2003; 69:701-707 62. Dobrinski I, Ogawa T, Avarbock M, RL. B. Computer assisted image analysis to assess colonization of recipient seminiferous tubules by spermatogonial stem cells from transgenic donor mice. Mol Reprod Dev. 1999; 53:142-148. 63. Jahnukainen K, Ehmcke J, Hou M, Schlatt S. Testicular function and fertility preservation in male cancer patients. Best Pract Res Clin Endocrinol Metab 2011; 25:287-302. 64. Kanatsu-Shinohara M, Miki H, Inoue K, Ogonuki N, Toyokuni S, Ogura A, Shinohara T. Long- term culture of mouse male germline stem cells under serum-or feeder-free conditions. Biol Reprod. 2005; 72:985-991. 65. Kubota H, Avarbock MR, Brinster RL. Growth factors essential for self-renewal and expansion of mouse spermatogonial stem cells. Proc Natl Acad Sci U S A 2004; 101:16489-16494. 66. Kanatsu-Shinohara M, Muneto T, Lee J, Takenaka M, Chuma S, Nakatsuji N, Horiuchi T, Shinohara T. Long-term culture of male germline stem cells from hamster testes. Biol Reprod. 2008; 78:611-617. 67. Ryu BY, Kubota H, Avarbock MR, Brinster RL. Conservation of spermatogonial stem cell self- renewal signaling between mouse and rat. Proc Natl Acad Sci U S A 2005; 102:14302-14307. 68. Lim J, Sung S, Kim H, Song S, Hong J, Yoon T, Kim J, Kim K, Lee D. Long-term proliferation and characterization of human spermatogonial stem cells obtained from obstructive and non- obstructive azoospermia under exogenous feeder-free culture conditions. Cell Prolif. 2010; 43:405-417. 69. Kossack N, Terwort N, Wistuba J, Ehmcke J, Schlatt S, Scholer H, Kliesch S, Gromoll J. A combined approach facilitates the reliable detection of human spermatogonia in vitro. Hum Reprod 2013; 28:3012-3025. 70. Sadri-Ardekani H, Mizrak SC, van Daalen SK, Korver CM, Roepers-Gajadien HL, Koruji M, Hovingh S, de Reijke TM, de la Rosette JJ, van der Veen F, de Rooij DG, Repping S, et al. Propagation of human spermatogonial stem cells in vitro. JAMA 2009; 302:2127-2134. 71. Akhondi M, Mohazzab A, Jeddi-Tehran iM, Sadeghi M, Eidi A, Khodadadi A, Piravar Z. Propagation of human germ stem cells in long-term culture. Iran J Reprod Med. 2013; 11:551-558. 72. Guo Y, Liu L, Sun M, Hai Y, Li Z, He Z. Expansion and long-term culture of human spermatogonial stem cells via the activation of SMAD3 and AKT pathways. Exp Biol Med (Maywood) 2015; 240:1112-1122. 73. Sadri-Ardekani H, Akhondi MA, van der Veen F, Repping S, van Pelt AM. In vitro propagation of human prepubertal spermatogonial stem cells. JAMA 2011; 305:2416-2418. 74. Nagano M, Patrizio P, Brinster RL. Long-term survival of human spermatogonial stem cells in mouse testes. Fertil Steril 2002; 78:1225-1233. Prospects of spermatogonial stem cell transplantation 133

75. Hendriks S, Dancet EA, Meissner A, Van der Veen F, Mochtar M, Repping S. Perspectives of infertile men on future stem cell treatments for nonobstructive azoospermia. Reprod Biomed Online. 2014; 28:650-657. 76. van der Meer Y, Huiskamp R, Davids J, Van der Tweel I, De Rooij D. The sensitivity to X rays of mouse spermatogonia that are committed to differentiate and of differentiating spermatogonia. Radiat Res. 1992; 130:296-302. 77. Tomlinson M, Meadows J, Kohut T, Haoula Z, Naeem A, Pooley K, Deb S. Review and follow- up of patients using a regional sperm cryopreservation service: ensuring that resources are targeted to those patients most in need. Andrology 2015; 3:709-716. 78. Ragni G, Somigliana E, Restelli L, Salvi R, Arnoldi M, Paffoni A. Sperm banking and rate of assisted reproduction treatment: insights from a 15-year cryopreservation program for male cancer patients. Cancer 2003; 97:1624-1629. 79. Keel B, Webster B. Semen analysis data from fresh and cryopreserved donor ejaculates: comparison of cryoprotectants and pregnancy rates. Fertil Steril. 1989; 52:100-105. 80. Boitrelle F, Albert M, Theillac C, Ferfouri F, Bergere M, Vialard F, Wainer R, Bailly M, Selva J. Cryopreservation of human spermatozoa decreases the number of motile normal spermatozoa, induces nuclear vacuolization and chromatin decondensation. J Androl. 2012; 33:1371-1378. 81. Valcarce D, Cartón-García F, Riesco M, Herráez M, Robles V. Analysis of DNA damage after human sperm cryopreservation in genes crucial for fertilization and early embryo development. Andrology 2013; 1:723-730. 82. Hendin B, Falcone T, Hallak J, Nelson D, Vemullapalli S, Goldberg J, Thomas AJ, Agarwal A. The effect of patient and semen characteristics on live birth rates following intrauterine insemination: a retrospective study. J Assist Reprod Genet. 2000; 17:245-252. 83. Mathur R, Akande A, Keay S, Hunt L, Jenkins J. Distinction between early and late ovarian hyperstimulation syndrome. Fertil Steril. 2000; 73:901-907. 84. Gelbaya TA. Short and long-term risks to women who conceive through in vitro fertilization. Human Fertility 2010; 13:19-27. 85. Hansen M, Kurinczuk JJ, Milne E, de Klerk N, Bower C. Assisted reproductive technology and birth defects: a systematic review and meta-analysis. Hum Reprod Update 2013; 19:330-353. 86. Helmerhorst FM, Perquin DA, Donker D, Keirse MJ. Perinatal outcome of singletons and twins after assisted conception: a systematic review of controlled studies. BMJ 2004; 328:261. 87. Page D, Silber S, Brown L. Men with infertility caused by AZFc deletion can produce sons by intracytoplasmic sperm injection, but are likely to transmit the deletion and infertility. Hum Reprod. 1999; 14:1722-1726. 88. Wu Y, Zhou H, Fan X, Zhang Y, Zhang M, Wang Y, Xie Z, Bai M, Yin Q, Liang D, Tang W, Liao J, et al. Correction of a genetic disease by CRISPR-Cas9-mediated gene editing in mouse spermatogonial stem cells. Cell Res 2015; 25:67-79. 134 Chapter 5

89. Chapman KM, Medrano GA, Jaichander P, Chaudhary J, Waits AE, Nobrega MA, Hotaling JM, Ober C, Hamra FK. Targeted Germline Modifications in Rats Using CRISPR/Cas9 and Spermatogonial Stem Cells. Cell Rep 2015; 10:1828-1835. 90. Fanslow DA, Wirt SE, Barker JC, Connelly JP, Porteus MH, Dann CT. Genome editing in mouse spermatogonial stem/progenitor cells using engineered nucleases. PLoS One 2014; 9:e112652. 91. Sato T, Sakuma T, Yokonishi T, Katagiri K, Kamimura S, Ogonuki N, Ogura A, Yamamoto T, Ogawa T. Genome Editing in Mouse Spermatogonial Stem Cell Lines Using TALEN and Double- Nicking CRISPR/Cas9. Stem Cell Reports 2015; 5:75-82. 92. Hilton IB, D'Ippolito AM, Vockley CM, Thakore PI, Crawford GE, Reddy TE, Gersbach CA. Epigenome editing by a CRISPR-Cas9-based acetyltransferase activates genes from promoters and enhancers. Nat Biotechnol 2015. 93. Ul Ain Q, Chung JY, Kim YH. Current and future delivery systems for engineered nucleases: ZFN, TALEN and RGEN. J Control Release 2015; 205:120-127. 94. Vloeberghs V, Verheyen G, Haentjens P, Goossens A, Polyzos NP, Tournaye H. How successful is TESE-ICSI in couples with non-obstructive azoospermia? Hum Reprod 2015; 30:1790- 1796. 95. Yuan Y, Zhou Q, Wan H, Shen B, Wang X, Wang M, Feng C, Xie M, Gu T, Zhou T, Fu R, Huang X, et al. Generation of fertile offspring from Kit(w)/Kit(wv) mice through differentiation of gene corrected nuclear transfer embryonic stem cells. Cell Res 2015; 25:851-863. 96. Aston K, Krausz C, Laface I, Ruiz-Castané E, Carrell D. Evaluation of 172 candidate polymorphisms for association with oligozoospermia or azoospermia in a large cohort of men of European descent. Hum Reprod. 2010; 25:1383-1397. 97. Teng Y, Chang Y, J T, Kuo P, Lee I, Lee M, Kuo P. A single-nucleotide polymorphism of the DAZL gene promoter confers susceptibility to spermatogenic failure in the Taiwanese Han. Hum Reprod. 2012; 27:2857-2865. 98. Parada-Bustamante A, Lardone M, R V, Ebensperger M, López P, Madariaga M, Piottante A, A. C. Analysis of 6 single-nucleotide polymorphisms in the androgen receptor gene in Chilean patients with primary spermatogenic failure. J Androl. 2012; 33:88-95. 99. He X, Ruan J, Du W, Chen G, Zhou Y, Xu S, Zuo X, Cao Y, Zhang X. PRM1 variant rs35576928 (Arg>Ser) is associated with defective spermatogenesis in the Chinese Han population. Reprod Biomed Online. 2012 Dec;25(6):627-34. 2012; 25:627-634. 100. Hu Z, Xia Y, Guo X, Dai J, Li H, Hu H, Jiang Y, Lu F, Wu Y, Yang X, Li H, Yao B, et al. A genome-wide association study in Chinese men identifies three risk loci for non-obstructive azoospermia. Nat Genet 2012; 44:183-186. 101. Singh P, Schimenti JC. The genetics of human infertility by functional interrogation of SNPs in mice. Proc Natl Acad Sci U S A 2015; 112:10431-10436. 102. Shalem O, Sanjana NE, Hartenian E, Shi X, Scott DA, Mikkelsen TS, Heckl D, Ebert BL, Root DE, Doench JG, Zhang F. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 2014; 343:84-87. Prospects of spermatogonial stem cell transplantation 135

