Phosphoregulation of Sfi1 Is Required for Yeast Centrosome Duplication and Separation

Phosphoregulation of Sfi1 is Required for Yeast Centrosome Duplication and Separation

Jennifer S. Avena

B.A., Ohio Wesleyan University, 2005

A dissertation submitted to the Faculty of the Graduate School of the University of Colorado in

partial fulfillment of the requirements for the degree of Doctor of Philosophy

Department of Molecular, Cellular, and Developmental Biology

2014

This thesis entitled: Phosphoregulation of Sfi1 is Required for Yeast Centrosome Duplication and Separation written by Jennifer Susan Avena has been approved for the Department of Molecular, Cellular, and Developmental Biology

Tin Tin Su, Ph.D., Committee Chair

Gia Voeltz, Ph.D., Committee Member

Date

The final copy of this thesis has been examined by the signatories, and we Find that both the content and the form meet acceptable presentation standards Of scholarly work in the above mentioned discipline. iii

Abstract

Avena, Jennifer Susan (Ph.D., Molecular, Cellular, and Developmental Biology)

Phosphoregulation of Sfi1 is Required for Yeast Centrosome Duplication and Separation

Thesis directed by Professor Mark Winey

Centrosomes serve as the main microtubule-organizing centers in many eukaryotic cells.

Duplication of centrosomes once per cell cycle is essential for bipolar spindle formation and genome maintenance and requires control by protein kinases, including cyclin-dependent kinases

(Cdks). However, an understanding of the mechanisms by which centrosome duplication is regulated and, in particular, restricted to once per cell cycle is still lacking. Sfi1, a conserved component of centrosomes, is a phospho-protein that is a key candidate for establishing the site of new centrosome assembly in the budding yeast Saccharomyces cerevisiae, likely via domain- specific functions. I have examined the role of phosphorylation of the Sfi1 amino (N) and carboxy (C) termini at the spindle pole body (SPB), the budding yeast centrosome. I have established that a nonphosphorylatable sfi1 N-terminal allele shows genetic interactions with protein-encoding genes of the new SPB precursor. This suggests that these proteins are promising candidates as physical interactors of the Sfi1 N terminus and may be recruited by the

Sfi1 N terminus for assembly of the new SPB during duplication. I additionally propose that this step of new SPB assembly requires the protein kinase Mps1. Importantly, I also have identified

Sfi1 as the first Cdk substrate required to restrict centrosome duplication to once per cell cycle. I found that reducing Cdk1 phosphorylation by changing Sfi1 C-terminal phosphorylation sites to nonphosphorylatable residues leads to defects in separation of duplicated SPBs and to inappropriate SPB reduplication during mitosis. These cells also display defects in bipolar iv spindle assembly, chromosome segregation, and growth. My findings lead to a model whereby phosphoregulation of Sfi1 by Cdk1 has the dual function of promoting SPB separation for spindle formation and preventing premature SPB duplication. In addition, I provide evidence that the protein phosphatase Cdc14 has the converse role of activating licensing, likely via dephosphorylation of Sfi1. This work could provide crucial insight into the initiation of centrosome duplication.

Dedication

I would like to dedicate this work to my husband for his unending support and to my parents and siblings for their support and motivation throughout the years. vi

Acknowledgements

I thank my advisor, Mark Winey for invaluable guidance, feedback, and advice. I thank

Shelly Jones and the Winey lab for helpful feedback and discussion. I thank Alex Stemm-Wolf for assistance with experimental design of the mps1 and domain overexpression work. I thank

Tom Giddings Jr. for immuno-EM imaging, EM sample preparation and feedback, Christina

Clarissa, Courtney Ozzello, and Janet Meehl Fox for assisting with EM, and Eileen O’Toole for

EM feedback and modeling. I thank Sue Jaspersen for helpful feedback and advice, SIM assistance, reagants, and equipment. I thank Shannon Burns and Zulin Yu for SIM assistance and imaging. I thank Brian Slaughter and Jay Unruh for analysis advice on SIM and assistance with live timelapse imaging. I thank Jennifer Gardner and Christine Smoyer for assistance with live timelapse imaging and analysis. I thank the Stowers Institute Molecular Biology facility for mutant alleles. I thank William Old and Christopher Ebmeier for assistance with mass spectrometry. I thank Zach Wilson, Jaimee Hoefert, Melissa Phillips, and Tara Peters for their contributions to data acquisition and analyses in the domain overexpression, dosage suppressor screen, and yeast two-hybrid projects, respectively. I thank the Blumenthal, Cech, Copley,

Detweiler, Espinosa, Han, Klymkowsky, Leinwand, Odorizzi, Pace, Xue, and Yi laboratories for use of reagants and/or equipment. I thank John Kilmartin, Orna Cohen-Fix, Fred Cross, Ivan

Rayment, Kevin Hardwick, Greg Odorizzi, Stanley Fields, and the Yeast Resource Center for sharing strains, plasmids and/or reagants.

This work was supported by NIH 5 T32 GM007135 (Jennifer Avena), NIH R01

GM51312 (Mark Winey), the Stowers Institute for Medical Research (Sue Jaspersen), and

University of Colorado BURST and UROP grants (Tara Peters). vii

Contents

Chapter One: Introduction

I. Centrosome structure, function, and duplication 1

II. Yeast centrosome, spindle pole body (SPB), structure, function, and duplication 3

III. Cell cycle regulation of centrosome duplication 7

IV. Sfi1 10

V. Model for phosphoregulation of Sfi1 involved in licensing of SPB duplication 11

VI. Synopsis of results 13

Chapter Two: A nonphosphorylatable sfi1 N-terminal allele shows genetic interactions with genes encoding satellite components

I. Introduction 15

II. Results 16

III. Discussion 25

IV. Materials and Methods 29

Chapter Three: Cdk1 phosphorylation of the Sfi1 C terminus is required for appropriate growth, bipolar spindle assembly, and chromosome segregation

I. Introduction 36

II. Results 37

III. Discussion 87

IV. Materials and Methods 95 viii

Chapter Four: Licensing of SPB duplication requires phosphoregulation of Sfi1

I. Introduction 110

II. Results 111

III. Discussion 128

IV. Materials and Methods 131

Chapter Five: Conclusions and future directions

I. Summary 139

II. A potential role for combinatorial control of phosphorylation in satellite formation 139

III. Mechanism for licensing of SPB duplication 141

IV. Analogies between the licensing events of SPB duplication and DNA replication 142

V. Model for conservation of Cdk regulation of Sfi1 in licensing of centrosome

Duplication 145

Appendices

A. Analysis of Sfi1 yeast two-hybrid interactions 148

B. Identification of SFI1 genetic interactions via a dosage suppressor screen 154

C. Sfi1 domain overexpression and localization 164

References 176

Tables

2-1 Yeast strains 33

2-2 Plasmids 34

2-3 Primers 35

3-1 Growth phenotype summary of nonphosphorylatable sfi1 alleles with alteration of Cdk sites within the Sfi1 C terminus 45

3-2 Phenotype summary of sfi1 alleles 47

3-3 Ploidy determination for sfi1 alleles via genetic crosses 78

3-4 sfi1-C3A SPB measurements via EM 86

3-5 sfi1-C3A SPB measurements via EM 86

3-6 Yeast strains 103

3-7 Plasmids 107

3-8 Primers 109

4-1 Yeast strains 137

4-2 Plasmids 138

B-1 Genomic region of candidate 2-µm suppressors 161

B-2 Yeast strains 162

B-3 Plasmids 163

C-2 Yeast strains 174

C-3 Plasmids 175

Figures

1-1 Centrosome and SPB structure 2

1-2 SPB composition 4

1-3 SPB duplication pathway 5

1-4 Model for Sfi1 in SPB duplication 12

2-1 Validation of GFP-SFI1-TRP1 sfi1Δ::HIS5 mps1-1 17

2-2 The protein kinase Mps1 is required for satellite formation 19

2-3 The Sfi1 N terminus is essential for cell viability 20

2-4 Control cells are phenotypically similar to wild-type cells 21

2-5 An sfi1 nonphosphorylatable N terminus allele shows normal spindle morphology 22

2-6 An sfi1 nonphosphorylatable N terminus allele shows genetic interactions with the satellite protein-encoding genes SPC42 and SPC29 24

2-7 An sfi1 nonphosphorylatable N terminus allele shows genetic interactions with the satellite protein-encoding genes CDC31 26

3-1 Clb2/Cdk1 phosphorylates Sfi1 at sites within Cdk motifs in vitro 39

3-2 Growth phenotypes of sfi1 C terminus mutant alleles with alteration of non-Cdk sites 41

3-3 Growth phenotypes of sfi1 C terminus mutant alleles with alteration of Cdk sites 42

3-4 Growth phenotypes of sfi1 C terminus mutant alleles with alteration of Cdk sites 43

3-5 Growth phenotypes of sfi1 C terminus mutant alleles with alteration of Cdk sites 46

3-6 Growth phenotypes of sfi1 alleles with alteration of S801 49

3-7 sfi1-S939A shows no growth defects 51

3-8 Validation for a control strain with integration at the SFI1 locus 52 xi

3-9 Multiple sfi1 alleles are dominant 53

3-10 sfi1-S855A displays spindle and chromosome segregation defects 58

3-11 sfi1-C2A displays spindle and chromosome segregation defects 59

3-12 sfi1-C3A spontaneously diploidizes 61

3-13 sfi1-C3A displays defects in viability and ploidy 62

3-14 sfi1-C3A displays spindle and chromosome segregation defects 63

3-15 sfi1-C4A displays defects in viability and ploidy 65

3-16 sfi1-C4A displays spindle and chromosome segregation defects 67

3-17 sfi1-C3A+S923A displays spindle and chromosome segregation defects 69

3-18 sfi1-C4A+S923A displays defects in ploidy and viability 70

3-19 sfi1-C4A+S923A displays spindle and chromosome segregation defects 71

3-20 sfi1-C3E displays spindle and chromosome segregation defects 72

3-21 sfi1-C4E displays defects in ploidy and viability 74

3-22 sfi1-C4E displays spindle and chromosome segregation defects 75

3-23 sfi1-C4E shows multiple spindle morphology phenotypes via EM 76

3-24 sfi1-C4E+S923E displays spindle and chromosome segregation defects 77

3-25 sfi1-C4A displays extra SPBs 79

3-26 sfi1-C3A displays unseparated SPBs or bipolar spindles with enlarged or reduplicated

SPBs 82

3-27 sfi1-C4A displays unseparated SPBs or bipolar spindles with enlarged or reduplicated

SPBs 84

4-1 sfi1-C4A displays multiple phenotypes upon release from G1 arrest that differ from esp1-1 112 xii

4-2 In a single cell cycle, sfi1-C4A displays multiple phenotypes that differ from esp1-1 114

4-3 sfi1-C3A is able to form a satellite at G1 phase 118

4-4 sfi1-C4A and sfi1-C3A cells arrest in mitotis upon induced overexpression of nondegradable Pds1 121

4-5 Reduplicated SPBs are enhanced in sfi1-C4A GAL1-pds1-mdb upon mitotic arrest 122

4-6 Reduplicated SPBs are present in sfi1-C4A GAL1-pds1-mdb upon mitotic arrest 123

4-7 Extra SPBs are present in sfi1-C3A GAL1-pds1-mdb upon mitotic arrest 125

4-8 Reduplicated SPBs result from cdc14 release from the nucleolus 126

4-9 Model for licensing of yeast centrosome duplication via phosphoregulation of

Sfi1 132

5-1 Model for Sfi1 in SPB duplication 140

5-2 Analogy between the licensing events of SPB duplication and DNA replication 143

5-3 Model for a conserved mechanism of licensing of centrosome duplication 147

B-1 Candidate dosage suppressors of sfi1-3A and sfi1-C3A+S923A 159

C-1 Growth phenotypes resulting from overexpression of individual Sfi1 domains 170

C-2 GFP-tagged Sfi1 domains are expressed 171

C-3 DNA content of cells overexpressing GFP-tagged Sfi1 domains 172

C-4 Localization of Sfi1 domains 173 1

Chapter One: Introduction

I. Centrosome structure, function, and duplication

Centrosomes are the main microtubule-organizing centers in many eukaryotes and they anchor and nucleate microtubules and duplicate to serve as the poles of the mitotic spindle

(Bettencourt-Dias and Glover, 2007). Centrosome duplication only once per cell cycle is critical for bipolar spindle organization and proper chromosome segregation. While not all eukaryotic cells require centrosomes for spindle formation (reviewed in (Compton, 2000)), when present, centrosome function and number are integral to appropriate spindle formation. Indeed, aberrant centrosome numbers are linked to chromosomal instability and are commonly observed in cancers (Ganem et al., 2009); thus, mechanisms that restrict duplication to a single event are essential.

Human centrosomes consist of two orthogonally-oriented centrioles composed of triplet microtubules and the surrounding pericentriolar matrix (PCM), which contains many proteins including the microtubule-nucleating –tubulin ring complex (-TuRC) (Fig. 1-1A-B)

(Bettencourt-Dias and Glover, 2007; Moritz et al., 2000). At the beginning of the cell cycle in

G1, cells contain a single centrosome with two centrioles tethered to each other via a fibrous link. At S phase, centriole duplication occurs, in which a new daughter centriole (procentriole) assembles orthogonally to each of the two mother centrioles. This tight orthogonal association is termed “engagement.” The daughter centrioles will elongate, and in G2, centrosome maturation occurs, in which increased recruitment of the -TuRC occurs, thus increasing the centrosome’s microtubule nucleation capabilities. At the end of G2, centrosome separation occurs, in which the tether connecting the two mother centrioles is severed. The centrosomes, each containing a mother and daughter centriole, can then form the bipolar spindle in mitosis. At mitotic exit, the 2

Figure 1-1. Centrosome and SPB structure A-B. Electron microscopy (EM) image of a human centrosome, which consist of two orthogonally-oriented centrioles (centriole and procentriole) (A) composed of nine (1-9) triplet microtubules (A-, B-, and C-microtubules), as shown in cross-section (B) at a region similar to that of the dashed line in A. Adapted from (Gönczy, 2012). Images obtained from (Paintrand et al., 1992). C. EM image of SPB with elongated half-bridge (HB) containing a satellite (S). C: cytoplasm. N: nucleus. cMT: cytoplasmic microtubules. nMTs: nuclear microtubules. Adapted from (Jaspersen and Winey, 2004). Images obtained from (Giddings et al., 2001). Bar: 100 nm.

3 mother and daughter centriole will then disengage, losing their tight orthogonal association, in preparation for centrosome duplication in the next cell cycle (Moritz et al., 2000; Azimzadeh and

Bornens, 2007; Bettencourt-Dias and Glover, 2007; Nigg, 2007; Lim et al., 2009).

II. Yeast centrosome, spindle pole body (SPB), structure, function, and duplication

In the budding yeast Saccharomyces cerevisiae, the spindle pole body (SPB) is the centrosome-equivalent. The SPB is embedded within the nuclear envelope and nucleates both nuclear and cytoplasmic microtubules. The SPB is composed of a central core, the microtubule- nucleating –tubulin complex, linkers connecting the central core to the –tubulin complex, membrane anchors, and a half-bridge, a specialized region of the nuclear envelope on only one side of the SPB (Figs. 1-1C and 1-2) (Jaspersen and Winey, 2004; Keck et al., 2011). All of the components of the SPB have been identified, with fifteen of the eighteen components being essential (Jaspersen and Winey, 2004; Araki et al., 2006), and ten of these proteins have human homologs or homologous domains (Keck et al., 2011).

While the SPB is structurally distinct from that of higher vertebrates, it is a valuable model in which to study centrosome duplication because it is well-characterized, and many structural components and protein kinase regulators are conserved in vertebrates. SPB duplication begins during the G1 phase of the cell cycle when the half-bridge elongates and a density called the satellite forms at the cytoplasmic distal tip. The satellite is composed of four core SPB proteins (Spc42, Spc29, Cnm67, and Nud1) and develops into the new mature SPB, which is linked to the mother SPB via a complete bridge. After duplication is complete, the bridge is severed, and the SPBs separate and move to opposite sides of the nucleus as they form the bipolar spindle (Fig. 1-3) (Jaspersen and Winey, 2004). 4

Figure 1-2. SPB composition SPB schematic. NE: nuclear envelope. cMT: cytoplasmic microtubules. nMT: nuclear microtubules. ■: satellite.

Figure 1-3. SPB duplication pathway A. SPB duplication pathway. (1) Half-bridge elongation, (2) satellite formation, (3) new SPB expansion, full bridge formation, and SPB insertion into the nuclear envelope (NE), (4) SPB separation and bipolar spindle formation, (5) licensing of SPB duplication. cMT: cytoplasmic microtubules. nMT: nuclear microtubules. ■: satellite. B. SPB duplication pathway with respect to the cell cycle and budding morphology in budding yeast. Duplication begins in unbudded cells in G1. Duplicated centrosomes are present by the time bud formation begins. They will then separate and move to opposite sides of the nucleus to serve as the poles in the mitotic spindle in large-budded cells. 6

Requirements for protein components at each step of the SPB duplication pathway are known (Jaspersen and Winey, 2004). However, an understanding of the mechanisms by which these proteins function is still generally lacking. For example, previous work has indicated that the half-bridge proteins, Sfi1, Cdc31 (centrin), Kar1, and Mps3, are required prior to satellite formation, as mutations in these proteins lead to cells with single SPBs lacking a half-bridge

(Byers 1981) (Rose and Fink, 1987; Winey et al., 1991; Jaspersen et al., 2002; Kilmartin, 2003).

However, it is not clear whether these proteins are required for maintenance of the half-bridge, half-bridge elongation, and/or satellite formation. Additionally, two cdc31 mutants (cdc31-5 and

CDC31-16) show an uncommon phenotype of an elongated half-bridge lacking a satellite (M.

Winey, unpublished data; Vallen et al., 1994). This suggests that Cdc31 may be required for satellite formation, but further examination of this proposed function is needed.

In addition to a requirement for SPB components in SPB duplication, it is also known that multiple protein kinases are required for SPB duplication. The kinase Mps1 is required both before and after satellite formation and is known to phosphorylate several SPB components, including the half-bridge component Cdc31 and the satellite proteins, Spc42 and Spc29 (Winey et al., 1991; Castillo et al., 2002; Araki et al., 2006; Holinger et al., 2009). Cyclin-dependent kinase 1 (Cdk1; Cdc28) is also known to promote SPB duplication (Haase et al., 2001; Jaspersen et al., 2004; Byers and Goetsch, 1974), and many SPB components are known to be Cdk1 substrates in vitro and/or in vivo, including the half-bridge protein, Sfi1, and the four satellite proteins (reviewed in Keck et al., 2011).

Once SPB duplication is complete, side-by-side SPBs are present, in which the mother

SPB and new SPB are connected via a bridge. Separation of the two SPBs at S phase is known 7 to require the half-bridge protein Sfi1 (Anderson et al., 2007; Elserafy et al., 2014) and cyclin/Cdk1 (Fitch et al., 1992; Lim et al., 1996; Haase et al., 2001).

III. Cell cycle regulation of centrosome duplication

Cyclin-dependent (Cdk), separase, and polo-like kinase 1 (Haase et al., 2001; Vidwans et al., 2003; Tsou and Stearns, 2006; Tsou et al., 2009) are known regulators of centrosome duplication implicated in limiting this process to once per cell cycle. However, an understanding of the mechanisms through which they ensure that duplication is tightly coupled with other cell cycle events is still lacking.

In order to be active, Cdks bind cyclin proteins, and the targets and functions of Cdk change throughout the cell cycle depending on which cyclin it binds since different cyclins are present at different cell cycle stages (i.e., G1, G1/S, S, and mitotic cyclins). Cdk activity is required for centrosome duplication, DNA replication and mitotic spindle assembly at different times during the cell cycle. Importantly, not only does the specificity of Cdk change throughout the cycle, but its level of activity also changes based on the presence or absence of cyclins. Cdk activity is low at the start of the cell cycle at G1, but rises later in G1 and remains high until mid- mitosis, at which point Cdk activity drops to allow for mitotic exit and progression to the next cell cycle (Morgan, 1997; Morgan, 2006; Enserink and Kolodner, 2010).

The oscillation in Cdk activity level is key to ensuring DNA replication occurs only once per cell cycle in a conserved process. When Cdk activity is low at early G1, proteins can assemble onto the DNA to form the pre-replication complex (pre-RC), which is composed of the origin of recognition complex (ORC), Cdc6, Cdt1, and the Mcm2-7 helicase complex. Pre-RC assembly makes the DNA competent for replication to proceed, i.e., licensing DNA replication 8

(origin licensing). Licensing only occurs when cells have low Cdk activity and its targets are therefore not phosphorylated (Nguyen et al., 2001; Kearsey and Cotterill, 2003; Mimura et al.,

2004; Liku et al., 2005; Morgan, 2006; Sclafani and Holzen, 2007; Enserink and Kolodner,

2010). Following pre-RC formation, both Cdk and DDK (Dbf4-dependent kinase) are required for formation of the pre-initiation complex (pre-IC) at the DNA’s origin of replication. The pre-

IC is a multi-protein complex that recruits DNA polymerase to the DNA, and activation of the

Mcm2-7 helicase complex follows to unwind the DNA for subsequent DNA synthesis (reviewed in Morgan, 2006; Sclafani and Holzen, 2007; Enserink and Kolodner, 2010). Cdk ensures that another round of licensing is blocked by phosphorylating substrates (ORC, Cdc6, Mcm2-7 complex) required for licensing. This enables DNA replication, but not re-replication, meaning a second round of licensing, to proceed. Thus not until Cdk activity is low again at the transition to the next cell cycle can the proteins required for licensing be dephosphorylated and assemble to once again make cells competent for DNA replication (Elsasser et al., 1999; Nguyen et al., 2000;

Nguyen et al., 2001; Morgan, 2006; Sclafani and Holzen, 2007; Makise et al., 2009; Enserink and Kolodner, 2010).

Previous work has shown that G1/S cyclin/Cdk is required for centrosome duplication in vertebrates and yeast (cyclin E/Cdk2 in Xenopus; cyclin A/Cdk2 in mammalian cells; Cdk1 in budding yeast ) (Byers and Goetsch, 1974; Hinchcliffe et al., 1999; Lacey et al., 1999;

Matsumoto et al., 1999; Meraldi et al., 1999; Jaspersen et al., 2004). Additionally, Cdk is proposed to play a role in licensing of centrosome duplication (i.e., rendering the centrosome competent for duplication; Drosophila melanogaster, budding yeast) (Haase et al., 2001;

Vidwans et al., 2003). However, while it is known that Cdk1 is required for centrosome duplication and Cdk is implicated in licensing, no protein targets of Cdk required for licensing of 9 centrosome duplication have been clearly identified, although some previous work has suggested a role for nucleophosmin (Okuda et al., 2000; Wang et al., 2005; Tsou and Stearns, 2006) and

Sfi1 (Elserafy et al., 2014). Similar to licensing of DNA replication, licensing in centrosome duplication should render the centrosome competent for duplication and should occur only once at a discrete point within the cell cycle. Centrosome duplication should thus occur only once per cell cycle due to a block to centrosome reduplication at all remaining points throughout the cell cycle (Tsou and Stearns, 2006; Nigg, 2007).

In budding yeast, G1 cyclins (Cln1, 2, 3) promote SPB duplication, and S and mitotic cyclins promote SPB separation. Importantly, deletion of all mitotic cyclins (Clb1, 2, 3, and 4) leads to reduplication of spindle pole bodies (SPBs, the yeast centrosome), indicating that mitotic cyclin/Cdk1 blocks SPB reduplication. Thus a decrease in Cdk1 activity levels is needed to eliminate the block to SPB duplication, allowing SPBs to become licensed, or competent, for duplication (Haase et al., 2001). However, even in budding yeast, in which mapping of Cdk1 substrates has been studied at a genome-wide level (Ubersax et al., 2003), no targets of Cdk1 that restrict licensing of SPB duplication have been identified.

Additional work has shown that SPB reduplication occurs only upon separation of duplicated SPBs (Simmons Kovacs et al., 2008), providing evidence for a centrosome-intrinsic

(in comparison to Cdk-regulated) mode to ensure duplication occurs only once per cell cycle. Specifically, the presence of side-by-side SPBs with a full bridge appears to not be permissive for SPB reduplication, while SPB separation, resulting in SPBs with short half- bridges, is permissive for reduplication. Similarly, a centrosome-intrinsic mode has been identified in human cells, in which centriole disengagement, which is regulated by separase and 10

Plk1, is required for licensing of centrosome duplication (Wong and Stearns, 2003; Tsou et al.,

2009).

IV. Sfi1

Sfi1 is a conserved component of centrosomes and SPBs that is required for SPB duplication in both budding and fission yeast and is proposed to be required for centriole formation in humans (Ma et al., 1999, 1; Kilmartin, 2003; Balestra et al., 2013; Elserafy et al.,

2014; Lee et al., 2014). Sfi1 has three domains: an N-terminal domain (185 residues in S. cerevisiae), an elongated alpha-helical, conserved repeat domain (616 residues; 21 repeats in S. cerevisiae) in which each repeat binds one molecule of centrin (Cdc31), and a C-terminal domain (145 residues in S. cerevisiae). Sfi1 is known to be phosphorylated in vivo at 19 residues, 11 within the N terminus and 8 within the C terminus (Chi et al., 2007; Keck et al.,

2011). Interestingly, four of these residues at the C terminus are within full Cdk consensus motifs (S/T*-P-x-K/R), making these sites excellent targets for Cdk1 phosphorylation.

Additionally, three residues, two at the N terminus and one at the C terminus, are within minimal

Cdk consensus motifs (S/T*-P). Full-length Sfi1 is a substrate of Cdk1 in vitro (Ubersax et al.,

2003), and recent work has found that Sfi1 is phosphorylated by Cdk1 in vitro specifically at the four C-terminal residues within full Cdk consensus motifs (Elserafy et al., 2014). Furthermore,

Sfi1 is dephosphorylated by Cdc14 (Bloom et al., 2011; Elserafy et al., 2014), which counteracts

Cdk1 activity (Visintin et al., 1998; Stegmeier and Amon, 2004). While it is not known whether

Sfi1 is a substrate of Mps1, it is known that Mps1 associates in a complex with Sfi1 (Breitkreutz et al., 2010). 11

Sfi1 is positioned on the cytoplasmic side of the half-bridge in an orientation-specific manner with the N terminus proximal and the C terminus distal to the SPB (Li et al., 2006).

Based on this topology, Kilmartin and colleagues put forth a model in which Sfi1 is a central player in both SPB duplication and separation (Li et al., 2006). They proposed that SPB duplication initiates when Sfi1 molecules associate via C-terminal end-to-end interactions with

Sfi1 molecules already present at the half-bridge, thus doubling the half-bridge length. A new

SPB can then assemble at the free SPB-distal N-terminal domain created by this arrangement, in which the N terminus recruits components of the satellite. After completion of SPB duplication, dissociation of the Sfi1 C termini at the bridge from each other would then allow SPB separation to occur (Fig. 1-4).

As previously discussed, Sfi1 has been shown to be required prior to satellite formation, as mutations in the repeat domain lead to cells arrested with only one SPB (Kilmartin, 2003).

Additionally, the Sfi1 C terminus is required for SPB separation (Anderson et al., 2007; Elserafy et al., 2014; Strawn and True, 2006). Additional research suggests that the repeat domain and C terminus of Sfi1 are required for SPB duplication after satellite formation (Kilmartin, 2003;

Elserafy et al., 2014). Interestingly, Cdk1 phosphorylation of the Sfi1 C terminus is specifically required for SPB separation, as nonphosphorylatable mutations within Cdk1 sites in this domain lead to cells arrested with side-by-side SPBs (Anderson et al., 2007; Elserafy et al., 2014).

V. Model for phosphoregulation of Sfi1 involved in licensing of SPB duplication

Based on the model by Kilmartin and colleagues that SPB duplication initiates when Sfi1 molecules associate via C-terminal end-to-end interactions with Sfi1 molecules already present at the half-bridge, it has been proposed that Sfi1 may serve as a licensing factor for SPB 12

Figure 1-4. Model for Sfi1 in SPB duplication Proposed model for Sfi1 in SPB duplication. SPB duplication initiates when Sfi1 molecules associate via C-terminal end-to-end interactions with Sfi1 molecules already present at the half- bridge, thus doubling the half-bridge length. The free SPB-distal Sfi1 N terminus then recruits components of the satellite. After completion of SPB duplication, dissociation of the Sfi1 C termini at the bridge from each other allows SPB separation to occur. NE: nuclear envelope. cMT: cytoplasmic microtubules. nMT: nuclear microtubules. ■: satellite. 13 duplication (Li et al., 2006; Jones and Winey, 2006; Bloom et al., 2011). Since the Sfi1 C terminus is phosphorylated by Cdk1 (Elserafy et al., 2014), and Sfi1 is dephosphorylated by

Cdc14 (Bloom et al., 2011; Elserafy et al., 2014), which is required for mitotic exit (Visintin et al., 1998; Stegmeier and Amon, 2004), it has been proposed by myself and others (Elserafy et al.,

2014) that Cdk1 and Cdc14 phosphoregulation of Sfi1 contribute to licensing of SPB duplication whereby mitotic Cdk1 phosphorylation of Sfi1 blocks and Cdc14 dephosphorylation enables

SPB duplication. A recent study showed that cells with Sfi1 containing phosphomimetic Cdk1 sites arrest with a single SPB, supporting the idea that phosphorylation of Sfi1 inhibits the process of duplication (Elserafy et al., 2014). However, no previous studies of Cdk1 targets have mimicked the reduplication phenotype revealed by depletion of Cdk1 activity (Haase et al.,

2001).

VI. Synopsis of results

In the current work, I explore the model that phosphoregulation of the Sfi1 C terminus is required for both the licensing of SPB duplication as well as for SPB separation and that the Sfi1

N terminus recruits components of the new SPB. I report that a nonphosphorylatable sfi1 N- terminal allele shows genetic interactions with protein-encoding genes of the satellite, indicating that these proteins are promising candidates as physical interactors of the Sfi1 N terminus and may be recruited by the Sfi1 N terminus for satellite formation. I additionally propose that the

SPB duplication step of satellite formation requires the protein kinase Mps1, possibly via regulation of the Sfi1 N terminus. Upon examination of the Sfi1 C terminus, I found that reducing Cdk1 phosphorylation by changing Sfi1 C-terminal phosphorylation sites to nonphosphorylatable residues leads to defects in separation of duplicated SPBs and to 14 inappropriate SPB reduplication during mitosis. These findings lead to a model whereby phosphoregulation of Sfi1 by Cdk1 has the dual function of promoting SPB separation for spindle formation and preventing premature SPB duplication. In addition, I propose that the protein phosphatase Cdc14 has the converse role of activating licensing, likely via dephosphorylation of Sfi1.

Chapter Two: A nonphosphorylatable sfi1 N-terminal allele shows genetic interactions with genes encoding satellite components

I. Introduction

The initial steps in SPB duplication at G1 phase are half-bridge elongation and satellite formation. Prior to SPB duplication initiation, the SPB contains a short half-bridge. Previous work showed that one mps1 mutant allele, mps1-8, terminates with a short half-bridge (Castillo et al., 2002), suggesting that Mps1 is required for half-bridge elongation. Given the proposed role of the Sfi1 C terminus in half-bridge elongation, Mps1 may regulate the Sfi1 C terminus in this step. However, whether the elongated half-bridge lacking a satellite is a true intermediate has been unknown. The only known previous work that identified an elongated half-bridge lacking a satellite is work with the mps1 mutant, mps1-1 (Winey et al., 1991), and the cdc31 mutants, cdc31-5 (M. Winey, unpublished data) and CDC31-16 (Vallen et al., 1994). I was therefore interested in determining whether the elongated half-bridge is a true intermediate or an aberrant mutant-specific structure by examining Sfi1 orientation. If it is a true intermediate, I would expect Sfi1 to be oriented properly, with the Sfi1 C termini at the center of the elongated half-bridge and the N termini proximal to the mother SPB and at the distal tip of the elongated half-bridge (Li et al., 2006).

There is still a paucity of information regarding the process of assembly of the satellite, the new SPB precursor, at the cytoplasmic distal tip of the elongated half-bridge. It has been shown by immuno-electron microscopy (EM) that the satellite is composed of Spc42, Spc29,

Cnm67, and Nud1, which are all protein components of the core SPB (Adams and Kilmartin,

1999). Given the localization of the Sfi1 N terminus at the cytoplasmic distal tip of the elongated half-bridge, the location at which the satellite assembles, it seems plausible that the N terminus may recruit satellite components, as proposed by Kilmartin and colleagues (Li et al., 16

2006). However, there are currently no published reports of genetic or physical interactions between Sfi1 and the satellite proteins. Furthermore, while genetic interactions between genes encoding other half-bridge protein components (Cdc31, Kar1, Mps3) and genes encoding satellite proteins have been identified (Elliott et al., 1999; Pereira et al., 1999; Jaspersen et al.,

2002), no direct physical interactions between half-bridge and satellite proteins have been seen.

Therefore, further examination of the mechanism of satellite recruitment, including protein interactions, order of satellite proteins assembled, and regulation of assembly, is needed.

II. Results

The elongated half-bridge is a true intermediate, and Mps1 is required for the subsequent step of satellite formation

To identify proteins required for satellite formation, I examined an mps1 mutant with an elongated half-bridge lacking a satellite. I investigated whether the elongated half-bridge was a true intermediate or an aberrant mutant-specific structure by examining Sfi1 orientation.

Specifically, I looked at localization of the Sfi1 N terminus at the elongated half-bridge. To do this, I utilized a strain containing the mps1-1 mutation with Sfi1 tagged at its N terminus with

GFP (trp1::GFP-SFI1-TRP1 sfi1Δ::HIS5 mps1-1). I first validated that this strain behaved as does mps1-1 via analysis of budding and DNA content (Fig.2-1). As expected, at 37°C, GFP-

SFI1 mps1-1 cells display defects in chromosome segregation, which occurs in the first cell cycle following release from G1 arrest (Winey et al., 1991). Additionally, both strains are lethal at

37°C (see Fig. 2-1 legend) (Winey et al., 1991).

Using immuno-EM, I found that the mps1-1 mutant contains Sfi1 molecules positioned in the correct end-to-end orientation along the elongated half bridge, with the N terminus proximal 17

Figure 2-1. Validation of GFP-SFI1-TRP1 sfi1Δ::HIS5 mps1-1 GFP-SFI1-mps1-1 and mps1-1 cells were grown to early-log phase (asynchronous) in YPD at 25°C and treated with the mating pheromone α-factor to arrest cells in G1 (3.5 h and 3 h, respectively). Cells were then released into YPD at 37°C and examined at multiple timepoints after release: 1.5 h (GFP-SFI1-mps1-1) or 45 minutes (mps1-1), 2 h, 3 h, and 4 h post-release. DNA content via flow cytometry and budding analysis are shown. UB: unbudded. SB: small- budded. LB: large-budded. Note: each strain was examined in an independent experiment. Following the final timepoint, cells at 37°C were plated and grown on YPD plates for 3 days at 25°C. 0.2% oGFP-SFI1-mps1-1 cells from 37°C were viable compared to the same strain grown at 25°C (100%). 0.3% of mps1-1 cells from 37°C were viable compared to the same strain grown at 25°C (100%).

