Defining Roles for Cyclin-Dependent Kinases and a Transcriptional
Oscillator in the Organization of Cell Cycle Events
by
Laura Anne Simmons Kovacs
University Program in Genetics and Genomics
Duke University
Date:_______________________
Approved:
___________________________
Steven B. Haase, Ph.D., Advisor
___________________________
Sally Kornbluth, Ph.D.
___________________________
Daniel J. Lew, Ph.D.
___________________________
Dennis J. Thiele, Ph.D.
___________________________
Gregory A. Wray, Ph.D.
Dissertation submitted in partial fulfillment of
the requirements for the degree of Doctor of Philosophy in the
University Program in Genetics and Genomics
in the Graduate School of Duke University
2009
Abstract
Defining Roles for Cyclin-Dependent Kinases and a
Transcriptional Oscillator in the Organization of Cell Cycle
Events
by
Laura Anne Simmons Kovacs
University Program in Genetics and Genomics
Duke University
Date:_______________________
Approved:
___________________________
Steven B. Haase, Ph.D., Advisor
___________________________
Sally Kornbluth, Ph.D.
___________________________
Daniel J. Lew, Ph.D.
___________________________
Dennis J. Thiele, Ph.D.
___________________________
Gregory A. Wray, Ph.D.
An abstract of a dissertation submitted in partial fulfillment of
the requirements for the degree of Doctor of Philosophy in the
University Program in Genetics and Genomics
in the Graduate School of Duke University
2009
Copyright by
Laura Anne Simmons Kovacs
2009
Abstract
The cell cycle is a series of ordered events that culminates in a single cell dividing
into two daughter cells. These events must be properly coordinated to ensure the
faithful passage of genetic material. How cell-cycle events are carried out accurately
remains a fundamental question in cell biology. In this dissertation, I investigate
mechanisms orchestrating cell-cycle events in the yeast, Saccharomyces cerevisiae.
Cyclin dependent kinase (CDK) activity is thought to both form the fundamental
cell-cycle oscillator and act as an effector of that oscillator, regulating cell-cycle events.
By measuring transcript dynamics over time in cells lacking all CDK activity, I show that
transcriptional oscillations are not dependent on CDK activity. This data indicates that
CDKs do not form the underlying cell-cycle oscillator. I propose a model in which a
transcription factor network rather than CDK activity forms the cell-cycle oscillator. In
this model, CDKs are activated by the periodic transcription of cyclin genes and feedback
on the network, increasing the robustness of network oscillations in addition to
regulating cell-cycle events.
I also investigate CDK-dependent and –independent mechanism regulating the
duplication of the yeast centrosome, the spindle pole body (SPB). It is critical for the
formation of a bipolar spindle in mitosis that the SPB duplicates once and only once per
cell cycle. Through a combination of genetic and microscopic techniques I show that
iv
three distinct mechanisms regulate SPB duplication, ensuring its restriction to once per
cell cycle.
Together, the data presented in this dissertation support a model in which CDKs,
periodic transcription, and a TF-network oscillator are all important cell-cycle regulatory
mechanisms that collaborate to regulate the intricate collection of events that
constitute the cell cycle.
v
Contents
Abstract ...............................................................................................................................iv
List of Tables ...................................................................................................................... xii
List of Figures .................................................................................................................... xiii
Acknowledgements............................................................................................................ xv
1. Introduction: Cyclin-dependent kinases and the regulation of the cell cycle ................ 1
1.1 Events of the Cell Cycle ........................................................................................... 1
1.2 What are the factors that regulate cell-cycle events?............................................ 3
1.3 What are CDKs and what do they do? .................................................................... 4
1.3.1 How is the cell cycle regulated in somatic cells? ........................................... 6
1.4 Saccharomyces cerevisiae as a model system for the study of cell-cycle regulation
..................................................................................................................................... 10
1.4.1 Cyclin homologues in S. cerevisiae .............................................................. 11
1.4.2 Regulation of G1 events by G1-cyclin-CDK activity...................................... 13
1.4.3 S-phase cyclin-CDK activity .......................................................................... 15
1.4.4 Roles for mitotic B-cyclins ............................................................................ 16
1.4.5 Mitosis and mitotic exit in yeast .................................................................. 18
1.5 Checkpoints: mechanisms that ensure the order of cell cycle events ................. 19
1.6 Common threads in cell cycle regulation ............................................................. 22
1.7 What is the role of B-cyclins in the regulation of the yeast cell cycle? ................ 23
1.7.1 Independent oscillations of G1 events in the absence of B-cyclins ............ 24
1.7.2 Periodic transcription in the absence of cyclin-CDK activity ....................... 26
vi
1.7.3 SPB reduplication in the absence of mitotic B-cyclins ................................. 28
1.7.4 Questions raised by studies of yeast cells lacking B-cyclin activity ............. 30
1.8 What is the function of CDKs in cell cycle oscillations? ........................................ 30
1.8.1 Biological oscillators..................................................................................... 30
1.8.2 Cell cycle oscillators ..................................................................................... 32
1.8.3 CDK-independent cell cycle oscillations?..................................................... 34
1.9 How do B-cyclins ensure that SPB duplication occurs once and only once per cell
cycle?........................................................................................................................... 35
1.9.1 Restriction of DNA replication to once per cell cycle .................................. 35
1.9.2 Regulation of centrosome duplication ........................................................ 36
1.9.3 SPB duplication and SPB licensing ............................................................... 39
1.10 Concluding remarks ............................................................................................ 43
1.10.1 Dissertation outline ................................................................................... 43
2. The role of CDKs in cell-cycle oscillations ..................................................................... 44
2.1 Introduction .......................................................................................................... 44
