NUP2 AND A NEWLY DISCOVERED NUCLEAR PORE COMPLEX PROTEIN, NUPA, FUNCTION AT MITOTIC CHROMATIN CONTROLLED BY THE NIMA KINASE DISSERTATION

Transcription

1 NUP2 AND A NEWLY DISCOVERED NUCLEAR PORE COMPLEX PROTEIN, NUPA, FUNCTION AT MITOTIC CHROMATIN CONTROLLED BY THE NIMA KINASE DISSERTATION Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University By Sarine W. Markossian Graduate Program in Molecular Genetics The Ohio State University 2011 Dissertation Committee: Stephen A. Osmani, Ph.D., Advisor Anita K. Hopper, Ph.D. Iris Meier, Ph.D. Harold A. Fisk, Ph.D.

2 Copyright by Sarine W. Markossian 2011

3 ABSTRACT In Aspergillus nidulans the mitotic NIMA kinase is required for the transition from G2 into mitosis. How NIMA regulates this process is not well understood. The nucleoporin Nup2 is of interest because it is essential and has a unique and conserved ability to transfer from nuclear pore complexes (NPCs) to chromatin specifically at mitosis. This unique transition suggests that Nup2 might play mitotic roles downstream of NIMA conserved from fungi to humans. We have determined that NIMA promotes Nup2 phosphorylation and relocation from NPCs onto chromatin. Affinity purifications followed by Mass spectroscopy identified a novel nucleoporin, NupA, which copurified with Nup2. NupA transfers from NPCs to mitotic chromatin and is also essential. Deletion analysis and heterokaryon rescue revealed that deletion of either does not cause major defects in nuclear transport but causes mitotic abnormalities that engage the spindle assembly checkpoint (SAC). We have also defined the essential function of NupA which is to tether Nup2 to both the NPC and mitotic chromatin where we propose Nup2 has a role in promoting efficient generation and function of the mitotic apparatus. In addition, Nup2 and NupA have dynamic locations within mitotic chromatin during different phases of mitosis. They are associated throughout chromatin at early prophase but move to locate at the periphery of the chromatin during the transition from anaphase to telophase. This movement at anaphase is likely functionally significant as deletion of either Nup2 or NupA causes defects in mitotic exit into G1. Hence, Nup2 and NupA may have dual roles in mitosis, the first being during early mitosis, defects in which activates the SAC, and the latter in mitotic exit and early G1. Because Nup2 co- purifies with ii

4 importins α and, we propose that both roles might be linked to proper localizations and functions of importins α and during metaphase and mitotic exit. These studies collectively demonstrate that Nup2 and the newly discovered NupA have important functions during mitosis that are controlled by NIMA. iii

5 To my mother, Iskouhi Shamikian, for providing me the opportunity of higher education and teaching me to believe that everything is possible through hard work. iv

6 ACKNOWLEDGEMENTS I am deeply indebted to my advisor Dr. Stephen A. Osmani for his continual guidance and constant support throughout the years of my graduate studies. I thoroughly enjoyed his mentorship and his friendship. I would also like to thank my current and previous committee members, Dr. Anita K. Hopper, Dr. Iris Meier, Dr. Harold A. Fisk, and Dr. Berl R. Oakley for their valuable inputs into my work and all the encouragement during my graduate studies. In addition, I would like to thank all current and former members of the Osmani lab for their friendship, the assistance provided with experiments, the rewarding discussions about my work, and the continuous and critical feedback they have provided throughout the years. I would like to particularly thank Aysha H. Osmani, Subbulakshmi Suresh and Jessica S. El-Hallal because this work would not have been possible without their fruitful collaborations. I would also like to thank all the Molecular Genetics family, particularly my classmates for all the fun times, their friendship, and the extreme encouragement and support they provided throughout this experience. Finally, I would like to thank my parents, my sister, and all the rest of my extended family, including friends that I was lucky enough to encounter throughout the years in Lebanon and in Ohio. It is through their constant and unconditional love, faith, support and encouragement that this achievement was possible. v

7 VITA 2001 B.S. Biology, American University of Beirut 2003 M.S. Biology, American University of Beirut 2004 to present..graduate Research Associate, Department of Molecular Genetics, The Ohio State University PUBLICATIONS Liu HL, Osmani AH, Ukil L, Son S, Markossian S, Shen KF, Govindaraghavan M, Varadaraj A, Hashmi SB, De Souza CP, Osmani SA Single step affinity purification for fungal proteomics. Eukaryot Cell 9: Kreydiyyeh S.I., Markossian S., Hodeify R.F PGE2 exerts dose-dependent opposite effects on net water and chloride absorption from the rat colon. Prostaglandins and Other Lipid Mediators 79: vi

8 Kreydiyyeh S.I., Markossian S Tumor necrosis factor alpha down-regulates the Na+-K+ ATPase and the Na+-K+2Cl- cotransporter in the kidney cortex and medulla. Cytokine 33: Markossian S., Kreydiyyeh S.I TNF-alpha down-regulates the Na+-K+ ATPase and the Na+-K+-2Cl-cotransporter in the rat colon via PGE2. Cytokine 30: Markossian S.W Effect of TNF-α on electrolyte transport across epithelia : possible mechanism of action. M.S. Thesis. Beirut: American University of Beirut. Kreydiyyeh S.I., Usta J., Knio K., Markossian S., Dagher S Aniseed oil increases glucose absorption and reduces urine output in the rat. Life Sciences 74: FIELDS OF STUDY Major Field: Molecular Genetics vii

9 TABLE OF CONTENTS ABSTRACT... ii ACKNOWLEDGEMENTS... v VITA... vi PUBLICATIONS... vi FIELDS OF STUDY... vii TABLE OF CONTENTS... viii LIST OF TABLES... xiv LIST OF FIGURES... xv LIST OF ABBREVIATIONS... xviii CHAPTER INTRODUCTION Project Goal Aspergillus nidulans as a model system to study mitosis Biology Genome assembly and annotation Advantages The Nucleus Overview Differences in nuclear structure between lower and higher eukaryotes The nuclear envelope (NE) The Nuclear Pore Complex Overview Isolation of Nups NPC biogenesis into an intact NE Nucleoporin 2 (Nup2) Protein structure viii

10 Isolation and characterization Nucleo-cytoplasmic transport Overview Role of Nup2 in transport Nuclear chromatin Overview Nups and gene regulation Nup2 and gene regulation: Mitosis Overview Types of mitosis Regulation of mitosis Overview G2/M transition The mitotic spindle Spindle assembly and function The spindle assembly checkpoint (SAC) Post-mitotic NPC assembly An evolving relationship between nuclear transport and mitosis Overview Role of RanGTP and transport receptors in mitosis Role of Nups in mitosis Evidence from the literature connecting Nup2 with cell division The hypothesis CHAPTER MATERIALS AND METHODS General Aspergillus nidulans techniques Specific media for culture Preparation of A. nidulans conidia stock suspensions Long term storage and stock preparation of A. nidulans Strain generation by meiotic crossing Transformation ix

11 2.1.6 Counter selection on 5-FOA Genomic DNA extraction Small scale protein preparation alca driven protein expression Cell fixing and DAPI Staining Fusion PCR and gene targeting in Aspergillus nidulans Gene deletion constructs Endogenous C-terminal tagging constructs Nup2 antibody generation and immunofluorescence Western blot analysis Heterokaryon rescue technique for the identification and analysis of essential genes in Aspergillus nidulans Microscopy and image acquisition Fixed samples Live cell imaging CHAPTER NUP2 AND A NEWLY DISCOVERED NUCLEAR PORE COMPLEX PROTEIN, NUPA, PLAY ESSENTIAL ROLES AT MITOTIC CHROMATIN CONTROLLED BY THE NIMA KINASE Introduction Material and Methods Single-Step Affinity Purification of Nup2 in G2 and M Synchronization of cells and generation of G2/M samples Protein preparation for purifications S-Tag purification Coomassie staining of protein gels In vitro λ-phosphatase assay NIMA induction experiments NupA sequencing and 3 RACE for NupA Genomic and cdna sequencing of NupA Identification of the intron in the Aspergillus fumigatus NupA Quantification x

12 Quantification of dispersal and import of NLS-DsRed Quantification of time spent in mitosis Quantification of nuclear movement Results Nup2 is phosphorylated and locates to chromatin during mitosis Identification of a Novel Nucleoprin, NupA Affinity purification of Nup2 during G2 and mitosis and isolation of NupA cdna sequencing of NupA reveals an unusually long intron which is conserved in Aspergillus fumigatus, Aspergillus niger, Aspergillus aculeatus and Aspergillus carbonarius NupA is an essential gene whose deletion does not affect short term growth but does cause mitotic defects NupA translocates from NPCs to chromatin during mitosis NupA is required for Nup2 location at NPCs and mitotic chromatin Nup2 and NupA transfer from NPCs to chromatin at early prophase and move back to the nuclear periphery in anaphase Nup2 and NupA are not essential for nuclear transport Deletion of either Nup2 or NupA causes mitotic defects that engage the spindle assembly checkpoint Nup2 and NupA null mutants form polyploid and/or aneuploid nuclei Nup2 and NupA null mutants display defects in mitotic exit Nup2 and NupA null mutants exhibit defects in reassembling normal postmitotic daughter nuclei Nup2 and NupA null mutants are not capable to localize Mad1 into the nuclear periphery after the first mitosis NupA null mutants show defects in post-mitotic membrane fission NupA null nuclei exhibit additional interphase defects in Ima1-localization Discussion and future direction Identification of NupA, a previously uncharacterized Nup Presence of an unusually large intron in nupa Why would NupA, a gene essential for mitosis, diverge this rapidly during evolution? NIMA regulates the movement of Nup2 to chromatin during mitosis Nup2 and NupA have roles in metaphase spindle assembly and/or stability xi

13 3.4.6 What could the essential functions of Nup2 and NupA be in mitosis? Nup2 and NupA have roles in mitotic exit and rebuilding of post-mitotic daughter nuclei Nup2 and NupA orchestrate reversible NPC disassembly with DNA segregation and formation of daughter nuclei CHAPTER ISOLATION OF SEPERATION OF FUNCTION ALLELES, AND CHARACHTERIZATION OF THE ESSENTIAL FUNCTION OF, NUP2 AT THE NUCLEAR PORE COMPLEX AND ON MITOTIC CHROMATIN Introduction Material and methods Domain study Generation and analysis of Nup2ΔDomain4-GFP mutants Generation and analysis of the Nup2 fusion mutants Results A domain (domain 4) within Nup2 is sufficient to locate it to both NPCs and mitotic chromatin Domain 4 is necessary for Nup2 s translocation to both the NPC and mitotic chromatin and is essential for Nup2 s function Efforts to artificially tether Nup2 exclusively to the NPC Efforts to artificially tether Nup2 exclusively to DNA Discussion and future directions Mechanism of Nup2 s translocation to chromatin Separation-of-function alleles of Nup Can Nup2 s dynamic locations carry one essential function? CHAPTER EFFORTS TO DEFINE IF MAMMALIAN NUP50 IS A FUNCTIONAL HOMOLOGUE OF A. nidulans NUP Introduction Material and Methods Generation and analysis of strains carrying alca-egfp2-nup Complementation Study Endogenous replacement of Nup2 with Nup50 with or without domain Results xii

14 5.3.1 Induction of Nup50 expression under the alca promoter in Aspergillus nidulans does not complement Nup2 s function alca induced Nup50 is unable to locate to either the NPC nor to mitotic chromatin in the presence or absence of endogenous Nup Discussion and future directions Nup2 targets to nuclear pores and to DNA in distinct ways in A. nidulans and mammals How to test for complementation in the future? CHAPTER FINAL DISCUSSION Overview The proposed model (Figure 6.1) Mitotic roles of Nups are not vertebrate specific Lessons learned from Aspergillus nidulans BIBLIOGRAPGHY xiii

15 LIST OF TABLES Table A-1: Aspergillus nidulans Haploid strains used in this study Table A-2: Aspergillus nidulans Heterokaryons used in this study Table A-3: List of primers xiv

16 LIST OF FIGURES Figure 1.1 Cartoon depicting the life cycle of Aspergillus nidulans...33 Figure 1.2 Cross-sectional views of a typical mammalian and an Aspergillus nidulans nucleus...35 Figure 1.3 The Architecture of the Nuclear Pore Complex (NPC)...36 Figure 1.4 Nup2 protein structure...37 Figure 1.5 The classical nuclear import cycle...39 Figure 1.6 Nup2 facilitates the importin / mediated nuclear import...40 Figure 1.7 Cartoon depicting NIMA s function at the G2/M transition in Aspergillus nidulans...41 Figure 1.8 The composition of the Aspergillus nidulans NPC during G2 and mitosis...42 Figure 2.1 Gene deletion constructs...55 Figure 2.2 Endogenous C-terminal tagging constructs...56 Figure 2.3 Expression and purification of an antigen against Nup Figure 3.1 Untagged Nup2 relocates from NPCs to DNA during mitosis...87 Figure 3.2 Identification of a novel Nup via affinity purification of A. nidulans Nup2..88 Figure 3.3 Induction of NIMA promotes phosphorylation and relocation of Nup2 from NPCs onto chromatin in S phase arrested cells...89 Figure 3.4 NupA is essential and its deletion, similar to Nup2, does not affect germination but causes mitotic DNA segregation defects...91 Figure 3.5 NupA colocalizes with Nup2 at the Nuclear Pore Complex during interphase and on chromatin during mitosis...93 Figure 3.6 NupA is required for Nup2 location to NPCs and to mitotic chromatin...95 xv

17 Figure 3.7 Nup2 and NupA associate with chromatin at the start of prophase and move back to the periphery of chromatin in early anaphase...96 Figure 3.8 Nup2 and NupA are dispensable for nucleocytoplasmic transport...99 Figure 3.9 Nup2 and NupA deleted nuclei exhibit defects in mitotic spindle structure..102 Figure 3.10 Nup2 and NupA deleted cells undergo slow mitosis due to activation of the spindle assembly checkpoint Figure 3.11 Nup2 and NupA deleted cells undergo slow mitosis due to activation of the... spindle assembly checkpoint Figure 3.12 Nup2 and NupA null mutants form polyploid and/or aneuploid nuclei Figure 3.13 Nup2 and NupA null mutants exhibit excessive and prolonged nuclear movement upon mitotic exit which is dependant on microtubules Figure 3.14 Nup2 and NupA null mutants are defective in postmitotic Mad1 localization to the nuclear periphery Figure 3.15 NupA null mutants show defects in post-mitotic membrane fission Figure 3.16 NupA null mutants exhibit defects in double pinching of the nuclear envelope during mitotic exit Figure 3.17 NupA null nuclei exhibit an interphase defect in Ima1-localization Figure 4.1 Method used to integrate GFP tagged domains 1-3 into the A. nidulans genome under the alca promotor at the pyroa locus Figure 4.2 Method used to integrate GFP tagged domains 3-5 into the A. nidulans genome under the alca inducible promotor at the pyrg locus Figure 4.3 Generation of the Nup2ΔDomain4-GFP protein Figure 4.4 Generation of the Nup2-Nup37 chimeric protein Figure 4.5 Generation of the Nup2-Histone H1 chimeric protein Figure 4.6 A single domain within Nup2, Domain 4 (aa ), is responsible for its translocation to the NPC during interphase and to mitotic chromatin Figure 4.7 Domain 4 is sufficient to locate Nup2 to the NPC at interphase and to mitotic chromatin Figure 4.8 Overexpression of domains 1-3 does not cause growth defects xvi

18 Figure 4.9 Overexpression of domains 4 and 5 does not cause growth defects Figure 4.10 Domain 4 is necessary to locate Nup2 to both the NPC at interphase and to mitotic chromatin, and is essential for Nup2 s function Figure 4.11 The Nup2::GFP::Nup37 protein is functional Figure 4.12 Artificially NPC tethered Nup2ΔD4 is not functional Figure 4.13 The Nup2::GFP::H1 protein is functional and locates to the NPC at interphase and to mitotic chromatin Figure 4.14 The Nup2ΔD4::GFP::H1 protein is not functional and does not locate to chromatin in the absence of endogenous Nup Figure 5.1 Generation of EGFP2-alcA-mmNup50 cassette targeted to the pyrg locus of Aspergillus nidulans Figure 5.2 mmnup50 locates to the Aspergillus nidulans nucleus in interphase and disperses throughout the cell in mitosis Figure 5.3 Replacement of endogenous A. nidulans Nup2 with EGFP2-mmNup Figure 5.4 Replacement of endogenous A. nidulans Nup2 with EGFP2-mmNup50 containing domain Figure 6.1 Model of the regulation, mechanism of translocation and the function of Nup2 at chromatin during Aspergillus nidulans mitosis xvii

19 LIST OF ABBREVIATIONS An APC CDK CR Aspergillus nisulans anaphase promoting complex cyclin dependent kinase Cherry red fluorescent protein D4 domain 4 DAPI DNA 5-FOA FG INM GFP M mm mrfp MTOC NE NIM NLS 4',6-diamidino-2-phenylindole deoxyribonucleic acid 5-fluoroorotic acid phenylalanine inner nuclear membrane green fluorescent protein mitosis millimolar monomeric red fluorescent protein microtubule organizing center nuclear envelope never-in-mitosis nuclear localization signal xviii

20 NPC nuclear pore complex Nup2 nucleoporin 2 NupA ONM PCR SAC Sc WT Δ nucleoporin A outer nuclear membrane polymerase chain reaction spindle assembly checkpoint Saccharomyces cerevisiae wild type deletion µm micrometer xix

21 CHAPTER 1 INTRODUCTION 1.1 Project Goal Regulation of mitosis is essential for human growth and development. Knowledge about the regulatory events during mitosis is of critical importance to our understanding of the progression and treatment of many human diseases such as cancer and birth defects. Although much is known about mitosis, many basic questions remain and the potential for new discoveries still exists. For instance, the reversible mitotic disassembly of nuclear pore complexes (NPCs), which are gateways between the nucleus and the cytoplasm, and the mitotic function of the nuclear pore complex proteins, nucleoporins (Nups) are not well understood. Nucleoporins were thought to have exclusive roles in nuclear transport but recent evidence has shown that many nucleoporins have dual roles dependent on the stage of the cell cycle: a NPC dependent conventional transport role during interphase and a NPC independent role in mitosis. Those Nups are shown to dynamically locate at distinct locations during mitosis (De Souza and Osmani, 2007; Guttinger et al., 2009; Kutay and Hetzer, 2008). One such Nup is Nup2 which is the focus of my research. Nup2 undergoes perhaps the most distinctive mitotic translocation of any Nup described so far. It completely translocates from the NPC in interphase to chromatin during mitosis in the filamentous fungus Aspergillus nidulans 1

22 (Osmani et al., 2006a), the model genetic system used in my research. This mitotic translocation also occurs in mammals (Dultz et al., 2008) indicating that Nup2 proteins have conserved roles at chromatin during mitosis. Moreover, Nup2 is essential in Aspergillus nidulans and its deletion causes mitotic DNA segregation defects (Osmani et al., 2006a). Together, these findings suggest that Nup2 has a distinct mitotic function, independent from its function at the pore during interphase. The main objective of this study is to reveal the function and the mechanism of action of Nup2 at chromatin during mitosis. Our immediate goal is to help clarify how stepwise disassembly and reassembly of NPCs during mitosis are coordinated with spindle formation and chromosome segregation. My findings will then hopefully provide a better understanding of how mitosis is regulated and the opportunity to develop novel therapeutic approaches for the treatment of human diseases that are associated with the cell cycle, such as cancer. 1.2 Aspergillus nidulans as a model system to study mitosis Biology Aspergillus nidulans, (teleomorph Emericella nidulans) is a filamentous fungus of the phylum Ascomycota. Aspergillus nidulans was the first homothallic fungus explored and chosen for experimental breeding (Pontecorvo et al., 1953). In the early 1950s, Pontecorvo and his team found Aspergillus nidulans very suitable for a genetic approach to study spatial organization of the cell and laid the foundation for genetic analysis in Aspergillus nidulans (Pontecorvo et al., 1953). Since then, additional progress has been made in establishing it as a powerful biochemical and genetic system, and it has become one of the important model organisms for the study of cell biology. Aspergillus nidulans undergoes sexual, asexual and parasexual life cycles (Figure 1.1). It is mainly haploid but can also form heterokaryons, hyphae containing two different types of nuclei in one cytoplasm, by hyphal fusion and vegetative diploids by nuclear fusion (karyogamy). It is homothallic, i.e. self-fertile. Sexual crosses thus can occur between two organisms of the same genotype (self-crossing) as well as between 2

23 two organisms of distinct genetic make-ups (out-crossing). These diverse life cycles make it an ideal organism for diverse set of experiments. In the asexual cycle, a haploid uninucleate conidiospore undergoes first isotropic growth. Polarized growth is then initiated leading into the formation of the germtube. This is followed by multiple para-synchronous mitoses leading to the formation of vegetative hyphae. The vegetative hyphae then differentiate to produce a sporogenic body, the conidiophore which generates multiple conidiospores. Differentiation is initiated by formation of a foot cell from which a stalk emerges. A vesicle is then formed at the end of the stalk. The nuclei in the vesicle undergo rounds of simultaneous mitosis followed by specialized budding growth giving rise to two layers of uninucleate sterigmata: the metulae and the phialides. The phialides, acting like a stem cell, in their turn then undergo multiple rounds of budding and mitosis producing many conidiospores (Casselton and Zolan, 2002; Todd et al., 2007a; Todd et al., 2007b). Because the asexual cycle produces haploid homokaryon hyphae and haploid spores, it is ideal for gene targeting and replacement experiments. In addition, the fact that the germling undergoes fast and para-synchronous mitosis makes it an ideal place to follow protein location dynamics throughout the cell cycle by live cell confocal microscopy. A. nidulans has a G1 phase of 15 min, an S phase of 40 min, a clearly defined G2 phase of 40 min, and a mitotic phase (M phase) of only 5 min, under ideal growth conditions giving a nuclear division generation time of 100 min (Oakley, 1993). In the parasexual cycle, hyphal fusion occurs between two genetically distinct cells to form heterokaryons. Heterokaryons can then enter the sexual cycle after karyogamy. Heterokaryons can also produce vegetative diploids which are usually very stable. Haploidization can occur after drug treatments that interfere with the function of the mitotic spindle, leading back to the homokaryon haploid vegetative state. Heterokaryons spontaneously form during gene replacement of essential genes by heterokaryon rescue and they may be used to propagate the deletion mutant nucleus, demonstrate the gene to be essential, and allow determination of the phenotype of the mutant allele (Osmani et al., 2006b). The sexual life cycle of A. nidulans is divided into 3 main steps: heterokaryon formation by hyphal fusion, diploid formation by karyogamy, and haploidization through 3

24 meiotic cell division. As stated previously, A. nidulans can either undergo self-crossing or out-crossing. Out crosses are initiated by hyphal fusion between two vegetative hyphae, each carrying nuclei of a particular genetic type. The resulting heterokaryon can undergo a septation event resulting in the separation of two nuclei of different genetic type into one septal compartment. The pair of nuclei destined for meiosis then divide in synchrony to form a mass of cells known as ascogenous hypha. Karyogamy then occurs in the penultimate cell to give rise into a diploid nucleus which then undergoes meiosis. This is followed by one round of mitosis, resulting in the formation of eight nuclei, each destined to become an ascospore in an ascus. Each ascospore nucleus then undergoes one more round of mitosis without cytokinesis to form a mature binucleate ascospore. Asci are contained in closed sexual fruiting bodies (cleistothecia), and surrounded by Hülle cells that are presumed to function as nurse cells (Casselton and Zolan, 2002; Todd et al., 2007a; Todd et al., 2007b). The sexual cycle of A. nidulans allows for conventional genetic analysis of meiotic progeny and generation of strains with specific genotypes needed Genome assembly and annotation The Aspergillus nidulans genome was sequenced by the Broad Institute and comparative analysis with the genomes of A. fumigatus and A. oryzae was performed (Galagan et al., 2005). The genome was sequenced through the whole-genome shotgun approach, and generated sequences were assembled using the Arachne software package. The size of the genome is approximately 31Mb and is distributed among 8 chromosomes. Current assembly represents about 96.3% of the A. nidulans genome. The genome coverage is to a depth of 13X, meaning that sequencing was repeated 13 times to ensure the accuracy of the data obtained. Annotation of the genome was performed in two steps. In the first step, three different gene prediction algorithms were used for automated gene annotation. In the second step, manual annotation was performed at The Institute for Genomic Research (TIGR, Rockville, MD) using the PASA pipeline to align all available EST data to the genome, and compare it to the annotated data sat. The current annotation release (Release 4, March 7, 2006) contains 10,701 genes, corresponding to 10,701 4

25 transcripts. About 58.8% of the genome corresponds to gene sequences, the remaining 41.2% of the genome corresponds to intergenic regions and repeats (Galagan, 2008) Advantages Aspergillus nidulans is a powerful model genetic system that has many unique characteristics making it an advantageous model to study different aspects of biology. The life cycles of this interesting filamentous fungus offers many technical advantages. These include the Heterokaryon Rescue technique to study essential genes (Osmani et al., 2006a; Osmani et al., 2006b), the ability of efficiently targeting linear gene replacement cassettes with an extremely high percentage of homologous recombination using the Ku deleted strains (Nayak et al., 2006; Osmani et al., 2006b; Yang et al., 2004), the S-tag protein affinity purification system developed for A. nidulans which enables the highly specific purifications of many multi-protein complexes from large quantities of protein extracts generated from this fungus (Liu et al., 2009; Liu et al., 2010), and the availability of high quality genome sequences for A. nidulans and many other Aspergilli species that can be used for comparative genomics (Osmani and Goldman, 2008). Moreover, A. nidulans has been used historically for genetic screens, therefore many cell cycle mutants (Morris, 1975; Simchen, 1978), and mutants deficient in processes like DNA damage response (Goldman et al., 2002) and nuclear migration (Morris et al., 1995) are available. Genetic screens, which are a powerful way of identifying new genes and/or new functions for genes, can be carried out with ease in this model organism (Casselton and Zolan, 2002). In addition, as stated previously, A. nidulans has a short life cycle (Oakley, 1993). It is possible then in this fungus to follow proteins by live cell imaging throughout one or consecutive cell cycles. Because A. nidulans can undergo self and out-crosses, it is thus easy to generate strains carrying different combinations of protein markers and deletions as required for the experiment. In addition, there are no mating types in this fungus so any two strains can be mated. Furthermore, genes can be endogenously tagged in A. nidulans with GFP, chrfp and other fluorophores through homologous recombination. Finally, inducible promoters like the alcohol dehydrogenase promotor (alca) are also well characterized, and can be used 5

26 to study the effect of altering the levels of gene expression of specific genes during different stages of the cell cycle. In addition to its unique technical advantages, this fungus undergoes unique and interesting biological processes. These processes are considered to be the intermediate in complexity between higher eukaryotes and many model yeast systems. A remarkable example of such a process is the regulation of nuclear pore complexes during A. nidulans mitosis (De Souza et al., 2004; Liu et al., 2009; Osmani et al., 2006a). Another interesting process pertinent to this argument is the regulation of nucleolar segregation in A. nidulans (Ukil et al., 2009). These processes are very reminiscent to the NPC and nucleolar disassembly and reassembly of higher eukaryotes during mitosis making this fungus again an ideal system to study them. Because A. nidulans is a powerful model genetic system that has many unique characteristics, and the fact that it is only through the utilization of a diverse set of model systems that comparative biological insights to many processes will be possible, I chose to conduct my Ph.D. research using this fungus. 1.3 The Nucleus Overview The nucleus is one of the defining features of eukaryotic cells. It carries the genetic material, the DNA, which is transmitted from cell to cell and parent to offspring. The DNA is organized into a highly compact form called chromatin which is surrounded and protected by the nuclear envelope, a double membrane layer containing many integral and peripheral membrane proteins essential for its function. The nucleus is the site of transcription, RNA processing, mrna maturation and ribosome biogenesis. Transcription factors that carry signals from the cellular environment need to access the nucleus from the cytoplasm to activate gene expression. Also, proteins of the DNA replication apparatus and the DNA damage response should access the nucleus from the cytoplasm. Moreover, although gene transcription happens in the nucleus, mrna needs 6

27 to shuttle from the nucleus to the cytoplasm where it is translated into protein by ribosomes which themselves are built in the nucleolus and are transported into the cytoplasm. Thus, proper shuttling of proteins, mrna and other nucleic acids in and out of the nucleus is of upmost importance for cellular function. To accomplish this, the nuclear envelope contains pores which constitute one of the biggest macromolecular complexes in eukaryotic cells, the nuclear pore complexes (NPCs). The nucleus also contains the nucleolus which is the site of ribosome biogenesis and RNA processing. In this section, I would like to introduce the different parts of the nucleus relevant to my studies and would like to emphasize their function and importance in maintaining a living cell. Most importantly, I would like to introduce the nuclear pore protein, Nucleoporin 2, since understanding its regulation and function in Aspergillus nidulans is the major focus of my research Differences in nuclear structure between lower and higher eukaryotes The nuclear structure, components and function are highly conserved from lower to higher eukaryotes (Figure 1.2). The fungal nucleus though differs from the mammalian one in various ways. In this section, I focus on the most important differences in nuclear structure between mammalian and fungal nuclei. Fungi lack lamins, thus the nuclear lamina is absent. The microtubule organizing centers in fungi are called spindle pole bodies and are either attached to (such as in S. pombe) or imbedded (A. nidulans, S. cerevisiae) in the nuclear envelope. In addition, the fungal nucleus is 8 to 10 times smaller than the mammalian one. Moreover, in fungi, the centromeric heterochromatin is tethered to the region next to the spindle pole body. This region is called the MTOC attachment site (MAS) in S. pombe (King et al., 2008) The nuclear envelope (NE) The nuclear envelope is an extension of the endoplasmic reticulum (ER) which is flattened around the chromatin and forms one continuous double layer of lipid membranes, the inner nuclear membrane (INM) and the outer nuclear membrane (ONM) 7