103. Sanjana NE, Shalem O, Zhang F. Improved vectors and genome-wide libraries for CRISPR screening. Nat Methods 2014; 11:783-784. 104. Jan SZ, Hamer G, Repping S, de Rooij DG, van Pelt AM, Vormer TL. Molecular control of rodent spermatogenesis. Biochim Biophys Acta 2012. 105. Handyside A, Lesko J, Tarín J, Winston R, MR. H. Birth of a normal girl after in vitro fertilization and preimplantation diagnostic testing for cystic fibrosis. N Engl J Med. 1992; 327:905-909. 106. Kuliev A, Rechitsky S, Verlinsky O, Ivakhnenko V, Cieslak J, Evsikov S, Wolf G, Angastiniotis M, Kalakoutis G, Strom C, Verlinsky Y. Birth of healthy children after preimplantation diagnosis of thalassemias. J Assist Reprod Genet. 1999; 16:207-211. 107. Sermon K, Goossens V, Seneca S, Lissens W, De Vos A, Vandervorst M, Van Steirteghem A, I. L. Preimplantation diagnosis for Huntington's disease (HD): clinical application and analysis of the HD expansion in affected embryos. Prenat Diagn. 1998 Dec;18(13):1427-36. 1998; 18:1427-1436. 108. Wu Y, Liang D, Wang Y, Bai M, Tang W, Bao S, Yan Z, Li D, Li J. Correction of a genetic disease in mouse via use of CRISPR-Cas9. Cell Stem Cell 2013; 13:659-662. 109. Long C, McAnally JR, Shelton JM, Mireault AA, Bassel-Duby R, Olson EN. Prevention of muscular dystrophy in mice by CRISPR/Cas9-mediated editing of germline DNA. Science 2014; 345:1184-1188. 110. Yin H, Xue W, Chen S, Bogorad RL, Benedetti E, Grompe M, Koteliansky V, Sharp PA, Jacks T, Anderson DG. Genome editing with Cas9 in adult mice corrects a disease mutation and phenotype. Nat Biotechnol 2014; 32:551-553. 111. Xie F, Ye L, Chang JC, Beyer AI, Wang J, Muench MO, Kan YW. Seamless gene correction of beta-thalassemia mutations in patient-specific iPSCs using CRISPR/Cas9 and piggyBac. Genome Res 2014; 24:1526-1533. 112. Li HL, Fujimoto N, Sasakawa N, Shirai S, Ohkame T, Sakuma T, Tanaka M, Amano N, Watanabe A, Sakurai H, Yamamoto T, Yamanaka S, et al. Precise correction of the dystrophin gene in duchenne muscular dystrophy patient induced pluripotent stem cells by TALEN and CRISPR-Cas9. Stem Cell Reports 2015; 4:143-154. 113. Schwank G, Koo BK, Sasselli V, Dekkers JF, Heo I, Demircan T, Sasaki N, Boymans S, Cuppen E, van der Ent CK, Nieuwenhuis EE, Beekman JM, et al. Functional repair of CFTR by CRISPR/Cas9 in intestinal stem cell organoids of cystic fibrosis patients. Cell Stem Cell 2013; 13:653- 658. 114. Berger SL, Kouzarides T, Shiekhattar R, Shilatifard A. An operational definition of epigenetics. Genes Dev 2009; 23:781-783. 115. Dawson MA, Kouzarides T. Cancer epigenetics: from mechanism to therapy. Cell 2012; 150:12-27. 116. Ordovas JM, Smith CE. Epigenetics and cardiovascular disease. Nat Rev Cardiol 2010; 7:510-519. 117. MacFarlane AJ, Strom A, Scott FW. Epigenetics: deciphering how environmental factors may modify autoimmune type 1 diabetes. Mamm Genome 2009; 20:624-632. 136 Chapter 5

118. de Groote M, Verschure P, Rots M. Epigenetic Editing: targeted rewriting of epigenetic marks to modulate expression of selected target genes. Nucleic Acids Res. 2012; 40:10596-10613. 119. Falahi F, Sgro A, Blancafort P. Epigenome engineering in cancer: fairytale or a realistic path to the clinic? Front Oncol 2015; 5:22. 120. Chen H, Kazemier H, de Groote M, Ruiters M, Xu G, Rots M. Induced DNA demethylation by targeting Ten-Eleven Translocation 2 to the human ICAM-1 promoter. Nucleic Acids Res. 2014; 42:1563-1574. 121. Hammoud S, Low D, Yi C, Carrell D, Guccione E, Cairns B. Chromatin and transcription transitions of mammalian adult germline stem cells and spermatogenesis. Cell Stem Cell. 2014 Aug 7;15(2):239-53. 2014; 15:239-253. 122. Gan H, Wen L, Liao S, Lin X, Ma T, Liu J, Song C, Wang M, He C, Han C, Tang F. Dynamics of 5-hydroxymethylcytosine during mouse spermatogenesis. Nat Commun. 2013;4:1995. 2013; 1995:1-11. 123. An J, Zhang X, Qin J, Wan Y, Hu Y, Liu T, Li J, Dong W, Du E, Pan C, Zeng W. The histone methyltransferase ESET is required for the survival of spermatogonial stem/progenitor cells in mice. Cell Death Dis 2014; 5:e1196. 124. Poplinski A, Tüttelmann F, Kanber D, Horsthemke B, Gromoll J. Idiopathic male infertility is strongly associated with aberrant methylation of MEST and IGF2/H19 ICR1. Int J Androl. 2010; 33:642-649. 125. Kläver R, F T, Bleiziffer A, Haaf T, Kliesch S, Gromol lJ. DNA methylation in spermatozoa as a prospective marker in andrology. Andrology 2013; 1:731-740. 126. Richardson M, Bleiziffer A, Tüttelmann F, Gromoll J, Wilkinson M. Epigenetic regulation of the RHOX homeobox gene cluster and its association with human male infertility. Hum Mol Genet. 2014; 23:12-23. 127. Urdinguio RG, Bayon GF, Dmitrijeva M, Torano EG, Bravo C, Fraga MF, Bassas L, Larriba S, Fernandez AF. Aberrant DNA methylation patterns of spermatozoa in men with unexplained infertility. Hum Reprod 2015; 30:1014-1028. 128. Ferfouri F, Boitrelle F, Ghout I, Albert M, Molina Gomes D, Wainer R, Bailly M, Selva J, Vialard F. A genome-wide DNA methylation study in azoospermia. Andrology 2013; 1:815-821. 129. Lumey LH. Decreased birthweights in infants after maternal in utero exposure to the Dutch famine of 1944-1945. Paediatr Perinat Epidemiol 1992; 6:240-253. 130. Heijmans BT, Tobi EW, Stein AD, Putter H, Blauw GJ, Susser ES, Slagboom PE, Lumey LH. Persistent epigenetic differences associated with prenatal exposure to famine in humans. Proc Natl Acad Sci U S A 2008; 105:17046-17049. 131. Painter RC, Osmond C, Gluckman P, Hanson M, Phillips DI, Roseboom TJ. Transgenerational effects of prenatal exposure to the Dutch famine on neonatal adiposity and health in later life. BJOG 2008; 115:1243-1249. Prospects of spermatogonial stem cell transplantation 137

132. Veenendaal MV, Painter RC, de Rooij SR, Bossuyt PM, van der Post JA, Gluckman PD, Hanson MA, Roseboom TJ. Transgenerational effects of prenatal exposure to the 1944-45 Dutch famine. BJOG 2013; 120:548-553. 133. Pembrey ME, Bygren LO, Kaati G, Edvinsson S, Northstone K, Sjostrom M, Golding J. Sex- specific, male-line transgenerational responses in humans. Eur J Hum Genet 2006; 14:159-166. 134. Buiting K, Gross S, Lich C, Gillessen-Kaesbach G, el-Maarri O, Horsthemke B. Epimutations in Prader-Willi and Angelman syndromes: a molecular study of 136 patients with an imprinting defect. Am J Hum Genet 2003; 72:571-577. 135. Suter CM, Martin DI, Ward RL. Germline epimutation of MLH1 in individuals with multiple cancers. Nat Genet 2004; 36:497-501. 136. Daxinger L, Whitelaw E. Understanding transgenerational epigenetic inheritance via the gametes in mammals. Nat Rev Genet 2012; 13:153-162. 137. Wei Y, Schatten H, Sun QY. Environmental epigenetic inheritance through gametes and implications for human reproduction. Hum Reprod Update 2015; 21:194-208. 138. Nickkholgh B, Mizrak SC, van Daalen SK, Korver CM, Sadri-Ardekani H, Repping S, van Pelt AM. Genetic and epigenetic stability of human spermatogonial stem cells during long-term culture. Fertil Steril 2014; 102:1700-1707 e1701. 139. Geens M, Van de Velde H, De Block G, Goossens E, Van Steirteghem A, Tournaye H. The efficiency of magnetic-activated cell sorting and fluorescence-activated cell sorting in the decontamination of testicular cell suspensions in cancer patients. Hum Reprod 2007; 22:733-742. 140. Sadri-Ardekani H, Homburg CH, van Capel TM, van den Berg H, van der Veen F, van der Schoot CE, van Pelt AM, Repping S. Eliminating acute lymphoblastic leukemia cells from human testicular cell cultures: a pilot study. Fertil Steril 2014; 101:1072-1078 e1071. 141. Hermann BP, Sukhwani M, Salati J, Sheng Y, Chu T, Orwig KE. Separating spermatogonia from cancer cells in contaminated prepubertal primate testis cell suspensions. Hum Reprod 2011; 26:3222-3231. 142. Zhang Z, Shao S, Meistrich ML. Irradiated mouse testes efficiently support spermatogenesis derived from donor germ cells of mice and rats. J Androl 2006; 27:365-375. 143. Kanatsu-Shinohara M, Toyokuni S, Shinohara T. Genetic selection of mouse male germline stem cells in vitro: offspring from single stem cells. Biol Reprod 2005; 72:236-240. 144. Zeng W, Tang L, Bondareva A, Luo J, Megee SO, Modelski M, Blash S, Melican DT, Destrempes MM, Overton SA, Gavin WG, Ayres S, et al. Non-viral transfection of goat germline stem cells by nucleofection results in production of transgenic sperm after germ cell transplantation. Mol Reprod Dev 2012; 79:255-261. 145. D'Astolfo DS, Pagliero RJ, Pras A, Karthaus WR, Clevers H, Prasad V, Lebbink RJ, Rehmann H, Geijsen N. Efficient intracellular delivery of native proteins. Cell 2015; 161:674-690. 138 Chapter 5