2.2 Oscillations in G1-cyclin are not necessary for B-cyclin independent oscillations.
..................................................................................................................................... 47
2.3 Transcriptional dynamics in α-factor treated cells. .............................................. 49
2.3.1 α-factor specific transcriptional regulation ................................................. 52
2.3.2 Periodic transcription in α-factor treated cells............................................ 54
2.4 Periodic transcription in cells lacking functional CDK........................................... 58
2.5 TF-networks may maintain oscillations in the absence of CDK activity. .............. 61
2.6 What is the role of CDKs in transcriptional oscillations? ...................................... 66
vii
2.7 Discussion.............................................................................................................. 71
2.7.1 A new model for cell-cycle oscillations ........................................................ 71
2.7.2 Are cell-cycle events entrained to CDK-independent oscillations? ............. 73
2.8 Future Directions .................................................................................................. 76
2.8.1 Damped oscillations in the absence of CDK activity? .................................. 76
2.8.2 Roles of B-cyclins in enforcing robustness to CDK-independent oscillations
............................................................................................................................... 77
2.8.3 Transcriptional response to α-factor treatment.......................................... 78
2.9 Experimental Details ............................................................................................. 79
2.9.1 Plasmid and strain construction .................................................................. 79
2.9.2 Cell growth and synchronization ................................................................. 79
2.9.3 Microscopy ................................................................................................... 80
2.9.4 RNA isolation and microarray analysis ........................................................ 80
2.9.5 CLOCCS model fitting ................................................................................... 81
2.9.6 Parameter estimation by expectation maximization .................................. 82
2.9.7 Correlating expression data ......................................................................... 84
2.10 Chapter 2 collaborator contributions ................................................................. 86
3. Continuing studies into the nature of the CDK-independent transcriptional
oscillator….. ....................................................................................................................... 87
3.1 Is a TF-network oscillator responsible for CDK-independent transcriptional
oscillations?................................................................................................................. 87
3.1.1 Are mRNA levels a good proxy for TF function? .......................................... 88
3.1.2 Are CDK-independent oscillations regulated by a TF network? .................. 91
3.1.3 Which transcriptional regulators compose the TF network oscillator? ...... 94
viii
3.2 How does the TF-network oscillator interact with checkpoints? ......................... 95
3.2.1 What is the role of B-cyclin activity in arresting transcriptional dynamics
during checkpoint arrest? ..................................................................................... 97
3.2.2 Are there B-cyclin independent mechanisms whereby checkpoints affect
global transcriptional dynamics? ........................................................................ 101
3.2.3 A model for checkpoint regulation of periodic gene expression .............. 102
3.3 Experimental details ........................................................................................... 104
3.3.1 Plasmid and strain construction ................................................................ 104
3.3.2 Cell growth and synchronization ............................................................... 104
3.3.3 CLOCCS model fitting ................................................................................. 105
3.3.4 Protein isolation and immunoblotting....................................................... 106
3.3.5 RNA isolation and microarray analysis. ..................................................... 106
3.4 Chapter 3 collaborator contributions ................................................................. 107
4. Intrinsic and CDK-dependent control of spindle pole body duplication.................... 108
4.1 Introduction ........................................................................................................ 108
4.2 SPB separation is required for SPB reduplication. .............................................. 113
4.3 A mitotic cyclin expressed in G1 can inhibit SPB duplication. ............................ 118
4.3.1 Clb2 expression in early G1 inhibits satellite assembly. ............................ 120
4.4 Role of B-cyclin localization in the regulation of SPB duplication ...................... 125
4.4.1 Cytoplasmic localization cannot confer the ability to inhibit SPB duplication
to Clb5 ................................................................................................................. 127
4.5 Discussion............................................................................................................ 129
4.5.1 The SPB-intrinsic mechanism preventing SPB reduplication ..................... 129
4.5.2 Inhibition of SPB licensing.......................................................................... 130
ix
4.5.3 Multiple half-bridges.................................................................................. 132
4.5.4 Cyclin specificity in SPB licensing ............................................................... 133
4.5.5 Model for restriction of SPB duplication to once per cell cycle ................ 133
4.5.6 Implications for the centrosome duplication cycle ................................... 134
4.6 Future Directions ................................................................................................ 135
4.6.1 What CDK targets are important for SPB separation? .............................. 135
4.6.2 What are the CDK targets involved in SPB licensing? ................................ 136
4.7 Experimental details ........................................................................................... 139
4.7.1 Plasmid and strain construction. ............................................................... 139
4.7.2 Cell growth and synchronization ............................................................... 143
4.7.3 Microscopy ................................................................................................. 144
4.7.4 Immunoblotting ......................................................................................... 144
4.7.5 Flow Cytometry .......................................................................................... 145
4.8 Chapter 4 collaborator contributions ................................................................. 145
5. Periodic transcription regulates spindle pole body duplication ................................. 146
5.1 Introduction ........................................................................................................ 146
5.1.1 Expression of SPB components during the cell cycle. ............................... 147
5.2 SPB duplication requires new protein synthesis ................................................ 149
5.3 The role of MBF in SPB duplication..................................................................... 152
5.4 G1 transcription and inhibition of SPB reduplication ......................................... 153
5.5 Discussion............................................................................................................ 158
5.5.1 Just-in-time transcription of SPB component genes ................................. 159
x
5.5.2 Cyclin specificity in G1 events .................................................................... 159
5.5.3 Three mechanisms for inhibition of SPB duplication................................. 162
