Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. IDENTIFYING THE MALE MEIOTIC FUNCTIONS OF THE SPINDLE MATRIX PROTEIN MEGATOR A thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Biochemistry School of Fundamental Sciences Massey University, Palmerston North, New Zealand. Daniela Bianchi 2022 Abstract Male infertility is an emerging global concern and investigating the molecular events accountable for healthy sperm formation is crucial to the design of diagnostic assays and curative therapeutics. The work here proposed provides the first characterisation of the nucleopore protein Megator - the Drosophila homologue of the human Tpr - throughout the meiotic divisions responsible for sperm formation. Megator/Tpr performs highly conserved roles at the nuclear pore complex during in- terphase. In somatic cells mitosis is also part of the spindle matrix, a structure that surrounds the spindle and promotes accurate chromosome segregation as a crucial com- ponent of the spindle assembly checkpoint (SAC). Through genetic crossing, a combination of antibodies and fluorescent protein-expressing cells, Megator was found to localise in the spindle during meiosis I, consistent with forming a matrix. Moreover, Megator depletion by in vivo RNAi led to chromosome segregation defects suggesting loss of spindle assembly checkpoint function, as it has been seen in depleted mitotic cells. Remarkably, Megator was not detectable beyond background levels in meiosis II cells. Protein depletion in these cells induced abnormal chromatin masses and some cells were entirely devoid of nuclei. Examination of other spindle matrix proteins revealed that their distributions were altered by Megator de- pletion. These data suggest that Megator’s roles in male meiosis are semi-conserved with mitosis. ii Acknowledgements First, I would like to thank my supervisor Matthew Savoian for his microscope exper- tise. My gratitude goes also to the Manawatu Microscopy and Imaging Centre (MMIC) for the free use of all the microscopy equipment. I’m grateful to Dr Helen Fitzsimons for the use of her Drosophila Neurogenetics labo- ratory for the flies crossing, the donation of one of the Drosophila strains used in this study and, above all, for her kindness. Thank you to the lab technicians in the MMIC: Yaniu and Raoul. Raoul, thank you for all the time you helped me in setting up the microscopes, and Yanyu... I did really enjoy our moments of laughing and “girls time”. To all my friends in the statistics department whose academic achievements have been a great inspiration, thank you for your example. To my SFS colleagues, thank you for the time spent together ironically laughing over failed experiments or reassuring each other during these two tough years of research and pandemia. I’m grateful to my family who has supported me, despite being an ocean and twelve hours distant. Last but not least, I cannot thank my lovely husband Gabriele enough for his patience, love and encouragement. I couldn’t have done all of this without you. iii Coronavirus statement Due to the Covid-19 pandemic, Massey University closed from 23rd March 2020 to 18th May 2020 and 17th August to 13rd September 2021. This closure severely affected the research as not only it was not possible to produce any work during the lockdown, but any fly crossing and stock produced before the lockdown had to be destroyed. Once the university reopened, time was also lost to allow for the maintenance of the stock and the reset of the crossing needed for the experiments. The pandemic also severely affected the supply chain for months, preventing the deliv- ery of fundamental reagents such as transgenic flies and antibodies. When they arrived, some transgenic lines were dead or, after analysis through different microscopes, they had lost the transgene. Subsequent attempts to procure the flies were repeatedly unsuc- cessful. Further time was then lost when the fly stocks became ill and all experiments had to be halted until they fully recovered months later. Finally, along with the delays, the project encountered a host of technical challenges in the live cells analysis, fundamental for the achievement of objective 3. These unex- pected problems involved both the imaging conditions (radiation and cells becoming flattened) and drug efficacy (some drugs documented to be efficient in mitotic cells had no observable effect in meiotic cells). As a result of these challenges, part of objective 1 was unable to be completed as planned and objective 3 was removed completely. Here below a detailed description of how these objectives were affected: iv Objective 1: Due to both late delivery and time constraints, the examinations of the consequences of Megator depletion using additional independent RNAi strains were not performed. This was a key experiment as, despite the fly lines being previously validated, the use of additional RNAi lines would have identified if any of the observed phenotypes were due to off-target effects. Objective 3: Following Megator depletion, the study of Megator’s role in the spindle assembly checkpoint was entirely removed due to problems in the live cell imaging conditions. During the process of extraction from the testis, the spermatocyte cysts tend to disrupt upon manipulation. It has been observed that the disruption of the cyst affects the viability of the meiotic cells, resulting in low chances of cells going through both cell divisions. Also, despite the use of different culturing conditions, cells were constantly observed to flatten, resulting in the formation of aberrations in spindle morphology, making the examination of spindle and chromosome segregation defects impossible to be examined. v Table of contents Abstract ii Acknowledgements iii Coronavirus statement iv List of Figures ix List of Tables xi List of Abbreviations xii 1 Introduction 1 1.1 The impact of infertility . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Drosophila spermatogenesis: a model system for studying male fertility 3 1.3 The GAL4/UAS system in Drosophila melanogaster . . . . . . . . . . . 5 1.4 Overview of meiotic stages . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.4.1 Exceptional meiotic features in male Drosophila melanogaster . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.5 Mechanism and components of chromosome movements during cell division 12 1.5.1 Kinetochore functions: spindle attachment . . . . . . . . . . . . 13 1.5.2 Kinetochore functions: Spindle Assembly Checkpoint activity . 15 1.5.3 Kinetochore functions: generating or transducing the forces for chromosome segregation . . . . . . . . . . . . . . . . . . . . . . 18 1.6 The Spindle matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 vi 1.6.1 The conserved spindle matrix protein Tpr/Mlp1/Mlp2/Megator . . . . . . . . . . . . . . . . . . . . . 22 1.6.2 The spindle matrix protein Tpr during cell division . . . . . . . 26 1.6.3 The Drosophila spindle matrix protein Megator . . . . . . . . . 27 1.7 Megator’s role during meiosis . . . . . . . . . . . . . . . . . . . . . . . 29 2 Materials and Methods 31 2.1 Drosophila melanogaster strains . . . . . . . . . . . . . . . . . . . . . . 31 2.2 Fly strains maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.3 Genetic crosses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.4 Drosophila testes isolation and fixation . . . . . . . . . . . . . . . . . . 34 2.5 Immunostaining of Drosophila testes . . . . . . . . . . . . . . . . . . . 34 2.6 Microscopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 2.7 Quantification of Megator depletion. . . . . . . . . . . . . . . . . . . . 36 2.8 Counting of onion stage spermatids. . . . . . . . . . . . . . . . . . . . . 37 3 Results 38 3.1 Characterisation of Megator distribution during meiosis I and meiosis II 38 3.1.1 Validation of Megator’s localisation during meiosis: Chromator and Skeletor distributions . . . . . . . . . . . . . . . 41 3.1.2 Confirmation of Megator’s distribution during Meiosis II: Megator- mCherry distribution . . . . . . . . . . . . . . . . . . . . . . . . 45 3.2 Identifying Megator meiotic functions . . . . . . . . . . . . . . . . . . . 48 3.2.1 Quantifying Megator’s depletion by quantitative microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.2.2 Megator promotes matrix localisation of Chromator and Skeletor throughout meiosis . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.2.3 An examination of Megator’s relationship with the Drosophila’s Lamin B/Dm0 . . . . . . . . . . . . . . . . . . . . 56 3.2.4 Megator is required for accurate chromosome segregation in meiosis 60 vii 4 Discussion and future directions 67 4.1 The spindle matrix is a highly conserved feature of dividing cells . . . . 68 4.2 The spindle matrix composition is not conserved between mitosis and the two meiotic divisions . . . . . . . . . . . . . . . . . . . . . . . . . . 70 4.3 Megator may be a master controller of spindle matrix composition . . . 71 4.4 The lack of a Megator-defined spindle matrix in meiosis II is in contrast with studies in mitosis . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 4.5 Megator absence at the nuclear rim at prophase II opens a question about the NPC composition during meiosis II . . . . . . . . . . . . . . 75 4.6 Megator may regulate correct chromosome segregation by playing a con- served role in the Spindle Assembly Checkpoint . . . . . . . . . . . . . 76 4.7 Megator may control correct chromosome segregation by playing a role in spindle morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 4.8 Megator is predicted to affect male fertility . . . . . . . . . . . . . . . 80 4.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Bibliography 82 Appendix 116 viii List of Figures 1.1 Overview of spermatogenesis. . . . . . . . . . . . . . . . . . . . . . . . 4 1.2 The GAL4/UAS system and in vivo gene knockdown in Drosophila melanogaster. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.3 Schematic representation of meiotic stages. . . . . . . . . . . . . . . . . 10 1.4 Spindle and kinetochore architecture during cell division. . . . . . . . . 13 1.5 Schematic representation of kinetochore-microtubules attachment dur- ing cell division. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 1.6 Schematic representation of the spindle assembly checkpoint. . . . . . . 17 1.7 The poleward forces generated during anaphase. . . . . . . . . . . . . . 19 1.8 Chromosome movement after microtubules severing. . . . . . . . . . . . 21 1.9 The nucleopore complex and Tpr architecture. . . . . . . . . . . . . . . 24 2.1 Schematic representation of Drosophila crossing performed in the study. 33 3.1 Megator shows distinct distribution throughout meiosis I and meiosis II. 39 3.2 Chromator and Skeletor are spindle matrix proteins during meiosis I and II. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.3 Megator-mCherry localisation during MI and MII confirm Megator an- tibody distributions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3.4 Megator shRNA expressing cells have significantly reduced Megator pro- tein fluorescence’s depletion efficiency. . . . . . . . . . . . . . . . . . . . 50 3.5 Megator shRNA efficiently depletes Megator in MI cells. . . . . . . . . 51 3.6 Megator depletion decreases the recruitment of the spindle matrix pro- teins Chromator and Skeletor. . . . . . . . . . . . . . . . . . . . . . . . 54 ix 3.7 Megator depletion induces Lamin Dm0 mislocalisation during meiosis. . 58 3.8 Onion Stage spermatids reveal multiple defects in Megator depleted cells. 61 3.9 Megator is crucial for correct chromosome segregation in both meiosis I and meiosis II. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 x List of Tables 2.1 List of primary antibodies and corresponding dilutions. . . . . . . . . . 35 2.2 List of secondary antibodies and corresponding dilutions. . . . . . . . . 35 3.1 Fluorescence intensity is significantly reduced in the spindles of Megator depleted cells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.2 Megator depletion leads to high rates of meiotic defects. . . . . . . . . 