146. Yang L, Guell M, Niu D, George H, Lesha E, Grishin D, Aach J, Shrock E, Xu W, Poci J, Cortazio R, Wilkinson RA, et al. Genome-wide inactivation of porcine endogenous retroviruses (PERVs). Science 2015; 350:1101-1104. 147. Wang H, Yang H, Shivalila CS, Dawlaty MM, Cheng AW, Zhang F, Jaenisch R. One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering. Cell 2013; 153:910-918. 148. Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE, Norville JE, Church GM. RNA-guided human genome engineering via Cas9. Science 2013; 339:823-826. 149. Fu Y, Foden JA, Khayter C, Maeder ML, Reyon D, Joung JK, Sander JD. High-frequency off- target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nat Biotechnol 2013; 31:822- 826. 150. Slaymaker IM, Gao L, Zetsche B, Scott DA, Yan WX, Zhang F. Rationally engineered Cas9 nucleases with improved specificity. Science 2015. 151. Kleinstiver BP, Pattanayak V, Prew MS, Tsai SQ, Nguyen NT, Zheng Z, Joung JK. High- fidelity CRISPR-Cas9 nucleases with no detectable genome-wide off-target effects. Nature 2016. 152. Zetsche B, Gootenberg JS, Abudayyeh OO, Slaymaker IM, Makarova KS, Essletzbichler P, Volz SE, Joung J, van der Oost J, Regev A, Koonin EV, Zhang F. Cpf1 Is a Single RNA-Guided Endonuclease of a Class 2 CRISPR-Cas System. Cell 2015; 163:759-771. 153. Nishimasu H, Cong L, Yan WX, Ran FA, Zetsche B, Li Y, Kurabayashi A, Ishitani R, Zhang F, Nureki O. Crystal Structure of Staphylococcus aureus Cas9. Cell 2015; 162:1113-1126. 154. Liang P, Xu Y, Zhang X, Ding C, Huang R, Zhang Z, Lv J, Xie X, Chen Y, Li Y, Sun Y, Bai Y, et al. CRISPR/Cas9-mediated gene editing in human tripronuclear zygotes. Protein Cell 2015; 6:363- 372. 155. Kang X, He W, Huang Y, Yu Q, Chen Y, Gao X, Sun X, Fan Y. Introducing precise genetic modifications into human 3PN embryos by CRISPR/Cas-mediated genome editing. J Assist Reprod Genet 2016. 156. Lanphier E, Urnov F, Haecker S, Werner M, Smolensk iJ. Don’t edit the human germ line. Nature 2015; 519:410-411. 157. Baltimore D, Berg P, Botchan M, Carroll D, Charo R, Church G, Corn J, Daley G, Doudna J, Fenner M, Greely H, Jinek M, et al. A prudent path forward for genomic engineering and germline gene modification. Science 2015; 348:36-38. 158. Ishii T. Germline genome-editing research and its socioethical implications. Trends Mol Med 2015; 21:473-481. 159. Miller HI. Germline gene therapy: We're ready. Science 2015; 348:1325. 160. Pollack R. Eugenics lurk in the shadow of CRISPR. Science 2015; 348:871. 161. Porteus MH, Dann CT. Genome editing of the germline: broadening the discussion. Mol Ther 2015; 23:980-982. 162. Vassena R, Heindryckx B, Peco R, Pennings G, Raya A, Sermon K, Veiga A. Genome engineering through CRISPR/Cas9 technology in the human germline and pluripotent stem cells. Hum Reprod Update 2016.

Chapter 6

General discussion and implications for future research 140 Chapter 6

General discussion

In the current thesis we have shed light on the dynamic response to DNA damage during spermatogonial development, with a specific focus on the role of the structural maintenance of chromosomes (SMC) 5/6 complex. In addition, we have addressed the broad and potentially large clinical prospects and implications of using CRISPR-Cas9 to treat spermatogenic failure or to prevent the transmission of genetic abnormalities to human offspring by transplantation of genetically modified human spermatogonial stem cells (SSCs). In the current chapter, we put our results into perspective and provide guidance for future research.

The role of the SMC5/6 complex during spermatogonial proliferation and differentiation One striking characteristic of spermatogenesis is the continuous change in chromatin architecture and function. Regulation of chromatin dynamics is not only required to guide processes like meiosis and chromatin compaction during spermiogenesis but also essential to safeguard genomic stability during spermatogonial development. Previous studies in model organisms such as yeast and C. elegans have clearly demonstrated the involvement of the SMC5/6 complex in chromatin modeling and genome stability maintenance during meiosis [1]. Also in mouse and human testes, the localization of SMC5/6 subunits in meiotic cells has been reported [2-4]. Interestingly, initial protein expression of SMC6 coincides exactly with spermatogonial differentiation [4]. Yet, the function of SMC5/6 during spermatogenesis remains largely unknown, highlighting the need for loss-of-function studies. To this end we applied the novel CRISPR-Cas9 technique to the cultured primary mouse SSCs, namely male germline stem (GS) cells, to generate spermatogonial cell lines lacking a functional SMC5/6 subunit. Using our optimized protocol we successfully generated a stable GS cell line devoid of non-SMC element 2 (NSMCE2) [5]. As a subunit of the SMC5/6 complex, NSMCE2 is a SUMO ligase essential for DNA damage repair in yeast [6]. In mammalian somatic cell types NSMCE2 has significant phenotypes. In mice, NSMCE2 knockout (KO) leads to embryonic lethality, while it suppresses cancer and aging in adult mice [7]. Patients with decreased expression of NSMCE2 display primordial dwarfism, extreme insulin resistance and gonadal failure [8]. We employed CRISPR-Cas9 to generate a Nsmce2-null human osteosarcoma cell line (U2OS), and found that NSMCE2 helps to resolve topological stress, for instance replication-induced chromosome entanglements [6]. Somewhat surprisingly, when we switched to GS cells, Nsmce2-/- GS cells did not show any significant phenotype compared with their Nsmce2+/+ counterparts, and the transcriptome was only marginally affected by NSMCE2 deprivation [5]. This is in sharp contrast General discussion 141 with what we and others have previously reported in somatic cell types [6-8]. It has long been known that germ cells differ from somatic cells in relation to cell biology and (epi)genetic attributes [9, 10]. Germ cells also have their unique ways to preserve DNA integrity [11]. Our results again illustrate the unique genomic integrity maintenance pathways active in the germline. Notably, a trivial role of NSMCE2 at the spermatogonial stage does not mean that this protein is absolutely dispensable for the entire process of spermatogenesis. NSMCE2 could be involved in later developmental stages such as meiosis. This is not unlikely, considering that yeast meiotic progression requires NSMCE2 function [12], and that patients with decreased expression of NSMCE2 display primary ovarian failure [8]. Transplantation of Nsmce2-/- GS cells into sterile recipient testes, which enables further development of these cells, could provide an answer to this issue. We have also endeavored to generate Smc5- and Smc6-null spermatogonial cell lines. However, we observed cell death for ˃90% of GS cells after transfection with Smc5/6-targeting plasmids. Coincidentally, we also failed to generate Smc6-null U2OS cell lines. We therefore propose that, despite the trivial role for NSMCE2, the SMC5/6 complex as a whole is indispensable for spermatogonial and perhaps also somatic cell viability. Given that reliable conditional KO systems are not available for the earliest stages of spermatogenesis due to the lack of suitable promoters, applying an inducible Cas9 system to GS cells for controlled loss-of-function studies, as reported recently by Chen et al. [13], or performing rescue experiments after gene KO, could help to elucidate the role of SMC5/6 in spermatogonia and spermatogenesis once combined with SSC transplantation.