5.5.4 Periodic transcription regulates cell-cycle events ..................................... 163
5.6 Future Directions ................................................................................................ 163
5.6.1 Does SPB component genes expression correlate with SPB duplication? 164
5.6.2 How does Clb2Δdb drive SPB duplication in α-factor arrested cells? ....... 164
5.7 Experimental Details ........................................................................................... 165
5.7.1 Plasmid and strain construction ................................................................ 165
5.7.2 Cell growth and synchronization ............................................................... 167
5.7.3 Microscopy ................................................................................................. 168
5.7.4 Immunoblotting ......................................................................................... 168
6. Discussion: new perspectives on the role of CDKs in the regulation of the cellcycle…….. ......................................................................................................................... 169
6.1 CDK-independent mechanisms regulating the cell-cycle. .................................. 170
6.2 CDK-dependent mechanisms regulating the cell-cycle. ..................................... 174
6.3 The evolution of cell-cycle control...................................................................... 176
6.3.1 Conservation and variation among eukaryotic cell-cycle networks .......... 178
6.4 The role of CDKs in the metazoan cell-cycle ....................................................... 180
6.5 Concluding Remarks............................................................................................ 183
References ...................................................................................................................... 184
Biography ........................................................................................................................ 214
xi
List of Tables
Table 1.1: Cyclin genetics in S. cerevisiae ......................................................................... 12
Table 1.2: Genes important for SPB structure and duplication........................................ 42
Table 2.1: Yeast strains used in Chapter 2 ........................................................................ 80
Table 3.1: Transcription factor stability during the cell cycle ........................................... 90
Table 3.2: Yeast strains used in Chapter 3 ...................................................................... 105
Table 4.1: Yeast strains used in Chapter 4. ..................................................................... 140
Table 5.1: Yeast strains used in Chapter 5. ..................................................................... 166
xii
List of Figures
Figure 1.1: Cyclin dynamics in three different cell cycles depicted as cyclin-CDK activity
versus time.. ........................................................................................................................ 7
Figure 1.2: Regulation of yeast cell-cycle events by cyclin-CDK activity.. ........................ 11
Figure 1.3: TF network oscillator proposed to maintain oscillations in the absence of Bcyclins. ............................................................................................................................... 28
Figure 1.4: Centrosome duplication cycles. ...................................................................... 38
Figure 2.1: Oscillations in Cln2 are not essential for B-cyclin-independent oscillations .. 48
Figure 2.2: Experimental controls for α-factor treatment experiments. ......................... 50
Figure 2.3: α-factor regulated expression. ....................................................................... 53
Figure 2.4: Periodic expression persists in α-factor treated cells..................................... 55
Figure 2.5: Oscillations in α-factor are not dependent on oscillations of the negative
regulators SST2 and MSG5.. .............................................................................................. 56
Figure 2.6: Experimental controls for cdc28-4 experiments. ........................................... 59
Figure 2.7: Periodic expression persists in cdc28-4 cells. ................................................. 60
Figure 2.8: Transcription factor networks may produce CDK-independent transcriptional
oscillations. ....................................................................................................................... 63
Figure 2.9: CDK activity adds robustness to cell-cycle oscillations. .................................. 67
Figure 2.10: Model for cell-cycle regulation. .................................................................... 70
Figure 2.11: Expectation maximization analysis ............................................................... 85
Figure 3.1: TF protein dynamics........................................................................................ 89
Figure 3.2: Protein synthetic rates affect B-cyclin-independent oscillations. .................. 92
Figure 3.3: Periodic transcript dynamics during the DNA-replication checkpoint. .......... 96
Figure 3.4: Transcript during the replication checkpoint. ................................................ 98
xiii
Figure 3.5: Model of the relationship between cell-cycle-control mechanisms. ........... 103
Figure 4.1: The SPB duplication cycle is coordinated with the cell cycle. ...................... 111
Figure 4.2: SPB separation is required for SPB duplication. ........................................... 115
Figure 4.3: Ectopic expression of Clb2 in early G1 inhibits SPB duplication................... 119
Figure 4.4: Ectopic expression of G1 or S-phase cyclins in early G1 cannot inhibit SPB
duplication.. .................................................................................................................... 121
Figure 4.5: Clb2 inhibits early steps in SPB duplication.. ................................................ 122
Figure 4.6: Clb2 influences the structure of the SPB. ..................................................... 124
Figure 4.7: Clb2 must be cytoplasmic in order to inhibit SPB duplication.. ................... 126
Figure 4.8: Cytoplasmic Clb5Δdb cannot inhibit SPB duplication.. ................................ 128
Figure 5.1: Dynamics of SPB component genes in wild-type cells. ................................ 148
Figure 5.2 New protein synthesis is important for SPB duplication.. ............................. 151
Figure 5.3: Clb2Δdb driven SPB duplication in α-factor is more efficient in Δmbp1 cells....
......................................................................................................................................... 153
Figure 5.4: SPB reduplication is concurrent with re-budding and transcription of SPB
component genes.. ......................................................................................................... 154
Figure 5.5: SPB reduplication in Δclb1-3; Δmbp1 cells.. ................................................ 156
Figure 6.1: Model for cell-cycle regulation. .................................................................... 171
Figure 6.2: Regulatory network regulating the yeast cell-cycle. .................................... 173
Figure 6.3: The cell-cycle in Caulobacter is regulated by a TF-network. ........................ 177
Figure 6.4: Generalized structure of a eukaryotic cell-cycle regulatory network. ......... 179
xiv
Acknowledgements
I would first like to express my earnest gratitude to my advisor, mentor, and
friend, Steve Haase. From Steve, I have learned so much, not only about science, but
also about life, and about myself. Steve, thank you for all the advice, support, proverbs,
laughs, lessons, sports analogies, wisdom, encouragement, sarcasm, honesty, stories,
and long discussions. You truly make the lab an enjoyable place to be, to learn, and to
do science. I have been grateful, not just today, but every day of graduate school, that I
am lucky enough to have you as my advisor. I can only hope that someday I will be able
to teach, advise, council, and lead students with as much thought, skill, and enthusiasm
as you have me.