62 A.1 List of genotypes and sources of Drosophila melanogaster fly lines used in this study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 A.2 Total fluorescence counting in the spindle/cytoplasm regions of control and Megator depleted cells. . . . . . . . . . . . . . . . . . . . . . . . . 123 A.3 Total counting of nebenkern to nuclei ratio in the onion stage spermatids in the homozygous parental lines and progeny. . . . . . . . . . . . . . . 126 xi List of Abbreviations APC/C Anaphase-promoting complex/cyclosome Apc2 Adenomatous Polyposis Coli tumor suppressor homolog 2 Bam Bag-of-Marbles-protein BDC Bloomington Drosophila Center Bub1 Budding uninhibited by benzimidazole 1 CCAN Centromere-associated network proteins Cdc20 Cell Division Cycle 20 Cdk Cyclin Dependent Kinase CENP-A Centromere protein A Chro Chromator CID Centromeric histone H3-like protein identifier Covid-19 Coronavirus disease Crm1 Chromosomal maintenance 1 CySCs Somatic cyst stem cells DIC Differential Interference Contrast microscopy DNA Deoxyribonucleic acid dsRNA Double Strand RNA EGFP Enhanced Green Fluorescence Protein ERK Extracellular signal-Regulated Kinase FG-Nups Phenylalanine-Glycine enriched nups F1 First generation G1 Gap 1 xii G2 Gap 2 GSCs Germline Stem Cells GTPase Guanosine Triphosphatase HGPS Hutchinson-Gilford Progeria Syndrome KD Knock Down kDa Kilodalton KMN network Knl1/Mis12 complex/Ndc80 complex K-fiber Kinetochore fibres KLP10A Kinesin Like Protein 10A KLP59C Kinesin Like Protein 59C kMTs Kinetochore microtubules KPNA2 Karyopherin alpha 2 Mad1 Mitotic Arrest Deficient protein 1 Mad2 Mitotic Arrest Deficient protein 2 MAPK Mitogen-Activated Protein Kinase MCC Mitotic Checkpoint Complex Mlp1 Myosin-like protein 1 Mlp2 Myosin-like protein 2 MTs Microtubules MTOCs Microtubule Organizing Centers Mtor Megator MI First meiotic division MII Second meiotic division NEB Nuclear Envelope Breakdown NLS Nuclear Localization Signal NPC Nuclear Pore Complex Nups Nucleoporins P1 Parental line RISC RNAInduced Silencing Complex xiii RNAi RNA interference ROI Region of interest SAC Spindle Assembly Checkpoint SAFs Spindle Assembly Factors SEM Surface-epitope masking shRNA Short hairpin RNA siRNA Small interfering RNA Skel Skeletor SMC1 Structural Maintenance of Chromosome protein 1 SPB Spindle Pole Body TFR Total Fertility Rate Tpr Translocated Promoter Region UAS Upstream activation sequence UV ultraviolet VDRC Vienna Drosophila Resource Center WHO World Health Organization 1n haploid cell 2n diploid cell + plus ends microtubules - minus ends microtubules xiv Chapter 1 Introduction 1.1 The impact of infertility Industrial revolutions, medical and agricultural progress have always contributed to the world’s population growth. However, in the past few decades the rise of infertility across the globe has changed the scenario (Roser, 2014). Fertility is defined as the ability to initiate a clinical pregnancy, while infertility is the couple’s incapability to do so for more than 12 months (Zegers-Hochschild et al., 2017). In mammals, a pregnancy is established after fusion of male and female gametes: sperm and egg cells, respectively. The merging of these haploid cells (fertilisation) results in the formation of a cell with the full diploid complement (2n) of chromosome, characteristics of humans, most mammals and many animals. Fertility is measured by the Total Fertility Rate (TFR), which counts the number of children per woman. A low TFR has an impact on the economy, as the reduction in childbirths creates countries demographically older and less economically competi- tive (International Strategic Analysis, 2019). The World Health Organisation (WHO) has included infertility in the list of the diseases that cause disability (Burns, 2007). Furthermore, it has also established a minimal 2.1 TFR to guarantee a successful re- placement, which is defined as the population’s ability to replace itself from an existing generation to the next one, without migration. A recent study estimated that in the 1 CHAPTER 1. INTRODUCTION 2 next 80 years there will be a lower TFR than the replacement rate for 183 countries (Vollset et al., 2020). Some of these, such as Japan, have already experienced a low TFR for decades (World Bank Data, 2019), while other countries are facing the issue for the first time after years of demographic expansion. New Zealand, for example, has recently reached its lowest TFR to date, 1.6 (StatsNz, 2021). Based on TFR data, it was calculated that infertility affects almost 48.5 million couples (Ombelet et al., 2008; Mascarenhas et al., 2012; Agarwal et al., 2015). Aging, especially for women, combined with delays in choosing to have children is one of the reasons for couple infertility (Battaglia et al., 1996; Kidd et al., 2001). However, infertility can be linked to factors which negatively impact male and female gametogenesis (Vander Borght and Wyns, 2018). Therefore, both men and women can contribute to infertility (Turner et al., 2020). Agarwal et al. (2015) have indeed determined that males are responsible for 20− 30% infertility cases worldwide. Male fertility and formation of genetically stable sperm (spermatogenesis) requires two specialised and sequential cell divisions, termed meiosis I (MI) and meiosis II (MII), followed by a complex postmeiotic development pathway (Holstein et al., 2003). When errors occur during meiotic chromosome segregation, the daughter cells can receive an improper number of chromosomes (aneuploidy), which can cause various degrees of infertility or results in genetically abnormal progeny post fertilisation (Boveri, 1902; Bond, 1987; Nagaoka et al., 2012). Due to the great socio-economic burden caused by male infertility, a deeper under- standing of the molecular events responsible for male meiosis is needed. These data will provide insights for the development of target therapies or prognosticative assays that will help tackle the problem of low global fertility rate. CHAPTER 1. INTRODUCTION 3 1.2 Drosophila spermatogenesis: a model system for studying male fertility Humans and flies share most of the key developmental stages occurring during sperm formation, or spermatogenesis (Preston et al., 2019) (Figures 1.1 A and B). Therefore, the fruit fly Drosophila melanogaster is one of the most used model systems (Bonilla and Xu, 2008). These insects are easily genetically manipulated and their short repro- ductive cycles create a high number of offspring within a few weeks. Additionally, the Drosophila has only four chromosomes: chromosome I as sex chromosome, two auto- somes (chromosomes II and III) and a small chromosome IV which contains few genes. This limited Drosophila karyotype provides easier analyses of chromosome behaviour during cell division. Drosophila spermatogenesis occurs in the testis, a 2 mm long tube with a diameter of approximately 100 microns (Bairati 1967 Figure 1.1 D). Each male Drosophila fly contains two coiled testes which are connected to the seminal vesicles and the ejac- ulatory duct (Figure 1.1 C). During Drosophila spermatogenesis, the involved cells progressively differentiate and migrate from the apical tip to the end of the testis, which facilitates their identification (Fuller, 1993) (Figures 1.1 D and F). The Drosophila testis contains two different classes of stem cells: the germline stem cells (GSCs) and the cyst somatic cells (CySCs) (Bairati, 1967). The GSCs are surrounded by two CySCs and both these types of stem cells are maintained and anchored to the testis’s apical tip by the hub cells, a cluster of non-dividing cells (Hardy et al., 1979). The spermatogenesis begins when GSCs start to divide. CHAPTER 1. INTRODUCTION 4 Figure 1.1: Overview of spermatogenesis. Similarities between mammalian (A) and Drosophila (B) spermatogenesis (original drawings, generated with reference to Preston et al. (2019)). Schematic representation of male Drosophila reproductive system (C), Drosophila testis (D) and mature Drosophila sperm (E). Detailed diagram of Drosophila melanogaster spermatogenesis (F). GsCs is the germline stem cells and CyScs is cyst somatic stem cells. This process will generate two different types of cells. The first one will remain at- tached to the hub and will keep the stem cells feature (such as self-renewal Cuevas and Matunis 2011), the other sister cell – called a gonialblast - will start to differentiate asymmetrically through four cycles of mitosis to create cyst of 2, 4, 8 and then 16 interconnected spermatogonial cells, or primary spermatocytes. At the end of mitosis, CHAPTER 1. INTRODUCTION 5 the primary spermatocytes undergo a single round of DNA replication and enter the first of the two meiotic divisions (Cenci et al., 1994). After two cycles of meiosis (MI and MII), these now haploid (1n) spermatocytes dif- ferentiate into 64 individual ’onion-stage’ spermatids interconnected by ring canals. Onion stage spermatids are characterised by a nucleus and equally sized round mito- chondrial derivatives, called a nebenkern (Tates, 1971). The ratio nebenkern to nuclei is 1:1 and variations in the ratio can be used as read outs of meiotic errors (González et al., 1989). Following extensive morphological changes, such as nuclear elongation and tail formation, the 64 spermatids mature into spermatozoa (Tokuyasu et al., 1974). During sperm individualisation, the last step of spermatogenesis, each spermatozoon become fully independent and migrates to the seminal vesicle, where it will be stored until fertilisation (Fuller, 1993; Tokuyasu et al., 1972a,b). If properly formed, a mature sperm is constituted by an elongated head, a flagellum or sperm tail, and a basal body connecting the two (Fabian and Brill, 2012), (Figure. 1.1 E). The head restrains the nucleus and the acrosome. While the latter is important for the fusion with the female gamete during fertilisation, the nucleus carries the genetic information or DNA. The sperm tail is essential for sperm movement and contains an axoneme core that extends from the basal body (Alberts et al., 2002; Mohri et al., 2012). 1.3 The GAL4/UAS system in Drosophila melanogaster Drosophila melanogaster is a powerful model system to study gene expression and function due to its ability of being easily genetically manipulated through a large range of techniques. These have allowed the production of engineered mutants and transgenic fly lines, many of which are available from stock centres such as Vienna Drosophila Resource Center (VDRC), Kyoto Drosophila Genetic Resource Center and Bloomington Drosophila Center (BDC). The GAL4/UAS system allows the expression or “driving” of an engineered target gene in a specific tissue or cell-type (Brand and Perrimon, 1993). This system is based on CHAPTER 1. INTRODUCTION 6 the GAL4 yeast transcriptional activator under the control of an upstream activation sequence (UAS) and a target gene with a promoter containing a GAL4 binding site (Fischer et al., 1988). Depending on the absence or presence of its activator, the gene of interest will remain silent or get activated. The use of this system requires the creation of two lines of transgenic flies. One of these lines carries the GAL4 gene fused to a tissue specific enhancer, also called a GAL4 driver, the other line contains the UAS fused to the transgene of interest. The crossing of these two transgenic parental lines (P1) yield progeny (F1) that will express the GAL4 gene. This will bind to the UAS regulatory region, inducing the expression of the gene of interest (Figure 1.2 A). The GAL4/UAS system is a versatile technique capable not only of driving the expres- sion of a specific transgene, but also its tissue specific mediated knockdown. Indeed, it is an extremely valuable tool for the functional analysis of any unknown gene involved in a particular process in which its mutants are lethal (Qi et al., 2004). For this pur- pose, it is sufficient to combine the GAL4/UAS system’s properties to the endogenic gene silencing pathway present in many eukaryotic cells and termed RNA interference (RNAi) (Fire et al., 1998; Paddison et al., 2002). This pathway is normally triggered by the production of double stranded RNA (dsRNA) after either bidirectional tran- scription or transcription of an inverted repeated sequence. The dsRNA formed