SSC differentiation A cornerstone of lifelong spermatogenesis is the perfect balance between SSC self- renewal and differentiation. Too much self-renewal will lead to tumor-like spermatogonial clusters while excessive differentiation will deplete the stem cell pool eventually leading to a total loss of germ cells and infertility [14]. In the mouse, the SSCs are part of the greater pool of undifferentiated spermatogonia and can divide to form two single spermatogonia or stay connected via intercellular bridges to eventually form chains of usually 8 or 16 cells [15]. At a fixed stage of spermatogenesis, and regulated via retinoic acid (RA) signaling, the undifferentiated spermatogonia develop into differentiating spermatogonia. These differentiating spermatogonia are no longer dividing “freely” but instead are committed towards meiosis and subsequent spermiogenesis via a fixed program of divisions and development. Because differentiation of spermatogonia is strongly negatively correlated with 142 Chapter 6 stem cell capacity [15, 16], unravelling the mechanisms for spermatogonial differentiation is of great interest and importance. In vitro cultured SSCs, termed GS cells, provide an advantageous system to interrogate this matter. We and others have demonstrated that both self-renewal and differentiation can occur in GS cell culture, and that exogenous factors such as RA can drive GS cell differentiation in a controlled manner [5, 17, 18]. Previous in vivo studies have shown that RA plays crucial roles in spermatogonial differentiation and the initiation of meiosis [19]. Consistently, we also found that exposure of GS cells to RA led to the significant downregulation of SSC markers and genes involved in self-renewal, whereas differentiation genes were upregulated [5]. The molecular alterations induced in vitro resemble those occurred during normal spermatogenesis in vivo. Nonetheless, few target genes of RA have been identified in spermatogenic cells. By high-throughput RNA-sequencing (RNA-seq), we unraveled the transcriptomic profile of murine differentiating GS cells. Our results are in line with a recent study that also used RA to induce differentiation of mouse spermatogonia in vitro [18]. Our RNA-seq data showed a significant transcriptomic change induced by RA- mediated spermatogonial differentiation. Specifically, differentiation affected a large number of genes involved in diverse biological processes, e.g. cell adhesion, cell migration, cell cycle, cell proliferation/differentiation, apoptosis/cell death and transcription regulation. Using our RNA-seq data, novel markers that can separate undifferentiated from differentiating spermatogonia can be identified. For instance, using our RNA-seq data, we have identified and validated PDX1 as a novel marker for mouse undifferentiated spermatogonia. Furthermore, by way of prevailing genetic modification tools such as RNA interference (RNAi) or CRISPR-Cas9, and SSC transplantation, the exact function of genes found to be involved in spermatogonial development can be investigated. In the future, a similar strategy, i.e. the establishment of a reliable culture system for human SSCs followed by in vitro differentiation and RNA-seq, could be employed to probe the mechanisms for human spermatogonial development. Because human spermatogonia are thought to be very diverse, consisting of various subtypes with differential stem cell capacity, modern single-cell sequencing techniques need to be optimized and used [20]. Human spermatogonial biology is currently not sufficiently understood, and more knowledge is required for the future use of human spermatogonia in the clinic. We have recently successfully determined the transcriptomes of six successive stages of human spermatogenesis through a combination of laser-capture microdissection (LCM) and RNA-seq (Jan et al., Development, in press). Further optimization of our established protocol to allow single-cell analysis would greatly benefit our pursuit of understanding human spermatogonial biology. General discussion 143

As an alternative for SSC auto-transplantation, in vitro differentiation of SSCs to functional haploid gametes can potentially be used to treat male infertility. Nonetheless, this has remained challenging for decades, mainly due to the inability to reproduce complete meiosis and spermiogenesis in vitro. A breakthrough in this respect was achieved recently, when complete in vitro meiosis and formation of functional spermatid-like cells from embryonic stem (ES) cell-derived primordial germ cell-like cells (PGCLCs) were demonstrated [21]. However, no expression of SSC markers was detected during the differentiation process, suggesting the failure of those PGCLCs to differentiate into SSCs, probably due to the absence of growth factors essential for SSC self-renewal (e.g. GDNF and bFGF). In the future, it would be intriguing to investigate whether SSCs could initiate and maintain spermatogenesis in this culture system. Alternatively, instead of SSCs, induced- pluripotent stem cells (iPSCs) could in theory be used to generate male gametes. Either way, the development and optimization of in vitro gametogenesis methods hold great promise for use in both fundamental research and the clinic.

The impact of DNA damage on spermatogonia and male fertility Male cancer patients subjected to chemo- or radiotherapy are typically faced with a declined number of spermatogenic cells and therefore subfertility, largely due to the therapy- induced DNA lesions and the resulting germ cell apoptosis. Especially, spermatogonia are very sensitive to DNA damage [11] and severe spermatogonial apoptosis can result in SSC depletion leading to permanent infertility. On the other hand, failure to repair DNA damage or induce apoptosis can lead to transmission of genetic mutations or chromosomal aberrations to the next generation. Hence, in spermatogenic cells, a prompt DNA damage response is of overriding importance to the maintenance of transgenerational genomic stability. It is therefore thought that, in comparison to somatic cells, germ cells show differential sensitivity to DNA damage and hold unique mechanisms for DNA repair [11]. Specifically, they tend to be more prone to undergo apoptosis, displaying a slower rate of DNA repair as well as a higher incidence of unrepaired DNA damage in the surviving cells [22, 23]. Interestingly, the undifferentiated spermatogonia, including the SSCs, are more resistant to DNA damage than differentiating spermatogonia [24-27]. However, the mechanism that causes differentiating spermatogonia to undergo apoptosis more readily is currently not known. Our data revealed that, at the transcriptome level, undifferentiated and differentiating spermatogonia exhibited a highly similar response to IR. Therefore, it is likely that the increase in radiosensitivity of differentiating spermatogonia is largely caused by intrinsic properties, such as chromatin architecture, proliferation activity, gene expression 144 Chapter 6 and protein content that are already induced upon differentiation, rather than the additional genes that are induced by irradiation in both cell types. Differentiating spermatogonia may readily undergo apoptosis in response to irradiation to prevent the accumulation of genomic alterations and thus reduce the hereditary risk. In contrast, apoptosis of the undifferentiated spermatogonia will lead to permanent infertility if the SSCs are eliminated. Undifferentiated spermatogonia are likely to be “programmed” to ensure continuation of gamete production, even with a risk of genomic aberrations being transmitted to the offspring. Despite this, the underlying molecular mechanisms modulating the balance between spermatogonial apoptosis and survival, particularly in SSCs, remain largely elusive. Various genetically modified animal models, of which the generation can now be greatly facilitated by novel techniques such as CRISPR-Cas9, could be employed to further study the unique DNA damage response and DNA repair pathways in different spermatogenic cells. In addition, our optimized protocol for genome modification of mouse GS cells has enabled a more efficient generation of GS cell lines with modified/removed genes involved in the spermatogonial DNA damage response, facilitating further studies in this field.

SSC transplantation and its clinical translation In 1994, Brinster and colleagues reported successful SSC transplantation [28, 29]. They demonstrated that, upon transplantation into recipient testes, donor SSCs can go through the blood-testis barrier and relocate to the basement membrane of seminiferous tubules from where they can initiate and maintain complete spermatogenesis. Later, transplantation became the standard assay to demonstrate the stem cell capacity of SSCs. SSC transplantation has now successfully been recapitulated in a wide array of species, including rodents, domestic animals and even non-human primates, with the generation of healthy descendants or embryos from donor-derived sperm [30, 31]. The success of SSC transplantation in various animal models, especially in non-human primates, has become an important step to consider future use of SSC transplantation in a clinical setting. For instance, SSC transplantation is currently being investigated as a promising treatment option for childhood or adult cancer survivors who are at high risk of losing their fertility due to chemo- or radiotherapy [32]. In the future, SSC transplantation might also be considered to improve fertility of subfertile men, e.g. males suffering from oligozoospermia or azoospermia. Moreover, in combination with currently available genome editing tools, SSC culture and transplantation could be used to prevent transmission of genomic diseases from the affected father to the offspring [33]. At the moment, apart from ethical and clinical considerations, clinical translation of SSC transplantation is mainly hindered by technical hurdles. Above all, as shown by previous General discussion 145 animal studies, a sufficient number of transplanted SSCs is a prerequisite for successful transplantation [34, 35]. Unfortunately, in contrast to most animal studies, typically only a small testicular biopsy can be obtained from patients. As a consequence, SSCs from the obtained biopsy must be propagated in vitro prior to transplantation. In rodents, long-term culture of SSCs is well-established and rodent SSCs can proliferate in vitro for years while maintaining stem cell capacity [36]. Several groups, including ours, have reported the culture of human SSCs [37-42]. However, from these studies it is still impossible to determine if truly functional, i.e. spermatozoa-producing, spermatogonia have been cultured in vitro. First, there is no reliable molecular marker to distinguish SSCs from other testicular cells in culture. Second, after xeno-transplantation into immunodeficient mice, only migration of cultured human cells to the basal membrane of the recipient seminiferous tubules was reported [37, 38]. To unequivocally prove the presence of bona fide human SSCs in culture, de novo testicular morphogenesis using cultured cells [43, 44] or auto-transplantation to the human testis will most likely be required. Once long-term culture of human SSCs has been established and optimized, the (epi)genetic stability of cultured SSCs before and after transplantation needs to be guaranteed. Moreover, for cancer patients, lingering cancer cells must be eliminated from the primary SSC culture before auto-transplantation. Hence, extensive preclinical research is needed to ensure the safety and efficacy of SSC transplantation.

SSCs and CRISPR-Cas9-mediated germline genomic editing Since SSCs can both initiate and maintain lifelong production of sperm that can be used to generate offspring, they are an ideal target of gene manipulation to generate genetically modified animal models. Genome editing used to depend on homologous recombination which is rather inefficient and time-consuming. Yet, over the last decade, more efficient techniques such as ZFNs, TALENs and, most recently, the CRISPR-Cas9 system have emerged. These tools use engineered nucleases that are able to trigger DNA double-strand breaks (DSBs) at specific genomic loci, thereby greatly enhancing the DNA mutation rate. Of these, CRISPR-Cas9 is currently most extensively used, enabling rapid and efficient genome editing in a wide range of species and cell types including SSCs. To date, several groups have reported precise genome modification of rodent SSCs by way of CRISPR-Cas9 [45-47]. Moreover, after transplantation of genetically modified SSCs, a genetic disease has been corrected in the offspring [46]. These accomplishments in rodents bring closer the future clinical use of genetically modified human SSCs to cure spermatogenic failure, or prevent transmission of genetic disorders to the offspring 146 Chapter 6