I would like to acknowledge my committee members past and present. Sally,
Dennis, Greg, Ken and Danny, thank you for helping and challenging me throughout this
process. For help with various experiments and analyses, I would like to thank Sara
Miller, Phillip Christopher, Sam Johnson, Yasheng Gao, Holly Dressman, Laura-Leigh
Rowlette, Mark Winey, Sayan Mukherjee, and many others. I would also like to thank
Andrea Lanahan, Carol Richardson, Anne Lacey and all the other administrative staff in
the Biology Department for all that they do.
I would like to thank all members of the Haase lab, past and present. Mark,
thank you for your camaraderie over the years. We are opposites in many ways, but I
truly admire your attention to detail have learned so much from it. Dave, I am so
xv
grateful to have you as a friend, accomplice, and collaborator. Thank you for your
analytical assistance, for helping me learn to think like a systems biologist, and for your
patience in teaching me the skills I need to do my own analysis. Charles, thanks for
taking the blame like a champion. Sara, thanks for being my little sister in the lab. To
everyone else, thanks for the laughs, discussions, and for putting up with me.
While at Duke, I have been fortunate to meet and become close to so many
wonderful people from a variety of schools and departments. For that opportunity, and
for enough memories to last a lifetime, I have to thank the Basketball Committee. To all
my friends, thank you for listening, sharing, commiserating, and celebrating all that
graduate school and the life outside has to offer. I wish you all the utmost happiness
and cannot wait to share it with you in the years to come. To the girls, thanks for lunch.
Last but not least, I have to thank my family. Mom and Dad, thank you for all the
love and encouragement over the years. You have been unwavering in your support
while I follow my dreams, even though you might not understand them, and I truly
appreciate it. Mike, you may not know it, but you are an inspiration. I am so grateful to
have you as a brother. Jeff, you are my rock and my best friend. You entered my life at
the beginning of graduate school and have been there for me every step of the way. I
am so thankful to have you as both a colleague and a confidant.
xvi
Chapter 1
Introduction: Cyclin-dependent kinases and the
regulation of the cell cycle
“…Double or nothing. With few exceptions a living cell either
reproduces or dies; the principle is so simple that no one has bothered
to call it a principle. A cell is born in the division of a parent cell. It then
doubles in every respect: in every part, in every kind of molecule, even
in the amount of water it contains. Thereafter it divides with such equal
justice that each new daughter cell is an identical copy of the parent.
This doubling and halving, the cycle of growth and division, is known
generally as the cell cycle...”
- Daniel Mazia [1]
The cell is the most basic unit of life that includes all the necessary information
and machinery to make an exact copy of itself: to reproduce. In order to reproduce a
cell must undergo a complex, highly regulated series of events that culminates in a
single cell dividing into two identical cells. This process is termed the cell cycle. In this
dissertation, I will ask fundamental questions about cell-cycle regulation and present
evidence that both challenges and advances the current understanding of this important
process.
1.1 Events of the Cell Cycle
In order to complete a cell cycle, cells must undergo two fundamental processes:
duplication and segregation. Before segregation occurs, the cell must duplicate its
1
contents. The cell grows to double its initial size, making enough of its essential
contents to support two cells. Importantly, a cell must make an exact copy of its
genome to pass the information necessary for life onto the next generation. Once the
cell has duplicated its contents, the cell must undergo segregation, dividing itself and its
contents in two. Repeated cycles of duplication and segregation lead to exponential
accumulation of identical cells and are at the heart of organismal growth and
reproduction.
In eukaryotic cells, the fundamental duplication and segregation processes of the
cell cycle are divided into four phases: G1, S-phase, G2, and mitosis. S-phase and
mitosis are defined by the essential processes that take place during these phases. S-,
or synthesis, phase is defined as the discrete period when DNA replication takes place.
Mitosis is the phase of the cell cycle in which cellular contents begins to be segregated
between the mother and daughter cells. During mitosis cells construct an elaborate
structure, the mitotic spindle, which partitions condensed chromosomes into the two
daughter cells. Once the genome is divided, the spindle breaks down and the cell
undergoes cytokinesis, physically dividing the mother cell into two daughter cells. In
addition to S-phase and mitosis, there are two gap phases, G1 and G2. These phases
were initially defined as gaps in time between the essential events of S-phase and
mitosis [2], however, the cell is not idle during these phases. During the gap phases, the
cell grows, interacts with the environment and prepares for the essential processes in
2
the next phase. Perhaps the most important event during the gap phases takes place in
G1, when cells make an irreversible commitment to proceed through the cell cycle and
cell division. This point of commitment is called Start in yeast and the restriction point
in metazoans.
The relationship between the different phases of the cell cycle were initially
investigated by fusing cells in different phases of the cell cycle together and measuring
cell cycle progression of the resulting binucleate cell. While fusion of S-phase cells with
G2 cells or G1 cells with G2 cells did not affect the progression of DNA replication in
either nucleus, G1 cells rapidly entered S-phase when fused to an S-phase cell [3]. In
addition, when G2 cells were fused with G1 or S-phase cells, both nuclei went through
mitosis at the same time [3]. These experiments demonstrate that cell-cycle events are
regulated by soluble factors that both promote and inhibit cell cycle events. The big
question regarding the cell cycle became: what are these factors and how do they
regulate the cell cycle?