18 to the mother SPB and at the distal tip of the elongated half-bridge (Fig. 2-2). This clarifies the current model of SPB duplication, indicating that the elongated half-bridge is a true intermediate before satellite formation and that Mps1 is required for satellite formation.

The Sfi1 N terminus is essential for cell viability

To determine whether the N terminus is essential, I integrated an SFI1 N terminus deletion construct into the endogenous SFI1 locus. This sfi1-∆N strain is inviable, indicating that the N terminus is essential (Fig. 2-3). I was able to show this via two different methods. In one method, I created a heterozygous diploid (JA107) and found that in 97% (n=33) of tetrads, spores segregated 2:2 for viable:inviable. In the second method, I utilized a mutant or wild-type version of SFI1 with a rescuing SFI1 plasmid. Loss of the plasmid on 5-fluoroorotic acid (5-

FOA) resulted in lethality for sfi1-∆N, but SFI1 grew normally (Fig. 2-3).

An sfi1 nonphosphorylatable N terminus allele shows genetic interactions with mutant alleles of SPC42 and SPC29

To determine the significance of Sfi1 N terminus phosphorylation, I constructed a mutant allele of Sfi1 via gene synthesis. All 11 N terminus in vivo phosphorylation sites were converted to nonphosphorylatable residues (S/T to A, Y to F) to create the allele sfi1-NnonP. I first validated a control SFI1 strain (see materials and methods) to ensure that integration of an SFI1 plasmid did not have an effect on cell phenotype, as shown in Fig. 2-4. The sfi1-NnonP allele results in normal growth, budding profile (16±3% large-budded cells versus 20±5% for control at

30°C), correct DNA content throughout the cell cycle, and proper bipolar spindle formation (Fig.

2-5, 2-6, and 2-7). The growth phenotype of sfi1-NnonP was confirmed in yeast strains with the 19

Figure 2-2. The protein kinase Mps1 is required for satellite formation A, B. GFP-SFI1-mps1-1 cells were grown to log phase and shifted to restrictive temperature (37°C 4 h) and then prepared for immuno-EM with nanogold secondary label. Eight cells with SPBs were identified: five with labeling at the SPB and three with both distal and proximal labeling. N terminus labeling of two separate cells at the SPB (A; 5/19 gold dots) and the distal tip of the elongated half-bridge (B; 10/19 gold dots). Note: some cells also contained labeling closer to the center of the half-bridge (A; 4/19 gold dots in two cells). +: label representation. C. New model of SPB duplication. Half-bridge elongation and satellite formation are distinct steps. NE: nuclear envelope. cMT: cytoplasmic microtubules. nMT: nuclear microtubules. ■: satellite.

Figure 2-3. The Sfi1 N terminus is essential for cell viability Strains with wild-type SFI1 or the N terminus deletion, sfi1-∆N grown at 24°C. sfi1-∆N, is lethal when a rescuing autonomously-replicating SFI1 plasmid (YCp-SFI1-URA3) is lost at permissive temperature on 5-fluoroorotic acid (5-FOA). 5-FOA selects against strains containing URA3. Therefore, only strains which can survive without the rescuing SFI1 plasmid will grow on 5- FOA.

Figure 2-4. Control cells are phenotypically similar to wild-type cells sfi1∆::SFI1-TRP1-KANMX MATa cells containing SFI1 wild-type (JA50) or control (JA98). A. Dilution series on YPD. B. Cells grown to early-log phase at 24°C then shifted to 37°C for 4 h. DNA content via flow cytometry at 24°C and 30°C. Note: % large-budded cells at 37°C for wild-type (17%) and control (21%) similar.

Figure 2-5. An sfi1 nonphosphorylatable N terminus allele shows normal spindle morphology Conversion of all 11 N terminus in vivo phosphorylation sites to nonphosphorylatable residues (S/T to A, Y to F; sfi1-NnonP) resulted in no spindle morphology defects when multiple copies of the allele are present. Cells grown to early- to mid-log phase shifted to 30°C for 4 h. Immunofluorescent staining (α-tubulin: green, DNA: blue) shows bipolar spindles for control and sfi1-NnonP. Scale bar: 5 µm.

following: multiple alleles at the SFI1 locus and a single allele at the LEU2 locus. The first

strain indicated was used for all phenotypic analysis. These results indicate that phosphorylation

of the converted sites on the Sfi1 N terminus on its own is not required for growth and

appropriate bipolar spindle formation.

In order to determine whether the Sfi1 N terminus phosphorylation status impacts

interactions with other SPB proteins, I examined whether the sfi1-NnonP mutant allele shows

synthetic growth defects with mutant alleles of satellite proteins. I found that the sfi1-NnonP

mutant allele shows synthetic growth defects with mutant alleles of satellite proteins Spc42

(spc42-11; N58Y, E71G, L121P) and Spc29 (spc29-2; R161S) at 30°C (Fig. 2-6). These alleles

have previously been shown to lead to defects in SPB duplication at 37°C (Donaldson and

Kilmartin, 1996; Elliott et al., 1999). No synthetic growth defect was seen with spc42-S4T6A

(Jaspersen et al., 2004) or cnm67∆Z (Schaerer et al., 2001), suggesting that the synthetic growth

defects we do see are a result of the combination of mutations in both proteins. The double 1C 2C 1C 2C 1C 2C mutant sfi1-NnonP spc42-11 at nonpermissive temperature (30°C) shows characteristics of

monopolar spindles, suggesting an SPB duplication pathway defect, and have chromosome

segregation defects, with DNA localized to one region and increased DNA levels per cell

compared to controls and single mutants. The double mutant sfi1-NnonP spc29-2 at 30°C also

shows characteristics of monopolar spindles with DNA localized to one region. However, the

single mutant spc29-2 commonly shows this defect as well at 30°C. Nonetheless, a growth

defect is present in the double mutant, and we see increased DNA levels per cell compared to the

control and single mutants. Overall, these results suggest that Spc42 and Spc29 are promising

candidates as Sfi1 N terminus physical interactors.

Figure 2-6. An sfi1 nonphosphorylatable N terminus allele shows genetic interactions with the satellite protein-encoding genes SPC42 and SPC29 Conversion of all 11 N terminus in vivo phosphorylation sites to nonphosphorylatable residues (S/T to A, Y to F; sfi1-NnonP) resulted in no growth phenotype when multiple copies of the allele are present. However, spc42-11 sfi1-NnonP and spc29-2 sfi1-NnonP show synthetic growth defects. A. Dilution series on YPD. B, C. Cells grown to early- to mid-log phase then shifted to 30°C for 4 h. B. Immunofluorescent staining (α-tubulin: green, DNA: blue). The double mutants display single microtubule asters and DNA localized to one region while control cells and single mutants display a bipolar spindle. Percentages are indicated for the phenotype described. The remaining percentage not indicated is the second of these two phenotypes. Notes: Control image is at 24°C. spc29-2 strains were imaged separately from the remaining images. The spc29-2 single mutant has 70% single microtubule asters with DNA localized to one region. Scale bar: 5 µm. C. DNA content via flow cytometry at 24°C and 30°C. Double mutants have increased DNA levels per cell compared to controls and single mutants. 25

An sfi1 nonphosphorylatable N terminus allele shows genetic interactions with a mutant

allele of CDC31

I also saw that the sfi1-NnonP mutant allele shows synthetic growth defects with a mutant allele

of CDC31, cdc31-2 (E133K; Fig. 2-7), but not with a different allele, CDC31-16 (Vallen et al.,

1994). At 36°C, the cdc31-2 allele leads to a defect in SPB duplication prior to satellite

formation (Winey et al., 1991). The double mutant sfi1-NnonP cdc31-2 at nonpermissive

temperature (30°C) shows characteristics of monopolar spindles, suggesting an SPB duplication

pathway defect, and has chromosome segregation defects, with DNA localized to one region and

an increased percentage of cells with DNA content corresponding to the G2/M phase of the cell

cycle (2N for control and sfi1-NnonP, 4N for diploidized cdc31-2 strains) compared to control or

single mutants. These results provide the foundation to examine whether Cdc31 may regulate 1C 2C 1C 2C 1C 2C the Sfi1 N terminus.

III. Discussion

The elongated half-bridge is a true intermediate, and Mps1 is required for the subsequent

step of satellite formation

I showed that the mps1-1 mutant, which arrests with an elongated half-bridge lacking a

satellite, contains Sfi1 molecules positioned correctly at the half-bridge. This supports the

current model of SPB duplication, indicating that the elongated half-bridge is an authentic

intermediate. Its formation serves as the first step of SPB duplication (Fig. 2-1) and may require

the Sfi1 C terminus.

This finding also indicates that the protein kinase Mps1 is required for the subsequent

step of satellite formation, potentially via regulation of the Sfi1 N terminus. Mps1 associates in a

Figure 2-7. An sfi1 nonphosphorylatable N terminus allele shows genetic interactions with the satellite protein-encoding genes CDC31 Conversion of all 11 N terminus in vivo phosphorylation sites to nonphosphorylatable residues (S/T to A, Y to F; sfi1-NnonP) resulted in no growth phenotype when multiple copies of the allele are present. However, cdc31-2 sfi1-NnonP shows slight synthetic growth defects at 30°C. A. Dilution series on YPD. B, C. Cells shifted to 30°C for 4 hrs. B. Immunofluorescent staining (α-tubulin: green, DNA: blue). The single cdc31-2 mutant commonly displays a bipolar spindle. The double mutant displays single microtubule asters and DNA localized to one region. Percentages are indicated for the phenotype described. The remaining percentage not indicated is the second of these two phenotypes. Scale bar: 5 µm. C. DNA content via flow cytometry. The double mutant has more cells in G2/M (2N for control and sfi1-NnonP, 4N for diploidized cdc31-2 strains) compared to the single mutant.

27 complex with Sfi1 (Breitkreutz et al., 2010). Therefore, Mps1 is a good candidate for regulating

Sfi1 in the process of satellite recruitment. However, no previous research has determined whether Sfi1 is phosphorylated by Mps1. It is known that the satellite proteins Spc42 and Spc29 are Mps1 substrates, both in vivo and in vitro (Castillo et al., 2002; Holinger et al., 2009; Araki et al., 2010). The localization of the Sfi1 N terminus at the distal tip of the elongated half-bridge, the site of satellite assembly, makes it an excellent candidate to recruit Spc42 and/or Spc29. This recruitment may require phosphoregulation of multiple proteins, Sfi1 and/or the satellite proteins.

Further work examining a C-terminally tagged Sfi1 with GFP in an mps1-1 background would prove fruitful, as it would be expected that the Sfi1 C terminus will localize to the center of the elongated half-bridge in mps1-1. Additionally, it would useful to examine Sfi1 N and C terminus localization in cdc31-5 and CDC31-16 backgrounds. These mutants display an elongated half-bridge lacking a satellite, and examination of Sfi1 at the half-bridge would show whether the half-bridge in these mutants is also intact in regards to Sfi1’s positioning. This could then suggest that Cdc31 is involved in satellite recruitment, possibly via regulation of Sfi1.

An sfi1 nonphosphorylatable N terminus allele shows genetic interactions with mutant alleles of SPC42 and SPC29

The identification of genetic interactions between SFI1 and SPC42 and SPC29, separately, suggests that the satellite proteins Spc42 and Spc29 are promising candidates as Sfi1

N terminus physical interactors. Additionally, this suggests that phosphorylation of the Sfi1 N terminus may play a role in the process of recruitment of Spc42 and/or Spc29, as the nonphosphorylatable allele of SFI1 at its N terminus perturbed the SPB duplication process when 28 combined with mutant alleles of SPC42 and SPC29. While there was no synthetic growth defect with the nonphosphorylatable allele of SPC42, spc42-S4T6A, phosphorylation sites on Spc42 other than S4 and T6 may be important in satellite formation. Spc42 residues S4 and T6 are both within Cdk1 consensus motifs and are phosphorylated by Cdk1 (Jaspersen et al., 2004). I propose that phosphorylation by Mps1 is important in satellite formation, and thus other phosphorylation sites on Spc42 may play a role. Future research examining whether a direct physical interaction exists between the Sfi1 N terminus and satellite proteins is needed.

Additionally, an understanding of the phosphorylation status and regulator(s) required for this interaction will prove key.

An sfi1 nonphosphorylatable N terminus allele shows genetic interactions with a mutant allele of CDC31

I found that sfi1-NnonP shows synthetic growth defects with cdc31-2. Previous research suggests that SFI1 genetic interactions with CDC31 are Sfi1-domain specific, with mutations in the Sfi1 repeat domain (shown by synthetic growth defects and dosage suppression) (Kilmartin,

2003), but not the C terminus (tested by dosage suppression) (Anderson et al., 2007), leading to genetic interaction with CDC31. However, no previous work to this point had examined the Sfi1

N terminus in this manner. Kilmartin (2003) did show that the Sfi1 N terminus does not directly bind Cdc31 repeats in vitro. However, given that the Sfi1 repeats bind Cdc31 (Kilmartin, 2003),

Cdc31 could play a role in regulation of the Sfi1 N terminus.

Additionally, Schiebel and colleagues found that Cdc31 is phosphorylated by Mps1 in vitro. These authors suggest that alteration of one of the Mps1 sites (T110) to a nonphosphorylatable residue leads to a failure in SPB duplication, as assessed by alpha-tubulin 29

(Tub1) and Spc42 fluorescent signals (Araki et al., 2010). It will be interesting to determine whether Mps1 regulation of Cdc31 may impact the Sfi1 N terminus. Future examination of whether Cdc31 affects Sfi1 N terminus function, including the potential function of satellite recruitment, will prove useful.

IV. Materials and Methods

Strain Construction

The W303 strain background was used for all experiments (Table S1; except Y3165).

The Sfi1 N terminus is defined as containing residues 1 to 185, all residues prior to the first Sfi1 repeat as defined by Kilmartin (2003). To obtain appropriate yeast strains for examining an Sfi1

N terminus deletion, I created sfi1Δ::SFI1-TRP1-KANMX YCp-SFI1-URA3 (JA105) by integration of pRS316-SFI1 (E2226) into sfi1Δ::SFI1-TRP1-KANMX MATa (JA50). pRS316-

SFI1 was created using pRS316 and pRS314-SFI1 (gift of John Kilmartin) cut with BamHI and

XhoI. JA50 was created by isolating a spore from the diploid resulting from transforming pRS304-KANMX-SFI1 (E2225), cut with NcoI, into sfi1Δ::KANMX/SFI1 MATa/MATα (JA43) via standard protocol (Gietz and Woods, 2002). E2225 was created using pRS304-KANMX cut at BamHI and XhoI with the SFI1 sequence inserted from pRS314-SFI1. JA43 was constructed as previously described (Longtine et al., 1998). pRS304-KANMX-SFI1 was constructed from pRS304-KANMX and pRS314-SFI1 cut with BamHI and XhoI.

pRS304-KANMX-SFI1 (E2225) was mutagenized via PCR as previously described

(Schaerer et al., 2001) to create pRS304-KANMX-sfi1-∆N (E2290), which was transformed into sfi1∆::KANMX YCp-SFI1-URA3 MATa (JA54). JA54 was a spore obtained from dissection of

JA51 (sfi1Δ::KANMX/SFI1 YCp-SFI1-URA3 MATa/MATα; JA43 transformed with pRS316- 30

SFI1). sfi1-∆N-KANMX YCp-SFI1-URA3 (JA108) was obtained by transformation of pRS304-

KANMX-sfi1-∆N into JA54. sfi1Δ:: sfi1-∆N –TRP1-KANMX/SFI1 MATa/MATα (JA107) was created by transformation of pRS304-KANMX-sfi1-∆N into JA43.

Control and mutant strains used to study sfi1-NnonP (except JA50 and JA1-001) contain silent mutations for nucleotide changes of T-195G (upstream SFI1 region), C99G, A394T,

G395C, A2130G, and T2131C in order to develop additional restriction enzyme sites within the

SFI1 sequence for plasmid manipulation purposes. Standard PCR mutagenesis was used to create these sites in plasmid E2225 resulting in the plasmid pRS304-KANMX-SFI1 Control

(E2249). To create sfi1∆::sfi1-NnonP-TRP1-KANMX (JA103), all 11 Sfi1 N terminus in vivo phosphorylation sites (T7, S10, T11, T23, S24, T26, T30, S80, Y103, S109, Y117) were converted to nonphosphorylatable residues (S/T to A, Y to F). Nucleotide changes are as follows: A19G, T28G, A31G, A67G, T70G, A76G, A88G, T238G, A308T, T325G, and A350T.

This altered sequence (created by GenScript, Piscataway, NJ, USA) was cloned into pRS304-

KANMX-SFI1 Control (E2249) to create pRS304-KANMX-sfi1-NnonP (E2250). To obtain appropriate yeast strains, the appropriate plasmids with a pRS304-KANMX base were cut with

NcoI and integrated into sfi1Δ::KANMX/SFI1 (JA43). Resulting heterozygotes were then dissected to select appropriate spores, JA50, JA98, and JA103. Integration of the appropriate sequence was confirmed via PCR. Sequencing showed the yeast strains did contain the expected mutation, although additional SFI1 sequence appeared to be present in the strains via sequencing.

Additional alleles of sfi1-NnonP were also created via different integration techniques.

The SFI1 wild-type and control and sfi1-NnonP sequences from the appropriate pRS304-

KANMX plasmids (E2225, E2249, and E2250, respectively, cut with BamHI and XhoI) were placed into the pRS305 vector base, and the resultant plasmids (E2322, E2320, and E2321, 31 respectively) were integrated at the LEU2 locus in sfi1∆::KANMX YCp-SFI1-URA3 MATa

(JA54). The resultant strains were leu2::SFI1-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 (wild- type, JA1-001; control, JA272), and leu2::sfi1-NnonP-LEU2 sfi1∆::KANMX YCp-SFI1-URA3

(JA1-026). Integration of a single copy of the allele was confirmed via southern blot as previously described (Brown, 2004) using Amersham gene images AlkPhos direct labeling and CDP-

Star detection system (GE Healthcare Life Sciences) with genomic DNA (isolated as described in

Amberg et al., 2005) digested with BglII. The following two single-stranded DNA probes were used: a LEU2 probe (498 bp) constructed from primers LEU386F (P103) LEU884R (P106) and an SFI1 probe (491 bp) constructed from primers SFI1372F (P107) and SFI1863R (P108) from pRS305-SFI1 (E2320).

Double mutants were created via standard mating with sfi1::sfi1-NnonP-TRP1-KANMX

Mata (JA103) or Matα (JA104) and dissection of resulting diploids to select appropriate spores.

The spc42, spc29, and cdc31 strains are from the following sources: YUMY001 from Paul

Straight, JVK1782x1380-1a from Michele Jones, Y3165 from John Kilmartin, SLJ715, SLJ810,

SLJ907, SLJ1422 from Sue Jaspersen, and FSY182 from Schaerer et al. (2001).

Five-fold dilution series with 5 µL of the same initial OD/mL for all strains (0.114 in Fig.

2-4; 0.113 in Fig. 2-6; 0.148 in Fig. 2-7) in the left column were placed on YPD plates and grown at the indicated temperatures for 2 days.

Flow cytometry

Cells were fixed with 70% ethanol for 1 h at room temperature or at 4°C overnight.

Fixed cells were then treated with 0.1% RNase in 0.2M Tris-HCl pH 7.5, 20 mM EDTA for 2 h at 37°C, and the DNA was stained with 50 µg/mL propidium iodide (Sigma Chemical Co., St.

Louis, MO) in PBS for 1 h at room temperature or at 4°C overnight. DNA content was analyzed 32 using the CyAn ADP analyzer (Beckman Coulter, Indianapolis, IN, USA) with a 488nm laser.

30,000 events per sample were taken. DNA content (propidium iodide area) is indicated on X axes and Count on Y axes.

Cytological techniques

For immunofluorescence, cultures were fixed with 4% formaldehyde for 45 minutes, subjected to zymolyase and prepared on slides. YOL1/34 rat anti-tubulin antibody (1/150), FITC goat anti-rat secondary antibody (1/200), and Hoescht dye for DNA were used. Imaging of fixed cells was performed at room temperature on an Eclipse Ti inverted microscope (Nikon, Japan) fitted with a CFI Plan Apo VC 60× H numerical aperture 1.4 objective (Nikon, Japan) and a

CoolSNAP HQ2 charge-coupled device camera (Photometrics, Tuscon, AZ). The 1.5X intermediate magnification was used for fixed immunofluorescent images. Metamorph Imaging software (Molecular Devices, Sunnyvale, CA) was used to collect images. Maximum projections using Metamorph Imaging software or ImageJ software (National Institutes of

Health, Bethesda, MD, USA) are shown.

Immuno-EM

Log phase cells were high pressure frozen in a Wohlwend Compact 02 HPF and freeze- substituted in 0.25% glutaraldehyde, 0.1% uranyl acetate in acetone and embedded in Lowicryl

HM20 (Giddings et al., 2001). The anti-GFP antibody was a gift from Jason Kahana/Pam Silver at Harvard University. Imaging was conducted using a FEI Phillips CM100 electron microscope.

Table 2-1. Yeast strains

Strain Genotype JA50 sfi1Δ::SFI1-TRP1-KANMX MATa JA98 sfi1∆::SFI1-TRP1-KANMX MATa JA103 sfi1∆::sfi1-NnonP-TRP1-KANMX MATa JA105 sfi1∆::SFI1-TRP1-KANMX YCp-SFI1-URA3 MATa JA107 sfi1∆::sfi1-∆N-TRP1-KANMX/SFI1 MATa/MATα JA108 sfi1∆::sfi1-∆N-TRP1-KANMX YCp-SFI1-URA3 MATa JA272 leu2::SFI1-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-001 leu2::SFI1-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-026 leu2::sfi1-NnonP-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa YUMY001 mps1-1 ura3-1 his3-11 leu2-3,112 trp1-1 ade2-1 can1-100 MATα JVK1782x1380-1a trp1::GFP-SFI1-TRP1 sfi1Δ::HIS5 mps1-1MATa SLJ715 bar1 spc42-11 MATα JA110 spc42-11 sfi1∆::sfi1-NnonP-TRP1-KANMX MATa Y3165 spc29-2 his3Δ200 trp1Δ1 ura3-52 leu2-3,112 pMPS1-URA MATα JA113 spc29-2 sfi1∆::sfi1-NnonP-TRP1-KANMX MATa SLJ810 bar1 cdc31-2 pURA3-CDC31 MATα JA116 cdc31-2 sfi1∆::sfi1-NnonP-TRP1-KANMX pURA3-CDC31 MATa SLJ907 bar1 CDC31-16 pURA3-CDC31 lys2Δ MATα JA118 CDC31-16 sfi1∆::sfi1-NnonP-TRP1-KANMX pURA3-CDC31 MATa SLJ1422 spc42Δ::LEU2 his3::spc42-S4T6A-HIS3 MATa JA1-027 spc42Δ::LEU2 his3::spc42-S4T6A-HIS3 sfi1∆::sfi1-NnonP-TRP1- KANMX MATa FSY182 cnm67::cnm67∆Z-HIS3MX6 trp1-289 ura3-52 leu2-3,112 MATa JA1-028 cnm67::cnm67∆Z-HIS3MX6 sfi1∆::sfi1-NnonP-TRP1-KANMX MATa

Table 2-2. Plasmids

Identifier Name Genotype E2226 pRS316-SFI1 YCp-SFI1-URA3 E2225 pRS304-KANMX-SFI1 TRP1-KANMX-SFI1 E2249 pRS304-KANMX-SFI1 TRP1-KANMX-SFI1 E2250 pRS304-KANMX-sfi1-NnonP TRP1-KANMX-sfi1-NnonP E2290 pRS304-KANMX-sfi1-∆N TRP1-KANMX-sfi1-∆N E2320 pRS305-SFI1 SFI1-LEU2 E2321 pRS305-sfi1-NnonP sfi1-NnonP-LEU2 E2322 pRS305-SFI1 SFI1-LEU2

Table 2-3. Primers

Name Identifier Sequence LEU386F P103 5’-GGTACTGACTTCGTTGTTGTCAG-3’ LEU884R P106 5’-CAAATCTGGAGCAGAACCGTG-3’ SFI1372F P107 5’-GGTTCATTTCAAACTATCCAGATC-3’ SFI1863R P108 5’-CCTGATCTGCTAACCTTGACTGC-3’

Chapter Three: Cdk1 phosphorylation of the Sfi1 C terminus is required for appropriate growth, bipolar spindle assembly, and chromosome segregation

I. Introduction

The Sfi1 C terminus is phosphorylated at 8 residues in vivo (T816, S855, T876, T877,

S882, S892, S920, S923) (Chi et al., 2007; Keck et al., 2011). Four of these residues are within full Cdk consensus motifs (S/T*-P-x-K/R; T816, S882, S892, S892) and one is within a minimal

Cdk consensus motif (S/T*-P; S923). Recent work has found that Sfi1 is phosphorylated by

Cdk1 in vitro specifically at the four C-terminal residues within full Cdk consensus motifs

(Elserafy et al., 2014). Based on the model by Kilmartin and colleagues, it has been proposed that the Sfi1 C terminus is involved in both the initiation of SPB duplication and separation (Li et al., 2006). However, only the latter has been clearly shown (Strawn and True, 2006; Anderson et al., 2007; Elserafy et al., 2014), in which Cdk1 phosphorylation of the Sfi1 C terminus is specifically required for SPB separation, as nonphosphorylatable mutations within Cdk1 sites in this domain lead to cells arrested with side-by-side SPBs (Anderson et al., 2007; Elserafy et al.,

2014).

Given the model that mitotic Cdk1 phosphorylation of Sfi1 blocks and Cdc14 dephosphorylation enables licensing of SPB duplication, several predictions can be made regarding the phenotypes of Sfi1 C terminus alleles. As previously seen, I would expect some nonphosphorylatable alleles to have side-by-side SPBs, indicating a requirement for Sfi1 C terminus phosphorylation in SPB separation. In addition, these alleles would be predicted to permit duplication at the S and G2/M stages of the cell cycle, and thus, as seen when all mitotic cyclins are deleted (Haase et al., 2001), reduplicated SPBs are expected. A recent study found that conversion of multiple Sfi1 C-terminal Cdk sites to nonphosphorylatable residues led to cells that often arrested with side-by-side SPBs, indicating a requirement for phosphorylation of 37

Sfi1 in SPB separation (Elserafy et al., 2014). They did not see reduplicated SPBs; however, they only examined cells at one timepoint, and a later timepoint may be needed to allow time for reduplication events to occur, as was previously suggested when examining reduplication events upon deletion of mitotic cyclins (Haase et al., 2001).

In addition, I would expect some phosphomimetic alleles to arrest with single SPBs with a short half-bridge, indicating a requirement of dephosphorylation for half-bridge elongation, the first identified step of SPB duplication. Schiebel and colleagues recently showed that cells with

Sfi1 containing phosphomimetic Cdk1 sites arrest with a single SPB, supporting the idea that phosphorylation of Sfi1 inhibits the process of duplication and dephosphorylation promotes duplication (Elserafy et al., 2014). However, their experimental procedure only allowed them to look at the effects of the mutation in a single cell cycle after satellite formation; thus, the role of phosphorylation of the Sfi1 C terminus in the initiation of SPB duplication is not yet clear.

In order to examine these phenotypic predictions, a mutant allele collection containing alterations in individual and combination of phosphorylation sites can be used. Using this collection, the importance of specific sites and the importance of the number of sites converted, and thus the requirement for combinatorial control of phosphorylation on a single protein, can be determined.

II. Results

Mitotic cyclin Clb2/Cdk1 phosphorylates the Sfi1 C terminus at sites within Cdk motifs in vitro

Sfi1 contains five residues within full Cdk1 consensus motifs (S/T*-P-x-K/R), four at the

C terminus known to be phosphorylated in vivo (T816, S855, S882, S892) (Chi et al., 2007; 38

Keck et al., 2011) and one immediately preceding the C terminus (S801; Fig. 3-1). Using a recombinant fragment of Sfi1, I found that four of these residues (S801, T816, S855, S882) and

S923 (an in vivo phosphorylation site within a minimal Cdk consensus motif (S/T*-P)) are phosphorylated by mitotic cyclin Clb2/Cdk1 in vitro (Fig. 3-1, B-G). In vitro Clb2/Cdk1 phosphorylation of three of these residues (T816, S855, S882) as well as S892 has recently been shown (Elserafy et al., 2014). I also identified phosphorylation of non-Cdk sites, including two previously found to be phosphorylated in vivo (see Fig. 3-1 for all sites).

Characterization of an sfi1 C terminus mutant allele collection

In order to determine the importance of individual and combinations of phosphorylated residues, a nonphosphorylatable mutant allele collection was created with individual and combinations of the C-terminal phosphorylated residues converted to alanine. For initial analysis of growth phenotypes, alleles were integrated at the LEU2 locus into a haploid strain containing a rescuing YCp-SFI1-URA3 plasmid. Growth was assessed on 5-FOA and YPD following loss of the rescuing plasmid. Select alleles were integrated at the SFI1 locus.

Alteration of residues that are not within Cdk consensus motifs resulted in no growth defects

(Fig. 3-2). Alteration of individual residues within Cdk1 consensus motifs to nonphosphorylatable residues resulted in a temperature-sensitive (ts) growth defect at 37°C or no defect (Figs. 3-3, 3-4, 3-5, and Tables 3-1 and 3-2). The most severe single mutation was S855A.

Combinations of nonphosphorylatable mutations resulted in varying phenotypes, depending on the residue and number of sites mutated, with more mutations often correlating with a more severe growth phenotype (Figs. 3-3, 3-4, 3-5, and Tables 3-1 and 3-2). The ts sfi1-C2A allele

(T816A S892A) shows a slight growth defect at 37°C. The alleles sfi1-C3A (T816A S882A 39

Figure 3-1. Clb2/Cdk1 phosphorylates Sfi1 at sites within Cdk motifs in vitro A. Sfi1 schematic with N-terminal, Sfi1 repeat, and C-terminal domains. The Sfi1 repeat domain contains 21 conserved Sfi1 repeat sequences (Kilmartin, 2003; Li et al., 2006). The C terminus is defined as all residues (802-946) immediately following the final Sfi1 repeat sequence (Anderson et al., 2007). All residues within full Cdk1 consensus motifs and the single known phosphorylated residue within a C-terminal minimal Cdk1 consensus motif (S923) are identified. B. Identification of phosphorylated residues from two replicate mass spectrometry runs of in vitro phosphorylated recombinant Sfi1 C terminus fusion protein by Clb2/Cdk1. Coverage of Sfi1 was 88% for both runs. If multiple phosphorylated isoforms are probable, highest ions score is in bold, with brackets identifying the region of sites potentially phosphorylated. Sites previously identified in vivo are indicated (Chi et al., 2007; Keck et al., 2011). pS/T: phosphorylated residue. S/T(P): Cdk site. C-G. Annotated spectra with defining ions only of in vitro phosphorylated residues within full Cdk1 consensus and minimal C-terminal consensus motifs. All identified b and y ions, summarized for all ions with or without neutral, water, and/or ammonia loss at any charge state, are indicated for each peptide sequence, with the phosphorylated residue in red as “pS/T.” –H3PO4: neutral loss, -H20: water loss, -NH3: ammonia loss. C. S801. Parent m/z 848.76; z=3. D. T816. Parent m/z 1041.00; z=2. E. S855. Parent m/z + 768.40; z=2. F. S882. Parent m/z 742.38; z=3. Ion y13 –H3PO4 -NH3 at m/z 1480.19. G. S923. Parent m/z 780.39; z=3.

5-FOA B

YPD C

YPD

Figure 3-2. Growth phenotypes of sfi1 C terminus mutant alleles with alteration of non- Cdk sites sfi1 C terminus mutant alleles containing non-Cdk site alterations were integrated at the LEU2 locus and assessed upon loss of the rescuing autonomously-replicating SFI1 plasmid (YCp-SFI1- URA3) on 5-FOA (A) and YPD (B-C). A. Growth on 5-FOA after 4 days. Plated 5 µL of OD/mL 0.1 in first column. B-C. Single colonies from 5-FOA at 24°C were plated on YPD. B. Growth on YPD after 2 days at 24°C, 30°C, 37°C and after 8 days at 16°C. Plated 5 µL of OD/mL 0.1 in first column. C. Growth on YPD after 2 days at 24°C, 30°C, 37°C and after 6 days at 16°C. Plated 5 µL of OD/mL 0.047 in first column.

5-FOA B

5-FOA

Figure 3-3. Growth phenotypes of sfi1 C terminus mutant alleles with alteration of Cdk sites sfi1 C terminus mutant alleles with alteration of Cdk sites to nonphosphorylatable (S/T to A) or phosphomimetic (S/T to E) residues were integrated at the LEU2 locus and assessed upon loss of the rescuing autonomously-replicating SFI1 plasmid (YCp-SFI1-URA3) on 5-FOA. A. Growth after 4 days with the exception of alleles identified with an asterisk, indicating growth after 5 days at 24°C, 30°C, 37°C and 6 days at 16°C. Wild-type SFI1-LEU2 shown in first row, JA272 shown in second and last row. Plated 5 µL of OD/mL 0.1 in first column. B. Growth after 4 days. Plated 5 µL of OD/mL 0.071 in first column.

YPD

Figure 3-4. Growth phenotypes of sfi1 C terminus mutant alleles with alteration of Cdk sites sfi1 C terminus mutant alleles with alteration of Cdk sites to nonphosphorylatable (S/T to A) or phosphomimetic (S/T to E) residues were integrated at the LEU2 locus and assessed upon loss of the rescuing autonomously-replicating SFI1 plasmid (YCp-SFI1-URA3). Single colonies from 5-FOA at 24°C were plated on YPD. Plated 5 µL of OD/mL 0.1 in first column. A. Growth after 2 days. SFI1 (JA1-001) shown in first row, Control (JA272) shown in second row. B. Growth at 16°C for indicated time. Day 2: JA1-001 shown in first row, JA272 shown in second row. Days 4 and 6: JA272 shown as control.

YPD Figure 3-4 (continued).

Table 3-1. Growth phenotype summary of nonphosphorylatable sfi1 alleles with alteration of Cdk sites within the Sfi1 C terminus. Summary of alleles integrated at the LEU2 locus for growth on both 5-FOA and YPD.