[33]. Our optimized protocol to generate genetically modified mouse GS cell lines more efficiently may contribute to the realization of this goal in the future. Although one can expect that human SSCs should be amenable to genome modification, SSCs from large animals such as boars [48] and goats [49] are more refractory to transfection/transduction than their rodent counterparts. Thus, it is likely that also human SSCs are very refractory to gene transfection/transduction. In addition, while novel electroporation devices are increasingly harnessed to transfect SSCs with adequate efficiency, electroporation typically generates considerable cell death. Moreover, transfection of SSCs most likely reduces cell proliferation and induces senescence [50]. Together, these factors weaken the use of transfection/electroporation on the already sparse human SSCs in culture. Genome modification of human SSCs could also be achieved via viral transduction [51]. However, residual viral sequences can integrate into the genome of SSCs, raising serious safety concerns and excluding its clinical application. For males who wish to prevent the transmission of genomic disorders to the offspring via transplantation of genetically modified SSCs, more issues need to be taken into consideration. First, the disease-causing mutation needs to be corrected in SSCs. This can only be achieved when the CRISPR-Cas9-mediated DSB is repaired via homology-directed repair (HDR). Unfortunately, the incidence of HDR is rather low [52]. Second, the repaired SSCs, if any, need to be isolated from the mixed cell population and then expanded from single cells to establish stable and healthy SSC lines. Even in rodents, this procedure is technically difficult and time-consuming. We and others have demonstrated that in rodents, SSC expansion from single cells to a 6-well plate takes 2-3 months [5, 45]. It is expected to be more time-consuming for human SSCs that have a longer doubling time [53]. Moreover, although the established healthy SSC lines can be screened for (epi)genetic stability and the absence of off-target effects, long-term cultured SSCs have been reported to have a substantially impaired ability to regenerate full spermatogenesis after transplantation [54]. For males who wish to prevent transmission of genetic disorders to their offspring by transplantation of corrected SSCs, another major barrier remains which is the depletion of endogenous mutated germ cells before transplantation. Besides being necessary to prevent mutated sperm from still being produced, endogenous spermatogonia may need to be eliminated to provide access to the niche for transplanted and genetically corrected spermatogonia. In rodents, the depletion of endogenous spermatogenesis is typically achieved by injecting alkylating antineoplastic agents such as busulfan. However, for non- rodent species the administration of busulfan at dosages to sufficiently eliminate germ cells is General discussion 147 usually lethal [31], making testicular irradiation the only option. However, irradiation of a man’s testis is accompanied by serious side-effects [33]. Thus, auto-transplantation of genetically modified human SSCs to produce healthy offspring is not a straightforward procedure. Because transplantation of genetically modified SSCs results in germline transmission of modified genes to the offspring, also ethical issues remain an important topic of discussion. Many of these discussions are likely to be similar to those that have been vigorously held over the genome modification of human embryos. Initially, two groups from China reported genome editing in human zygotes by CRISPR-Cas9, albeit at low efficiency and with a high incidence of off-target effects [55, 56]. Recently, a study from the US demonstrated highly effective modification of human embryos virtually without off-target effects [57], although the validity of this study has been debated [58]. Collectively, these publications received worldwide media attention and raised intensive social and ethical debates. Interestingly, recent years witnessed a dramatic switch in terms of the public attitude towards human germline editing, from the initial complete prohibition [59] to the recent conditional permission at laboratory levels [60]. Debates on this matter are still ongoing and a broader audience including patients, clinicians, researchers, ethicists and politicians need to be involved in future discussions.

Implications for future research A major focus of this study is how genomic integrity maintenance is regulated during spermatogonial development. To study the role of the SMC5/6 complex in this process we optimized the use of CRISPR-Cas9 to edit the mouse spermatogonial genome [5]. Because CRISPR-Cas9-mediated genome editing in combination with SSC transplantation could be used to cure spermatogenic failure or prevent genetic diseases in the offspring, we also addressed the clinical prospects of SSC transplantation, with or without germline genomic editing [33]. Indeed, in vitro propagation of SSCs followed by transplantation could be a promising treatment option for subfertile men. Nevertheless, unlike the establishment of long-term culture systems for rodent SSCs, human SSC culture remains inefficient and requires optimization. Recent publications about long-term culture of undifferentiated spermatogonia from large animals such as pigs [61] and cattle [62] may provide clues on how to optimize the culture of human SSCs. Besides, given that human SSCs transplanted to a recipient mouse testis fail to undergo further differentiation due to their phylogenetic distance, the spermatogenic ability of cultured human SSCs still needs to be investigated. This could for instance be done by in vitro differentiation or in vivo developmental assays such as de novo 148 Chapter 6 testicular morphogenesis or auto-transplantation. Future studies should also focus on the (epi)genetic stability of cultured human SSCs, the fitness of the recipient males subjected to SSC transplantation, as well as the follow-up of SSC transplantation-generated offspring, if any. For SSC transplantation with germline genomic editing, much more fundamental and preclinical work needs to be conducted to guarantee the efficiency and safety of this technique. Moreover, ethical and societal implications of germline editing need to be probed. In addition, alternatives for SSC transplantation, such as three-dimensional (3D) or organ culture, which may potentially allow in vitro differentiation and maturation of human SSCs to functional sperm, could be developed and optimized. An issue that cannot be ignored is that the etiology of spermatogenic failure is unknown in most cases, largely due to the intricate regulatory mechanisms of spermatogenesis. Over the last decades, research that uses (transgenic) mouse models has greatly extended the knowledge about the regulatory mechanisms for SSC self-renewal and differentiation and further spermatogenesis. Nevertheless, due to the phylogenetic distance, these data gained via mouse studies may not fully translate to the human. An illustration of this point are the difficulties in culturing human SSCs when using protocols initially developed for the mouse. An important reason for this is the lack of knowledge about the signals or growth factors that promote human SSC self-renewal. Establishment of an immortalized human spermatogonial cell line could provide adequate cells for fundamental mechanistic studies [51]. Also, a recent paper from our department, which unraveled the transcriptomes of human male germ cells at distinct developmental stages, could provide clues for the optimization of culture systems for human SSC proliferation/differentiation (Jan et al., Development, in press). In addition, previous studies employing mouse models have well demonstrated that epigenetic factors [10], non-coding RNAs such as miRNA [63] and long non-coding (lnc) RNA [64] play important roles in regulating SSC fate and spermatogenesis. Single-cell sequencing could be further exploited to uncover the expression profiles of non-coding RNAs in sub-clusters of human germ cells. By using the novel CRISPR-Cas9 system or other genetic modification strategies, their functions may be further annotated. Collectively, it is only when fundamental studies and preclinical trials are well combined that the development of treatment options for male subfertility can be accelerated. General discussion 149

References

1. Verver DE, Hwang GH, Jordan PW, Hamer G. Resolving complex chromosome structures during meiosis: versatile deployment of Smc5/6. Chromosoma 2016; 125:15-27. 2. Gomez R, Jordan PW, Viera A, Alsheimer M, Fukuda T, Jessberger R, Llano E, Pendas AM, Handel MA, Suja JA. Dynamic localization of SMC5/6 complex proteins during mammalian meiosis and mitosis suggests functions in distinct chromosome processes. J Cell Sci 2013; 126:4239-4252. 3. Verver DE, Langedijk NSM, Jordan PW, Repping S, Hamer G. The SMC5/6 Complex Is Involved in Crucial Processes During Human Spermatogenesis. Biology Of Reproduction 2014; 91. 4. Verver DE, van Pelt AM, Repping S, Hamer G. Role for rodent Smc6 in pericentromeric heterochromatin domains during spermatogonial differentiation and meiosis. Cell Death Dis 2013; 4:e749. 5. Zheng Y, Jongejan A, Mulder CL, Mastenbroek S, Repping S, Wang Y, Li J, Hamer G. Trivial role for NSMCE2 during in vitro proliferation and differentiation of GS cells. Reproduction 2017. 6. Verver DE, Zheng Y, Speijer D, Hoebe R, Dekker HL, Repping S, Stap J, Hamer G. Non-SMC Element 2 (NSMCE2) of the SMC5/6 Complex Helps to Resolve Topological Stress. Int J Mol Sci 2016; 17. 7. Jacome A, Gutierrez-Martinez P, Schiavoni F, Tenaglia E, Martinez P, Rodriguez-Acebes S, Lecona E, Murga M, Mendez J, Blasco MA, Fernandez-Capetillo O. NSMCE2 suppresses cancer and aging in mice independently of its SUMO ligase activity. EMBO J 2015; 34:2604-2619. 8. Payne F, Colnaghi R, Rocha N, Seth A, Harris J, Carpenter G, Bottomley WE, Wheeler E, Wong S, Saudek V, Savage D, O'Rahilly S, et al. Hypomorphism in human NSMCE2 linked to primordial dwarfism and insulin resistance. J Clin Invest 2014; 124:4028-4038. 9. Manku G, Culty M. Mammalian gonocyte and spermatogonia differentiation: recent advances and remaining challenges. Reproduction 2015; 149:R139-157. 10. Tseng YT, Liao HF, Yu CY, Mo CF, Lin SP. Epigenetic factors in the regulation of prospermatogonia and spermatogonial stem cells. Reproduction 2015; 150:R77-91. 11. Marjault HB, Allemand I. Consequences of irradiation on adult spermatogenesis: Between infertility and hereditary risk. Mutat Res 2016; 770:340-348. 12. Pebernard S, McDonald WH, Pavlova Y, Yates JR, 3rd, Boddy MN. Nse1, Nse2, and a novel subunit of the Smc5-Smc6 complex, Nse3, play a crucial role in meiosis. Mol Biol Cell 2004; 15:4866- 4876. 13. Chen J, Cai T, Zheng C, Lin X, Wang G, Liao S, Wang X, Gan H, Zhang D, Hu X, Wang S, Li Z, et al. MicroRNA-202 maintains spermatogonial stem cells by inhibiting cell cycle regulators and RNA binding proteins. Nucleic Acids Res 2017; 45:4142-4157. 14. Silber SJ. Evaluation and treatment of male infertility. Clin Obstet Gynecol 2000; 43:854-888. 15. De Rooij DG, Russell LD. All you wanted to know about spermatogonia but were afraid to ask. Journal Of Andrology 2000; 21:776-798. 150 Chapter 6