1.2 What are the factors that regulate cell-cycle events?
The earliest insight into the activity of factors that regulate the cell cycle came
from studies of oocyte maturation and early embryonic cell divisions in frog and marine
invertebrate embryos. In 1971, it was shown that injecting cytoplasm from previously
activated embryos could activate frog oocyte maturation [4]. These experiments
demonstrated that active oocytes produce a cytoplasmic element named the
3
“maturation promoting factor”, MPF, that can activate meiosis and is present during
mitosis in cleaving embryos [4, 5, reviewed in 6]. In subsequent years, MPF activity was
found in other embryonic systems as well as in mitotic extracts from somatic
mammalian cells and the yeast Saccharomyces cerevisiae [7-10, reviewed in 11]. Many
properties of MPF activity were described prior to the identification of its active
components. It was shown that MPF activity peaked during mitosis in developing
embryos and that its peak is dependent on protein synthesis and is associated with a
peak in protein phosphorylation [7, 12-14, also see 15].
When a protein whose abundance oscillates with cleavage cycles in sea urchin
embryos was discovered in 1983, it was hypothesized that this protein, designated
cyclin, may be an important component of MPF [16]. Later, several studies purified a
34kD protein from MPF that was homologous to the cdc2+ gene of Schizosaccharomyces
pombe, which encodes a protein kinase that is essential for entry into mitosis [17-22].
MPF was finally purified into two identifiable components: a 47kD cyclin and the 34kD
cdc2 protein kinase (later termed CDK) that were present in a stoichiometric complex
and whose activity oscillated with embryonic cleavage cycles [23, 24].
1.3 What are CDKs and what do they do?
Cyclin dependent kinases (CDKs) are serine/threonine protein kinases that are
known to be essential for cell-cycle events in eukaryotes. Binding of a cyclin subunit is
required for the activity of CDKs. All cyclins share sequence similarity in a distinct
4
domain called the cyclin box which is necessary for CDK binding and activation
[reviewed in 25, 26]. Cyclins not only bind and activate CDKs, they have been proposed
to direct the kinase activity of CDKs to specific targets [27, 28, reviewed in 29, 30, 31].
In general, CDKs phosphorylate serines or threonines followed by a proline, however,
the full consensus sequence [K/R][S/T]PX[K/R] is required for phosphorylation of many
proteins [32-34]. Through protein phosphorylation, cyclin-CDK activity modulates the
activity of its targets.
The discovery of the cyclin-CDK complex brought about a CDK-centered model of
the cell-cycle. Throughout the cell cycle, cyclin levels steadily rise, reaching a peak at
mitosis, activating CDK to form MPF and driving mitosis through phosphorylation of
target proteins [16, 35] (Figure 1.1 A). At the end of mitosis B-cyclin-CDK promotes
cyclin proteolysis, which allows exit from mitosis and the beginning of a new cell cycle
[16, 36]. This cycle of cyclin synthesis and destruction turns with each cell cycle and was
proposed to be the fundamental oscillator that drives the cell cycle in studies of
embryonic cells [35, 36, reviewed in 37, 38]. Thus, CDKs are proposed to have two
fundamental roles in cell-cycle regulation. They are thought to both form the
underlying cell-cycle oscillator and serve as effectors of that oscillator, phosyphorylating
proteins involved in cell-cycle events.
In early embryonic systems, cyclin synthesis is fueled by a large store of maternal
mRNAs and proteins. Early cleavage division consists of rapid cycles of DNA replication
5
followed by mitosis. However, after several cleavage divisions, embryos reach the midblastula transition (MBT). This transition is marked by several changes in the cell cycle
including desyncronization of cell division, lengthening of the cell cycle and initiation of
zygotic transcription [39, 40, reviewed in 41]. These fundamental changes give rise to
the distinct somatic cell cycle, upon which the organism will rely throughout the rest of
development and its adult life.
Although embryonic cleavage cycles consist of only DNA replication and mitosis,
the somatic cell cycle includes the G1 and G2 cell-cycle phases, which allows for
additional control of cell cycle progression. The activation of zygotic transcription may
be fundamental to these changes in cell cycle regulation. During cleavage divisions, the
cell cycle is driven by cyclin accumulation through constitutive translation of maternal
mRNA followed by cyclin proteolysis [16]. At the MBT, cyclin accumulation becomes
dependent on transcription. In addition, simple cyclin-CDK oscillations become more
complicated as multiple cyclins and CDKs are produced that regulate not only mitosis,
but also other cell-cycle transitions (Figure 1.1).
1.3.1 How is the cell cycle regulated in somatic cells?
There are four major categories of cyclins that activate CDKs to regulate the
somatic cell cycle in metazoans. These different types of cyclins, denoted A, B, D and E,
are each important for the regulation of different parts of the cell cycle (Figure 1.1 B).
The D-type cyclins are important for regulation of the early events in the cell cycle. They
6
are transcriptionally activated in response to mitogenic signals that stimulate cells to
enter the cell cycle. D-type cyclins then bind to and activate Cdk4 and Cdk6 [reviewed in
42]. One important target of these active CDK complexes is the pRB family of
Figure 1.1: Cyclin dynamics in three different cell cycles depicted as cyclin-CDK activity
versus time. Different classes of cyclins are shown in different colors. (A) Early
embryonic cleavage divisions. (B) Metazoan somatic cells. (C) S. cerevisiae cells.