28 (Watson, 1955). These two membranes connect at the nuclear pores. The nuclear envelope differs tremendously from the ER due to the proteins it contains. The INM and the ONM are enriched with a diverse set of proteins which are typically not enriched in the ER (Hetzer et al., 2005). These proteins are important and essential for many functions of the nucleus such as chromatin organization, gene expression, and DNA metabolism (Hetzer, 2010). One proteinacious complex of the nuclear envelope, the NPC, will be discussed in section Other proteins specifically locate to the INM. These proteins include a large number of yet uncharacterized nuclear envelope integral membrane proteins (NETS) (Schirmer et al., 2003), lamin B receptor (LBR) and lamin associated polypeptides (LAP) in higher eukaryotes (Akhtar and Gasser, 2007; Dorner et al., 2007; Schirmer and Foisner, 2007), in addition to Man1 (LEMD3) (Lin et al., 2000), also called Heh1 (King et al., 2006) or Src1 (Rodriguez-Navarro et al., 2002) in yeast, and a highly conserved protein, Ima1 (King et al., 2008). Most importantly, malfunction of many of those proteins are linked to many human diseases (Vlcek and Foisner, 2007). ONM proteins are specifically enriched at the ONM and have many essential functions. One group of ONMs are the KASH (klarsicht, ANC-1, Syne Homology) domaincontaining proteins which form bridges with Sad1p/UNC-84 (SUN)-domain INM proteins linking the outer and the inner membranes through the periplasmic space of the NE (Starr and Han, 2003; Wilhelmsen et al., 2006). This SUN-KASH complex links the cytoplasm and specifically MTOCs to the inner nucleoplasm and chromatin. In S. pombe, the Sad1-kms2 bridging complex links the MTOC to the centromeric heterochromatin through Ima1 which resides at the INM and contributes to the coupling of heterochromatin to the MAS (King et al., 2008). In mammals, other related ONM proteins, the Nesprins, also carry similar functions. Nesprins directly interact with the actin cytoskeleton and also with SUN proteins forming the LINC complex which establishes connections between the nucleus and the actin cytoskeleton (Hetzer, 2010). These proteins are mainly characterized for their roles in nuclear positioning, migration and signaling between the cytoplasm and the nucleus (Crisp et al., 2006; Luxton et al., 2010). 8

29 1.3.4 The Nuclear Pore Complex Overview Nuclear pore complexes are large macromolecular assemblies that span the double layer of the nuclear envelope and are made of multiple copies of different proteins termed nucleoporins, commonly abbreviated as Nups. Because they span the nuclear envelope, they form channels bridging the nucleus to the cytoplasm, allowing the diffusion of small molecules and the active transport of larger proteins and nucleic acids (Alber et al., 2007a; D'Angelo and Hetzer, 2008; Stewart, 2007). NPCs have an overall conserved structure. The central transport channel is made up of transmembrane Nups in combination with multiple subcomplexes of core Nups such as the ynup84/mnup complex. Peripheral Nups reside in the center of the transport channel. Some of these peripheral Nups are specific to the nuclear side, forming the nuclear basket. Some are more specific to the cytoplasmic side and form the cytoplasmic fibrils (Figure 1.3 and 1.8). Transport through NPCs is highly regulated through active processes, although molecules smaller than 40kD can pass through the pore by passive diffusion. Active transport will be discussed in more detail in section NPCs form doughnut-shaped structures and have an 8-fold symmetry. They are massive structures with an estimated mass of 66 to 125 megadaltons (mda). Each NPC has been proposed to contain about 456 individual proteins of approximately 30 different Nups. The central core Nups mostly contain α-solenoid and -propeller domains which act as a scaffold for NPC assembly (Alber et al., 2007a; Alber et al., 2007b; Hetzer et al., 2005; Hetzer and Wente, 2009; Hsia et al., 2007; Wente and Rout, 2010). The peripheral Nups contain Phenylalanine-Glycine (FG) repeats. The regions in these Nups that have the FG repeats exhibit properties of natively unfolded proteins which display high flexibility and lack of ordered secondary structure (Frey et al., 2006). FG Nups form a barrier in the central transport channel of the NPC by reversible cross linking of their FG repeats forming a 3- dimensional meshwork with hydrogel-like properties (Frey and Gorlich, 2007; Frey et al., 2006). Cargo-carrier complexes are docked and translocate through the NPC in a process mediated by interactions between transport receptors and the FG repeats of the peripheral Nups (Bayliss et al., 2000). 9

30 Isolation of Nups The first Nups identified and characterized were the FG Nups. They were identified through an immunological approach using monoclonal antibodies raised against rat liver nuclei isolated with non-ionic detergents that kept intact NPCs (Davis and Blobel, 1986). These antibodies were then used to identify Nups in vertebrates and in yeast. The genes of the first Nups were then cloned and sequenced in yeast by immunoscreening (Davis and Fink, 1990; Fabre and Hurt, 1997; Hurt, 1988; Nehrbass et al., 1990). Later, genetic screens helped the identification of many other Nups in different organisms. For example, synthetic lethal screens identified many of the S. cerevisiae Nups (Fabre and Hurt, 1997). In addition extragenic suppressor screens identified SONA (Gle2p/Rae1) and SONB, the first Nups identified in A. nidulans (De Souza et al., 2003; Wu et al., 1998). In the last decade, the availability of proteomic analysis in different organisms from lower to higher eukaryotes has enabled and is enabling researchers to purify these massive complexes and perform mass spectrometry to identify and classify all of its protein components and their modifications (Alber et al., 2007b; Cronshaw et al., 2002; Liu et al., 2009; Liu et al., 2010; Rout et al., 2000) NPC biogenesis into an intact NE In organisms that undergo open mitosis, NPCs assemble into a reforming NE at the exit of mitosis and into growing intact NEs during interphase via two distinct mechanisms (Doucet et al., 2010). In higher eukaryotes, de novo nuclear pore biogenesis into intact NEs occurs in proliferating cells, during cell differentiation and due to changes in metabolism (Hetzer and Wente, 2009). Presumably, there are also distinct mechanisms of nuclear pore reassembly in fungi that undergo semi-open mitosis, such as Aspergillus nidulans (Liu et al., 2009). On the other hand, in S. cerevisiae, that undergoes closed mitosis, NPCs continuously assemble during the cell cycle into intact NEs. Post-mitotic NPC assembly will be discussed in section Increased evidence in the last decade favors the hypothesis that NPC biogenesis into an intact NE occurs de novo by INM/ONM fusion and not by duplication and splitting of existing NPCs. First, distinct cytoplasmic Nups are preassembled on the cytoplasmic ONM face, and Nups of the 10

31 nuclear basket preassembled on the INM face. The Sc-Nup84/ mammaliannup complex is required on both sides of the NE. The next step is the fusion of INM and ONM. This is accomplished by the orchestration of different events. It has been speculated that luminal domains of transmembrane Nups facilitate this step; actually the transmembrane Nup Pom121 is required for the incorporation of the Sc- Nup84/mNup complex into new assembly sites in metazoans. Recent work also shows that ER membrane curving proteins have a role in this process. In addition, many Nups, specifically members of the Sc-Nup84/mNup complex, Sc-Nup170, and Sc-Nup53, have amphipathic α-helical domains which are putative membrane interaction domains that help maintain the curved pore membrane by oligomerizing to form a coat. Once the Sc-Nup84/mNup complex is incorporated, then the sequential recruitment and insertion of the full cohort of peripheral Nups forms the mature NPC (Doucet and Hetzer, 2010; Doucet et al., 2010; Hetzer and Wente, 2009; Hsia et al., 2007; Wente and Rout, 2010) Nucleoporin 2 (Nup2) Protein structure Nucleoporin2 (Nup2) is a peripheral and mobile nucleoporin which resides mostly on the nucleoplasmic side of the NPC and is part of the nuclear basket. It is a highly conserved nucleoporin with a similar overall domain structure from yeast to humans. The Nup2 protein can be divided into 3 distinct conserved domains (Figure 1.4). An N- terminal importin-α binding domain, a domain rich in FG repeats, and a C-terminal Ran GTPase binding domain. The yeast Nup2p also contains a NPC targeting domain which follows the N-terminal importin- binding domain (Booth et al., 1999; Hood et al., 2000). The importin-α binding domain is conserved in the first 50aa of yeast Nup2p and its vertebrate orthologue Nup50 (Booth et al., 1999; Hood et al., 2000; Matsuura et al., 2003; Matsuura and Stewart, 2005). Crystal structures of this domain bound to importinα were resolved using both the yeast and the mouse proteins. The crystal structures showed that Nup2 contains two distinct conserved importin-α binding segments (BS1 and 11

32 2) which are separated by an acidic linker (Matsuura et al., 2003; Matsuura and Stewart, 2005). The FG repeats within Nup2 are characteristic of nucleoprins that interact with karyopherins and facilitate transport of karyopherin-cargo complexes (Denning et al., 2003; Dilworth et al., 2001). Finally, Nup2 contains a conserved C-terminal Ran binding domain typical to domains found in many Ran binding proteins such as Ran binding protein 1/Yrb1p, Yrb2p and the mammalian Nup358. The Aspergillus nidulans Nup2, similar to many of the filamentous fungi Nup2, is a significantly larger protein than its yeast and human counterparts (Figure 1.4B). The reason why An-Nup2 is a much larger protein is unknown, although the assumption is that An-Nup2 might carry additional functional domains not present in the other proteins. An-Nup2 has a predicted N-terminal importin-α binding domain (aa1-50) followed by 21 FG repeats (3 FXFGs; 3 GXFGs; and 15 FGs) distributed throughout the protein, a predicted α-helical coiled coil domain (aa ), and a predicted C-terminal Ran GTPase binding domain (aa ). An-Nup2 also carries 2 predicted classical NLSs and a bipartite NLS (Figure 1.4A) Isolation and characterization Nup2 was first isolated in budding yeast through a screen conducted to isolate nucleoprins using a monoclonal antibody against a yeast phage expression library (Loeb et al., 1993). This screen isolated a subset of Nups that contain the repetitive domain FXFG including Nup2p, Nup1p and Nsp1p (Davis and Fink, 1990). NUP2 is not required for growth in budding yeast but genetically interacts with NUP1 and NSP1. In vitro physical and structural characterizations of the S. cerevisiae Nup2p showed that it is a natively unfolded protein that exhibits little secondary structure, high flexibility and low compactness (Denning et al., 2002). Because Sc-Nup2p contains extensive regions of structural disorder, it is capable of simultaneous interactions with multiple protein partners and displays rapid association and dissociation rates with its binding partners. Sc-Nup2p was then classified as a bona fide nucleoporin by (Dilworth et al., 2001) who found that it associates with the distal regions of the NPC and is extremely mobile moving on and off the NPC in a Ran dependant manner. They showed that in normal 12

33 conditions where RanGTP is high, Sc-Nup2p associates with Nup60p which is responsible for Nup2 s localization at the nuclear basket (Denning et al., 2001; Dilworth et al., 2001). In parallel with work in yeast, the mammalian orthologue of Nup2, Npap60/Nup50, was isolated by screening a rat testis cdna expression library with a monoclonal antibody. The gene was then cloned and characterized by immunostaining to be a NPC protein in Rat fibroblast cells RAT1A and in differentiating germ cells (Fan et al., 1997). Nup50 was also identified as a two-hybrid interacting partner of p27 kip1, a member of the Cip-Kip family of cyclin-dependant kinase (cdk) inhibitors, and was shown to be a widely expressed mouse nuclear pore associated protein. Yeast two hybrid screens using Nup50 as bait against mouse embryo, brain, and testis libraries, showed that it interacts with many NPC proteins including Nup153. In addition, Nup50 was shown to coimmunoprecipitate with Nup153 in transfected cells. Endogenous Nup50 also coimmunoprecipitates with endogenous Nup153 from rat liver nuclei. Targeted disruption of Nup50 caused complex neural tube and CNS abnormalities and growth retardation resulting in late embryonic lethality (Smitherman et al., 2000). Nup50 was also found to be highly concentrated at the nuclear envelope of rat liver nuclei, and immunogold electron microscopy shows that it specifically localized at the nuclear basket of the NPC (Guan et al., 2000). Nup50, like Nup2p, was then shown to exhibit dynamic interactions with the NPC as it shuttles between the nuclear and cytoplasmic sides of the NPC. In addition, in the binucleated ciliate Tetrahymena thermophila, Nup50 was isolated by surfing its genome for proteins that contain phenylalanine (FG) repeats, and shown to localize not only to the periphery of both macronucleus (MAC) and micronucleus (MIC), but also significant levels of the protein was shown to be present in the nucleoplasm. The Aspergillus nidulans Nup2 was identified by sequence similarity to the yeast Nup2p. Interestingly, unlike the S. cerevisiae Nup2p, the Aspergillus nidulans Nup2 was shown to be essential, and its deletion leads to DNA segregation defects. Moreover, using live cell confocal microscopy and following the endogenous An-Nup2-GFP throughout the cell cycle, Osmani et. al showed that An-Nup2 localizes at the nuclear periphery as 13

34 expected but had a very unique and distinct translocation to chromatin during mitosis (Osmani et al., 2006a) Nucleo-cytoplasmic transport Overview Transport of macromolecules between the nucleus and the cytoplasm is an essential process in all eukaryotes. A considerable amount of energy is spent in this process which regulates fundamental cellular processes such as gene expression and signal transduction. Nucleo-cytoplasmic transport occurs through the NPC and is mediated by multiple families of soluble transport factors, termed karyopherins. There are several pathways for nuclear transport, each specific for a range of macromolecules. Most pathways though use the homologous super family of -karyopherins. The import carriers are called importins and the export ones, exportins. Although numerous pathways that use different karyopherins exist, the basic concepts for nuclear import and export are the same. Cytoplasmic proteins that need to access the nucleus carry nuclear localization sequences (NLSs) that are recognized by a specific karyopherin and form an importincargo complex. This complex then moves through the central channel of the NPC through interactions of the importin with FG-repeat containing Nups (Bayliss et al., 2000). Once inside the nucleus, nuclear Ran, a small GTPase of the Ras superfamily, binds to the karyopherin and causes a conformational switch releasing the cargo from the complex. In order to exit the nucleus, proteins need to contain a nuclear export sequence (NES) which are recognized by exportins. Nuclear RanGTP promotes the formation of the exportincargo complex, and binds to it, forming a trimeric complex. This trimeric complex passes through the pore through the interaction of exportin with FG-Nups. Once in the cytoplasm, RanGTP is converted to RanGDP by the cytoplasmic RanGAP promoting the dissociation of the complex and release of the cargo. RanGDP is then imported back into the nucleus where it is converted into RanGTP by RCC1 (RanGEF) which resides in the nucleus and is associated with chromatin. Therefore, the nuclear transport pathway relies on the difference of RanGTP levels between the nuclear and cytoplasmic compartments. 14

35 RanGTP is high in the nucleus but low in the cytoplasm. This is achieved by the presence of RanGEF in the nucleus and RanGAP in the cytoplasm. The most common import pathway for soluble cytoplasmic proteins is the classic import pathway which uses importin- as a carrier (Figure 1.5). The proteins carried through this pathway carry a classical NLS which is recognized by importin-α which in its turn binds to importin-. So the classical pathway uses a trimeric complex for the import of cargo. Importin- is then recycled back to the cytoplasm with RanGTP. Importin alpha on the other hand uses an exportin called CAS in addition to RanGTP (King et al., 2006; Osmani, 2008; Stewart, 2007; Weis, 2003; Wente and Rout, 2010). Recently, it has been shown that integral INM proteins also possess basic sequences very similar to the classical NLSs, and use also the importin / heterodimer to be targeted to the INM from the ER. Alterations in the nuclear pore complex allow the passage of these integral INM proteins along the pore membrane (King et al., 2006). Nup2 has a function in the classical import pathway as it catalyses cargo dissociation from the importin / heterodimer and release into the nucleus. Nup2 also facilitates export of importin alpha into the cytoplasm (Figure 1.6) (Stewart, 2007). The function of Nup2 in this process will be further discussed in the next section Role of Nup2 in transport Genetic and physical interactions between Sc-Nup2p and the yeast importin- (Srp1p) was the first hint for a function for Nup2 in the classical nuclear import pathway (Belanger et al., 1994). Further studies showed that Sc-Nup2p functions in nuclear export of importin by acting as a scaffold that facilitates formation of the importin- export complex (Figure 1.6). An N-terminal 175aa region within Sc-Nup2p was shown to bind importin- and target Sc-Nup2p to the NPC (Booth et al., 1999; Hood et al., 2000). Nup2 was also shown to function in nuclear import. It helps the release of the NLS bearing cargo from the importin / heterodimer by strongly binding to this dimer through importin (Figure 1.6) (Solsbacher et al., 2000). In yeast, Nup2p forms a multiprotein complex with Nup60p, Kap60 (importin ), Kap95 (importin ), Prp20p/RCC1 and Gsp1p/RanGTP (Allen et al., 2001; Allen et al., 2002; Denning et al., 2001; Denning et 15

36 al., 2002; Dilworth et al., 2001). The RBD of Sc-Nup2p is important for its function in Kap60 export and localization to the NPC. Moreover, Nup60 is required to locate Sc- Nup2p at the nuclear periphery and RanGTP levels modulate this interaction (Denning et al., 2001; Dilworth et al., 2001). Nup50 also interacts with the importin / -RanGTP complex and stimulate nuclear import. In fact, Nup50 binds to importin-α through its N- terminus, RanGTP through its C-terminus and to importin- through a central domain and alternates between binding modes and shuttles between the cytoplasm and the nucleus showing that it is a cofactor for importin / mediated nuclear import (Lindsay et al., 2002). Moreover, kinetic work studying the rates of disassembly of different importin-nls-cargo intermediates showed that Sc-Nup2p accelerates the rate of disassembly of Kap60p-NLS-cargo complexes in yeast (Gilchrist et al., 2002; Gilchrist and Rexach, 2003). The structural basis of Nup2 s function in cargo release was then revealed and the crystal structure of a complex between both the yeast and the vertebrate Nup2 and the armadillo repeat domain of importin-α was determined. The crystal structures revealed that Nup2 binds to two sites on importin-α, with one site overlapping with the NLS binding site. Nup2 s affinity for importin-α is higher than for the NLS (Matsuura et al., 2003). Humans reportedly have two isoforms of Nup50/Npap60, a long isoform Npap60L (1-469aa) and its alternatively spliced form Npap60S (29-469aa). Recent work has detected that these two isoforms differentially regulate import. They showed that they have opposing roles, Npap60S promotes the binding of NLS-cargo to importin-α and the Npap60L promotes the release of cargo from importin-α making their expression levels very critical for efficient transport (Ogawa et al., 2009) Nuclear chromatin Overview The nuclear envelope is probably evolved with many of its components to protect genomic chromatin. Nuclear chromatin, which is made of DNA and its associated proteins, is the genetic material carried by any eukaryotic cell and transmitted from parent to offspring during mitosis and meiosis. The cell packages its DNA into highly 16

37 organized structures called chromosomes. Although tightly compact, chromosomes are dynamic structures that change into different states depending on the stage of the cell cycle. Chromosomes replicate during interphase, they become highly condensed during mitosis and then they are separated to two daughter nuclei. The most abundant proteins present in chromatin and mostly responsible for its organization are the histones. In fact, the basic unit of chromosome organization is the nucleosome. Each nucleosome consists of a group of eight histone proteins: two molecules of each H2A, H2B, H3 and H4 and double stranded DNA wrapped around this complex. Nucleosomes are then packed together to form a compact chromatin fiber. This is achieved partly by the linker histone H1 that pulls the nucleosomes together (Alberts, 2002). Detailed mechanism on how this higher order structure is achieved is still poorly understood. It is becoming clear that understanding dynamic chromatin organization is crucial because of its importance to nuclear self-organization (Misteli, 2001). The mobility and post-translational modifications of many chromatin binding proteins, especially of the core histones, contribute to both chromatin condensation and transcriptional regulation (Woodcock and Ghosh, 2010). The amino-terminal tails of many histones are subject to many posttranslational modifications such as methylation, acetylation and phosphorylation (Strahl and Allis, 2000). The most ordered chromatin structure is the mitotic chromosome. The metaphase chromatin is highly condensed and its histones are highly phosphorylated. Chromosome condensation is a highly regulated process and is required for proper mitotic progression. Although many of the roles of mitotic specific histone phosphorylation remain largely unknown, progress is under way. For example, it has been shown in Tetrahymena that the mitotic phosphorylation of the H3 on serine 10 is required for condensation. This phosphorylation is involved in initiation of chromosome condensation but not its maintenance (Van Hooser et al., 1998; Wei et al., 1998; Wei et al., 1999). In Aspergillus nidulans, the NIMA kinase was shown to be the kinase responsible for this phosphorylation (De Souza et al., 2000). 17

38 Nups and gene regulation During the recent years, it has become evident that Nups have additional transport-independent functions including roles in mitosis, chromatin organization and gene regulation. Nups are known to play critical transport-based roles in gene expression mainly by controlling the access of transcription factors and the exit of RNA molecules (Capelson and Hetzer, 2009). In multiple cell types, it has been observed by Electron Microscopy that the NE is underlined with electron-dense heterochromatin, interrupted by NPC-associated light-staining euchromatin (Akhtar and Gasser, 2007). These observations led to the gene gating hypothesis, which proposes that nuclear pores specifically interact with active genes and co-regulate transcription with mrna export (Blobel, 1985). Several studies performed in yeast support this hypothesis. In fact, a genome-wide analysis in S. cerevisiae confirmed that multiple Nups indeed occupy regions of highly transcribed genes (Casolari et al., 2004). Interestingly, conserved DNA sequences required to target NPC-regulated genes to the nuclear periphery were identified in yeast (Ahmed et al., 2010).This further proves that NPC components play an active role in gene regulation in yeast. Work in higher eukaryotes also shows a role of Nups in gene regulation. In Drosophila, Nups were shown to bind to promoters of active genes and regulate gene expression (Capelson et al., 2010; Kalverda et al., 2010; Vaquerizas et al., 2010). Surprisingly though, the Nups bound to chromatin were found in the nucleoplasm, away from the NE-embedded NPCs. Most of the Nups identified to have such roles are peripheral Nups which have been shown to be dynamic components of the NPC. These Nups include Tpr, Nup153, Nup98, Nup62 and Nup50 (Capelson et al., 2010; Kalverda et al., 2010; Vaquerizas et al., 2010). Furthermore, it has also been suggested that the intranuclear chromatin binding of Nups may regulate the threedimensional organization of the genome. Because emerging evidence exist suggesting a regulatory role of chromatin organization in gene regulation and nuclear organization, Nups may influence gene expression and nuclear function through genome organization. Interestingly, although the majority of the Nups show a preference in association with active genes, some are also found to interact with repressed loci (Capelson et al., 2010; Kalverda et al., 2010; Vaquerizas et al., 2010). This implies that the role of peripheral Nups in gene regulation is complicated. Moreover, in Drosophila embryonic cells, 18

39 Nup98, Nup62 and Nup50 associate with genes that regulate development and the cell cycle (Mendjan et al., 2006). In addition, Nup98, Sec13 and FG-Nups were later shown to be recruited to developmentally induced genes in Drosophila salivary glands, and this association was shown to be required for the expression of these genes (Capelson et al., 2010). Together, these observations suggest a role for Nups in the regulation of developmental programs in multi-cellular organisms Nup2 and gene regulation: As mentioned above, Nup2 is one of the peripheral Nups that have been implicated in gene regulation both in yeast and Drosophila (Liang and Hetzer, 2011). Nup2 was first shown in yeast to be a physical tether for chromatin boundary activities (BAs), displayed by proteins involved in nucleo-cytoplasmic transport, to the NPC. A Boundary activity is a function displayed by proteins that bind to boundary elements and act as insulators to protect genes from the repressive or activating effects of the nearby chromatin environment. Boundary elements are specific DNA sequences that partition the genome into functional domains. When flanked by these elements, genes are insulated from repressive or activating effects of nearby heterochromatin or enhancer elements, respectively (Gerasimova and Corces, 2001). This implied a role for Nup2 in altering epigenetic activity, by tethering the boundary elements to the NPC, and thus organizing the genome. This also provided the first evidence that the activity of chromatin boundary/insulator elements may mechanistically be linked to nuclear structure (Ishii et al., 2002). In fact, Nup2 and RanGEF/Prp20p were shown to interact at specific chromatin regions and play an active role in helping the transition of chromatin between activity states (Dilworth et al., 2005). Later, Nup2 was shown to bind to a set of promotors in yeast and activate gene expression (Casolari et al., 2004; Schmid et al., 2006). In Drosophila, Nup50 was shown to interact with transcriptionally active genes, inside the nucleoplasm, especially those involved in developmental regulation and the cell cycle (Kalverda et al., 2010). 19

40 1.4 Mitosis Overview In order for the cell to multiply, its nucleus must be divided into two. This nuclear division is termed mitosis. Mitosis is then followed by the division of the cytoplasm during cytokinesis. Mitosis is a highly regulated process during which cells undergo dramatic changes leading to separation of their duplicated DNA content into two daughter nuclei. This involves the partial (Aspergillus nidulans) or complete (metazoans) disassembly of the NPCs, the nuclear envelope break down (NEBD) in metazoans, and the assembly of a very elaborate and dynamic machine, the mitotic spindle, which is responsible for proper segregation of the duplicated chromosomes so that daughter nuclei acquire a copy of each chromosome. The mitotic spindle is formed when microtubules emanating from the microtubule organizing centers (MTOCs) attach to the kinetochores of chromosomes. At the kinetochores, a safety mechanism called the Spindle Assembly Checkpoint (SAC) ensures that proper bipolar attachments are made from the chromosomes to the two MTOCs before initiation of sister chromatid separation during anaphase. Then, with the help of microtubule motor proteins, such as kinesins and dynein, the spindle segregates the chromosomes. This is followed by reassembly of the NPCs and the NE in metazoans. Any defects in this process will lead to aneuploidy, abnormal number of chromosomes, which is thought to be a causing factor for initiation and progression of cancer. Thus, mitosis is an extremely essential and highly conserved process without which there would be no continuity of eukaryotic life. In this section, I will discuss the regulation of mitosis and its differences between organisms, some basic key features for mitotic progression and most importantly about recent work linking nuclear transport components to mitosis thus leading to my hypothesis Types of mitosis Although basic mitotic events are conserved, mitotic variations between eukaryotes exist. In higher eukaryotes, mitosis is termed open. During open mitosis, the nuclear envelope completely breaks down allowing the interacting of microtubules 20

41 emanating from MTOCs, which are cytoplasmic, with the kinetochores of nuclear chromosomes. NEBD is preceded by a stepwise disassembly of the NPCs. Stepwise reassembly of the NPCs and reforming of the NE follows chromosomal segregation upon mitotic exit. On the other hand, some lower eukaryotes such as S. cerevisiae, which have their MTOC embedded in the NE, undergo a closed mitosis. In closed mitosis, the NPCs and the NE stay intact. In fact, the spindle forms in the nucleus and the two daughter nuclei are separated due to the fission of the NE with a single pinch directly following chromosome segregation. Variations at this level of mitosis exist (De Souza and Osmani, 2007; De Souza and Osmani, 2009; Guttinger et al., 2009). In Drosophila early embryos that undergo syncytial mitosis, the NEBD occurs initially only around the region of the centrosomes and the NE does not completely break down until after the chromosomes segregate (Kiseleva et al., 2001). In the basidiomycete Ustilago maydis, the MTOCs, which are embedded in the nuclear envelope during interphase, are extracted into the cytoplasm at the onset of mitosis tearing the NE at the region of the MTOC thus resulting in a form of open mitosis, although a complete NEBD is not observed (Kiseleva et al., 2001; Theisen et al., 2008). The nuclear envelope of the filamentous fungus Aspergillus nidulans undergoes a more subtle change during mitosis. This fungus undergoes what is termed a semi-open mitosis in which the NPCs partially disassemble allowing the NE, although intact, to become permeable (Figure 1.7 and 1.8) (De Souza et al., 2004; De Souza and Osmani, 2009; Osmani et al., 2006a). This allows mitotic regulators such as tubulin to access the nucleus through passive diffusion and eventually promote spindle assembly in an environment where the cytoplasm and nucleoplasm have been mixed. Because the spindle in A. nidulans is assembled in an environment similar to higher eukaryotes, semi-open mitosis is thought to be an intermediate between the closed mitosis of budding yeast and open mitosis of vertebrate cells. In organisms in which the NPCs disassemble, Nups have been shown to locate to various mitotic structures such as kinetochores, spindle pole bodies and chromatin (Galy et al., 2006; Liu et al., 2009; Osmani et al., 2006a; Rasala et al., 2006). Increasing evidence is revealing new roles of Nups at these mitotic structures suggesting that in organisms with open or semi-open mitosis Nups have evolved to carry mitotic functions when away from the pores (De 21