is cut into small interfering RNAs (siRNAs) which are ultimately degraded by the enzyme Dicer and by the RNA-induced silencing complex (RISC) (Hannon, 2002; Wilson and Doudna, 2013). Therefore, in the GAL4/UAS mediated knockdown, the expression of a transgene specific inverted repeated sequence will create short hairpin RNAs (shRNAs) that will be cleaved by the Dicer machinery, inducing the transgene silencing (Figure 1.2 B). CHAPTER 1. INTRODUCTION 7 Figure 1.2: The GAL4/UAS system and in vivo gene knockdown in Drosophila melanogaster. The GAL4/UAS system allows the expression (A) and the silencing (B) of a desiderable transgene in a tissue-specific manner. In the parental lines (P1) Drosophila virgin female flies carrying the GAL4 gene fused downstream to a tissue specific enhancer (GAL4 driver) are crossed to male flies carrying an Upstream activation sequence (UAS) fused to the sequence of a target gene (A) or an inverted repeat sequence of the transgene (B). In progeny flies (F1), the product of GAL4 will bind to UAS, allowing the tissue specific expression of the transgene (A) or the formation of short hairpin RNAs for gene silencing (B). Original drawings generated with reference to Brand and Perrimon (1993) and Perrimon et al. (2010). CHAPTER 1. INTRODUCTION 8 1.4 Overview of meiotic stages Mitosis and meiosis are two types of division responsible for cell replication. While mitosis occurs in the somatic cells and ensures the production of two genetically iden- tical daughter cells, meiosis occurs in the germline and ultimately halves the number of chromosomes in the gametes. This meiotic reduction is achieved by two consequent rounds of chromosome segregation, termed meiosis I (MI) and meiosis II (MII), without an intervening DNA replication event (Figure 1.3). While MII is similar to mitosis, MI is distinctive to the germ cells. All divisions are composed by different stages and their onset or phase-transition are mediated by fluctuation in the levels of cyclin and their Cdk1 complexes (Fisher et al., 2012; Webster et al., 2019). Meiosis I starts with prophase I, which is characterised by extensive changes in the cells. Striking among these is the reorganisation of the microtubule (MTs) cytoskele- ton. Unlike interphase, during early stages of division, these polymers of α- and β- tubulin are found exclusively radiating from the two centrosomes, the dominant sites of nucleation within the cell. The two centrosomes migrate around the still intact nuclear envelope to opposite sides through the actions of motor proteins (Whitehead et al., 1996; Gönczy et al., 1999; Robinson et al., 1999; Sharp et al., 1999, 2000; Mitchison et al., 2005). During prophase I the DNA condenses into homologous chromosomes that become paired. Each condensed chromosome consists of four chromatids, two sisters from the replicated maternal chromosome and two from the father. Like mitosis, cohe- sion is responsible for adhesion between sister chromatids (Watanabe and Nurse, 1999; Buonomo et al., 2000; Watanabe, 2005; Nasmyth and Haering, 2009; Ishiguro et al., 2011), while in many systems DNA recombination is responsible for pairing between the homologoues (Von Wettstein, 1971; Webber et al., 2004). In addition to compact- ing, each homologous chromosome begins to assemble a pair of kinetochores, protein super-complexes positioned at the two opposingly positioned centromeres. Due to their diverse composition, kinetochores perform the key events of division and are responsible CHAPTER 1. INTRODUCTION 9 for chromosome movement as well as for cell cycle exit (described in section 1.5.3). As cells transit into prometaphase I, microtubules radiating from the two centrosomes penetrate into the nuclear space in the vicinity of the chromosomes. At prometaphase I, as MTs contact the kinetochores, they are “captured” leading to attachment of the kinetochore/chromosome and spindle formation (Kirschner and Mitchison, 1986; Hayden et al., 1990; Rieder and Alexander, 1990). The opposing placement of the two MTs-nucleating centrosomes along with that of each bivalents’ centromere-bound kine- tochores drives the formation of a bipolar spindle, capped at each end by a centrosome. Each kinetochore progressively acquires more MTs from its closest centrosome, leading to a robust bundle of kinetochore fibre that exclusively attaches each kinetochore in the bivalent to one pole (Holy and Leibler, 1994; Khodjakov et al., 2003; Maiato et al., 2004). This leads to a “tug-of-war” of forces towards each spindle pole as chromosomes “congress” to the spindle equator (Kapoor et al., 2006; Cai et al., 2009). Alignment of all of the chromosomes at the equator marks metaphase I. If all of the kinetochores have become properly attached to the spindle, cells satisfy the spindle assembly checkpoint (SAC), a regulatory system which halts the cell cycle in case of improper attachment, and advances into anaphase I (described in section 1.5.2). At onset of anaphase I, the cohesin complexes are removed only from the chromosome arms (Waizenegger et al., 2000; Watanabe, 2005). Therefore, the chromosomes are pulled away from each other (anaphase A) and then polewards, allowing spindle elongation (anaphase B). The last two stages of meiosis I are telophase I and cytokinesis I. During telophase I chromosomes decondense and nuclei reform. At cytokinesis I the cytoplasm is also partitioned through the formation and constriction of the cytoplasm by an acto-myosin contractile ring that assembles at the spindle equator (Matsumura, 2005). Thus, at the end of MI, the two daughter cells carrying half of the original genetic content are ready to enter the second meiotic division. Meiosis II repeats the same events as meiosis I. During prophase II the chromatin CHAPTER 1. INTRODUCTION 10 condenses into distinct chromosomes each of which now consists of two sister chro- matids. At prometaphase II, spindle formation occurs allowing the chromosomes to move as kinetochores interact with microtubules. Their congression to the equator identifies the beginning of metaphase II. During anaphase II, the remaining cohesin complexes are removed allowing for chromatid separation and segregation (Watanabe and Nurse, 1999). This is followed by chromosome decondensation in telophase II and the formation of daughter nuclei with the subsequent cleaving of the cytoplasm during cytokinesis II. The net result of MI and MII is the formation of four haploid daughter cells which will ultimately be transformed into eggs or sperm, depending on the tissue. Figure 1.3: Schematic representation of meiotic stages. Homologous chromosomes are separated during meiosis I (MI) and chromatids are separated in meiosis II (MII). See text for more info. G1 is for Gap1 or cellular growth phase. S represents the synthesis phase where the genome is replicated. G2 or Gap2 is the second cellular growth phase, needed before meiosis (M).This cartoon represents stages of generic (not male Drosophila flies) meiosis. CHAPTER 1. INTRODUCTION 11 1.4.1 Exceptional meiotic features in male Drosophila melanogaster As described in section 1.2, meiosis in male Drosophila flies begins after the production of cyst of 16 interconnected primary spermatocytes. After the last cycle of mitosis, DNA replication occurs for 3 hours, followed by an extensive growth period (G2) of almost 80 hours, during which the cells increase their volume and produce the cellular content needed for the progression into spermatogenesis (Cenci et al., 1994). For almost the entirety of prophase I, the DNA is decondensed and chromatin masses occupy the nuclear lumen. Homologous chromosomes are believed to be paired in some way during early prophase I, but the mechanisms responsible for pairing remain subject to debate (Vazquez et al., 2002). By contrast, it is well established that recombination and, by extension, crossing over events or chiasma do not occur in male Drosophila flies (Morgan, 1914; Hawley, 2002). Furthermore, the “conjunction” of the homologoues may be genetically controlled (Thomas et al., 2005). The cohesin complexes in Drosophila, which are vital for chromosomes (during MI) and chromatids (in MII) segregation, do not contain the specific and conserved mei- otic core subunit Rec8. Instead, Drosophila appears to rely on two unique cohesion proteins (McKee et al., 2012; Yan and McKee, 2013). In addition, male meiosis in Drosophila, like most insects, is characterised by having a semi-open division. Probing by centrosome-derived microtubules (for kinetochore interaction at prometaphase) oc- curs through large holes in the envelope that occur near the centrosomes. This differs from many other animals, including mammals, where the nuclear envelope completely disintegrates during spindle formation. In Drosophila the residual envelope membranes surround the spindle as a “spindle envelope” that may serve to concentrate molecules within the spindle region (Yao et al., 2012). CHAPTER 1. INTRODUCTION 12 1.5 Mechanism and components of chromosome movements during cell division Successful chromosome segregation in cell division occurs through the regulated in- terplay of a spindle and chromosome-bound kinetochores (Rieder and Salmon, 1998) (Figures 1.4 A and B). Microtubules are inherently dynamic structures that can grow when subunits of tubu- lin are added (polymerisation) or shrink when removed (depolymerisation) (Gorbsky et al., 1988; Mitchison, 1989; Zhai et al., 1995). Microtubules within the spindle have two ends: the ones which protrude away from their nucleation site are called plus ends (+), and the others are the minus ends (-). In cells, MTs are nucleated from the micro- tubule organising centres (MTOCs), commonly the centrosomes, organelles consisting of centrioles and an amorphous cloud of nucleating material (Bornens and Azimzadeh, 2007; Blachon et al., 2008; Kim et al., 2013; Arquint and Nigg, 2016). Three classes of microtubules are present in the spindle: astral microtubules which extend from the centrosomes to the cell cortex, interpolar microtubules that radiate towards the opposite pole, sustain the spindle structure and, through interaction, push the spindle poles apart during anaphase B. The last class of spindle microtubules are called kinetochore microtubules (kMTs) and are organised into bundles or fibres (k- fiber) which are bound by kinetochores, (Figure 1.4 A). Kinetochores are key determinants of cell division. They perform three major functions: (i) attachment to the spindle, (ii) generating or transducing the forces for chromosome movement and (iii) serving as a catalytic platform for the spindle assembly check- point (SAC). These functions are possible due to its diverse composition and structure (De Wulf et al., 2003; Przewloka and Glover, 2009; Akiyoshi and Gull, 2014) (Figure 1.4 C and D). Traditionally kinetochore has been defined as having an inner and outer plate enriched in over 100 proteins of different classes (Musacchio and Desai, 2017) (Figure 1.4 B). CHAPTER 1. INTRODUCTION 13 The inner plate - composed of the centromere-associated network (CCAN) - binds centromeric chromatin proteins, or nucleosomes, which are characterised by unconven- tional histones (Earnshaw and Rothfield, 1985; Mellone et al., 2011). The kinetochore’s outer plate - generally called the KMN network (Knl1/Mis12complex/Ndc80) - binds the plus ends kMTs and recruit the SAC proteins (Cheeseman et al., 2006). Figure 1.4: Spindle and kinetochore architecture during cell division. Three classes of microtubules (MTs) radiate from the centrosome to assemble the bipolar spindle (A). The kinetochore microtubules (kMTs) bind the centromere region of the chromosomes through the kinetochore, a complex structure of inner and outer plate of different classes of proteins (B). Examples of kinetochore compositions: humans (C) and Drosophila (D). Original drawings created with reference to Singleton (2016) and Musacchio and Desai (2017). 