T816 S855 S882 S892 S923 24°C 37°C + + A + + A + +/- A + + A + + A + + A A + + A A + +/- A A +/- - A A + +/- A A + +/- A A + +/- A A + + A A + + A A + + A A + + A A A + + A A A +/- - A A A + +/- A A A +/- - A A A -/+/- - A A A +/- - A A A + - A A A + +/- A A A +/- - A A A A +/- - A A A A +/- -

YPD B

YPD

Figure 3-5. Growth phenotypes of sfi1 C terminus mutant alleles with alteration of Cdk sites sfi1 C terminus mutant alleles with alteration of Cdk sites to nonphosphorylatable (S/T to A) or phosphomimetic (S/T to E) residues were integrated at the SFI1 locus, and growth was assessed on YPD. Plated 5 µL of OD/mL 0.1 in first column. A. Growth after 2 days. JA162 (haploid) shown in first row; JA196 (diploid) shown in second row. B. Growth at 16°C for indicated time. JA162 shown in first row. Day 4: JA196 shown in second row.

Table 3-2. Phenotype summary of sfi1 alleles. Summary of alleles integrated at the SFI1 locus. Growth on YPD. +: normal; +/-: 47 slow; -: no growth. R: recessive; D: dominant. H: haploid; Di: diploid; H/Di: mixed ploidy. N/A: not assessed. LB: large-budded cells. MT: microtubule; SS: short bipolar metaphase spindle. Yes: for 24°C and/or 37°C for at least one temperature shift duration, significant difference compared to control (haploid JA162 or diploid JA196) at the same temperature; Yes1: for 24°C and/or 37°C, difference compared to control at the same temperature based on one experiment; Trend: for 24°C and/or 37°C for at least one temperature shift duration, trend towards a difference compared to control at the same temperature (0.05

Strain Strain Growth Recessive/ Ploidy Increased Increased Cell Increased % Identifier Dominant % LB % G2/M viability spindle defect phenotype 24°C 37°C at 37°C 1 SS MT aster sfi1-S801A JA291 + + R N/A N/A N/A N/A N/A N/A sfi1-S855A JA217 + +/- R H No Yes No Yes No sfi1-C2A JA192 + +/- R H Trend Yes No Yes1 No sfi1-C3A JA188 +/- +/- R H/Di No No Yes Yes Yes sfi1-C4A JA249 +/- - D Di Trend No Yes Yes Yes sfi1-C3A+S923A JA184/ +/- +/- R H/Di# N/A# N/A N/A# N/A# N/A# JA273# sfi1-C4A+S923A JA266 +/- - D Di Trend No Yes Yes1 Yes1 sfi1-C5A JA293 +/- - D N/A N/A N/A N/A N/A N/A sfi1-S855E JA221 + +/- R N/A N/A N/A N/A N/A N/A sfi1-C3E JA231 + +/- R H No Yes N/A Yes No sfi1-C4E JA270 +/- - R Di Trend Yes Yes Yes& No& sfi1-C4E+S923E JA268 +/- - R Di No No Yes Yes1 Yes1 48

S892A) and sfi1-C4A (T816A S855A S882A S892A) show growth defects at 24°C and little or no growth, respectively, at 37°C. Mutation of the minimal Cdk consensus site (S923A) in combination with sfi1-C3A or sfi1-C4A does not enhance the growth defect, sfi1-C3A+S923A and sfi1-C4A+S923A, respectively. In summary, alteration of certain Sfi1 C-terminal phosphorylation sites within full Cdk consensus motifs to nonphosphorylatable residues individually (i.e., S855A) or in combination (i.e., T816A T892A alone or in combination with additional altered residues; S855A in combination with additional altered residues) leads to growth defects of varying severity, and conversion of the site within the minimal Cdk consensus motif does not generally appear to enhance growth defects.

In addition to examining the requirement for phosphorylation of the Sfi1 C terminus via nonphosphorylatable alleles, I also examined the requirement for loss of phosphorylation.

Multiple phosphomimetic alleles were created with residues converted to glutamic acid. sfi1-

S855E, sfi1-C3E (T816E S882E S892E), and sfi1-C3E+923E showed a slight ts growth defect at

37°C. sfi1-C4E (T816A S855A S882A S892A) and sfi1-C4E+S923E grew similarly to sfi1-C4A and sfi1-C4A+S923A, respectively (Fig. 3-3, 3-4, and 3-5 and Table 3-2). In summary, alteration of Sfi1 C-terminal phosphorylation sites to phosphomimetic residues leads to less severe or similar defects compared to conversion of these sites to nonphosphorylatable residues.

In order to determine the significance of residue S801, which I showed to be phosphorylated in vitro, I also examined the following alleles: sfi1-S801A, sfi1-C5A (S801A

T816A S855A S882A S892A), and sfi1-C6A (S801A T816A S855A S882A S892A 923A) (Fig. 3-

6). When integrated at the SFI1 locus, sfi1-S801A showed no growth defects, and sfi1-C5A showed only a slightly more severe growth defect than sfi1-C4A at 24°C and lethality at 37°C

(Fig. 3-6A and Table 3-2). I was not able to integrate sfi1-C6A at the SFI1 locus but was able to 49

Figure 3-6. Growth phenotypes of sfi1 alleles with alteration of S801 A. sfi1 mutant alleles with alteration of Cdk sites to nonphosphorylatable residues (S/T to A) were integrated at the SFI1 locus, and growth was assessed after 2 days on YPD. Plated 5 µL of OD/mL 0.055 in first column. JA162 shown in first row; JA196 shown in last row. B. sfi1-C6A integrated at the LEU2 locus is lethal at all temperatures upon loss of the rescuing autonomously- replicating SFI1 plasmid (YCp-SFI1-URA3) on 5-FOA after 4 days. C. leu2::sfi1-C6A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 and leu2::sfi1-C4A+S923A-LEU2 sfi1∆::KANMX YCp-SFI1- URA3 were crossed to JA146, and a haploid spore (lacking YCp-SFI1-URA3) obtained from dissection of the resultant heterozygous diploid was plated on YPD for 2 days. Control: JA272.

50 integrate sfi1-C6A at the LEU2 locus. Loss of a rescuing YCp-SFI1-URA3 plasmid, upon growth on 5-FOA, resulted in lethality at all temperatures (Fig. 3-6B). However, upon backcrossing, I was able to create a strain of sfi1-C6A with integration at the LEU2 locus that was not reliant on a rescuing SFI1 plasmid. This strain was viable and showed a similar phenotype to sfi1-

C4A+S923A (Fig. 3-6C). Overall, conversion of residue S801 to alanine may slightly enhance the growth defects of sfi1-C4A or sfi1-C4A+S923A. However, detailed phenotypic analysis of strains containing this mutation was not performed, as phosphorylation of S801 has not been identified in vivo.

In addition to S923, one other residue, S939, at the Sfi1 C terminus is within a minimal

Cdk consensus motif. I therefore created a nonphosphorylatable allele sfi1-S939A integrated at the SFI1 locus and found that it shows no growth defects (Fig. 3-7). This residue has previously been covered via mass spectrometry but was not found to be phosphorylated in vivo. Therefore, this allele was not further pursued.

Validation of the control strain used for analysis of alleles integrated at the SFI1 locus was performed. This control strain contains a single silent mutation (see Materials and

Methods), which does not appear to affect the phenotype of the cells when compared to a strain containing SFI1 lacking the silent mutation integrated at the SFI1 locus (Fig. 3-8).

Multiple sfi1 alleles are dominant

Given the severity of growth defects in some sfi1 alleles, I further examined whether certain alleles were dominant. Examination of heterozygous diploids of sfi1 alleles revealed that multiple alleles are dominant, as revealed by a slight defect in growth at 37°C (Fig. 3-9, A-C and

Table 3-2). For the alleles which displayed a dominant phenotype, dissection of the 51

Figure 3-7. sfi1-S939A shows no growth defects sfi1-S939A-NATMX MATa (JA243) integrated at the SFI1 locus after 2 days of growth on YPD. Siblings from a single tetrad from dissection of sfi1-S939A-NATMX/SFI1 MATa/MATα (JA233) are shown.

Figure 3-8. Validation for a control strain with integration at the SFI1 locus A control strain with SFI1 integrated at the SFI1 locus containing a single silent mutation (SFI1 Control, JA162) and a strain containing SFI1 lacking the silent mutation integrated at the SFI1 locus (SFI1 Wild-type, JA158) were grown at 24°C and shifted at early-log phase to 37°C for 4 h in YPD for one experiment. DNA content via flow cytometry is shown. % G2/M DNA content, % large-budded cells (LB; n≥100 cells per group), and % of large-budded cells with bipolar spindles or separated poles, determined via immunofluorescent staining of α-tubulin and DNA (BPS/SP; n≥100 cells per group), is indicated for each condition. Note: both strains had a doubling time of 2.2 h. 53

Figure 3-9. Multiple sfi1 alleles are dominant A-B. Heterozygous diploids of sfi1 alleles integrated at the SFI1 locus. A. Cells were grown on YPD for 2 days. Strains: JA144, JA152, JA154, JA156, JA157, JA190, JA235, JA236, JA237, JA239, JA241, JA260, JA263, JA264. Note: two sfi1-C4A heterozygous diploids (JA235, JA236) are shown to more clearly demonstrate dominance. B. Cells containing the altered S801 residue were grown on YPD for 1 day. Strains: control haploid from dissection, JA196, JA287, JA288, JA289, JA290. A-B. Notes on spore viability resulting from dissection of heterozygous diploids: sfi1-C4A: 6/20 tetrads had four viable spores, 14/20 had at least one inviable spore; sfi1-C4A+S923A: 3/18 tetrads had four viable spores, 15/18 had at least one inviable spore; sfi1- C5A: 6/20 tetrads had four viable spores, 14/20 had at least one inviable spore. C. At the LEU2 locus, dominance of sfi1-C4A+S923A was confirmed, and sfi1-C6A was also dominant. The presence of a rescuing SFI1 plasmid in these two strains rescued the growth defect at 37°C. Heterozygous diploids of sfi1 alleles integrated at the LEU2 locus after 2 days of growth on YPD. Strains with (+) or without (-) a rescuing SFI1 plasmid (YCp-SFI1): sfi1-C6A/SFI1: JA325 (-), JA326 (+); sfi1-C4A+S923A/SFI1: JA1-024 (-), JA333 (+); SFI1 Control/SFI1: JA1-025 (-), JA334 (+). Notes on spore viability resulting from dissection of heterozygous diploids lacking YCp-SFI1: sfi1-C4A+S923A: 1/8 tetrads showed expected segregation; sfi1-C6A: 1/10 showed expected segregation; Control: 8/8 tetrads showed expected segregation.

55 corresponding heterozygous diploids showed poor segregation and spore viability regardless of the spore genotype, which may suggest a meiotic defect (see Fig. 3-9 legend).

sfi1 mutant proteins localize to the SPB

In order to determine whether mutations in SFI1 affect Sfi1’s localization at the SPB, I examined the localization of sfi1 in sfi1-C3A and sfi1-C4A. The sfi1-C3A and sfi1-C4A mutant proteins fused to GFP localize to the SPB at levels similar or greater than that of GFP-Sfi1 (Figs.

3-14, C and D and 3-16, C and D, respectively), indicating that the phenotypes observed are not caused by reduced levels of the mutant proteins but rather are likely due to loss of Cdk1 phosphorylation.

Phosphoregulation of Sfi1 is required for appropriate bipolar spindle assembly and chromosome segregation

I sought to further characterize select nonphosphorylatable and phosphomimetic alleles to better understand the effects of loss of Cdk1 phosphorylation on the Sfi1 C terminus (see Figs. 3-

10 through 3-24 and Table 3-2 for alleles and characterization). Some mutants that show defects in SPB duplication or separation arrest in mitosis due to activation of the spindle assembly checkpoint (SAC) (e.g., Weiss and Winey, 1996; Anderson et al., 2007). Therefore, in sfi1 mutants, I examined whether the prevalence of large-budded cells and G2/M DNA content was enhanced compared to control, which would suggest a mitotic arrest. Analysis of sfi1 alleles showed that some mutants had a trend towards an increased percentage of large-budded cells versus control at 37°C (Figs. 3-11, 3-15, 3-18, and 3-21 and Table 3-2). Additionally, some mutants had an increased percentage of cells with G2/M DNA content (Figs. 3-10, 3-11, 3-20, 56 and 3-21 and Table 3-2). However, multiple mutants did not display a significant increase in large-budded cells and several did not display increased G2/M DNA content, and not all cells which contained increased G2/M DNA content displayed a statistically significant increase in the percentage of large-budded cells compared to control (see Table 3-2). Nonetheless, while no trend or significance was present (i.e., p≥0.1), the majority of cells did have a larger percentage of large-budded cells than control (Figs. 3-10, 3-17, 3-20, and 3-24). For cells which do display an increase in G2/M DNA content or large-budded cells, these phenotypes would suggest that the cells may arrest in mitosis with G2/M DNA content. Cells which do not show these phenotypes may not arrest in mitosis but instead proceed through the cell cycle.

Since increased ploidy and cell lysis defects have previously been seen in mutant alleles of genes encoding other SPB proteins (Schild et al., 1981; Rose and Fink, 1987; Ivanovska and

Rose, 2001; Jaspersen et al., 2002), I examined whether sfi1 alleles also displayed these phenotypes. All sfi1 mutants that show growth defects at 24°C diploidize (Figs. 3-12, 3-13, 3-

15, 3-17, 3-18, 3-21, and 3-24 and Tables 3-2 and 3-3). Diploidization was confirmed for several of these mutants via genetic crosses (Table 3-3). The mutants that show growth defects at 24°C also display defects in cell viability (as assessed using a vital stain) at 37°C (Figs. 3-13,

3-15, 3-17, 3-18, 3-21, 3-24 and Table 3-2). For sfi1-C2A, sfi1-C3A+S923A, and sfi1-C3E, I also assessed viability of cells on solid media following a temporary shift to restrictive temperature in liquid media (see legend for Figs. 3-11, 3-17, and 3-20). Via this technique, all strains showed decreased viability after exposure to higher temperatures. In summary, multiple sfi1 alleles show increases in ploidy and cell viability defects, both phenotypes which have previously been seen in mutant alleles of genes encoding other SPB proteins. 57

To examine whether mutations in SFI1 lead to defects in spindle morphology and chromosome segregation, I examined microtubules and DNA via immunofluorescence in large- budded cells. Upon examination of spindle morphology, all alleles examined had more single microtubule asters with chromosome missegregation than control at 37°C, suggestive of a defect in SPB duplication or separation (Figs. 3-10, 3-11, 3-14, 3-16, 3-17, 3-19, 3-20, 3-22, and 3-24 and Table 3-2). The alleles that show growth defects at 24°C also displayed this spindle morphology defect at 24°C. In addition, large-budded cells in several mutants that show growth defects at 24°C also displayed increases in short metaphase bipolar spindles at 24°C and/or 37°C

(Figs. 3-14, 3-16, 3-17, 3-19, and 3-24 and Table 3-2). Conditional alleles did not show this latter phenotype. Overall, all examined alleles showed spindle morphology defects and chromosome missegregation.

Cdk1 phosphorylation of Sfi1 is required for SPB separation and to block SPB reduplication

In order to determine the origins of the spindle defects observed in sfi1 alleles, I examined the morphology of the SPBs in sfi1-C3A and sfi1-C4A. The phenotypes of sfi1-C4A in large-budded cells were examined using structured illumination microscopy (SIM; Fig.3-25).

Spc42-GFP was used to label both the SPB and the assembly intermediate, the satellite (Adams and Kilmartin, 1999). Using SIM, it was possible to resolve side-by-side SPBs and a satellite versus mother SPB into two Spc42-GFP foci (190±40 nm distance apart in control unbudded and small-budded cells, n=18 from an asynchronous culture of control cells; see Fig. 3-25A for satellite and mother SPB in a G1-arrested cell). In contrast to control cells containing mainly two separated SPBs, sfi1-C4A cells at 37°C commonly displayed one of the following three 58

Figure 3-10. sfi1-S855A displays spindle and chromosome segregation defects A-E. sfi1-S855A (JA217) or control (JA162) cells grown at 24°C and shifted at early-log phase to 37°C for 4 h in YPD. In liquid culture, sfi1-S855A grew similarly to control at both 24°C (Doubling time (DT): 2.2±0.0 and 2.0±0.1, respectively, not significant, NS) and 37°C (DT: 2.1±0.1 and 1.8±0.2, respectively, NS). Significance is only shown for comparisons between strains at each temperature. A. Budding content. Error bars: SD. n=200 cells per group from 2 experiments. B-C. DNA content via flow cytometry. C. Quantification of B. At 24°C, sfi1- S855A has more cells with G2/M DNA content than control. **: p<0.01. Error bars: SD from 2 experiments. D-E. Immunofluorescent staining (α-tubulin: green, DNA: blue) of fixed large- budded cells. MT: microtubule. D. Representative cells at 24°C and 37°C. Upper three panels: bipolar spindles. Lower panel: 1 MT aster. Bar: 5 µm. E. Quantification of D. At 24°C, sfi1- S855A trends toward more single microtubule asters versus control (p=0.07). **: p<0.01. Error bars: SD. n≥199 cells per group from 2 experiments. 59

Figure 3-11. sfi1-C2A displays spindle and chromosome segregation defects A-E. sfi1-C2A (JA192) or control (JA162) cells grown at 24°C and shifted at early-log phase to 37°C for 4 h in YPD. Significance is only shown for comparisons between strains at each temperature. A. Budding content. At 37°C, sfi1-C2A trends toward more large-budded cells versus control (p=0.06). Error bars: SD. n≥300 cells per group from 3 experiments. B-C. DNA content via flow cytometry. C. Quantification of B. At 37°C, sfi1-C2A has more cells with G2/M DNA content than control. *: p<0.05. Error bars: SD for 3 experiments. D-E. Immunofluorescent staining (α-tubulin: green, DNA: blue) of fixed large-budded cells. MT: microtubule, SS: short metaphase bipolar spindle, BPS/SP: bipolar spindle or separated poles. D. Representative cells at 24°C and 37°C. Bar: 5 µm. E. Quantification of D. n≥100 cells per group from 1 experiment. Note on cell viability: After exposure to the restrictive temperature, sfi1-C2A viability was similar to that at 24°C but less than control cells. Specifically, sfi1-C2A was incubated in liquid culture at 24°C or 37°C for 4 h and then plated on solid rich media at 24°C. After two days of growth, 58% and 68% of sfi1-C2A (vs. 100% viability for controls) plated colonies grew from 24°C and 37°C, respectively.

Figure 3-12. sfi1-C3A spontaneously diploidizes A-B. DNA content via flow cytometry of an isolated diploid strain of sfi1-C3A. A. Upper panels from left to right: Haploid control (JA162), tetraploid resulting from cross between JA180 and H111αα, heterozygous diploid spore obtained from dissection of tetraploid. Lower panels: Spores from one tetrad obtained from dissection of heterozygous diploid spore from upper right panel described above. sfi1-C3A spontaneously diploidizes upon germination. B. G1-arrested cells using α-factor. Left to right: JA162, JA188.

Figure 3-13. sfi1-C3A displays defects in viability and ploidy A-D. sfi1-C3A (JA188) or control diploid (JA196) cells grown at 24°C or shifted at early-log phase to 37°C for 4 h in YPD. A. Trypan blue staining to assess cell viability. **: p<0.01. Error bars: SD. B. Budding content of viable cells. No significant differences for % large-budded cells. LB: large-budded, SB: small-budded, UB: unbudded. n=200 cells per group from 2 experiments (A-B). C-D. DNA content via flow cytometry. D. *p<0.05. Error bars: SD from 2 experiments.

Figure 3-14. sfi1-C3A displays spindle and chromosome segregation defects A-B. Immunofluorescent staining (α-tubulin: green, DNA: blue) of fixed large-budded sfi1-C3A (JA188) or control (JA196) cells grown at 24°C or shifted at early-log phase to 37°C for 4 h in YPD. MT: microtubule, SS: short metaphase bipolar spindle, BPS/SP: bipolar spindle or separated poles. Bar: 5 µm. B. Quantification of A. Asterisks indicate a statistically significant difference using the Student’s t test between sfi1-C3A and control for each phenotypic category at 24°C (red) and 37°C (black). **p<0.01. *: p<0.05. Significance is only shown for comparisons between strains at each temperature. Error bars: SD. n≥200 cells per group from 2 experiments. C-D. Asynchronous cultures GFP-sfi1-C3A pLEU-HIS-SPC42-mCherry (JA308, JA309) or GFP-SFI1 pLEU-HIS-SPC42-mCherry (JA310, JA311) were grown at 24°C in SC- Leu and shifted at early-log phase to 37°C for 4 h. Cells were briefly fixed prior to imaging or imaged live. C. Localization of GFP-Sfi1 (green) and Spc42-mCherry (red) and a merged image for fixed control at 37°C and sfi1-C3A cells at 24°C and 37°C. Bar: 5 µm. D. Quantification of C. *p<0.05 via Student’s t test. Error bars: SD. n≥14 cells per group from 2 experiments.

Figure 3-15. sfi1-C4A displays defects in viability and ploidy A-D. sfi1-C4A (JA249) or control diploid (JA196) cells grown at 24°C or shifted to 37°C for 4h (left panels) or 8.25 h (right panels) in YPD. A. Trypan blue staining to assess cell viability. *: p<0.05. Error bars: SD. B. Budding content of viable cells. Student’s t test performed on % large-budded of viable cells only: for 4 h shift, NS; for 8.25 h shift, p=0.08 for sfi1-C4A versus control at 37°C. LB: large-budded, SB: small-budded, UB: unbudded. For 4h and 8.25 h shifts, independently, n≥200 cells per group from 2 experiments (A-B). C-D. DNA content via flow cytometry. D. Quantification of C with cells containing 0C removed from quantification. **p<0.01. *: p<0.05. Error bars: SD from 2 experiments. 67

Figure 3-16. sfi1-C4A displays spindle and chromosome segregation defects A-B. sfi1-C4A (JA249) or control diploid (JA196) cells grown at 24°C or shifted to 37°C for 8.25 h (A-B) or 4h (B) in YPD. A. Immunofluorescent staining (α-tubulin: green, DNA: blue) of fixed large-budded cells. MT: microtubule, SS: short metaphase bipolar spindle, BPS/SP: bipolar spindle or separated poles. Bar: 5 µm. B. Quantification of A and of same experiment with a 4 h shift. Asterisks: statistically significant difference via Student’s t test between sfi1- C4A and control for each phenotypic category at 24°C (red) and 37°C (black). **p<0.01. *: p<0.05. Significance only shown for comparisons between strains at each temperature. Error bars: SD. n≥192 cells per group from 2 experiments for 8.25 h shift. n≥200 cells per group from 2 experiments for 4 h shift. C-D. GFP-sfi1-C4A pLEU-HIS-SPC42-mCherry (JA353, JA354) or GFP-SFI1 pLEU-HIS-SPC42-mCherry (JA310, JA311) grown in SC-Leu at 24°C or shifted to 37°C 9 h were fixed prior to localization (C) of Sfi1 (green) and Spc42 (red) and quantification of Sfi1 levels in arbitrary units (a.u.) (D). Error bars: SD. n≥62 cells per group from 2 experiments. Bar: 5 µm.

Figure 3-17. sfi1-C3A+S923A displays spindle and chromosome segregation defects A-B, D. All alleles were integrated at the LEU2 locus: Wild-type SFI1 (DT 1.7 h), Control SFI1 (DT: 1.6 h), and sfi1-C3A+S923A at LEU2 (DT: 2.1 h). Cells were grown at 24°C or shifted to 37°C for 4 h in YPD (A,B) or YM-1 (D). A. Trypan blue staining to assess cell viability. n≥100 cells per group from 1 experiment. B. Propidium iodide (PI; binds DNA) staining to assess DNA content via flow cytometry. % large-budded of viable cells (LB) indicated, n≥100 cells per group from 1 experiment. C. Cells were treated with DNase, and PI staining was assessed via flow cytometry. sfi1-C3A+S923A shows increased PI staining via flow cytometry. However, this was due to the residual staining due to the large size of the cells (unbudded cell area: 39±9 µm2 and 16±3 µm2 in mutant and control, respectively, n=5 per group, p<0.001), as cells treated with DNase still showed increased PI staining compared to control. Left panels: forward scatter on X axis, side scatter on Y axis; cell size does not change with DNase staining. Right panels: histogram of PI staining. There is a rightward shift in the PI peak from control to mutant both with and without DNase. D. Immunofluorescent staining (α-tubulin: green, DNA: blue) of fixed large-budded cells. Bar: 5 µm. Note on cell viability: sfi1-C3A+S923A at the LEU2 locus shows decreased viability. sfi1-C3A+S923A was incubated in liquid culture at 24°C or 37°C for 4 h and then plated on solid media at 24°C. After two days of growth, only 25% and 34% of sfi1- C3A+S923A (vs. 100% viability for control) plated colonies grew from 24°C and 37°C, respectively. 70

Figure 3-18. sfi1-C4A+S923A displays defects in ploidy and viability A-D. sfi1-C4A+S923A (JA266) or control diploid (JA196) cells grown at 24°C and shifted at early-log phase to 37°C for 4 h in YPD. Significance is only shown for comparisons between strains at each temperature. A. Trypan blue staining to assess cell viability. *: p<0.05. Error bars: SD. n≥200 cells per group from 2 experiments. B. Budding content. At 37°C, sfi1-C5A trends toward more large-budded cells versus control (p=0.06). Error bars: SD. n≥200 cells per group from 2 experiments. C,D. DNA content via flow cytometry. D. Quantification of C with cells containing 0C DNA content removed from quantification. Error bars: SD for 2 experiments.

Figure 3-19. sfi1-C4A+S923A displays spindle and chromosome segregation defects A-B. Immunofluorescent staining (α-tubulin: green, DNA: blue) of fixed large-budded cells. MT: microtubule, SS: short metaphase bipolar spindle, BPS/SP: bipolar spindle or separated poles. A. Representative cells at 24°C and 37°C. Bar: 5 µm. B. Quantification of A. n≥100 cells per group from 1 experiment.

Figure 3-20. sfi1-C3E displays spindle and chromosome segregation defects sfi1-C3E (JA231) or control (JA162) cells grown at 24°C and shifted at early-log phase to 37°C for 4 h in YPD. Significance is only shown for comparisons between strains at each temperature. A. Budding content. No significant differences. Error bars: SD. n≥300 cells per group from 3 experiments. B-C. DNA content via flow cytometry. C. Quantification of B. *: p<0.05. Error bars: SD from 3 experiments. D-E. Immunofluorescent staining (α-tubulin: green, DNA: blue) of fixed large-budded cells. MT:microtubule. D. Representative cells at 24°C and 37°C. Bar: 5 µm. E. Quantification of D. *: p<0.05. **: p<0.01. Error bars: SD. n≥199 cells per group from 2 experiments. Note on cell viability: sfi1-C3E shows decreased viability at restrictive temperature. sfi1-C3E was incubated in liquid culture at 24°C or 37°C for 4 h and then plated on solid media at 24°C. After two days of growth, 100±34% and 64±7% (normalized to 100% viability for control at each temperature) of sfi1-C3E plated colonies grew from 24°C and 37°C, respectively (p=0.02 via Student’s t test for sfi1-C3E versus control at 37°C).

Figure 3-21. sfi1-C4E displays defects in ploidy and viability A-D. sfi1-C4E (JA270) or control (JA196) cells grown at 24°C and shifted at early-log phase to 37°C for 4 h (left panels) or 8.25 h (right panels) in YPD. Significance is only shown for comparisons between strains at each temperature. A. Trypan blue staining to assess cell viability. *: p<0.05. No significant differences (NS) for 8.25 h shift. Error bars: SD. B. Budding content. Student’s t test performed on % large-budded of viable cells only: for 4 h shift, NS; for 8.25 h shift, p=0.09 for sfi1-C4A versus control at 37°C. Error bars: SD. n≥300 cells per group from 3 experiments for 4 h shift (A,B). n≥200 cells per group from 2 experiments for 8.25 h shift (A,B). C-D. DNA content via flow cytometry. D. Quantification of C with cells containing 0C removed from quantification. **: p<0.01. *: p<0.05. NS for 4 h shift. Error bars: SD from 3 (4 h shift) or 2 (8.25 h shift) experiments. 75

Figure 3-22. sfi1-C4E displays spindle and chromosome segregation defects sfi1-C4E (JA270) or control (JA196) cells grown at 24°C and shifted at early-log phase to 37°C for 4 h in YPD. Immunofluorescent staining (α-tubulin: green, DNA: blue) of fixed large- budded cells. MT: microtubule. A. Representative cells at 24°C and 37°C. Bar: 5 µm. B. Quantification of A. Asterisks: statistically significant difference via Student’s t test between sfi1-C4A and control for each phenotypic category at 24°C (red) and 37°C (black). **p<0.01. *: p<0.05. Significance only shown for comparisons between strains at each temperature. Error bars: SD. n≥200 cells per group from 2 experiments. 76

Figure 3-23. sfi1-C4E shows multiple spindle morphology phenotypes via EM A-C. sfi1-C4E (JA270) cells grown at 24°C and shifted at early-log phase to 37°C for 4 h in YPD were prepared for EM. Serial sections from three cells were examined. The following three phenotypes were seen: a single SPB at one location in the nuclear envelope (A; 2 adjacent serial sections from same cell shown; n=1; half-bridge not visible), two SPBs at one location within the nuclear envelope (side-by-side SPBs; B; n=1), and a short bipolar spindle with one SPB per pole (C; n=1). Scale bars: 100 nm (A-B), 500nm (C). Note: Upon temperature shift for 8.75 h, serial sections from one cell revealed two SPBs at one location within the nuclear envelope.

Figure 3-24. sfi1-C4E+S923E displays spindle and chromosome segregation defects sfi1-C4E+S923E (JA268) or control (JA196) cells grown at 24°C and shifted at early-log phase to 37°C for 4 h in YPD. Significance is only shown for comparisons between strains at each temperature. A. Trypan blue staining to assess cell viability. *: p<0.05. Error bars: SD. B. Budding content. No significant differences. Error bars: SD. n≥154 cells per group from 2 experiments (A-B). C-D. DNA content via flow cytometry. D. Quantification of C with cells containing 0C removed from quantification. **p<0.01. Error bars: SD from 2 experiments. E-F. Immunofluorescent staining (α-tubulin: green, DNA: blue) of fixed large-budded cells. MT: microtubule. E. Representative cells at 24°C and 37°C. Bar: 5 µm. F. Quantification of E. n≥100 cells per group from 1 experiment. 78

Table 3-3. Ploidy determination for sfi1 alleles via genetic crosses. Select sfi1 strains were mated to haploid (MATα) and diploid strains (MATα/MATα) to determine ploidy. Dissection of a strain resulting from mating of two strains with the same ploidy should result in viable spores with generally appropriate segregation. Number of tetrads with viable spores and generally appropriate segregation out of the total number of tetrads tested (Total/Total tetrads tested) are indicated. NT: No tetrads for dissection were identified following conditions that induce spore formation of three and nine appropriate isolates for sfi1-C4A and sfi1-C3A+S923A, respectively, obtained from cross of sfi1 allele to MATα or MATα/MATα, respectively. For SFI1, data for wild-type and control haploid strains are combined for each locus. Alleles are integrated at the SFI1 locus unless otherwise indicated. sfi1-C3A, sfi1-C4A, and sfi1-C4E diploidize. sfi1- C3A+S923A integrated at the LEU2 locus is haploid, while two isolates were haploid and one was diploid when integrated at the SFI1 locus. *: Contains wild-type SFI1 sequence lacking silent mutations (see Materials and Methods).

Strain Strain Tetrads with Tetrads with viable Identifier viable spores and spores and generally generally appropriate appropriate segregation from segregation from dissection of cells dissection of cells resulting from resulting from cross of sfi1 allele cross of sfi1 allele to MATα to MATα/MATα SFI1 (integrated at LEU2) JA1-001*/ 60/70 1/16 JA272 SFI1 JA158*/JA162 55/71 0/36 sfi1-C3A JA188 0/40 25/40 sfi1-C4A JA249 NT 13/13 sfi1-C3A+S923A (integrated JA273 46/61 NT at LEU2) sfi1-C3A+S923A (haploid JA184 24/28 0/38 isolates) sfi1-C3A+S923A (diploid JA184 0/35 16/20 isolate) sfi1-C4E JA270 0/16 22/22

Figure 3-25. sfi1-C4A displays extra SPBs A. G1-arrested SFI1 SPC42-GFP (JA254) using α-factor at 24°C in YPD. Fixed cells were imaged by SIM. GFP on left (inset bar: 1 µm) and merge with transmitted image on right. Bar: 5 µm. B-F. Asynchronous sfi1-C4A SPC42-GFP (JA302) and SFI1 SPC42-GFP (JA254) control cells grown at 24°C and shifted to 37°C for 9 h in YPD. Fixed large-budded cells were imaged by SIM. B. GFP on left (inset bar: 1 µm) and merge with transmitted image on right. In sfi1- C4A large-budded cells that contained at least three GFP foci, two foci were located adjacent to one another (260±50nm, n=37), while a third GFP focus was located at a distance (2.15±2.82 µm, n=33; 13.71±6.92% at 24°C, 22.57±1.18% at 37°C). Of these cells with reduplicated SPBs, some had four SPBs total, indicating two reduplicated SPBs (n=1 of 11 at 24°C, n=3 of 21 at 37°C). Bar: 5 µm. C. Quantification of B. Asterisks: statistically significant difference via Student’s t test between sfi1-C4A and control for each phenotypic category at 24°C (red) and 37°C (black). **p<0.01. *: p<0.05. Significance only shown for comparisons between strains at each temperature. Error bars: SD. n≥77 cells per group from 2 experiments. D. Schematic of large-budded cell with a reduplicated SPB. SPB 1: Single SPB at one pole, SPB 2: Paired SPB, SPB 3: Reduplicated SPB. E-F. Histogram of ratios of Spc42-GFP intensity as indicated. a.u.: arbitrary units. E. Ratio of GFP intensity of reduplicated to paired SPB: 0.7±0.2, n=24. F. Ratio of GFP intensity of single SPB at one pole to paired SPB: 0.9±0.6, n=28. 81 phenotypes: a single (Spc42-GFP) focus, two adjacent foci (280±70nm from each other, n=78) suggestive of side-by-side SPBs, or at least three foci (Fig. 3-25, B and C). At 24°C, side-by- side SPBs were significantly enriched in sfi1-C4A versus control. Commonly, in sfi1-C4A large- budded cells that contained at least three GFP foci, two foci were located adjacent to one another, while a third GFP focus was located at a distance (Fig. 3-25B and legend). The observation of more than two SPBs suggests that an aberrant reduplication event has occurred.