42 Souza and Osmani, 2007; De Souza and Osmani, 2009; Guttinger et al., 2009). The roles of Nups in mitosis will be discussed in section Regulation of mitosis Overview The decision to enter mitosis is one of the most critical and dynamic steps in the cell cycle. Therefore, it is under strict control by a network of regulatory pathways known as checkpoints. For example, cells need to ensure that DNA replication is complete and the cell volume is big enough to proceed into mitosis. To identify genes that regulate cell cycle events, genetic screens were conducted in three simple eukaryotes all aimed to identify temperature-sensitive cell cycle mutants. The mutants were scored in Saccharomyces cerevisiae by bud size (Hartwell et al., 1970), and cell size for Schizosaccharomyces pombe (Nurse et al., 1976) and named cell-division-cycle (cdc) genes. In Aspergillus nidulans however, Ron Morris scored the mitotic events of chromosome condensation, spindle formation, and the number and spacing of nuclei. Mutants that were arrested in interphase were named never in mitosis (nim), and the ones arrested in mitosis were termed blocked in mitosis (bim) (Morris, 1975). Many key cell cycle regulators were discovered from these screens and many of them were shown to be involved in regulation of protein phosphorylation. These regulators were then identified to have functions either in mitotic entry, progression or exit (Fleig and Gould, 1991; Osmani, 2010; Reinhard Fischer et al., 2008) G2/M transition The major regulator identified to have a role in mitotic entry was the Cyclindependent kinase Cdk1/Cyclin B complex (NimX/NimE in A. nidulans, Cdc2/Cdc13 in S. pombe). Cdk1 is an essential kinase that functions in mitotic progression. It requires a cyclin binding partner for it to function and its role is universally preserved in all eukaryotes. During G2, phosphorylation of the Cdk1/Cyclin B complex by the Wee1 kinase at Tyr 15 keeps it inactive. Upon the G2/M transition, this complex is 22

43 dephosphorylated by the Cdc25 phosphatase (NimT in A. nidulans) and leads to its activation (Fleig and Gould, 1991; Osmani et al., 1991; Osmani, 2010; Reinhard Fischer et al., 2008). Mutant nimt23 strains therefore arrest in G2 with the presence of a phosphorylated and inactive Cdk1 kinase. However, in A. nidulans, the activation of the Cdk1 kinase alone is not sufficient to promote mitotic entry. Another distinct mitotic kinase, NIMA, was identified in A. nidulans and showed to have an essential and parallel role in mitotic entry alongside the Cdk1/CyclinB complex. nima5, nima7 and nima1 mutants cause late G2 cell cycle arrest at the restrictive temperature (42 C) (Oakley and Morris, 1983; Osmani et al., 1991; Osmani et al., 1987; Osmani et al., 1988b). Homology based screens have uncovered a large family of NIMA-related kinases (Nrks) in human cells, tremed the Nek genes. In Aspergillus nidulans, NIMA is tightly regulated with the cell cycle with its mrna and protein levels peaking during late G2 where its activity is maximal (Osmani et al., 1991; Osmani et al., 1987). NIMA was shown to carry a role in mitotic entry independent of the Cdk1/cylin B complex kinase activation (Osmani et al., 1991). Interestingly, NIMA is itself regulated by phosphorylation and its activity is dependant on Cdk1 kinase which hyperphosphorylates NIMA in late G2 (Ye et al., 1995). Insights regarding how NIMA mutants arrest in G2 although Cdk1 is active came first from the location of CyclinB. CyclinB was unable to access the nucleus in nima5 cells arrested at the restrictive temperature (Wu et al., 1998). This suggested that NIMA had a role in allowing the cyclinb/cdk1 complex to access the nucleus and phosphorylate its nuclear substrates. Exactly how the functions of both of these kinases are interconnected during mitotic entry and progression is still not very well understood. Later, it was shown that NIMA genetically interacts with two nuclear pore complex proteins SONA and SONB which are the homologs of yeast Gle2/Rae1 and the human Nup98/Nup96 precursor respectively (De Souza et al., 2003; Wu et al., 1998). SONB was shown to be hyperphosphorylated during mitosis in a NIMA dependant manner (De Souza et al., 2004). Furthermore, ectopic expression of NIMA promotes the release of SONB from the pore in S phase cells and also allows the free diffusion of tubulin into the nuclei (De Souza et al., 2004). Thus NIMA promotes the G2/M transition by promoting partial disassembly of the NPCs and allowing mitotic regulators including itself, and tubulin to access the nucleus through diffusion (Figure 1.7 and 1.8). Recently, it has been reported 23

44 that the NPC disassembly is a phosphorylation related process even in mammalian cells. In fact, it has just been published that phosphorylation of Nup98 by multiple kinases including Cdk1, and the Nek family kinases, promotes NPCs disassembly (Laurell et al., 2011). It has also been shown that ectopic expression of NIMA promotes mitotic events such as DNA condensation and spindle formation even in cells arrested in S phase or without Cdk1 activity (Osmani et al., 1988a; Ye et al., 1995). NIMA has a dynamic localization during the cell cycle (De Souza et al., 2004; De Souza et al., 2000). It is cytoplasmic in interphase, moves to the nuclear periphery at the onset of mitosis where it phophorylates Nups and opens the transport channel, it then accesses the nucleus and phosphorylates H3 serine 10 residue promoting chromatin condensation (De Souza et al., 2000). It then moves to the spindle in metaphase and accumulates at the spindle pole bodies in late mitosis and gets degraded during mitotic exit. This dynamic localization of NIMA suggests that it has many roles throughout mitosis and interacts with different proteins which might be potential substrates at different locations. Although insights into how NIMA regulates G2/M transition and promotes DNA condensation have been discovered, there is still much more to be understood. In fact, the mechanism by which it might regulate spindle assembly is unknown The mitotic spindle Spindle assembly and function The mitotic spindle is an intricate macromolecular machine that segregates chromosomes to two daughter cells during mitosis. While differences exist in spindle assembly between different cell types and organisms, all spindles share common structural features. The spindle is made of microtubule polymers. In most mammalian cell types, spindle microtubule nucleation occurs primarily at two centrosomes. In fungi, the spindle microtubules emanate from spindle pole bodies which are embedded in the nuclear envelope. These MTOCs contain -tubulin which forms a multi-subunit ring complex ( -TURC) and works as a template for microtubule formation (Walczak and 24

45 Heald, 2008; Zheng et al., 1995). During mitosis, spindle poles recruit more of these complexes thus increasing their nucleation capacity (Khodjakov and Rieder, 1999; Piehl et al., 2004). The classical search and capture model is the major pathway by which chromosomes become properly attached to the spindle (Kirschner and Mitchison, 1986; McIntosh et al., 2002). In this model, microtubules emanating from the MTOCs randomly probe the cytoplasm until they are captured and stabilized by one of the kinetochores on sister chromatids (Holy and Leibler, 1994). At this stage, these chromosomes are termed monooriented because they are attached to a single spindle pole and remain close to one pole until they become bioriented when they interact with microtubules from the opposite pole. Once bioriented, chromosomes then quickly move toward the spindle equator. The attachments made by the kinetochores to the spindle microtubules are end-on attachments and these spindle microtubules are called kinetochore fibers (K-fibers). In budding yeast, a single microtubule is attached to each kinetochore, although in vertebrates, a single k-fiber is made of a bundle of about 20 to 30 microtubules (Biggins and Walczak, 2003; Cleveland et al., 2003). An opposing model for the classical search and capture is self-assembly. This model proposes that microtubule nucleation starts around chromosomes and occurs independently of centrosomes. These microtubules are then arranged into an antiparallel array which builds the bipolar spindle (Khodjakov et al., 2000; Khodjakov and Rieder, 2001; McKim and Hawley, 1995; Walczak and Heald, 2008). -tubulin complexes have been shown to contribute to both centrosome- and spindle-mediated microtubule nucleation (Haren et al., 2006; Luders et al., 2006). Chromosomes associate with many important regulators of spindle dynamics and promote spindle assembly by regulating both microtubule dynamics and motor proteins (Karsenti and Vernos, 2001; Mitchison and Salmon, 2001; Scholey et al., 2003; Wittmann et al., 2001). This is mediated at least in part by the small GTPase Ran (Dasso, 2002; Hetzer et al., 2002). A RanGTP gradient exists around mitotic chromosomes. This is due to the association of its guanine exchange factor RCC1 to chromatin and the presence of RanGAP, which promotes RanGTP hydrolysis, only in the cytoplasm. RanGTP generated around chromatin stimulates microtubule polymerization and thus the assembly of the spindle at the vicinity of chromosomes (Caudron et al., 2005; Kalab et al., 2006; Kalab et al., 2002). Like its role in transport, RanGTP releases 25

46 its cargoes at the vicinity of chromosomes during mitosis, some of which are spindle assembly factors (SAFs) (Dasso, 2002; Hetzer et al., 2002). The role of transport in mitosis will be discussed further in section Lastly, various motor proteins function in the spindle to cross-link microtubules, move kinetochores and chromosome arms, and regulate microtubule dynamics at the plus and minus ends. Aspergillus nidulans has 11 kinesins and one dynein. Three of the kinesins identified: BimC, KlpA and KipB have been shown to have mitotic roles (Reinhard Fischer et al., 2008) The spindle assembly checkpoint (SAC) Missegregation of chromosomes results in aneuploidy, which can lead to genomic instability and promote cancer (Weaver and Cleveland, 2006; Weaver et al., 2006). Therefore, it is crucial for cells to avoid errors in this process. To accomplish this, the cells use the spindle assembly checkpoint (SAC) to monitor whether all chromosomes are properly attached to the spindle before allowing the cell to proceed into anaphase (Musacchio and Hardwick, 2002). Genetic screens in S. cerevisiae identified mutations in genes that bypass mitotic arrest in cells treated with microtubule poisons. These genes include the MAD (mitotic-arrest deficient) genes, MAD1, MAD2 and MAD3 (BUBR1 in humans) and the BUB (budding uninhibited by benzimidazole) gene BUB1. These genes are conserved in all eukaryotes and are involved in the SAC pathway. An additional major player in the SAC pathway is Mps1p (Monopolar spindle 1). Mps1, Bub1, and BubR1 are kinases whose activation is thought to be an early event in checkpoint signaling. The SAC is activated if the mitotic chromosomes do not establish proper bipolar attachments. Although the exact event which activates SAC is still under debate, being either lack of tension or microtubule occupancy at the kinetochores, the targets of SAC regulation are better understood. The SAC has been shown to target Cdc20, a cofactor of the anaphase-promoting complex/cyclosome (APC/C) which is an ubiquitin ligase. The SAC inhibits the ability of Cdc20 to activate the APC/C-mediated polyubiquitination of many key mitotic substrates including cyclinb, securin and Mps1. This prevents their destruction by the 26S proteosome. Securin is an inhibitor of Separase which is required at the onset of anaphase to cleave the cohesion complex that holds the 26

47 two sister chromatids together. The degradation of cyclinb inactivates Cdk1 activity thus helping promote exit from mitosis. The SAC proteins Mad1, Mad2, BubR1, Bub3 and Mps1 locate at unattached kinetochores in prophase and recruit Cdc20 in addition to the APC/C, inhibiting their functions (Lew and Burke, 2003; Musacchio and Salmon, 2007) Post-mitotic NPC assembly In organisms that undergo open mitosis, the mechanism of post-mitotic NPC assembly is distinct from the de novo assembly of NPCs into an intact membrane at interphase. As discussed in , the de novo assembly of NPCs at interphase requires the fusion of INM and ONM. Conversely, post-mitotic pore reassembly occurs on chromatin before membrane recruitment. Thus, ELYS, a nucleoporin critical for the recruitment of the essential Nup107/160 complex to chromatin, is required for postmitotic NPC assembly but not for the de novo assembly during interphase. On the other hand, the transmembrane Nup, Pom121 is required for the recruitment of the Nup107/160 complex to NPC assembly sites at interphase. In addition, a membrane curvature-sensing domain within Nup133, a member of the Nup107/160 complex, is required for NPC assembly during interphase but not during mitotic exit. As stated previously, the first step of post-mitotic pore reassembly is then the binding of several Nups to chromatin. First, the Nup Mel28/ELYS binds to chromatin and recruits the mnup complex. Two additional Nups are known to associate with chromatin, Nup50 and Nup153, during mitosis and prior to membrane recruitment during mitotic exit (Dultz et al., 2008). These Nups though, are thought to be incorporated later into the NPC as they are part of the nuclear basket. The next step in the reassembly is the recruitment of membrane. In this process, chromatin associated Nups then bind to transmembrane Nups mpom121 and mndc1 forming contacts with membrane. At this point, pore intermediates consisting of mnup , mnup53, mpom121, and mndc1 will be formed which will potentially span the INM and the ONM. In the third step of assembly, additional Nups such as mnup93 and mnup62 subcomplexes are incorporated. The final step of post-mitotic NPC assembly is the incorporation of peripheral Nups such as mnup214, mnup153, Tpr, and mnup50 (Antonin et al., 2008; Doucet and Hetzer, 2010; Hetzer and Wente, 2009). 27

48 Further work is needed to understand how this process is regulated and what the complex protein-protein interactions that occur during the assembly process are An evolving relationship between nuclear transport and mitosis Overview Research conducted in the last decade to decipher the complex series of mitotic events has revealed an important role for the nuclear transport machinery during mitosis. Proteins that have dual roles, one in transport and the other in mitosis, include Nups, Karyopherins, RanGTP and molecules that regulate Ran Role of RanGTP and transport receptors in mitosis In recent years, it has been apparent that spatial and temporal coordination of mitotic events requires Ran, which has been shown to be involved in centrosome duplication, microtubule dynamics, chromosome alignment, kinetochores-microtubule attachments, and nuclear envelope dynamics (Clarke and Zhang, 2008). Major discoveries deciphering the role of Ran in the cell cycle came in 1999, when several studies showed that affecting RanGTP/RanGDP levels in X. laevis egg extracts disrupted microtubule aster and spindle formation (Carazo-Salas et al., 1999; Kalab et al., 1999; Ohba et al., 1999; Wilde and Zheng, 1999; Zhang et al., 1999). These studies provided the first solid evidence of Ran s function in mitosis as the effects observed in microtubule dynamics were clearly distinct from nucleocytoplasmic transport. Later importins and, and more recently another karyopherin, transportin, were implicated in spindle assembly. During mitosis in metazoans, the importin / complex or importin alone binds to NLSs located in spindle assembly factors (SAFs) thereby inhibiting their activities. Ran-GTP then helps the release of SAFs from the importin / complex and thus promotes spindle assembly. Because RCC1 remains chromatin-bound during mitosis, the levels of Ran-GTP are highest around chromatin. This helps the guided release of SAFs in the vicinity of mitotic chromosomes where their activity is needed to 28

49 assemble the spindle. The chromosome high gradient of RanGTP also acts as a spatial signal controlling reactions at a distance from chromatin (Ciciarello et al., 2007; Harel and Forbes, 2004; Kalab and Heald, 2008; Walczak and Heald, 2008; Wozniak et al., 2010). Several SAFs have been identified, mainly using Xenopus egg extracts including TPX2 and NuMA (nuclear mitotic apparatus protein) as well as the kinesin family member HSET/XCTK2, which are mainly involved in spindle pole formation. Other SAFs, including NuSAP and HURP, promote microtubule bundling and stabilization in chromatin proximal regions of the spindle (Cai et al., 2009; Koffa et al., 2006; Ribbeck et al., 2006; Walczak and Heald, 2008; Wong and Fang, 2006; Wozniak et al., 2010). The nuclear pore protein Rae1 (Aspergillus nidulans SONA) was also shown to be a Ran regulated importin- cargo that contributes to spindle assembly. Rae1 exists in a large ribonucleoprotein complex that binds to MTs and controls microtubule dynamics in a RanGTP/importin -regulated manner (Blower et al., 2005; Kraemer et al., 2001). In fact, interaction of Rae1 with NuMA specifically during mitosis was shown to be critical for normal mitotic spindle formation (Wong et al., 2006). The role of Nups in mitosis will be further discussed in the next section. Furthermore, the nuclear export factor CRM1 has been shown to locate at kinetochores through RCC1. CRM1 is proposed to have a role in mitosis through the recruitment of RanBP2 and RanGAP1 to kinetochores. Once on kinetochores, the RanBP2 RanGAP1 complex has been shown to have a role in the interaction of kinetochores with k-fibres (Arnaoutov et al., 2005). CRM1 has been also shown to interact with survivin, a protein of the chromosome passenger complex (CPC), and recruit CPC, including its proteins Aurora B, borealin and INCENP, to centromeres (Knauer et al., 2006). In addition, Ran, RanBP1 and CRM1 also localize to centrosomes, suggesting additional roles for RanGTP in centrosome function during mitosis (Peloponese et al., 2005; Wang et al., 2005) Role of Nups in mitosis There is increasing evidence that, in organisms that undergo open or a semi-open mitosis, the nuclear pore complex proteins have specific mitotic locations once released from the pores. Many Nups have been shown to carry key functions in regulating 29

50 different mitotic processes including spindle formation and anaphase onset. The roles of Rae1 and Nup98 in spindle assembly have been discussed in the previous section. In addition to Rae1, the Nup107/160 subcomplex also locates to spindles and is involved in spindle formation in X. laevis egg extracts (Orjalo et al., 2006). In mitotic mammalian cells, the Nup107/160 complex can also locate to spindle poles and proximal microtubules during prometaphase (Orjalo et al., 2006). Interestingly, a big fraction of the Nup107/160 complex also localizes to kinetochores in both C. elegans embryos (Galy et al., 2006) and mammalian cells (Belgareh et al., 2001; Harel et al., 2003; Loiodice et al., 2004). Depletion of this complex from cells leads to a prolonged prometaphase, chromosome missegregation and delays in onset of anaphase (Zuccolo et al., 2007). In fact, kinetochores stripped of this complex show defects in chromosome attachments, and maintenance of stable kinetochores-mt interactions. Little is known about the molecular mechanisms underlying these defects. Studies have shown that this complex acts as a scaffold at the kinetochores to attract important regulators that are effectors of this complex. These proteins include CENP-F, the RanBP2-RanGAP complex and CRM1 (Arnaoutov et al., 2005; Joseph et al., 2002; Pichler et al., 2002; Zuccolo et al., 2007). Moreover, further studies in mice have revealed that Rae1 and Nup98 do not only have a role in spindle assembly but are also involved in proper timing of securin degradation during mitotic exit. Thus, reduced levels of Rae1 and Nup98 lead to premature sister chromatid separation that results in aneuploidy. This is accomplished by an intimate link between Rae1 and APC/C activity (Guttinger et al., 2009). Rae1 and Nup98 are not the only Nups that have a connection with SAC. Very interestingly, the SAC proteins, Mad1 and Mad2 locate to NPCs in mammalian cells (Campbell et al., 2001), S. cerevisiae (Iouk et al., 2002), and A. nidulans (De Souza et al., 2009), suggesting a functional relationship between the SAC and the nuclear pore. In fact, in A. nidulans, the nuclear pore complex protein Mlp-1 acts as a scaffold to locate Mad1 and Mad2 near kinetochores and the telophase spindle suggesting that it provides a spatial and temporal regulation for these SAC components throughout the cell cycle (De Souza et al., 2009). Recently, the peripheral mammalian Nup, Nup153 has also been shown to play dual roles in both early mitotic progression and resolution of membrane abscission during cytokinesis in HeLa cells (Mackay et al., 2009). Nup153 depletion leads to defects 30

51 in NPC basket assembly which activates an Arurora B-mediated abscission checkpoint (Mackay et al., 2010). Lastly, it seems that there are close ties between the NPC and mitotic progression. Although some of the functions of Nups in mitosis have been discovered, much is yet to be understood. In fact, scientists in the field anticipate more links between Nups and mitosis to be discovered in the future. Mechanistic studies deciphering these links will lead to a better understanding on how mitotic events such as NPC disassembly, spindle assembly, and chromosome segregation are coordinated Evidence from the literature connecting Nup2 with cell division In the semi-open mitosis of A. nidulans, Nup2 undergoes a very unique and exclusive translocation from NPCs to chromatin during mitosis. This was observed by endogenous tagging of An-Nup2 with GFP at its C-terminal (Osmani et al., 2006a). Since then, its mammalian orthologue Nup50 was also shown to be able to translocate to the vicinity of chromatin during metaphase. A fraction of the ectopically expressed MmNup50 tagged with two EGFPs at its N-terminus was shown to localize at the vicinity of chromatin in metaphase and anaphase using NRK cells (Dultz et al., 2008). Nup2 is an essential gene in A. nidulans and its deletion leads to mitotic defects (Osmani et al., 2006a). In addition, an abstract at the 2008 ASCB meeting has reported that RNAi knockdown of Drosophila Nup50 causes mitotic defects and delays in anaphase (S.A. Ribeiro, A.R. Martins, C. Chandsawangbhuwana, R.D. Vale, E.R. Griffis The role of Nups in mitosis: a Systematic RNAi Screen in Drosophila cells. Mol. Biol. Cell 19 (suppl) abstract 2095). Recently, a multiclassifier combinatorial proteomics approach identified Nup50 as one of many Nups present in the protein composition of mitotic chromosomes (Ohta et al., 2010). This approach also identified many uncharacterized proteins purified with isolated mitotic chromosomes, suggesting that our knowledge regarding the molecular mechanisms of mitotic chromosome structure and function is very limited, and our understanding of the intricate coordination of mitotic events with chromosome dynamics is still very poor. Further studies in the field are needed to uncover all the mechanisms that regulate mitotic progression. 31

52 1.4.8 The hypothesis Previous results (section 1.4.7) suggest that Nup2 has a role on mitotic chromosomes conserved from lower to higher eukaryotes. In this study, we undertake the tasks to further understand whether the essential function of A. nidulans Nup2 lies in its ability to associate to mitotic chromosomes away from the NPC, and the challenge to differentiate between its interphase and mitotic functions. Because Nup2 has a function in accelerating cargo release from karyopherins during interphase, we hypothesize that Nup2 s essential function in Aspergillus nidulans is to promote cargo release from karyopherins at the vicinity of chromatin during spindle or NPC assembly. Thus, we propose Nup2 has a role in metaphase spindle assembly and stability or in proper building of daughter nuclei during mitotic exit. In addition, because NIMA regulates the reversible disassembly of NPCs during mitosis, we propose that Nup2 s translocation is regulated by NIMA. Alternatively, we also entertain the thought that Nup2 might carry a function in proper regulation of the 3D structure and organization of chromosomes in daughter nuclei once cells exit mitosis. Proper chromosome organization is crucial for nuclear function and is needed to build proper daughter nuclei. Hence, this study will provide important new insights into the mechanisms by which the NIMA mitotic kinase regulates the transition from G2 into mitosis and additionally extend our understanding of the mitotic roles of NPC associated proteins. 32

53 Figure 1.1 Cartoon depicting the life cycle of Aspergillus nidulans. Aspergillus nidulans is a filamentous fungus that undergoes asexual, sexual and parasexual cycles. Yellow boxes represent parts of the life cycle used for some of the experimental purposes in this study. Image adapted from (Todd et al., 2007a). 33

54 Figure

55 Figure 1.2 Cross-sectional views of a typical mammalian and an Aspergillus nidulans nucleus. Although very similar and highly conserved in structure, components, and function, the Aspergillus nidulans nucleus differs from the mammalian one in various ways. The Aspergillus nidulans nucleus lacks nuclear lamina, its MTOC, called the spindle pole body, is imbedded in the nuclear envelope, and it is 8 to 10 times smaller than the mammalian one. Moreover, its centromeric heterochromatin is tethered to the region next to the spindle pole body. Part A adapted from (Alberts, 2002). 35

56 Figure 1.3 The Architecture of the Nuclear Pore Complex (NPC). NPCs are multiprotein channels that span the nuclear envelope. They are made up of about 30 different nucleoporins that constitute a central core, a nuclear basket and cytoplasmic fibrils. Nup2 is a peripheral Nup which resides in the nuclear basket. 36

57 Figure 1.4 Nup2 protein structure. (A) Cartoon depicting the Aspergillus nidulans Nup2 with various predicted domains and motifs. An-Nup2 contains an N-terminal importin- binding domain, a C-terminal RanGTP binding domain, FG repeats, and several classical NLSs. (B) Protein structures of many Nup2 orthologues. Nup2 is highly conserved from lower to higher eukaryotes. Interestingly, the filamentous fungus Nup2 is larger than its orthologues in other organisms. 37

58 Figure

59 Figure 1.5 The classical nuclear import cycle. NLS bearing cytoplasmic cargo bind to the importin / complex and are transported throughout the pore. In the nucleus, RanGTP binds to importin dissociating the complex and releasing the cargo into the nucleus. Importins are then recycled back into the cytoplasm. Importin is recycled by forming a trimeric complex with Ran and the exportin CAS. 39

60 Figure 1.6 Nup2 facilitates the importin / mediated nuclear import. The Nup2 N- terminus displaces the NLS-bearing cargo from importin. By transferring RanGTP to importin and recruiting CAS and RanGTP to importin, Nup2 further facilitates NLScargo release as well as export of importin. Thus Nup2 facilititates both the termination of nuclear import and the initiation of export. 40

61 Figure 1.7 Cartoon depicting NIMA s function at the G2/M transition in Aspergillus nidulans. Upon mitotic entry, NIMA phosphorylates many of the Nups promoting their release from the pore and the partial disassembly of NPCs. This then allows passive diffusion of mitotic regulators and access of tubulin into the nucleus, and thus leads to DNA condensation and mitotic spindle assembly. 41

62 Figure 1.8 The composition of the Aspergillus nidulans NPC during G2 and mitosis. The NPCs partially disassemble during Aspergillus nidulans mitosis. NIMA phosphorylates many of the Nups and promotes their disassociation from the pore. The core Nups stay at the nuclear periphery during mitosis while the peripheral Nups disperse. Nup2 shows a unique and exclusive translocation to chromatin during mitosis and might be under the regulation of NIMA phosphorylation. Image adapted from (Osmani et al., 2006a). 42

63 CHAPTER 2 MATERIALS AND METHODS 2.1 General Aspergillus nidulans techniques Specific media for culture YG media: (56 mm dextrose, 5 g/l yeast extract, 10 mm magnesium sulfate, supplemented with 1 μg/ml p-aminobenzoic acid (paba), 0.5 μg/ml pyrodoxine HCL (pyro), 2.5 μg/ml riboflavin HCL (ribo), 2 μg/ml nicotinic acid, 20 μg/ml choline, 20ng/ml D-biotin and 1 ml/l trace elements). Strains carrying the pryg89 auxotrphic mutation were grown in YGUU (YG media supplemented with 1.2 g/l uridine and 1.12 g/l uracil). YAG media: (YG media with 15g/L agar). MAG media: (20 g/l malt extract, 20 g/l bacto peptone, 56 mm dextrose, supplemented with 1 μg/ml p-aminobenzoic acid (paba), 0.5 μg/ml pyrodoxine HCL (pyro), 2.5 μg/ml riboflavin HCL (ribo), 2 μg/ml nicotinic acid, 20 μg/ml choline, 20 ng/ml D-biotin, 50 mg/l adenine sulfate, 50 mg/l leucine, 50 mg/l L-methionine, 100 mg/l arginine, 200 mg/l L-lysine HCL, 1 ml/l trace elements and 2% agar). Strains carrying uncomplemented pryg89 auxotrphic mutation were grown on MAGUU (MAG media supplemented with 1.2 g/l uridine and 1.12 g/l uracil). Minimal Media Urea: (10 mm urea, 7 mm potassium chloride, 1 mm magnesium sulfate, 1 ml/l trace elements, and supplements as required). Glucose (final concentration 1% w/v) or glycerol (final 43

64 concentaration 0.47 % v/v) was added prior to autoclaving. Ethanol (final concentration 1% v/v) was added after autoclaving. In addition, potassium phosphate [ph 6.8] (12 mm) and sodium thiosulfate (3.2 mm) were added after autoclaving. For solid media 1.5% w/v agar was added prior to autoclaving. Minimal media Low Nitrate: (82 mm sodium nitrate, 7 mm potassium chloride, 2 mm magnesium sulfate, 11 mm potasssium phosphate monobasic, 111 mm dextrose, 1 ml/l Clive Roberts Trace Elements, additional supplements as required, and 1.5% w/v agar [ph 6.7]). Minimal media Yeast Lactose: (10 mm urea, 7 mm potassium chloride, 1 mm magnesium sulfate, 5 g/l yeast extract, 20 g/l lactose, 1 ml/l trace elements, and supplements as required). Potassium phosphate [ph 6.8] (12 mm) and sodium thiosulfate (3.2 mm) were added after autoclaving. 40 mm threonine was used for alca:: based protein induction Preparation of A. nidulans conidia stock suspensions A. nidulans conidiospores were inoculated at 1X10 7 spores/ml into 4mL of MAG or MAGUU media containing only 0.75% agar at 48 C. Media containing the spores was then vortexed and overlayed onto MAG or MAGUU plates. These plates were incubated at 32 C for 30 to 40 hours until conidia were ready to harvest. Fresh conidiospores on the surface of the plate were harvested in 10mL of 0.2% Tween 20 using a sterile glass spreader. Suspended conidia were transferred to sterile 15mL falcon tubes (Corning). The suspensions were centrifuged at 3,800 rpm for 2 minutes to sediment the conidiospores. Hyphal debris was removed by gently resuspending and recovering only the top conidial layer of the pellet. The collected conidia were washed two times in 10mL of 0.2% Tween 20. After the final wash the conidia were resuspended in stock storage solution (8.5mM sodium chloride, 200μM Tween 80). Concentration of conidiospores in suspension were quantitated to allow for accurate inoculate densities for germination in growth media. Conidial suspensions were quantitated by counting 10 μl of a 1 x 10-3 dilution of conidia in 0.2% Tween 20 using a Bright-Line hemocytometer (Reichert-Jung). Three fields of conidia were counted for each sample, and the average value used for quantitation. The 44