1.5.1 Kinetochore functions: spindle attachment In centrosome-containing cells such as those of the male germline and soma, spindle formation initiates at prometaphase. Dynamic astral MTs probe the cell until contacting a kinetochore (Kirschner and Mitchison, 1986). Light and electron microscopy studies have shown that the first CHAPTER 1. INTRODUCTION 14 contacts occur laterally with a single MT contracting and extending beyond kineto- chore (Hayden et al., 1990; Merdes and De Mey, 1990; Rieder and Alexander, 1990; Tanaka et al., 2005). The presence of the minus end directed motor dynein on the kinetochore causes rapid gliding of the chromosome towards the centrosome (Rieder and Khodjakov, 2003). Attachment to both centrosomes and biorientation requires that the chromosome move away from the bound centrosome or spindle pole, towards the unattached centrosome. This is accomplished by a variety of mechanisms all of which utilise MT plus end di- rected kinesin (Desai et al., 1999; Mimori-Kiyosue et al., 2005; Manning et al., 2007; Brouhard et al., 2008; Al-Bassam et al., 2010; Walczak et al., 2013). This facilitates MT capture by the unattached kinetochore and a proper attachment of each kineto- chore to an opposing centrosome or spindle pole. Attachment of each kinetochore in a bivalent (MI) or sister chromatid (MII, mitosis) is necessary for equal DNA distribution (Goldstein, 1981; Lee et al., 2000; Petronczki et al., 2003; Corbett et al., 2010). Although the layout of the kinetochores and bipolar nature of the spindle help to ensure the exclusive attachment of each kinetochore to a single and opposing spindle pole (amphitelic attachment, Figure 1.5 A) errors can still occur. These aberrant attachments are called syntelic, monotelic and merotelic. Syntelic at- tachments occur when both kinetochores are bound to the same pole (Figure 1.5 B). In monotelic attachments, just one kinetochore is linked to one pole (Figure 1.5 C), while in merotelic attachment a single kinetochore is attached to both poles (Figure 1.5 D). Anaphase onset in the presence of such errors would lead to genetic instabil- ity. Accordingly, a highly conserved surveillance system dubbed “the spindle assembly checkpoint” (SAC) has evolved to inhibit anaphase onset. The mechanisms that perceive and correct the improper kinetochore-microtubules at- tachment have been the subject of much investigation. Studies in mantids (Li and Nicklas, 1995, 1997) and Drosophila spermatocytes (Chaurasia and Lehner, 2018) re- vealed that at the most fundamental level, it is driven by mechanical tension sensing: CHAPTER 1. INTRODUCTION 15 a lack of tension on kinetochores, as it occurs during most malorientations, prevents anaphase onset until it is correct. However, merotelic malorientations, which introduce tension, are invisible to the SAC and a prominent cause of aneuploidy (Cimini et al., 2001, 2002). More recent studies have identified the molecular players and events which this pathway operates (Tanaka et al., 2002; Gruneberg et al., 2004; Kapoor et al., 2006; Liu et al., 2009). Figure 1.5: Schematic representation of kinetochore-microtubules attachment during cell division. During amphitelic attachment each kinetochore is bound to kineto- chore microtubules nucleated from each pole (A). In syntelic attachment both kinetochores are bound to the same pole (B). During monotelic attachments a single kinetochore is bound to one pole (C). In merotelic attachments a kinetochore is bound to both poles (D). 1.5.2 Kinetochore functions: Spindle Assembly Checkpoint activity The spindle assembly checkpoint (SAC) is a surveillance system that ensures genomic stability in the daughter cells during cell division (Musacchio and Salmon, 2007; Foley and Kapoor, 2013; Sacristan and Kops, 2015). First discovered in yeast, it is highly conserved across eukaryotes (Musacchio and Hardwick, 2002; Musacchio and Salmon, 2007; Kops et al., 2020). It eludes the premature disjoining of the chromosomes, thus aneuploidy, by preventing the activation of the Anaphase Promoting Complex/Cyclo- some (APC/C), a multi-subunit ligase that leads to the degradation of multiple targets such as Securin for cohesion removal and the cyclins thereby leading to chromosomes segregation and cell division exit (Musacchio and Salmon, 2007). CHAPTER 1. INTRODUCTION 16 Under normal conditions, the presence of a single unattached kinetochore is sufficient to trigger a robust response of the SAC which induces the so-called “wait anaphase” signal, until correction of the improper kinetochore attachments (Rieder et al., 1995). In agreement with this, disruption of the spindle leads to prolonged arrest (Hoyt et al., 1991; Rieder and Maiato, 2004), while loss of SAC protein function leads to exit and aneuploidy (Hardwick et al., 1999; Shonn et al., 2003). The precise mechanism by which the SAC operates remains to be elucidated. Tanaka et al. (2002) and Liu et al. (2009) have shown that these mechanical tensions involve Au- rora B, a serine-threonine kinase found at the centromeres from prophase to metaphase (Gruneberg et al., 2004). Unattached or improperly attached centromeres lack tension, which places Aurora B proximal to the kinetochore. This position triggers the Aurora B kinase dependent phosphorylation of KMN network proteins with a consequent loss of microtubules binding. By contrast, bipolar amphitelic attachment generates tension that makes Aurora B kinase distal to the kinetochore and unable to phosphorylate key sites needed for MTs release. In addition, this tensionless placement of Aurora B makes it available to phosphorylate other substrates. Unattached kinetochores or attachments that fail to generate tension triggers the re- cruitment of specific SAC proteins to form the APC inhibiting MCC (mitotic check- point complex). This complex is constituted by the Cdc20 (Cell division cycle 20), Mad2 (mitotic-arrest deficient 2) and the Bub proteins (budding uninhibited by ben- zimidazole) Bub3 and BubR1 (Sudakin et al., 2001). Its formation is catalysed by Mad1, another SAC protein, that recruits Mad2 at the kinetochore for SAC signal amplification (Mapelli and Musacchio, 2007). The amplification is also triggered by the Aurora B kinase through mechanical kMts tension signaling (Tanaka et al., 2002; Liu et al., 2009) and depends on the recruitment of the SAC protein Mps1 (Monopolar Spindle 1), which phosphorylates the KNL1 proteins present at the outer plate of the kinetochore (Saurin et al., 2011). CHAPTER 1. INTRODUCTION 17 After the MCC assembly, the complex inhibits the APC/C making it inactive and un- able to add polyubiquitin chains to both Securin and Cyclin B for their degradation by the proteasome (Fang et al., 1998; Wassmann and Benezra, 1998; Morrow et al., 2005). This results in the Securin being linked to the protease Separase which is unable to remove the cohesin complex that held the chromatids, thus preventing their disjunc- tion (Thornton and Toczyski, 2004) (Figure 1.6 B). By contrast, when the chromosome biorientation and kMTs tension are established, the MCC is shed. Therefore, Cdc20 is able to bind the APC/C, triggering its activation. Securin and Cyclin are then de- graded which promote the chromatid disjunction and anaphase transition (Figure 1.6 A). Figure 1.6: Schematic representation of the spindle assembly checkpoint. When sister kinetochores become bioriented, the mitotic checkpoint complex (MCC) is removed (A). Cdc20 binds and activates the Anaphase Promoting complex/Cyclosome (APC/C) with consequent Securin and Cyclin B ubiquitination. These events promote chromatid disjunction upon removal of the cohesin (black circle), and anaphase transition. Unattached kinetochore (B) triggers the recruitment of SAC proteins to form the MCC. MCC inhibits the APC/C and prevents consequent degradation of Separase and CyclinB. In this state, the cell cycle is halted until errors in attachments are corrected. Original drawing generated with reference to Musacchio and Hardwick (2002) . CHAPTER 1. INTRODUCTION 18 Most of the SAC proteins discovered in mitosis are conserved in many eukaryotes and their localisation/functions have been studied for decades (Musacchio and Hardwick, 2002; Vleugel et al., 2012). By contrast, SAC proteins in meiosis have been studied just recently. In budding yeast, the loss or disruption of Mad1 and Mad2 during meio- sis leads to high rates of chromosome segregation defects, when compared to mitosis (Hardwick et al., 1999; Shonn et al., 2003). Furthermore, time-lapse microscopy anal- ysis and MTs depolymerisation experiments by Savoian et al. (2000) and Rebollo and González (2000) revealed complete segregation failure, proving for the first time the activity of a meiotic spindle checkpoint in Drosophila. Furthermore, the meiotic spindle checkpoints are generally believed to be feeble (Malmanche et al., 2006; Buffin et al., 2007; Gorbsky, 2015). Therefore, in Drosophila spermatocytes the meiotic SAC activ- ity operates with less efficiency or be regulated by different mechanisms than mitosis (Church and Lin, 1988). 1.5.3 Kinetochore functions: generating or transducing the forces for chromosome segregation The molecular details that dictate chromosome segregation are not fully understood. Through studies in mitosis, what is known is that the chromosome movements are the results of forces generated by different elements. Due to the binding and insertion of the kMT plus ends into the kinetochore’s outer plate (Goldstein, 1981; Church and Lin, 1988), poleward movement requires MT depolymerisation. CHAPTER 1. INTRODUCTION 19 Figure 1.7: The poleward forces generated during anaphase. Anaphase is char- acterised by the disjunction of the chromatids and their movement polewards through the kinetochore-bound microtubules (A). In the Pacman model, the kinesins 13 (or KLP59C in Drosophila) trigger the depolymerisation of the kMTs plus-ends, while the motor protein dynein generates the pulling forces on chromosomes by walking towards the kMTs minus ends (B). In the Poleward flux model: the kMTs’ minus ends at the spindle poles are de- polymerised by the kinesin 13, or by the kinesin like KLP10A in Drosophila (C). Interpolar microtubules sliding through bipolar kinesins (D) and the action of chromokinesins (E) fa- vor the chromatids movement poleward and spindle elongation during anaphase B. Original drawings generated with reference to Rogers et al. (2004). At the onset of anaphase, the plus ends polymerisation activity decreases (Rogers et al., 2004) and while the kinesin 13 (KLP59C in Drosophila) trigger the kMts de- polymerisation, the minus end directed motor protein dynein “walks” on kinetochore microtubules, generating pulling forces on the chromosomes, (Pfarr et al., 1990; Steuer et al., 1990; Antonio et al., 2000; Li et al., 2007). This process is called the Pacman model (Cassimeris et al., 1987) (Figure 1.7 B). Perturbation of dynein in vertebrates CHAPTER 1. INTRODUCTION 20 (Vorozhko et al., 2008) and Drosophila embryos (Sharp et al., 2000) has displayed attenuation in the rate of chromosome poleward segregation during mitosis. This at- tenuation has been also indirectly observed in meiosis, using dynein binding mutant Drosophila spermatocytes (Savoian et al., 2000). During anaphase, microtubules also start to be shortened at their minus ends, through removal of tubulin subunits by kinesin 13 (KLP10A in Drosophila) present at the poles, in a process called Poleward flux (Figure 1.7 C). Photobleaching studies in Drosophila spermatocytes have shown that Flux-Pacman model is conserved in meiosis (Gorbsky et al., 1988; Sharp et al., 2000; LaFountain Jr et al., 2004; Savoian, 2015). Other forces that assist chromosome movement are generated by the crosslink and consequent sliding of the interpolar microtubules through bipolar kinesins (Figure 1.7 D) and the binding of the interpolar microtubules with chromosome-associated kinesins (Figure 1.7 E). These events mark the beginning of spindle elongation or anaphase B. 1.6 The Spindle matrix Several pieces of data have suggested that the spindle does not act as an independent or isolated cell division structure. Electron microscopic analysis of isolated sea-urchin mitotic spindles showed the presence of additional components that are different to spindle microtubules, and which appear to remain in their absence and help maintain spindle structure (Goldman and Rebhun, 1969; Leslie et al., 