EM was used to further study the phenotype of sfi1 mutant cells. Control large-budded cells display a bipolar spindle with two SPBs, one at either side of the nucleus (Fig. 3-27A). In sfi1-C3A and sfi1-C4A, large-budded cells displayed either unseparated SPBs at the same position within the nuclear envelope (Figs. 3-26A and 3-27B, respectively), suggesting a defect in SPB separation, or short bipolar spindles with aberrant SPBs (Figs. 3-26, B and C and 3-27,

C-E, respectively). Unseparated SPBs were positioned in either a side-by-side orientation (53% and 9% of unseparated SPBs in sfi1-C3A and sfi1-C4A, respectively; Fig. 3-26A, upper panel) or at abnormal orientations to one another (Figs. 3-26A, lower panel, and 3-27B for sfi1-C3A and sfi1-C4A, respectively). Some cells with short spindles have two SPBs, with one being enlarged

(Figs. 3-26B and 3-27C for sfi1-C3A and sfi1-C4A, respectively). However, the remaining cells contain three or four SPBs, in which two SPBs are adjacent, and the other SPB (or two adjacent

SPBs when there are four total) is connected by the short spindle and lies on the other side of the nuclear envelope (Figs. 3-26C and 3-27, D and E for sfi1-C3A and sfi1-C4A, respectively).

Given that a bridge connecting two adjacent SPBs is typically seen in these cells (n=3 of 3 cells for sfi1-C3A; n=4 of 5 cells for sfi1-C4A), and the distance between these SPBs is approximately that expected in side-by-side SPBs in wild-type cells (Figs. 3-26C and 3-27, D and E and legends

Figure 3-26. sfi1-C3A displays unseparated SPBs or bipolar spindles with enlarged or reduplicated SPBs A-C. Asynchronous sfi1-C3A (JA188) cells shifted at early-log phase to 37°C for 4 h in YPD were prepared for EM. Serial sections were examined for 21 cells. A. Representative cells containing two SPBs at one pole with SPBs in a side-by-side configuration (n=8; upper panel) or SPBs with an abnormal orientation (n=7; lower panel). Scale bar: 100 nm. B-C. Cells contain short bipolar spindles with aberrant SPBs. Left panels: full spindle, 500 nm scale bar. Right panels: SPB(s), 100 nm scale bar. B. Representative cell (n=3) with an abnormally large SPB in the upper panel. C. Representative cell (n=3) containing a short bipolar spindle with three SPBs. The upper panel identifies a single SPB, while the lower panel shows a mature SPB connected via a bridge to a partial SPB. Average distance between adjacent SPBs: 148±22 nm.

Figure 3-27. sfi1-C4A displays unseparated SPBs or bipolar spindles with enlarged or reduplicated SPBs A. Asynchronous control cells (JA196) shifted to 37°C for 4 h in YPD and prepared for EM. Bipolar spindle with two SPBs (1, 2). Bar: 500 nm. B-E. Asynchronous sfi1-C4A (JA249) cells shifted to 37°C for 9 h in YPD and prepared for EM. Serial sections were examined by EM for 18 cells. B. Representative cell (n=11) with two SPBs at one position. SPBs (1, 2) are at abnormal orientations to one another with a bridge. Bar: 100 nm. C-E. Multiple sections from the same cell are shown. Left panel(s): full or portion of nucleus, 500 nm bar. Right panels: SPB(s), 100 nm bar. C. Representative cell (n=2) containing a short bipolar spindle with SPB 1 at an orientation different from that of abnormally large SPB 2. D-E. Representative cells (n=5) containing a short bipolar spindle with extra SPBs. Average distance between adjacent SPBs: 108±10 nm. D. Cell with 3 SPBs (n=4). SPB 1 is located at a region of the nuclear envelope opposite SPBs 2 and 3, which are connected via a bridge. Left panel shows the location of SPB 1, shown in the upper right panel at higher magnification, and SPBs 2 and 3. In the lower right panel, both SPBs 2 and 3 are clearly seen in a higher magnification view from a different serial section of the same cell. E. Cell with 4 SPBs (n=1). Adjacent SPBs 1 and 2 are located at a region of the nuclear envelope opposite adjacent SPBs 3 and 4. 86

Table 3-4. sfi1-C3A SPB measurements via EM Phenotype SPB ∆SPB Bridge Spindle diameter diameter: length / length (nm) Opposing Distance (nm) poles (nm) between SPBs(nm) Side-by-side SPBs (n=8) 171±28 144±19(nm)

Two abnormal SPBs at one pole 231±25 (n=7) Bipolar spindle with enlarged 262±13 51±44 2,248±611 SPBs (n=3) Bipolar spindle with reduplicated SPB (n=3) Paired SPB 228±56 148±22

Reduplicated SPB 78±18

Single SPB at one pole 202±64 92±50 2,602±434

∆SPB: Paired - 150±37 Reduplicated SPBs (nm)

Table 3-5. sfi1-C4A SPB measurements via EM Phenotype SPB ∆SPB Bridge Spindle diameter diameter: length / length (nm) Opposing Distance (nm) poles (nm) between SPBs(nm) Side-by-side SPBs (n=1) 130±39 107±4

Two abnormal SPBs at one pole 189±73 110±16 (n=10) Bipolar spindle with enlarged 228±59 75±72 1,486±399 SPBs (n=2) Bipolar spindle with reduplicated 2,257±605 SPB (n=5) Paired SPB 199±64 108±10 (n=5) Reduplicated SPB 106±36 (n=5) Single SPB at one pole 199±111 72±75

∆SPB: Paired - 93±48 Reduplicated SPBs (nm) (n=5 pairs) 87 and Tables 3-4 and 3-5 for sfi1-C3A and sfi1-C4A, respectively) (Winey et al., 1991; Li et al.,

2006), we conclude that an aberrant reduplication event has occurred.

In cells in which SPB reduplication occurs, SPB duplication proceeds multiple times within a single cell cycle. SPB separation is required for reduplication (Simmons Kovacs et al.,

2008), so it is unlikely the sfi1-C4A mutants proceed to a second cell cycle with unseparated

SPBs. It is also unlikely that the extra SPB is present in these cells via SPB missegregation, as seen in a ts Separase mutant (esp1-1), in which a spindle defect occurs in the first cell cycle at restrictive temperature. The daughter cell then receives both SPBs from the mother cell, and these SPBs in the esp1-1 daughter cell then undergo duplication, resulting in extra SPBs in a cell in the second cell cycle (Baum et al., 1988; McGrew et al., 1992). While additional SPBs in esp1 cells appear normal structurally via EM (Baum et al., 1988; McGrew et al., 1992), in sfi1-

C3A and sfi1-C4A, the additional reduplicated SPB differs from the other SPBs within the cell.

Specifically, the reduplicated SPB has a smaller diameter by EM (ΔSPB diameter: 150±37 for sfi1-C3A, n=3; 93±48 for sfi1-C4A, n=5; Figs. 3-26C and 3-27, D and E and Tables 3-4 and 3-5, respectively) and lower intensity of Spc42-GFP by SIM (Fig. 3-25, B and D-F and legend) compared to its paired SPB at the same position in the nuclear envelope. Therefore, it appears that SPB reduplication has occurred.

III. Discussion

Cdk1 phosphorylates the Sfi1 C terminus at sites within Cdk motifs in vitro

This work, in combination with previous research (Elserafy et al., 2014), showed that 6 residues within Cdk consensus motifs at (T816, S855, S882, S892, S923) or immediately preceding (S801) the Sfi1 C terminus are phosphorylated by Clb2/Cdk1 in vitro. All but one of 88 these sites (S801) are known to be phosphorylated in vivo (Chi et al., 2007; Keck et al., 2011).

No previous research to our knowledge has identified a peptide containing S801 in vivo, so this peptide is an excellent candidate to be phosphorylated in vivo as well. Previous research has shown that S-phase cyclin Clb5/Cdk1as1 (an analog-sensitive allele of Cdk1) is also able to phosphorylate the Sfi1 C terminus, but to a lesser degree than mitotic cyclin Clb2/Cdk1as1. They additionally showed that a C-terminal region of Sfi1 interacts via yeast two-hybrid with mitotic cyclins Clb2, 3 and 4. Thus, multiple cyclins likely associate with Cdk1 to phosphorylate the

Sfi1 C terminus. Importantly, Schiebel and colleagues showed that converting all six identified

Cdk1 sites to nonphosphorylatable residues led to a complete inability of Cdk1 (in association with Clb2 or Clb5) to phosphorylate this altered Sfi1 C-terminal region in vitro (Elserafy et al.,

2014). This suggests that all the key sites at the C-terminal region of Sfi1 phosphorylated by

Cdk1 have been identified.

Alteration of Cdk1 phosphorylation sites leads to varying growth defects

I constructed a mutant allele collection through alteration of the eight phosphorylation sites at the Sfi1 C terminus, including the five Cdk1 sites (T816A S855A S882A S892A S923A).

I also examined the effects of altering the in vitro Cdk1 site immediately upstream of the C terminus (S801). I saw differential growth depending on the specific site and combination of residues altered, demonstrating the importance of specific phospho-sites and of combinatorial control of phosphorylation. In particular, I found that only the sites within Cdk consensus motifs, and not those outside of these motifs, were important for cell growth. Similarly, a recent study found that sfi1-T876A and sfi1-T876D showed no growth defects. However, they did find that a combination of non-Cdk site mutations led to a portion of cells displaying single SPBs. 89

Specifically, this was seen in sfi1-S826D T866D T876D. They showed that these sites are phosphorylated in vitro by the polo-like kinase Cdc5; however, only T876 is known to be phosphorylated in vivo (Elserafy et al., 2014).

Alteration of individual residues within Cdk1 consensus motifs consistently resulted in a temperature-sensitive growth defect at 37°C (sfi1-S855A) or no defect (sfi1-T816A, sfi1-S882A, sfi1- S892A, sfi1- S923A). While Schiebel and colleagues saw no growth defect in sfi1-S855A

(sfi1Cdk1-1A) (Elserafy et al., 2014), Hardwick and colleagues found that sfi1-S855N (sfi1-120) leads to a conditional temperature sensitive growth defect. While Schiebel and colleague’s results concerning growth of sfi1-S855A differ from mine, the use of different allele integration and assessment techniques were used, in which they integrated sfi1 alleles at the LEU2 locus and assessed the growth of mutants by the loss of a rescuing SFI1 plasmid (Elserafy et al., 2014).

While I performed this same technique, and did see defects, I also examined the sfi1-S855A phenotype via integration at the SFI1 locus.

Combinations of mutations resulted in varying phenotypes, depending on the residue and number of sites mutated, with more mutations often correlating with a more severe growth phenotype. It is interesting to note that conversion of the site within the minimal Cdk consensus motif (S923A) in combination with sfi1-C3A or sfi1-C4A does not enhance severity of the growth phenotype. sfi1-C3A and sfi1-C4A both showed growth defects at 24°C and little or no growth at 37°C. sfi1-C5A showed only a slightly more severe growth defect at 24°C than sfi1-

C4A, and while sfi1-C6A integrated at the LEU2 locus displayed a similar phenotype to sfi1-

C4A+S923A on rich media, it was lethal when grown on a more stringent media (5-FOA), and no transformants of sfi1-C6A at the SFI1 locus were produced. Together, these results suggest that a progressive reduction in Cdk phosphorylation of Sfi1 leads to greater defects, and the loss of 90 all Cdk phosphorylation at and immediately preceding the Sfi1 C terminus (confirmed by

(Elserafy et al., 2014)) may be lethal. Indeed, a previous study found that sfi1-C6A (sfi1Cdk1-6A; sfi1-S801A T816A S855A S882A S892A S923A) and sfi1Cdk1-6D were lethal when integrated at the LEU2 locus upon loss of wild-type SFI1 on 5-FOA plates. My work and that of Schiebel and colleagues (Elserafy et al., 2014) would suggest that all the key targets of Cdk on Sfi1 required in the SPB duplication process have been identified.

Multiple sfi1 C terminus mutant alleles spontaneously diploidize and show decreased viability

Several of the more severe sfi1 alleles diploidized at 24°C. Diploidization or tetraploidization was also seen in mutant alleles of the genes encoding the other three half-bridge proteins, Cdc31, Kar1, and Mps3 (Schild et al., 1981; Rose and Fink, 1987; Jaspersen et al.,

2002). The complete mechanism by which this diploidization occurs, however, is not clear. It is noted that upon addition of a GFP C-terminal tag on SPC42, the same sfi1 strains that typically diploidize upon germination remain haploids. The reasoning for this is not clear but does suggest that tagging Spc42 affects either Sfi1 function specifically or the general process that results in increases in ploidy.

Several sfi1 alleles also showed decreased viability at higher temperatures. The decrease in viability in these cells may suggest a cell lysis defect. This is not unexpected, as some cdc31 alleles display a cell lysis defect, with the lack of an intact plasmid membrane (Ivanovska and

Rose, 2001).

Cdk1 phosphorylation of Sfi1 is required for appropriate bipolar spindle assembly and chromosome segregation

sfi1-C4A and sfi1-C3A cells both display two main classes of phenotypes, indicating multiple execution points: two SPBs at one position or a bipolar spindle with abnormal, either in size or number, SPBs. The cells with two SPBs at one position appear to fail or stall in SPB separation. If the SPBs do separate, and a bipolar spindle is formed, the SPBs at the poles are aberrant. These defects appear to lead to the observed chromosome segregation defects.

Based on these defects, it could be expected that sfi1-C3A and sfi1-C4A would arrest in

G2/M due to activation of the spindle assembly checkpoint (SAC), which ensures that chromosomes are attached correctly to the mitotic spindle at metaphase before progression to anaphase (Musacchio and Hardwick, 2002). However, no significant increase in large-budded cells, indicative of a G2/M arrest, was seen in either mutant. Nonetheless, a trend for more large-budded cells in sfi1-C4A cells was seen, and SAC activation cannot be ruled out for either mutant. Indeed, previous studies have found that sfi1 mutants with altered C-terminal Cdk sites show synthetic lethality with strains deleted for a component of the SAC (mad1∆, bub1∆, mad3∆, or mad2∆) (Anderson et al., 2007; Elserafy et al., 2014). Future examination of whether sfi1-C3A and sfi1-C4A show synthetic growth defects with mutants of the SAC would prove useful.

Cdk1 phosphorylation of Sfi1 is required for SPB separation

Consistent with my findings, previous research has shown that SPB separation requires

B-type cyclins/Cdk1 (Fitch et al., 1992; Lim et al., 1996; Haase et al., 2001) and Cdk1 phosphorylation of the Sfi1 C terminus (Anderson et al., 2007; Elserafy et al., 2014). Previous 92 work has shown that sfi1-S855N (sfi1-120) cells arrested with side-by-side SPBs, indicating a defect in SPB separation (Anderson et al., 2007). Additionally, a previous study mutated the prolines within these Cdk motifs (sfi1-1; P817S P893S) and found that the synthetically lethal double mutants sfi1-1 bik1Δ, displayed side-by-side SPBs, indicating a defect in SPB separation

(Strawn and True, 2006). Schiebel and colleagues found that mutating six Cdk sites (sfi1Cdk1-6A; sfi1-S801A T816A S855A S882A S892A S923A) to nonphosphorylatable residues led to a block in

SPB separation (Elserafy et al., 2014). However, they did not see cells with reduplicated SPBs, whereas I did see reduplicated SPBs when only three or four of these sites were converted to alanines. Multiple reasons could explain this difference. Their study used a mutant allele in which sfi1Cdk1-6A was integrated at the LEU2 locus, and they induced the degradation of wild- type SFI1 in their experimental procedures. Secondly, they only analyzed synchronized cells at one timepoint; whereas I examined an asynchronous population of cells. Nonetheless, both my work and the Schiebel lab found that Sfi1 C terminus phosphorylation is required for SPB separation.

Schiebel and colleagues suggest that the mitotic cyclin Clb4 specifically, in association with Cdk1, may be required for Sfi1 phosphorylation to permit SPB separation. They showed that the Sfi1 C terminus shows a yeast two-hybrid interaction with Clb4 (Elserafy et al., 2014), and Clb4 localizes to the cytoplasmic side of the SPB during S phase (Maekawa and Schiebel,

2004), which correlates with the timing of SPB separation. S-phase cyclins (Clb 5, 6) may also be involved in phosphorylation of Sfi1 for SPB separation, as S-phase cyclins are proposed to promote separation (Haase et al., 2001).

Cdk1 phosphorylation of Sfi1 is required to block SPB reduplication 93

The second phenotypic class displayed by sfi1-C3A and sfi1-C4A cells is a bipolar spindle with abnormal, either in size or number, SPBs. In sfi1-C3A and sfi1-C4A, one subset of cells displays bipolar spindles with at least one abnormally large SPB. The mechanism for incorporation of SPB components at mitosis is not yet well known. However, it is known that

SPB size and component organization are cell cycle regulated (Byers and Goetsch, 1974; Yoder et al., 2003). Since there is a trend for sfi1-C4A cells to be large-budded, a portion of these cells may arrest or stall in mitosis. Thus, additional incorporation of SPB components, as seen in

CDC20 depletion (O’Toole et al., 1997), upon a prolonged mitosis could occur to result in enlarged SPBs. It could also be hypothesized that if cells with this phenotype arrest at mitosis via activation of the SAC, they may proceed to form extra SPBs, as discussed below, if assessed at a later timepoint.

A key finding is that some cells with bipolar spindles contain reduplicated SPBs, in which it appears premature licensing of SPB duplication may have occurred. The EM data is consistent with the SIM results and indicates that loss of Cdk1 phosphorylation of Sfi1 leads to

SPB reduplication. Specifically, in cells with reduplicated SPBs, as seen previously in cells in which all mitotic cyclins are deleted (Haase et al., 2001), each extra SPB has duplicated using one of the original SPBs present in the spindle as a “template” during a single cell cycle. Thus, the nonphosphorylatable sfi1 mutations lead to premature licensing of SPB duplication during mitosis. It has previously been shown that mitotic cyclin/Cdk1 activity is required to block reduplication (Haase et al., 2001), which is consistent with a role for phosphorylation of Sfi1 by

Cdk1 in blocking reduplication. Excitingly, this now points to Sfi1 as the first target of Cdk1 in blocking reduplication. 94

It is interesting to note that cells with an SPB reduplication event typically had one reduplicated SPB (three SPBs total). It will be of future interest to determine whether there exists a preference for reduplication of one of the two SPBs. In particular, whether the new or old SPB is preferentially reduplicated could be examined. Previous work has shown that the

RFP-labeled SPB components Spc42 and Spc110 show no or less fluorescence in the new versus the old SPB (Pereira et al., 2001; Yoder et al., 2003), as RFP has slow folding properties (Knop et al., 2002). Using this technique, they found that in wild-type cells, the old SPB is distributed to the bud, while the new SPB remains in the mother (Pereira et al., 2001; Yoder et al., 2003). In sfi1-C3A and sfi1-C4A cells, examination of fluorescent intensity levels would be a useful method to determine old versus new SPB. In addition to cells with one reduplication event, cells with two reduplicated SPBs (four SPBs total) were also present in sfi1-C4A. Thus, it is predicted that, with additional time at restrictive temperature and/or in mitotic arrest, both original SPBs may be reduplicated, resulting in four SPBs total.

Mimicking Cdk1 phosphorylation of Sfi1 is expected to block G1 SPB duplication

Since I predict that a lack of phosphorylation on the Sfi1 C-terminal Cdk sites is required for half-bridge elongation, it is expected that a phosphomimetic allele in which the Sfi1 C- terminal Cdk1 sites are altered would arrest with a single SPB with a short half-bridge.

However, multiple spindle morphology phenotypes were seen upon examination of three phosphomimetic alleles, sfi-C3E, sfi1-C4E, and sfi1-C4E+S923E. Closer examination of sfi1-

C4E via EM revealed a cell with a monopolar spindle, as expected, but the following phenotypes were also seen: duplicated side-by-side SPBs and a short bipolar spindle with one SPB per pole.

The latter two phenotypes have also been seen in the corresponding nonphosphorylatable allele, 95 sfi1-C4A, and do not suggest a defect in the initial step of SPB duplication. Thus, these three alleles may actually fail to mimic phosphorylation and instead simply mimic loss of a phosphorylatable residue at the altered sites.

However, the phosphomimetic alleles did show differences compared to the corresponding nonphosphorylatable alleles in growth (for sfi1-C3E), dominance (for all alleles), and ploidy (for sfi1-C3E). Also, a monopolar spindle with a single SPB was seen in sfi1-C4E, but not sfi1-C3A or sfi1-C4A. Additionally, unlike sfi1-C3A and sfi1-C4A, no significant increase in short metaphase bipolar spindles was seen in sfi1-C3E and sfi1-C4E compared to control, and no short spindles were seen in sfi1-C4E+S923E at 37°C (from one experiment).

However, all alleles with conversion of phosphorylated residues to alanine or glutamic acid displayed predominantly single microtubule asters, except sfi1-C3E, which predominantly had anaphase bipolar spindles or separated SPBs. It is interesting to note that, unlike sfi-C3E, sfi1-

C4E, and sfi1-C4E+S923E, the phosphomimetic allele sfi1-C6D (sfi1Cdk1-6D; sfi1-S801D T816D

S855D S882D S892D S923D) examined by Schiebel and colleagues showed only single SPBs by

EM (n=9) (Elserafy et al., 2014). However, these authors only examined synchronized cells at a short timepoint after release from a G1 arrest. Thus, one cannot rule out the possibility of additional phenotypes when examined using a different experimental strategy. Overall, examination of an allele in which both the phosphorylated residue (T816 S855 S882 S892) as well as the adjacent proline within the Cdk consensus motif are altered to glutamic acid, sfi-C8E, would better mimic the -2 charge state resulting from phosphorylation and thus is best to further pursue (Strickfaden et al., 2007).

IV. Materials and Methods 96

Strain Construction

The W303 strain background was used for all experiments (Table 3-1; except select strains used for genetic crosses). Yeast transformation (Gietz and Woods, 2002) with pRS402 was used to create ADE2 strains where indicated in the strain list. Standard PCR mutagenesis, including use of the Quikchange II mutagenesis kit (Agilent) by the Stowers Institute Molecular

Biology facility, was used to create all mutations within the SFI1 sequence using a pRS305 vector base. All alleles integrated at the LEU2 locus contain silent mutations for SFI1 nucleotide changes of T-195G (upstream SFI1 region), C99G, A394T, G395C, A2130G, and T2131C in order to develop additional restriction enzyme sites within the Sfi1 sequence for plasmid manipulation purposes. A nucleotide change of 2804 A to G is found in the SFI1 sequence obtained from John Kilmartin versus Sanger sequencing, for an amino acid change of E935G

(Kilmartin, 2003). Nucleotide changes for the mutations in SFI1 are as follows (amino acid change: nucleotide change): 801A: T2401G; 816A: A2446G; S855A: A2563G G2564C; T876A:

A2626G G2628T; T877A: A2629G; S882A: T2644G; S892A: T2674G; S920A: A2758G

G2759C C2760T; S923A: T2767G; S939A: T2815G; 816E: A2446G C2447A T2448A; S855E:

A2563G G2564A T2565A; T876E: A2626G C2627A G2628A; T877E: A2629G C2630A;

S882E: T2644G C2645A T2646A; S892E: T2674G C2675A T2676A; S920E: A2758G G2759A

C2760A; S923E: T2767G C2768A T2769A.

The SFI1 wild-type and control (containing the described silent mutations) sequences from the appropriate pRS304-KANMX plasmids (E2225 and E2249, respectively, cut with

BamHI and XhoI; see Chapter 2) were placed into the pRS305 vector base, and the resultant plasmids (E2322, E2320, respectively) were digested with BsrGI and integrated at the LEU2 locus in sfi1∆::KANMX YCp-SFI1-URA3 MATa (JA54). The resultant strains were leu2::SFI1- 97

LEU2 sfi1∆::KANMX YCp-SFI1-URA3 (wild-type, JA1-001; control, JA272). The same integration technique was used for integration of all sfi1 alleles, obtained via mutations of E2320, at the LEU2 locus. All strains in chapter except JA1-001, JA54, JA152, JA158, and ploidy test strains contain silent mutations.

Integration of a single copy of the alleles at the LEU2 locus was confirmed via southern blot as previously described (Brown, 2004) using Amersham gene images AlkPhos direct labeling and CDP-Star detection system (GE Healthcare Life Sciences) with genomic DNA (isolated as described in Amberg et al., 2005) digested with BglII. The following two single-stranded DNA probes were used: a LEU2 probe (498 bp) constructed from primers LEU386F (P103) LEU884R

(P106) and an SFI1 probe (491 bp) constructed from primers SFI1372F (P107) and SFI1863R

(P108) from pRS305-SFI1 (E2320). All alleles were confirmed as single integrants except JA1-

002 and JA1-013, which contain two integrants.

Creation of single integrant mutants with the NATMX marker (integrated 43 bp downstream of the SFI1 sequence) via PCR is as previously described (Tong and Boone, 2006), with integration at the SFI1 locus in a wild-type diploid (JA144) and SFI1 containing only silent mutations from nucleotide changes A2130G and T2131C (residues E710 and L711).

Heterozygote diploids were then dissected, and the appropriate spore containing the NATMX marker and mutation in SFI1 was obtained. To amplify the sfi1 region from the appropriate pRS305-sfi1 plasmids, primers SFI11686F (P92) and MXSFI1+42R (P110) were used. To amplify the

NATMX region from the pNATMX plasmid (obtained from Sue Jaspersen), primers MXF (P94) and

SFI1+43DSMXR (P111) were used. Integration of a single copy of the sfi1 allele at the SFI1 locus was confirmed for JA158, JA162, and JA180 via southern blot. The DNA probe (699 bp) was constructed from primers Seq2SFI1F (P97) and STOPSFIR (P95), and the genomic DNA was digested with HindIII and XhoI. Integration of a single copy of the sfi1 allele at the SFI1 locus 98 was confirmed for JA249 and JA270 via PCR using primers NATMXF (P165) and SFI12307R

(P127).

The technique described (Tong and Boone, 2006) to create single integrant sfi1 mutants with the NATMX marker via PCR was used to create N-terminally tagged GFP-Sfi1 strains by transforming trp1::N-GFP-SFI1 TRP1 sfi1Δ::HIS5 (gift of John Kilmartin) (Kilmartin, 2003).

The CEN-based plasmid pLEU2-SPC42-mCherry-HIS was then transformed into these strains and maintained by growth in SC-LEU media.

C-terminal tagging of SPC42 with yeGFP1 using pFA6-yeGFP1-KANMX-HISMX

(E2642) via PCR was performed as previously described (Longtine et al., 1998; Sheff and Thorn,

2004). Specifically, SFI1-NATMX SPC42-GFP-KANMX ade2::ADE2 MATa (JA254) was a spore from dissection of SFI1-NATMX/SFI1-NATMX SPC42-GFP-KANMX/SPC42 ade2::ADE2/ade2::ADE2 MATa/MATα (JA253). Transformation of the yeGFP1 PCR product into SFI1-NATMX/SFI1-NATMX ade2::ADE2/ade2::ADE2 MATa/MATα (JA196) was performed to obtain JA253. sfi1-C4A-NATMX SPC42-GFP-KANMX ade2::ADE2 MATa

(JA302) was obtained by dissection of sfi1-C4A-NATMX/SFI1 SPC42-GFP-KANMX/SPC42 ade2::ADE2/ade2::ADE2 MATa/MATα (JA1-030). JA1-030 was obtained by transformation of the yeGFP1 PCR product into sfi1-C4A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα

(JA235, JA236).

The strains sfi1-S801A and sfi1-C5A integrated at the endogenous locus with a NATMX marker were created as described above. For leu2::sfi1-C6A-LEU2 sfi1∆::KANMX YCp-SFI1-

URA3 (JA312, JA313) and leu2::sfi1-C4A+S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3

(JA314, JA315), integration of a single allele was confirmed via PCR, and sequencing confirmed the mutation. Strains were backcrossed to JA146, and a haploid spore (lacking YCp-SFI1- 99

URA3) from dissection of the resultant heterozygous diploid was obtained (JA322, JA324).

Heterozygous diploids of sfi1 alleles integrated at the LEU2 locus were used to examine sfi1-

C6A. Strains with (+) or without (-) a rescuing SFI1 plasmid (YCp-SFI1): sfi1-C6A/SFI1: JA325

(-; JA312 x JA146), JA326 (+; JA313 x JA146); sfi1-C4A+S923A/SFI1: JA1-024 (-; JA314 x

JA146), JA333 (+; JA315 x JA146); SFI1 Control/SFI1: JA1-025 (-; leu2::SFI1-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 x JA146), JA334 (+; leu2::SFI1-LEU2 sfi1∆::KANMX YCp-

SFI1-URA3 x JA146).

The strains used for genetic crosses are from the following sources: H111αα from

Hawthorne, H243-13-2αα from Culbertson, WX266-2d, WX257-4C, and WX257-8b from Mark

Winey, w303 1b from Pillus, and Y3912 from Michele Jones. sfi1-C3A/sfi1-C3A MATa/MATa

(JA188) was isolated from JA180, which showed mixed ploidy.

Five-fold dilution series with 5 µL of the same initial OD/mL in the left column were placed on plates and grown at the indicated temperatures for 2 days on YPD and 4 days on 5-

FOA unless otherwise indicated. For temperature shift experiments of asynchronous cultures, cultures were shifted at early-log phase. All dilution series shown contain the control with the silent mutation unless otherwise indicated. Growth defects that are present only early in growth

(e.g., YPD Day1, 5-FOA prior to day 4) are not included in Table 3-1. To assess cell viability, the vital stain trypan blue (Sigma-Aldrich) was used.

Flow cytometry

Cells were fixed with 70% ethanol for 1 h at room temperature or at 4°C overnight.

Fixed cells were then treated with 0.1% RNase in 0.2M Tris-HCl pH 7.5, 20 mM EDTA for 2 h at 37°C, and the DNA was stained with 50 µg/mL propidium iodide (Sigma Chemical Co., St.

Louis, MO) in PBS for 1 h at room temperature or at 4°C overnight. DNA content was analyzed 100 using the CyAn ADP analyzer (Beckman Coulter, Indianapolis, IN, USA) with a 488nm laser.

30,000 events per sample were taken. Unless otherwise indicated, DNA content (propidium iodide area) is indicated on X axes and Count on Y axes. For DNAse treatment, 10 units DNAse was added to the sample for 1 h at room temperature in PBS/DNase reaction buffer following

RNAse treatment.

Cytological techniques

Imaging of live and fixed cells was performed at room temperature on an Eclipse Ti inverted microscope (Nikon, Japan) fitted with a CFI Plan Apo VC 60× H numerical aperture 1.4 objective (Nikon, Japan) and a CoolSNAP HQ2 charge-coupled device camera (Photometrics,

Tuscon, AZ). Metamorph Imaging software (Molecular Devices, Sunnyvale, CA) was used to collect images, and maximum projections are shown. The 1.5X intermediate magnification was used for fixed immunofluorescent images. Quantification of cell size was performed on immunofluorescent images using ImageJ software (National Institutes of Health, Bethesda, MD,

USA).

For immunofluorescence, cultures were fixed with 4% formaldehyde for 45 minutes, subjected to zymolyase and prepared on slides. YOL1/34 rat anti-tubulin antibody (1/150), FITC goat anti-rat secondary antibody (1/200), and Hoescht dye for DNA were used. For imaging of live GFP cells, cells were briefly centrifuged and washed and imaged in 1x PBS. For brief fixation of GFP cells, cultures were fixed with 3.7% formaldehyde for 15 minutes and resuspended and imaged in 1xPBS or KPO4/sorbitol.

SIM images were acquired at the Stowers Institute for Medical Research at room temperature on an Applied Precision OMX Blaze microscope (Issaquah, WA, USA) equipped with a PCO Edge sCMOS camera (Kelheim, Germany). The objective used was an Olympus 101

(Center Valley, PA, USA) 60x 1.42NA Plan Apo N oil objective. Image stacks were acquired at

125 nm intervals. SIM reconstruction was performed with the Applied Precision software package utilizing optical transfer functions measured with 100 nm green fluospheres in Prolong

Gold on a coverslip surface (Life Technologies, Carlsbad, CA, USA) following the Applied

Precision protocols. After reconstruction, SIM images were scaled 2 by 2 with bilinear interpolation through ImageJ software for future quantification.

Quantification of GFP intensities in live and fixed GFP cells was performed with

Metamorph imaging software using maximum projections, in which the average pixel intensity was determined for the fluorescent focus and subtracted from the average pixel intensity for the immediate background border. Distances were measured using ImageJ software. GFP intensity

(arbitrary units; a.u.) was averaged for each strain at each temperature, and the average of this value from two experiments for each condition was taken. A Student’s t test was then performed on the average from both experiments.

Transmission electron microscopy

Log phase cells were high pressure frozen in a Wohlwend Compact 02 HPF and freeze- substituted in 2% osmium tetroxide, 0.1% uranyl acetate in acetone and embedded in Spurr’s epoxy (sfi1-C3A) or in 0.25% glutaraldehyde, 0.1% uranyl acetate in acetone and embedded in

Lowicryl HM20 (sfi1-C4A and sfi1-C4E strains) (Giddings et al., 2001). Imaging was conducted using a FEI Phillips CM100 electron microscope.

Distances in electron micrographs were measured using ImageJ software. Enlarged SPBs were classified as >213.28 nm diameter, which is >33.33% deviance from the predicted average diameter of 160nm in a diploid (Byers and Goetsch, 1974).

Protein Techniques and kinase reaction 102

Sfi1 residues 800 to 946 were amplified via PCR from pRS305-SFI1, digested with NheI-

HF and cloned into pKLD116 plasmid (digested with StuI and NheI-HF; gift of Ivan Rayment) immediately downstream of the rTEV site and in frame with the N-terminal 6xHis and maltose binding protein tags. The resulting plasmid was transformed into C+ (DE3) competent E. coli cells (gift of Greg Odorizzi). IPTG induction was performed using 0.3 mM IPTG for 2.5 h at

23°C. The sample was flash frozen, treated with lysozyme, sonicated, and column purification using Talon metal affinity resin (Clontech Laboratories, Inc.) was performed using 200 mM imidazole for elution. An in vitro kinase reaction was performed using 1 µg of the Sfi1 C terminus fusion protein (6xHis-MBP-rTEVsite-Sfi1C), 1 mM ATP, and Clb2 (purified from bacteria)/Cdc28 (purified from baculovirus co-infected with the Cdk activating kinase (CAK))

(Jaspersen et al., 2004).

Mass spectrometry

The 20 µL kinase reaction was digested with 2 µg GluC (no denaturation nor reduction/alkylation) and incubated at room temperature for 1 h. 2 µL or 10% of the reaction was loaded directly onto a Waters nanoAcquity 75 µm X 250 mm 1.7 µm BEH130 C18 column

(no trapping nor desalting). Peptides were eluted with a gradient from 8% acetonitrile, 0.1% formic acid to 32% acetonitrile, 0.1% formic acid at a flow rate of 300 nL/min. Spectra were searched using Mascot v2.2 (Matrix Science) against a small custom database with the protein sequence of the recombinant Sfi1 fusion protein included. Phosphorylated peptides were identified by manual inspection of all MS/MS spectra with Mascot ions scores of at least 20.