16. Lord T, Oatley JM. A revised Asingle model to explain stem cell dynamics in the mouse male germline. Reproduction 2017; 154:R55-R64. 17. Dann CT, Alvarado AL, Molyneux LA, Denard BS, Garbers DL, Porteus MH. Spermatogonial stem cell self-renewal requires OCT4, a factor downregulated during retinoic acid-induced differentiation. Stem Cells 2008; 26:2928-2937. 18. Wang S, Wang X, Ma L, Lin X, Zhang D, Li Z, Wu Y, Zheng C, Feng X, Liao S, Feng Y, Chen J, et al. Retinoic Acid Is Sufficient for the In Vitro Induction of Mouse Spermatocytes. Stem Cell Reports 2016; 7:80-94. 19. Busada JT, Geyer CB. The Role of Retinoic Acid (RA) in Spermatogonial Differentiation. Biol Reprod 2016; 94:10. 20. Neuhaus N, Yoon J, Terwort N, Kliesch S, Seggewiss J, Huge A, Voss R, Schlatt S, Grindberg RV, Scholer R. Single-cell gene expression analysis reveals diversity among human spermatogonia. Molecular Human Reproduction 2017; 23:79-90. 21. Zhou Q, Wang M, Yuan Y, Wang X, Fu R, Wan H, Xie M, Liu M, Guo X, Zheng Y, Feng G, Shi Q, et al. Complete Meiosis from Embryonic Stem Cell-Derived Germ Cells In Vitro. Cell Stem Cell 2016; 18:330-340. 22. Paris L, Cordelli E, Eleuteri P, Grollino MG, Pasquali E, Ranaldi R, Meschini R, Pacchierotti F. Kinetics of gamma-H2AX induction and removal in bone marrow and testicular cells of mice after X-ray irradiation. Mutagenesis 2011; 26:563-572. 23. Rube CE, Zhang S, Miebach N, Fricke A, Rube C. Protecting the heritable genome: DNA damage response mechanisms in spermatogonial stem cells. DNA Repair (Amst) 2011; 10:159-168. 24. van der Meer Y, Huiskamp R, Davids JA, van der Tweel I, de Rooij DG. The sensitivity to X rays of mouse spermatogonia that are committed to differentiate and of differentiating spermatogonia. Radiat Res 1992; 130:296-302. 25. van Beek ME, Meistrich ML, de Rooij DG. Probability of self-renewing divisions of spermatogonial stem cells in colonies, formed after fission neutron irradiation. Cell Tissue Kinet 1990; 23:1-16. 26. Aloisio GM, Nakada Y, Saatcioglu HD, Pena CG, Baker MD, Tarnawa ED, Mukherjee J, Manjunath H, Bugde A, Sengupta AL, Amatruda JF, Cuevas I, et al. PAX7 expression defines germline stem cells in the adult testis. J Clin Invest 2014; 124:3929-3944. 27. Komai Y, Tanaka T, Tokuyama Y, Yanai H, Ohe S, Omachi T, Atsumi N, Yoshida N, Kumano K, Hisha H, Matsuda T, Ueno H. Bmi1 expression in long-term germ stem cells. Sci Rep 2014; 4:6175. 28. Brinster RL, Avarbock MR. Germline transmission of donor haplotype following spermatogonial transplantation. Proc Natl Acad Sci U S A 1994; 91:11303-11307. 29. Brinster RL, Zimmermann JW. Spermatogenesis following male germ-cell transplantation. Proc Natl Acad Sci U S A 1994; 91:11298-11302. 30. Dores C, Alpaugh W, Dobrinski I. From in vitro culture to in vivo models to study testis development and spermatogenesis. Cell Tissue Res 2012; 349:691-702. General discussion 151

31. Gonzalez R, Dobrinski I. Beyond the mouse monopoly: studying the male germ line in domestic animal models. ILAR J 2015; 56:83-98. 32. Struijk RB, Mulder CL, van der Veen F, van Pelt AM, Repping S. Restoring fertility in sterile childhood cancer survivors by autotransplanting spermatogonial stem cells: are we there yet? Biomed Res Int 2013; 2013:903142. 33. Mulder CL, Zheng Y, Jan SZ, Struijk RB, Repping S, Hamer G, van Pelt AM. Spermatogonial stem cell autotransplantation and germline genomic editing: a future cure for spermatogenic failure and prevention of transmission of genomic diseases. Hum Reprod Update 2016; 22:561-573. 34. Dobrinski I, Ogawa T, Avarbock MR, Brinster RL. Computer assisted image analysis to assess colonization of recipient seminiferous tubules by spermatogonial stem cells from transgenic donor mice. Mol Reprod Dev 1999; 53:142-148. 35. Nagano MC. Homing efficiency and proliferation kinetics of male germ line stem cells following transplantation in mice. Biol Reprod 2003; 69:701-707. 36. Kanatsu-Shinohara M, Ogonuki N, Iwano T, Lee J, Kazuki Y, Inoue K, Miki H, Takehashi M, Toyokuni S, Shinkai Y, Oshimura M, Ishino F, et al. Genetic and epigenetic properties of mouse male germline stem cells during long-term culture. Development 2005; 132:4155-4163. 37. Sadri-Ardekani H, Mizrak SC, van Daalen SK, Korver CM, Roepers-Gajadien HL, Koruji M, Hovingh S, de Reijke TM, de la Rosette JJ, van der Veen F, de Rooij DG, Repping S, et al. Propagation of human spermatogonial stem cells in vitro. JAMA 2009; 302:2127-2134. 38. Sadri-Ardekani H, Akhondi MA, van der Veen F, Repping S, van Pelt AM. In vitro propagation of human prepubertal spermatogonial stem cells. JAMA 2011; 305:2416-2418. 39. Lim JJ, Sung SY, Kim HJ, Song SH, Hong JY, Yoon TK, Kim JK, Kim KS, Lee DR. Long-term proliferation and characterization of human spermatogonial stem cells obtained from obstructive and non-obstructive azoospermia under exogenous feeder-free culture conditions. Cell Prolif 2010; 43:405- 417. 40. Guo Y, Liu L, Sun M, Hai Y, Li Z, He Z. Expansion and long-term culture of human spermatogonial stem cells via the activation of SMAD3 and AKT pathways. Exp Biol Med (Maywood) 2015; 240:1112-1122. 41. Kossack N, Terwort N, Wistuba J, Ehmcke J, Schlatt S, Scholer H, Kliesch S, Gromoll J. A combined approach facilitates the reliable detection of human spermatogonia in vitro. Hum Reprod 2013; 28:3012-3025. 42. Akhondi MM, Mohazzab A, Jeddi-Tehrani M, Sadeghi MR, Eidi A, Khodadadi A, Piravar Z. Propagation of human germ stem cells in long-term culture. Iran J Reprod Med 2013; 11:551-558. 43. Gassei K, Orwig KE. Experimental methods to preserve male fertility and treat male factor infertility. Fertil Steril 2016; 105:256-266. 44. Honaramooz A, Megee SO, Rathi R, Dobrinski I. Building a testis: formation of functional testis tissue after transplantation of isolated porcine (Sus scrofa) testis cells. Biol Reprod 2007; 76:43-47. 152 Chapter 6

45. Chapman KM, Medrano GA, Jaichander P, Chaudhary J, Waits AE, Nobrega MA, Hotaling JM, Ober C, Hamra FK. Targeted Germline Modifications in Rats Using CRISPR/Cas9 and Spermatogonial Stem Cells. Cell Rep 2015; 10:1828-1835. 46. Wu Y, Zhou H, Fan X, Zhang Y, Zhang M, Wang Y, Xie Z, Bai M, Yin Q, Liang D, Tang W, Liao J, et al. Correction of a genetic disease by CRISPR-Cas9-mediated gene editing in mouse spermatogonial stem cells. Cell Res 2015; 25:67-79. 47. Sato T, Sakuma T, Yokonishi T, Katagiri K, Kamimura S, Ogonuki N, Ogura A, Yamamoto T, Ogawa T. Genome Editing in Mouse Spermatogonial Stem Cell Lines Using TALEN and Double- Nicking CRISPR/Cas9. Stem Cell Reports 2015; 5:75-82. 48. Zeng W, Tang L, Bondareva A, Honaramooz A, Tanco V, Dores C, Megee S, Modelski M, Rodriguez-Sosa JR, Paczkowski M, Silva E, Wheeler M, et al. Viral transduction of male germline stem cells results in transgene transmission after germ cell transplantation in pigs. Biol Reprod 2013; 88:27. 49. Zeng W, Tang L, Bondareva A, Luo J, Megee SO, Modelski M, Blash S, Melican DT, Destrempes MM, Overton SA, Gavin WG, Ayres S, et al. Non-viral transfection of goat germline stem cells by nucleofection results in production of transgenic sperm after germ cell transplantation. Mol Reprod Dev 2012; 79:255-261. 50. Laible G, Alonso-Gonzalez L. Gene targeting from laboratory to livestock: current status and emerging concepts. Biotechnol J 2009; 4:1278-1292. 51. Hou J, Niu M, Liu L, Zhu Z, Wang X, Sun M, Yuan Q, Yang S, Zeng W, Liu Y, Li Z, He Z. Establishment and Characterization of Human Germline Stem Cell Line with Unlimited Proliferation Potentials and no Tumor Formation. Sci Rep 2015; 5:16922. 52. West J, Gill WW. Genome Editing in Large Animals. J Equine Vet Sci 2016; 41:1-6. 53. Guo Y, Hai Y, Gong Y, Li Z, He Z. Characterization, isolation, and culture of mouse and human spermatogonial stem cells. J Cell Physiol 2014; 229:407-413. 54. Helsel AR, Oatley MJ, Oatley JM. Glycolysis-Optimized Conditions Enhance Maintenance of Regenerative Integrity in Mouse Spermatogonial Stem Cells during Long-Term Culture. Stem Cell Reports 2017; 8:1430-1441. 55. Liang P, Xu Y, Zhang X, Ding C, Huang R, Zhang Z, Lv J, Xie X, Chen Y, Li Y, Sun Y, Bai Y, et al. CRISPR/Cas9-mediated gene editing in human tripronuclear zygotes. Protein Cell 2015; 6:363- 372. 56. Kang X, He W, Huang Y, Yu Q, Chen Y, Gao X, Sun X, Fan Y. Introducing precise genetic modifications into human 3PN embryos by CRISPR/Cas-mediated genome editing. J Assist Reprod Genet 2016; 33:581-588. 57. Ma H, Marti-Gutierrez N, Park SW, Wu J, Lee Y, Suzuki K, Koski A, Ji D, Hayama T, Ahmed R, Darby H, Van Dyken C, et al. Correction of a pathogenic gene mutation in human embryos. Nature 2017; 548:413-419. 58. Egli D, Zuccaro M, Kosicki M, Church G, Bradley A, Jasin M. Inter-homologue repair in fertilized human eggs? bioRxiv 2017. General discussion 153

59. Lanphier E, Urnov F, Haecker SE, Werner M, Smolenski J. Don't edit the human germ line. Nature 2015; 519:410-411. 60. Kaiser J. A yellow light for embryo editing. Science 2017; 355:675. 61. Zhang P, Chen X, Zheng Y, Zhu J, Qin Y, Lv Y, Zeng W. Long-Term Propagation of Porcine Undifferentiated Spermatogonia. Stem Cells Dev 2017. 62. Oatley MJ, Kaucher AV, Yang QE, Waqas MS, Oatley JM. Conditions for Long-Term Culture of Cattle Undifferentiated Spermatogonia. Biol Reprod 2016; 95:14. 63. Chen X, Li X, Guo J, Zhang P, Zeng W. The roles of microRNAs in regulation of mammalian spermatogenesis. J Anim Sci Biotechnol 2017; 8:35. 64. Li L, Wang M, Wang M, Wu X, Geng L, Xue Y, Wei X, Jia Y, Wu X. A long non-coding RNA interacts with Gfra1 and maintains survival of mouse spermatogonial stem cells. Cell Death Dis 2016; 7:e2140.