1987). In diatoms, addi- tional components were able to aid chromosome segregation even after microtubules depolymerisation by colchicine (Pickett-Heaps et al., 1980). Such observations point toward the existence of a spindle matrix, a structure surrounding the spindle that could interact with microtubules and, if disrupted, compromise the spindle function (Pickett-Heaps et al., 1984, 1996). Microsurgery studies using ultraviolet microbeams in mitotic newt fibroblasts (Spurck et al., 1997), Drosophila S2 cells (Maiato et al., 2004) as well as meiotic crane fly sper- matocytes (Sillers and Forer, 1983; Forer et al., 1997, 2003) led to the hypothesis that CHAPTER 1. INTRODUCTION 21 the matrix could have elastic properties that affect chromosome segregation. In these latter studies, ultraviolet microbeams were used to sever a k-fiber immediately be- fore anaphase onset (Figure 1.8 A). Despite cutting their spindle attachments in half, the chromosome still segregated as the kinetochore associated microtubules “stubs” translocated polewards (Figure 1.8 B). Moreover, the velocity of the chromosome at- tached to the kinetochore stubs increased compared to the chromosomes with uncut k-fibers (Forer et al., 2015). An alternative explanation proposes that following kMTs severing, the motor protein dynein is recruited to the minus end of the kinetochore stubs where it interacts with adjacent non-kMTs to drag the kMTs stub poleward (Figure 1.8 C). Consistent with this, inhibition of dynein prevents kMTs stub movement (Sikirzhytski et al., 2014). However, these findings have been recently challenged (Forer et al., 2018). Thus, the role of the matrix in chromosome movement remains unclear. Figure 1.8: Chromosome movement after microtubules severing. Kinetochore mi- crotubules (kMTs) are severed by ultraviolet microbeam (UV) creating kinetochore stubs that are no longer attached to the spindle poles (A). In the spindle matrix model, the pole- ward movement of kinetochore stubs (and chromosomes), is promoted by forces generated by the spindle matrix, rather than microtubules (B). In the dynein model, this motor protein gathers to the minus end of the kinetochore stubs, interacts with near microtubules to propel the poleward chromosome movement (C). Original drawing generated with reference to Forer et al. (2018). CHAPTER 1. INTRODUCTION 22 The composition of the matrix is still elusive. Studies in different systems and cell types have revealed a host of molecules that collect around the spindle. Some spindle matrix proteins includes Lamin B (Hayes, 2006; Tsai et al., 2006), cell cycle regulators like Cyclin B and Polo kinase (Yao et al., 2018), actin and Titin (Janmey et al., 1995; Chang et al., 2005; Fabian et al., 2007a,b; Pickett-Heaps and Forer, 2009). Others are nucleoporins (Nups), proteins of the Nuclear Pore Complex (NPC) (Joseph and Dasso, 2008; Katsani et al., 2008; De Souza et al., 2009; Hashizume et al., 2010). In Drosophila, co-immunoprecipitation experiments have identified multiple interacting proteins subsequently shown be part of the spindle matrix: Megator (Lince-Faria et al., 2009), Chromator (Rath et al., 2004), Skeletor (Walker et al., 2000), EAST (Qi et al., 2005) and Asator (Qi et al., 2009). These proteins form a spindle-like structure during prophase which persists at metaphase, even after microtubules are removed by drug treatment. Moreover, their perturbations through RNAi mediated proteins depletions lead to malformed microtubules and chromosome missegregation (Walker et al., 2000; Rath et al., 2004; Qi et al., 2005; Ding et al., 2009; Lince-Faria et al., 2009). However, only one has been identified so far as a spindle matrix component across several specices: the nucleoporin Tpr/Mlp1/Mlp2/Megator (Cordes et al., 1997; Zimowska et al., 1997; Qi et al., 2004; De Souza et al., 2009). 1.6.1 The conserved spindle matrix protein Tpr/Mlp1/Mlp2/Megator Nucleoporins (Nups) are a highly conserved family of 30 different proteins that assemble into the well-characterised nuclear pore complex (Grossman et al., 2012) (Figure 1.9 A and B). In this complex they act as a selectively permeable barrier for the cargoes trafficking between nucleus and cytoplasm (Anderson and Hetzer, 2007; Wente and Rout, 2010; Zhang et al., 2010; Grossman et al., 2012). However, recent research has highlighted that nucleoporins have multiple functions such as gene regulation (Ishii et al., 2002; Brickner and Walter, 2004; Taddei et al., 2006), DNA repair (Galy et al., CHAPTER 1. INTRODUCTION 23 2004; Zhao et al., 2004) and cell division (Hashizume et al., 2010; Lussi et al., 2010; Mishra et al., 2010; Nakano et al., 2010; Wozniak et al., 2010; Salsi et al., 2014). Tpr (Translocated promoter region) is a highly conserved nucleoporin which has homo- logues in trypanosomes (Holden et al., 2014), budding yeast: Mlp1 and Mlp2 (Myosin like protein) (Kölling et al., 1993; Strambio-de Castillia et al., 1999), fission yeast: Nup211 and Alm1 (Jiménez et al., 2000; Bae et al., 2009; DeGrasse et al., 2009) , Arabidopsis: NUA (Nuclear pore anchor, Xu et al. (2007)), Aspergillus: An-Mlp1 (De Souza et al., 2009; De Souza and Osmani, 2009) and in Drosophila melanogaster : Megator (Zimowska et al., 1997; Qi et al., 2004; Lince-Faria et al., 2009). Tpr was first identified as an activator of the protooncogenes met, raf and trk (Park et al., 1986; Soman et al., 1991; Greco et al., 1992), but subsequent studies have shown that is a 265 kDa mammalian protein, found during interphase in the nuclear basket of the nucleopore complex (Cordes et al., 1997; Zimowska et al., 1997; Frosst et al., 2002; Krull et al., 2004; Hüve et al., 2008) and at the mitotic spindle during metaphase (Sauer et al., 2005; Lince-Faria et al., 2009). Tpr is composed of an α-helical coiled-coil NH2-terminal domain with almost 1,600- residues, and a non-coiled COOH-terminal domain enriched of acidic residues (Mitchell and Cooper, 1992; Hase et al., 2001). Its anchoring to the NPC is mediated by the Nup153 (Hase and Cordes, 2003) through the amino acids 436–606, while its nuclear localisation is mediated by the 1812-1867 residues in the Carboxyl-terminal domain (Cordes et al., 1998) (Figure 1.9 C). CHAPTER 1. INTRODUCTION 24 Figure 1.9: The nucleopore complex and Tpr architecture. The nuclear pore complex (NPC) comprises a central channel surrounded by multiple parts, commonly classified in three ring-like structures: the cytoplasmic ring, a central ring and nucleoplasmic ring, plus two additional structures: the cytoplasmic filaments and the nuclear basket (A). Color coded tables indicate the nucleoporins (Nups) that localise in these regions (B). Most of the Nups are followed by a number that states their molecular mass and the FG-Nups are Nups enriched in phenylalanine-glycine. The structure and protein interactions of the nucleoporin Tpr (Translocated promoter region) are shown in (C). NLS is the nuclear localisation signal, Crm1 is Chromosomal Maintenance 1 and KPNA2/Importin-β is Karyopherin subunit alpha 2/Importin-β. Original drawings generated with reference to Grossman et al. (2012) (A and B), Krull et al. (2004) and Snow and Paschal (2014) (C). CHAPTER 1. INTRODUCTION 25 Tpr is implicated in several functions. Consistent with being a nucleopore protein, it mediates protein import/export (Frosst et al., 2002; Shibata et al., 2002; Vomas- tek et al., 2008; Ben-Efraim et al., 2009; Snow et al., 2013). Tpr perturbation by specific knockdown or antibody injections leads to disruption in its ability to bind the Crm1 (Chromosomal Maintenance 1), a mammalian cargo export protein (Frosst et al., 2002). Furthermore, analysis of fibroblasts from patients affected by Hutchinson- Gilford Progeria Syndrome (HGPS) shows defects in Tpr-mediated cargo import and implicate Tpr in premature aging (Kelley et al., 2011; Snow et al., 2013). Similarly, Tpr has been demonstrated to be essential for mRNA export (Bangs et al., 1998; Xu et al., 2007; Rajanala and Nandicoori, 2012; Lee et al., 2020) and overexpres- sion or knockdown triggers the nuclear accumulation of unspliced RNA (Shibata et al., 2002; Coyle et al., 2011; Rajanala and Nandicoori, 2012). Similar observations have been made in yeast (Green et al., 2003; Galy et al., 2004; Vinciguerra et al., 2005). Several data also implicate Tpr in gene regulation. Under stress conditions, it associates with the HSP70 promoter (Heat Shock Protein 70 kilodaltons) and the HSF1 (Heat Shock Transcription Factor 1), suggesting an involvement in the regulation of inducible genes (Skaggs et al., 2007). This idea was strengthened by observations of the x-linked genes’ expression decreasing after knockdown in Drosophila of both Tpr and Nup153 (Mendjan et al., 2006). Consistent with this finding, Tpr was also hypothesised to play roles in chromatin organisation by keeping heterochromatin exclusion zones near the NPC. Indeed, knockdown of Tpr in transfected HeLa cells shows heterochromatin presence in proximity to the NPC (Krull et al., 2010). Tpr has also been implicated in cell signalling through phosphorylation of the MAP- K/ERK (mitogen-activated protein kinase/extracellular signal-regulated kinase) and McCloskey et al. (2018) found that the Tpr-ERK complex negatively controls the number of NPC per cell nucleus. CHAPTER 1. INTRODUCTION 26 1.6.2 The spindle matrix protein Tpr during cell division Studies in different systems have underlined the role of Tpr and its homologues during mitosis (Qi et al., 2004; Niepel et al., 2005; Lee et al., 2008; De Souza et al., 2009; Lince-Faria et al., 2009; Nakano et al., 2010; Schweizer et al., 2013). In budding yeast, Mlp2 regulates the Spindle pole body (SPB), a microtubule nucleating structure functionally similar to the centrosome. When Mlp2 is depleted, smaller SPBs are formed with consequent SPBs components erroneously positioned in the nucleus (Niepel et al., 2005). In fission yeast, Jiménez et al. (2000) reported that functional disruption of the Alm1 gene is linked to elongated cells and impairment in the germination process. Tpr and its homologues have a highly conserved SAC function during mitosis (De Souza et al., 2009; Lince-Faria et al., 2009; Qi et al., 2009; Nakano et al., 2010). As described in 1.5.3, Mad1 and Mad2, essential proteins for SAC function, are recruited to unattached kinetochores by Mps1 where they bring about the formation of the MCC complex and subsequent APC/C inhibition. The SAC activity starts after the releasing of Mad1 and Mad2 from the Tpr (Lee et al., 2008) by the coordinated action of the Cyclin B-CDK1 and MPS1 (Morin et al., 2012; Cunha-Silva et al., 2020; Jackman et al., 2020). Lee et al. (2008) found that human Tpr binds to Mad1 and Mad2, through residues 1–774 of the N-terminal region and 1700–2350 in the C-terminus, respectively. Phosphorylation on Tpr residue S2059 is needed for its Mad1-Mad2 interactions (Rajanala et al., 2014). Once Mad1-Mad2 are released, they are recruited to the unattached kinetochore by BUB1 to start the assembly of the MCC and SAC response amplification (Saurin et al., 2011; Faesen et al., 2017; Ji et al., 2017; Allan et al., 2020; Jackman et al., 2020). Knockdown of Tpr by siRNA prevents the correct recruitment of Mad1-Mad2 at kine- tochore, causing inactivation of SAC and chromosome segregation defects. This sug- gests that Tpr could act as a scaffold for Mad1/Mad2 localisation (Lee et al., 2008). CHAPTER 1. INTRODUCTION 27 Consistent with this, Schweizer et al. (2013) show that depletion of Tpr leads to im- pairment in SAC proteins levels at kinetochore, in particular Mad1 and Mad2. Fur- thermore, Nakano et al. (2010) reported that Tpr association with the motor pro- tein dynein spatiotemporally regulates the recruitment of Mad1 and Mad2 during metaphase/anaphase transition. Accordingly, knockdown of either Tpr or dynein leads to lagging chromosomes (Nakano et al., 2010). Although Tpr has been demonstrated to promote Mad1/2 loading on mitotic kineto- chores, they colocalise to interphase nuclear pore complexes as well. Two independent studies found that Mad1 construct lacking Tpr binding domain is still able to interact with Mad2 in interphase, which challenged the idea of Tpr being needed for the Mad1- Mad2 interaction (Rodriguez-Bravo et al., 2014; Lara-Gonzalez et al., 2019). However, Rodriguez-Bravo et al. (2014) show that Tpr is important for MCC formation even prior to the beginning of mitosis. Displacing Mad1-Mad2 from the NPC during in- terphase, prevents MCC formation with subsequent accelerated anaphase, leading to malorientations and lagging chromosomes. Furthermore, a recent study in Drosophila shows that the Mps1 kinase is crucial for Tpr phosphorylation and consequent releasing of Mad1 and Mad2 from the NPC to the nucleoplasm (Cunha-Silva et al., 2020). 