65 number of conidia obtained from this count was multiplied by 1 x 10 7 to determine the concentration of the original suspension in spores/ml. Conidiospore suspensions were stored for up to two weeks at 4 C Long term storage and stock preparation of A. nidulans A single colony was replica plated on a selective media plate using a sterile toothpick and incubated at 32 C for 48 hours and then at room temperature for 3 to 5 days. 5mL of sterile 7.5% milk (7.5g of Carnation Nonfat Dry Milk in 100mLs dh2o and autoclaved for 20 minutes) was added onto the plate and conidiospores were harvested by gently rubbing the top of the fungal lawn to release conidia into suspension using a sterile glass spreader. Then, 250μl of the suspension was transferred into each of two glass vials containing baked, sterile silica and placed on ice for one hour. The silica was then vortexed briefly to evenly distribute the spores and returned to ice for 30 minutes. Subsequently, the silica was left at room temperature for 2 to 3 days with lids loosened to promote complete drying. After 3 days, the silica was again vortexed and then the vials placed in a room temperature desiccator. Strains were re-grown when needed from silica stocks by placing 5-10 silica pieces onto appropriate solid media plates, and incubating at 32 C for several days in an air incubator Strain generation by meiotic crossing A. nidulans strains were forced to undergo meiosis by letting each strain carry at least one forcing auxotrophic marker which is complemented in the other strain. In addition, the parental strains were chosen to be of different conidial colors when possible, providing a visual screen of progeny indicative of successful meiotic crossing events. Parental strains were then alternately spotted on a MAGUU plate, with approximately 2 cm distance between each spots. The plates were incubated at 32 C until the edges of colonies were in proximity of each other. Then, a strip of hyphae at the interface of the two colonies was removed, sliced into pieces, and placed on the surface of a minimal media low nitrate plate. The plate was sealed with tape and incubated at 32 C for a 45

66 minimum of 2 weeks to allow cleistothecia to form. Cleistothecial maturation was monitored using a dissecting microscope (Bausch and Lomb). Once mature cleistothecia were formed, they were collected with a sterilized glass pipette, and rolled across the surface of a 4% water agar plate to clean the surface of the cleistothecia of any hyphal debris. Then, cleistothecia were crushed in 0.2% Tween 20 in a 1.5 ml Eppendorf tube to release ascospores. Ascospores were plated on MAGUU to determine whether the strains had crossed. Individual colonies were selected and tested on a range of minimal media plates lacking various supplements to identify strains with desired genotypes Transformation This technique was previously described by (Osmani et al., 1987). Briefly, 1 x 10 9 fresh conidia were inoculated into 50 ml YGUU, and grown at 32 C in an air incubator with 200rpm for about 5.5 hours to 6 hours or until conidia began to germinate. Germlings were harvested by centrifugation in a swinging bucket rotor at 2,000 rpm for 2 minutes, and resuspended in a protoplasting mix containing 20 ml Solution 1 (105.6 g/l ammonium sulfate, 19.2 g/l citric acid, [ph 6.0]), 20 ml Solution 2 (10 g/l yeast extract, 20 g/l sucrose, 1 μg/ml acid paba, 0.5 μg/ml pyro, 2.5 μg/ml ribo, trace elements, 4.92 g/l MgSO4), 80 mg bovine serum albumin (BSA), 10 mg/ml of the enzyme VINOFLOW FC. The resuspended germlings were transferred to a clean sterile flask, and incubated in an air incubator at 32 C for 2 to 3 hours, or until protpolasts were formed. Protoplasts were detected because their cell walls were degraded making their large vacuoles visible with a light microscope. The protoplasts were then collected by centrifugation in a swinging bucket rotor for 2 minutes at 2,000 rpm, and washed two times in Solution 3 (52.8 g/l ammonium sulfate, 10 g/l sucrose, 9.6 g/l citric acid, [ph 6.0]), and resuspended in 1 ml Solution 5 (44.7 g/l KCl, 7.35 g/l CaCl2, 2.09 g/l MOPS, [ph 6.0]). Transformation was then followed by combining 2-4 μg DNA, 100μL protoplasts, and 50 μl of room temperature Solution 4 (250 g/l PEG 8000, 7.35 g/l CaCl2, 44.7 g/l KCl, 10 ml 1 M Tris [ph 7.5]). The transformation reaction was first incubated on ice for 20 minutes, then an additional 1mL of Solution 4 was added to the mix and incubated at room temperature for 20 minutes. If the selection marker was the 46

67 auxotrophic mutation pyrg89, then 10 μl, 25μL, 50 μl, 100 μl, 250 μl, and 500 μl volumes of the transformation mix was added in 4 ml YAG sucrose (same as below except 7.5 g/l agar) and overlayed onto YAG sucrose plates (5 g/l yeast extract, 3.6 g/l dextrose, g/l sucrose, 2.47 g/l MgSO4, 1μg/mL PABA, 500 ng/ml pyro, 2.5 μg/ml ribo, trace elements, 15 g/l agar). If the selection markers was pyroa4, then different volumes of the transformation mix were added to 4mL mmurea sucrose (as previously described except with 3.6 g/l dextrose, g/l sucrose, and 7.5g/L agar) and overlayed onto mmurea sucrose plates (same as above except 15g/L agar). Transformation plates were incubated in an incubator at 32 C for 70 hours or until colonies emerged. Very importantly, transformations were conducted in nkua ku70 Δ strains (mainly SO451) which have been shown to have a very high frequency of homologous gene targeting (Nayak et al., 2006) Counter selection on 5-FOA Counterselection on 5-FOA was performed when selecting transformants based on the absence of the wild type pyrg gene. The replacement DNA cassette was prepared targeting the pyrg locus. After transformation, different volumes of the transformation mix were added to 4 ml top agar containing YAGUU with sucrose, and transferred into empty Petri dishes. These plates were then left overnight at room temperature. Then the plates were overlaid with YAGUU + 1mg/ml 5-FOA (5FOA (1mg/ml 5FOA) which was added directly to the medium after autoclaving, while still hot. Once all FOA was dissolved, the bottle was placed in 50 ºC water bath). The plates were then incubated at 32 ºC for at least 3-5 days for transformants to emerge Genomic DNA extraction Aspergillus nidulans genomic DNA can be extracted in two ways depending on experimental usage: A small scale extraction or a large scale extraction. For a small scale extraction, a small amount of spores were inoculated in petri dishes with appropriate media, and mycelia harvested by vacuum filtration of the media through Miracloth 47

68 (Calbiochem), and washed twice with cold Stop Buffer (9 g/l sodium chloride, 65 mg/l sodium azide, 20 ml 0.5 M EDTA [ph 8.0], 2.1 g/l sodium fluoride). Excess liquid was pressed out of mycelia which were placed in an eppendorf, and immediately immersed in liquid nitrogen. The mycelia were then removed from the liquid nitrogen and lyophilized overnight. The dried mycelia were crushed, and then 100 μl of Miniprep Lysis Solution (Promega) was added to it and vortexed to mix. 100 μl Miniprep Neutralization Solution was then added and vortexed to mix. Samples were centrifuged at 14,000 rpm in a Model 5420 table top refrigerated centrifuge (Eppendorf) for 10 minutes. The supernatant was processed to isolate the DNA using a Miniprep Purification Kit (Promega) according to the manufacturer s instructions. Genomic DNA was eluted from the column in 50 μl dh2o. For large scale extractions, A. nidulans conidia were inoculated into appropriate media, and incubated overnight in an air incubator at 32 C at 200 rpm. Cells were allowed to grow until a 10 ml sample yielded a packed cell volume of 0.5 ml after undergoing centrifugation at 7,000 rpm. The mycelium was then harvested by vacuum filtration of the media through Miracloth (Calbiochem), and washed twice with cold Stop Buffer (9 g/l sodium chloride, 65 mg/l sodium azide, 20 ml 0.5 M EDTA [ph 8.0], 2.1 g/l sodium fluoride). Excess liquid was pressed out of mycelia which were placed in a 50 ml polypropylene tube (Corning) and immediately immersed in liquid nitrogen for at least 2 minutes. The mycelium was then removed from the liquid nitrogen and lyophilized overnight. Lyophilized mycelia samples were stored at -80 C until DNA extraction. To extract genomic DNA, 20 mg of lyophilized mycelia was thoroughly ground with a disposable pestle in a 1.5 ml Eppendorf microcentrifuge tube. After grinding, 250 μl 0.5% SDS DNA Extraction Buffer (200mM Trizma Base [ph 8.5], 250 mm NaCl, 25 mm EDTA, 0.5% SDS), 175μL phenol, and 75 μl chloroform were added directly to 40 mg of ground mycelia. Tubes were rocked for 15 minutes at room temperature, after which, debris was pelleted by centrifugation at 14,000 rpm for 20 minutes. 400 μl of chloroform was added to the supernatant and spun at 14,000 rpm for 10 minutes. The supernatant was then combined with an equal volume of 5 M lithium chloride, placed on ice for 10 minutes to precipitate RNA, and centrifuged at 14,000 rpm 48

69 for 10 minutes. DNA was precipitated from the supernatant using 2-propanol, and suspended in 50 μl TE Small scale protein preparation Proteins were extracted from A. nidulans in two ways depending on the experiment conducted. Small scale protein preparations were used mainly to check for protein expression by western blotting. On the other hand, large scale protein preparations were used for protein affinity purifications. For small scale protein preparations, conidia were inoculated, at roughly 1x10 6 conidia/ml concentration, into 30 ml of YG or minimal media in sterile petri dishes. Cultures were incubated overnight at 30 C until just before hyphae began to conidiate at the media-air interface. Mycelia were harvested through Miracloth as described above for genomic DNA extraction. Mycelia were then frozen in liquid nitrogen and dried overnight in a lyophilizer (Savant). The next day samples were crushed with toothpicks. The mycelia were then weighed and mixed with 6M Urea Sample buffer (250mM Trizma Base [ph6.8], 7 M Urea, 100 μl/ml β- mercaptoethanol, 200 μl/ml glycerol, 4% sodium dodecyl sulfate) at 40 μl/mg of dried mycelia. Samples were boiled for 5 min prior to analysis by SDS-PAGE. These steps are also explained in (Liu et al., 2010) alca driven protein expression alca driven protein expression was performed either in liquid cultures for protein preparation or on solid media for phenotypic analysis. To induce protein expression, cells were grown either in minimal media yeast extract lactose supplemented with 40mM threonine, or minimal media with 1% ethanol. Minimal media with 1% glucose was used for repressing the promotor. Minimal media with 0.47 % glycerol was used for nonrepressive/ non-inductive conditions. 49

70 Cell fixing and DAPI Staining 4',6-diamidino-2-phenylindole (DAPI) staining of DAN was performed as described earlier with some modifications (Oakley, 1993). Briefly, germlings were fixed on culture dishes using 4% gluteraldehyde, 0.2% NP40 and 50mMPO4 for 15 to 20 minutes. The cells were then washed once with 1xPHEM (45mM PIPES, 45mM HEPES, 10mM EGTA, 5mM MgCl, [ph 6.9]), and twice with distilled H2O, and stained with 0.015μg/ml DAPI in distilled water. 2.2 Fusion PCR and gene targeting in Aspergillus nidulans The fusion PCR-based method of amplifying gene-targeting constructs eliminates the need to perform ligation reactions required in conventional cloning strategies. This method is used to produce many different types of gene-targeting constructs used to generate A. nidulans strains. Here is a brief description of the two most common targeting constructs used. Detailed description of this method can be found in (Szewczyk et al., 2006; Yang et al., 2004). Specific constructs made by this method will be discussed in individual chapters Gene deletion constructs To delete the entire open-reading frame of nup2 or nupa, three different DNA fragments were amplified (Figure 2.1). The first fragment was the nutritional marker cassette consisting of the promotor, the ORF and the terminator of a heterologous (but functionally complementary) A. fumigatus pyrg gene (Af-pyrG) to minimize the chance of the integration of the construct into the native A. nidulans pyrg locus. In addition, two fragments of the genome sequences immediately upstream and downstream of either nup2 or nupa (~1.0KB in length) were PCR amplified as targeting regions. 3-way fusion PCR was then performed to construct the deletion cassettes. Details about how the fusion PCR works is described in (Szewczyk et al., 2006; Yang et al., 2004). Upon transformation of these constructs, transformants were tested for proper replacement by 50

71 making small scale genomic preparations and performing diagnostic PCRs using appropriate primers that target sequences outside of these constructs Endogenous C-terminal tagging constructs The DNA constructs used for endogenous C-terminal protein tagging were all generated using 3-way fusion PCR (Figure 2.2). GFP and ChRFP tagging cassettes, as well as the affinity S-tagging cassette, were amplified with the primer pair HP116 and FN-01-PyrG. This was made possible due to the common GA5 linker and a stretch of AfpyrG sequence available in all plasmids generated that carry different combinations of tags and markers. Gene specific primers were used to amplify the 5 -flanking as well as the 3 -flanking sequences of specific genes whose encoded proteins were tagged. Once the tagging constructs were amplified, transformation was carried out according to the standard protocol. 2.3 Nup2 antibody generation and immunofluorescence A rabbit polyclonal antibody was generated against the aa region of Aspergillus nidulans Nup2 which was expressed and purified from E. coli (Figure 2.3). Briefly, an antigenic, hydrophobic domain of Nup2 (aa ), uniquely found in A. nidulans, was cloned into the bacterial expression vector Pet14b which carries a His.tag (Novagen, Merck KGaA, Darmstadt, Germany) and expressed in E. coli strain BL21(DE3). The expression of the Nup2 domain was induced with IPTG, and the fusion protein was extracted from E. coli by sonication. The lysate was centrifuged and (Figure 2.1B) His-tag purification of the Nup2 antigen from the supernatant was performed using Ni-NTA His.bind resin containing columns (Figure 2.1C and D). The purified Nup2 was used to immunize male rabbits. The α-nup2 affinity-purified antiserum was produced by Bethyl laboratories, Inc., Montgomery, Tex. Immunofluorescence was performed as described in (Oakley, 1993) with some modifications. Briefly, cells were grown on rich YG media then fixed in 1xPHEM buffer (45mM PIPES, 45mM HEPES, 10mM EGTA, 51

72 5mM MgCl, [ph 6.9]) containing 6% paraformaldehyde (EM grade; Electron Microscopy Sciences). Digestion solution contained 10mg/ml Vinoflow FCE. (Novozymes A/S, Bagsvaerd, Denmark). Cells were digested for 50 mins at 30 C. The antibody generated against Nup2 was used at a dilution of 1/5, Western blot analysis Western blot analysis was performed to estimate protein expression levels and to verify proper integration of targeting constructs by confirming the protein size of chimeric proteins. Small scale or large scale prepared proteins in SDS protein sample buffer were loaded on 7-10% SDS-polyacrylamide gels along with rainbow recombinant protein molecular weight markers (GE Healthcare). Gel electrophoresis was run at 30mAmp, and proteins subsequently transferred to nitrocellulose membrane using a protein gel transfer apparatus at 180mAmp for 3 hours. The membrane was further processed for protein detection by blocking in 5% milk (5.0 g Carnation Nonfat Dry Milk in 100mL of TBS buffer 200mM Tris base, 5M NaCl ph 7.5) for 1 hour followed by 1 hour incubation with primary antibody made in 5% milk. The blot was then washed 3 timeswith TBST buffer (TBS buffer with 0.5% Tween 20). Secondary antibody (in 5% milk) incubation was performed for one hour followed by 3X washes in TBST buffer. An enhanced chemiluminescence reagent kit for western blot analysis was used to detect proteins. The primary antibodies used were anti-nup2 (1:5000), anti- GFP (Living colors 1:5000), anti-dsred (Living colors - 1:10,000), anti-s (Immunology consultancy laboratory 1:5000) and anti-tubulin (Sigma B512 1:5000). Secondary antibodies respectively used were anti-rabbit (1:5000), anti-mouse (1:5000), and anti-rabbit (1:5000). All secondary antibodies are ECL peroxidase labeled antibody from GE healthcare. 52

73 2.5 Heterokaryon rescue technique for the identification and analysis of essential genes in Aspergillus nidulans The heterokaryon rescue technique was generated in our laboratory to positively identify essential genes and to perform phenotypic characterization of such null alleles. The technique is fully described in (Osmani et al., 2006b). Briefly, heterokaryons were generated by transforming pyrg89 recipient strains carrying nkua ku70 Δ (Nayak et al., 2006), with Nup2::pyrG Af or NupA::pyrG Af linear constructs as described in section Heterokaryons, which produce two types of uninucleate conidia, were identified by streaking transformants on rich YG media with or without Uridine and Uracil. Only Nup2::pyrG Af or NupA::pyrG Af spores could germinate in the absence of uridine and uracil (UU). Media lacking UU was used to study the phenotypes of Nup2 and NupA nulls. These heterokaryons were propagated by cutting a small portion of mycelia from the growing edge of a colony, and placing it onto YAG plates. To maintain and store heterokaryons, sections of the leading edge of heterokaryons, including the agar, were removed, placed in a sterile microfuge tube, and stored at -80 C. 2.6 Microscopy and image acquisition Fixed samples Fixed samples were examined using an E800 microscope (Nikon, Inc.) with DAPI, FITC, and Texas Red filters (Omega Optical, Inc.). Image capture was performed using an UltraPix digital camera (Life Science Resources, Ltd.) and analyzed by Ultraview image capture software (Perkin Elmer) Live cell imaging To examine fluorescently tagged proteins in living cells, cells were germinated in 35 mm glass bottom Petri dishes (MatTek Cultureware). Wild type haploid strains were germinated in 3 ml of minimal media at room temperature (21-23 C) for 24 hours or until 53

74 the 4 to 8 nuclear stage was reached. For nup2 and nupa null mutant analysis, spores from heterokaryons were germinated at room temperature (around C) in YG medium with or without UU for about 8 hours (to image first mitosis) or overnight (12 to 16 hours, to image the 2 nd, 3 rd mitosis and terminal phenotypes). Visualization of GFP/chRed fusion proteins was performed using inverted microscopes (Nikon), configured with an Ultraview spinning disk confocal system, and controlled by Ultraview software (Perkin Elmer) utilizing Nikon Plan Apo 60XA/1.40 oil objectives. All live cell imaging was carried out at room temperature (microscope room temperature was around 24 C). Image analysis, kymograph generation and pixel intensity profiles were carried out using Image J freeware (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, ). All images are represented as maximum intensity projections captured at 2x2 binning as recently described in (Ukil et al., 2009). 54

75 Figure 2.1 Gene deletion constructs. To construct gene deletion cassettes, the pyrg AF gene is amplified from the pfno3 plasmid (Piece 1). Then, 1 kb upstream and downstream of the gene to be deleted is amplified using gene specific primer pairs GSP1/2 (Piece 2) and GSP3/4 (Piece 3). 3 way fusion PCR of the 3 pieces makes the gene deleted cassette which is transformed in a strain that carries pyrg89 mutation in the background. The cassette then replaces the gene of interest by homologous recombination. 55

76 Figure 2.2 Endogenous C-terminal tagging constructs. Each tag with a specific marker is amplified using the primers HP116/FN-01-PyrG (Piece 1). Then, 1 kb upstream and downstream of the end of the gene to be tagged, is amplified using gene specific primer pairs GSP1/2 (Piece 2) and GSP3/4 (Piece 3). 3 way fusion PCR of the 3 pieces makes the cassette which then tags the gene of interest by homologous recombination. 56

77 Figure 2.3 Expression and purification of an antigen against Nup2. (A) Coomassie stained SDS-PAGE gel showing the domain of An-Nup2 (aa ) expressed in E. coli at the predicted size of 24kd. Small fractions of bacterial cultures have been run before (NI: Not Induced) and after induction (I: Induced) of the Nup2 domain with IPTG. (B) Coomassie stained SDS-PAGE gel showing a fraction of the soluble supernatant isolated (SN) and a fraction of the bacterial culture (BC) run next to each other from induced E. coli samples. (C) His-tag purification of the Nup2 antigen using a Ni-NTA His.bind resin containing columns. 1-3μgs of the flow through (FT), the first 2 washes (W1 and 2), and the elution fractions 3, 4 and 5 were run on an SDS-PAGE gel stained with Coomassie blue. (D) Graph depicting different elution fractions vs concentration of the Nup2 purified protein. 57

78 CHAPTER 3 NUP2 AND A NEWLY DISCOVERED NUCLEAR PORE COMPLEX PROTEIN, NUPA, PLAY ESSENTIAL ROLES AT MITOTIC CHROMATIN CONTROLLED BY THE NIMA KINASE. 3.1 Introduction During nuclear division, the cell undergoes a highly coordinated series of events to properly segregate its components between two daughter nuclei. This immensely intricate network of events, and how they are regulated and coordinated, is not fully understood. Recent evidence shows that in organisms that undergo open or semi-open mitosis, several Nup proteins carryout additional functions at distinct mitosic locations in addition to their previously described conventional nuclear transport roles. Once released from the pores, many Nups have been shown to not only have distinct localizations at mitotic structures, such as the spindle poles and kinetochores, but also fulfill key functions in regulating different mitotic processes at those locations (De Souza and Osmani, 2009; Guttinger et al., 2009). Future discoveries regarding new links between Nups and mitosis, particularly mechanistic insights that help decipher how mitotic events such as NPC disassembly, spindle assembly, and chromosome segregation are coordinated, would provide essential new knowledge regarding our understanding of how the cell cycle functions. Since the isolation of the first Nups in late 1980s and early 1990s 58

79 (Davis and Blobel, 1986; Davis and Fink, 1990; Hurt, 1988), the last two decades has seen an increased body of work introducing previously unidentified Nups, and deciphering their functions. Although the more recent availability of proteomic analysis has enabled researchers to identify the majority of Nups present in S cerevisiae and mammals (Alber et al., 2007b; Cronshaw et al., 2002; Rout et al., 2000), very recent discoveries of previously uncharacterized Nups (Chadrin et al., 2010) is making it clear that a comprehensive list of Nups present in all organisms is yet not achieved, indicating further Nups might yet be discovered. We have previously identified by sequence similarities the Aspergillus nidulans Nup2 orthologue and have shown that it has a unique and exclusive translocation from the interphase NPCs to mitotic chromatin. We have also shown that unlike its S. cerevisiae orthologue, An-Nup2 is essential and its deletion causes defects in DNA segregation (Osmani et al., 2006a).The mammalian orthologue (Nup50) was also recently shown to locate at the vicinity of chromatin only during mitosis (Dultz et al., 2008). This suggests that Nup2 might have a unique mitotic function at chromatin which might be conserved in higher eukaryotes. Because Nup2 has a known function in accelerating cargo release from karyopherins during interphase, I suggest Nup2 s essential function in Aspergillus nidulans might lie in its ability to promote cargo release from karyopherins at the vicinity of chromatin in spindle formation during mitosis, and/or in NPC assembly during exit from mitosis into G1. We have also previously described the identification of most of the components of the Aspergillus nidulans NPC based on sequence similarities to S. cerevisiae, S. pombe, C. elegans and mammals (Osmani et al., 2006a). This approach though, would not identify very weakly conserved Nups. Indeed, recent affinity purifications of the conserved Nup subcomplex identified two new fungal Nups which were shown to be orthologues of Nups that was previously thought to be vertebrate specific, Nup37 and ELYS (Liu et al., 2009). In the same work, Liu et al. showed that all three characterized transmembrane Nups are dispensable in Aspergillus nidulans suggesting the presence of additional novel transmembrane Nups yet to be discovered. These data suggest that the NPC component inventory of Aspergillus nidulans may not yet be complete. Here, using affinity purifications of the Aspergillus nidulans Nup2, we identify a previously uncharacterized essential A. nidulans nucleoporin, we call NupA. NupA, like Nup2, translocates from 59

80 interphase NPCs to mitotic chromatin suggesting that both of these proteins might carry essential functions at chromatin during mitosis. In this study we show that both Nup2 and NupA are essential for normal mitosis and that their deletions do not cause any defects in the classical nuclear import pathway but cause marked mitotic defects that engage the SAC. NupA s essential function is found to lie in its ability to locate Nup2 to both the NPC and mitotic chromatin. We further show that Nup2 is a mitotic phosphoprotein which is regulated by NIMA suggesting that it might act as a downstream effecter of NIMA s function in promoting mitosis. We also observe that Nup2 and NupA associate with chromatin in a dynamic manner. They first get directly associated with all chromatin during prophase and then move from being on chromatin to locate around it at anaphase. This movement at anaphase might be functionally significant as deletion of either protein causes defects in mitotic exit and the regeneration of normal post-mitotic daughter nuclei. We therefore suggest Nup2 and NupA have dual roles. The first is during mitosis in a process which is monitored by the SAC, and a second function later during mitotic exit and early G1 involving generation of daughter nuclei. 3.2 Material and Methods Single-Step Affinity Purification of Nup2 in G2 and M Synchronization of cells and generation of G2/M samples For G2 arrest, 2.5x10 6 conidia/ml of the strain SO926 (Nup2-S-Tag) carrying the nimt23 mutation was inoculated in 1000 ml YG media in a 2000 ml siliconized (Gel Slick Solution, Rockland, Inc.) conical flask. The sample was then incubated with shaking (250 rpm) overnight (~ 14-16h) at 28 ºC until grown to a packed cell volume of ~ ml per 10 ml and then transferred into a water bath heated to 55 C. The cultures were then incubated in the water bath with mixing to bring it up to 42 C. Upon reaching that temperature, the cultures were immediately transferred to a 42 C air shaker. The cultures were then incubated for 3 hours at 42 C. After the 3 hour incubation time, 500μL of the cultures were set aside for measuring the chromosome mitotic index while the rest 60

81 was harvested and lyophilized using standard methods (As described in Chapter 2, section 2.1.7). The sample put aside was immediately fixed and DAPI stained. For mitotic arrest, cultures were shifted to 42 C in the same manner and benomyl (2.4μg/mL) added to the samples after 2 hours and 45 minutes of incubation at 42 C. After benomyl addition, the cultures were left for 15 more minutes at 42 C, and then cooled rapidly in an ice/water bath until they reached 30 C. The cultures were incubated at 30 C for 40 more minutes before being harvested as above. A 500μL sample of the mitotic culture was also set aside for measuring the chromosome mitotic index Protein preparation for purifications A 1000 ml culture typically yielded g of dry mycelia. The dry mycelia were ground in a mortar and pestle to a fine powder, and 0.1 g dry weight was added to 1.3 ml of cold HK extraction buffer. Samples were kept on ice after this step. The samples were vortexed for 3x 10 sec, placing samples on ice in between each vortexing step to keep samples ice cold. Then, the extract was centrifuged at 21,000 rpm (53,200 xg) at 4 C, for 30 min. The supernatant was removed and centrifuged at 21,000 rpm (53,200 xg) for 10 min. The supernatant was removed and protein concentration estimated using the Bradford method. The typical protein concentration was mg/ml. Samples were used directly for S-tag purification. These protocols are also explained in more detail in (Liu et al., 2010) S-Tag purification 300 µl of S-Protein Agarose slurry was added (150 μl packed bead volume) (Novagen) per 100 mg of protein. Sample were mixed on ice at 4 0 C for 1 hour and centrifuged to pellet the beads. The beads were then washed 1x with an equal volume of HK buffer (containing 0.2% NP40 and 50μg/ml PMSF). Additional 6x washes were followed using an equal volume of HK buffer. At this point, the beads were transferred to Eppendorf tubes and washed 2x in 1ml HK buffer. Then, ¼ the final bead volume of 4X SDS-PAGE sample buffer was added to the beads (e.g. for 150 μl beads add 50 μl), 61

82 vortexed to mix, and boiled for 5 min. The samples were then centrifuged to pellet the beads and the supernatant loaded on a SDS-PAGE gel Coomassie staining of protein gels Polyacrylamide protein gels were fixed in 50% ethanol/10% acetic acid (minimum 4 hr). The gels were then washed in 50% ethanol for 30 min and rinsed 3x with dh 2 O for 5 min each. The residual water was removed and enough Bio-Safe Coomassie (Biorad) added to cover the gel. Staining was continued for 1 hour with gentle agitation, and background staining was removed by destaining in water for 30 minutes In vitro λ-phosphatase assay After S-tag affinity purification, three tubes with equal amount of beads were prepared for no phosphatase, phosphatase and phosphatase + phosphatase inhibitor treatments. To initiate the λ-phosphatase assay (New England Biolab), the individual samples were washed in 2X phosphatase buffer (i.e. volume of phosphatase buffer equal to volume of beads) which included MnCl2 and protease inhibitors (prepared according to manufacturer s protocol). After washing, the three samples were treated with control buffer and equal volumes of either 1x phosphatase buffer without phosphatase, or 1x phosphatase buffer with phosphatase (40 units λ-phosphatase), or 1x phosphatase buffer with phosphatase and phosphatase inhibitors (1mM Na Vanadate and 50mM NaFluoride). The beads were incubated at 30 C for 30 minutes (mixing occasionally). Once the treatment was completed, sample buffer was added to a final concentration of 1X and boiled to stop the reaction. The supernatant was collected after pelleting of the beads by centrifugation at 14,000rpm and analyzed by SDS-PAGE NIMA induction experiments To follow the state of Nup2 upon NIMA induction, conidia from strains HA377 (WT for NIMA) and SO1030 (alca-nima-δc-stag) were inoculated overnight in 1L minimal medium containing glucose at 26 C. The cells were grown until 10ml of culture 62