Chapter 7

Summary Samenvatting 156 Chapter 7

Summary

The aim of this thesis was to unravel the mechanisms that determine and regulate the dynamic response to DNA damage during spermatogonial development, with a specific focus on the role of the structural maintenance of chromosomes (SMC) 5/6 complex. To this end we used CRISPR-Cas9-mediated genome editing to remove genes of interest from spermatogonial stem cells (SSCs) in culture. Because cultured and genetically modified SSCs can be (auto-)transplanted to recipient donor testes, our optimized method of germline genome editing evoked clinical and ethical questions that were reviewed at the end of the thesis. Following is the summary of our findings.

In Chapter 1 we provided an overview of spermatogenic failure, spermatogenesis, SSCs, DNA damage repair, chromatin dynamics, the SMC5/6 complex and genome editing. In Chapter 2 we aimed to gain more insights into the molecular mechanisms by which the SMC5/6 complex works in mammalian cells. We therefore harnessed CRISPR-Cas9 to generate a human osteosarcoma cell line (U2OS) that lacks NSMCE2 (non-SMC element 2), a subunit of the SMC5/6 complex described to be essential for DNA repair. Using this cell line, we were able to find that treatment with the topoisomerase II inhibitor etoposide triggered an increased sensitivity in cells lacking NSMCE2. In contrast, NSMCE2 appeared not essential for a proper DNA damage response or cell survival after DNA double-strand break (DSB) induction by ionizing irradiation (IR). By way of immunoprecipitation and mass spectrometry, we found that the SMC5/6 complex physically interacts with the DNA topoisomerase TOP2A. We therefore propose that the SMC5/6 complex functions in resolving TOP2A-mediated DSB-repair intermediates that are generated during replication. In Chapter 3 we reported an optimized protocol to generate genetically modified mouse male germline stem (GS) cell lines using CRISPR-Cas9. By this we generated a GS cell line devoid of NSMCE2. We found that NSMCE2 was dispensable for proliferation, differentiation and topological stress relief in mouse GS cells. Moreover, RNA-sequencing analysis demonstrated that the spermatogonial transcriptome was only minimally affected by the absence of NSMCE2. Only differential expression of Sgsm1 appeared highly significant, but with SGSM1 protein levels being unaffected without NSMCE2. Hence, despite the essential roles of NSMCE2 in somatic cells, genome integrity maintenance in the germline seems to be regulated differently, without the requirement of NSMCE2. Summary 157

In Chapter 4 we used in vitro cultured GS cells to investigate the influence of spermatogonial differentiation on the response to DNA damage. Interestingly, the sensitivity of spermatogonia to DNA damage increases during differentiation but the mechanisms responsible for this have not been revealed before. We analyzed the transcriptomes of in vitro differentiating and undifferentiated spermatogonia with and without IR treatment. At the RNA level, both spermatogonial subtypes showed a very similar response to IR. When comparing undifferentiated and differentiating spermatogonia, there were no IR-induced differentially expressed genes (DEGs) that could be attributed to spermatogonial differentiation. Nevertheless, in response to IR, differentiating spermatogonia did not upregulate the DNA damage response protein p53 as much as undifferentiated spermatogonia. Hence, the difference in radiosensitivity between undifferentiated and differentiating spermatogonia is largely caused by intrinsic properties, like chromatin architecture, gene expression, proliferation rate and protein content that are already induced upon differentiation, rather than additional genes induced by irradiation in both cell types. In Chapter 5 we reviewed the state of the art with respect to SSC transplantation and genomic editing using CRISPR-Cas9, followed by envisioning the clinical prospects of SSC transplantation, with or without genomic editing, to restore male fertility or prevent transmission of genomic disorders. The clinical and technical hindrances, as well as ethical issues, were disclosed as well. Despite the requirement of optimization at various levels, SSC transplantation, with or without genomic editing, is still a promising platform that may be used in the clinic in the future. In Chapter 6 we discussed the results obtained in this thesis and uncovered implications for future (pre)clinical and fundamental research. 158 Chapter 7

Samenvatting

Het doel van dit proefschrift is het ontrafelen van de mechanismen die de veranderlijke reactie bepalen van zich ontwikkelende spermatogonia op DNA-schade, met een focus op de rol van het “structural maintenance of chromosomes” (SMC) 5/6 complex in dit proces. Om dit te bereiken hebben we de CRISPR-Cas9-methode voor genoomediting gebruikt om relevante genen te verwijderen uit spermatogoniale stamcellen (SSCs). Omdat gekweekte en genetisch gemodificeerde SSCs getransplanteerd kunnen worden naar een donortestis roept onze geoptimaliseerde methode voor kiembaanmodificatie ook klinische en ethische vragen op die beschreven staan aan het eind van dit proefschrift. Nu volgt een samenvatting van onze bevindingen.

In hoofdstuk 1 geven we een overzicht van spermatogenese, spermatogenetisch falen, SSCs, reparatie van DNA-schade, dynamiek van chromatine, het SMC5/6 complex en genoom-modificatie. In hoofdstuk 2 onderzoeken we de moleculaire mechanismen die de functie van het SMC5/6 complex bepalen in zoogdiercellen. Hiertoe hebben we CRISPR-Cas9 gebruikt om non-SMC element 2 (NSMCE2, een onderdeel van SMC5/6 dat essentieel is voor reparatie van DNA-schade) te verwijderen uit humane osteosarcoma cellen. Gebruik makende van deze cellijn hebben we gevonden dat gebrek aan NSMCE2 tot gevolg heeft dat cellen gevoeliger worden voor topologische stress, in dit geval veroorzaakt door de topoisomerase II-remmer etoposide. NSMCE2 bleek echter niet essentieel voor het repareren van DNA- schade veroorzaakt door ioniserende straling. We hebben bovendien gevonden dat het SMC5/6 complex fysiek bindt aan de DNA topoisomerase TOP2A. We stellen daarom voor dat het SMC5/6 complex bijdraagt aan het repareren van TOP2A-gemedieerde DNA-schade die veroorzaakt wordt door topologische stress die optreedt gedurende het repliceren van DNA. In hoofdstuk 3 beschrijven we een geoptimaliseerd protocol voor het gebruik van CRISPR-Cas9 voor het maken van genetisch gemodificeerde muizenzaadcel-stamcellijnen (GS-cellen). Met dit protocol hebben we GS-cellen gemaakt die geen NSMCE2 meer hebben. Het blijkt dat in deze cellen NSMCE2 niet nodig is voor proliferatie, differentiatie of het oplossen van topologische stress. Uit transcriptoom-analyse blijkt bovendien dat het spermatogoniale transcriptoom bijna niet beïnvloed wordt door het afwezig zijn van NSMCE2. Alleen expressie van het gen Sgsm1 leek significant te verschillen, ware het niet dat het eiwit SGSM1 onveranderlijk aanwezig bleef. Ondanks de essentiële rol van NSMCE2 in Samenvatting 159 somatische cellen lijkt het er dus op dat het bewaken van genomische integriteit anders gereguleerd is in voortplantingscellen, niet noodzakelijkerwijs via NSMCE2. In hoofdstuk 4 hebben we GS-cellen gebruikt om het gevolg van spermatogoniale differentiatie op de reactie van deze cellen op DNA-schade te onderzoeken. Het is bekend dat de stralingsgevoeligheid van spermatogoniën toeneemt als ze differentiëren. Wat hieraan ten grondslag ligt is echter onbekend. We hebben het transcriptoom geanalyseerd van in vitro differentiërende en ongedifferentieerde spermatogonia, met en zonder stralingsschade. Beide spermatogoniale subtypen lieten een sterk vergelijkbaar genexpressiepatroon zien in reactie op ioniserende straling. In deze vergelijking kwamen er geen differentieel tot expressie gebrachte genen (DEGs) aan het licht veroorzaakt door spermatogoniale differentiatie. Desondanks lieten differentiërende spermatogonia een minder sterke verrijking van het DNA-schade eiwit p53 zien. In het algemeen bleek het verschil in stralingsgevoeligheid tussen differentiërende en ongedifferentieerde spermatogonia vooral bepaald te worden door intrinsieke eigenschappen zoals chromatine-architectuur, genexpressie, proliferatie en eiwitten die al geïnduceerd waren door differentiatie, en niet door genen die geïnduceerd werden door stralingsschade in beide celtypen. In hoofdstuk 5 geven we een overzicht van de stand van zaken met betrekking tot SSC-transplantatie en genoommodificatie middels CRISPR-Cas9, gevolgd door een beschrijving van de klinische vooruitzichten van SSC-transplantatie (met of zonder genoommodificatie) voor het herstellen van mannelijke vruchtbaarheid of het voorkomen van het doorgeven van erfelijke ziekten aan het nageslacht. Klinische en technische problemen en ethische kwesties zijn hierbij meegenomen. Ondanks dat verdere optimalisatie nog op alle vlakken nodig is, blijkt SSC-transplantatie, met of zonder genoommodificatie, een veelbelovende methode die wellicht ooit toepast zal worden in de kliniek. In hoofdstuk 6 worden de resultaten beschreven in dit proefschrift bediscussieerd, inclusief implicaties voor mogelijk toekomstig (pre)klinisch en fundamenteel onderzoek.