1.6.3 The Drosophila spindle matrix protein Megator The Drosophila spindle matrix protein Megator (Mtor), first identified as Bx34, is a protein of 2346 amino acids with a total mass of 260 kDa (Mitchell and Cooper, 1992; Byrd et al., 1994; Bangs et al., 1998). 70% of Megator’s amino-terminus is a predicted coiled-coil region, while the remaining 30% carboxy-terminus is enriched in negatively charged amino-acid residues. Although sequence comparisons reveal 28% identity, the domain structures and functions are highly conserved (Zimowska et al., 1997; Lince-Faria et al., 2009). Studies in mitosis show that Megator localises to the nuclear rim at interphase and in the spindle at metaphase (Zimowska et al., 1997; Qi et al., 2004; Lince-Faria et al., CHAPTER 1. INTRODUCTION 28 2009). This dynamic distribution is not affected by the depolymerisation of micro- tubules, therefore Megator is considered a spindle matrix protein (Qi et al., 2004; Lince-Faria et al., 2009). Megator’s localisation to the NPC and spindle matrix is me- diated by its amino-terminus coiled-coil domain, while the carboxy-terminus domain is responsible for Megator’s localisation during interphase (Yao et al., 2012). Early mutational studies by Qi et al. (2004) demonstrated that Megator is an essential gene and homozygous Megator mutants are lethal. Similar to Tpr, depletion of Megator in Drosophila S2 tissue cells showed significant reduction of both Mad2 and Mps1 recruitment at kinetochores, resulting in a premature entry into anaphase (Qi et al., 2004; Lince-Faria et al., 2009). Lince-Faria et al. (2009) also reported that Mad2 depleted S2 cells have the same premature exit and lagging chromosome phenotype as Megator depleted cells, underlying Megator’s role for the SAC function. This raises the possibility that loss of Megator function in the germline may affect fertility, similarly to Drosophila Mad2-null mutants, which experience a 50% reduction in fertility and viable progeny (Buffin et al., 2007). As mentioned in 1.6, Megator interacts with other Drosophila spindle matrix proteins: Skeletor and Chromator. Skeletor is an 81 kDa protein that lacks structural motifs (Walker et al., 2000). The 101 kDa Chromator protein is composed of an amino- terminus chromodomain with a site for Skeletor interaction present in the carboxy- terminus at residues 601-926 (Rath et al., 2004; Wasser et al., 2007). During interphase, Megator is present at the nuclear rim in polytene containing chromosomes and occupies the intranuclear space around the chromosomes. In contrast, Chromator and Skeletor localise to the chromosome’s interband regions (Walker et al., 2000; Qi et al., 2004). Despite these different localisation, all three proteins surround the spindle to form the spindle matrix during mitosis, and retain their organisation after spindle removal through MTs depolymerising drugs (Walker et al., 2000; Qi et al., 2004; Rath et al., 2004; Lince-Faria et al., 2009). Furthermore, protein depletion studies yield pheno- types with mitotic defects, such as chromosome misalignment and missegregation being CHAPTER 1. INTRODUCTION 29 prevalent (Rath et al., 2004; Ding et al., 2009; Lince-Faria et al., 2009). 1.7 Megator’s role during meiosis Haploid sperm formation and male fertility require two specialised male meiotic divi- sions to reduce the number of chromosomes in each daughter cell. To do so, each chro- mosome interacts and segregates at anaphase via the spindle and kinetochores. Errors in chromosome segregation cause genetic instability and can generate various patholo- gies, from cancer to male sterility or developmental abnormalities in the progeny. Much of the understanding of meiosis comes from studies of somatic cell mitosis. However, despite common molecular components, the two processes are not entirely interchange- able. Recent works in somatic cells mitosis have shown that proteins of the nuclear pore com- plex also perform extra-nuclear functions during cell division. Megator, the Drosophila homologue of vertebrate Tpr, is a conserved nucleoporin present in the nuclear basket. Along with two additional proteins, Skeletor and Chromator - Megator redistributes around the spindle during cell division forming a spindle matrix, a hydrogel-like struc- ture proposed to supply mechanical support to microtubules for chromosomes segrega- tion. Megator/Tpr is also involved in the spindle assembly checkpoint (SAC) during somatic cell mitosis. Indeed, RNAi-mediated depletions of Megator/Tpr in Drosophila and vertebrate tissue culture cells lead to chromosome segregation defects due to its role in recruitment of the essential SAC protein Mad2 to kinetochores. In Drosophila testes, Megator is also involved in the regulation and maintenance of the asymmetric division of the stem cells (Liu et al., 2016). The asymmetric division of GSCs required Apc2 and E-cadherin, two proteins that work together to anchor the spindle (Yamashita et al., 2003). Liu et al. (2016) show that the depletion of Megator alters the localisation/expression of Apc2 and produces a significant reduction of E- cadherin, resulting in increased frequency of lagging chromosomes and spindles with CHAPTER 1. INTRODUCTION 30 half-formed or not clear spindle poles. Furthermore, Megator depletion in the CySCs triggers defects in GSCs differentiation, suggesting that Megator is essential for the maintenance of GSCc in the testes. Also, CySCs-Megator mutants started to move away from the niche and differentiate in cyst cells, indicating that the Megator’s role in CySCs is to regulate their attachment to the niche. Despite its importance in mitosis, no information is available on Megator’s function during male meiosis or on the existence of a spindle matrix. Therefore, this project aims to: 1. Characterise Megator distribution during meiosis I and II. 2. Determine if Megator is required for correct chromosome segregation in meiosis I and II. 3. Identify if Megator regulates correct chromosome segregation by performing a conserved role in the SAC. Chapter 2 Materials and Methods 2.1 Drosophila melanogaster strains The Drosophila fly strains used in this thesis are listed and desribed in the Appendix, Table A.1. Shorthand names are used, following the nomenclature approved by Fly- Base1. Italics is only used for gene names and genotypes, not for proteins and pheno- types. Relative crossing between the original fly lines to produce progeny with the desire genotype, are illustrated in section 2.3. 2.2 Fly strains maintenance Flies were raised at 25◦C. Standard fly media was prepared from 10 g agar, 40 g yeast, and 110 g polenta dissolved in 1 L of distilled water. The mixture was brought to boil and then let to simmer for two minutes while stirring. The resulting product was removed from the heat to add 130 g granulated white sugar, 20 mL molasses and 3.3 g of Moldex (methyl 4-hydroxybenzoate) diluted in 37 mL 95% ethanol. Approximately 8 mL of the final mixture was poured into 30 mL vials (for fly crossing). Volumes of 40 mL were poured into 100 mL bottles (for stock maintenance). Once cooled, the media 1https://wiki.flybase.org/wiki/FlyBase:Nomenclature 31 CHAPTER 2. MATERIALS AND METHODS 32 was supplemented with yeast sprinkled on food and each vial or bottle was sealed with a foam plug or sponge plug. 2.3 Genetic crosses In order to carry out genetic crossing for functional studies, virgin female flies were col- lected from stocks in the morning. Virgin females β-tubulinEGFP//;bam-GAL4// and males carrying genotypes in accordance to experiments were mated together maintain- ing a 3:1 ratio of females to males. The mating process took place in the vials prepared with the fly media as stated above. Adult flies were removed after five days, while the progeny was due to eclose after ten days at 25°C. A schematic representation and descriptions of the crosses utilised in this study are shown (Figure 2.1). Control flies were generated by crossing β-tubulinEGFP//;bam-GAL4// virgin female flies to w(CS10) fly. The functional analysis of Megator was performed by using tissue- specific RNAi-based protein depletion under the GAL4/UAS system as homozygous Megator mutants are lethal (Qi et al., 2004). Virgin female flies carried the tissue specific bam-GAL4 driver which drives Megator knockdown only in Drosophila late spermatogonia and early spermatocytes (Bunt et al., 2012). For further studies on Megator function, virgin βtubulinEGFP//;bam-GAL4// female flies were crossed with either of two different lines expressing the fluorescent protein tagged Megator-mCherry (Yao et al., 2018). CHAPTER 2. MATERIALS AND METHODS 33 Figure 2.1: Schematic representation of Drosophila crossing performed in the study. Typical crosses generated to produce: control flies (A), flies with Megator (Mtor) tissue-specific knockdown (B) and flies expressing two different lines of the fluorescent protein Mtor-mCherry (C and D). Female and male sex chromosomes are indicated with x and y, respectively. + indicates a wild-type chromosome. CHAPTER 2. MATERIALS AND METHODS 34 2.4 Drosophila testes isolation and fixation Drosophila testes were dissected in a drop of phosphate-buffered saline (PBS, pH 7.4) with dissecting needles under a dissecting microscope. After the removal of unwanted material, six testes were transferred to frosted slides in 3 µL of PBS supplemented with 5% glycerol. Testes were punctured a few times and covered with coverslip (22x22 mm, thickness No.1) to allow meiotic cells to be released. Slides were plunged into liquid nitrogen until frozen. Slides were then removed and the coverslips were rapidly discarded before placement in dry ice cooled methanol for thirty minutes for fixation. After several rinses in PBS, testes were extracted for ten minutes with a solution of 0.5% Triton X-100 in PBS (Pisano et al., 1993; Cenci et al., 1994). Slides were then washed in PBS for three times (five minutes each) and left in PBS for one hour at room temperature for subsequent staining. 2.5 Immunostaining of Drosophila testes Primary antibodies (Table 2.1) were diluted in a solution of 0.05% Triton X-100 in PBS supplemented with 1% bovine serum albumin (w/v) BSA. Slides were pat dry around the sample and a square was drawn with a hydrophobic barrier pen. Then, 50 µL of the solution containing the primary antibodies was loaded onto the testes and the slides were incubated overnight at 4◦C in a moist chamber. The following day samples were washed three times (five minutes each) with PBS. Subsequently, secondary antibodies were applied (Table 2.2) and the slides incubated at room temperature for three hours. After this period, testes were washed again three times in PBS (five minutes each) and two times with distilled water. Vectashield mounting media containing DAPI was applied on the samples. A coverslip (22x22mm, thickness No. 1) was placed on top of the slide in alignment with the square previously drawn. Excess liquids were pat dry around the edges before sealing with nail polish. Samples were kept in the freezer at −20◦C until analysis with Zeiss LSM900 Airyscan 2 super-resolution scanning confocal CHAPTER 2. MATERIALS AND METHODS 35 microscope (Zeiss). Name Target Class Host Source Code Dilution 12F10-5F11 Megator Monoclonal mouse DSHB AB 2721935 1:100 12H9-4A2 Chromator Monoclonal mouse DSHB AB 2721936 1:25 1A1-3C2 Skeletor Monoclonal mouse DSHB AB 2721937 1:25 ADL101 Lamin Dm0 Monoclonal mouse DSHB AB 528332 1:50 Chicken anti-CID CID Monoclonal chicken Przewloka and Glover (2009) 1:500 Table 2.1: List of primary antibodies and corresponding dilutions. DSHB stands for Developmental Studies Hybridoma Bank. Name Target Class Origin Species Source Code Dilution Goat anti-chicken IgY Alexa Fluor 647 Chicken Polyclonal Goat ThermoFisher A-21449 1:500 Goat anti-chicken IgY Alexa Fluor 633 Chicken Polyclonal Goat ThermoFisher A-21103 1:500 Goat anti-mouse IgG CF568 Mouse Polyclonal Goat Merck SAB4600312 1:500 Table 2.2: List of secondary antibodies and corresponding dilutions. 