83 gave a PCV of about 0.1 to 0.2ml. The cultures were then centrifuged for 12 minutes at 10,000 rpm, washed twice with minimal media without carbon, and shifted into minimal medium containing 0.5% ethanol. Mycelial samples were taken every 15 minutes, washed with stop buffer and frozen in liquid nitrogen. Extracted whole cell protein samples were then run on SDS-PAGE and western blot analysis performed with NIMA antibody (F68) at a dilution of 1/2000, and the Nup2 antibody generated (Chapter 2) at a dilution of 1/5000. To follow Nup2-GFP upon induction of NIMA, strains HA377 and SO1030 were grown overnight at room temperature in 3 ml minimal medium containing glucose. The next day, the medium was replaced with either 3 ml minimal medium Glucose (with or without 20mM Hydroxyurea to synchronize cells at S phase) for repressing conditions or they were washed 2 times with minimum medium without carbon and supplied with 3 ml minimum medium containing 0.5% ethanol (with or without hydroxyurea) to induce NIMA expression. The cells are then left for 2 hours. After 2 hours, Nup2-GFP and NLS-DsRed were followed in these cells using a 60 /1.4 NA objective lens (Nikon) on a spinning-disk confocal microscope (UltraVIEW ERS; PerkinElmer) with a 488-nm and 568-nm argon ion lasers, and a cooled charge-coupled device camera (ORCA-AG; Hamamatsu). Graphs were plotted depicting the cumulative number of cells that had Nup2 on DNA with time NupA sequencing and 3 RACE for NupA 5 and 3 RACE of NupA was performed following the Marathon-Ready cdna user manual (Clontech, CA) with minor modifications. The 5 mrna end was first amplified using AP1 and SM61 from adapter ligated A. nidulans cdna. The 3 end was first amplified with AP1 and SM64. Touch down PCR conditions were used for a hot-lid thermal cycler as described in the manual, except 30 cycles were run in the 4 th step using the Expand long template enzyme (Roche Applied science, Mannheim, Germany). 5μl of this reaction was then used for a consecutive PCR, using the nested primers AP2 and SM62 (5 end) or SM63 (3 end). The PCR products were run on an agarose gel, 63

84 individual bands purified and cloned in pcr 2.1-TOPO vector (Invitrogen, CA). Selected clones were sequenced at the Plant-Microbe genomics facility, OSU, OH Genomic and cdna sequencing of NupA cdna and genomic DNA of NupA was amplified from either A. nidulans cdna libraries or genomic DNA with primers SM43/45. For both reactions, Pfu-turbo DNA polymerase (Stratagene, CA) was used for high-fidelity PCR amplification. For cdna isolation, 5 samples of two independent libraries of 200ng/l concentration were used in 50μl reactions with an annealing temperature of 60 C and an elongation time of 6 minutes. For the isolation of genomic DNA, 2μl of genomic DNA was used in a 50μl PCR reaction with an annealing temperature of 60 C and an elongation time of 6 minutes. The resulting PCR products were run on an agarose gel, bands purified and cloned in pcr 2.1-TOPO vector (Invitrogen, CA). Clones were chosen and sequenced at the Plant-Microbe genomics facility, OSU, OH Identification of the intron in the Aspergillus fumigatus NupA To identify the the 5 intron in A. fumigatus nupa, primers SM67/68 were used to PCR amplify the region that spans the intron using an A. fumigatus cdna library (a gift from Gregory S. May, Professor, Department of Laboratory Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX). The PCR products were run on an agarose gel, bands purified and cloned into pcr 2.1-TOPO vector (Invitrogen, CA). Clones were chosen and sequenced Quantification Quantification of dispersal and import of NLS-DsRed To calculate the rate of nuclear dispersal and import, cells were grown in YG with (for WT) or without UU (for nup2 and nupa) at room temperature (21-23 C) overnight (12-16hrs), and the NLS-DsRed signal was followed during mitosis. Images 64

85 were captured every 10 seconds using a 60 /1.4 NA objective lens (Nikon) on a spinning-disk confocal microscope (UltraVIEW ERS; PerkinElmer) with a 568-nm argon ion laser, and a cooled charge-coupled device camera (ORCA-AG; Hamamatsu). Maximum pixel intensity (MPI) of z sections spaced at 0.8 μm was measured using Image J freeware (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA). The normalized fluorescence intensity for each time point captured was measured as: Mean MPI in an area (A) (nucleus) - Mean MPI in A (cytoplasm) X 100 highest mean MPI in A Graphs were plotted (using Excel, Microsoft, WA, USA) as mean of fluorescence intensity measured (8 to 13 nuclei for each genotype) Quantification of time spent in mitosis To calculate time spent in mitosis, cells were grown on YG with (for WT, mad2) or without UU (for nup2, nupa, nup2 mad2, and nupa mad2) at room temperature (21-23 C) overnight (12-16hrs), and tubulin-gfp signal was followed throughout mitosis. Images were captured every 30 seconds using a spinning-disk confocal microscope (details above). Time in mitosis was measured starting the time point where the spindle is visible and until it disassembles. Number of nuclei measured varied according to the sample. For wild type, N= 56, Mad2 deletion N=58, nup2 N=165, nup2 mad2 N=83, nupa N=88, nupa mad2 N=48. Statistical analysis was conducted using PASW statistical software, SPSS Inc., Chicago, Illinois, USA. P values were calculated using the student T-test considering a p value 0.05 as significantly different with 95% confidence intervals and a p value 0.01 as highly significant with 99% confidence. Δmad2 nup2 and Δmad2 nupa mutants showed no statistical difference when compared to WT or Δmad2 mutant. On the other hand, nup2 and nupa cells spend significantly longer time in mitosis compared to WT (with a p value 0.01). 65

86 Quantification of nuclear movement Cells were grown in YG with UU at room temperature (21-23 C) overnight (12-16hrs) and nuclear movement was quantified following H1-chRFP every 45 seconds using a 60 /1.4 NA objective lens (Nikon) on a spinning-disk confocal microscope (UltraVIEW ERS; PerkinElmer) with a 568-nm argon ion laser, and a cooled chargecoupled device camera (ORCA-AG; Hamamatsu). Graphs were plotted (using KaleidaGraph, Synergy Software, PA, USA) by measuring the distance from a set point, usually at the head of the germling, to the center of the nucleus during time. To depolymerize microtubules, YG media was replaced with YG + benomyl (2.4μg/mL). 3.3 Results Nup2 is phosphorylated and locates to chromatin during mitosis During A. nidulans mitosis NPCs undergo partial disassembly and peripheral Nups disperse throughout the cell as nuclear division proceeds. Amongst the Nups that are removed from NPCs during mitosis, Nup2 is unique as it probably all translocates from NPCs onto chromatin during mitosis. During completion of mitosis, chromatin associated Nup2 relocates back to NPCs (Osmani et al., 2006a). Prior Nup2 localization studies utilized functional endogenously GFP-tagged versions of Nup2. To confirm that the endogenous untagged Nup2 undergoes these dynamic mitotic relocations, an antibody against A. nidulans Nup2 was generated (see Chapter 2: materials and methods) and used in immunofluorescence microscopy to confirm that endogenous untagged Nup2 behaves in the same manner as the Nup2-GFP version (Figure 3.1A and B). To biochemically monitor Nup2, an endogenously C-terminally S-Tagged version (Liu et al., 2010) was generated. Nup2-S-Tag (as well as the Nup2-GFP version) was found to migrate at an unexpected high molecular weight (predicted size is kd; actual size is between 160 and 250 kd), but the basis for this discrepancy is currently unknown. Notably, both mitotic Nup2 (Figure 3.1A) and Nup2-S-Tag (Figure 3.2A) appeared as a smear during SDS PAGE when compared to the defined band from G2 samples, which suggested Nup2 might be phosphorylated during mitosis. After treatment 66

87 in vitro with phosphatases the Nup2 smear collapsed into a single band showing that Nup2 is indeed a mitotic phosphoprotein (Figure 3.2A). Hence, Nup2 translocates from the NPC to mitotic chromatin in the phosphorylated state. This suggests that Nup2 is under the regulation of one or many mitotic kinases such as Cdk1 and NIMA. NIMA is an essential kinase in A. nidulans required to initiate mitosis in parallel with Cdk1 (Oakley and Morris, 1983; Osmani et al., 1991; Osmani et al., 1987). The mechanism by which it triggers mitosis lies in its ability to induce partial disassembly of NPCs (De Souza et al., 2003; De Souza et al., 2004). We have previously shown that ectopic NIMA expression promotes NPC disassembly (De Souza et al., 2004). To investigate whether NIMA also induces Nup2 s phosphorylation and relocation from the NPC to chromatin, we ectopically induced NIMA and detected Nup2 protein phosphorylation state by the Nup2 antibody using western blot analysis. Cells were grown in the presence or absence of hydroxyurea (HU), which synchronizes cells at the S phase. NIMA induced Nup2 s phosphorylation and relocation from the NPC to chromatin even in cells blocked at the S phase, showing that NIMA s function is independent of the cell cycle phase. Upon induction of NIMA, the Nup2 band upshifts (Figure 3.3A) and shows a smear very similar to what we have observed before during mitosis (Figures 3.2A and 3.1A). A delay is observed between the induction of NIMA and the phosphorylation of Nup2 especially in the presence of HU, which raises the question whether the phosphorylation is direct (Figure 3.3A). We then monitored Nup2-GFP by live cell microscopy and found that induction of NIMA promotes the relocation of Nup2 from the pore to DNA even in S phase arrested cells (Figure 3.3B and C). These results show that induction of NIMA not only promotes Nup2 s phosphorylation but also its relocation from the NPCs to DNA independent of the cell cycle phase Identification of a Novel Nucleoprin, NupA Affinity purification of Nup2 during G2 and mitosis and isolation of NupA Affinity purifications followed by Mass Spectroscopy analysis of Nup2-S-Tag and its copurifying proteins identified importins KapA (ANID_2142.1), KapB (ANID_0906.1), as well as to two hypothetical proteins ANID_ and 67

88 ANID_ These proteins copurified with Nup2 from G2 samples as well as mitotic samples (Figure 3.2A). Within the annotated genome sequence ( ANID_ falls between and bp at the end of contig 51 ( ) and ANID_ falls between 160 and 2077bp at the start of the contig 52 ( ). pblast with the coding sequences of these two genes gives a single hit in Aspergillus fumigatus: Afu3g1330, a putative 1299aa protein with a predicted molecular weight of 140.5Kd. This suggests that ANID_ and ANID_ are the 5 and 3 ends of a single gene that falls between two consecutive contigs. Confirming this we were able to amplify a fragment of DNA from genomic DNA that includes sequences from both ANID_ and ANID_ We then isolated and sequenced cdna to define a new protein we termed NupA, a 768aa protein with a predicted molecular weight of 84.2Kd, a much smaller protein than its predicted orthologue in A. fumigatus (Figure 3.2C). In addition 5 RACE and 3 RACE were conducted to confirm the start and stop codons of the NupA ORF. During SDS PAGE, NupA, like Nup2, ran as a smear in the mitotic sample and in vitro phosphatase treatment showed this to be dependent on mitotic specific phosphorylation. NupA was endogenously S-Tagged and affinity purified, to reveal the same sets of proteins as those purified in association with An-Nup2 in the G2 as well as the mitotic sample (data not shown) cdna sequencing of NupA reveals an unusually long intron which is conserved in Aspergillus fumigatus, Aspergillus niger, Aspergillus aculeatus and Aspergillus carbonarius. Comparing the cdna sequence to genomic sequence for nupa identified three introns, the first being rather large (800 bp) for fungal introns. To investigate whether A. fumigatus nupa contains a similar large intron, we sequenced the region spanning the predicted intron from an A. fumigatus cdna library. Similar to A. nidulans, A. fumigatus nupa also contains a very large (1386 bp) intron (Figure 3.2B) and, somewhat surprising, maintains the reading frame of NupA. Very interestingly, Aspergillus niger, Aspergillus aculeatus and Aspergillus carbonarius nupa s also show evidence of a large ~1kb intron in the same region. In fact, there is RNA-sequence data supporting the existence of this 68

89 intron in each of these species devoid of in frame stop codons (personal communication with Scott Baker, Pacific Northwest National Laboratory, Richland, WA). In addition, tblastn with the A. fumigatus coding sequence of the intron using the Aspergillus comparative database with no filters showed that it has 90% identity with a transcript in N. fischeri (score:811 bits, Evalue 0.0), 47% identity with a transcript in A. clavatus (score: 318 bits, E value 2e-86) and 26% identity with a transcript in A. terreus (score: 65.1 bits, E value 3e-10), suggesting that this intron has some potential to be coding. A. nidulans, A. fumigatus and A. niger NupA contain several potential classical and bipartite nuclear localization signals (Figure 3.2C) but no other known motifs were found in either (Figure 3.2C). Optimal Global alignments of A. nidulans with the A. fumigatus NupA, and of the A. nidulans with the A. niger NupA proteins show 46% and 49.5% identity respectively (Figure 3.2 C) NupA is an essential gene whose deletion does not affect short term growth but does cause mitotic defects Deletion of nupa generated either diploids or heterokaryons but not a ΔnupA mutant haploid demonstrating it to be an essential gene. Using heterokaryon rescue (Osmani et al., 2006b), heterokaryons which carry two genetically distinct nuclei in one cytoplasm, one which is ΔnupA and the other wild type, were identified by streaking asexual uninucleate spores (conidia) from nupa deletion transformants on plates with or without Uridine and Uracil (UU) (Figure 3.4 A-a). Diagnostic PCR of the nupa locus using genomic DNA extracted from the heterokaryon confirmed the presence of two genetically distinct types of nuclei (Figure 3.4 A-b). The ΔnupA mutant shows no defect in germination, or short term growth, and grows similar as a wild type for one day at room temperature, but stops growing after several rounds of defective mitosis (Figure 3.4 B-C). The nuclear DNA of the ΔnupA mutant appeared mis-segregated and clustered into regions within the germlings. Some micronuclei are also observed (Figure 3.4 C). These phenotypes are very similar if not identical to the Δnup2 mutant (Osmani et al., 2006a) which suggests that Nup2 and NupA might have interdependent functions involving mitosis. 69

90 3.3.4 NupA translocates from NPCs to chromatin during mitosis To determine the localization of NupA we endogenously tagged it with GFP. The C-terminally GFP tagged version of NupA was fully functional as cells were viable and no growth defect was observed. We also generated strains carrying NupA-GFP in combination with NLS-dsRed, histone H1-mRFP or Nup2-mCherry (Nayak et al., 2006; Yang et al., 2004) for analysis by live cell spinning disc confocal microscopy. NLS- DsRed is a marker for importin α/ mediated nuclear transport and contains the NLS sequence of StuA (Suelmann et al., 1997). During mitosis, A. nidulans NPCs partially disassemble allowing the dispersal of NLS-DsRed out of the nucleus into the cytoplasm (De Souza et al., 2004; De Souza and Osmani, 2009; Osmani et al., 2006a). NLS-DsRed is then actively transported back into daughter nuclei as cells exit mitosis into G1 and reform complete NPCs and reestablish active nuclear transport. Thus dispersal of NLS- DsRed acts as a marker for mitosis. Notably, as cells entered mitosis and NLS-DsRED dispersed from nuclei, NupA-GFP translocates from the NPCs at the nuclear periphery and concentrates on mitotic chromatin. NupA-GFP remained at mitotic chromatin until cells were exiting mitosis and accumulated NLS-DsRed back into nuclei (Figure 3.5C). To confirm NupA-GFP locates to mitotic chromatin the location of NupA-GFP and histone H1 mrfp were followed during mitosis and colocalization was apparent specifically during mitosis (Fig. 3.5B). Finally, because Nup2 shows the same pattern of moving from NPCs to chromatin during mitosis we followed both NupA-GFP and Nup2- ChRFP. Live cell imaging confirmed that NupA and Nup2 co-localize during interphase at the NPCs and at chromatin during mitosis (Fig. 3.5A). Thus NupA is the second NPC associated protein defined in A. nidulans that displays dramatic translocation to mitotic chromatin NupA is required for Nup2 location at NPCs and mitotic chromatin The data presented above indicate Nup2 and NupA likely play similar roles. Biochemically, they co-purify the same set of proteins. Cell biologically, they display dynamic subcellular localizations which are regulated in the identical manner during cell 70

91 cycle progression. Genetically, they are both essential and cause the same mitotic defects when deleted. One obvious explanation for these similarities is that one of these proteins might be responsible for the function of the other, perhaps by targeting it to NPCs at interphase and to chromatin during mitosis. To investigate whether Nup2 is responsible to locate NupA to NPCs and/or mitotic chromatin, Nup2 was deleted in cells carrying endogenously tagged NupA-GFP and NLS-DsRed. Live cell imaging was completed to see if lack of Nup2 caused any defects in the dynamic localizations of NupA. In the absence of Nup2, NupA-GFP was able to locate normally to NPCs during interphase and to chromatin during mitosis (Figure 3.6A). Conversely, if nupa was deleted then Nup2- GFP was unable to locate to NPCs during interphase nor to chromatin during mitosis (Fig. 3.6B). Together these results suggest that NupA s essential function lies in its ability to tether Nup2 correctly to interphase NPCs and to mitotic chromatin Nup2 and NupA transfer from NPCs to chromatin at early prophase and move back to the nuclear periphery in anaphase. Nuclear DNA undergoes a cycle of condensation and decondensation corresponding to entry and exit of mitosis. Mitotic nuclei can be distinguished as they contain compact chromatin which can be detected by following histone H1-mRFP. We followed Nup2-GFP and NupA-GFP individually in cells carrying histone H1-mRFP and captured images at 10 second intervals (Figure 3.7A and C) to follow their location with high temporal resolution during mitosis. Nup2 and NupA transfer from NPCs to chromatin at early prophase and stay associated with chromatin as it undergoes mitotic condensation. They both then rapidly move more to the periphery of the condensed chromatin during anaphase, as soon as chromosomes start to segregate. Notably, both NupA and Nup2 locate around condensed chromatin during both anaphase and telophase such that they appear to coat the entire genome as mitosis proceeds from anaphase into G1. These data indicate NupA and Nup2 transfer first to mitotic chromatin in prophase as it begins to undergo condensation. They then undergo a second transition immediately after anaphase, during which they move from a general chromatin localization to a position around the condensed genome (Figure 3.7 B and D). 71

92 3.3.7 Nup2 and NupA are not essential for nuclear transport. Nup2 binds to the importin α/ complex to facilitate active nuclear transport through NPCs (Gilchrist et al., 2002; Lindsay et al., 2002; Matsuura and Stewart, 2005; Solsbacher et al., 2000; Stewart, 2007). Although Nup2 is not essential in Saccharomyces cerevisiae (Loeb et al., 1993), it is essential in A. nidulans (Osmani et al., 2006a). However, Nup2 deletion in A. nidulans allows short term growth suggesting that Nup2 s essential function is not due to an essential role in nuclear transport which is required even for short term growth in A. nidulans (Osmani et al., 2006a). To study whether nuclear transport is affected in Δnup2 or ΔnupA nuclei, we used NLS-DsRed to detect whether the Δnup2 and ΔnupA mutant nuclei are capable of nuclear import of NLS bearing cargo into the nucleoplasm. NLS-DsRed was actively transported into Δnup2 and ΔnupA nuclei at interphase with no signal apparent within the cytoplasm (Figure 3.8A). In addition, we also followed Heh1-GFP localization as a marker of INM protein import. Heh1-GFP is transported from ER to the INM in a RanGTP, importin- and impotin- dependant manner, thus classic import pathway is required for its nuclear import. Moreover, Nup2 was shown in yeast to be also required for its nuclear import (King et al., 2006). However, we found that in A. nidulans Heh1-GFP is actively transported to the INM of ΔnupA nuclei at interphase with no signal apparent in the ER (figure 3.8B). We then decided to study the kinetics of nuclear import in the Nup2 and NupA null mutants using mitosis. As cells entered mitosis NLS-DsRED disperses throughout the cell in the Nup2 and NupA strains as occurs during wildtype mitosis. Also as occurs normally, NLS-DsRED got actively transported back into daughter nuclei as cells exit mitosis and entered G1 (Figure 3.8C). Interestingly, NLS-DsRed stays dispersed longer in the mutants than the wild-type suggesting that either mitosis might be taking longer in these cells or the import rate of NLS-DsRed in those nuclei is significantly slower. To study the rate of nuclear dispersal and import in the mutants, we quantified the fluorescent pixel intensity of the NLS-DsRED signal at 10 seconds intervals of nuclei going in and out of mitosis (Figure 3.8D). The dispersal and the import rates of NLS- DsRed in the deletion mutants are comparable to wild-type and the curves of the dispersal 72

93 and import are essentially superimposable. These data indicate that the rate of NPC partial disassembly is not affected by the lack of Nup2 or NupA. In addition, the data suggest that the rate of NPC reassembly during exit from mitosis, and the rate of import of NLS-DsRED, are also not affected in the absence of Nup2 or NupA. However, although importin α/ mediated transport appears to be efficient in these mutants, the time spent in mitosis was increased indicative of mitotic defects Deletion of either Nup2 or NupA causes mitotic defects that engage the spindle assembly checkpoint To further investigate the delays in mitotic progression caused by deletion of Nup2 and NupA, we used GFP tagged tubulin to monitor the dynamics of mitotic spindle formation in Nup2 and NupA deleted cells using live cell imaging (Ovechkina et al., 2003). Two main defects were apparent. The first involves the architecture of mitotic spindles in both Nup2 and NupA deleted cells. We observed numerous defects in mitotic spindle structure including spindle bundling, splitting and fusion, none of which were observed in WT controls (Figure 3.9 A and B). In addition to these structural defects, the time spent in a mitotic state was far more variable in both the Nup2 and NupA deleted cells with the deleted cells taking 10 to 11 minutes on average to complete mitosis versus 5.5 minutes for wild type cells (Figures 3.10 and 3.11). If the delays in mitosis caused by the absence of Nup2 or NupA were caused by defects that are monitored by the SAC then these mitotic delays should be dependent on a functional SAC. To investigate this possibility Nup2 and NupA were deleted in a strain lacking the SAC protein Mad2. The strains also contained GFP tubulin to monitor mitotic spindle assembly. Live cell imaging revealed that the mitotic delays caused by absence of Nup2 and NupA were entirely dependent on an intact SAC. In the absence of Mad2 virtually all of the mitotic delays caused by Nup2 or NupA deletion were removed such that the average time of mitosis was similar to WT controls, and all of the variability in the timing of mitosis was removed (Figures 3.10 and 3.11). We also followed Mad1-GFP in Nup2 and NupA null mutants. In these mutants, Mad1 translocates to kinetochores during the first mitosis and stays connected to kinetochores for about 11 minutes (Figure 73

94 3.14A), indicating that these mutants exhibit defects in the first mitotic progression which leads to SAC activation and delays in mitosis. We therefore conclude that the absence of Nup2 or NupA causes defects in mitosis which stimulate activation of the SAC and subsequent mitotic delays Nup2 and NupA null mutants form polyploid and/or aneuploid nuclei Following Tubulin-GFP and a Histone 1A-mRFP marker throughout multiple mitosis, we discovered that spindle defects in Nup2 and NupA null mutants are observed as early as the second mitosis. We also observed that spindle splitting leads to the generation of uneven number of nuclei which might be an indication of aneuploidy (Figure 3.12 B). In addition, Bundled and fused spindles may also be an indication of either polyploid nuclei (Figure 3.12 A) or nuclei which are bundled perhaps due to problems in nuclear envelope fission (Figures 3.8B, 3.15 A and B and 3.16C). We then counted the number of cells that had uneven number of nuclei (Figure 3.12B) in addition to bundled spindles (Figure 3.12 A). We thought these numbers would be suggestive of abnormal ploidy (either aneuploid or polyploid). Nup2 and NupA null mutants generated nuclei reminiscent of abnormal ploidy 28% and 34% of the time respectively. The amount of nuclei reminiscent of abnormal ploidy significantly increased as expected in the absence of the essential SAC protein Mad2 (Figure 3.12C) Nup2 and NupA null mutants display defects in mitotic exit The fact that Nup2 and NupA move from a general chromatin localization at metaphase to a position surrounding the condensed genome at anaphase (Figure 6B and D), suggests that they might have functions during mitotic exit at this location. We assessed the movement of nuclei following Histone 1A-mRFP as a marker for proper mitotic exit and transition into G1. During interphase, in wild type cells, nuclei exhibit a slow constant movement that keeps nuclei evenly distributed throughout the length of the cell. In contrast, after mitotic exit and entry into early G1, nuclei transiently move in a rapid manner until they become anchored to one position at early G1 (Figure 3.13 A). 74

95 Interestingly, we discovered that Nup2 and NupA deleted nuclei display excessive and prolonged nuclear movement upon mitotic exit and fail to anchor at stable positions in the cytoplasm (Figure 3.13 A). The abnormal movement of Nup2 and NupA deleted nuclei is dependant on microtubules (Figure 3.13 B). This suggests that the Nup2 and NupA deleted nuclei fail to properly get past early G1 phase and become delayed in the period of rapid nuclear movement at early G1. We also observed a prolonged DNA condensation in those Nup2 and NupA nuclei (Figure 3.12 A and B), further confirming a possible function of Nup2 and NupA in mitotic exit. It is necessary to mention, that these mitotic exit defects observed could be a consequence of a defective mitosis. For example, problems in completing DNA segregation might lead to lagging DNA that can form bridges between the two separated daughter DNA. It is then not difficult to imagine how such a problem might create downstream defects in mitotic exit Nup2 and NupA null mutants exhibit defects in reassembling normal post-mitotic daughter nuclei Nup2 and NupA null mutants are not capable to localize Mad1 into the nuclear periphery after the first mitosis. To study whether post-mitotic nuclear basket assembly is affected in Nup2 or NupA deleted nuclei, Mad1-GFP was followed in these null mutants throughout the first cell cycle. Mad1-GFP is transported into the nucleus and then locates at the nuclear periphery at the NPCs. Upon mitotic entry it moves to kinetochores, stays there until telophase when it disperses from kinetochores. Upon exit from mitosis, once daughter nuclei are rebuilt, Mad1-GFP is transported back into the nucleus and incorporated to the nuclear periphery (De Souza et al., 2009). In both mutants, Mad1-GFP was localized to the nuclear periphery in the first interphase as expected. Upon mitotic entry, Mad1-GFP moved to kinetochores as it co-localized with the inner kinetochore marker Ndc80-CR. It then dispersed from kinetochores at telophase, and stayed dispersed throughout the cells in the subsequent interphase. Thus, in the Nup2 and NupA null mutant nuclei Mad1-GFP 75

96 could neither locate around the nuclear periphery nor accumulate within nuclei upon exit from first mitosis (Figure 3.14A). This suggested that probably post-mitotic nuclear import of Mad1-GFP into the newly built daughter nuclei of both mutants was defective, and thus post-mitotic nuclear basket assembly of Mad1 was then hindered in those mutants. This was observed in 100% of Nup2 and NupA deleted cells followed (Figure 3.14 B and C). Interestingly though, in nupa and nup2 nuclei, Mad1-GFP stayed at the kinetochores at least double amount of time as the wild type (Figure 3.14A) which is consistent with the fact that SAC is activated in those cells due to mitotic defects caused by Nup2 and NupA deletion NupA null mutants show defects in post-mitotic membrane fission To further characterize the nature of the mitotic exit defects seen, we observed predicted INM marker proteins Heh1-GFP and AN0162-GFP (Bqt4 in S. pombe (Chikashige et al., 2009) in ΔnupA nuclei. Both INM proteins showed the presence of linked nuclei in the NupA null mutant suggesting that these nuclei cannot properly separate after mitosis (Figure 3.8B. b, c, d and 3.16C). We then followed AN0162-GFP, as a nuclear envelope marker, throughout the cell cycle and discovered that 20% of NupA null mutant nuclei fail to exit the first mitosis and separate into 2 daughter nuclei. Instead, the daughter nuclei collapse back together upon mitotic exit and form a polyploid nucleus (Figure 3.15A). This nucleus then divides into 4 upon entry into the second mitosis (Figure 3.15B). This suggests that NupA deleted nuclei show defects in post-mitotic nuclear membrane fission and explains the presence of polyploid nuclei in Nup2 and A null mutants. We were then interested in understanding whether the rest of ΔnupA nuclei that seem to exit mitosis, show any defects in membrane fission. We followed these nuclei throughout the first and second mitosis and discovered that they exhibit defects in double pinching of the nuclear envelope during exit from mitosis leading to 2 daughter nuclei which may be linked (Figure 3.16 A and B). 76