7
transcriptional repressors. Prior to cyclin D activation, pRB binds to and inhibits the
E2F family of transcription factors. When cyclin D-Cdk4/6 complexes phosphorylate pRB
this repression is relieved and E2F promotes the transcription of a large number of
genes that are important for progression from G1 into S-phase [reviewed in 43]. This
transition marks passage through the restriction point and commitment to the cell cycle.
Among the genes whose transcription is activated by E2F are the E-type and Atype cyclins [reviewed in 44]. Cyclins A and E activate Cdk2 and have partially
overlapping roles during the transition from G1 to S-phase and in DNA replication.
Although both complexes have been implicated in both functions, Cyclin E-Cdk2 is
thought to be important to further phosphorylate pRB proteins stimulating further E2F
dependent transcription, while cyclin A-Cdk2 is thought to be important for activating
DNA replication [reviewed in 42, 45]. Even though these two complexes have similar
roles, the regulation of cyclin E and cyclin A differs in regards to protein degradation.
Cyclin E-Cdk2 complexes phosphorylate cyclin E and target it for proteolysis by the
SCFCdc4 (Skp1, Cullin, F-box Cdc4) ubiquitin ligase [46]. This auto-inhibition imposes a
short lifespan on cyclin E. Cyclin A, however, persists until mitosis where it binds and
activates Cdk1.
The final family of cyclins, the B-type cyclins accumulate in G2 and peak in
mitosis. Like cyclin A, cyclin B activates Cdk1. Cyclin A-Cdk1 and cyclin B-Cdk1 also
8
share roles in activating mitosis. Cyclin A-Cdk1 may be important for chromosome
condensation prior to mitosis and in activation of cyclin B [47]. Cyclin B-Cdk1 is
responsible for stimulating the major events of mitosis including spindle formation and
nuclear envelope breakdown [48-50]. Following these events, cyclin B-Cdk1 activity
triggers spindle elongation at anaphase, the second irreversible transition of the cell
cycle. Starting at anaphase both cyclin A and cyclin B undergo ubiquitin-dependent
proteolysis by the anaphase promoting complex (APC) [reviewed in 51]. Destruction of
these mitotic activators allows the cell to complete mitosis, break down the mitotic
spindle and undergo cytokinesis. The new daughter cells are born with low CDK activity
and are poised to receive signals to enter the cell cycle again.
Throughout the cell cycle, there are a plethora of additional regulators of CDK
activity that add more complexity to this basic cyclin-CDK model of the cell cycle. CyclinCDK complexes are regulated not only by synthesis and degradation but also by
phosphorylation, localization and specific protein inhibitors. These additional levels of
control fine tune the cell-cycle and allow for responses to cellular stresses and the
environment. Despite the large number of cell-cycle regulators that have been
characterized in metazoan systems, oscillations of cyclin-CDK activity are thought to
remain at the core of cell-cycle regulation.
9
1.4 Saccharomyces cerevisiae as a model system for the study
of cell-cycle regulation
The budding yeast S. cerevisiae (hereinafter referred to as yeast) has been widely
studied as a model for all aspects of cell biology, with many proteins and mechanisms
being conserved from yeast to humans. In particular, budding yeast have proven to be
an excellent model for study of the cell cycle. The utility of yeast is due not only to the
unsurpassed advantages of yeast as a genetic system (the so-called “awesome power of
yeast genetics”), and to the relative rapidity of the cell cycle, but also to the unique
morphology of budding yeast as they traverse the cell cycle. In G1, yeast cells are small
and round. Once they pass through Start, yeast cells polarize growth and form a bud at
the G1/S transition. This bud will grow in size throughout the remainder of the cell cycle
and in mitosis it will receive a copy of the genome and become a daughter cell (Figure
1.2). Therefore, by microscopically observing yeast cells, it is possible to approximate
the cell cycle position of each cell by observing whether or not a cell has a bud and how
large the bud is.
These principles allowed for the classical mutational analysis of the cell cycle by
Leland Hartwell in the 1970’s [52-56]. In these studies, yeast cells were mutagenized
and mutants that arrested with distinct budding morphologies were analyzed. Mutants
that cannot complete G1 uniformly arrest as unbudded cells, while mutations in genes
essential later in the cell cycle arrest as budded cells. These mutants were denoted cell
division cycle (cdc) mutants. The cloning and characterization of CDC genes and their
10
Figure 1.2: Regulation of yeast cell-cycle events by cyclin-CDK activity. Different cyclinCDK complexes regulate important cell-cycle duplication and segregation events. SPB,
green; DNA, red.
gene products has given rise to a large field of study, which has resulted in a formidable
understanding of the budding yeast cell cycle, furthering its use as a model system. In
the sections that follow, I will summarize the vast literature relating to cell-cycle
regulation in yeast and point to interesting unanswered questions that will be addressed
in this dissertation.
1.4.1 Cyclin homologues in S. cerevisiae
Like the metazoan somatic cell cycle, the yeast cell cycle is regulated by multiple
cyclin-CDK complexes that are important for the regulation of cell cycle events (Figure
1.2). There are nine cyclins that bind to and activate a single CDK, Cdc28 or Cdk1
(henceforth referred to as CDK), at different points of the cell cycle (Figure 1.1 C).