2.6 Microscopy. All images were acquired using a Zeiss 980 Airyscan 2 super-resolution scanning con- focal microscope (Zeiss) using a 63x (N.A. 1.4) lens. Images in scanning confocal (without super-resolution mode) were acquired with the following settings: DAPI excitation wavelength: 405 nm, signal collected from 405 nm to 605 nm, detector gain 699V, 0.8% laser power. EGFP excitation laser wave- length: 488 nm, signal collected between 488-575 nm, detector gain 232V, 0.7% laser power. Megator-mCherry (Mtor-mCherry) and Megator antibody signals excitation wavelength: 561 nm, signal collected from 561 nm to 650 nm, detector gain 670V, 2.6% laser power. For all z-series step size was 0.21 µm and a zoom of 1.3 was used. Images in super-resolution scanning confocal mode were acquired with the following settings: DAPI excitation wavelength: 405 nm, signal collected from 405 nm to 517 nm, CHAPTER 2. MATERIALS AND METHODS 36 detector gain 748V, 2.7% laser power. EGFP excitation wavelength: 488 nm, signal collected between 488-575 nm, detector gain 738V, 1.8% laser power. Megator/Skele- tor/Chromator antibody signals excitation wavelength: 561 nm, signal collected from 561 nm to 650 nm, detector gain 784V, 1.4% laser power. Cid (Far-red) excitation wavelength: 640 nm, signal collected between 640-700 nm, detector gain 715V, 3.7% laser power. Step size was 0.14 µm and a zoom of 1.3 was used. Images were assembled in Fiji-Imagej (Schneider et al., 2012) and texts on the panels were added with GNU Image Manipulation Program (Team, 2021). All the images acquired in super-resolution scanning confocal mode, assembled in panels and presented in the study are z-projections of variable number of steps ranging between 76 to 125. All the assembled panels of images in scanning confocal only are single sections. 2.7 Quantification of Megator depletion. Megator function was assayed using tissue-specific RNAi-based protein depletion under the bam-GAL4/UAS system. Because Megator is not depleted testes-wide using this strategy (Insco et al., 2009; Bunt et al., 2012), protein levels in meiotic cells were de- termined with quantitative microscopy (in scanning confocal mode only) by measuring the fluorescence intensity of a specific region of interest (ROI) in both the spindle and cytoplasm regions. Single optical sections only were selected from control (n=28) and Megator depleted cells (n=24) during prometaphase I. Within these single sections, a circle with an area of 7.62µm2 was positioned in the largest region of the spindle that did not contain chromosomes or microtubules, abstractions which would affect the accuracy of the measurements. An identical circle was also placed in a spindle free region. Mean intensity values of these areas were tabled (Appendix, A.2) and consequently presented in a bar plot. To evaluate the statistical significance of the difference in intensity counts between cytoplasm and spindle in both control and Megator depleted cells, a T-test for difference of means was used. Confidence intervals at 95% and p-value are provided with the CHAPTER 2. MATERIALS AND METHODS 37 calculation. Because the assumptions of the T-test are not fully met, due to the small sample size, the Wilcoxon sign rank test was employed. Both bar plot and test were produced using R (R Core Team, 2017). 2.8 Counting of onion stage spermatids. Testes from each homozygous parental fly line and the heterozygous progenies (Figure 2.1 A and B) were dissected in a drop of PBS with dissecting needles, under a dissecting microscope. Four testes for each genotype were then transferred to a slide in a drop of PBS and covered with a 22x22 mm coverslip. Filter paper was used to blot liquid from the coverslip edges to allow the rupture of the testes, with consequent outflow of the content by pressure of the coverslip on the specimens. Slides were visualised using a 40x lens (0.75 NA) on a Leica DMRBE Microscope, Phase contrast optics (Leica microsystems). The ratio nebenkern to nuclei was determined and tabulated (Appendix, Table A.3). A ratio of 1:1 indicates correct chromosome segregation and cytokinesis, while 1:0, 0:1, 2:0, 0:2, 2:1, 1:2 and any combinations of these (termed “others”) were considered indicative of failed division. Chapter 3 Results 3.1 Characterisation of Megator distribution during meiosis I and meiosis II As a first step in the evaluation of the role played in meiosis by the conserved nu- cleoporin Tpr/Megator, its distribution during both meiosis I and meiosis II was de- termined. Control male flies (flies with endogenous levels of Megator) were generated by crossing virgin female flies expressing both the β-tubulinEGFP which labelled mi- crotubules (Inoue et al., 2004) and the testes-specific driver bam-GAL4 (Bunt et al., 2012), to w(CS10) male flies (crossing scheme: Figure 2.1 A). This cross was performed to ensure that the progeny were heterozygous for both transgenes. Male progeny were then collected three days after eclosion and their testes were dis- sected and fixed according to the methanol fixation procedure (described in section 2.4). Subsequently, immunofluorescence staining was performed using previously es- tablished antibodies raised against Megator and CID (the centromeric histone H3-like protein identifier), a marker for kinetochore position. Additionally, DNA was counter- stained with DAPI. Their signals were visualised by super-resolution scanning confocal microscopy (parameters described in section 2.6). 38 CHAPTER 3. RESULTS 39 Figure 3.1: Megator shows distinct distribution throughout meiosis I and II. Super-resolution scanning confocal microscope images of Megator (red; Mtor) distribution in meiosis I (A-F) and II (G-L) in control cells, compared with DNA in blue, microtubules (MTs) in green and the centromere marker CID in white. Corresponding images of Megator only, during meiosis I (a-f) and meiosis II (g-l). White arrowheads indicate Megator aggregates in the cytoplasm not corresponding to spindle microtubules or centrosomes. All images are z-projections. Scale bars 10µm. At prophase I, during chromosome condensation, Megator was present in aggregates of variable size throughout the nuclear volume and at the nuclear rim (Figure 3.1 A and a). This speckled nuclear rim appearance is consistent with Megator being part CHAPTER 3. RESULTS 40 of the nucleopore complex (Zimowska et al., 1997). Additionally, a few Megator spots were observed in the cytoplasm (Figure 3.1 a, white arrowheads). However, this signal did not correspond to microtubules or centrosomes, as revealed by the β-tubulinEGFP signal (Figure 3.1 A). After microtubule invasion into the nuclear space and consequent spindle formation, Megator moves into the space of the spindle (Figure 3.1 b), similar to previous findings of Megator being a component of the spindle matrix during mitotic cell division (Qi et al., 2004, 2005; Lince-Faria et al., 2009; Yao et al., 2018). During metaphase I, Megator robustly concentrates in the spindle region as revealed by the bright red signal in the merged figure (Figure 3.1 C). Furthermore, Megator forms aggregates which are excluded from the space occupied by the chromosomes, as shown by the regions without staining in the Megator channel only (Figure 3.1 c). Consistently, when chromosomes separate during anaphase I, Megator is still visible in the spindle region as agglomerates excluded from the space occupied by the DNA (Figure 3.1 d). Following spindle elongation, in accordance to previous findings in mitosis (Qi et al., 2004, 2005), Megator aggregates redistribute in the central spindle during telophase I and, close to cytokinesis I onset, it redistributes back near the daughter nuclei. Throughout meiosis I, minimal Megator signal is detected in the cytoplasm (Figure 3.1 a-f). At the beginning of meiosis II, Megator does not occupy the nuclear volume and is not present at the nuclear rim (Figure 3.1 g). This may suggest that nucleopores are either not present or are different at this stage. More surprisingly, in contrast to meiosis I, the Megator signal remains at background levels in the form of small, scattered granules. This is also consistent throughout all of meiosis II. Following spindle formation, the cytoplasm region displayed Megator aggregates that appeared to be excluded from the spindle microtubules as shown in the merged figures by the absence of red speckles in the spindle (Figure 3.1 H-J), as well as by the regions without staining in Megator channel only (Figure 3.1 h-j). Likewise, during telophase II and cytokinesis CHAPTER 3. RESULTS 41 II, numerous Megator particles were still detected in the cytoplasm (Figure 3.1 k and l). Together, these data reveal a different distribution of Megator between meiosis I and meiosis II. This suggests that Megator may be a spindle matrix protein in meiosis I but not in meiosis II or have other different functions. 3.1.1 Validation of Megator’s localisation during meiosis: Chromator and Skeletor distributions The unexpected dissimilarities in Megator distributions between meiosis I and II raised the possibility that the spindle matrix composition may substantially differ between the two division types. Therefore, the distributions of two additional mitotic spindle matrix proteins, Chromator and Skeletor (Walker et al., 2000; Rath et al., 2004; Ding et al., 2009),were investigated. Both these proteins interact with Megator (Qi et al., 2004; Yao et al., 2012). To compare Chromator and Skeletor localisation, testes from control male flies (crossing scheme visible in Figure 2.1 A) were dissected and fixed. Subsequently, testes were counterstained with DAPI and immunostained against CID, Chromator or Skeletor and analysed through super-resolution scanning confocal microscopy (Figure 3.2). CHAPTER 3. RESULTS 42 CHAPTER 3. RESULTS 43 Figure 3.2: Chromator and Skeletor are spindle matrix proteins during meiosis I and meiosis II. Comparative super-resolution scanning confocal microscope images of control cells stained for Chromator and Skeletor (both in red) during meiosis I (A-E and K- O) and meiosis II (F-J and P-T), relative to DNA in blue, microtubules (MTs) in green and CID in white. Corresponding images of Chromator and Skeletor only, during meiosis I (a-e and k-o) and II (f-j and p-t). Arrowhead indicates an unstained region non corresponding to DNA. All images are Z-projection. Scale bars 10µm. CHAPTER 3. RESULTS 44 During prophase I, both Chromator and Skeletor concentrate in the nuclear region (Fig- ure 3.2 a and k), similar to what was observed with Megator (Figure 3.1 a). However, in contrast to Megator, they do not localise to the nuclear rim which is consistent with Chromator and Skeletor not being nuclear envelope proteins. Following spindle forma- tion, in accordance to previous studies in mitosis (Walker et al., 2000; Rath et al., 2004), Chromator and Skeletor redistribute into the spindle during prometaphase/metaphase I and anaphase I (Figure 3.2 b and c, l and m). Their signals are excluded from the regions occupied by the chromosomes. In particular, during Chromator’s metaphase I, areas without staining (also non corresponding to the DNA) are observed (Figure 3.2 b, arrowhead). These Chromator and Skeletor localisation in the spindle matrix are consistent with studies of Megator in mitosis (Qi et al., 2004, 2005) as well as Megator distribution during meiosis I (Figure 3.1 b and c). However, Megator’s appearance in the spindle is heavily speckled, while both Chromator’s and Skeletor’s signals show much finer granules. The background signal of Chromator in the cytoplasm is low in comparison to its spindle’s localisation, during metaphase I and anaphase I (Figure 3.2 b and c). However, these signals appear considerably stronger in comparison