160 Acknowledgements

Acknowledgements

Over the last 4 years I was honored to join in this group as a PhD student and meet, communicate and work with these fabulous people. It's somewhat sad that the farewell time finally comes. Here I would like to convey my sincere gratitude to all the great people that I have met during my stay in the Netherlands. First, my promotor, prof. dr. Sjoerd Repping. Sjoerd, thank you very much for accepting me as a PhD student in your group and thus giving me an opportunity to learn, be trained and grow professionally. Although not occurring too often over the last 4 years, I do appreciate each time of our conversations, as they were always stimulating and refreshing. You always remind me of the big picture of my thesis. Besides, I am impressed by your dedication to “good science” in the current blundering scientific community. This kind of scientific attitude will always go along with me in my future journey. Another person to whom I would like to convey my heartfelt gratitude is my co-promotor, dr. Geert Hamer. Geert, I am happy that finally I can have an opportunity to let you know how much I appreciate you and how happy I work with you. Absolutely this thesis would never come true without you. I will never forget our first meeting in the airport, where you picked me up upon my first arrival in the Netherlands. I am happy that I could work with you and I was never pushed but had full liberty to arrange my experiments, which makes me think independently and grow professionally. Moreover, I got a lot of valuable scientific feedback from you, and was always reassured under tricky circumstance. Overall, I would be hard- pressed to express all my gratitude to your understanding and support during my PhD study. As I will stay in the field of spermatogenesis and stem cells, I am looking forward to our future collaboration. Absolutely I will acknowledge my doctorate committee members, dr. ir. W.M. Baarends, prof. dr. N. Zelcer, prof. dr. C.J.F. van Noorden, dr. N.A.P. Franken and prof. dr. D.G. de Rooij. Thank you for making up my thesis committee and spending precious time in reading and judging my thesis. The persons that I will not forget to acknowledge must include Saskia and Cindy. I am grateful to you for teaching me important experimental skills and your help in my project! Apart from technicians, I would like to convey my sincere gratitude to our dear secretary, Beatrix. Thank you so much for your help in all aspects during my stay and study in the Netherlands! Also, I appreciate the help and efforts from Aldo, our in-house bioinformatician. It is your support and help that makes my thesis finally come true! Acknowledgements 161

My dear colleagues, Callista, Jitske, Sabrina, Joana, Ieva, Myrthe, Robbert, Robin, Majid, Febilla, Vera, Arno, Kai Mee, Emma, Hajar, Louise, Jan Willem, etc., thank you for your accompanying during my PhD study! I miss the life that we discuss and have lunch together, and the memory of our dinners and parties always stays in my mind. I am proud of you and I wish you all the best in your future career! I would also not forget the help from other principal investigators in CVV and members of MORG: Ans, Sebastiaan, Gijs, Carrie, Marie, Remco, Truus and Souad. Thanks a lot for your useful suggestions and help in my project. To talk with you is always compelling. Indeed, what I benefit from our talks are not only scientific feedback, but also attitudes towards life and dilemma. Coming to a foreign country alone is always accompanied by some extent of nervousness and apprehension. It’s reassuring to meet and know some people who speak the same language and share the same culture. Thanks to my housemate Na Zhao, Gang Wang, Chao Ding and Zemin Ren, I genuinely felt at home when I live with you. I am also grateful to my dear friends Zongliang, Xiaoxia, Lihui, Xiuping, Jing Zhao, Wanhai, Xiao Yu, etc. I will never forget the time that we have dinners and travel together, and I sincerely wish you cheerfulness and success in your future journey! My scientific career started in 2010 when I became a master student; therefore, I would like to acknowledge prof. dr. Wenxian Zeng, who is the supervisor for my master’s study. It is under your guidance and supervision that I gained some basic knowledge about spermatogonial stem cells and spermatogenesis, which laid the groundwork for my PhD study. Furthermore, thank you so much for your efforts in my job application. Now it’s with great joy to rejoin your flourishing group. With your established platform and my experience gained in these years, I believe we have good chances for significant scientific accomplishments. Finally, I would like to devote my infinite gratitude to the most important persons in my life: my parents, for all the love and support during these years. It goes without saying that everything in my life would not have been possible without your support and understanding. Now my only wish for you is that you were healthy and happy every day! The 4-year PhD study in the Netherlands is a fantastic journey in my life. Not only am I trained to become an independent researcher, but also know the Dutch, the Netherlands, the Europe and the world outside of China. Yes, life is often associated with frustration and adversity. However, this journey is also surrounded by love, hope and a host of wonderful memories of times, places, people and emotions. With these staying in my heart I will definitely go ahead, no fear, no hesitation. 162 PhD portfolio

PhD portfolio

PhD period: 1.10.2013-30.9.2017 PhD supervisors: Prof. Dr. Sjoerd Repping Dr. Geert Hamer

PhD training Year Workload (Hours/ECTS) General courses Oral Presentation in English (AMC) 2014 22/0.8 The AMC World of Science (AMC) 2014 20/0.7 Scientific Writing in English for Publication (AMC) 2017 42/1.5 Specific courses Laboratory Animal Science (Utrecht University) 2014 80/3.0 DNA Technology (AMC) 2014 60/2.1 Advanced qPCR (AMC) 2015 20/0.7 Bioinformatics Sequence Analysis (AMC) 2016 30/1.1 Seminars, workshops and master classes Journal club 2013-2017 112/4.0 Weekly department seminars (ReproBio) 2013-2017 112/4.0 Progress report seminars 2013-2017 112/4.0 Oral presentations (invited) Bringing CRISPR/Cas9 to aneuploidy and refractory cells: 2015 14/0.5 practical issues and suggestions, AMC CRISPR Symposium (Inter)national conferences 7th Meeting of the International Network for Young 2014 16/0.5 Researchers in Male Fertility (Elsinore, Denmark) Symposium: Genome on Demand? Exploring the 2015 16/0.5 Implications of Human Genome Editing (Amsterdam) 9th Dutch Society for Stem Cell Research Meeting (Utrecht) 2016 8/0.25 10th Dutch Society for Stem Cell Research Meeting (Utrecht) 2017 8/0.25 Total 672/23.9

About the author 163

About the author

The author of this thesis, Yi Zheng, was born on September 8th, 1988, in Yaan, Sichuan, China, where is also the hometown of panda. From 2006 to 2010, as an undergraduate student, he studied Animal Science at Sichuan Agricultural University, in Sichuan, China. After his bachelor graduation in June 2010, he enrolled in Northwest A&F University, in Shaanxi, China, where he became a master student and studied Animal Genetics, Breeding and Reproduction Science. During the 3-year master study, he focused on characterization and in vitro culture of spermatogonial stem cells (SSCs) from pigs under the supervision of Prof. Wenxian Zeng. In 2013, he obtained his master’s degree and a scholarship from China Scholarship Council (CSC) to support his PhD study abroad. With the funding and a strong interest in spermatogenesis and SSCs, in October 2013 he joined Reproductive Biology Laboratory, Center for Reproductive Medicine, Academic Medical Center, University of Amsterdam, The Netherlands, as a PhD student under the supervision of Prof. Sjoerd Repping and Dr. Geert Hamer, where he focused on genome integrity maintenance during spermatogonial development. The results of the work during his PhD are presented in this thesis. 164 List of publications

List of publications

Mulder CL#, Zheng Y#, Jan SZ, Struijk RB, Repping S, Hamer G, van Pelt AM. Spermatogonial stem cell autotransplantation and germline genomic editing: a future cure for spermatogenic failure and prevention of transmission of genomic diseases. Human Reproduction Update. 2016 Sep;22(5):561- 73. (#equal contribution)

Zheng Y, Jongejan A, Mulder CL, Mastenbroek S, Repping S, Wang Y, Li J, Hamer G. Trivial role for NSMCE2 during in vitro proliferation and differentiation of male germline stem cells. Reproduction. 2017 Sep;154(3):81-95.

Verver DE#, Zheng Y#, Speijer D, Hoebe R, Dekker HL, Repping S, Stap J, Hamer G. Non-SMC Element 2 (NSMCE2) of the SMC5/6 Complex Helps to Resolve Topological Stress. International Journal of Molecular Sciences. 2016 Oct 26;17(11). pii: E1782. (#equal contribution)

Mulder CL, Catsburg LAE, Zheng Y, de Winter-Korver CM, van Daalen SKM, van Wely M, Pals S, Repping S, van Pelt AMM. Long-term health in recipients of transplanted in vitro propagated spermatogonial stem cells. Human Reproduction. 2017 Nov 18:1-10. doi: 10.1093/humrep/dex348.

Zhang P, Chen X, Zheng Y, Zhu J, Qin Y, Lv Y, Zeng W. Long-term propagation of porcine undifferentiated spermatogonia. Stem Cells and Development. 2017 Aug 1;26(15):1121-1131.

Zhang P, Qin Y, Zheng Y, Zeng W. Phospholipase D family member 6 (PLD6) is a surface marker for enrichment of undifferentiated spermatogonia in the pre-pubertal boars. Stem Cells and Development. 2017 Nov 7. doi: 10.1089/scd.2017.0140.

Zheng Y, Zhang Y, Qu R, He Y, Tian X, Zeng W. Spermatogonial stem cells from domestic animals: progress and prospects. Reproduction. 2014 Feb 3;147(3):R65-74.

Zheng Y, He Y, An J, Qin J, Wang Y, Zhang Y, Tian X, Zeng W. THY1 is a surface marker of porcine gonocytes. Reproduction, Fertility and Development. 2014;26(4):533-9.

Zheng Y, Tian X, Zhang Y, Qin J, An J, Zeng W. In vitro propagation of male germline stem cells from piglets. Journal of Assisted Reproduction and Genetics. 2013 Jul;30(7):945-52.