97 NupA null nuclei exhibit additional interphase defects in Ima1-localization. In order to further characterize the early G1 defects observed earlier in ΔnupA mutants, we followed the INM protein Ima-1-GFP throughout the cell cycle in those mutants. Interestingly, we discovered defects in Ima1 localization in NupA null nuclei as early as the first interphase (Figure 3.17A and C). Ima1-GFP localized into a single focus at the nuclear periphery in NupA deleted cells instead of a smooth INM localization. Because Ima-1 is involved in coupling of centromeric heterochromatin to the MTOC and bridging it to the cytoskeleton in S. pombe, the mislocalization of Ima1 in NupA null mutant might explain the earlier defects observed in nuclear organization and positioning during mitotic exit. Very interestingly, some of the Ima1-GFP signal redistributes around the nuclear periphery upon mitotic entry (Figure 3.17 A and B). This suggests that the Ima1-GFP focus observed is probably inside the nucleus at the INM and that NupA might be required for proper and even distribution of Ima1 at the INM. 3.4 Discussion and future direction This study describes the isolation of a novel and essential nucleoporin in A. nidulans, we call NupA. It also defines the first Nups, Nup2 and NupA, in A. nidulans that have essential roles in mitosis which is regulated by NIMA. Our work thus provides insights on the mechanism how NIMA regulate mitotic progression in A. nidulans and extends the understanding in the field of the mitotic roles of NPC associated proteins, strengthening the evolving connection between nuclear transport proteins and mitosis. Our discoveries weigh on the importance of intertwining NPCs proteins with the coordination of mitosis where massive reorganization of nuclear structure is achieved and is tightly regulated during space and time Identification of NupA, a previously uncharacterized Nup We have isolated a completely novel nucleoporin, NupA in Aspergillus nidulans as a copurifying partner of Nup2. NupA is the second Nup in A. nidulans shown to undergoe a unique and exclusive translocation to mitotic chromatin. It is conserved only in a subset of hyphal Ascomycota (subphylum Euascomycetes); class Eurotiomycota 77

98 which encompasses all Aspergilli, in addition other genera such as Paracoccidioides, Blastomyces, Coccidioides, Microsporum, Trichophyton, Histoplasma, and Uncinocarpus. There is no obvious ortholog of NupA in higher eukaryotes or in budding and fission yeast. pblast with the coding sequence of NupA shows a very low sequence homology to Nup1p (score:96; e 0.016) and Nup60p (score 82; e: 0.16) in Saccharomyces cerevisiae and Nup124p ( score: 130; e 2.9x ) and Nup60p (score:95; e 0.01) in Schizosaccharomyces pombe. Sc-Nup1, Sc-Nup60 and Sp-Nup124 are thought to be the orthologues of mammalian Nup153 in lower eukaryotes (Bapteste et al., 2005; Cronshaw et al., 2002; Hetzer et al., 2005; Mans et al., 2004; Sistla et al., 2007). Interestingly, many of these Nups have been shown to carry functionally similar roles to NupA. Similar to NupA, Sc-Nup60 is shown to bind and localize Sc-Nup2 to the NPC (Allen et al., 2001; Allen et al., 2002; Collins et al., 2007; Denning et al., 2001; Dilworth et al., 2001; Dilworth et al., 2005; Krogan et al., 2006). In addition, Nup153 exhibits two-hybrid interactions with Nup50, the mammalian orthologue of Nup2 (Smitherman et al., 2000). Nup153 also co-immunoprecipitates with Nup50 from rat liver nuclei and transfected cells (Guan et al., 2000; Smitherman et al., 2000). Moreover, posttranscriptional Nup153 gene silencing leads to mislocalization of Nup50 to the nuclear interior in HeLa cells (Hase and Cordes, 2003). Recently, Nup153 has also been shown to play roles in early mitosis and cytokinesis in HeLa cells (Mackay et al., 2009). These data suggest that NupA, although it has very low primary sequence similarity, might be the orthologue of mammalian Nup153, Sp-Nup124 and Sc-Nup60 in Aspergillus nidulans. On the other hand, NupA lacks key functional domains present in Nup153 such as the Zinc-finger domains that bind RanGTP which is required for Nup153 s function in transport (Ball and Ullman, 2005; Higa et al., 2007). The Zinc finger domain of Nup153 also functions in nuclear envelope break down of metazoans by binding to and recruiting the COPI complex to the nuclear envelope (Liu et al., 2003; Prunuske et al., 2006). The fact that NupA does not have an essential function in transport in A nidulans, in addition to the fact that the NE does not break down in this fungus, makes it easy to imagine how the Zinc finger domains might be evolutionary lost in the A. nidulans NupA. In addition, the heavily rich FG repeats present in Nup153, Sp- Nup124 and Sc-Nup1 are also missing in NupA. These observations suggest that NupA is 78

99 either a Eurotiomychota specific protein or its primary sequence conservation has been lost outside this group. The latter being very probable because many nucleoporins have very low primary sequence conservation. NupA can be then either the orthologue of Nup153 or of a yet unidentified vertebrate Nup. Further work, perhaps studying whether Nup153 can complement NupA s essential function in A. nidulans will shed light on its functional identity Presence of an unusually large intron in nupa We have found that nupa has an unusually large intron at its 5 end in A. nidulans which is conserved in many Aspergilli. This finding is further surprising because, in contrary to A. nidulans, this intron is open in A. fumigatus (our data), Aspergillus niger, Aspergillus aculeatus and Aspergillus carbonarius (personal communication with Scott Baker). What is more intriguing is the fact that tblastn with the A. fumigatus coding sequence of the intron using the Aspergillus comparative database shows 90% identity with a transcript in N. fischeri (score:811 bits, Evalue 0.0), 47% identity with a transcript in A. clavatus (score: 318 bits, E value 2e-86) and 26% identity with a transcript in A. terreus (score: 65.1 bits, E value 3e-10). This intron then has the potential to be coding, and NupA might be an alternatively spliced gene. Although we couldn t detect different transcripts reflected in cdnas of nupa in A. nidulans nor A. fumigatus, it is possible that NupA can be alternatively spliced depending on the environmental conditions the fungus faces or during asexual or sexual developmental programs. In addition, A. nidulans NupA protein shows 46% identity at the amino acid level with its A. fumigatus orthologue and 49.5% identity with its A. niger one. The percent amino acid identity between these genes is low compared to the degree of overall similarity between genes within the Aspergilli. For example, predicted gene orthologues shared by all three A. nidulans, A. fumigatus and A. oryzae species (three-way orthologues) display an average of 68% amino acid identity (Galagan et al., 2005). These observations then suggest that NupA is a gene that may have undergone a rapid divergent adaptive evolution. 79

100 3.4.3 Why would NupA, a gene essential for mitosis, diverge this rapidly during evolution? We have shown that NupA is an essential gene in A. nidulans and its deletion causes mitotic DNA segregation defects. Thus, it is very intriguing that a gene like NupA, required for such a basic and conserved process, is not evolutionary conserved at the primary protein sequence level. In fact, as mentioned earlier, its sequence is only highly conserved in Eurotiomycetes. Very interestingly, this class of fungi, which include all Aspergilla, are thought to undergo para-synchronous mitosis, unlike their relatives such as Neurospora crassa and Magnaporthe grisea which are hyphal ascomyctes that fall under the class Sordariomycota (Gladfelter, 2006). One then might think that NupA s function might be essential only in organisms that undergo synchronous or parasynchronous mitosis. However, this is inconsistent with the fact that it s binding partner, Nup2, whose function is dependant on NupA in A. nidulans, is a highly conserved protein at the primary sequence level and has a conserved localization to mitotic chromatin and presumably a conserved mitotic function (Dultz et al., 2008; Osmani et al., 2006a). NupA then might have a functional homolog in higher eukaryotes with low sequence homology to its A. nidulans counterpart. Alternatively, metazoans might have evolved a distinct mechanism to target Nup2 to the pore and to chromatin which is independent of NupA. Future work, in defining Nup2 and Nup50 domains required for NPC and or chromatin locations will help us understand the mechanisms by which these proteins target different sites during the eukaryotic cell cycle. Furthermore, the rapid evolutionary divergence of the NupA locus can be explained by the fact that it might encode a gene that confers Heterokaryon Incompatibility (HI) in A. nidulans. HI is a ubiquitous phenomenon in filamentous fungi that allows non-self recognition between genetically different isolates of the same fungal species and often results in compartmentalization and death of hyphae that undergoes fusion. Genes that regulate HI can be interacting proteins that are evolved as byproducts of divergent coevolution through a rapid adaptive evolution of those genes (Glass and Dementhon, 2006; Hall et al., 2010). Strong evidence shows that divergent coevolution among interacting components of the Drosophila nuclear pore complex results in multiple hybrid incompatibilities and promotes speciation (Presgraves and Stephan, 2007; Tang 80

101 and Presgraves, 2009). Speciation occurs upon the evolution of incompatible gene interactions that cause sterility or lethality in hybrids between populations. In fact, two D. simulans nucleoprins, Nup96 and Nup160 have been shown to be lethal hvbrid incompatibility genes that are incompatible with one or more factors on the D. melanogaster X chromosome. Nup153 has been suggested as a strong candidate for the X chromosome hybrid incompatibility factor and data shows that Nup153 have been subjected to recurrent adaptive evolution which might drive rapid functional divergence of this protein in different Drosophila lineages (Presgraves and Stephan, 2007; Tang and Presgraves, 2009). Thus, nupa is a strong candidate gene that could be involved in HI in A. nidulans. Further studies are needed to test whether nupa is actually involved in this process. Future work would focus on collecting different genetic isolates of A. nidulans from different places in the environment, the identification of genomic and cdna sequences of nupa for these isolates and the comparison of their percent identity with the A. nidulans strain used in the lab. These experiments would also detect whether alternatively spliced form of the NupA protein exists in those isolates. Moreover, studies can be performed to test whether NupA from the different isolates collected can complement its essential role in the laboratory strain NIMA regulates the movement of Nup2 to chromatin during mitosis We have previously shown that all of Nup2 translocates to mitotic chromatin. In this study we show that Nup2 is highly phosphorylated during mitosis and is under the regulation of the mitotic kinase NIMA. In fact, NIMA not only promotes Nup2 phophorylation, but drives Nup2 onto chromatin. This discovery makes Nup2 one of the very few mitotic targets of NIMA discovered till date and gives additional insights regarding the mechanism by which NIMA promotes the G2/M transition. Very importantly, these findings also present the first insight regarding the mechanism by which it might promote spindle assembly. 81

102 3.4.5 Nup2 and NupA have roles in metaphase spindle assembly and/or stability We have shown that absence of Nup2 and NupA do not cause major defects in importin / mediated nuclear transport but do cause defects that activate the spindle assembly checkpoint. Nup2 and NupA deleted cells form polyploid nuclei, and display abnormal mitotic spindles. This suggests that Nup2 and NupA play crucial roles during metaphase, perhaps directly for proper spindle assembly or potentially kinetochore function, absence of which activates the spindle assembly checkpoint. The SAC is fulfilled and inactivated in Nup2 and NupA deleted cells allowing progression of mitosis in the absence of these proteins. This can be explained with the possibility of either the defects being corrected with time thus allowing the inactivation of SAC or that SAC gets inactivated in those cells without proper spindle assembly and attachment. The former possibility cannot explain why Nup2 and NupA are essential. Moreover, data shows that even when SAC is overridden Nup2 and NupA deleted cells undergo defective mitosis leading to missegragated DNA and abnormal nuclei. This suggests the SAC is indeed inactivated in these cells prior to the correction of all the defects induced by the absence of Nup2 and NupA. This observed phenomenon constitutes an inactivation of SAC, and not mitotic slippage (Brito and Rieder, 2006; Rieder, 2010), because the SAC protein Mad1 leaves the kinetochores upon mitotic exit. In fact, data from our lab shows that the SAC is inactivated in cells arrested in mitosis when microtubules are depolymerized by benomyl after an average of 50 mins mitotic SAC arrest (unpublished, Colin De Souza). The level of benomyl used in those experiments induces full depolymerization of the mitotic spindle. Nup2 and NupA deleted cells inactivate SAC after only 10 to 11 minutes. This might be explained with the fact that in those cells less extreme defects in spindle assembly is predicted compared to benomyl treatment. It will be interesting to see whether inducing less severe defects in spindle assembly by using titrated levels of benomyl, will promote SAC inactivation at time points earlier than 50 minutes. The prediction is that as the amount of benomyl added to cells is increased, which leads to gradual levels of spindle microtubule depolymerization, a gradual increase in the time that cells stay arrested in mitosis with an active SAC will be apparent. This then might explain why in Nup2 and NupA deleted cells, where less severe spindle defects are 82

103 observed, SAC is inactivated after only about 10 to 11 minutes compared to being inactivated after a 50 mins mitotic arrest when mitotic spindles are fully depolymerized What could the essential functions of Nup2 and NupA be in mitosis? NupA is required to locate Nup2 both to the NPC at interphase and to chromatin during mitosis. Because Nup2 and NupA play similar essential roles, and cause the same mitotic defects when deleted, this suggests that NupA s essential function is to target Nup2 both to the NPC during interphase and to chromatin during mitosis, where Nup2 might have a role in making mitosis more efficient by concentrating mitotic effectors around chromatin. Because Nup2 co-purifies with importins α and in mitosis, Nup2 might achieve this task perhaps by accelerating the release of mitotic cargo from the importin / complex only at the vicinity of the chromatin during the A. nidulans semiopen mitosis. On the other hand, NupA s essential function might not lie solely on the fact of it being responsible to locate Nup2 to both locations. Perhaps, NupA functions not only to locate Nup2 at mitotic chromatin but also helps Nup2 in cargo release from the classical importin / complex. One way to further assess whether NupA s essential function is to target Nup2 to both locations is by artificially tethering Nup2 to either the chromatin, or the NPC, in the absence of NupA, and asking whether any one of these proteins will rescue the NupA deletion phenotype (see Chapter 4) Nup2 and NupA have roles in mitotic exit and rebuilding of post-mitotic daughter nuclei Evidence suggests that Nup2 and NupA are not only involved in proper spindle assembly but also carry functions during mitotic exit and early G1 involving rebuilding of postmitotic daughter nuclei. We observe a prolonged chromatin condensation in Nup2 and NupA nulls upon mitotic exit. In addition, we also observe a prolonged and excessive movement of Nup2 and NupA deleted nuclei upon mitotic exit and a defect in Ima-1 localization in those nuclei. Also, Mad1 and Nup2 were found to be unable to localize to the NPC after the first mitosis in NupA deleted nuclei. This suggests that the nuclear basket assembly might be altered in those cells. However, further work, following other 83

104 nuclear basket proteins such as Mlp1 would further clarify if this is the case. Nuclear basket proteins and particularly Nup2 has been shown to be responsible to tether chromatin to the nuclear periphery altering the 3D structure of the genome, thus regulating gene transcription (Ishii et al., 2002). Regulation of the genome organization has also been implicated to have a more global effect on nuclear organization and function (Misteli, 2001). In fact, proteins such as Ima1 and the Sun-Kash complex bridge the nuclear DNA to the cytoplasmic MTOCs and the actin cytoskeleton. Because the nuclear basket assembly is altered in NupA deleted nuclei, these nuclei might have problems in organizing their chromosomes in a proper 3 D structure upon mitotic exit which might explain why prolonged chromatin condensation is observed in Nup2 and NupA nulls. Furthermore, Ima-1 is mislocalised in NupA deleted nuclei which suggests that these nuclei have problems forming bridges between the nucleoplasm and the cytoplasm. Moreover, dramatic failure in membrane fission is observed in NupA deleted cells. This suggests that NupA might carry active roles in membrane fission. On the other hand, because A. nidulans carries multiple nuclei in one cytoplasm, a checkpoint mechanism might exist at the level of nuclear fission reminiscent of the abscission checkpoint in human cells during cytokinesis (Steigemann et al., 2009). This checkpoint might monitor whether DNA is properly segregated, decondensed, and that nuclei are properly built. Thus, NupA deleted nuclei might activate such a checkpoint explaining why they stay linked. However, it is very intriguing how these nuclei can then accomplish active nuclear transport. Nevertheless, it is easy to imagine that these nuclei are probably assembled into a closed entity but their genome is improperly organized, the connection between the nucleus and the cytoplasm compromised and the membrane linking the two daughter nuclei unresolved. In conclusion, Nup2 and NupA seem to play a crucial role in post mitotic nuclear assembly and organization and bridging the nucleus to the cytoplasm. 84

105 3.4.8 Nup2 and NupA orchestrate reversible NPC disassembly with DNA segregation and formation of daughter nuclei Nup2 and NupA display dynamic cell cycle specific locations and appear to carry multiple roles during the A. nidulans cell cycle. They are at the NPC during interphase where presumably they have roles in nuclear transport and perhaps gene expression. Nup2 of yeast and mammalian cells have been shown to have roles in facilitating cargo release in the classic import pathway and in gene regulation. In A. nidulans, these two proteins move from the nuclear pores to chromatin during mitosis. At metaphase, they are tightly associated with chromatin where they play a crucial role in either proper spindle assembly or promote microtubule-kinetochore attachments. At anaphase, they move from being on, to around, the condensed chromosomes, where they have functions in proper mitotic exit and building of daughter nuclei. It is clear today, that the one gene-one enzyme hypothesis proposed earlier this century is an extreme oversimplification of events that happen in a cell. Evidence shows that each protein can carry multiple roles. However, it is inevitable to ask, why would nucleoporins such as NupA and Nup2 carry so many functions? The answer is simple. Nuclear division involves the regulation of extremely complex and intricate series of events that happen in a stepwise manner. These events induce dramatic changes in the cell including NPC disassembly, chromosome dynamics and spindle formation. The proper coordination of these events is crucial for a successful mitosis. Nucleoporins such as Nup2 and NupA are great candidates to accomplish such coordination. Nup2 for example is an intrinsically mobile natively unfolded protein that exhibits little secondary structure, and has high flexibility and low compactness (Denning et al., 2002). This makes it capable of simultaneous and interchangeable interactions with multiple protein partners in a rapid manner. Thus, upon G2/M transition, NIMA and probably CDK1 promote Nup2 and NupA phosphorylation and translocation to chromatin thus coordinating NPC disassembly with chromosome condensation and spindle formation. Evidence also shows that these proteins might also coordinate DNA decondensation, genome reorganization and nuclear reassembly. Moreover, because Nup2 and NupA co-purify with importins α and in mitosis and G2, we propose that both roles of these proteins in mitosis and early G1 might be linked to proper localizations and functions of importins α and during metaphase and mitotic 85

106 exit. In fact, these importins have been shown to play roles in both mitosis and mitotic exit during NPC and daughter nuclei assembly. The importin / complex was shown to bind to NLSs located on many cargo such as spindle assembly factors (SAFs) and nuclear pore proteins thereby inhibiting their activities until they encounter high levels of Ran- GTP which then helps the release of those cargo from the importin / complex and thus promotes spindle and post-mitotic NPC assembly (Wozniak et al., 2010). Future work, for example following these importins throughout mitosis in Nup2 or NupA deleted cells, will clarify whether Nup2 and NupA function in several steps of nuclear division through those importins. 86

107 Figure 3.1 Untagged Nup2 relocates from NPCs to DNA during mitosis. (A) Western blotting of A. nidulans total cell extract with antibody raised against Nup2, pretreated with or without the Nup2 antigen. Equal loading was tested by staining for tubulin (data not shown). The newly generated antibody specifically recognizes Nup2 in both G2 and mitotic samples. (B) Cells fixed and stained with the Nup2 antibody and DAPI to visualize DNA. Nup2 is at the NPCs during interphase but colocalizes with DNA during mitosis. Scale bar, ~ 5μm. 87

108 Figure 3.2 Identification of a novel Nup via affinity purification of A. nidulans Nup2. (A) Coomassie stained SDS-PAGE gel showing affinity purified Nup2 with importins KapA, KapB and a novel nucleoporin that copurifies with Nup2 identified by Mass Spectroscopy analysis. The first two samples depict affinity purifications performed for G2 and M (mitosis) samples. In the third raw, the mitotic sample was treated with phosphatase prior to electophoresis. Also shown in a table are the % peptide coverage of different samples in G2 and M. (B) Cartoon depicting the genomic structure of A. nidulans, A. fumigatus, and A. niger nupa. All genes contain a large intron between exon1 and 2. (C) Cartoon depicting the protein structure of A. nidulans, A. fumigatus, and A. niger NupA. There is a 46% sequence identity between both NupA proteins. We identified many putative classical NLSs in both proteins (marked with black bars). No other known motifs wer identified in NupA. The experiment in 3.2 A was performed by Aysha H. Osmani. 88

109 Figure 3.3 Induction of NIMA promotes phosphorylation and relocation of Nup2 from NPCs onto chromatin in S phase arrested cells. (A) Western Blot showing an upshift in the Nup2 band upon induction of stable NIMA (NIMA-ΔC) depicted by the increase in its levels. The upshift is observed even in S-phase arrested cells (+HU). * depicts endogenous NIMA which is auto-phosphorylated as we observe an upshift in its band. (B) Graph depicting the time when Nup2 moves to DNA in a population of cells, upon induction of stable NIMA in presence or absence of HU. A copy of NIMA was introduced under the alca (alcohol dehydrogenase promotor) which was induced with ethanol (EtOH) (see materials and methods for details). (C) Maximum intensity projections of Nup2-GFP and NLS-DsRed signals from live cell imaging of one nucleus with or without induction of NIMA at the time points where cumulative percentage of cells with Nup2 on DNA is 50% respectively in the presence or absence of HU. Bar, ~ 5μm. The experiments listed in this figure were performed by Aysha H. Osmani. 89

110 Figure

111 Figure 3.4 NupA is essential and its deletion, similar to Nup2, does not affect germination but causes mitotic DNA segregation defects. (A-a) Conidia from transformants carrying the nupa::pyrg cassette were streaked on media with or without UU. Heterokaryons were defined as producing some spores which grow on +UU but not on -UU. (A-b) Diagnostic PCR of a heterokaryons showing 2 bands corresponding to the wt NupA gene and the nupa::pyrg allele. (B) Bright field images of spores from the wild-type, and nupa grown for 1 or 3 days at 22 C on the indicated media. Bar, ~ 20μm. (C) DAPI staining of a wild type germlings with 1, 2, 4 and 8 nuclei compared to nupa germlings of approximately the same length. nupa germlings show misssegregated DNA (arrows) or single large nuclei (arrowheads). Bar, ~ 5μm. 91

112 Figure

113 Figure 3.5 NupA colocalizes with Nup2 at the Nuclear Pore Complex during interphase and on chromatin during mitosis. (A) Maximum intensity projections of NupA-GFP and Nup2-ChRFP signals from live cell imaging of one nucleus in G2, mitosis (M), and G1 as indicated. The third panel shows a merge image and the fourth panel shows pixel profiles for both signals through the dividing nucleus. Bar, ~ 5μm. (B) Maximum intensity projections of NupA-GFP and Histone1-ChRFP signals from live cell imaging of one nucleus in G2, mitosis (M), and G1 as indicated. The third panel shows a merge image and the fourth panel shows pixel profiles for both signals through the dividing nucleus. Bar, ~ 5μm. (C) Live cell imaging of a nucleus in mitosis followed by NupA-GFP and NLS-DsRed using time lapse confocal microscopy. Also shown are pixel profiles of every time point as well as kymographs. NupA-GFP locates to chromatin during mitosis (M) marked by the dispersal of NLS-DsRed from the nucleus. During exit from mitosis NupA relocates back to NPCs and active nuclear transport is reestablished. Bar, ~ 5μm. 93

114 Figure

115 Figure 3.6 NupA is required for Nup2 location to NPCs and to mitotic chromatin. (A-B) Still images of nupa and nup2 germlings (outlined in white) consecutively in G2, M (mitosis) and G1 as indicated. The Nup2-GFP, NupA-GFP and NLS-DsRed signals are shown as maximum intensity projections from time lapse imaging. Also outlined are wild type spores within the same field of view. (A) In the absence of Nup2, NupA can locate to the pore and to mitotic chromatin. (B) Conversely, in the absence of NupA, Nup2 can neither locate to the pore nor to chromatin. Bar, ~ 5μm 95

116 Figure 3.7 Nup2 and NupA associate with chromatin at the start of prophase and move back to the periphery of chromatin in early anaphase. (A and C) Series of still images captured at 10 sec intervals during live cell imaging using a spanning disc confocal microscope following Nup2-GFP or NupA-GFP as indicated in combination with Histone1-mRFP during mitosis. Nup2 and NupA move to chromatin at the onset of prophase and chromosome condensation. (B and D) Z-Stack confocal images of Nup2- GFP or NupA-GFP in combination with Histone1-mRFP in metaphase and anaphase/telophase as indicated. Nup2 and NupA move to around chromatin in anaphase. Bar, ~ 5μm. 96

117 Figure 3.7 Continued 97

118 Figure 3.7 continued 98

119 Figure 3.8 Nup2 and NupA are dispensable for nucleocytoplasmic transport. (A-B) NLS-dsRed (A) and membrane bound Heh1-GFP (B) are actively imported to the nucleoplasm and the inner nuclear membrane respectively in nup2 and nupa nuclei. Nup49 also locates around nuclear periphery in nup2 and nupa nuclei. Shown are Bright Field and confocal images of either nupa or nup2 and Wt spores isolated from heterokaryons and grown on selective media. The Nup49-GFP, Heh1-GFP and NLS- DsRed signals are shown as maximum intensity projections. (B-C) NLS-DsRed disperses and is actively transported back into the mutant nuclei in a manner and rate similar to the wild-type. (C) Plotted are series of still images from typical movies captured following NLS-DsRed in wild-type, nup2, and nupa nuclei through mitosis. (D) Graphs depicting the rate of dispersal and re-import of NLS-dsRed in wild type, nup2, and nupa nuclei entering and exiting mitosis. Graphs depict average normalized pixel intensity vs time (N=10). The pixel intensity of the red signal was quantified in a defined area within the nucleus every 10 seconds and the highest pixel intensity normalized to 100%. Bar, ~ 5μm. 99

120 Figure 3.8 Continued 100

121 Figure 3.8 continued 101

122 Figure 3.9 Nup2 and NupA deleted nuclei exhibit defects in mitotic spindle structure. (A) Graph depicts percentage of Nup2 null and NupA null mutants with defective spindles. About 40% of spindles were defective for each null mutant. (B) Still images from typical movies captured following tubulin-gfp in wild-type, nup2, and nupa nuclei through mitosis. Defective spindles were categorized into 2 distinct classes including spindle bundling and splitting, none of which were observed in WT controls. Arrows represent bundled spindles, arrowheads represent spindle splitting and the red arrow represents a spindle that elongates and resists disassembling. 102

123 Figure

124 Figure 3.10 Nup2 and NupA deleted cells undergo slow mitosis due to activation of the spindle assembly checkpoint. (A) Bar graph representing average time spent in mitosis for each indicated genotype. The number on each bar represents the actual average. nup2 and nupa cells spend significantly longer time in mitosis compared to WT (with a p value 0.01, highly significant when comparing the means). Δmad2 nup2 and Δmad2 nupa mutants show no statistical difference when compared to WT or Δmad2 mutant (statistics conducted using PASW statistical software, SPSS Inc., Chicago, Illinois, see materials and methods for details). (B) Maximum intensity projections of tubulin-gfp signal from live cell imaging of germlings with 4 nuclei during G2-mitosis- G1. nup2 and nupa cells display a prolonged mitosis only when the SAC is functional. Bar, ~ 5μm. 104

125 Figure

126 Figure 3.11 Nup2 and NupA deleted cells undergo slow mitosis due to activation of the spindle assembly checkpoint. Histograms showing the distribution in the length of mitosis for each indicated genotype. The graphs also show the mean, number of samples and the standard deviation for each. 106

127 Figure

128 Figure 3.12 Nup2 and NupA null mutants form polyploid and/or aneuploid nuclei. (A-B) Still images form live cell imaging of a NupA null mutant germlings in mitosis followed by tubulin-gfp and H1-mRFP using time lapse confocal microscopy. (A) Depicts a NupA null mutant that has a polyploid nucleus which forms bundled spindles and divides from one into at least 6 nuclei. (B) Depicts a NupA null mutant that carries a nucleus which forms a spindle which splits and divides into three nuclei. The daughter nuclei could either be aneuploid or polyploid depending on the state of the mother nucleus. Bar, ~ 5μm. (C) Graph depicting percent cells with abnormal ploidy (exhibiting either defect 9A or B) in each genotype stated. Graph on the left shows that 28% and 34% of Nup2 and NupA null mutants carry either polyploid or aneuploid nuclei. Graph on the right depict that the percentage of nuclei with abnormal ploidy in a Nup2 null increases significantly when the essential SAC protein mad2 is deleted. 108

129 Figure

130 Figure 3.13 Nup2 and NupA null mutants exhibit excessive and prolonged nuclear movement upon mitotic exit which is dependant on microtubules. (A) Graphs depicting distance from a set point at the head of the germling to the center of nuclei throughout mitosis in wt, nup2, and nupa nuclei. At mitotic exit, the 2 daughter wt nuclei move for about 10 mins then stop moving once positioned in the germtube. In G2 and then consecutive G1, the wt nuclei exhibit a constant flow from the head to the tip of the germling. nup2 and nupa nuclei do not exhibit any constant flow from the head to the tip. More dramatically, they exhibit excessive and prolonged nuclear movement upon mitotic exit and lack of nuclear positioning. (B) Graphs depicting movement of nup2 and nupa nuclei vs. time in 3 different cells in presence or absence of benomyl. Upon addition of the microtubule depolemerizing drug benomyl, the nup2 and nupa nuclei stop moving. 110

131 Figure

132 Figure 3.14 Nup2 and NupA null mutants are defective in postmitotic Mad1 localization to the nuclear periphery. (A) Series of still images captured at 1min intervals during live cell imaging of wt, Nup2 and NupA null nuclei in first mitosis using a spanning disc confocal microscope following Mad1-GFP in combination with Ndc80- CR. In the first mitosis of nupa and nup2 nuclei, Mad1-GFP stays at the kinetochores at least double amount of time as the wild type then disperses throughout the spore, stays dispersed and does not to the nuclear periphery. In some cases, such as depicted in the Nup2 null, Mad1-GFP starts to localize faintly around the nuclear periphery after 3 to 4 hours. Bar ~ 2.5 μm. (B) Shown are nupa and nup2 germlings carrying Mad1-GFP and Ndc80-CR. Mad1 is dipersed throughout the germling in both Nup2 and NupA null mutants, although some Nup2 null germlings, show faint Mad1 localization around the nuclear periphery which is cell cycle dependant. Bar, ~ 5μm. 112