with those of Skeletor and Megator at the same stages (Figure 3.2 l and m, Figure 3.1 c and d). As reported for mitosis (Walker et al., 2000; Qi et al., 2004), during telophase I both Chromator and Skeletor colocalised with the DNA, as shown by the merged figures (Figure 3.2 D and N). However, in contrast to Chromator and Megator, Skeletor’s signal does not trail in the spindle’s, but is close to the DNA (Figure 3.2 n). At cytokinesis I onset, both Chromator and Skeletor colocalise with the DNA, as shown by the merged figures (Figures 3.2 E and O). At the beginning of the second meiotic division, consistent with meiosis I distribution, both Chromator and Skeletor displayed nuclear localisation during prophase II with low signal intensity in the cytoplasmic areas (Figure 3.2 f and p). During metaphase II as well as anaphase II, Chromator’s and Skeletor’s distributions mirror meiosis I: CHAPTER 3. RESULTS 45 both proteins mainly accumulate in the spindle with their signal excluded from the areas occupied by the chromosomes (Figure 3.2 g and h, q and r). Furthermore, Chromator exhibits a background signal much more pronounced than Skeletor (Figure 3.2 g and h, q and r). During telophase II and cytokinesis II, as seen in meiosis I and studies in mitosis (Walker et al., 2000; Qi et al., 2004), both Chromator and Skeletor colocalise with the DNA (Figure 3.2 I and J, S and T). Consistently, Chromator forms cytoplasmic aggregates, something not seen when staining for Skeletor. It is unclear if this represents lack of specificity due to the antibody. The super-resolution scanning confocal analysis of Chromator and Skeletor revealed that their distriburions during meiosis I and meiosis II are largely similar. Chromator and Skeletor localise in the nuclear space at the beginning of meiosis I and meiosis II and near the nuclei during cytokinesis I and II. More importantly, they redistribute at the spindle at metaphase and anaphase, during both meiosis. These findings are consistent with Chromator and Skeletor being spindle matrix proteins during mitosis. However, they are also in striking contrast with Megator which localise in the spindle region only during meiosis I. Together, these data suggest that only Chromator and Skeletor are spindle matrix proteins during both meiosis I and meiosis II. 3.1.2 Confirmation of Megator’s distribution during Meiosis II: Megator-mCherry distribution Megator placements suggest the existence of a spindle matrix in meiosis I but not in meiosis II (Figure 3.1). To exclude the possibility that the absence of a Megator spindle signal in meiosis II was due to surface-epitope masking effects blocking the Megator antibody or other technical artefacts, the fluorescent protein tagged Megator was studied. Two different previously described fly lines expressing Megator fused to mCherry (Mtor-mCherry) (Yao et al., 2018) were crossed, as with the antibody-based studies, with β-tubulinEGFP//;bam-GAL4// flies to allow imaging of microtubules (crossing scheme: Figure 2.1 C and D). Testes from the progeny were dissected, fixed, CHAPTER 3. RESULTS 46 counterstained with DAPI and analysed by scanning confocal microscopy (parameters described in section 2.4). CHAPTER 3. RESULTS 47 Figure 3.3: Megator-mCherry localisation during MI and MII confirm Megator antibody distributions. Comparative single sections of scanning confocal images of cells isolated from flies expressing the fluorescent protein Megator mCherry (Mtor-mCherry 1M or 2M) with microtubules (MTs) in green and DNA in blue, throughout MI (A-C and G- I) and MII (D-F and J-L). The corresponding Mtor-mCherry 1M and 2M single channel distributions in meiosis I (a-c and g-i) and MII (d-f and j-l).. Scale bars 10µm. CHAPTER 3. RESULTS 48 In both Mtor-mCherry lines, Megator accumulates in aggregates first around the nu- clear rim during prophase I and then in the nuclear spindle region throughout the rest of meiosis I (Figure 3.3 a-c and g-i). These distributions are identical with that observed using Megator antibody (Figure 3.1 a-f). Interestingly, Megator spots were again observed in the cytoplasm with both of the Mtor-mCherry expressing lines. These speckles did not correspond to the spindle microtubules or chromosomes (Figure 3.3 B and H). This indicates that during MI, there is no significant Megator antibody surface-epitope masking (SEM). An examination of Mtor-mCherry distribution in meiosis II confirmed the results of the fixed cell antibody-stained experiments (Figure 3.1 g-l). At no time during MII did either Mtor-mCherry expressing line have a signal at the nuclear rim during prophase II or concentrate in the spindle matrix during prometaphase/metaphase II (Figure 3.3 d and e, j and k). Rather, as seen with the antibody staining, it remains at, or possibly less than, background levels throughout all meiosis II (Figure 3.3 d-f and j-l). Together these data suggest that the absence of Megator signal in meiosis II is not due to a Megator antibody SEM. This contrasts with the mitotic cells examined to date, which display Megator homologues in a matrix-like localisation. It raises questions of Megator and spindle matrix function in the testes. 3.2 Identifying Megator meiotic functions As described in section 1.6.2, Megator roles have been studied only in mitosis, and its functions in meiosis, if any, are unknown. Because Megator is an essential gene (Qi et al., 2004) which also plays roles in the earliest stages of Drosophila spermato- genesis prior to meiosis (Liu et al. (2009), Figure 1.1 F), traditional mutants and gene knockouts cannot be used to test protein functions. Therefore, the GAL4/UAS system was employed to drive tissue specific Megator depletion in the late spermato- gonia and early spermatocytes through the bam-GAL4 driver (Insco et al., 2009; Bunt et al., 2012). β-tubulinEGFP//;bam-GAL4// virgin female flies were crossed with CHAPTER 3. RESULTS 49 males carrying UAS-Megator for Megator’s tissue specific knockdown (crossing scheme: Figure 2.1 B. Validation of Megator’s depletion efficiency is described below). The β- tubulinEGFP//;bam-GAL4// virgin female flies were also crossed to w(CS10) flies, as a control, to ensure that any phenotypes observed did not result from unequal copy num- bers of the β-tubulinEGFP transgene (crossing scheme: Figure 2.1 A). Male progeny from both crosses were collected and their testes dissected, fixed and immunostained. The effects of Megator depletion were then analysed by super-resolution scanning con- focal microscopy. 3.2.1 Quantifying Megator’s depletion by quantitative microscopy The use of bam-GAL4 to drive protein depletion in a subset of cells in the testis makes Western blot-based assays non-representative of the actual protein levels in the cells being studied. Therefore, Megator depletion efficiency was carried out by quantitative microscopy and the endogenous Megator fluorescence accumulation was analysed in both the spindle and cytoplasm regions of control and Megator shRNA expressing cells using antibody staining (Figure 3.4). CHAPTER 3. RESULTS 50 Figure 3.4: Megator shRNA expressing cells have significantly reduced Mega- tor protein fluorescence’s depletion efficiency. An example of single section scanning confocal images used to determine Megator intracellular levels in controls (A) and Megator depleted cells (B) during prometaphase I. Scale bar 10µm. A qualitative examination of single section scanning confocal images acquired under identical conditions revealed a striking difference in the brightness of the Megator sig- nal between treatments. In control cells Megator appeared as a robust prometaphase I spindle matrix signal with the presence of large aggregates (Figure 3.4 A). By contrast, in the Megator shRNA cells the Megator signal was consistently dimmer, sometimes difficult to see above background, and characterised by small speckles (Figure 3.4 B). Thus, the depletion strategy is able to reduce protein levels in meiotic cells. In order to quantify the depletion efficiency, the fluorescence intensity of both the spindle and cytoplasm was measured and compared between control and knockdown cells accord- ing to the methodology described in section 2.7. The resulting fluorescence counts (Appendix, Table A.2) were then plotted (Figure 3.5). CHAPTER 3. RESULTS 51 Figure 3.5: Megator shRNA efficiently depletes Megator in MI cells. Differences in Megator’s spindle and cytoplasmic accumulations in fixed prometaphase I control (A) and Megator depleted cells (B). The bar plot describes the relative fluorescence intensity of a circle of 7.62µm2 positioned in the maximum chromosome-free spindle or microtubules-free cytoplasmic area in control (n=28) and Megator depleted cells (n=24) during prometaphase I. Stars indicate the level of significant difference between group averages, calculated with Wilcoxon sign rank test. The bar plot confirms the fluorescence accumulation in the spindle of Megator depleted cells is considerably reduced compared to control cells. As expected from a visual examination of the images, the difference in fluorescence levels between spindle and cytoplasm in control cells is very large, with the average relative fluorescence intensity in the spindle of control cells being tenfold the average fluorescence in the cytoplasm. Quantification of fluorescence intensity between control and Megator shRNA expressing cells reveals an 85% drop in signal intensity at the spindle matrix. Indeed, matrix and cytoplasm intensities are significantly different following depletion (p < 0.0001, Wilcoxon sign rank test), with the fluorescence intensity in the spindle matrix being threefold the one in the cytoplasm. Notably, the large confidence intervals in the spindle regions of the control cells reflect CHAPTER 3. RESULTS 52 the much larger variability of fluorescence that can occur due to the variation in Mega- tor aggregates numbers and sizes (standard error: 3.29, Table 3.1). By contrast, the presence of less aggregates in the Megator depleted cells reflect the reduced variability in the spindle (standard error: 0.35, Table 3.1). The variability in the spindle of Mega- tor depleted cells is a tenth of the variability of fluorescence in the spindle of control cells. The difference in variability is also reduced between cytoplasm and spindle in the depleted cells, as the spindle region has only twofold the fluorescence of the cytoplasm. Location Mean Standard deviation Standard error Cytoplasm 3.85 1.63 0.31 Control Spindle 40.60 17.40 3.29 Cytoplasm 2.08 0.79 0.16 Megator depleted Spindle 6.85 1.73 0.35 Table 3.1: Fluorescence intensity is significantly reduced in the spindles of Mega- tor depleted cells. Because of the small size of the sample analysed, the statistical significance of the results was further confirmed with the Wilcoxon sign rank test. This is a non-parametric test used to evaluate the significance of an hypothesis when the standard t-test cannot be performed due to non-satisfied assumptions (e.g., normality, as in this case). Each test produced a p-value: if small (p < 0.05), it confirms the significant difference between groups. The test confirmed the existence of a difference between spindle and cytoplasm in the control cells, while there was no significant difference between these pools in the case of Megator shRNA, at a 95% confidence level. Together, these data show that the Megator shRNA can efficiently deplete Megator protein. 3.2.2 Megator promotes matrix localisation of Chromator and Skeletor throughout meiosis The interaction between Chromator, Skeletor and Megator during mitosis has been documented (Walker et al., 2000; Qi et al., 2004; Rath et al., 2004), but no information is available in meiosis. Therefore, Megator depleted cells (crossing scheme Figure 2.1 CHAPTER 3. RESULTS 53 B) were stained for Chromator and Skeletor and the distribution of each was analysed by super-resolution scanning confocal microscopy (Figure 3.6). CHAPTER 3. RESULTS 54 Figure 3.6: Megator depletion decreases the recruitment of the spindle matrix proteins Chromator and Skeletor. Comparative super-resolution scanning confocal mi- croscope images of Megator depleted cells stained for Chroma