Multiple phosphorylated isoforms were identified as possible if the delta score was less than 4

(difference between the top two scoring phosphorylated positional isomers).

103

Table 3-6. Yeast strains

Strain Genotype Integration at LEU2 locus JA54 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-001 leu2::SFI1-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-002 leu2::sfi1-T816A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-003 leu2::sfi1-S882A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-004 leu2::sfi1-S892A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-005 leu2::sfi1-S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-006 leu2::sfi1-T816A S882A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-007 leu2::sfi1-T816A S892A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-008 leu2::sfi1-T816A S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-009 leu2::sfi1-S882A S892A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-010 leu2::sfi1-S882A S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-011 leu2::sfi1-S892A S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-012 leu2::sfi1-C3A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-013 leu2::sfi1-T816A S892A S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-014 leu2::sfi1-T816A S882A S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-015 leu2::sfi1-S822A S920A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-016 leu2::sfi1-T816A S882A S920A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-017 leu2::sfi1-T876A T877A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-018 leu2::sfi1-876A 877A S920A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-019 leu2::sfi1-876E 877E-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-020 leu2::sfi1-876E 877E S920E-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-021 leu2::sfi1-T816A T876A S892A S920A-LEU2 sfi1∆::KANMX YCp-SFI1- URA3 MATa JA1-022 leu2::sfi1-T816A S892A S920A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA1-023 leu2::sfi1-C3E+S923E-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA272 leu2::SFI1-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA273 leu2::sfi1-C3A+S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA274 leu2::sfi1-S855A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA275 leu2::sfi1-C4A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA276 leu2::sfi1-T816A S855A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA277 leu2::sfi1-S855A S882A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa 104

JA279 leu2::sfi1-S855A S892A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA280 leu2::sfi1-S855A S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA281 leu2::sfi1-T816A S855A S882A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA282 leu2::sfi1-T816A S855A S892A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA283 leu2::sfi1-T816A S855A S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA284 leu2::sfi1-S855A S882A S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA285 leu2::sfi1-S855A S892A S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA286 leu2::sfi1-S855A S882A S892A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA312, JA313 leu2::sfi1-C6A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA314, JA315 leu2::sfi1-C4A+S923A-LEU2 sfi1∆::KANMX YCp-SFI1-URA3 MATa JA322 leu2::sfi1-C6A-LEU2/leu2::sfi1-C6A-LEU2 sfi1∆::KANMX/sfi1∆::KANMX YCp-SFI1-URA3 MATa/MATa JA324 leu2::sfi1-C4A+S923A-LEU2/ leu2::sfi1-C4A+S923A-LEU2 sfi1∆::KANMX/sfi1∆::KANMX YCp-SFI1-URA3 MATa/MATa JA325 leu2::sfi1-C6A-LEU2/leu2 sfi1∆::KANMX/SFI1 ade2::ADE2/ade2 MATa/MATα JA326 leu2::sfi1-C6A-LEU2/leu2 sfi1∆::KANMX/SFI1 ade2::ADE2/ade2 YCp- SFI1-URA3 MATa/MATα JA1-024 leu2::sfi1-C4A+S923A-LEU2/leu2 sfi1∆::KANMX/SFI1 ade2::ADE2/ade2 MATa/MATα JA333 leu2::sfi1-C4A+S923A-LEU2/leu2 sfi1∆::KANMX/SFI1 ade2::ADE2/ade2 YCp-SFI1-URA3 MATa/MATα JA1-025 leu2::SFI1-LEU2/leu2 sfi1∆::KANMX/SFI1 ade2::ADE2/ade2 MATa/MATα JA334 leu2::SFI1-LEU2/leu2 sfi1∆::KANMX/SFI1 ade2::ADE2/ade2 YCp- SFI1-URA3 MATa/MATα Integration at SFI1 locus JA144 ade2::ADE2/ade2::ADE2 MATa/MATα JA152 SFI1-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA154 SFI1-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA156 sfi1-C3A+S923A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA157 sfi1-C3A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA158 SFI1-NATMX ade2::ADE2 MATa JA162 SFI1-NATMX ade2::ADE2 MATa JA180 sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 MATa/MATa JA184 sfi1-C3A+S923A-NATMX ade2::ADE2 MATa 105

JA188 sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 MATa/MATa JA190 sfi1-C2A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA192 sfi1-C2A-NATMX ade2::ADE2 MATa JA196 SFI1-NATMX/SFI1-NATMX ade2::ADE2/ade2::ADE2 MATa/MATα JA217 sfi1-S855A-NATMX ade2::ADE2 MATa JA221 sfi1-S855E-NATMX ade2::ADE2 MATa JA231 sfi1-C3E-NATMX ade2::ADE2 MATa JA233 sfi1-S939A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA235, JA236 sfi1-C4A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA237 sfi1-C3E-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA239 sfi1-S855A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA241 sfi1-S855E-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA243 sfi1-S939A-NATMX ade2::ADE2 MATa JA249 sfi1-C4A-NATMX/sfi1-C4A-NATMX ade2::ADE2/ade2::ADE2 MATa/MATa JA254 SFI1-NATMX SPC42-GFP-KANMX ade2::ADE2 MATa JA260 sfi1-C4E-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA263 sfi1-C4E+S923E-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA264 sfi1-C4A+S923A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA266 sfi1-C4A+S923A-NATMX/sfi1-C4A+S923A-NATMX ade2::ADE2/ade2::ADE2 MATa/MATa JA270 sfi1-C4E-NATMX/sfi1-C4E-NATMX ade2::ADE2/ade2::ADE2 MATa/MATa JA268 sfi1-C4E+S923E-NATMX/sfi1-C4E+S923E-NATMX ade2::ADE2/ade2::ADE2 MATa MATa/MATa JA289, JA290 sfi1-C5A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA287,288 sfi1-S801A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα JA291 sfi1-S801A-NATMX ADE2 MATa JA293 sfi1-C5A-NATMX/sfi1-C5A-NATMX ade2::ADE2/ade2::ADE2 MATa/MATa JA302 sfi1-C4A-NATMX SPC42-GFP-KANMX ade2::ADE2 MATa GFP-Sfi1 strains JA308, JA309 trp1::GFP-sfi1-C3A-NATMX-TRP1 sfi1Δ::HIS5/trp1::GFP-sfi1-C3A- NATMX-TRP1 sfi1Δ::HIS5 pLEU-HIS-SPC42-mCherry MATa/MATa JA310, JA311 trp1::GFP-SFI1-NATMX-TRP1 sfi1Δ::HIS5/trp1::GFP-SFI1-NATMX- TRP1 sfi1Δ::HIS5 pLEU-HIS-SPC42-mCherry MATa/MATα JA353, JA354 trp1::GFP-sfi1-C4A-NATMX-TRP1 sfi1Δ::HIS5/trp1::GFP-sfi1-C4A- NATMX-TRP1 sfi1Δ::HIS5 pLEU-HIS-SPC42-mCherry MATa/MATa Ploidy test strains H111αα ade5/ade5 lys2/lys2 ino1-13/ino1-13 ino4-8/ino4-8 Matα/Matα 106

H243-13-2αα trp1/trp1 ura1/URA1 Matα/Matα WX266-2d ura3-52/ura3-52 trp1∆1/trp1∆1 his3∆200/his3∆200 leu2-3,112/leu2- 3,112 Matα/Matα WX257-4C ura3-52 leu2-3,112 trp1∆1 Matα WX257-8b ura3-52 leu2-3,112 trp1∆1 his3∆200 Matα w303 1b leu2-3,112 trp1-1 can1-100 ura3-1 ade2-1 his3-11,15 Matα Y3912 tub4∆::TUB4-KANMX-TRP1 Matα JA67 cdc31∆::CDC31-KANMX-TRP1 Matα JA146 ade2::ADE2 Matα

107

Table 3-7. Plasmids

Identifier Name Genotype E697 pRS402 ADE2 E2330 pNATMX NATMX E2595 pKLD116 6xHIS-MBP-TEVsite E2603, E2693, E2712, pKLD116-SFI1-C 6xHIS-MBP-TEVsite-SFI1-C E2713 E2642 pFA6-yeGFP1-KANMX-HISMX yeGFP1-KANMX-HISMX pSJ906 pLEU2-SPC42-mCherry-HIS LEU2-SPC42-mCherry-HIS (CEN) E2322 pRS305-SFI1 SFI1-LEU2 p96SFI1-II-B6 pRS305-sfi1-T816A sfi1-T816A-LEU2 p96SFI1-II-A10 pRS305-sfi1-S882A sfi1-S882A-LEU2 p96SFI1-II-B4 pRS305-sfi1-S892A sfi1-S892A-LEU2 C1 pRS305-sfi1-S923A sfi1-S923A-LEU2 p96SFI1-II-A1 pRS305-sfi1-T816A S882A sfi1-T816A S882A-LEU2 p96SFI1-II-B12 pRS305-sfi1-T816A S892A sfi1-T816A S892A-LEU2 E2415 pRS305-sfi1-T816A S923A sfi1-T816A S923A-LEU2 E2416 pRS305-sfi1-S882A S892A sfi1-S882A S892A-LEU2 p96SFI1-II-G1 pRS305-sfi1-S882A S923A sfi1-S882A S923A-LEU2 E2418 pRS305-sfi1-S892A S923A sfi1-S892A S923A-LEU2 H9 pRS305-sfi1-C3A sfi1-C3A-LEU2 D3 pRS305-sfi1-T816A S892A S923A sfi1-T816A S892A S923A-LEU2 E2416 pRS305-sfi1-T816A S882A S923A sfi1-T816A S882A S923A-LEU2 p96SFI1-II-B10 pRS305-sfi1-S822A S920A sfi1-S822A S920A-LEU2 p96SFI1-II-D2 pRS305-sfi1-T816A S882A S920A sfi1-T816A S882A S920A-LEU2 E2420 pRS305-sfi1-T876A T877A sfi1-T876A T877A-LEU2 E2421 pRS305-sfi1-876A 877A S920A sfi1-876A 877A S920A-LEU2 E2422 pRS305-sfi1-876E 877E sfi1-876E 877E-LEU2 E2423 pRS305-sfi1-876E 877E S920E sfi1-876E 877E S920E-LEU2 p96SFI1-II-G10 pRS305-sfi1-T816A T876A S892A sfi1-T816A T876A S892A S920A S920A-LEU2 p96SFI1-I-A5 pRS305-sfi1-T816A S892A S920A sfi1-T816A S892A S920A-LEU2 pC3E923E pRS305-sfi1-C3E+S923E sfi1-C3E+S923E-LEU2 E2320 pRS305-SFI1 SFI1-LEU2 pC3A923A pRS305-sfi1-C3A+S923A sfi1-C3A+S923A-LEU2 E2641 pRS305-sfi1-S855A sfi1-S855A-LEU2 E2605 pRS305-sfi1-C4A sfi1-C4A-LEU2 E2644 pRS305-sfi1-T816A S855A sfi1-T816A S855A-LEU2 108

E2645 pRS305-sfi1-S855A S882A sfi1-S855A S882A-LEU2 E2646 pRS305-sfi1-S855A S892A sfi1-S855A S892A-LEU2 E2647 pRS305-sfi1-S855A S923A sfi1-S855A S923A-LEU2 E2648 pRS305-sfi1-T816A S855A S882A sfi1-T816A S855A S882A-LEU2 E2649 pRS305-sfi1-T816A S855A S892A sfi1-T816A S855A S892A-LEU2 E2650 pRS305-sfi1-T816A S855A S923A sfi1-T816A S855A S923A-LEU2 E2651 pRS305-sfi1-S855A S882A S923A sfi1-S855A S882A S923A-LEU2 E2652 pRS305-sfi1-S855A S892A S923A sfi1-S855A S892A S923A-LEU2 E2653 pRS305-sfi1-S855A S882A S892A sfi1-S855A S882A S892A-LEU2 E2655 pRS305-sfi1-C4A+S923A sfi1-C4A+S923A-LEU2 E2640 pRS305-sfi1-C3E sfi1-C3E-LEU2 E2639 pRS305-sfi1-S939A sfi1-S939A-LEU2 E2606 pRS305-sfi1-S855E sfi1-S855E-LEU2 E2654 pRS305-sfi1-C4E sfi1-C4E-LEU2 E2656 pRS305-sfi1-C4E+S923E sfi1-C4E+S923E-LEU2 E2676 pRS305-sfi1-S801A sfi1-S801A-LEU2 E2677 pRS305-sfi1-C5A sfi1-C5A-LEU2 E2678 pRS305-sfi1-C6A sfi1-C6A-LEU2

109

Table 3-8. Primers

Name Identifier Sequence LEU38 P103 5’-GGTACTGACTTCGTTGTTGTCAG-3’ 6F LEU88 P106 5’-CAAATCTGGAGCAGAACCGTG-3’ 4R SFI137 P107 5’-GGTTCATTTCAAACTATCCAGATC-3’ 2F SFI186 P108 5’-CCTGATCTGCTAACCTTGACTGC-3’ 3R SFI116 P92 5'-GAGGGTCTCGTAAATGAGTGTCTAGC-3' 86F MXSFI P110 5’-CAGATCTGGCGCGCCTTAATTAACCCGGGGATCCG 1+42R CGACTACATATGCACACATACATACG-3’ MXF P94 5’-CGGATCCCCGGGTTAATTAA-3’ SFI1+4 P111 5’- 3DSM GTTGTGATGTTTTCATAGACCTGGAGAATATTATTAGTAATT XR GTATGCATCTCAGAAGCAAGAAAGGTTAGAATTCGAGCTCG TTTAAAC-3’ Seq2SF P97 5’-GTTCGCGAAGAATTTGTGTTAGTCAAGAC-3’ I1F STOPS P95 5’-CTATTGACGTTTACGACTTAACGGGGA-3’ FIR NATM P165 5’-GCTTCGTGGTCATCTCGTACTC-3’ XF SFI123 P127 5’-CTCGACTGTCTCATCATTACG-3’ 07R

110

Chapter Four: Licensing of SPB duplication requires phosphoregulation of Sfi1

I. Introduction

I and others have shown that the sites within C-terminal Cdk1 consensus motifs phosphorylated in vivo (Chi et al., 2007; Keck et al., 2011) are also phosphorylated in vitro by

Cdk1 (Elserafy et al., 2014). A key finding in my work is that mutating these Cdk sites in sfi1-

C3A and sfi1-C4A results in bipolar spindles containing extra SPBs. I propose that these extra

SPBs are reduplicated SPBs due to premature licensing of SPB duplication. Reduplicated SPBs were seen previously in cells in which all mitotic cyclins are deleted (Haase et al., 2001), suggesting that mitotic cyclin/Cdk1 activity blocks reduplication of the SPB, ensuring appropriate licensing of SPB duplication. A recent study showed that cells with Sfi1 containing phosphomimetic Cdk1 sites arrest with a single SPB, supporting the idea that phosphorylation of

Sfi1 inhibits the process of duplication (Elserafy et al., 2014). However, until my current work, no studies of Cdk1 targets have mimicked the reduplication phenotype revealed by depletion of

Cdk1 activity (Haase et al., 2001). My current work suggests that Sfi1 is the first target of Cdk1 in restricting licensing. It is therefore important to confirm that SPB reduplication, in which SPB duplication occurs multiple times within a single cell cycle, and not SPB missegregation, has occurred. One method in which to do this is to examine sfi1-C4A cells arrested in mitosis, with the expectation that reduplication events would be enhanced within a single cell cycle in mitotically-arrested cells.

In addition to Sfi1 regulation by Cdk1, Sfi1 is dephosphorylated by the protein phosphatase Cdc14 (Bloom et al., 2011; Elserafy et al., 2014), which is activated during anaphase and required for mitotic exit. Cdk1 activity is counteracted by Cdc14. Cdc14 is thought to remove Cdk1-dependent phosphates from a number of targets (Visintin et al., 1998; 111

Stegmeier and Amon, 2004) and has been shown to dephosphorylate Sfi1 (Bloom et al., 2011;

Elserafy et al., 2014). Sfi1 contains four Cdk sites at or immediately preceding the Sfi1 C terminus that are within a Cdc14 consensus motif (S-P-x-K/R; S801, S855, S882, S892)

(Bremmer et al., 2012; Eissler et al., 2014; Elserafy et al., 2014), thus these specific sites are excellent candidates for Cdc14 dephosphorylation. Based on the model of licensing of SPB duplication via Sfi1 phosphoregulation, Cdc14 is an excellent candidate in having a role in licensing via regulation of Sfi1. Specifically, I propose that dephosphorylation of Sfi1 by Cdc14 would license SPBs for duplication.

II. Results sfi1-C4A cells display multiple phenotypes via live timelapse imaging

In order to attempt to confirm that SPB reduplication, and not missegregation, had occurred in sfi1-C4A, I observed the SPB using Spc42-GFP in sfi1-C4A over a time course at

37°C after release from the mating pheromone α-factor-induced G1 arrest. For comparison, I also observed cells mutated in the gene encoding Separase (esp1-1). While esp1-1 displayed multiple phenotypes, a majority of cells displayed a SPB missegregation phenotype, with only the bud, and not the mother cell, containing two Spc42-GFP foci, and after cell division, no GFP foci in the mother cell and multiple GFP foci in the resultant large-budded daughter cell were present (Fig. 4-1, B and E). In contrast, sfi1-C4A either remained as large-budded cells with one

GFP focus or proceeded through cell divisions similar to control cells, though with a much slower cell cycle (Fig. 4-1, A, C-E).

In a separate experiment, I reintroduced α -factor after an initial release from the G1 arrest at 37ºC (Fig. 4-2). As expected, esp1-1 typically had no GFP foci in the mother cell and 112

E 6

# cells # 2

0 normal 1 focus, remain as LB 2 foci, remain as LB 0 foci in mother, 0 foci in daughter, multiple foci in LB multiple foci in LB from daughter from mother

Control esp1-1 sfi1-C4A

Figure 4-1. sfi1-C4A displays multiple phenotypes upon release from G1 arrest that differ from esp1-1 A-D. Cultures containing SPC42-GFP were grown to early- to mid-log phase in YPD at 24°C and arrested in G1 using α-factor. Cells were then released into YNB+CAA media and imaged using a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) in a 37°C chamber. Time indicated is time from release. Bar: 5 µm. A. Sfi1 control cells (JA254) displayed normal cell divisions, with two independent α-factor arrested cells indicated in white and yellow in the left panel. The subsequent cell divisions are tracked with respective colors. See video 1. B. esp1-1 SPC42-GFP (JA365) cells showed a SPB missegregation phenotype, with all GFP foci distributed to the daughter cell, followed by additional SPB duplication in the daughter (see right panel). See video 2. C. sfi1-C4A SPC42-GFP (JA302) cell that appears to divide like a control cell. The initial two cell divisions are tracked. See video 3. D. sfi1-C4A SPC42-GFP (JA302) cell that remains as a large-budded cell with a single GFP focus throughout the remainder of the experiment. See video 4. E. Summary of resulting cell phenotypes. LB: large bud. 113

Fig4-1Avideo1. SFI1 SPC42-GFP upon release from G1 arrest SFI1 SPC42-GFP (JA254) cells were grown to early/mid-log phase in YPD at 24°C and arrested in G1 using α-factor. Cells were then released into YNB+CAA media and imaged using a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) in a 37°C chamber. Time indicated is time from release. Images were taken on average every 4 minutes from 53 to 281 minutes after release. Bar: 5 µm. See Fig. 4-1A.

Fig4-1Bvideo2. esp1-1 SPC42-GFP upon release from G1 arrest esp1-1 SPC42-GFP (JA365) cells were grown to early/mid-log phase in YPD at 24°C and arrested in G1 using α -factor. Cells were then released into YNB+CAA media and imaged using a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) in a 37°C chamber. Time indicated is time from release. Images were taken on average every 5.3 minutes from 67 to 226 minutes after release. Bar: 5 µm. See Fig. 4-1B

Fig4-1Cvideo3. sfi1-C4A SPC42-GFP upon release from G1 arrest sfi1-C4A SPC42-GFP (JA302) cells were grown to early/mid-log phase in YPD at 24°C and arrested in G1 using α -factor. Cells were then released into YNB+CAA media and imaged using a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) in a 37°C chamber. Time indicated is time from release. Images were taken on average every 5.2 minutes from 59 to 423 minutes after release. Bar: 5 µm. See Fig. 4-1C.

Fig4-1Dvideo4. sfi1-C4A SPC42-GFP upon release from G1 arrest sfi1-C4A SPC42-GFP (JA302) cells were grown to early/mid-log phase in YPD at 24°C and arrested in G1 using α -factor. Cells were then released into YNB+CAA media and imaged using a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) in a 37°C chamber. Time indicated is time from release. Images were taken on average every 5.2 minutes from 59 to 365 minutes after release. Bar: 5 µm. See Fig. 4-1D.

114

E 10 8 6

4 # cells # 2 0 1 focus in mother, 1 in 0 foci in mother, 1 focus in 1 focus, remain as LB 2 foci, remain as LB daughter daughter

Control esp1-1 sfi1-C4A

115

Figure 4-2. In a single cell cycle, sfi1-C4A displays multiple phenotypes that differ from esp1-1 A-D. Cultures containing SPC42-GFP were grown to early- to mid-log phase in YPD at 24°C and arrested in G1 using α-factor. Cells were then released into YNB+CAA media and imaged using a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) in a 37°C chamber. α- factor was reintroduced 55 minutes after release to arrest the cells in the G1 phase of the subsequent cell cycle. Time indicated is time from release. Bar: 5 µm. A. Sfi1 control cells (JA254) displayed normal cell divisions, with the resulting mother and daughter from each cell containing one GFP focus. See video 5. B. esp1-1 SPC42-GFP (JA365) cells showed a SPB missegregation phenotype, with all GFP foci distributed to the daughter cell. See video 6. C. sfi1-C4A SPC42-GFP (JA302) cells. Top cell distributes one GFP focus to the mother and daughter, and then the daughter dies. The lower left cell distributes one GFP focus to the mother and daughter. The lower right cell distributes the GFP focus to the daughter only, and the mother lacking GFP then proceeds to die. See video 7. D. sfi1-C4A SPC42-GFP (JA302) cell that remains as a large-budded cell with a single GFP focus and does not appear to progress to the second cell cycle. See video 8. E. Summary of resulting cell phenotypes.

116

Fig4-2Avideo5. SFI1 SPC42-GFP in a single cell cycle SFI1 SPC42-GFP (JA254) cells were grown to early/mid-log phase in YPD at 24°C and arrested in G1 using α -factor. Cells were then released into YNB+CAA media and imaged using a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) in a 37°C chamber. α-factor was reintroduced 55 minutes after release to arrest the cells in the G1 phase of the subsequent cell cycle. Time indicated is time from release. Images were taken on average every 4 minutes from 43 to 203 minutes after release. Bar: 5 µm. See Fig. 4-2A.

Fig4-2Bvideo6. esp1-1 SPC42-GFP in a single cell cycle esp1-1 SPC42-GFP (JA365) cells were grown to early/mid-log phase in YPD at 24°C and arrested in G1 using α -factor. Cells were then released into YNB+CAA media and imaged using a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) in a 37°C chamber. α- factor was reintroduced 55 minutes after release to arrest the cells in the G1 phase of the subsequent cell cycle. Time indicated is time from release. Images were taken on average every 8.3 minutes from 50 to 373 minutes after release. Bar: 5 µm. See Fig. 4-2B.

Fig4-2Cvideo7. sfi1-C4A SPC42-GFP in a single cell cycle sfi1-C4A SPC42-GFP (JA302) cells were grown to early/mid-log phase in YPD at 24°C and arrested in G1 using α -factor. Cells were then released into YNB+CAA media and imaged using a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) in a 37°C chamber. α- factor was reintroduced 55 minutes after release to arrest the cells in the G1 phase of the subsequent cell cycle. Time indicated is time from release. Images were taken on average every 4.9 minutes from 32 to 264 minutes after release. Bar: 5 µm. See Fig. 4-2C.

Fig4-2Dvideo8. sfi1-C4A SPC42-GFP in a single cell cycle sfi1-C4A SPC42-GFP (JA302) cells were grown to early/mid-log phase in YPD at 24°C and arrested in G1 using α -factor. Cells were then released into YNB+CAA media and imaged using a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) in a 37°C chamber. α- factor was reintroduced 55 minutes after release to arrest the cells in the G1 phase of the subsequent cell cycle. Time indicated is time from release. Images were taken on average every 6.2 minutes from 51 to 509 minutes after release. Bar: 5 µm. See Fig. 4-2D.

117 one GFP focus in the resultant daughter cell (Fig. 4-2, B and E). The single GFP focus likely represents two SPBs, as the bud during the first cell cycle contained two GFP foci. sfi1-C4A displayed three main phenotypes (Fig. 4-2, C-E), with the majority of cells acting similar to control cells (Fig. 4-2A), in which one GFP focus was distributed to the mother and one to the daughter cell. Additionally, some sfi1-C4A cells did display a phenotype similar to esp1-1 in which the mother cell had no GFP foci and the daughter had one GFP focus. However, in contrast to esp1-1, three out of four of these sfi1-C4A cells always showed one, not two, GFP focus as large-budded cells during the first cell cycle. A final class of sfi1-C4A cells did not appear to progress through the cell cycle but remained as large-budded cells with one (n=3) or two (n=1) GFP foci. In conclusion, sfi1-C4A generally displayed a phenotype different from that of esp1-1. However, while these results do not rule out the model that the SPBs were reduplicated during the first cell cycle, they also do not fully rule out SPB missegregation.

sfi1-C3A is able to form a satellite at G1 phase

To demonstrate that SPB duplication occurs appropriately, and the reduplication phenotype present in sfi1-C3A and sfi1-C4A occurs within a single cell cycle, I examined whether sfi1-C3A can form a satellite. Examination of cells arrested in G1 with α-factor at 37ºC showed that sfi1-C3A was able to form a satellite (n=1 of 3; Fig. 4-3). In this same culture, we did also observe one cell with a single SPB and no visible satellite and one cell with side-by-side

SPBs, which may result due to the prolonged hold in G1 at high temperatures. Nonetheless, I was able to conclude that one of the initial steps in SPB duplication, satellite formation, can occur in sfi1-C3A.

118

Figure 4-3. sfi1-C3A is able to form a satellite at G1 phase An sfi1-C3A SPC42-GFP culture (JA247) was grown to early-log phase in YPD at 24°C and then cells were simultaneously treated with α-factor and shifted to 37°C for 3 h until they arrested in G1 (87% unbudded/shmoo of viable cells; 49% 1C DNA content via flow cytometry if cells with 0C DNA content are excluded). Sequential serial sections from the same cell are shown. Asterisk: Satellite. Arrow: SPB.

119

Reduplicated SPBs are enhanced in sfi1-C4A upon mitotic arrest

Since I propose that the aberrant SPB reduplication events observed in sfi1-C4A occur during mitosis within a single cell cycle, it would be expected that mitotically-arrested sfi1-C4A cells would display reduplicated SPBs. Therefore, I arrested cells in mitosis by inducing overexpression of a nondegradable Pds1 (GAL1-pds1-mdb) (Cohen-Fix et al., 1996) and then shifted to 37°C (Fig. 4-4). Pds1 (securin) binds separase (Esp1), and the degradation of Pds1 by the anaphase-promoting complex (APC) is required to allow sister chromatid separation and for progression to anaphase (Cohen-Fix et al., 1996; Ciosk et al., 1998).

The cells were initially arrested in G1, in which both SFI1 and sfi1-C4A GAL1-pds1-mdb cells commonly had two Spc42-GFP foci (>98%, Fig. 4-5, A and B), indicative of a mother SPB and satellite. As expected, SFI1 GAL1-pds1-mdb cells released from G1 and arrested in mitosis at 37°C displayed two separated (Spc42-GFP) foci, indicative of a bipolar spindle (86±2%; Fig.

4-5, C and D). In contrast, sfi1-C4A GAL1-pds1-mdb cells arrested in mitosis at 37°C only occasionally displayed two separated foci (15±1%). sfi1-C4A GAL1-pds1-mdb cells sometimes showed two foci at one position (30±3%), indicating that SPB duplication occurred, but not separation, but predominantly displayed at least three foci (41±3%) (Fig. 4-5, C and D). A majority of sfi1-C4A cells with at least three Spc42-GFP foci contained at least two adjacent foci with additional SPB(s), indicative of reduplicated SPBs (36.8±1.4%; Fig. 4-5C). These findings support the conclusion that SPB reduplication within a single cell cycle is responsible for the presence of additional SPBs in cells lacking Cdk1 phosphorylation sites in Sfi1.

The reduplication phenotype was also seen via EM. In sfi1-C4A GAL1-pds1-mdb

SPC42-GFP cells, the following phenotypes were seen: side-by-side SPBs, a bipolar spindle with a single SPB per pole, and a bipolar spindle with at least four SPBs (Fig. 4-6, A, B, and C, 120 respectively). sfi1-C4A GAL1-pds1-mdb cells without SPC42-GFP displayed side-by-side SPBs or a bipolar spindle with at least four SPBs (Fig. 4-6, D and E, respectively). In sfi1-C3A GAL1- pds1-mdb cells, the same experimental procedure was performed, and cells with three poles with one SPB per pole were seen, as well as cells with a monopolar spindle or two SPBs at one position within the nuclear envelope (Fig. 4-7 and legend).

Cdc14 is required to license SPB duplication

To determine whether Cdc14 acted on Sfi1 to control licensing of SPB duplication, I utilized an experimental strategy previously used by Cross and colleagues to examine another

Cdc14-regulated dephosphorylation event in mitosis (Bloom et al., 2011). Specifically, MET-

CDC20 cdh1Δ GALS-ESP1 cells arrested in metaphase via repression of CDC20 were driven into anaphase by induction of separase (Esp1), which leads to Cdc14 release from the nucleolus and activation. Lack of both Cdc20 and Cdh1, which are APC activators, prevents mitotic exit into G1 (Visintin et al., 1997; Sullivan and Uhlmann, 2003; Lu and Cross, 2009; Bloom et al.,

2011). Mitotic arrest of these cells was evident, with most cells displaying large buds and G2/M

DNA content (Fig. 4-8, A-D). This engineered strain, both with and without Esp1 induction, includes a population of large-budded cells with three or more dispersed Spc42-GFP foci, which may arise from SPB missegregation events (Fig. 4-8, E, G, and H). However, notably, significantly more actual SPB reduplication events (two adjacent SPBs with additional SPBs in large-budded cells) were seen in cells in which Esp1 overexpression was induced and Cdc14 was activated than in cells in which Esp1 was not induced (Fig. 4-8, F and H), indicating that active

Cdc14 promotes the presence of reduplicated SPBs.

121

Figure 4-4. sfi1-C4A and sfi1-C3A cells arrest in mitotis upon induced overexpression of nondegradable Pds1 Cells containing GAL1-pds1-mdb were grown to early log phase (T0) in YEP media containing 2% raffinose, then arrested in G1 with α-factor (T1), then released into YEP media containing 2% galactose to arrest cells in mitosis at 24°C (T2; 3.25 h). Once at mitotic arrest, they were shifted to 37°C for an estimated 2.5 h (T3). A. Experiment schematic, with samples taken for analysis at each timepoint shown. B-C. SFI1 (JA295) or sfi1-C4A (JA297) cells containing GAL1-pds1-mdb SPC42-GFP. D-E. SFI1 (JA298), sfi1-C3A (JA299; haploid isolated from mixed ploidy population), or sfi1-C4A (JA300) cells containing GAL1-pds1-mdb. B, D. Budding content. LB: large-budded, SB: small-budded, UB: unbudded. Error bars: SD. n≥91 per group from 2 experiments (B). n≥150 per group from 3 experiments (D). C, E. DNA content via flow cytometry from 1 experiment. %G2/M from 2 experiments (C).

122

Figure 4-5. Reduplicated SPBs are enhanced in sfi1-C4A GAL1-pds1-mdb upon mitotic arrest Experiment performed as described in Fig. 4-4. A-D. SFI1 (JA295) or sfi1-C4A (JA297) cells containing GAL1-pds1-mdb SPC42-GFP. Fixed cells were imaged using SIM, with GFP on left (insets bar: 1 µm) and merge with transmitted image on right. Bar: 5 µm. A-B. G1-arrested (T1) cells and quantification (B). No significant differences (Student’s t test) for each category between strains. n≥219 per group from 2 experiments. C-D. Mitotically-arrested large-budded cells following 37°C shift (T3). Examples of a single reduplication (middle panel) or two reduplications (lower panel) for sfi1-C4A. D. Quantification of C. Asterisks: statistically significant difference via Student’s t test **p<0.01. *: p<0.05. Error bars: SD. n≥185 cells per group from 2 experiments. 123

124

Figure 4-6. Reduplicated SPBs are present in sfi1-C4A GAL1-pds1-mdb upon mitotic arrest Experiment performed as described in Fig. 4-4. A-C. Mitotically-arrested sfi1-C4A GAL1-pds1- mdb SPC42-GFP (JA297) cells following 37°C shift (T3) were prepared for EM. Serial sections from four cells were examined. A. A cell containing side-by-side SPBs at one position within the nuclear envelope (n=1). 100 nm bar. B. Representative cell (n=2) with a bipolar spindle with a single SPB at each pole. Left panel shows location of SPBs, 500 nm bar. Right: enlarged panels of SPBs, 100 nm bar. C. Images from a cell containing at least two reduplicated SPBs, with at least four SPBs (1-4) total (n=1). Left panels: entire nucleus, 500 nm bar. Right: enlarged panels of SPBs, 100 nm bar. H. D-E. Mitotically-arrested sfi1-C4A GAL1-pds1-mdb (JA300) cells following 37°C shift (T3) were prepared for EM. Serial sections from five cells were examined. D. Representative cell (n=4) containing side-by-side SPBs at one position within the nuclear envelope. 100 nm bar. E. Images from a cell containing two reduplicated SPBs, with four SPBs total (n=1). Left panel shows position of SPBs, 500 nm bar. Right: enlarged panels of SPBs, 100 nm bar.

125

Figure 4-7. Extra SPBs are present in sfi1-C3A GAL1-pds1-mdb upon mitotic arrest Experiment performed as described in Fig. 4-4. Mitotically-arrested sfi1-C3A GAL1-pds1-mdb cells (JA299; haploid isolated from mixed ploidy population) following 37°C shift (T3) were prepared for EM. Serial sections from seven cells were examined. Representative cell (n=2) with 3 poles, with a single SPB at each pole. Left panels shows location of SPBs, 500 nm bar. Middle and right: enlarged panels of SPBs, 100 nm bar. Upper midde panel displays a different serial section than the upper left and right panels. Note: Cells containing one SPB (n=2) or two SPBs at one position within the nuclear envelope (n=3) were also seen.