133 Figure 3.14 Continued 113

134 Figure 3.14 continued 114

135 Figure 3.15 NupA null mutants show defects in post-mitotic membrane fission (A) Series of still images captured at 1min intervals during live cell imaging of a wt nucleus and NupA null nucleus throughout the first mitosis using a spinning disc confocal microscope following AN0162-GFP in combination with NLS-DsRed. For a wild type nucleus refer to Figure 16A. Two types of observations were seen. In type 1, a nucleus fails to divide and two daughter nuclei fuse back together to form one polyploid nucleus. This defect is exhibited 20% of the time. In type 2, a nucleus generates 2 nuclei which seem to be linked. (B) In the second mitosis, the polyploid nuclei generate 4 nuclei from a single nucleus. Bar, ~ 5μm. 115

136 Figure

137 Figure 3.16 NupA null mutants exhibit defects in double pinching of the nuclear envelope during mitotic exit. (A) Live cell imaging of wt and NupA null germling in 1st mitosis followed by AN0162-GFP and NLS-DsRed using time lapse confocal microscopy. AN0162-GFP does not seem to locate to the periphery of the nucleolar remnant as obviously as the wild type. Compare structure at arrows. (B) Live cell imaging of wt and NupA null germling in 2 nd mitosis followed by AN0162-GFP and NLS-DsRed using time lapse confocal microscopy. AN0162-GFP does not seem to locate to the periphery of the nucleolar remnant as obviously as the wild type. (C) Terminal phenotype showing linked nuclei following AN0162-GFP. Bar, ~ 5μm. 117

138 Figure 3.16 Continued 118

139 Figure 3.16 continued 119

140 Figure 3.17 NupA null nuclei exhibit an interphase defect in Ima1-localization. Series of still images captured from a live cell imaging of a NupA null nucleus during first (A) and second (B) mitosis following Ima1-GFP. Ima1-GFP localizes into a prominent single focus in a nupa nuclei especially in G2 compared to wt. During mitosis, this focus is partially dispersed and the signal partly smoothened around the nuclear periphery. (C) Maximum intensity projections of NupA null germlings following Ima1-GFP and NLS-DsRed. Ima1-GFP is mislocalised in these germlings into either a prominent single focus or multiple foci compared to the smooth distribution of Ima1 observed in wt (B). Bar, ~ 5μm. 120

141 Figure

142 CHAPTER 4 ISOLATION OF SEPARATION OF FUNCTION ALLELES, AND CHARACTERIZATION OF THE ESSENTIAL FUNCTION OF, NUP2 AT THE NUCLEAR PORE COMPLEX AND ON MITOTIC CHROMATIN 4.1 Introduction Mitosis encompasses highly complex events that need proteins which carry specific functions at different locations that coordinate these intertwined events. Increasing evidence is showing that Nups, particularly during open mitosis, carry mitotic functions once away from the pores which are independent of their interphase roles in nuclear transport. Once released from the pores, these Nups have distinct localizations to mitotic structures such as the spindle poles and kinetochores, where they either assist in the formation of bipolar spindles or help ensure that this process is completed before chromosome segregation. Later, during mitotic exit, Nups form the seeds to rebuild NPCs and work to coordinate this process with reformation of the nuclear envelope, thereby ensuring the formation of functional daughter nuclei (De Souza and Osmani, 2009; Guttinger et al., 2009). One underlying problem encountered when trying to decipher the mitotic functions of these proteins is the difficulty to 122

143 differentiate between their interphase and mitotic roles. A key approach to distinguish between the mitotic and interphase functions is the generation of separation-of-function alleles. Thus, it is highly advantageous that several model genetic organisms exist, such as A. nidulans, that undergo intermediate modes of mitosis and in which separation-offunction alleles can be generated relatively easily. In A. nidulans, the NPCs partially disassemble during mitosis. Many of the Nups that disassemble have shown to have specific mitotic locations, such as spindle pole bodies and the spindle matrix (De Souza et al., 2009; Liu et al., 2009). Only one Nup though, Nup2, shows an exclusive translocation to mitotic chromatin (Osmani et al., 2006a). Very interestingly Nup2 is essential in A. nidulans and its deletion causes defects in mitotic progression but not in importin / mediated nuclear transport (Osmani et al., 2006 and Chapter 3). Nup2 therefore is a strong candidate protein in A. nidulans that carries a mitotic function at its chromatin location. We therefore hypothesize that the essential function of A. nidulans Nup2 lies in its ability to associate to mitotic chromosomes. Here, we undertake the task to differentiate between its interphase and mitotic functions by generating separation-offunction alleles. We first perform a domain study to define the domain within Nup2 responsible for its NPC vs chromatin locations, only to discover that a single domain (Domain 4) within Nup2 is necessary and sufficient for its translocation to both NPCs and mitotic chromatin. As expected, Domain 4 is found to be essential for Nup2 s function. Efforts are also described to generate two types of Nup2 chimeric proteins that either exclusively tethers Nup2 to the pore or to chromatin throughout the cell cycle. 4.2 Material and methods Domain study The integrating psm5 plasmid was generated by subcloning the ¾ pyroa gene into the integrating plasmid pal5-gfp (puc19 backbone) by cutting out Pyr4 and replacing it with ¾ pyro (Figure 4.1A). To cut out pyr4, pal5-gfp was digested with SacII and EcorI. The ¾ pyroa gene was amplified from the pjw52 plasmid (gift from 123

144 Nancy Keller, Ph.D., Professor of Bacteriology and Medical Microbiology and Immunology, The University of Wisconsin, Madison, WI), using Pfu-turbo DNA polymerase (Stratagene, CA) for high-fidelity PCR amplification. The ¾ pyroa was amplified using primers SM1 (carrying a BamHI site) and SM2 (carrying an EcoRI site), and digested by BamHI and EcoRI. ¾ pyro was then ligated into pal5-gfp at the Pyr4 site, between the SacII and EcoRI sites, using a linker (made using primers SM5 and SM6, its one end attached to a digested BamHI site on ¾ pyro A, and its other end attached to a digested SacII site on the pal5-gfp plasmid, linking the insert to the plasmid). Domains 1, 2 and 3 were cloned in frame with GFP under the alcohol dehydrogenase (alca) inducible promoter into psm5 (~ 9254bps), between the NheI and NotI digestion sites, producing the plasmids psm6 (~12483bps), psm7 (~11063bps), and psm8 (~11651bps) respectively (Figure 4.1A and B). Primer pairs SM7/8 (Domain1), SM11/12 (Domain 2), and SM13/14 (Domain3) were used to amplify these domains from genomic DNA. The amplified pieces were then digested with AvrII and BamHI (Domain1), NotI and NheI (Domain 2 and 3) and ligated into psm5. To clone domain 1, a linker made by primers SM9 and SM10 was used, whose one end attached to a digested BamHI site on Domain1, and the other attached to a digested NotI site on the psm5 plasmid, linking domain1 to the plasmid. The plasmids generated were then transformed individually into the GR5 (pyroa4) strain of A. nidulans, and were integrated at the pyroa locus. Transformants were selected on pyridoxine (pyro) (Figure 4.1B). For Domains 3, 4 and 5, on the other hand, cassettes carrying alca-domain X-GFP were generated by fusion PCR and targeted to replace the pyrg locus in strain TN02. The transformants were counter-selected on 5-FOA plates (Figure 4.2). For details on how each individual cassette was made refer to figure legend 4.2. Proper integration of all the domains by homologous recombination was confirmed by Diagnostic PCR and the expression of the fusion proteins was confirmed for domains 1, 2 and 3 by western blotting (Figure 4.7E). Live cell imaging was performed as described in section of chapter 2, except that spores carrying the different domains were grown in 3 ml of minimal media lacking ribo in the presence of 0.47% glycerol and UU. alca driven expression of these domains were conducted as described in section of chapter

145 4.2.2 Generation and analysis of Nup2ΔDomain4-GFP mutants To generate the Nup2ΔD4 mutants, Nup2ΔD4::GFP::PyrG construct was generated by fusion PCR and targeted into the Nup2 locus to replace the endogenous Nup2 gene by homologous recombination (Figure 4.3). For details on how the cassette was constructed refer to figure legend 4.3. This cassette was then transformed in SO451 (pyrg89), and transformants were selected on UU. Transformants were then streaked on YAG plates with or without UU to assess whether they are heterokaryons. The heterokaryon rescue technique described previously (chapter 2, section 2.5) was used to further analyze those mutants and maintain them. For live cell imaging, cells were grown and imaged in the same conditions as the Nup2 null mutants, described in section of chapter Generation and analysis of the Nup2 fusion mutants Nup2::GFP::Nup37, Nup2ΔD4::GFP::Nup37, Nup2::GFP::H1, and Nup2ΔD4::GFP::H1 proteins were expressed in A. nidulans under the endogenous Nup2 promoter. Two constructs, Nup2ΔD4::GFP::Nup37::PyrG, and Nup2ΔD4::GFP::H1::PyrG were made that target and replace the Nup2 locus by homologous recombination. Details on how individual cassettes are made are present in Figure legends 4.4 and 4.5. These cassettes were then transformed into SO451 (pyrg89) and transformants were selected on UU. According to the homologous recombination event that occurred, each construct had the potential to result in forming two different types of fusions, either with or without domain 4 (Figure 4.4 and 4.5). The type of integration present in the cells was confirmed by diagnostic PCR. Transformants were then streaked on YAG plates with or without UU to assess whether they were heterokaryons. If heterokaryons were formed, then further analysis of those strains was performed according to the heterokaryon rescue technique described previously in chapter 2, section 2.5. Haploid and diploid strains isolated were also confirmed by diagnostic PCR. For live cell imaging, spores from heterokaryons were grown and imaged in the same conditions as the Nup2 null mutants, described in section of 125

146 chapter 2. Spores from haploid and diploid strains were grown at room temperature (around C) in YG medium overnight (about hours). 4.3 Results A domain (domain 4) within Nup2 is sufficient to locate it to both NPCs and mitotic chromatin In an attempt to investigate whether Nup2 s translocation to mitotic chromatin is required for its essential function, we planned to define the domain within Nup2 responsible for its chromatin location. Deletion of such a domain within Nup2 will prevent its normal function on chromatin at mitosis but not its interphase function at the pore, thus helping us generate a separation-of-function allele of Nup2. Hence, a deletion analysis of the Nup2 protein was performed by tagging different truncations of Nup2 with the green fluorescent protein (GFP), and expressing them under the inducible promoter alca. The localization of these domains was then followed throughout the cell cycle (Figure 4.6A). Domain 4 which is a stretch of 401 aa (from aa 401 to aa 802) was the smallest domain found which located not only to the pore during interphase, but also to chromatin during mitosis (Figure 4.6B). The endogenous nup2 gene was then deleted in the background of cells carrying alca-domain4-gfp. Domain 4 was able to locate to both locations independently of the endogenous Nup2 (Figure 4.7). Thus, domain 4 is sufficient to locate GFP to the pore at interphase and to mitotic chromatin. This data suggest that domain 4 probably binds to either one protein that carries Nup2 to both the NPC and mitotic chromatin, or to two distinct proteins at each location. It was also observed that Domain 5 requires the endogenous Nup2 to access the nucleus (data not shown). Moreover, overexpression of these domains, and specifically domain 4, did not show any dominant negative effect on cell growth as A. nidulans strains overexpressing these domains did not show any defects in growth (Figure 4.8 and 4.9). 126

147 4.3.2 Domain 4 is necessary for Nup2 s translocation to both the NPC and mitotic chromatin and is essential for Nup2 s function Replacement of nup2 with nup2 d4::gfp::pyrg (Nup2 specifically lacking domain 4) generated either diploids or heterokaryons but not a nup2 d4::gfp::pyrg mutant haploid demonstrating that domain 4 carries an essential function of Nup2. Using the heterokaryon rescue technique (Osmani et al., 2006b), heterokaryons were identified by streaking asexual uninucleate spores (conidia) from transformants on plates with or without Uridine and Uracil (UU) (Figure 4.10A). Diagnostic PCR of the nup2 locus, using genomic DNA extracted from the heterokaryon, confirmed the presence of two genetically distinct types of nuclei (Figure 4.10A). Similar to the Nup2 null mutant, the nup2 d4 mutant shows no defect in germination, or short term growth, but stops growing after several rounds of defective mitosis (Figure 4.10B). Also like Δnup2, DAPI staining of nup2 d4 germlings showed defects in DNA segregation (Figure 4.10C). Because Domain 4 can locate to both the pore and to DNA, these data suggests that the essential function of this domain is to locate Nup2 to both locations. To test this, we followed Nup2 d4::gfp signal, by live cell microscopy, through a time span of a cell cycle in mutant germlings. Nup2 d4::gfp was found dispersed throughout the cell all the time. Nup2 d4::gfp thus could neither locate to the pore, nor to mitotic chromatin (Figure 4.10D). These results suggest that Domain 4 may be essential for Nup2 s function probably because it is necessary to locate Nup2 either to the pore at interphase or to chromatin during mitosis or to both sites. On the other hand, these results also suggest that Nup2 d4::gfp could be misfolded and thus a non-functional protein Efforts to artificially tether Nup2 exclusively to the NPC In an attempt to investigate whether Nup2 s translocation to mitotic chromatin is required for its essential function, we sought to generate a separation-of-function allele of Nup2 by constructing a chimera that will locate Nup2 exclusively to NPCs throughout the cell cycle. We thus replaced the endogenous nup2 with a version fused to Nup37 (nup2::gfp::nup37::pyrg). Nup37 is a small unessential core nucleoporin which stays 127

148 at the nuclear periphery during mitosis (Liu et al., 2009). Because Nup37 should locate Nup2 to the NE during interphase and mitosis, and thus prevent Nup2 s chromatin location, we expected to obtain heterokaryons from this study. On the contrary, we isolated haploids showing that the Nup2-GFP-Nup37 chimeric protein was functional (Figure 4.11 A and B). We then followed Nup2-GFP-Nup37 throughout the cell cycle to find that it locates to the nuclear pore during interphase and it appears to stay around nuclear periphery during mitosis in some cases (Figure 4.11C a) and does seem to translocate to DNA during mitosis in other cases, although not as clearly as a wild type Nup2 would do (Figure 4.11C b). These results were confusing and suggest that Nup2 s location at the pore might carry its essential function assuming that the chimera generated did really only locate at the pore throughout mitosis. The results also suggested that Nup2 may partially force Nup37 on chromatin during mitosis instead of it being forced to stay at the NE. We then hypothesized that if we delete the sequence within Nup2 responsible for its proper localization in cells, Nup37 will then be able to carry Nup2 to the NE exclusively throughout the cell cycle. So we attempted to artificially tether Nup2 to the NE by replacing the endogenous nup2 with a version of the previous chimera that lacks domain 4: nup2δdomain4-gfp-nup37. This time, we obtained heterokaryons as expected, showing that the Nup2ΔDomain4-GFP-Nup37 chimeric protein is nonfunctional (Figure 4.12 A and B). We also obtained diploids (Figure 4.12 C). Spores from both heterokaryons and diploids were grown on YG medium lacking UU and the Nup2ΔDomain4-GFP-Nup37 chimera followed by live cell microscopy either in diploid, carrying a copy of the wt Nup2 gene, or haploid germlings. Nup2ΔDomain4-GFP- Nup37 locates exclusively to the NPC throughout the cell cycle in both germlings (Figure 4.12 C and E). This suggests that the mitotic location of Nup2 is necessary for its function assuming that the Nup2ΔDomain4-GFP-Nup37 chimera is otherwise a functional protein at the pore and that deletion of domain 4 only alters Nup2 s localization but not its function once at the proper location. This may also suggest that domain 4 carries an essential function which is independent of locating Nup2 to its proper location in the cell. On the other hand, Nup2ΔDomain4-GFP-Nup37 chimera could locate at the NPC but be a non-functional protein. 128

149 4.3.4 Efforts to artificially tether Nup2 exclusively to DNA We also attempted to generate a Nup2 chimera that will only locate Nup2 to chromatin throughout the cell cycle. We hypothesized that if the chromatin location of Nup2 carries its essential function then this chimera should be functional. This would also answer that Nup2 s location to chromatin is not only necessary, as shown in the previous section, but also sufficient for Nup2 s essential function. To target Nup2 exclusively to chromatin throughout the cell cycle without interfering with chromatin functions, we chose to use histone H1 as the chromatin targeting protein. Histone H1 is not essential in A. nidulans (Ramon et al., 2000) and stays associated to chromatin during mitosis. We thus replaced the endogenous nup2 with nup2::gfp::histone H1::pyrG. The Nup2-GFP-H1 chimera was indeed functional but located to the nuclear pore during interphase and only moved to DNA during mitosis (Figure 4.13). This showed that also in this case Nup2 forces histone H1 onto the NPC during interphase instead of it being forced to stay at chromatin. As described in the previous section, we then attempted to artificially tether Nup2 to chromatin by replacing the endogenous nup2 with a version of the previous chimera that lacks domain 4: nup2δdomain4::gfp::h1::pyrg. We expected that this version of the chimera to locate only to chromatin and be functional. However, we obtained heterokaryons from this study showing that this chimera is non-functional (Figure 4.14 A and B). We also obtained diploids (Figure 4.14 C). We then followed Nup2ΔDomain4-GFP-H1 and found that it is unable to locate properly to neither the pore nor DNA in those germlings (Figure 4.14 D). This explained why the cells expressing this chimera were viable. In contrast, Nup2ΔDomain4-GFP-H1 could locate exclusively to DNA in diploids which carry a copy of wt Nup2 (Figure 4.14 E). The basis of this discrepancy is unknown. It is probable that Nup2 functions as a dimer specifically on DNA, thus wt Nup2 dimerises with Nup2ΔDomain4-GFP-H1 and this dimerization is required only to locate Nup2 to chromatin. 129

150 4.4 Discussion and future directions In this work, a significant amount of effort has been extended to generate separation-of-function alleles of Nup2 to differentiate between its interphase and mitotic functions. Although yet unsuccessful in generating a functional allele of Nup2 that resides at chromatin, or at the NPC, throughout the cell cycle, which will be a definite proof that Nup2 s essential function is on mitotic chromatin or NPC, our work gives insights into the mechanism by which Nup2 locates to the pore and to chromatin. Our findings thus not only start to elucidate the mechanism by which Nup2 is targeted to its proper locations in the cell, but also builds a solid base upon which future experiments should depend on to finally accomplish the challenge to differentiate between its interphase and mitotic functions Mechanism of Nup2 s translocation to chromatin We had set to define a domain within Nup2 that is responsible for its DNA location, with the hope of generating a separation-of-function allele of Nup2, and understanding the mechanism of its translocation to chromatin. However, we have discovered a stretch of 401 aa within Nup2, Domain 4, that is sufficient and necessary to translocate Nup2 to both the NPC and mitotic chromatin. It is intriguing that a single domain be responsible for both locations. Nevertheless, this domain is quite large and further investigation is needed to know whether we can differentiate two different motifs within domain 4 that can independently target either to the pore or to DNA. In this case, Nup2 will have two different binding partners at each site, and each motif within domain 4 will be responsible for its binding to a specific binding partner. One alternate explanation is that Nup2 s location to the pore and to chromatin is dictated by one binding partner, which tethers Nup2 to these locations through its interaction with domain 4. Data from chapter 3 strongly favors the latter hypothesis. NupA which copurifies with Nup2 at both G2 and M, is required for Nup2 s location to the NPC at interphase and to mitotic chromatin. These data then suggest that NupA targets Nup2 to either the pore or to DNA via domain 4. In experiments planned to study whether NupA targets Nup2 to the pore and to DNA through domain 4, the endogenous Nup2 can be 130

151 replaced with nup2δdomain4::gfp::nupa. Experiments then can be conducted to assess whether this chimera is capable to locate to the pore and to DNA and rescue the lethal phenotype observed in cells carrying nup2δdomain4::gfp. This will show that domain 4 s essential function is to target Nup2 to the pore and to mitotic DNA, through its interaction with NupA. Moreover, further work to define the domains within NupA responsible for its NPC vs chromatin locations, will elucidate the mechanism by which NupA targets to both locations. The hope is to define a particular domain within NupA that targets only to mitotic DNA or only to NPCs. Affinity purifications of such a domain using mitotic samples might purify mitotic specific binding partners of the Nup2-NupA complex helping us better understand the mechanism by which this complex binds to mitotic DNA and its function at chromatin during mitosis. Moreover, it is very interesting that the Aspergillus nidulans Nup2, similar to many of the filamentous fungi Nup2, is a significantly larger protein than its yeast and human counterparts. We have previously assumed that the reason why An-Nup2 is a much larger protein is that it carries additional functional domains not present in its yeast and humans orthologues. In fact, Domain 4, which is conserved in other filamentous fungi, seems to be absent from its yeast and vertebrate orthologues. This then, strengthens the argument stated in chapter 3, that metazoans might have evolved a distinct mechanism to target Nup2 to the pore and to chromatin which is independent of NupA. Moreover, this domain is absent in yeast probably because the yeast NPCs do not disassemble at all during mitosis and Sc-Nup2 does not translocate to DNA. It is likely then, that the more complex filamentous fungi have evolved this domain to target Nup2 to both locations. Metazoans must have then lost this domain throughout evolution as they probably evolved a distinct mechanism to accomplish this function Separation-of-function alleles of Nup2 In this chapter we have worked to isolate separation of function alleles of Nup2. To accomplish this, different Nup2 chimeric proteins were made that can either target to the NPC or to chromatin. We have isolated haploid strains that carry the Nup2-GFP- Nup37, and Nup2-GFP-Histone H1 chimeric proteins. In both cases, we were 131

152 unsuccessful to generate the separation-of-function alleles, as Nup2 was able to locate both to the pore and to DNA via domain 4. We have however successfully generated a chimeric protein Nup2ΔDomain4-GFP-Nup37 that tethers Nup2 exclusively to the NPC throughout the cell cycle as domain 4 is not able to compete and force Nup2 on chromatin during mitosis. This protein is always at the pore and is nonfunctional. Thus, artificially tethering Nup2 to the pore throughout the cell cycle is lethal. This suggests that chromatin location of Nup2 is essential for its function. One major problem with this experiment is the fact that the Nup2ΔDomain4-GFP-Nup37 chimera can be nonfunctional, not because it is unable to locate to the NPC, but because the absence of Domain 4 within Nup2 forms a nonfunctional protein, either due to the loss of a function that domain 4 carries or changes in the properties of the protein, such as proper folding. Subsequent efforts were then made to generate a functional allele of Nup2 that resides at chromatin throughout the cell cycle, by replacing endogenous nup2 with nup2δdomain4::gfp::h1. These efforts have not yet been successful. These experiments will test whether chromatin location is in fact sufficient for Nup2 s essential function and would provide clear-cut evidence that Nup2 s essential function is on mitotic chromatin. Thus, there is still much to be done to be able to generate separation-of-function alleles of Nup2. The goal is to isolate a chimera that exclusively locates Nup2 to one location which is functional. Many approaches can be taken in future experiments. First, the deletion of nup37 in the nup2::gfp::nup37 haploid strain, and the deletion of histone H1 in nup2::gfp::histone H1, might more effectively force Nup2 to either location without competition for each site by the endogenous version of the ectopic targeting protein. Moreover, the deletion of nupa in both of these strains might better force Nup2 to either the pore or to DNA, as without NupA, Nup2 will loose its intrinsic signal to locate at both places and will follow the lead of the protein it is fused to. These experiments will also answer whether NupA s essential function is only to tether Nup2 to the pore or to DNA. An alternate way to generate the separation-of-function allele is to define a domain within NupA responsible to tether NupA to the pore for example. One can imagine that deletion of such a domain will prevent NupA and Nup2 location at the pore during interphase but allow its mitotic chromatin location. This will allow us then to test whether a strictly chromatin location of NupA and Nup2 is sufficient for their essential function. 132

153 4.4.3 Can Nup2 s dynamic locations carry one essential function? The experiments suggested above plan to generate a separation-of-function allele of Nup2 depending on its location, thus assuming that each location of Nup2 carries a separate function. What if that is not the case? One location might be important for multiple functions, or both locations might be important for one essential function. Because Nup2 is suggested to be a candidate protein that coordinates different mitotic events, in addition to it having such a dynamic localization within cells, it is inevitable to think that Nup2 s dynamic movement between locations might be important for its essential function. For example, Nup2 might function as a messanger that transmits a signal from the pore to DNA upon mitotic entry to promote DNA condensation, thus coordinates those events. One way we can test that in the future is by generating cells that lack NupA and carry two different types of Nup2 chimeric proteins, Nup2-Nup37, and Nup2-H1 instead of the endogenous Nup2. In case the movement of Nup2 between the pore and chromatin is important, then these cells will not survive as the movement of Nup2 between these two locations will be hindered. Furthermore, if the actual movement between the NPC and chromatin is important for Nup2 s essential function, then generation of separation-of-function alleles that differentiate between its transport and mitotic functions depending on its location will not be possible. Further work on the mechanism by which Nup2 is targeted to DNA, the insights about its function in mitosis and the identification of its mitotic specific binding partners will help to generate an allele that separates its mitotic function from its nuclear transport role. 133

154 Figure 4.1 Method used to integrate GFP tagged domains 1-3 into the A. nidulans genome under the alca promotor at the pyroa locus. (A) Cartoon diagram of the integrating plasmid psm5 showing the cloning site where the Nup2 domains were inserted. psm5 was made by subcloning the 3/4 pyro gene into the integrating plasmid pal5-gfp. (B) 3 different plasmids psm6, 7 or 8 containing domains 1 (bps:1-3231), 2 (bps: ) and 3 (bps: ) respectively were made. These plasmids were integrated into the A. nidulans genome at the pyroa locus. The recipient strain GR5 carries a mutation at the pyroa locus (pyroa4). This mutation is depicted by the green bar. Homologous recombination which occurs only beyond the mutation will give rise to a wt pyro gene. Transformants that have the plasmid integrated are selected on pyro. 134

155 Figure

156 Figure 4.2 Method used to integrate GFP tagged domains 3-5 into the A. nidulans genome under the alca inducible promotor at the pyrg locus. (A) Primers SM24 and SM25 were used to amplify the alca-domain3-gfp-h2a from the psm8 plasmid. Sequence 1 KB upstream and downstream of pyrg were also amplified. pyrg deletion cassette was then made by 3-way fusion PCR using these 3 pieces, and the pyrg locus was replaced by this cassette by homologous recombination in strain TN02. Transformants were then counter-selected on 5-FOA. Only the ones that are deleted for pyrg could survive. (B) To introduce alca-d4-gfp-h2a at the pyrg locus, primers SM21/28 and SM29/23 were used to amplify 2 pieces from the genome of the transformant obtained from (A). 2-way fusion PCR using these 2 pieces was then performed to produce the pyrg targeting cassette. (C) For alca-d5-gfp-h2a, the same technique was used. Primer pairs SM21/30, 31/32, and 29/23 were used to amplify 3 pieces which were then fused by 3-way fusion PCR constructing the pyrg targeting cassette. 136

157 Figure

158 Figure 4.3 Generation of the Nup2ΔDomain4-GFP protein. nup2δd4::gfp::pyrg was placed under the endogenous nup2 promotor in the A. nidulans genome. To accomplish that, a Nup2 replacement cassette was made that targets the nup2 locus, and replaces the endogenous nup2 in the genome with nup2δd4::gfp::pyrg. To construct the cassette, primer pairs JD151/SM52 and SM53/JD156 were used to amplify two pieces from the genome of SO593, whose nup2 locus was previously tagged with GFP::PyrG. 2- way fusion PCR of these two pieces created the nup2δd4::gfp::pyrg cassette. This cassette was then transformed in strain SO451 (pyrg89) and transformants were selected on UU. 138

159 Figure 4.4 Generation of the Nup2-Nup37 chimeric protein. nup2::gfp::nup37::pyrg and nup2δd4::gfp::nup37::pyrg were placed under the endogenous nup2 promotor in the A. nidulans genome. To accomplish this, a Nup2 deletion cassette carrying nup2δd4::gfp::nup37::pyrg was made that targets the nup2 locus, and replaces the endogenous nup2 in the genome. To construct the cassette, primer pairs JD151/SM52, SM53/SM75, and SM76/JD156 were used to amplify three pieces from the genome of SO593, whose nup2 locus was previously tagged with GFP::PyrG. Primers SM77 and SM78 were also used to amplify the nup37 locus from the genome. 4- way fusion PCR of the four pieces created the nup2δd4::gfp::nup37::pyrg cassette. This cassette was then transformed in SO451 (pyrg89) and transformants were selected on UU. If recombination events 1 and 3 happened then, nup2δd4::gfp::nup37::pyrg was introduced into the SO451 genome. On the other hand, if recombination events 2 and 3 happened, then nup2::gfp::nup37::pyrg was introduced into the SO451 genome. 139