11
Table 1.1: Cyclin genetics in S. cerevisiae
strain
12
cln3
GAL-CLN3
cln1,2
cln1-3
GAL-CLN2
cln1,2 clb5,6
clb5
clb6
clb5,6
clb3,4
clb3,4,5
clb1
clb2
clb1,2
clb2,3
clb2,4
clb1,3,4
clb2,3,4
clb1,2,3
clb1,2,4
clb1-4
clb1-5
clb1-6
GAL-CLB2Δdb
CLN3
+++
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
CLN1
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
CLN2
+
+
+++
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
CLB5
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
CLB6
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
CLB3
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
CLB4
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
CLB1
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
CLB2
+
+
+
+
+
+
+
+
+
+
+
+
+
+++
viable
phenotype(s) and reference(s)
yes
yes
yes
no
yes
no
yes
yes
yes
yes
no
yes
yes
no
no
yes
yes
no
no
no
no
no
no
no
large cell size [57, 58]
small cell size [57, 58]
[59]
unbudded [60, 61]
elongated buds [62]
[63]
delay in S-phase [63, 64]
[63, 64]
[63, 64]
[65, 66]
fails in spindle formation [63]
[65, 66]
delay in mitotic entry, elongated buds [65, 66]
elongated buds [65, 66]
[65, 66]
[65, 66]
[65, 66]
[65, 66]
elongated buds, rebuds [65, 66]
elongated buds [65, 66]
reduplicates SPBs, elongated buds, rebuds [65-68]
reduplicates SPBs, elongated buds, rebuds [67]
elongated buds, rebuds [68]
unbudded, mitotic exit failure [62, 69]
Based on sequence similarity, these cyclins fall into two classes, the G1 cyclins: Cln1,
Cln2, and Cln3, and the B-type cyclins, which are all homologous to metazoan cyclin B:
Clb1, Clb2, Clb3, Clb4, Clb5 and Clb6. Within both the G1-cyclins and the B-type cyclins,
there is considerable functional overlap among cyclins. This redundancy leads to
complicated genetic relationships among the yeast cyclins, with mutations in individual
cyclin genes having little effect on the cell cycle. These relationships are summarized in
Table 1.1 and will be detailed in the following sections.
1.4.2 Regulation of G1 events by G1-cyclin-CDK activity
Classical mutations in the yeast CDK, CDC28, arrest as unbudded cells, suggesting
that CDK is essential for G1 events [56]. During G1, CDK is activated by the three
partially redundant G1 cyclins; the deletion of all three G1 cyclins is lethal and
phenocopies a cdc28 arrest [60, 61]. Early in the cell cycle, Cln3-CDK is the only active
CDK complex. Its activity builds through G1 and triggers Start when Cln3-CDK activity
reaches a critical level. The key event of Start is induction of the G1 transcriptional
program which is controlled by two heterodimeric transcriptional activators that are
analogous to metazoan E2F: SBF (Swi4 cell cycle box [SCB] binding factor) and MBF
(Mlu1 cell cycle box [MCB] binding factor) [70, 71, reviewed in 72]. In early G1, the
functional homologue of metazoan pRb, Whi5, inhibits G1 transcription through
inhibitory binding of the transcriptional activators SBF and possibly MBF [73, 74]. Cln3CDK is known to trigger G1 transcription through phosphorylation of Whi5 [73, 75].
13
Hyper-phosphorylated Whi5 is exported from the nucleus where it can no longer inhibit
G1 transcription [73, 74].
The SBF and MBF transcription factors promote the transcription of a large
number of transcripts at the G1/S border. Many of these genes are factors involved in
budding, duplication of the yeast centrosome (the spindle pole body, SPB), DNA
replication and the cyclins CLN1, CLN2, CLB5 and CLB6. The cyclins CLN1 and CLN2 as
well as many genes important for budding are targets of SBF while the cyclins CLB5 and
CLB6 and genes important for SPB duplication and DNA replication are thought to be the
targets of the MBF transcriptional activator [63, 76, 77, reviewed in 78]. However, both
genetic and transcription factor localization studies indicate that there is significant
redundancy between SBF and MBF [77, 79-82].
There is evidence that activation of G1 transcription triggers a positive feedback
loop at Start in which newly synthesized Cln1/2-CDK activity further actives G1
transcription through phosphorylation of Whi5 [83, also see 84, 85, reviewed in 86].
This feedback loop is likely important to ensure rapid and robust entry into the cell cycle
at Start. In addition to the activation of Start, G1 cyclins have many other important
roles in G1. Cln1/2-CDK activity promotes apical growth, or directed growth towards
the bud tip, which is required for budding [62]. Cln1/2-CDK activity is also important for
SPB duplication [87-89]. In addition, Cln1/2-CDK promotes the transition into S-phase
through the activation of B-cyclins.
14
1.4.3 S-phase cyclin-CDK activity
The S-phase B-type cyclins CLB5 and CLB6 are transcribed along with CLN1 and
CLN2 in G1, however, they are not active until S-phase. In G1, Clb5 and Clb6 are
targeted for proteolysis by the E3 ubiquitin ligase APCCdh1 (Anaphase Promoting
Complex, activated by Cdh1) [90, reviewed in 91], and their associated CDK activity is
inhibited by the B-cyclin-specific CDK inhibitor, Sic1 [92, 93]. The APC activator Cdh1 is
inactivated in late G1 through phosphorylation by Cln1/2-CDK allowing the persistence
of B-cyclin proteins [94]. Sic1 is also phosphorylated by Cln1/2-CDK activity which
targets Sic1 for ubiquitination by the SCFCdc4 E3 ubiquitin ligase [95-97]. Upon
destruction of Sic1, and inactivation of APCCdh1, Clb5/6-CDK becomes active and S-phase
begins [63].