126

127

Figure 4-8. Reduplicated SPBs result from Cdc14 release from the nucleolus An asynchronous culture (T0) of MET-CDC20 cdh1Δ GALS-ESP1 SPC42-GFP (JA256) cells was grown to early log phase in SC-Met with 3% Raffinose. Methionine was then added to a final concentration of 2 mM to arrest cells in metaphase and added every two h for the experiment remainder. After metaphase arrest (T1; 4.5 h), Esp1 either was induced for 4 h using 3% galactose to release Cdc14 from the nucleolus or remained uninduced in raffinose (T2). A. Experiment schematic. B. Budding content. Error bars: SD. n≥199 cells per group from 2 experiments. C-D. DNA content via flow cytometry. D. Quantification of C from 2 experiments (except T1). Error bars: SD. E-H. Fixed large-budded cells from T1 (E) or T2 (F-G) were imaged via SIM, with GFP in left panels (insets bar: 1 µm) and merge with the transmitted image in right panels. Bar: 5 µm. F. The upper panel shows cells with no Esp1 induction with bipolar spindles. The lower panels show cells with Esp1 induction for 4 h with one or two reduplicated SPBs (3 or 4 Spc42-GFP foci total, respectively) in large-budded cells. G. A portion of cells contains at least three dispersed Spc42-GFP foci. H. Quantification of E-G. Asterisks: statistically significant difference via Student’s t test only assessed for T2 conditions. *: p<0.05. Error bars: SD. n≥211 per group from 2 experiments.

128

III. Discussion

Reducing Cdk1 phosphorylation of Sfi1 leads to reduplicated SPBs

Live timelapse imaging analysis of sfi1-C4A did not rule out the model that the SPBs were reduplicated during the first cell cycle, but they also did not fully rule out SPB missegregation. In order to examine single cells in a meaningful manner, the capability to use higher resolution would be needed in order to determine the precise number of SPBs and SPB intermediates present in the cells. However, by arresting a population of sfi1-C4A cells in mitosis, I was able to see an enhancement of cells with extra SPBs, suggesting that SPBs reduplicate within a single cell cycle in mitosis. In addition, I was able to show that these cells do not have extra SPBs at the beginning of the cell cycle during G1 arrest, in which they displayed a satellite and a SPB.

Since sfi1-C4A also shows defects at 24°C, examining whether SPB reduplication occurs in mitosis in a system in which the sfi1-C4A defect can be selectively induced would further confirm that SPB reduplication can occur in mitosis in a single cell cycle. If sfi1-C4A can act as a dominant negative allele and can be incorporated into the half-bridge, then a system in which sfi1-C4A overexpression can be induced upon mitotic arrest (e.g., GAL1-sfi1-C4A in SFI1 MET-

CDC20 cells) could provide fruitful.

Given the requirement of Mps1 for SPB duplication (Winey et al., 1991; Castillo et al.,

2002; Jaspersen et al., 2004; Holinger et al., 2009), it would be interesting to determine whether inhibition of Mps1 activity would lead to a nonpermissive state for reduplication in mitotically- arrested sfi1-C4A cells. This experimental strategy would help further clarify that the extra SPBs seen in sfi1-C4A and sfi1-C3A are due to reduplication, and not SPB missegregation, events.

Previous research has suggested a requirement of Mps1 for Sfi1 incorporation at the SPB 129

(Elserafy et al., 2014), which corresponds with previous work from the Winey lab showing that one mps1 mutant allele, mps1-8, terminates with a short half-bridge (Castillo et al., 2002), suggesting that Mps1 is required for half-bridge elongation.

Similarly, it would be interesting to determine whether loss of Mps1 phosphorylation on one of its substrates in which phosphorylation is required for SPB duplication would lead to a nonpermissive state for reduplication in mitotically-arrested sfi1-C4A cells. Specifically, Cdc31 is phosphorylated by Mps1 at residue T110, and converting this site to a nonphosphorylatable residue (cdc31-T110A) leads to single microtubule asters, suggesting a potential defect in SPB duplication (Araki et al., 2010). Thus, if examination of cdc31-T110A via SIM and/or EM shows a defect exists specifically in SPB duplication, not separation, then examining whether SPB reduplication is not permitted at a mitotic arrest in sfi1-C4A GAL1-pds1-mdb cells would prove useful.

Phosphorylation of Sfi1 blocks SPB reduplication

Via an execution point experiment, Schiebel and colleagues showed that cells with Sfi1 containing six phosphomimetic Cdk1 sites (sfi1Cdk1-6D) arrest with a single SPB in the first cell cycle following a G1 arrest. This would suggest a requirement for the loss of Sfi1 C terminal phosphorylation after satellite formation. However, it is possible that dephosphorylation of the

Sfi1 C terminus is required for half-bridge elongation, and thus the initiation of SPB duplication, and for maintaining the elongated and full bridge structure, which is predicted to contain Sfi1 molecules in an end-to-end configuration (Elserafy et al., 2014). In their experiment, the inability to maintain the bridge could correlate with this conclusion. Examination of these cells in asynchronous cultures or upon shift to restrictive temperature following a mitotic arrest could 130 prove useful in clarifying the phenotype of these cells. Nonetheless, their findings that sfi1Cdk1-

6D cells arrest with a single SPB suggests that Cdk1 phosphorylation of Sfi1 inhibits SPB duplication (Elserafy et al., 2014). This result supports our conclusion that the block to SPB duplication by Cdk1 phosphorylation of Sfi1 must be removed to license SPB duplication.

Using the same experimental strategy, Schiebel and colleagues propose that Cdc5 phosphorylation of Sfi1 is also required to block reduplication, as they show that Sfi1 residues

(S826 T866 T876) are phosphorylated by Cdc5 in vitro, and they see cells arrested with single

SPBs in a phosphomimetic allele (sfi1Cdc5-3D; sfi1-S826D T866D T876D) (Elserafy et al., 2014).

However, while Cdc5 may regulate Sfi1 to some extent, the role of this regulation is not clear.

Cdc14 is required to license SPB duplication

A recent study showed that cdc14-2 cells forced into G1 were delayed in SPB duplication

(Elserafy et al., 2014). The authors conclude that Cdc14 promotes timely SPB duplication, which is consistent with my work demonstrating that Cdc14 is specifically involved in licensing of SPB duplication, as evidenced by the presence of reduplicated SPBs with prolonged activation of Cdc14. Schiebel and colleagues did not observe an increase in Sfi1 incorporation into the

SPB, as measured by Sfi1 fluorescent signal intensity, with overexpression of Cdc14 in a reported metaphase arrest (Elserafy et al., 2014). Using the protocol from the Cross lab (Lu and

Cross, 2009; Bloom et al., 2011), I was able to arrest cells at a stage in the cell cycle that was permissive for Cdc14-driven reduplication and directly examined SPB assembly using SIM.

While Cln/Cdk1 activity is required for SPB duplication in G1 (Byers and Goetsch, 1974), in mitotically-arrested cells in the absence of mitotic cyclin degradation (i.e., upon overexpression of GAL1-pds1-mdb or in MET-CDC20 cdh1Δ cells upon CDC20 repression) (Schwab et al., 131

1997; Cohen-Fix and Koshland, 1999; Tinker-Kulberg and Morgan, 1999; Wäsch and Cross,

2002), it appears that mitotic cyclins permit SPB duplication when Sfi1 phosphorylation is diminished.

Phosphoregulation of Sfi1 is required for licensing of SPB duplication

In the current study, I have reproduced the reduplication phenotype seen upon depletion of Cdk activity (Haase et al., 2001), either by nonphosphorylatable versions of Sfi1 or by inducing prolonged Cdc14 activity. My findings support the model that Cdk1 phosphorylation of Sfi1 is required for SPB separation and a block to SPB reduplication (Fig. 4-9). Specifically, I propose that after SPB duplication is complete, Cdk1 phosphorylates at least four Sfi1 C- terminal residues, allowing for SPB separation and bipolar spindle formation during mitosis.

These C-terminal residues remain phosphorylated during mitosis, restricting licensing of SPB duplication until downregulation of mitotic cyclin/Cdk1 at the end of mitosis.

Dephosphorylation of Sfi1 by Cdc14 at the end of mitosis then licenses the SPBs for the next cycle of G1 SPB duplication. Thus, appropriate phosphoregulation of Sfi1 by Cdk1 and Cdc14 ensures SPB duplication occurs only once per cell cycle.

IV. Materials and Methods

Strain construction

Creation of single integrant mutants integrated at the SFI1 locus with the NATMX marker via PCR is as previously described (Tong and Boone, 2006) (see Chapter 3) using a standard yeast transformation protocol (Gietz and Woods, 2002). The following strains contain

132

Figure 4-9. Model for licensing of yeast centrosome duplication via phosphoregulation of Sfi1 Cyclin/Cdk1 activity is required for SPB separation, in which Cdk1 phosphorylates the C terminus of Sfi1. This phosphorylation ensures SPB duplication does not begin until completion of mitosis and downregulation of mitotic cyclin/Cdk1. Dephosphorylation of Sfi1, likely by Cdc14, licenses SPB duplication to allow the next cycle of SPB duplication to begin at G1. ■: Satellite. NE: nuclear envelope. cMT: cytoplasmic microtubules. nMT: nuclear microtubules. 133 the silent mutations from nucleotide changes A2130G and T2131C (residues E710 and L711):

JA247, JA254, JA297, JA299, JA300, JA302.

C-terminal tagging of SPC42 with yeGFP1 using pFA6-yeGFP1-KANMX-HISMX

(E2642) via PCR was performed as previously described (Longtine et al., 1998; Sheff and Thorn,

(JA302) was obtained by dissection of sfi1-C4A-NATMX/SFI1 SPC42-GFP-KANMX/SPC42 ade2::ADE2/ ade2::ADE2 MATa/MATα (JA1-030). JA1-030 was obtained by transformation of the yeGFP1 PCR product into sfi1-C4A-NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα

(JA235, JA236).

sfi1-C3A-NATMX/sfi1-C3A-NATMX SPC42-GFP-KANMX/SPC42 ade2::ADE2/ade2::ADE2 MATa/MATa (JA247) was created by transformation of the yeGFP1

PCR product into sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 MATa/MATa

(JA188). esp1-1 SPC42-GFP-KANMX ade2::ADE2 (JA365) was obtained by mating esp1-1 ura3-52 ade2 lys2 leu2-3,112 cyh2 (SN127-3B; Peter Baum) to JA254 and selecting an appropriate spore from dissection of the resultant diploid strain.

Strains JA295, JA297, JA298, JA299, and JA300 were created by integrating pOC70

(gift of Orna Cohen-Fix) (Cohen-Fix et al., 1996) digested with BstEII at the LEU2 locus of heterozygous diploid strains (JA1-029, JA1-030, JA157 for strains JA298 and JA299, and

JA235/JA236, respectively), followed by dissection and selection of appropriate haploids. JA1- 134

029 (sfi1-C3A-NATMX/SFI1 SPC42-GFP-KANMX/SPC42 ade2::ADE2/ ade2::ADE2

MATa/MATα) was obtained by transformation of the yeGFP1 PCR product into sfi1-C3A-

NATMX/SFI1 ade2::ADE2/ade2::ADE2 MATa/MATα (JA157).

JA256 (cdc20::MET-CDC20-TRP1 cdh1Δ::LEU2 esp1::GALS-ESP1-URA3 SPC42-

GFP-KANMX ade2::ADE2) was created via mating of YL165 (cdc20::MET-CDC20-TRP1 cdh1::LEU2 ESP1::GALS-ESP1-URA3 CDC14-YFP-HIS3 NET-CFP-KAN MATa; gift of Fred

Cross) (Lu and Cross, 2009) and SFI1-NATMX SPC42-GFP-KANMX ade2::ADE2 MATα

(JA255) followed by dissection and selection of the appropriate haploid.

To assess cell viability, the vital stain trypan blue (Sigma-Aldrich) was used.

Flow cytometry

Cells were fixed with 70% ethanol for 1 h at room temperature or at 4°C overnight.

Fixed cells were then treated with 0.1% RNase in 0.2M Tris-HCl pH 7.5, 20 mM EDTA for 2 h at 37°C, and the DNA was stained with 50 µg/mL propidium iodide (Sigma Chemical Co., St.

Louis, MO) in PBS for 1 h at room temperature or at 4°C overnight. DNA content was analyzed using the CyAn ADP analyzer (Beckman Coulter, Indianapolis, IN, USA) with a 488nm laser.

30,000 events per sample were taken. DNA content (propidium iodide area) is indicated on X axes and Count on Y axes.

Cytological techniques

Live timelapse imaging was performed at the Stowers Institute for Medical Research on a confocal laser scanning microscope (ConfoCor 3; Carl Zeiss) using the Avalanche photodiode

(APD) imaging module with a 63x, 1.4 NA objective, Zoom 5, a 488 nm Argon laser, and AIM software (version 4.2), with a 0.056 µm pixel size, 3.20 µs or 1.61 µs pixel dwell time, a 536 µm pinhole, and z-stacks (10 slices) over 9 µm. The CellAsic ONIX microfluidic perfusion system 135

(EMD Millipore) and haploid yeast cell microfluidic plates (EMD Millipore #Y04C-02-5PK) were used in a 37°C chamber. G1-arrested cells (using 10ug/mL α-factor) were released into pre-warmed YNB+CAA media and subsequently loaded into the microfluidics plates at 37°C for imaging. For experiments with a subsequent G1 arrest, media containing 10 µg/mL α-factor flowed into the microfluidics system at 55 minutes following release. A constant flow rate was used throughout all imaging. Interval times between images are indicated as average times.

Images were binned 2x2, smoothed, and the max projection was used for GFP and DIC using Fiji

(ImageJ; National Institutes of Health, Bethesda, MD, USA) analysis software.

For brief fixation of GFP cells, cultures were fixed with 3.7% formaldehyde for 15 minutes and resuspended and imaged in 1xPBS or KPO4/sorbitol. SIM images were acquired at the Stowers Institute for Medical Research at room temperature on an Applied Precision OMX

Blaze microscope (Issaquah, WA, USA) equipped with a PCO Edge sCMOS camera (Kelheim,

Germany). The objective used was an Olympus (Center Valley, PA, USA) 60x 1.42NA Plan

Apo N oil objective. Image stacks were acquired at 125 nm intervals. SIM reconstruction was performed with the Applied Precision software package utilizing optical transfer functions measured with 100 nm green fluospheres in Prolong Gold on a coverslip surface (Life

Technologies, Carlsbad, CA, USA) following the Applied Precision protocols. After reconstruction, SIM images were scaled 2 by 2 with bilinear interpolation through ImageJ software for future quantification.

Transmission electron microscopy

Log phase cells were high pressure frozen in a Wohlwend Compact 02 HPF and freeze- substituted in 2% osmium tetroxide, 0.1% uranyl acetate in acetone and embedded in Epon epoxy (JA247) or in 0.25% glutaraldehyde, 0.1% uranyl acetate in acetone and embedded in 136

Lowicryl HM20 (GAL1-pds1-mdb strains) (Giddings et al., 2001). Imaging was conducted using a FEI Phillips CM100 electron microscope.

137

Table 4-1. Yeast strains

Strain Genotype JA247 sfi1-C3A-NATMX/sfi1-C3A-NATMX SPC42-GFP-KANMX/SPC42 ade2::ADE2/ade2::ADE2 MATa/MATa JA254 SFI1-NATMX SPC42-GFP-KANMX ade2::ADE2 MATa JA256 cdc20::MET-CDC20-TRP1 cdh1Δ::LEU2 esp1::GALS-ESP1-URA3 SPC42-GFP- KANMX ade2::ADE2 MATα JA295 SFI1 leu2::GAL1-pds1-mdb-LEU2 SPC42-GFP-KANMX ade2::ADE2 MATa JA297 sfi1-C4A-NATMX leu2::GAL1-pds1-mdb-LEU2 SPC42-GFP-KANMX ade2::ADE2 MATa JA298 SFI1 GAL1-pds1-mdb MATa JA299 sfi1-C3A-NATMX leu2::GAL1-pds1-mdb-LEU2 ade2::ADE2 MATa JA300 sfi1-C4A-NATMX/sfi1-C4A-NATMX leu2::GAL1-pds1-mdb-LEU2/leu2::GAL1-pds1- mdb-LEU2 ade2::ADE2/ade2::ADE2 MATa/MATa JA302 sfi1-C4A-NATMX SPC42-GFP-KANMX ade2::ADE2 MATa JA365 esp1-1 SPC42-GFP-KANMX ade2::ADE2 MATa

138

Table 4-2. Plasmids

Identifier Name Genotype E2642 pFA6-yeGFP1-KANMX-HISMX yeGFP1-KANMX-HISMX pOC70 pRS305-GAL1-pds1-mdb GAL1-pds1-mdb-LEU2

139

Chapter Five: Conclusions and future directions

I. Summary

Duplication of centrosomes once per cell cycle is essential for bipolar spindle formation and genome maintenance. Thus, tight control of this process is critical. I have proposed that the kinase Mps1 is required for satellite formation and thus may play a key role in regulating recruitment of satellite components, potentially through the Sfi1 N terminus (Fig. 5-1). Through studies of a nonphosphorylatable Sfi1 N terminus allele (sfi1-NnonP), I found that it shows genetic interactions with protein-encoding genes of the satellite proteins Spc42 and Spc29, thus identifying these satellite components as promising candidates to be recruited by the Sfi1 N terminus.

I also analyzed the import of the Sfi1 C terminus by creating a mutant allele collection. I found that conversion of three or four Cdk1 sites at the Sfi1 C terminus to nonphosphorylatable residues leads to defects in bipolar spindle assembly, chromosome segregation, and growth.

These cells often showed one of two phenotypes: unseparated SPBs or reduplicated SPBs. I was also able to show that reduplicated SPBs were enhanced upon mitotic arrest in sfi1-C4A and were present upon prolonged activation of the protein phosphatase Cdc14 in a wild-type SFI1 background. These findings lead to the model that phosphorylation of Sfi1 by Cdk1 both promotes SPB separation for spindle formation and prevents premature SPB duplication.

Additionally, the protein phosphatase Cdc14 has the converse role of activating licensing, likely via dephosphorylation of Sfi1 (Figs. 4-9 and 5-1).

II. A potential role for combinatorial control of posttranslational modifications in satellite formation 140

Figure 5-1. Model for Sfi1 in SPB duplication Proposed requirements in the SPB duplication pathway based on this study. Mps1 is required for satellite formation and may regulate recruitment of satellite components, potentially through the Sfi1 N terminus. The satellite proteins Spc42 and Spc29 are promising candidates to be recruited by the Sfi1 N terminus. Cdk1 phosphorylation of the Sfi1 C terminus is required for SPB separation. This phosphorylation ensures SPB duplication does not begin until completion of mitosis and downregulation of mitotic cyclin/Cdk1. Cdc14 licenses SPB duplication, likely via dephosphorylation of Sfi1, to allow the next cycle of SPB duplication to begin at G1. NE: nuclear envelope. cMT: cytoplasmic microtubules. nMT: nuclear microtubules.

141

Since I propose that Mps1 may play a key role in satellite formation, it will prove useful to examine whether synthetic defects between sfi1-NnonP and phosphomutants of the satellite proteins and the potential regulator Cdc31 exist. While sfi1-NnonP shows no defects on its own, removal of phosphorylation from multiple protein components involved in satellite formation may prove detrimental, which would suggest that combinatorial control of phosphorylation is important in satellite formation. Furthermore, roles for other modifications may prove important. For example, the Sfi1 N terminus has been shown to be ubiquitinated in vivo at residue K18 (Radivojac et al., 2010). If direct interactions between Sfi1 and satellite components are identified, it will be interesting to determine whether an experimental system can be developed in which a satellite can be created de novo in the cell if the appropriate components and modifications are available.

III. Mechanism for licensing of SPB duplication

The Sfi1 C terminus is localized at the center of the bridge, while the N terminus has two distinct localizations, one proximal to the mother SPB and one at the daughter SPB (Li et al.,

2006). This suggests Sfi1's structure may be important in bridge formation and severing during

SPB separation, with the Sfi1 C termini directly or indirectly associating to form the bridge, and dissociation being needed to sever the bridge. Given the localization of Sfi1 and its functions in separation and licensing, a role for Sfi1 in both the SPB-intrinsic (Simmons Kovacs et al., 2008) and Cdk1-regulated modes (Haase et al., 2001) of ensuring licensing of SPB duplication occurs only once per cell cycle is plausible. Specifically, when Sfi1 C termini are not phosphorylated,

Sfi1 C termini can associate, and SPB duplication can proceed. The presence of a full bridge inhibits SPB reduplication via an SPB-intrinsic mode. Phosphorylation of the Sfi1 C terminus 142 by Cdk1 then enables Sfi1 molecules to dissociate, and SPB separation can thus occur. Phosphorylation of the Sfi1 C terminus then becomes the block to reduplication. This block will be maintained until dephosphorylation of Sfi1 allows for licensing to occur following mitotic cyclin/Cdk1 downregulation.

While Kilmartin and colleagues have proposed that SPB duplication initiates when Sfi1 molecules associate via C-terminal end-to-end interactions with Sfi1 molecules already present at the half-bridge (Li et al., 2006), no previous work has identified Sfi1 C-terminal interactions.

Therefore, future work examining whether the Sfi1 C termini directly interact with each other and/or interact with other components of the SPB and the impact of phosphorylation status on these interactions will be important in understanding the mechanism that initiates SPB duplication. It is expected that half-bridge proteins may be particularly good candidates as Sfi1 interactors given their location and thus proximity to Sfi1 and requirement in the initial steps of

SPB duplication (Byers 1981) (Rose and Fink, 1987; Winey et al., 1991; Jaspersen et al., 2002).

Additionally, two of the half-bridge proteins, Kar1 and Mps3, are membrane proteins (Vallen et al., 1992; Spang et al., 1995; Jaspersen et al., 2002) and could thus potentially anchor Sfi1 to the bridge.

IV. Analogies between the licensing events of SPB duplication and DNA replication

Analogies between the events of licensing of DNA replication and licensing of SPB duplication can be drawn based on the current work (Fig. 5-2). In SPB duplication, I have shown that dephosphorylation of the Sfi1 C terminus licenses this event in late mitosis. This dephosphorylation makes the SPB competent for duplication in G1. Specifically, I propose this dephosphorylation permits half-bridge elongation to occur in early G1 by addition of Sfi1 143

Figure 5-2. Analogy between the licensing events of SPB duplication and DNA replication A. In SPB duplication, dephosphorylation of Sfi1 C-terminal Cdk sites is the licensing event that then permits half-bridge elongation to occur in G1. Note: Previous work has shown that half- bridge elongation requires Mps1 (Castillo et al., 2002; Elserafy et al., 2014). B. In DNA replication, pre-RC components must be dephosphorylated in order to form the pre-RC on the DNA origin of replication, the licensing event of DNA replication (Nguyen et al., 2001; Morgan, 2006; Sclafani and Holzen, 2007; Enserink and Kolodner, 2010). In budding yeast, ORC is bound to the DNA throughout the cell cycle (Diffley et al., 1994) and is phosphorylated by Clb- Cdk (Nguyen et al., 2001; Makise et al., 2009), Cdk phosphorylation of Cdc6 leads to Cdc6 degradation (Elsasser et al., 1999), CDC6 transcription occurs at the end of mitosis (Piatti et al., 1995) and phosphorylated (at sites within Cdk consensus motifs) Cdc6’s interaction with Clb- Cdk is suggested to prevent pre-RC assembly (Mimura et al, 2004), and Clb-Cdk phosphorylation of the Mcm2-7 complex promotes export of the complex from the nucleus (Nguyen et al., 2000).

144 molecules to the half-bridge in an end-to-end fashion. As discussed above, this addition may result from direct binding of Sfi1 C termini or through association with other SPB components.

Similarly, in DNA replication, pre-RC assembly on the DNA early in G1 requires that the pre-

RC components (ORC, Cdc6, Mcm2-7) be dephosphorylated. This pre-RC formation is the licensing event that makes the DNA competent to proceed in the process of DNA replication

(Nguyen et al., 2001; Mimura et al., 2004; Liku et al., 2005; Morgan, 2006; Sclafani and Holzen,

2007; Enserink and Kolodner, 2010). Thus, Sfi1 in SPB duplication could be considered analogous to the pre-RC components ORC, Cdc6, Mcm2-7 in DNA replication, as dephosphorylated components are required for duplication competency and Cdk phosphorylation of these components blocks reduplication. Interestingly, while DNA re-replication requires disruption of the effects of Cdk phosphorylation on three pre-RC components in combination,

ORC, Cdc6, Mcm2-7 (Nguyen et al., 2001), SPB reduplication requires loss of Cdk phosphorylation of only one protein, Sfi1. Thus, while regulation of multiple licensing factors is important in restricting initiation of DNA replication to once per cell cycle, I propose phosphoregulation of a single licensing factor, Sfi1, is sufficient to ensure SPB duplication occurs only once per cell cycle.

In both SPB duplication and DNA replication, following licensing and initial formation of a protein complex (pre-RC) or structure (elongated half-bridge), additional proteins can then be recruited at the site of duplication. In SPB duplication, the satellite will form at the cytoplasmic distal tip of the elongated half-bridge, and this step involves Mps1 and possibly recruitment of satellite proteins by Sfi1 (this study). This satellite will then expand to form the new SPB, and this process requires Cdk1 (Byers and Goetsch, 1974). In DNA replication, once pre-RC formation occurs, additional proteins can be recruited to the origin of replication to form 145 the pre-IC, whose formation requires both Cdk and DDK later in G1. Recruitment of DNA polymerase and Mcm2-7 activation can then occur for DNA synthesis to proceed in S phase

(reviewed in Morgan, 2006; Sclafani and Holzen, 2007; Enserink and Kolodner, 2010).

V. Model for conservation of Cdk regulation of Sfi1 in licensing of centrosome duplication

In human cells, centriole engagement has been identified as a centrosome-intrinsic mode to block licensing of centrosome duplication, while centriole disengagement, which is regulated by separase and Plk1, is required for licensing of centrosome duplication (Tsou et al., 2009). No known involvement of Cdk in this process of centriole disengagement has been shown. In fact, in in vitro Xenopus egg extracts, it was shown that centriole disengagement does not require mitotic Cdk1 inactivation, as mitotically-arrested cells with nondegradable mitotic cyclin B still displayed disengaged centrioles (Tsou and Stearns, 2006). However, Haase and colleagues suggest that a Cdk-regulated block to licensing after disengagement and before centrosome duplication begins in G1 could be plausible (Simmons Kovacs et al., 2008). Interestingly, in D. melanogaster, inactivation of mitotic Cdk1 leads to centriole reduplication, reportedly without centriole disengagement (Vidwans et al., 2003). This result would suggest a role for Cdk in licensing of centrosome duplication. Further studies will prove useful in determining whether

Cdks are involved in licensing in metazoans and any potential targets.

Given that human Sfi1 also contains multiple Cdk consensus motifs at the C terminus, this study could provide crucial insight into the licensing of centrosome duplication. It is known that hSfi1 localizes to the centrosome in humans (Kilmartin, 2003). Through an siRNA screen, it has been suggested that human Sfi1 (hSfi1) is required for centriole formation (Balestra et al.,

2013); however, further analysis of a role for Sfi1 in centriole duplication is needed. Future 146 examination using mutations of the hSfi1 gene to alter the phosphorylation sites within Cdk consensus motifs in human cells could prove useful.

Indeed, the mechanism for licensing of SPB duplication could be conserved, in which phosphoregulation of Sfi1 by Cdk is required (Fig. 5-3). However, the downstream components required for centrosome duplication may differ. For example, two key components required for centriole duplication in humans as well as other organisms are polo-like kinase 4 (Plk4; SAK in

D. melanogaster) (O’Connell et al., 2001; Kleylein-Sohn et al., 2007; Bettencourt-Dias et al.,

2005) and Sas-6 (Leidel et al., 2005; Strnad et al., 2007; Kleylein-Sohn et al., 2007; Azimzadeh and Bornens, 2007; Nigg and Stearns, 2011). Indeed, Sas-6 is critical for the initial formation of the centriole via formation of a cartwheel or, in Caenorhabditis elegans, a cylinder structure that are thought to establish the ninefold symmetry of the centriole (reviewed in Azimzadeh and

Bornens, 2007; Nigg and Stearns, 2011). In C. elegans, it has been shown that phosphorylation of SAS-6 by ZYG-1, a kinase related to Plk4, is required for centriole assembly (Kitagawa et al.,

2009). Additionally, it has recently been suggested that the protein asterless (Cep152, which interacts with Plk4 (Hatch et al., 2010; Cizmecioglu et al., 2010)) is required for initiation of centriole duplication in D. melanogaster (Novak et al., 2014). These components and structures are not found in budding yeast. However, given the conservation of Sfi1 and its potential role in centrosome duplication initiation, examination of whether Sfi1 in relation to these key proteins is required for centriole duplication will be important.

147

Figure 5-3. Model for a conserved mechanism of licensing of centrosome duplication Sfi1 at the centrosome is not phosphorylated at late mitosis/G1 when Cdk activity levels are low, and this lack of phosphorylation enables the licensing of centrosome duplication, thus making the centrosome competent for duplication to continue. Once Sfi1 is phosphorylated, reduplication is prohibited, and centrosome duplication cannot begin again until Cdk activity levels are once again low at the end of mitosis, allowing for another round of centrosome duplication in the two new daughter cells.

148

Appendix A: Analysis of Sfi1 yeast two-hybrid interactions

I. Introduction

A model put forth by Kilmartin and colleagues based on Sfi1 localization suggests that

SPB duplication initiates when Sfi1 molecules associate with those already at the half-bridge through end-to-end interactions at the C termini, doubling the half-bridge length. The newly exposed Sfi1 N terminus at the distal tip of the elongated half-bridge can then recruit satellite components (Li et al., 2006). While predicted, it is not known whether the Sfi1 C termini directly interact. Furthermore, there are no previously published reports of genetic or physical interactions between Sfi1 and the satellite proteins, Spc42, Spc29, Cnm67, and Nud1 (Adams and Kilmartin, 1999). The genetic interactions I identified between SFI1, using an N terminus mutation, and the satellite protein encoding genes SPC42 and SPC29 provides a basis to examine physical interactions between the Sfi1 N terminus and satellite proteins (see Chapter 2).

Therefore, Tara Peters, an undergraduate student who completed an honors thesis on this work, and I sought to determine whether we could identify the interactions predicted for the Sfi1 N and

C termini based on the model by Kilmartin and colleagues (Li et al., 2006). Specifically, we aimed to determine whether the Sfi1 C termini directly interact and whether the Sfi1 N terminus interacts with the satellite proteins.

Yeast two-hybrid (Y2H) analysis is a useful way to initially examine potential physical protein interactors of full-length proteins or protein domains. In the Y2H assay used in this study, the functional domains of the Gal4 transcriptional activator protein are used. The Gal4

DNA binding domain (BD) is fused to one protein of interest, serving as the “bait”, and the Gal4 activation domain (AD) is fused to a second protein of interest, serving as the “prey”. If the two proteins of interest interact, transcription of a reporter gene under control of an inducible 149 promoter is activated (Fields and Song, 1989; James et al., 1996). We therefore took advantage of this Y2H system to examine interactors of the Sfi1 N- and C-terminal domains through both a directed approach with candidate interactors as well as a screening approach.

II. Results

Please refer to Tara Peters’ undergraduate honors thesis entitled “Identification of Sfi1 protein interactions through yeast two-hybrid analysis” for detailed results and figures.

Directed Y2H with candidate interactors

The Sfi1 N terminus fused to the GAL4 BD did not show interaction via Y2H with the satellite proteins Spc42 or Spc29, individually, fused to the GAL4 AD. The Sfi1 C terminus did not self-associate, as tested via fusion to both the BD and AD. Additionally, since Cdc31 is known to interact with the Sfi1 repeat domain, (Kilmartin, 2003), we included two Sfi1 repeats with the Sfi1 C terminus sequence. We therefore sought to determine whether Cdc31 may possibly regulate any Sfi1 C-terminal interactions, despite the fact that Cdc31 does not interact with the Sfi1 C terminus in vitro (Kilmartin, 2003). No self-association of the Sfi1 C terminus with two Sfi1 repeats was seen.

Y2H screen identifies Cse2 as an Sfi1 C terminus interactor

Expression of the Sfi1 C terminus-BD fusion protein was first confirmed via western blot. In a screen for Y2H interactors, yeast containing this Sfi1 C terminus-BD construct were mated to a pooled AD-ORF (open reading frame) array in yeast. A total of 4.84 million clones were screened. Two independent colonies were isolated to identify the same interactor, Cse2. 150

The plasmid containing AD-CSE2 was isolated and used to confirm a positive interaction between BD-Sfi1 C terminus and AD-Cse2 via Y2H.

III. Discussion

It is predicted that the Sfi1 N terminus recruits satellite components during the process of

SPB duplication (Li et al., 2006). The absence of a Y2H interaction between the Sfi1 N terminus and Spc42 and Spc29 could suggest that a physical interaction does not exist. However, expression of the bait and prey protein fusion constructs needs to be confirmed before concluding a lack of interaction of exists. Alternately, posttranslational modifications could be required for an interaction to occur. As discussed in Chapter Two, I propose that Mps1 may play a key role in satellite formation. Thus, phosphorylation of the Sfi1 N terminus and/or satellite proteins may be required, and conducting a Y2H assay using phosphomimetic versions of these proteins may prove useful. Additionally, the presence or absence of a Y2H interaction of the

Sfi1 N terminus with the other satellite proteins, Cnm67 and Nud1, could be examined.

Furthermore, Cdc31 may have a regulatory role in the Sfi1 N terminus interactions. Thus, while it does not interact with Cdc31 in vitro (Kilmartin, 2003), the Sfi1 N terminus may require the presence of Cdc31 in order to recruit satellite components. Future examination of whether the

Sfi1 N terminus with Sfi1 repeats can interact via Y2H with satellite proteins will be worthwhile.

The Sfi1 C terminus is predicted to interact with itself. However, we did not see a Y2H interaction between BD- and AD-fused Sfi1 C terminus fusion constructs, with or without the addition of two Sfi1 repeats. Initially, confirmation of expression of the AD-Sfi1 C terminus fusion construct is needed to determine whether the Y2H lack of interaction is accurate. While an unmodified Sfi1 C terminus may not self-associate, a nonphosphorylatable version of the C 151 terminus may interact with itself. As seen in my work (see Chapter 3) and others (Anderson et al., 2007; Elserafy et al., 2014), a decrease in Cdk1 phosphorylation at the Sfi1 C terminus leads to unseparated SPBs and thus may block dissociation of the Sfi1 C termini at the bridge.

Furthermore, I have shown that this lack of phosphorylation promotes SPB reduplication (see

Chapter 3), leading to formation of a bridge and thus potentially association of Sfi1 C termini.

As suggested from my work, this decrease in phosphorylation may be required for Sfi1 C termini to interact.

Additionally, the Sfi1 C terminus may interact with other SPB components, such as the half-bridge membrane proteins Kar1 and Mps3, which could anchor Sfi1 to the bridge.