160 Figure

161 Figure 4.5 Generation of the Nup2-Histone H1 chimeric protein. nup2::gfp::h1::pyrg and nup2δd4::gfp::h1::pyrg were placed under the endogenous nup2 promotor in the A. nidulans genome. To accomplish this, a Nup2 replacement cassette, carrying nup2δd4::gfp::h1::pyrg, was made that targets the nup2 locus, and replaces the endogenous nup2 in the genome. To construct the cassette, primer pairs JD151/SM52, SM53/SM75, and SM76/JD156 were used to amplify three pieces from the genome of SO593, whose nup2 locus was previously tagged with GFP::PyrG. Primers SM79 and SM80 were also used to amplify the histone H1 locus from the genome. 4-way fusion PCR of the four pieces created the nup2δd4::gfp::h1::pyrg. This cassette was then transformed in SO451 (pyrg89) and transformants were selected on UU. If recombination events 1 and 3 happened then nup2δd4::gfp::h1::pyrg was introduced into the SO451 genome. On the other hand, if recombination events 2 and 3 happened then nup2::gfp::h1::pyrg was introduced into the SO451 genome. 141

162 Figure

163 Figure 4.6 A single domain within Nup2, Domain 4 (aa ), is responsible for its translocation to the NPC during interphase and to mitotic chromatin. (A) Cartoon depicting the different truncations of Nup2 (domains 1-5) which were GFP tagged and expressed under the alca inducible promoter in a background containing wild type Nup2. (B-C) The location of GFP-Tagged domains of Nup2 was followed through mitosis using live cell spinning disk confocal microscopy. Bar, ~ 5μm. (B) While domains 2 and 5 disperse throughout the cell during mitosis, domains 1, 3 and 4 are at nuclear pores during interphase and translocate to DNA during mitosis (number of nuclei tested for each domain is 13). 143

164 Figure 4.7 Domain 4 is sufficient to locate Nup2 to the NPC at interphase and to mitotic chromatin. The location of GFP-Tagged Domain4 was followed through mitosis using live cell spinning disk confocal microscopy in the presence or absence of endogenous Nup2. Domain 4 is at the NPC and at mitotic chromatin in the presence or absence of endogenous Nup2. Bar, ~ 5μm. 144

165 Figure 4.8 Overexpression of domains 1-3 does not cause growth defects. (A-D) Strains carrying domains 1-3 under the alca promoter were spotted on repressive (Glucose (A), Yeast Extract Lactose (C)) and inducing (Ethanol (B), YEL + 40mM threonine (D)) media, and incubated at different (25 C, 32 C, 37 C, 42 C) temperatures for ~ 2 days. SO51 was used as a WT control. No growth defects were observed upon overexpression of the three domains. (E) Western blots detecting the expression of the three domains using anti-gfp. The cells were grown in YEL (-) and alca was induced with 40mM threonine (+). The lower panel shows equal loading, the blot was stained with anti- -tubulin. 145

166 Figure

167 Figure 4.9 Overexpression of domains 4 and 5 does not cause growth defects. (A-C) Strains carrying domains 3-5 under the alca promoter were spotted on repressing (Glucose (A)), non-repressing/ non-inducing (Glycerol (B)), and inducing (Ethanol (B)) media, and left at different (25 C, 32 C, 37 C, 42 C) temperatures for about 2 days. GR5, TN02 and R153 were used as WT controls. No growth defects were observed upon overexpression of these domains. 147

168 Figure

169 Figure 4.10 Domain 4 is necessary to locate Nup2 to both the NPC at interphase and to mitotic chromatin, and is essential for Nup2 s function. (A) Conidia from transformants carrying the nup2 d4::gfp::pyrg cassette were streaked on media with or without UU. Heterokaryons were defined as producing some spores which grow on +UU but not on -UU. Diagnostic PCR of a heterokaryon showing 2 bands corresponding to the wt Nup2 gene and the nup2 d4::gfp::pyrg allele. (B-C) Cells expressing the mutant Nup2ΔD4 protein do not survive and show defects in mitotic DNA segregation. (B) Depicted are bright field images of spores from heterokaryons containing either the wild type Nup2 or the domain 4 deleted Nup2 proteins for 3 days at 22 C. Bar, ~ 20μm. (C) DAPI staining of the wild type vs the mutant germling shows miss-segregated DNA due to the deletion of domain 4. (D) Nup2ΔD4 protein is dispersed throughout the cell. Bar, ~ 5μm. 149

170 Figure

171 Figure 4.11 The Nup2::GFP::Nup37 protein is functional (A) Cartoon depicting nup2::gfp::nup37::pyrg at the nup2 locus. (B) Endogenous replacement of nup2 with nup2::gfp::nup37::pyrg produces haploid A. nidulans strains that show no growth defect (data not shown). Diagnostic PCR of a haploid strain showing a single band corresponding to the nup2::gfp::nup37::pyrg allele. (C) Still images from a movie captured by live cell confocal microscopy following Nup2::GFP::Nup37 throughout mitosis. Nup2::GFP::Nup37 is at the NPC during interphase and seems to either stay around nuclear periphery during mitosis (a) or move at least partially to mitotic chromatin (b). Bar, ~ 5μm. This experiment was performed in collaboration with Subbulakshmi Suresh. 151

172 Figure

173 Figure 4.12 Artificially NPC tethered Nup2ΔD4 is not functional. (A) Cartoon depicting the nup2δd4::gfp::nup37::pyrg cassette replacing the endogenous nup2. (B- D) The Nup2ΔD4::GFP::Nup37 protein is non-functional as endogenous replacement of nup2 with nup2δd4::gfp::nup37::pyrg produces either heterokaryons (B) or diploids (D) in A. nidulans. Diagnostic PCR of either a heterokaryon (B) or a diploid strain (D) showing 2 bands corresponding to the wt nup2 gene and the nup2δd4::gfp::nup37::pyrg allele. (C-E) Still images from movies captured by live cell confocal microscopy following Nup2ΔD4::GFP::Nup37 throughout mitosis in either a haploid nup2δd4::gfp::nup37::pyrg germling from a heterokaryon or a diploid germling carrying nup2δd4::gfp::nup37::pyrg in addition to a wt copy of Nup2. Nup2ΔD4::GFP::Nup37 stays at the NPC during interphase and mitosis in the presence or absence of endogenous Nup2. Bar, ~ 5μm. This experiment was performed in collaboration with Subbulakshmi Suresh. 153

174 Figure

175 Figure 4.13 The Nup2::GFP::H1 protein is functional and locates to the NPC at interphase and to mitotic chromatin. (A) Cartoon depicting nup2::gfp::h1::pyrg at the nup2 locus. (B) Endogenous replacement of nup2 with nup2::gfp::h1::pyrg produces haploid A. nidulans strains that show no growth defect (data not shown). Diagnostic PCR of a haploid strain showing a single band corresponding to the nup2::gfp::h1::pyrg allele. (C) Still images from a movie captured by live cell confocal microscopy following Nup2::GFP::H1 throughout mitosis. Nup2::GFP::H1 is at the NPC during interphase and moves to chromatin during mitosis. Bar, ~ 5μm. This experiment was performed in collaboration with Subbulakshmi Suresh. 155

176 Figure 4.14 The Nup2ΔD4::GFP::H1 protein is not functional and does not locate to chromatin in the absence of endogenous Nup2. (A) Cartoon depicting the nup2δd4::gfp::h1::pyrg cassette replacing the endogenous nup2. (B-D) The Nup2ΔD4::GFP::H1 protein is non-functional as endogenous replacement of nup2 with nup2δd4::gfp::h1::pyrg produces either heterokaryons (B) or diploids (D) in A. nidulans. Diagnostic PCR of either a heterokaryon (B) or a diploid strain (D) showing 2 bands corresponding to the wt nup2 gene and the nup2δd4::gfp::h1::pyrg allele. (C) Bright field image and maximum intensity projection of the Nup2ΔD4::GFP::H1 signal in spores from a heterokaryon carrying either wt nup2 or nup2δd4::gfp::h1::pyrg nuclei. Nup2ΔD4::GFP::H1 is dispersed throughout the cell in germlings carrying nup2δd4::gfp::h1::pyrg nuclei. (E) Still images from a movie captured by live cell confocal microscopy following Nup2ΔD4::GFP::H1 throughout mitosis in a diploid germling carrying nup2δd4::gfp::h1::pyrg in addition to a wt copy of Nup2. Nup2ΔD4::GFP::H1 locates at chromatin during both interphase and mitosis in the presence of endogenous Nup2. Bar, ~ 5μm. This experiment was performed in collaboration with Subbulakshmi Suresh. 156

177 Figure

178 CHAPTER 5 EFFORTS TO DEFINE IF MAMMALIAN NUP50 IS A FUNCTIONAL HOMOLOGUE OF A. nidulans NUP2 5.1 Introduction In Aspergillus nidulans, all Nup2 transfers from nuclear pores to DNA during mitosis (Osmani et al., 2006a). Very interestingly, the mammalian orthologue of Nup2, Nup50 has been also shown to locate at the vicinity of chromatin during mitosis (Dultz et al., 2008). This suggests that Nup2 has a role at mitotic chromatin that is conserved from lower to higher eukaryotes. Thus, in this chapter, we attempted to investigate whether the mammalian orthologue of Nup2 translocates to chromatin at mitosis, and if it would complement the essential role of Nup2 in Aspergillus nidulans. This study then may provide evidence that the novel essential mitotic function of A. nidulans Nup2 is conserved in humans. Knowledge gained from these experiments might help us identify a potentially novel essential function of the mammalian Nup50 in mitosis. To accomplish this we first expressed mouse Nup50 in Aspergillus nidulans and found that it is a nuclear protein at interphase and disperses throughout the cell during mitosis. As expected, Nup50, which could not locate to the pore or to mitotic chromatin, could not complement Nup2 s essential function. This suggested that the mechanism by which Nup50 targets to the mammalian NPC and chromatin is distinct. Hence we thought that Nup2 s function at mitotic chromatin might still be conserved in mammals, although the mechanism of its 158

179 translocation to chromatin is probably not consereved. We then thought that if we force Nup50 to locate at the interphase NPC and mitotic chromatin in A. nidulans, then it might be able to complement An-Nup2. As discussed previously (chapters 1 and 4), Nup50 is a much smaller protein than An-Nup2. The domain which is missing from Nup50 is Domain 4 which was defined in chapter 4 as a domain sufficient and necessary to carry Nup2 to both nuclear pores and DNA. Thus, we are currently working on experiments to artificially locate Nup50 to the pore and mitotic chromatin in Aspergillus nidulans by tagging it to Domain 4 and asking whether this version of Nup50 will be able to complement Nup2 s essential function. 5.2 Material and Methods Generation and analysis of strains carrying alca-egfp2-nup50 EGFP2-mmNup50 was introduced under the alcohol dehydrogenase inducible promoter alca at the pyrg locus by homologous recombination (Figure 5.1 and 5.2 A). Four way fusion PCR was used to prepare the alca::egfp2-mmnup50 cassette which is targeted to the pyrg locus. For details on how the cassette was made refer to Figure legend 5.1. This cassette was then transformed in strain SO455, and transformants were selected on 5FOA+UU. Live cell imaging was performed as described in section of Chapter 2, except that spores carrying the cassette were grown in 3 ml of minimal media in the presence of 1% Ethanol and UU. alca driven expression of Nup50 was conducted as described in section of Chapter Complementation Study Nup2 was deleted in strains carrying alca::egfp2-mmnup50. Heterokaryons were identified by streaking conidia from transformants on YAG plates with or without Uridine and Uracil (UU). The heterokaryon rescue technique described previously (chapter 2, section 2.5) was used to maintain and analyze these mutants. Conidia from heterokaryons were streaked on minimal media plates without UU and with 1 % Ethanol 159

180 to induce alca expression of EGP2-Nup50. Cell growth was assessed after 1 or 3 days at 32 C Endogenous replacement of Nup2 with Nup50 with or without domain 4 EGFP2-Nup50, D4::EGFP2-Nup50, EGFP2-Nup50::D4 proteins were expressed in A. nidulans under the endogenous Nup2 promoter (Figure 5.4 and 5.5). Three Constructs EGFP2-mmNup50::pyrG, EGFP2-mmNup50::D4::pyrG, D4::EGFP2- mmnup50::pyrg are currently being made that target and replace the nup2 locus by homologous recombination. Details on how individual cassettes are made are present in Figure legends 4.4 and 4.5. These cassettes will then be transformed into strain SO451 (pyrg89) and transformants will be selected on UU. Transformants will be then streaked on YAG plates with or without UU to assess whether they are heterokaryons (expressed protein does not complement Δnup2) or haploid (expressed protein does complement Δnup2). 5.3 Results Induction of Nup50 expression under the alca promoter in Aspergillus nidulans does not complement Nup2 s function. In an attempt to test whether the Mus musculus(mm)- Nup50 complements the Aspergillus nidulans Nup2 we introduced mmnup50 into A. nidulans and expressed it under the alca inducible promotor. In a subsequent step, we deleted An-Nup2 in this background. Heterokaryons were identified by streaking conidia from transformants on YAG plates with or without Uridine and Uracil (UU). Conidia from either Δnup2; alca::egfp2-mmnup50 heterokaryons or Δnup2 heterokaryons were then streaked on minimal media dextrose (repressing) and ethanol (inducing) plates lacking Uridine/Uracil. The Δnup2; alca::egfp2-mmnup50 and Δnup2 mutant cells grow with similar rates (data not shown). Thus, these preliminary results suggest that high levels of mmnup50 may not complement Nup2 s function in Aspergillus nidulans. Future experiments, as discussed later in this chapter, including the replacement of the endogenous Nup2 with Nup50, will further solidify this result. 160

181 5.3.2 alca induced Nup50 is unable to locate to either the NPC nor to mitotic chromatin in the presence or absence of endogenous Nup2 To study whether Nup50 can translocate to mitotic chromatin in A. nidulans we followed EGFP2-Nup50 in the presence (Figure 5.2B) and absence (data not shown) of endogenous Nup2 throughout the cell cycle. In both cases, Nup50 was nuclear in interphase and dispersed throughout the cell during mitosis. This explains why Nup50 could not complement Nup2 s essential function and suggests that the mechanism by which Nup50 targets to the mammalian NPC and chromatin is distinct from its counterpart in A. nidulans. However, Nup2 s function at mitotic chromatin might still be conserved in mammals, thus it is possible that once targeted properly to mitotic chromatin (and/or NPCs), Nup50 might be able to complement Nup2 s essential mitotic function. 5.4 Discussion and future directions Here, we attempt to complement Nup2 s essential mitotic function with its mammalian orthologue Nup50. Interestingly, we find that mammals may have evolved a distinct mechanism to target the highly conserved nucleoporin Nup2 to either the NPC at interphase or chromatin during mitosis. This finding provides a solid direction for future attempts in asking whether Nup50 does complement Nup2 s mitotic function. If successful, this study will not only help us suggest a potentially novel function of the mammalian Nup50 in mitosis, but also provide the first evidence of a conserved mitotic role of a nucleoporin from fungi to humans Nup2 targets to nuclear pores and to DNA in distinct ways in A. nidulans and mammals We have shown that the mammalian ortholog of Nup2 targets to the NPC during interphase and to chromatin during mitosis in a manner distinct from An-Nup2. EGFP2- Nup50 when expressed in A. nidulans can neither locate to the pore nor to mitotic 161

182 chromatin. Thus, Nup50 is probably incorporated to the mammalian NPC and mitotic chromatin through a different mechanism. We have shown in chapter 3 that Nup2 is targeted to nuclear pores and to DNA through its binding partner NupA. Nup50 thus seems to be unable to bind to NupA neither at the pore nor at mitotic chromatin. This is expected as NupA has a very low sequence similarity to any of the Nups that bind Nup50 in higher eukaryotes. This then strengthens the hypothesis stated in chapter3 that NupA might not be conserved in mammals which have evolved a distinct mechanism to target Nup2 to nuclear pores and to DNA. On the other hand, this data might also suggest that N-terminally tagging Nup50 with two EGFPs might render Nup50 nonfunctional which explains why it is unable to complement Nup2 s function How to test for complementation in the future? Although, the mechanism by which Nup50 targets to the mammalian NPC and chromatin may be distinct, Nup2 s function at mitotic chromatin might still be conserved in mammals. To be able then to study whether Nup50 does complement Nup2 s essential function, we need to target Nup50 to its proper locations in A. nidulans. We have suggested in chapter 4 that NupA targets Nup2 to the pore and DNA through Nup2 s Domain 4. Domain 4 within Nup2 is sufficient to target GFP to NPCs and mitotic chromatin and necessary to locate Nup2 at both places. Thus, we are currently testing whether the endogenous replacement of Nup2 with Nup50 with or without Domain 4 complements An-Nup2 s function. In case Nup50 does complement Nup2 s function in A. nidulans, we then would like to investigate whether Nup50 recues the Nup2 deletion phenotype if only located at chromatin. To accomplish that, we are going to design experiments, as in Chapter 4, section 4.2.3, which will force Nup50 only to chromatin, probably by fusing it to histone H1. Because Nup50 does not carry an intrinsic signal that locates it either to nuclear pores or DNA, its targeting to DNA is expected to be more straightforward. 162

183 Figure 5.1 Generation of EGFP2-alcA-mmNup50 cassette targeted to the pyrg locus of Aspergillus nidulans. (A) mmnup50 cdna tagged with two EGFPs (piece 2) was amplified by primers SM57/SM59 from pegfp2-mmnup50 (gift from J. Ellenberg, EMBL, Heidelberg, Germany.) alca was amplified by primers SM24/SM65 from the plasmid pal5-gfp (piece 2), and 1KB upstream (piece 1) and downstream (piece 4) of pyrg were amplified with primer pairs SM21/20 and SM22/66 respectively. (B) Agarose gel showing PCR products of individual pieces (1 to 4) used to generate the fused cassette. (C) A cartoon depicting the cassette made by 4-way fusion PCR of the pieces 1 to 4. (D) One and five micro liters of the fusion PCR product were run on a 0.8% agarose gel. This experiment was performed in collaboration with Jessica S. El-Hallal. 163

184 Figure

185 Figure 5.2 mmnup50 locates to the Aspergillus nidulans nucleus in interphase and disperses throughout the cell in mitosis. (A) The alca::egfp2-mmnup50 generated cassette was transformed into the pyrg locus of the A. nidulans strain SO455. Transformants were counter selected on plates containing 5-Fluoroorotic acid (5-FOA) and Uracil/Uridine to select for the ones that have the pyrg gene deleted and the construct integrated at this locus. (B) Depicted are still images from time lapse confocal microscopy of living cells carrying either the endogenously tagged An-Nup2-GFP or EGFP2-mmNup50. Nup2 locates to the NPC at interphase and to DNA during mitosis, but its mammalian homolog is nuclear in A. nidulans at interphase and disperses throughout the cell during mitosis. Bar, ~ 5μm. This experiment was performed in collaboration with Jessica S. El-Hallal. 165

186 Figure

187 Figure 5.3 Replacement of endogenous A. nidulans Nup2 with EGFP2-mmNup50. EGFP2-mmNup50::PyrG will be placed under the endogenous nup2 promotor in the A. nidulans genome. To accomplish that, a Nup2 deletion cassette is made that targets the nup2 locus, and replaces the endogenous nup2 in the genome with EGFP2- mmnup50::pyrg. 1kb upstream of Nup2 is amplified with primers JD151/SM89; then another piece (carrying Pyrg + 1kb downstream of Nup2) is amplified using the primers LU233 and JD156 from the genome of SO593, whose nup2 locus was previously tagged with GFP::PyrG. EGFP2-mmNup50 is amplified using primers SM88 and SM82. 3-way fusion PCR of these three pieces will create the EGFP2-mmNup50::PyrG cassette. This cassette will be then transformed in SO451 (pyrg89) and transformants will be selected on UU. This experiment is being conducted in collaboration with Jessica S. El-Hallal. 167

188 Figure 5.4 Replacement of endogenous A. nidulans Nup2 with EGFP2-mmNup50 containing domain 4. (A-B) EGFP2-mmNup50::PyrG containing domain 4 will be placed under the endogenous nup2 promotor in the A. nidulans genome. To accomplish that, two different Nup2 deletion cassette are made that target the nup2 locus, and replace the endogenous nup2 in the genome with either EGFP2-mmNup50::D4::PyrG (A) or D4::EGFP2-mmNup50::PyrG (B). 1kb upstream of Nup2 is amplified with either primers JD151/SM89 or JD151/SM86; then another piece (carrying Pyrg + 1kb downstream of Nup2) is amplified using the primers LU233 and JD156 from the genome of SO593, whose nup2 locus was previously tagged with GFP::PyrG. EGFP2-mmNup50 is amplified using primers either SM88/SM83 or SM88/SM82. Domain 4 is amplified using primers either SM84/SM85 or SM84/SM90. 4-way fusion PCR of the specific four pieces will create the individual cassettes. These cassettes will be then transformed in SO451 (pyrg89) and transformants will be selected on UU. This experiment is being conducted in collaboration with Jessica S. El-Hallal. 168

189 Figure

190 CHAPTER 6 FINAL DISCUSSION 6.1 Overview Mitosis is a highly regulated and conserved mechanism by which cells divide their nucleus into two. It is thus essential for the survival of unicellular and multicellular organisms alike. Proper regulation of mitosis is crucial as any missregulation can lead to the development of human diseases such as cancer. Therefore, deciphering the intricate network of events that happen during mitosis, and how they are regulated, is a prerequisite for understanding the development and treatment of those diseases. Although there has been enormous progress in the field, the detailed mechanisms by which many individual mitotic events happen, and how these events are coordinated, are still very poorly understood. For instance, the mechanism and regulation of the reversible stepwise mitotic disassembly of NPCs and its coordination with mitotic events such as the cycle of DNA condensation and spindle formation is still not understood. The mechanism by which Ran coordinates different mitotic events and how it is regulated is also not fully understood. In fact, the relationship between transport proteins and mitosis is just emerging and much work should be done to further understand this connection. Moreover, many mitotic targets of the main kinases that regulate different events throughout mitosis still remain to be discovered. For example, how NIMA promotes spindle assembly is yet not understood. Hence, there is 170

191 still much to be discovered to achieve a comprehensive understanding on how cells divide their nucleus. In this work, we strive to understand the regulation of mitosis in Aspergillus nidulans in hopes that our findings will shed light on the regulation of mitosis in all eukaryotes. In chapter 3, we identify a novel and essential nucleoporin in A. nidulans, we have named NupA which, like Nup2, all translocates from NPCs to chromatin during mitosis. In addition, we define the first Nups, Nup2 and NupA, in A. nidulans that have essential roles in mitosis which are regulated by the mitotic NIMA kinase. In chapter 4, we provide insights into the mechanism by which Nup2 locates to nuclear pores and to chromatin. Moreover, efforts are described that aim to differentiate between Nup2 s interphase and mitotic functions. In chapter 5, we find that mammals have evolved a distinct mechanism to target the highly conserved nucleoporin Nup2 to the NPC at interphase and chromatin during mitosis. Efforts are attempted to complement Nup2 s essential function with the mammalian Nup50. Our work thus improves our understanding on how NIMA regulates mitotic progression in A. nidulans, and extends the knowledge in the field of the mitotic roles of NPC associated proteins. We thus contribute in fortifying the evolving connection between nuclear transport proteins and mitosis. Most importantly though, our work provides solid groundwork upon which future experiments can build to finally accomplish the challenge to differentiate between the interphase and mitotic functions of Nup2, and to decipher the functions and the mechanism of action of the NupA-Nup2 complex during mitosis. 6.2 The proposed model (Figure 6.1) NupA targets Nup2 to the NPCs during interphase via Domain4 where Nup2 is in a complex with NupA, in addition to the importins and. To initiate mitosis both CDK1 and NIMA are activated (Fleig and Gould, 1991; Ye et al., 1995), and promote phosphorylation of Nup2 and probably NupA and the importin / complex. Phosphorylation of Nup2 and NupA promote their release from the pores, and their movement to and association with chromatin. In fact, mitotic specific phosphorylation might regulate not only the dynamic transitions of these proteins in a spatial and temporal 171

192 manner, but also their function at each location. Further work will map and elucidate the function of these mitotic phosphorylation events in the regulation of these proteins. Once associated with chromosomes, Nup2, NupA and the importin / complex condense with chromatin at prophase, and stay associated with the highly condensed chromosomes throughout metaphase. At this location, Nup2 has a role in metaphase spindle assembly and stability by potentially promoting mitotic specific cargo release from the importin / complex by binding to RanGTP, whose levels are high at the vicinity of mitotic chromatin, and promoting the dissociation of the / complex and thus facilitating the release of mitotic specific cargos. Mitotic specific cargo could be either spindle assembly factors (SAFs) and/or motor proteins. Therefore, Nup2 might make mitosis more efficient by properly targeting such cargos to the right place at the right time, as fast as, and as efficiently as, possible. It is easy to imagine how essential this role can be if mitosis as a whole takes only 5 minutes. At anaphase, these proteins then move from being tightly associated with chromatin to residing around the periphery of the chromatin mass where they have a role in proper post-mitotic nuclei reassembly. The role of Nup2 and NupA in building post-mitotic nuclei might be in nuclear envelope fission upon mitotic exit, and/or proper post-mitotic nuclear basket reassembly, and/or DNA decondensation, thus aiding proper chromatin tethering to the nuclear pores, and correct nuclear threedimensional organization. We propose that Nup2 has a function during mitotic exit which is to promote cargo release from the importin / complex at the vicinity of chromatin during the NPC reassembly. Thus, Nup2 and NupA play multiple roles in the cell, a potentially non-essential interphase role in transport in concert with all other Nups, and exclusive and unique essential roles during mitotic progression. In fact, we propose both carry dual roles in mitosis, one leading up to metaphase and the other during mitotic exit and into G1. We further propose Nup2 and NupA also play a role in coordinating mitotic events such as NPC disassembly with spindle assembly and chromosome segregation with post-mitotic reassembly of daughter nuclei. 172

193 6.3 Mitotic roles of Nups are not vertebrate specific To date, most of the mitotic roles of Nups discovered have been in the open mitosis of vertebrate cells. In fact, such dual roled Nups carry multiple functions at different locations once away from the pores during the open mitosis of metazoans. Recent evidence shows that in fungi that undergo either open and semi-open mitosis, Ustilago maydis and A. nidulans respectively, Nups locate to specific mitotic structures once released from the pores suggesting that the mitotic roles of Nups are not vertebrate specific (De Souza et al., 2009; Liu et al., 2009; Osmani et al., 2006a; Theisen et al., 2008). Because A. nidulans undergoes a semi-open mitosis, with many of its Nups dispersing from the pore and residing at specific mitotic structures, it is possible that the Nups of this fungus might also carry mitotic roles. We have shown in chapter 3 that Nup2 and a novel nucleoporin NupA have essential mitotic functions in A. nidulans. In chapter 5, we start to inquire whether the mitotic function of one of those Nups, Nup2, is conserved throughout evolution. In this work thus, we show that the function of Nups in coordinating mitotic events and the involvement of the transport pathway in mitosis are evolved as early as in some fungi that have elaborate developmental stages, and are thus functionally conserved over a wide span of evolution. 6.4 Lessons learned from Aspergillus nidulans Aspergillus nidulans, as many of the filamentous fungi, is a fast growing saprophytic organism, which can divide its nucleus in a very fast and efficient manner (within 5 minutes). Thus, the highly tuned regulation of mitotic events in this organism is crucial for proper mitotic progression as there isn t much time to correct mistakes. This is where proteins such as Nup2 and NupA, that help make mitosis more efficient and coordinate different mitotic events, seem to play a crucial role. One would think then, a simple mistake will lead to the death of this fungus. However, that is not the case. A. nidulans is very resilient to mistakes as it keeps growing and multiplying its nuclei even when things are not normal, as seen in Nup2 and NupA deleted cells. This is possible probably because Aspergillus nidulans carries multiple nuclei in one cytoplasm which 173

194 undergo syncytial mitosis. Hence defects that occur in mitosis of one nucleus will not lead to the lethality of the cell, as it can count on other nuclei for proper division, function and eventually growth. This has also been obsereved in the syncytial mitosis that occurs in Drosophila embryos (Sibon et al., 2000). So there is much to be learned from this fungus regarding how it regulates mitotic progression. In addition, understanding mitotic regulation will help us decipher the etiology and progression of human diseases such as cancer and birth defects. For example, it has long been known that cancer is a disease in which the regulation of the cell cycle progression had gone awry. How cancerous cells misregulate their mitosis is also not very well understood. Progress in understanding the regulation of mitotic events in organisms such as A. nidulans and others, and discovery of proteins that regulate mitotic progression might be central in finding cures to cancer in the future. 174

195 Figure 6.1 Model of the regulation, mechanism of translocation and the function of Nup2 at chromatin during Aspergillus nidulans mitosis. During interphase Nup2 is targeted to the NPC through its interaction with domain 4 of NupA. Upon the G2 - M transition, NIMA and CDK1 promote phosphorylation of Nup2 and NupA. Once phosphorylated, Nup2 moves to and associates with mitotic chromatin via NupA as early as prophase. At metaphase, the Nup2-NupA complex stays associated with condensed chromatin where we propose it facilitates the release of mitotic regulators (sush as SAFs) in the vicinity of chromatin from the importin / complex. The Nup2-NupA complex then moves from being on, to locating around the chromatin mass periphery at anaphase - telophase where it has a role in mitotic exit. The Nup2-NupA complex helps regeneration of transport competent G1 daughter nuclei by regulating either nuclear fission and/or chromatin organization and tethering to the nuclear basket. We propose that both roles might be linked to proper localizations and functions of importins α and during metaphase and mitotic exit. 175

196 Figure