Interestingly, previous research has found that Kar1 overexpression could suppress the synthetic lethality of specific sfi1 C-terminal alleles with mad1∆, Mad1 being a SAC component

(Anderson et al., 2007), or bik1, an allele leading to reduced expression of the microtubule- associated protein Bik1 (Strawn and True, 2006). However, Anderson and colleagues did not see suppression of the synthetic lethality of sfi1 C-terminal alleles with mad1∆ upon overexpression of Mps3 (Anderson et al., 2007). Although, they did see suppression (using alleles that alter the Sfi1 C terminus or immediately upstream the C terminus) upon overexpression of the membrane anchor proteins Bbp1, Ndc1, and Mps2, proteins involved in

SPB insertion into the nuclear envelope (Winey et al., 1991, 1993; Schramm et al., 2000), leading the authors to propose these proteins may interact with the Sfi1 C terminus for SPB separation (Anderson et al., 2007). The half-bridge protein Cdc31 could also potentially interact with and/or regulate the Sfi1 C terminus. However, previous work has shown that the Sfi1 C terminus does not interact with Cdc31 in vitro, and genetic interactions have only been seen between CDC31 and SFI1 using sfi1 alleles that alter the repeat domain (Kilmartin, 2003; 152

Anderson et al., 2007) or the N terminus (this study; see Chapter Two), but not the C terminus.

Indeed, we did not observe an interaction of the Sfi1 C terminus with itself upon addition of two

Cdc31-binding Sfi1 repeats.

In a screen for Sfi1 C terminus Y2H interactors, we identified Cse2, a component of the mediator complex involved in transcription via RNA polymerase II (Gustafsson et al., 1998).

Cse2 is a promising candidate as a physical interactor with Sfi1. In a genome-scale analysis of synthetic genetic interactions, CSE2 and SFI1 were shown to genetically interact (using an sfi1 allele with alteration of the repeat domain) (Costanzo et al., 2010). Both Cse2 and Sfi1 have been shown to interact, either genetically or physically, with microtubule-related proteins. Via

Y2H, Cse2 interacts with Cin8 (Wong et al., 2007), a kinesin-related motor protein involved in separating duplicated SPBs (Hoyt et al., 1992; Roof et al., 1992). CIN8 also genetically interacts with SFI1 (Strawn and True, 2006, sfi1 C-terminal allele; Anderson et al., 2007, genetic interaction network). Furthermore, mutation of CSE2 leads to defects in chromosome segregation, and a majority of cells display a large-budded phenotype (Xiao et al., 1993).

Based on our findings and the body of research on Cse2, it will be interesting to determine whether Cse2 may associate with the SPB, microtubules, and/or microtubule-related proteins to play a role in appropriate bipolar spindle formation. It is interesting to note that immuno-EM indicates Sfi1 localizes to the cytoplasmic side of the half-bridge (Kilmartin, 2003;

Li et al., 2006), while Cse2 is nuclear (Dastidar et al., 2012). However, via immunoprecipitation, Sfi1 was previously shown to interact in vivo with Spc110, an SPB component located on the nuclear side of the nuclear envelope (Rout and Kilmartin, 1990).

Further validation of the Y2H interaction of Cse2 with the Sfi1 C terminus is first needed via analysis of expression of a second reporter, adenine expression, and examining BD-Cse2 153 interaction with AD-Sfi1. While Y2H is a valuable tool to initially identify potential physical interactors and to examine the interaction of domains and mutant constructs, identifying co- precipitating protein interactors of Sfi1 (full length, overexpressed domains, or mutants) in vivo via western blot or mass spectrometry will prove useful. Additionally, in vitro binding assays could be performed to assess direct binding for candidate interactors. If Y2H interactions are identified upon modification of proteins, these Y2H results will be informative in conducting the in vivo and in vitro interaction assays.

IV. Materials and Methods

Please refer to Tara Peters’ undergraduate honors thesis entitled, “Identification of Sfi1 protein interactions through yeast two-hybrid analysis,” for detailed materials and methods and strain information. The Gal4 DNA binding domain vector (pOBD2) constructs were transformed into PJ69-4a (trp1-901 leu2-3,112 ura3-52 his3-200 gal4∆ gal80∆ LYS2::GAL1-

HIS3 GAL2-ADE2 met2::GAL7-lacZ MATa) (James et al., 1996), and the Gal4 activating domain vector (pOAD) constructs were transformed into PJ69-4α (trp1-901 leu2-3,112 ura3-52 his3-200 gal4∆ gal80∆ LYS2::GAL1-HIS3 GAL2-ADE2 met2::GAL7-lacZ MATα). The AD-

ORF pool in PJ694a yeast contains an estimated 6,000 S. cerevisiae ORFs obtained by PCR and fused in frame with AD in pOAD. The transformed yeast were pooled to create the AD array

(Hudson et al., 1997; James et al., 1996). Vectors and the yeast AD-ORF pool were obtained from Stanley Fields through the Yeast Resource Center at the University of Washington. PJ69-

4a and PJ69-4α were a gift from Greg Odorizzi. Y2H assays described used the GAL1-HIS3 reporter to assess interactions via growth on media lacking histidine.

154

Appendix B: Identification of SFI1 genetic interactions via a dosage suppressor screen

I. Introduction

A model put forth by Kilmartin and colleagues based on Sfi1 localization suggests that

SPB duplication initiates when Sfi1 molecules associate with those already at the half-bridge through end-to-end interactions at the C termini, thus doubling the half-bridge length. After completion of SPB duplication, dissociation of the Sfi1 C termini at the bridge from each other would then allow SPB separation to occur (Li et al., 2006). While predicted, it is not known whether the Sfi1 C termini directly interact or interact with other SPB components, such as the half-bridge proteins Kar1 and Mps3. Additionally, while it is known that the Sfi1 C terminus is phosphorylated by Cdk1 (see Chapter 3) (Elserafy et al., 2014) and that Cdc14 dephosphorylates

Sfi1 and interacts with the Sfi1 C terminus via yeast two-hybrid (Elserafy et al., 2014), other potential regulators of the C terminus have not been identified.

A few studies examining sfi1 C-terminal alleles have identified proteins that may interact with Sfi1 or function in a process affected in these sfi1 alleles through examination of SFI1 genetic interactions. Previous research has found that Kar1 overexpression could suppress the synthetic lethality of specific sfi1 C-terminal alleles with mad1∆ (Anderson et al., 2007) or bik1

(Strawn and True, 2006). In regards to the half-bridge protein Cdc31, while overexpression of

CDC31 can suppress the phenotype of sfi1 alleles that alter the repeat domain (Kilmartin, 2003;

Anderson et al., 2007), it does not suppress that of sfi1 alleles that alter the C terminus (Anderson et al., 2007). Overexpression of Bbp1 and Ndc1, individually, but not Mps3, suppresses the synthetic lethality of sfi1 C-terminal alleles with mad1∆ (Anderson et al., 2007). Since Bbp1 and

Ndc1 are proteins involved in SPB insertion into the nuclear envelope (Winey et al., 1991, 1993; 155

Schramm et al., 2000), Anderson and colleagues suggested that these proteins may interact with the Sfi1 C terminus for SPB separation (Anderson et al., 2007).

Previous research has identified genetic interactions between SFI1 and microtubule- related protein-encoding genes (Strawn and True, 2006; Anderson et al., 2007, genetic interaction network). Studies have also found that sfi1 mutants with altered C-terminal Cdk sites show synthetic lethality with strains deleted for a component of the SAC (mad1∆, bub1∆, mad3∆, or mad2∆) (Anderson et al., 2007; Elserafy et al., 2014), suggesting that the SAC is activated in these alleles.

While multiple genetic interactions have been identified using sfi1 C-terminal alleles that result in a defect in SPB separation, genetic interactions of sfi1 alleles that also result in reduplicated SPBs have not previously been examined. The identification of genes that genetically interact via dosage suppression would provide a sound basis with which to further explore a potential physical interaction or regulatory relationship between proteins (Magtanong et al., 2011). In dosage suppression, overexpression of a gene leads to a decrease or loss of the defect associated with a specific allele. Thus, with the assistance of Melissa Phillips during her graduate rotation, I conducted a dosage suppressor screen using sfi1-C3A and examined dosage suppression of both sfi1-C3A and sfi1-C3A+S923A.

II. Results

Using sfi1-C3A, a dosage suppressor screen was performed using a pooled S. cerevisiae genomic library in a 2-µm LEU2 vector (high copy number) (Jones et al., 2008). While the library is reported to contain 1,588 clones, only 365 sfi1-C3A transformants containing a 2-µm vector were obtained. All transformants were tested for growth at 37°C. Twelve colonies in 156 which suppression of the growth defect at 37°C were identified. We were able to isolate and transform nine of the suppressing 2-µm plasmids into sfi1-C3A to confirm suppression by an individual plasmid.

As expected, the presence of an empty vector did not suppress the growth defect of sfi1-

C3A at 37°C, while overexpression of Sfi1 via a 2-µm SFI1 plasmid led to partial suppression of the growth defect at 37°C (Fig. B-1). Three of the nine 2-µm plasmids did not appear to suppress the sfi1-C3A growth defect. The six remaining plasmids (O23, T25, JJ29, SS34, TT40,

CCC44) were also transformed into sfi1-C3A+S923A. The genomic region contained in each of these six plasmids was identified (Table B-1). For each of the six 2-µm plasmids in both sfi1-

C3A and sfi1-C3A+S923A, growth of multiple yeast transformants using the same plasmid varied, and rescued growth at 37°C was typically not at the same level as that of a strain containing the 2-µm SFI1 (Fig. B-1). We do note that the 2-µm SFI1 plasmid in the sfi1-

C3A+S923A strain typically did not appear to rescue the growth defect (Fig. B-1). While generally similar between the two alleles, suppression of the growth defect by one 2-µm plasmid

(JJ29) was only seen in sfi1-C3A, not sfi1-C3A+S923A (Fig. B-1).

III. Discussion

Suppression was variable across transformants for each 2-µm plasmid and was only slight. We also noted that the 2-µm SFI1 plasmid in the sfi1-C3A+S923A strain typically did not appear to rescue the growth defect. However, sfi1-C3A+S923A is recessive (see Chapter 3), as determined via assessment of a heterozygous diploid. Thus, we would expect rescue of the growth defect using a high copy plasmid containing SFI1. We did however see rescue in sfi1-

C3A. In general, suppression results were similar between the two sfi1 alleles tested. 157

Nonetheless, some transformants do appear to show suppression, and thus further testing of suppression by individual ORFs may be of interest to pursue. For example, Nup145 (ORF contained in O23) is a nuclear pore protein (Wente and Blobel, 1994). Mad1 (ORF contained in

O23) is a component of the SAC (Li and Murray, 1991; Hardwick and Murray, 1995), and

MAD1 has previously been shown to genetically interact with SFI1 (using sfi1 C-terminal alleles;

Anderson et al., 2007). Sog2 (ORF contained in T25) is required for appropriate cell morphology (Nelson et al., 2003), and Ycg1 (ORF contained in JJ29) is suggested to be a component of the condensin complex (Ouspenski et al., 2000).

IV. Materials and Methods

pRS425-SFI1 (E2463) was created via cloning using pRS425 and pRS305-SFI1 (E2322) digested with BamHI-HF and XhoI. The empty vector was pGP564 (Thermo Scientific Open

Biosystems).

For the dosage suppressor screen, 100 ng of a 2-µm-LEU2 pooled yeast genomic library

(pGFP564 vector; Thermo Scientific Open Biosystems Yeast Genomic Tiling Collection Assay

Ready DNA; designed by Greg Prelich, Albert Einsten College of Medicine) (Jones et al., 2008) was transformed into sfi1-C3A (JA188) via standard protocol (Gietz and Woods, 2002). This library is reported to contain 1,588 clones covering 97% of the genome length. The entire transformation was plated on SC-Leu plates at 24°C. The total yield was 365 transformants.

The transformants were replica plated to YPD plates and grown at 37°C. After six days at 37°C, single colonies from these plates were patched to SC-Leu plates at 24°C and 37°C. These plates were then replica plated to YPD plates grown at 37°C to obtain candidate suppressors. Strains were preserved on SC-Leu plates at 24°C. 158

Plasmids from candidate suppressor yeast strains were recovered (Amberg et al., 2005) and transformed via electroporation into electrocompetent TOP10 E. coli cells. Sequencing with the following two primers was performed to identify the genomic sequence present in each candidate suppressor plasmid: pGP564seqF (P122): 5’-AGCGGATAACAATTTCACACAGGA-

3’ and pGP564seqF (P123): 5’-TAAGTTGGGTAACGCCAGGG-3’. Sequencing blasts were performed using the S. cerevisiae WU-BLAST2 search tool in the Saccharomyces genome database (Stanford, CA). The 2-µm plasmids were transformed into sfi1-C3A (JA188) and sfi1-

C3A+S923A (JA184) to confirm and examine suppression by growth on YPD and SC-Leu.

159

Figure B-1. Candidate dosage suppressors of sfi1-3A and sfi1-C3A+S923A Growth after two days on SC-Leu or YPD plates at 24°C and 37°C of eight yeast transformants for each sfi1 allele using a single 2-µm plasmid: O23 (A), T25 (B), JJ29 (C), SS34 (D), TT40 (E), and CCC44 (F). Schematic shows numbered identifiers of individual transformants as well 160

Figure B-1 (continued). as controls. Left panel: sfi1-C3A, C1: sfi1-C3A containing empty vector (Y4630), C2: sfi1-C3A containing 2-µm SFI1 (JA204). Right panel: sfi1-C3A+S923A, C-: sfi1-C3A+S923A containing empty vector (Y4647), C+: sfi1-C3A+S923A containing 2-µm SFI1 (Y4648). 161

Table B-1. Genomic region of candidate 2-µm suppressors “Library Identifier” indicates the identification of the corresponding clone number in the 2-µm library (Jones et al., 2008). Chromosome region indicated is that obtained through sequencing, which is very similar to that identified by Prelich and colleagues (Jones et al., 2008). ORFs identified in library clone. (-5’): missing 5’ end of gene. (-3’): missing 3’ end of gene. Note: A single genomic region for CCC44 could not be identified.

Name Library Chromosome Region ORFs Identifier

O23 YGPM23m09 VII: 338,555-349,221 NUP145 (-5'), NBP35, LIF1, MF(ALPHA)2, YGL088W, snR10, MMS2, MAD1 (-3')

T25 YGPM17c01 XV: 998,397-1,010,736 SOG2, MSC6, GDS1, YOR356W, GRD19, HAP5 (-3')

JJ29 YGPM7m05 IV: 1,112,412-1,122,278 PEP7, UTP4, YCG1, YDR326C (-5')

SS34 YGPM8g21 XII: 64,971-76,526 UBI4 (-3'), ENT4, YLL037W, PRP19, GRC3, RIX7, YLL033W, YLL032C (-5')

TT40 YGPM14h08 XIII: 369,151-381,506 ERB1 (-3'), tV(AAC)M1, YMR050C, YMR051C, tW(CCA)M, FAR3, YMR052C-A, STB2 (-5')

162

Table B-2. Yeast strains

Strain Genotype JA204 sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 2-µm- SFI1-LEU2 MATa/MATa Y4630 sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 2-µm- LEU2 MATa/MATa Y4634, Y4635 sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 2-µm-O23- LEU2 MATa/MATa Y4636, Y4637 sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 2-µm-T25- LEU2 MATa/MATa Y4638, Y4639 sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 2-µm- JJ29-LEU2 MATa/MATa Y4641, Y4642 sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 2-µm- SS34-LEU2 MATa/MATa Y4643, Y4644 sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 2-µm- TT40-LEU2 MATa/MATa Y4645, Y4646 sfi1-C3A-NATMX/sfi1-C3A-NATMX ade2::ADE2/ade2::ADE2 2-µm- CCC44-LEU2 MATa/MATa Y4647 sfi1-C3A+S923A-NATMX ade2::ADE2 2-µm-SFI1-LEU2 MATa Y4648 sfi1-C3A+S923A-NATMX ade2::ADE2 2-µm-LEU2 MATa Y4650 sfi1-C3A+S923A-NATMX ade2::ADE2 2-µm-O23-LEU2 MATa Y4651, Y4652 sfi1-C3A+S923A-NATMX ade2::ADE2 2-µm-T25-LEU2 MATa Y4653 sfi1-C3A+S923A-NATMX ade2::ADE2 2-µm-JJ29-LEU2 MATa Y4655, Y4656 sfi1-C3A+S923A-NATMX ade2::ADE2 2-µm-SS34-LEU2 MATa Y4657, Y4658 sfi1-C3A+S923A-NATMX ade2::ADE2 2-µm-TT40-LEU2 MATa Y4659, Y4660 sfi1-C3A+S923A-NATMX ade2::ADE2 2-µm-CCC44-LEU2 MATa

163

Table B-3. Plasmids

Identifier Name Genotype E2463 pRS425-SFI1 2-µm-SFI1 pGP564 pGP564 2-µm-LEU2 E2627 pGP564-O23 2-µm-O23-LEU2 E2628 pGP564-T25 2-µm-T25-LEU2 E2629 pGP564-JJ29 2-µm-JJ29-LEU2 E2631 pGP564-SS34 2-µm-SS34-LEU2 E2632 pGP564-TT40 2-µm-TT40-LEU2 E2633 pGP564-CCC44 2-µm-CCC44-LEU2

164

Appendix C: Sfi1 domain overexpression and localization

I. Introduction

The domains of Sfi1 are proposed to have distinct functions. Sfi1 is positioned on the cytoplasmic side of the half-bridge in an orientation-specific manner with the N terminus proximal and the C terminus distal to the SPB (Li et al., 2006). Based on this topology,

Kilmartin and colleagues have suggested that the Sfi1 C terminus is involved in the initiation of

SPB duplication as well as in SPB separation. The Sfi1 N terminus is proposed to recruit the new SPB. To date, no published work has reported a function for the Sfi1 N terminus, although the genetic interactions I observed between SFI1 using an N-terminal allele and the satellite protein-encoding genes SPC42 and SPC29 suggest a potential role in satellite recruitment (see

Chapter 2), supporting the proposed model by Kilmartin and colleagues (Li et al., 2006). The

Sfi1 repeat domain is a conserved, elongated alpha-helical domain containing 21 Sfi1 repeats, each repeat binding one molecule of centrin (Cdc31), and is proposed to have a structural role at the half-bridge (Li et al., 2006). The Sfi1 repeat domain has been shown to be required prior to satellite formation and may be required for SPB duplication after satellite formation as well

(Kilmartin, 2003). The Sfi1 C terminus is required for SPB separation (Kilmartin, 2003;

Anderson et al., 2007; Elserafy et al., 2014; Strawn and True, 2006) (see Chapter 3). Additional research suggests that it is also required for SPB duplication after satellite formation (Elserafy et al., 2014). My work also shows that the C terminus is required to block SPB reduplication (see

Chapter 3).

In order to obtain an initial insight into potential roles of or interactors with the individual domains of Sfi1, I, along with Zach Wilson and Jaimee Hoefert during their graduate rotations, 165 examined the effects of overexpression of Sfi1 domains. Using this experimental approach, we determined whether each domain or a modified version of the domain could act as a dominant negative. If a dominant negative effect is seen, examination of the processes affected and proteins potentially interacting with the overexpressed domain can be conducted.

II. Results

SFI1 domain overexpression constructs were integrated into the genome, with galactose- inducible expression under control of the GAL1 promoter. Expression is repressed in the presence of glucose. Using this system with or without a C-terminal GFP tag, we found that overexpression of the Sfi1 repeat domain (GAL1-SFI1-R) is toxic (Fig. C-1). Expression of GFP constructs was confirmed (Fig. C-2). Upon closer examination of the effect of overexpression of the Sfi1 repeat domain, we found that the majority of cells were unbudded and contained G1

DNA content (from one experiment; Fig. C-3). No effect on growth results from overexpression of the Sfi1 N and C termini, individually (Fig. C-1). Overexpression of a nonphosphorylatable version of the N terminus (sfi1-N-NnonP) or a version of the C terminus with reduced phosphorylation sites (sfi1-C-C3A+S923A) did not affect growth.

Examination of GFP-tagged overexpressed domains revealed that no single domain clearly localized to the SPB (Fig. C-4). Expression of both GAL1-GFP and GAL1-SFI1-

C-GFP was generally diffuse and appeared cytoplasmic, although DNA staining would be needed for a more clear determination of localization. The Sfi1 N terminus displayed both distinct GFP punctae as well as some localized diffuse GFP expression. The Sfi1 repeat domain displayed aggregates, which sometimes appeared crescent-shaped, in the cells. The majority of 166 large-budded cells with overexpression of SFI1-R-GFP had two SPBs. However, some cells did display a single SPB (Fig. C-4).

III. Discussion

Overexpression of the Sfi1 repeat domain is toxic to cells and thus has a dominant negative effect, while the Sfi1 N- and C-terminal domains, both wild-type as well as modified versions, do not appear to affect cells. Given that Sfi1 repeats bind Cdc31 (centrin) (Kilmartin,

2003), it is plausible that the overexpressed repeat domain could be sequestering Cdc31 from the

SPB, as the repeat domain does not appear to localize to the SPB. Indeed, a small portion of large-budded cells with overexpression of SFI1-R-GFP had a single SPB, suggesting a possible

SPB duplication or separation defect. Cdc31 is known to be required for SPB duplication

(Byers, 1981; Winey et al., 1991). However, the majority of GAL1-SFI1-R-GFP cells had two

SPBs. In addition to a role in SPB duplication, Cdc31 is also proposed to be involved in nuclear mRNA export through association with the Sac1-Thp1-Sus1 complex at the nuclear pore complexes and is required for cell integrity and morphogenesis (Sullivan et al., 1998; Fischer et al., 2004). Furthermore, in humans, centrin 2 is a component of the Xeroderma Pigmentosum

Group C Complex (XPC) involved in nucleotide excision repair (NER) (Araki et al., 2001).

Thus, the overexpressed Sfi1 repeat domain could potentially affect a function of Cdc31 other that of SPB duplication. Further examination of the phenotype of cells with overexpression of

SFI1-R is therefore needed. Examination of Cdc31 localization upon galactose induction of repeat domain overexpression would prove useful. Additionally, examining whether Cdc31 binds the overexpressed Sfi1 repeat domain should be examined. It would also be interesting to determine whether overexpression of Cdc31 could overcome the defect seen upon 167 overexpression of the Sfi1 repeat domain. Furthermore, given the localization of the repeat domain, colocalization experiments with membrane proteins may provide additional insight into whether the repeat domain is membrane-associated.

We do not see localization of the Sfi1 N and C termini at the SPB or defects in growth resulting from their overexpression. However, the N terminus does display distinct GFP punctae, so examination of localization with respect to SPB proteins other than Spc42 would be useful. While not known, if Cdc31 plays any regulatory role in the function of the Sfi1 termini, addition of multiple repeat domains to these constructs may prove useful. Furthermore, while one version of the C terminus with reduced phosphorylation sites, sfi1-C-C3A+S923A, did not affect growth, utilizing the sfi1-C4A, sfi1-C4A+S923A, or sfi1-C5A C terminus may prove useful, as these alleles are dominant using the full-length ORF at endogenous expression levels, while sfi1-C3A+S923A is recessive (see Chapter Three). Additionally, in order to determine the region of Sfi1 required for SPB localization, a deletion series could be constructed.

IV. Materials and Methods

Strain construction

The Sfi1 N terminus is defined as containing residues 1 to 185, all residues prior to the first Sfi1 repeat as defined by Kilmartin (2003). The Sfi1 repeat domain (residues 186 to 801) contains 21 conserved Sfi1 repeat sequences (Kilmartin, 2003; Li et al., 2006). The C terminus is defined as all residues (802 to 946) immediately following the final Sfi1 repeat sequence

(Anderson et al., 2007). SFI1-N-Control contains a nucleotide change of C99G, creating a silent mutation for plasmid manipulation purposes. This sequence change is also present in SFI1-N-

NnonP. 168

Plasmids E2425 through E2432 were obtained by inserting the appropriate PCR product

(using the appropriate plasmids described in Chapter 3) into pRS303-GAL1-GST using the restriction enzymes PacI and XmaI for GST constructs and PacI and Asc1 for constructs lacking

GST. E2438/E2439 was created using E2432 and E2434 digested with XmaI and AscI.

E2440/E2441 was created using E2431 and E2432 digested with XmaI and AscI. E2442/E2443 was created using pFA6-KANMX-GAL-yeGFP3 digested with BglII and SalI and pRS303-GAL1.

E2444/E2445 was created by inserting the appropriate PCR product into E2443 digested with

PacI and XmaI. All plasmids contain a linker (sequence PINPG), which contains an XmaI restriction enzyme site, downstream of the GAL1 promoter sequence and the inserted SFI1 domain sequence upstream of GST or GFP sequence, if present.

Relevant strains Y4465 through Y4480 were created via transformation via standard protocol (Gietz and Woods, 2002) of the appropriate plasmid into LPY5 (obtained from Sandy

Garcia). Strains Y4536 through Y4549 were created via transformation of the appropriate plasmid into SLJ2471 (obtained from Sue Jaspersen).

Growth techniques

Asynchronous cell cultures were grown in YEP-3% Raffinose at 30°C overnight to mid- log phase. Cells were then washed and resuspended in YEP-3% Galactose or YPD (2%

Glucose) an estimated 4 h (or 2 h for imaging).

Flow cytometry

Cells were fixed with 70% ethanol for 1 h at room temperature or at 4°C overnight.

Fixed cells were then treated with 0.1% RNase in 0.2M Tris-HCl pH 7.5, 20 mM EDTA for 2 h at 37°C, and the DNA was stained with 50 µg/mL propidium iodide (Sigma Chemical Co., St.

Louis, MO) in PBS for 1 h at room temperature or at 4°C overnight. DNA content was analyzed 169 using the CyAn ADP analyzer (Beckman Coulter, Indianapolis, IN, USA) with a 488nm laser.

30,000 events per sample were taken. DNA content (propidium iodide area) is indicated on X axes and Count on Y axes.

Cytological techniques

Imaging of cells was performed at room temperature on an Eclipse Ti inverted microscope (Nikon, Japan) fitted with a CFI Plan Apo VC 60× H numerical aperture 1.4 objective (Nikon, Japan) and a CoolSNAP HQ2 charge-coupled device camera (Photometrics,

Tuscon, AZ). Metamorph Imaging software (Molecular Devices, Sunnyvale, CA) was used to collect images.

Protein isolation and expression detection

Samples were prepared for western blot analysis as previously described (Horvath and

Riezman, 1994; Kushnirov, 2000). Samples were loaded on an 8.5% gel. A Benchmark prestained protein ladder (Life Technologies) was also loaded. A mouse monoclonal GFP antibody (Covance Inc.) at 1:5,000 and a goat anti-mouse Alexa 680 secondary antibody at

1:10,000 were used. The LI-COR Odyssey Imaging System (LI-COR Biosciences, Lincoln, NE) was used for detection.

170

Figure C-1. Growth phenotypes resulting from overexpression of individual Sfi1 domains A. Growth after two days at 24°C. Two transformants for GAL1-SFI1-N are shown. B. Growth after 2 days at 30°C. All strains contain SPC42-mCherry. Note: GAL1-SFI1 domain strains lacking the GFP tag showed similar growth to the corresponding tagged domain strain.

171

Figure C-2. GFP-tagged Sfi1 domains are expressed All strains contain SPC42-mCherry. Asynchronous cultures were grown in YEP-3% Raffinose at 30°C to mid-log phase. Cells were then washed and resuspended in YEP-3% Galactose or YPD (2% Glucose) for 4 h. Samples were then collected and used for western blot analysis using a GFP monoclonal antibody.

172

Figure C-3. DNA content of cells overexpressing GFP-tagged Sfi1 domains Asynchronous cultures were grown in YEP-3% Raffinose at 30°C to mid-log phase. Cells were then washed and resuspended in YEP-3% Galactose or YPD (2% Glucose) for 4 h. DNA content via flow cytometry from one experiment with G1 DNA content indicated. For GAL1- SFI1-R-GFP, budding analysis was performed on fixed samples. The percentage of unbudded, small-budded, and large-budded cells were 25%, 44%, and 31%, respectively, in glucose (n=108) and 60%, 26%, and 14%, respectively, in galactose (n=100).

173

Figure C-4. Localization of Sfi1 domains Asynchronous cultures were grown in YEP-3% Raffinose at 30°C to mid-log phase. Cells were then washed and resuspended in YEP-3% Galactose or YPD (2% Glucose) for 2 h. Cells grown in galactose were imaged from one experiment. The SPC42-mCherry strain grown in YPD was imaged. Localization of Spc42-mCherry (red) and Sfi1-GFP (green) and a merged image. GFP alone (n=24) and Sfi1-C-GFP (n=40) expression was diffuse and generally appeared cytoplasmic. Sfi1-N-GFP showed GFP punctae often (n=24 of 57 cells), and the remaining cells typically showed localized diffuse GFP. The Sfi1 repeat domain displayed GFP aggregates (n=41 of 42 cells). Of the large-budded GAL1-SFI1-R-GFP cells with Spc42-mCherry signal (n=15), four had 1 SPB and 11 had two SPBs. In GAL1-GFP cells, 100% of large-budded cells (n=10) had two SPBs. Note: scale not available for these images.

174

Table C-1. Yeast strains

Strain Genotype LPY5 leu2-3,112 trp1-1 can1-100 ura3-1 ade2-1 his3-11,15 MATa SLJ2471 SPC42::SPC42-mCherry-URA3 MATα Y4465, Y4466 his3::GAL1-SFI1-C-HIS3 MATa Y4469, Y4470 his3::GAL1-SFI1-R-HIS3 MATa Y4471, Y4472 his3::GAL1-SFI1-N-nonPN-HIS3 MATa Y4473, Y4474 his3::GAL1-SFI1-N-HIS3 MATa Y4475, Y4476 his3::GAL1-SFI1-N-Control-HIS3 MATa Y4477, Y4478 his3::GAL1-SFI1-C-C4A-HIS3 MATa Y4479, Y4480 his3::GAL1-SFI1-HIS3 MATa Y4536, Y4537 his3::GAL1-SFI1-C-HIS3 SPC42::SPC42-mCherry-URA3 MATα Y4538, Y4539 his3::GAL1-SFI1-R-HIS3 SPC42::SPC42-mCherry-URA3 MATα Y4540, Y4541 his3::GAL1-SFI1-N-HIS3 SPC42::SPC42-mCherry-URA3 MATα Y4542, Y4543 his3::GAL1-yeGFP3-HIS3 SPC42::SPC42-mCherry-URA3 MATα Y4544, Y4545 his3::GAL1-SFI1-N-yeGFP3-HIS3 SPC42::SPC42-mCherry-URA3 MATα Y4546, Y4547 his3::GAL1-SFI1-R-yeGFP3-HIS3 SPC42::SPC42-mCherry-URA3 MATα Y4548, Y4549 his3::GAL1-SFI1-C-yeGFP3-HIS3 SPC42::SPC42-mCherry-URA3 MATα

175

Table C-2. Plasmids

Identifier Name Genotype E2425 pRS303-GAL1-SFI1-C GAL1-SFI1-C-HIS3 E2426 pRS303-GAL1-SFI1-R GAL1-SFI1-R-HIS3 E2427 pRS303-GAL1-SFI1-N GAL1-SFI1-N-HIS3 E2428 pRS303-GAL1-sfi1-C-C4A GAL1-sfi1-C-C4A-HIS3 E2429 pRS303-GAL1-sfi1-N- GAL1-sfi1-N-NnonP-HIS3 NnonP E2430 pRS303-GAL1-SFI1-N- GAL1-SFI1-N-Control- Control HIS3 E2431 pRS303-GAL1-SFI1-C-GST GAL1-SFI1-C-GST-HIS3 E2432 pRS303-GAL1-SFI1-N-GST GAL1-SFI1-N-GST-HIS3 E2438, E2439 pRS303-GAL1-SFI1-N- GAL1-SFI1-N-yeGFP3- yeGFP3 HIS3 E2440, E2441 pRS303-GAL1-SFI1-C- GAL1-SFI1-C-yeGFP3- yeGFP3 HIS3 E2442, E2443 pRS303-GAL1-yeGFP3 GAL1-yeGFP3-HIS3 E2444, E2445 pRS303-GAL1-SFI1-R- GAL1-SFI1-R-yeGFP3- yeGFP3 HIS3

176

References

Adams, I.R., and J.V. Kilmartin. 1999. Localization of core spindle pole body (SPB) components during SPB duplication in Saccharomyces cerevisiae. J. Cell Biol.. 145:809–823.

Amberg, D.C., D.J. Burke, and J.N. Strathern, 2005. Methods in yeast genetics: A Cold Spring Harbor Laboratory course manual.Cold Spring Harbor Laboratory Press.

Anderson, V.E., J. Prudden, S. Prochnik, T.H. Giddings Jr, and K.G. Hardwick. 2007. Novel sfi1 alleles uncover additional functions for Sfi1p in bipolar spindle assembly and function. Mol. Biol. Cell. 18:2047–2056. doi:10.1091/mbc.E06-10-0918.

Araki, M., C. Masutani, M. Takemura, A. Uchida, K. Sugasawa, J. Kondoh, Y. Ohkuma, and F. Hanaoka. 2001. Centrosome protein centrin 2/caltractin 1 is part of the xeroderma pigmentosum group C complex that initiates global genome nucleotide excision repair. J. Biol. Chem.. 276:18665–18672. doi:10.1074/jbc.M100855200.

Araki, Y., L. Gombos, S.P.S. Migueleti, L. Sivashanmugam, C. Antony, and E. Schiebel. 2010. N-terminal regions of Mps1 kinase determine functional bifurcation. J. Cell Biol.. 189:41–56. doi:10.1083/jcb.200910027.

Araki, Y., C.K. Lau, H. Maekawa, S.L. Jaspersen, T.H. Giddings, E. Schiebel, and M. Winey. 2006. The Saccharomyces cerevisiae spindle pole body (SPB) component Nbp1p is required for SPB membrane insertion and interacts with the integral membrane proteins Ndc1p and Mps2p. Mol. Biol. Cell. 17:1959–1970. doi:10.1091/mbc.E05-07-0668.

Azimzadeh, J., and M. Bornens. 2007. Structure and duplication of the centrosome. J. Cell Sci.. 120:2139–2142. doi:10.1242/jcs.005231.

Balestra, F.R., P. Strnad, I. Flückiger, and P. Gönczy. 2013. Discovering regulators of centriole biogenesis through siRNA-based functional genomics in human cells. Dev. Cell. 25:555– 571. doi:10.1016/j.devcel.2013.05.016.

Barrett, J.G., B.D. Manning, and M. Snyder. 2000. The Kar3p kinesin-related protein forms a novel heterodimeric structure with its associated protein Cik1p. Mol. Biol. Cell. 11:2373– 2385.

Baum, P., C. Yip, L. Goetsch, and B. Byers. 1988. A yeast gene essential for regulation of spindle pole duplication. Mol. Cell. Biol.. 8:5386–5397.

Berlin, V., C.A. Styles, and G.R. Fink. 1990. BIK1, a protein required for microtubule function during mating and mitosis in Saccharomyces cerevisiae, colocalizes with tubulin. J. Cell Biol.. 111:2573–2586.

Bettencourt-Dias, M., and D.M. Glover. 2007. Centrosome biogenesis and function: centrosomics brings new understanding. Nat. Rev. Mol. Cell Biol.. 8:451–463. doi:10.1038/nrm2180. 177

Bettencourt-Dias, M., A. Rodrigues-Martins, L. Carpenter, M. Riparbelli, L. Lehmann, M.K. Gatt, N. Carmo, F. Balloux, G. Callaini, and D.M. Glover. 2005. SAK/PLK4 Is Required for Centriole Duplication and Flagella Development. Curr. Biol.. 15:2199–2207. doi:10.1016/j.cub.2005.11.042.

Bloom, J., I.M. Cristea, A.L. Procko, V. Lubkov, B.T. Chait, M. Snyder, and F.R. Cross. 2011. Global analysis of Cdc14 phosphatase reveals diverse roles in mitotic processes. J. Biol. Chem.. 286:5434–5445. doi:10.1074/jbc.M110.205054.

Breitkreutz, A., H. Choi, J.R. Sharom, L. Boucher, V. Neduva, B. Larsen, Z.-Y. Lin, B.-J. Breitkreutz, C. Stark, G. Liu, J. Ahn, D. Dewar-Darch, T. Reguly, X. Tang, R. Almeida, Z.S. Qin, T. Pawson, A.-C. Gingras, A.I. Nesvizhskii, and M. Tyers. 2010. A global protein kinase and phosphatase interaction network in yeast. Science. 328:1043–1046. doi:10.1126/science.1176495.

Bremmer, S.C., H. Hall, J.S. Martinez, C.L. Eissler, T.H. Hinrichsen, S. Rossie, L.L. Parker, M.C. Hall, and H. Charbonneau. 2012. Cdc14 phosphatases preferentially dephosphorylate a subset of cyclin-dependent kinase (Cdk) sites containing phosphoserine. J. Biol. Chem.. 287:1662–1669. doi:10.1074/jbc.M111.281105.

Brown, T. 2004. Analysis of DNA sequences by blotting and hybridization. Ausubel, F.M., R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, Eds. Current protocols in molecular biology. John Wiley and Sons, Inc. 2.9.1-2.9.20.

Byers, B. 1981. Molecular Genetics in Yeast. Alfred Benzon Symposium. 16:119-133.

Byers, B., and L. Goetsch. 1974. Duplication of spindle plaques and integration of the yeast cell cycle. Cold Spring Harb. Symp. Quant. Biol.. 38:123–131.

Castillo, A.R., J.B. Meehl, G. Morgan, A. Schutz-Geschwender, and M. Winey. 2002. The yeast protein kinase Mps1p is required for assembly of the integral spindle pole body component Spc42p. J. Cell Biol.. 156:453–465. doi:10.1083/jcb.200111025.

Chi, A., C. Huttenhower, L.Y. Geer, J.J. Coon, J.E.P. Syka, D.L. Bai, J. Shabanowitz, D.J. Burke, O.G. Troyanskaya, and D.F. Hunt. 2007. Analysis of phosphorylation sites on proteins from Saccharomyces cerevisiae by electron transfer dissociation (ETD) mass spectrometry. Proc. Natl. Acad. Sci. U. S. A.. 104:2193–2198. doi:10.1073/pnas.0607084104.

Ciosk, R., W. Zachariae, C. Michaelis, A. Shevchenko, M. Mann, and K. Nasmyth. 1998. An ESP1/PDS1 complex regulates loss of sister chromatid cohesion at the metaphase to anaphase transition in yeast. Cell. 93:1067–1076.

Cizmecioglu, O., M. Arnold, R. Bahtz, F. Settele, L. Ehret, U. Haselmann-Weiss, C. Antony, and I. Hoffmann. 2010. Cep152 acts as a scaffold for recruitment of Plk4 and CPAP to the centrosome. J. Cell Biol.. 191:731–739. doi:10.1083/jcb.201007107. 178

Cohen-Fix, O., and D. Koshland. 1999. Pds1p of budding yeast has dual roles: inhibition of anaphase initiation and regulation of mitotic exit. Genes Dev.. 13:1950–1959.

Cohen-Fix, O., J.M. Peters, M.W. Kirschner, and D. Koshland. 1996. Anaphase initiation in Saccharomyces cerevisiae is controlled by the APC-dependent degradation of the anaphase inhibitor Pds1p. Genes Dev.. 10:3081–3093.

Collins, S.R., K.M. Miller, N.L. Maas, A. Roguev, J. Fillingham, C.S. Chu, M. Schuldiner, M. Gebbia, J. Recht, M. Shales, H. Ding, H. Xu, J. Han, K. Ingvarsdottir, B. Cheng, B. Andrews, C. Boone, S.L. Berger, P. Hieter, Z. Zhang, G.W. Brown, C.J. Ingles, A. Emili, C.D. Allis, D.P. Toczyski, J.S. Weissman, J.F. Greenblatt, and N.J. Krogan. 2007. Functional dissection of protein complexes involved in yeast chromosome biology using a genetic interaction map. Nature. 446:806–810. doi:10.1038/nature05649.

Compton, D.A. 2000. Spindle Assembly in Animal Cells. Annu. Rev. Biochem.. 69:95–114. doi:10.1146/annurev.biochem.69.1.95.

Costanzo, M., A. Baryshnikova, J. Bellay, Y. Kim, E.D. Spear, C.S. Sevier, H. Ding, J.L.Y. Koh, K. Toufighi, S. Mostafavi, J. Prinz, R.P. St Onge, B. VanderSluis, T. Makhnevych, F.J. Vizeacoumar, S. Alizadeh, S. Bahr, R.L. Brost, Y. Chen, M. Cokol, R. Deshpande, Z. Li, Z.-Y. Lin, W. Liang, M. Marback, J. Paw, B.-J. San Luis, E. Shuteriqi, A.H.Y. Tong, N. van Dyk, I.M. Wallace, J.A. Whitney, M.T. Weirauch, G. Zhong, H. Zhu, W.A. Houry, M. Brudno, S. Ragibizadeh, B. Papp, C. Pál, F.P. Roth, G. Giaever, C. Nislow, O.G. Troyanskaya, H. Bussey, G.D. Bader, A.-C. Gingras, Q.D. Morris, P.M. Kim, C.A. Kaiser, C.L. Myers, B.J. Andrews, and C. Boone. 2010. The genetic landscape of a cell. Science. 327:425–431. doi:10.1126/science.1180823.

Dastidar, R.G., J. Hooda, A. Shah, T.M. Cao, R.M. Henke, and L. Zhang. 2012. The nuclear localization of SWI/SNF proteins is subjected to oxygen regulation. Cell Biosci.. 2:30. doi:10.1186/2045-3701-2-30.

Diffley, J.F., J.H. Cocker, S.J. Dowell, and A. Rowley. 1994. Two steps in the assembly of complexes at yeast replication origins in vivo. Cell. 78:303–316.

Donaldson, A.D., and J.V. Kilmartin. 1996. Spc42p: a phosphorylated component of the S. cerevisiae spindle pole body (SPD) with an essential function during SPB duplication. J. Cell Biol.. 132:887–901.

Eissler, C.L., G. Mazón, B.L. Powers, S.N. Savinov, L.S. Symington, and M.C. Hall. 2014. The Cdk/Cdc14 Module Controls Activation of the Yen1 Holliday Junction Resolvase to Promote Genome Stability. Mol. Cell. 54:80–93. doi:10.1016/j.molcel.2014.02.012.

Elliott, S., M. Knop, G. Schlenstedt, and E. Schiebel. 1999. Spc29p is a component of the Spc110p subcomplex and is essential for spindle pole body duplication. Proc. Natl. Acad. Sci. U. S. A.. 96:6205–6210.

Elsasser, S., Y. Chi, P. Yang, and J.L. Campbell. 1999. Phosphorylation controls timing of Cdc6p destruction: A biochemical analysis. Mol. Biol. Cell. 10:3263–3277. 179

Elserafy, M., M. Sarić, A. Neuner, T.-C. Lin, W. Zhang, C. Seybold, L. Sivashanmugam, and E. Schiebel. 2014. Molecular Mechanisms that Restrict Yeast Centrosome Duplication to One Event per Cell Cycle. Curr. Biol. CB. doi:10.1016/j.cub.2014.05.032.

Enserink, J.M., and R.D. Kolodner. 2010. An overview of Cdk1-controlled targets and processes. Cell Div.. 5:11. doi:10.1186/1747-1028-5-11.

Fields, S., and O. Song. 1989. A novel genetic system to detect protein-protein interactions. Nature. 340:245–246. doi:10.1038/340245a0.

Fischer, T., S. Rodríguez-Navarro, G. Pereira, A. Rácz, E. Schiebel, and E. Hurt. 2004. Yeast centrin Cdc31 is linked to the nuclear mRNA export machinery. Nat. Cell Biol.. 6:840– 848. doi:10.1038/ncb1163.

Fitch, I., C. Dahmann, U. Surana, A. Amon, K. Nasmyth, L. Goetsch, B. Byers, and B. Futcher. 1992. Characterization of four B-type cyclin genes of the budding yeast Saccharomyces cerevisiae. Mol. Biol. Cell. 3:805–818.

Ganem, N.J., S.A. Godinho, and D. Pellman. 2009. A mechanism linking extra centrosomes to chromosomal instability. Nature. 460:278–282. doi:10.1038/nature08136.

Giddings, T.H., E.T. O’Toole, M. Morphew, D.N. Mastronarde, J.R. McIntosh, and M. Winey. 2001. Using rapid freeze and freeze-substitution for the preparation of yeast cells for electron microscopy and three-dimensional analysis. Methods Cell Biol.. 67:27–42.

Gietz, R.D., and R.A. Woods. 2002. Transformation of yeast by lithium acetate/single-stranded carrier DNA/polyethylene glycol method. Methods Enzymol.. 350:87–96.

Gönczy, P. 2012. Towards a molecular architecture of centriole assembly. Nat. Rev. Mol. Cell Biol.. 13:425–435. doi:10.1038/nrm3373.

Gustafsson, C.M., L.C. Myers, J. Beve, H. Spåhr, M. Lui, H. Erdjument-Bromage, P. Tempst, and R.D. Kornberg. 1998. Identification of new mediator subunits in the RNA polymerase II holoenzyme from Saccharomyces cerevisiae. J. Biol. Chem.. 273:30851– 30854.

Haase, S.B., M. Winey, and S.I. Reed. 2001. Multi-step control of spindle pole body duplication by cyclin-dependent kinase. Nat. Cell Biol.. 3:38–42. doi:10.1038/35050543.

Hardwick, K.G., and A.W. Murray. 1995. Mad1p, a phosphoprotein component of the spindle assembly checkpoint in budding yeast. J. Cell Biol.. 131:709–720.

Hatch, E.M., A. Kulukian, A.J. Holland, D.W. Cleveland, and T. Stearns. 2010. Cep152 interacts with Plk4 and is required for centriole duplication. J. Cell Biol.. 191:721–729. doi:10.1083/jcb.201006049. 180

Hinchcliffe, E.H., C. Li, E.A. Thompson, J.L. Maller, and G. Sluder. 1999. Requirement of Cdk2-cyclin E activity for repeated centrosome reproduction in Xenopus egg extracts. Science. 283:851–854.

Holinger, E.P., W.M. Old, T.H. Giddings, C. Wong, J.R. Yates, and M. Winey. 2009. Budding yeast centrosome duplication requires stabilization of Spc29 via Mps1-mediated phosphorylation. J. Biol. Chem.. 284:12949–12955. doi:10.1074/jbc.M900088200.

Horvath, A., and H. Riezman. 1994. Rapid protein extraction from Saccharomyces cerevisiae. Yeast Chichester Engl.. 10:1305–1310. doi:10.1002/yea.320101007.

Hoyt, M.A., L. He, K.K. Loo, and W.S. Saunders. 1992. Two Saccharomyces cerevisiae kinesin- related gene products required for mitotic spindle assembly. J. Cell Biol.. 118:109–120.

Hudson, J.R., E.P. Dawson, K.L. Rushing, C.H. Jackson, D. Lockshon, D. Conover, C. Lanciault, J.R. Harris, S.J. Simmons, R. Rothstein, and S. Fields. 1997. The complete set of predicted genes from Saccharomyces cerevisiae in a readily usable form. Genome Res.. 7:1169–1173.

Ivanovska, I., and M.D. Rose. 2001. Fine structure analysis of the yeast centrin, Cdc31p, identifies residues specific for cell morphology and spindle pole body duplication. Genetics. 157:503–518.

James, P., J. Halladay, and E.A. Craig. 1996. Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast. Genetics. 144:1425–1436.

Jaspersen, S.L., T.H. Giddings, and M. Winey. 2002. Mps3p is a novel component of the yeast spindle pole body that interacts with the yeast centrin homologue Cdc31p. J. Cell Biol.. 159:945–956. doi:10.1083/jcb.200208169.

Jaspersen, S.L., B.J. Huneycutt, T.H. Giddings, K.A. Resing, N.G. Ahn, and M. Winey. 2004. Cdc28/Cdk1 regulates spindle pole body duplication through phosphorylation of Spc42 and Mps1. Dev. Cell. 7:263–274. doi:10.1016/j.devcel.2004.07.006.

Jaspersen, S.L., and M. Winey. 2004. The budding yeast spindle pole body: structure, duplication, and function. Annu. Rev. Cell Dev. Biol.. 20:1–28. doi:10.1146/annurev.cellbio.20.022003.114106.

Jones, G.M., J. Stalker, S. Humphray, A. West, T. Cox, J. Rogers, I. Dunham, and G. Prelich. 2008. A systematic library for comprehensive overexpression screens in Saccharomyces cerevisiae. Nat. Methods. 5:239–241. doi:10.1038/nmeth.1181.

Jones, M.H., and M. Winey. 2006. Centrosome duplication: is asymmetry the clue? Curr. Biol. CB. 16:R808–810. doi:10.1016/j.cub.2006.08.041.

Kearsey, S.E., and S. Cotterill. 2003. Enigmatic variations: divergent modes of regulating eukaryotic DNA replication. Mol. Cell. 12:1067–1075. 181

Keck, J.M., M.H. Jones, C.C.L. Wong, J. Binkley, D. Chen, S.L. Jaspersen, E.P. Holinger, T. Xu, M. Niepel, M.P. Rout, J. Vogel, A. Sidow, J.R. Yates 3rd, and M. Winey. 2011. A cell cycle phosphoproteome of the yeast centrosome. Science. 332:1557–1561. doi:10.1126/science.1205193.

Kilmartin, J.V. 2003. Sfi1p has conserved centrin-binding sites and an essential function in budding yeast spindle pole body duplication. J. Cell Biol.. 162:1211–1221. doi:10.1083/jcb.200307064.

Kitagawa, D., C. Busso, I. Flückiger, and P. Gönczy. 2009. Phosphorylation of SAS-6 by ZYG-1 Is Critical for Centriole Formation in C. elegans Embryos. Dev. Cell. 17:900–907. doi:10.1016/j.devcel.2009.11.002.

Kleylein-Sohn, J., J. Westendorf, M. Le Clech, R. Habedanck, Y.-D. Stierhof, and E.A. Nigg. 2007. Plk4-Induced Centriole Biogenesis in Human Cells. Dev. Cell. 13:190–202. doi:10.1016/j.devcel.2007.07.002.

Knop, M., F. Barr, C.G. Riedel, T. Heckel, and C. Reichel. 2002. Improved version of the red fluorescent protein (drFP583/DsRed/RFP). BioTechniques. 33:592, 594, 596–598 passim.

Kushnirov, V.V. 2000. Rapid and reliable protein extraction from yeast. Yeast Chichester Engl.. 16:857–860. doi:10.1002/1097-0061(20000630)16:9<857::AID-YEA561>3.0.CO;2-B.

Lacey, K.R., P.K. Jackson, and T. Stearns. 1999. Cyclin-dependent kinase control of centrosome duplication. Proc. Natl. Acad. Sci. U. S. A.. 96:2817–2822.

Lee, I.-J., N. Wang, W. Hu, K. Schott, J. Bähler, T.H. Giddings, J.R. Pringle, L.-L. Du, and J.-Q. Wu. 2014. Regulation of spindle-pole body assembly and cytokinesis by the centrin- binding protein Sfi1 in fission yeast. Mol. Biol. Cell. mbc.E13–11–0699. doi:10.1091/mbc.E13-11-0699.

Leidel, S., M. Delattre, L. Cerutti, K. Baumer, and P. Gönczy. 2005. SAS-6 defines a protein family required for centrosome duplication in C. elegans and in human cells. Nat. Cell Biol.. 7:115–125. doi:10.1038/ncb1220.

Lim, H.H., P.Y. Goh, and U. Surana. 1996. Spindle pole body separation in Saccharomyces cerevisiae requires dephosphorylation of the tyrosine 19 residue of Cdc28. Mol. Cell. Biol.. 16:6385–6397.

Lim, H.H., T. Zhang, and U. Surana. 2009. Regulation of centrosome separation in yeast and vertebrates: common threads. Trends Cell Biol.. 19:325–333. doi:10.1016/j.tcb.2009.03.008.

Li, R., and A.W. Murray. 1991. Feedback control of mitosis in budding yeast. Cell. 66:519–531.

Li, S., A.M. Sandercock, P. Conduit, C.V. Robinson, R.L. Williams, and J.V. Kilmartin. 2006. Structural role of Sfi1p-centrin filaments in budding yeast spindle pole body duplication. J. Cell Biol.. 173:867–877. doi:10.1083/jcb.200603153. 182

Liku, M.E., V.Q. Nguyen, A.W. Rosales, K. Irie, and J.J. Li. 2005. CDK phosphorylation of a novel NLS-NES module distributed between two subunits of the Mcm2-7 complex prevents chromosomal rereplication. Mol. Biol. Cell. 16:5026–5039. doi:10.1091/mbc.E05-05-0412.

Longtine, M.S., A. McKenzie 3rd, D.J. Demarini, N.G. Shah, A. Wach, A. Brachat, P. Philippsen, and J.R. Pringle. 1998. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast. 14:953– 961.

Lu, Y., and F. Cross. 2009. Mitotic exit in the absence of separase activity. Mol. Biol. Cell. 20:1576–1591. doi:10.1091/mbc.E08-10-1042.

Ma, P., J. Winderickx, D. Nauwelaers, F. Dumortier, A. De Doncker, J.M. Thevelein, and P. Van Dijck. 1999. Deletion of SFI1, a novel suppressor of partial Ras-cAMP pathway deficiency in the yeast Saccharomyces cerevisiae, causes G(2) arrest. Yeast.. 15:1097– 1109.

Maekawa, H., and E. Schiebel. 2004. Cdk1-Clb4 controls the interaction of astral microtubule plus ends with subdomains of the daughter cell cortex. Genes Dev.. 18:1709–1724. doi:10.1101/gad.298704.

Magtanong, L., C.H. Ho, S.L. Barker, W. Jiao, A. Baryshnikova, S. Bahr, A.M. Smith, L.E. Heisler, J.S. Choy, E. Kuzmin, K. Andrusiak, A. Kobylianski, Z. Li, M. Costanzo, M.A. Basrai, G. Giaever, C. Nislow, B. Andrews, and C. Boone. 2011. Dosage suppression genetic interaction networks enhance functional wiring diagrams of the cell. Nat. Biotechnol.. 29:505–511. doi:10.1038/nbt.1855.

Makise, M., M. Takehara, A. Kuniyasu, N. Matsui, H. Nakayama, and T. Mizushima. 2009. Linkage between phosphorylation of the origin recognition complex and its ATP binding activity in Saccharomyces cerevisiae. J. Biol. Chem.. 284:3396–3407. doi:10.1074/jbc.M804293200.

Matsumoto, Y., K. Hayashi, and E. Nishida. 1999. Cyclin-dependent kinase 2 (Cdk2) is required for centrosome duplication in mammalian cells. Curr. Biol. CB. 9:429–432.

McGrew, J.T., L. Goetsch, B. Byers, and P. Baum. 1992. Requirement for ESP1 in the nuclear division of Saccharomyces cerevisiae. Mol. Biol. Cell. 3:1443–1454.

Meraldi, P., J. Lukas, A.M. Fry, J. Bartek, and E.A. Nigg. 1999. Centrosome duplication in mammalian somatic cells requires E2F and Cdk2-cyclin A. Nat. Cell Biol.. 1:88–93. doi:10.1038/10054.

Mimura, S., T. Seki, S. Tanaka, and J.F.X. Diffley. 2004. Phosphorylation-dependent binding of mitotic cyclins to Cdc6 contributes to DNA replication control. Nature. 431:1118–1123. doi:10.1038/nature03024.

183

Morgan, D.O. 1997. CYCLIN-DEPENDENT KINASES: Engines, Clocks, and Microprocessors. Annu. Rev. Cell Dev. Biol.. 13:261–291. doi:10.1146/annurev.cellbio.13.1.261.

Morgan, D.O. 2006. The cell cycle: Principles of control. Sinauer Associates, Inc.

Moritz, M., M.B. Braunfeld, V. Guénebaut, J. Heuser, and D.A. Agard. 2000. Structure of the gamma-tubulin ring complex: a template for microtubule nucleation. Nat. Cell Biol.. 2:365–370. doi:10.1038/35014058.

Musacchio, A., and K.G. Hardwick. 2002. The spindle checkpoint: structural insights into dynamic signalling. Nat. Rev. Mol. Cell Biol.. 3:731–741. doi:10.1038/nrm929.

Nelson, B., C. Kurischko, J. Horecka, M. Mody, P. Nair, L. Pratt, A. Zougman, L.D.B. McBroom, T.R. Hughes, C. Boone, and F.C. Luca. 2003. RAM: a conserved signaling network that regulates Ace2p transcriptional activity and polarized morphogenesis. Mol. Biol. Cell. 14:3782–3803. doi:10.1091/mbc.E03-01-0018.

Nguyen, V.Q., C. Co, and J.J. Li. 2001. Cyclin-dependent kinases prevent DNA re-replication through multiple mechanisms. Nature. 411:1068–1073. doi:10.1038/35082600.

Nguyen, V.Q., C. Co, K. Irie, and J.J. Li. 2000. Clb/Cdc28 kinases promote nuclear export of the replication initiator proteins Mcm2-7. Curr. Biol. CB. 10:195–205.

Nigg, E.A. 2007. Centrosome duplication: of rules and licenses. Trends Cell Biol.. 17:215–221. doi:10.1016/j.tcb.2007.03.003.

Nigg, E.A., and T. Stearns. 2011. The centrosome cycle: Centriole biogenesis, duplication and inherent asymmetries. Nat. Cell Biol.. 13:1154–1160. doi:10.1038/ncb2345.

Novak, Z.A., P.T. Conduit, A. Wainman, and J.W. Raff. 2014. Asterless licenses daughter centrioles to duplicate for the first time in Drosophila embryos. Curr. Biol. CB. 24:1276– 1282. doi:10.1016/j.cub.2014.04.023.

O’Connell, K.F., C. Caron, K.R. Kopish, D.D. Hurd, K.J. Kemphues, Y. Li, and J.G. White. 2001. The C. elegans zyg-1 gene encodes a regulator of centrosome duplication with distinct maternal and paternal roles in the embryo. Cell. 105:547–558.

Okuda, M., H.F. Horn, P. Tarapore, Y. Tokuyama, A.G. Smulian, P.K. Chan, E.S. Knudsen, I.A. Hofmann, J.D. Snyder, K.E. Bove, and K. Fukasawa. 2000. Nucleophosmin/B23 is a target of CDK2/cyclin E in centrosome duplication. Cell. 103:127–140.

O’Toole, E.T., D.N. Mastronarde, T.H. Giddings, M. Winey, D.J. Burke, and J.R. McIntosh. 1997. Three-dimensional analysis and ultrastructural design of mitotic spindles from the cdc20 mutant of Saccharomyces cerevisiae. Mol. Biol. Cell. 8:1–11.

Ouspenski, I.I., O.A. Cabello, and B.R. Brinkley. 2000. Chromosome Condensation Factor Brn1p Is Required for Chromatid Separation in Mitosis. Mol. Biol. Cell. 11:1305–1313. doi:10.1091/mbc.11.4.1305. 184

Page, B.D., L.L. Satterwhite, M.D. Rose, and M. Snyder. 1994. Localization of the Kar3 kinesin heavy chain-related protein requires the Cik1 interacting protein. J. Cell Biol.. 124:507– 519.

Paintrand, M., M. Moudjou, H. Delacroix, and M. Bornens. 1992. Centrosome organization and centriole architecture: Their sensitivity to divalent cations. J. Struct. Biol.. 108:107–128. doi:10.1016/1047-8477(92)90011-X.

Pereira, G., U. Grueneberg, M. Knop, and E. Schiebel. 1999. Interaction of the yeast gamma- tubulin complex-binding protein Spc72p with Kar1p is essential for microtubule function during karyogamy. EMBO J.. 18:4180–4195. doi:10.1093/emboj/18.15.4180.

Pereira, G., T.U. Tanaka, K. Nasmyth, and E. Schiebel. 2001. Modes of spindle pole body inheritance and segregation of the Bfa1p-Bub2p checkpoint protein complex. EMBO J.. 20:6359–6370. doi:10.1093/emboj/20.22.6359.

Piatti, S., C. Lengauer, and K. Nasmyth. 1995. Cdc6 is an unstable protein whose de novo synthesis in G1 is important for the onset of S phase and for preventing a “reductional” anaphase in the budding yeast Saccharomyces cerevisiae. EMBO J.. 14:3788–3799.

Radivojac, P., V. Vacic, C. Haynes, R.R. Cocklin, A. Mohan, J.W. Heyen, M.G. Goebl, and L.M. Iakoucheva. 2010. Identification, analysis, and prediction of protein ubiquitination sites. Proteins. 78:365–380. doi:10.1002/prot.22555.

Roof, D.M., P.B. Meluh, and M.D. Rose. 1992. Kinesin-related proteins required for assembly of the mitotic spindle. J. Cell Biol.. 118:95–108.

Rose, M.D., and G.R. Fink. 1987. KAR1, a gene required for function of both intranuclear and extranuclear microtubules in yeast. Cell. 48:1047–1060.

Rout, M.P., and J.V. Kilmartin. 1990. Components of the yeast spindle and spindle pole body. J. Cell Biol.. 111:1913–1927.

Schaerer, F., G. Morgan, M. Winey, and P. Philippsen. 2001. Cnm67p Is a Spacer Protein of the Saccharomyces cerevisiae Spindle Pole Body Outer Plaque. Mol. Biol. Cell. 12:2519– 2533. doi:10.1091/mbc.12.8.2519.

Schild, D., H.N. Ananthaswamy, and R.K. Mortimer. 1981. An endomitotic effect of a cell cycle mutation of Saccharomyces cerevisiae. Genetics. 97:551–562.

Schramm, C., S. Elliott, A. Shevchenko, and E. Schiebel. 2000. The Bbp1p-Mps2p complex connects the SPB to the nuclear envelope and is essential for SPB duplication. EMBO J.. 19:421–433. doi:10.1093/emboj/19.3.421.

Schwab, M., A.S. Lutum, and W. Seufert. 1997. Yeast Hct1 is a regulator of Clb2 cyclin proteolysis. Cell. 90:683–693. 185

Sclafani, R.A., and T.M. Holzen. 2007. Cell cycle regulation of DNA replication. Annu. Rev. Genet.. 41:237–280. doi:10.1146/annurev.genet.41.110306.130308.

Sheff, M.A., and K.S. Thorn. 2004. Optimized cassettes for fluorescent protein tagging in Saccharomyces cerevisiae. Yeast Chichester Engl.. 21:661–670. doi:10.1002/yea.1130.

Simmons Kovacs, L.A., C.L. Nelson, and S.B. Haase. 2008. Intrinsic and cyclin-dependent kinase-dependent control of spindle pole body duplication in budding yeast. Mol. Biol. Cell. 19:3243–3253. doi:10.1091/mbc.E08-02-0148.

Spang, A., I. Courtney, K. Grein, M. Matzner, and E. Schiebel. 1995. The Cdc31p-binding protein Kar1p is a component of the half bridge of the yeast spindle pole body. J. Cell Biol.. 128:863–877. doi:10.1083/jcb.128.5.863.

Stegmeier, F., and A. Amon. 2004. Closing mitosis: the functions of the Cdc14 phosphatase and its regulation. Annu. Rev. Genet.. 38:203–232. doi:10.1146/annurev.genet.38.072902.093051.

Strawn, L.A., and H.L. True. 2006. Deletion of RNQ1 gene reveals novel functional relationship between divergently transcribed Bik1p/CLIP-170 and Sfi1p in spindle pole body separation. Curr. Genet.. 50:347–366. doi:10.1007/s00294-006-0098-6.

Strickfaden, S.C., M.J. Winters, G. Ben-Ari, R.E. Lamson, M. Tyers, and P.M. Pryciak. 2007. A Mechanism for Cell-Cycle Regulation of MAP Kinase Signaling in a Yeast Differentiation Pathway. Cell. 128:519–531. doi:10.1016/j.cell.2006.12.032.

Strnad, P., S. Leidel, T. Vinogradova, U. Euteneuer, A. Khodjakov, and P. Gönczy. 2007. Regulated HsSAS-6 levels ensure formation of a single procentriole per centriole during the centrosome duplication cycle. Dev. Cell. 13:203–213. doi:10.1016/j.devcel.2007.07.004.

Sullivan, D.S., S. Biggins, and M.D. Rose. 1998. The yeast centrin, cdc31p, and the interacting protein kinase, Kic1p, are required for cell integrity. J. Cell Biol.. 143:751–765.

Sullivan, M., and F. Uhlmann. 2003. A non-proteolytic function of separase links the onset of anaphase to mitotic exit. Nat. Cell Biol.. 5:249–254. doi:10.1038/ncb940.

Tinker-Kulberg, R.L., and D.O. Morgan. 1999. Pds1 and Esp1 control both anaphase and mitotic exit in normal cells and after DNA damage. Genes Dev.. 13:1936–1949.

Tong, A.H.Y., and C. Boone. 2006. Synthetic genetic array analysis in Saccharomyces cerevisiae. Methods Mol. Biol. Clifton NJ. 313:171–192.

Tsou, M.-F.B., and T. Stearns. 2006. Controlling centrosome number: licenses and blocks. Curr. Opin. Cell Biol.. 18:74–78. doi:10.1016/j.ceb.2005.12.008. 186

Tsou, M.-F.B., W.-J. Wang, K.A. George, K. Uryu, T. Stearns, and P.V. Jallepalli. 2009. Polo kinase and separase regulate the mitotic licensing of centriole duplication in human cells. Dev. Cell. 17:344–354. doi:10.1016/j.devcel.2009.07.015.

Ubersax, J.A., E.L. Woodbury, P.N. Quang, M. Paraz, J.D. Blethrow, K. Shah, K.M. Shokat, and D.O. Morgan. 2003. Targets of the cyclin-dependent kinase Cdk1. Nature. 425:859–864. doi:10.1038/nature02062.

Vallen, E.A., M.A. Hiller, T.Y. Scherson, and M.D. Rose. 1992. Separate domains of KAR1 mediate distinct functions in mitosis and nuclear fusion. J. Cell Biol.. 117:1277–1287.

Vallen, E.A., W. Ho, M. Winey, and M.D. Rose. 1994. Genetic interactions between CDC31 and KAR1, two genes required for duplication of the microtubule organizing center in Saccharomyces cerevisiae. Genetics. 137:407–422.

Vidwans, S.J., M.L. Wong, and P.H. O’Farrell. 2003. Anomalous centriole configurations are detected in Drosophila wing disc cells upon Cdk1 inactivation. J. Cell Sci.. 116:137–143.

Visintin, R., K. Craig, E.S. Hwang, S. Prinz, M. Tyers, and A. Amon. 1998. The phosphatase Cdc14 triggers mitotic exit by reversal of Cdk-dependent phosphorylation. Mol. Cell. 2:709–718.

Visintin, R., S. Prinz, and A. Amon. 1997. CDC20 and CDH1: A Family of Substrate-Specific Activators of APC-Dependent Proteolysis. Science. 278:460–463. doi:10.1126/science.278.5337.460.

Wang, W., A. Budhu, M. Forgues, and X.W. Wang. 2005. Temporal and spatial control of nucleophosmin by the Ran-Crm1 complex in centrosome duplication. Nat. Cell Biol.. 7:823–830. doi:10.1038/ncb1282.

Wäsch, R., and F.R. Cross. 2002. APC-dependent proteolysis of the mitotic cyclin Clb2 is essential for mitotic exit. Nature. 418:556–562. doi:10.1038/nature00856.

Weiss, E., and M. Winey. 1996. The Saccharomyces cerevisiae spindle pole body duplication gene MPS1 is part of a mitotic checkpoint. J. Cell Biol.. 132:111–123.

Wente, S.R., and G. Blobel. 1994. NUP145 encodes a novel yeast glycine-leucine- phenylalanine-glycine (GLFG) nucleoporin required for nuclear envelope structure. J. Cell Biol.. 125:955–969.

Winey, M., L. Goetsch, P. Baum, and B. Byers. 1991. MPS1 and MPS2: novel yeast genes defining distinct steps of spindle pole body duplication. J. Cell Biol.. 114:745–754.

Winey, M., M.A. Hoyt, C. Chan, L. Goetsch, D. Botstein, and B. Byers. 1993. NDC1: a nuclear periphery component required for yeast spindle pole body duplication. J. Cell Biol.. 122:743–751. 187

Wong, C., and T. Stearns. 2003. Centrosome number is controlled by a centrosome-intrinsic block to reduplication. Nat. Cell Biol.. 5:539–544. doi:10.1038/ncb993.

Wong, J., Y. Nakajima, S. Westermann, C. Shang, J. Kang, C. Goodner, P. Houshmand, S. Fields, C.S.M. Chan, D. Drubin, G. Barnes, and T. Hazbun. 2007. A Protein Interaction Map of the Mitotic Spindle. Mol. Biol. Cell. 18:3800–3809. doi:10.1091/mbc.E07-06- 0536.

Xiao, Z., J.T. McGrew, A.J. Schroeder, and M. Fitzgerald-Hayes. 1993. CSE1 and CSE2, two new genes required for accurate mitotic chromosome segregation in Saccharomyces cerevisiae. Mol. Cell. Biol.. 13:4691–4702.

Yoder, T.J., C.G. Pearson, K. Bloom, and T.N. Davis. 2003. The Saccharomyces cerevisiae spindle pole body is a dynamic structure. Mol. Biol. Cell. 14:3494–3505. doi:10.1091/mbc.E02-10-0655.