Either Clb5-CDK or Clb6-CDK can trigger DNA replication; however, Δclb5 cells
but not Δclb6 cells progress slowly through S-phase [63, 64]. This difference in mutant
phenotypes between the S-phase cyclins is due to a difference in regulation. While Clb5
is stable until mitosis, when it is targeted for degradation by the APC, Clb6 is
phosphorylated and targeted for degradation by the SCFCdc4 shortly after it becomes
active [98]. Despite these roles for Clb5 and Clb6 for S-phase entry, both CLB5 and CLB6
are dispensable for DNA replication, as Δclb5, 6 cells are viable [63]. In fact, any of the
six B-type-cyclins, Clb1-6, are capable of supporting S-phase, however a strain carrying
15
mutations in all six B-cyclin genes (Δclb1-6) arrests at the G1/S border with a single copy
of the genome [93].
1.4.4 Roles for mitotic B-cyclins
Late in S-phase the first set of mitotic B-cyclins, CLB3 and CLB4 are transcribed
[65, 66]. Cells lacking both CLB3 and CLB4 are viable and do not have any detectable
phenotypes. However, in wild-type cells, Clb3/4-CDK activity is thought to be important
in late S-phase for separation of the duplicated SPBs into a short spindle [65, 66]. While
CLB3 and CLB4 are not essential for spindle assembly, a Δclb3,4,5 mutant arrests
without a spindle indicating that Clb5-CDK activity can support spindle assembly in the
absence of CLB3 and CLB4 [63].
The remaining mitotic cyclins, CLB1 and CLB2 are expressed after CLB3 and CLB4
but before mitotic entry [65, 66]. Together, the four mitotic B-cyclins are important for
mitosis, however, Clb2-CDK activity is the most potent activator of mitosis. Strains
carrying a mutation in CLB2 are viable, but accumulate as budded cells with replicated
DNA, suggesting Clb2-CDK activity has a key role in mitotic entry [65, 66]. The double
mutants Δclb1,2 and Δclb2,3 are lethal as are all combinations of three or more mitotic
B-cyclin deletions containing Δclb2 (see Table 1.1). In addition, mitotic B-cyclins are also
important to oppose the apical growth promoted by Cln1/2-CDK and promote isotropic
growth, or growth in all directions [62]. Accordingly, most mitotic B-cyclin mutants
16
display elongated buds while over-expression of CLB1 or CLB2 inhibits apical growth and
prevents budding [62, 90].
Mitotic B-cyclins, in particular Clb2-CDK activity, have important roles in the
regulation of periodic transcription. In cells lacking mitotic B-cyclins, the expression of
CLN2 and other SBF-regulated transcripts remains high outside of G1, suggesting that
mitotic B-cyclins have a role in turning off SBF transcriptional activity [69, 99]. Mitotic Bcyclins do not seem to have an effect on MBF regulated transcription [99], however,
MBF has been shown to be down-regulated by a negative feedback loop involving the
product of a MBF transcript, NRM1 [100].
Similar to the positive feedback loop involving G1 cyclins and G1 transcription,
mitotic B-cyclins participate in a positive feedback loop that controls their expression.
CLB2 is a member of a cluster of genes periodically expressed in G2/M [101, 102]. The
expression of these genes is thought to be regulated by a complex of transcription
factors termed SFF (Swi-Five Factor) [103]. Phosphorylation of the SFF regulatory
subunit by Clb2-CDK is involved in the activation of transcription at SFF-regulated genes
[104-106]. Thus, Clb2 promotes its own transcription forming positive feedback loop.
Like the positive feedback loop involved with CLN1/2 expression, this positive feedback
loop is thought to ensure a sharp increase in Clb2-CDK activity and a robust transition
into mitosis [reviewed in 78].
17
The essential activity of mitotic B-cyclins (Clb1-4) in yeast, as in metazoans, is
activation of mitosis [107]. Mitotic B-cyclin-CDK activity promotes the phosphorylation
of several components of the APC which allows binding of the APC activator Cdc20
[108]. APCCdc20 activates mitosis by promoting the ubiquitination and subsequent
proteolysis of the anaphase inhibitor, Pds1 (also called securin) [109]. Pds1 binds to and
inhibits the protease, Esp1 (also called separase) [110]. Once Pds1 inhibition of Esp1 is
relieved, Esp1 cleaves cohesion, a protein that holds sister chromatids together [110].
In the absence of cohesion, the spindle can elongate as cells enter anaphase [111].
1.4.5 Mitosis and mitotic exit in yeast
Pds1 is not the only important target of the APCCdc20 [112]. APCCdc20 also
promotes the ubiquitnation of Clb2 and presumably the other mitotic B-cyclins,
although APCCdc20 alone cannot eliminate mitotic B-cyclin activity [113]. Through the
activation of APCCdc20, which in turn triggers the proteolysis of mitotic B-cyclins and the
progression of mitosis, mitotic CDK activity promotes its own destruction.
The next steps in cell-cycle progression are completion of anaphase and exit
from mitosis. Exit from mitosis requires complete loss of mitotic B-cyclin-CDK activity
which is regulated by the phosphatase Cdc14 [reviewed in 114]. Mutants in CDC14
arrest in late anaphase with high mitotic B-cyclin activity [52, 115]. Until mitosis, Cdc14
is sequestered in the nucleolus by its inhibitor Net1 [116, 117]. During mitosis Cdc14 is
18
released in two different waves once in early anaphase and a second time when the
nucleus proceeds through the mother-bud neck [118, reviewed in 119].
Cdc14 is important for the critical events of mitotic exit, spindle disassembly,
chromosome decondensation and most importantly the destruction of the remaining
mitotic B-cyclins. The critical targets of Cdc14 for the inhibition of mitotic CDK activity
are Cdh1, Sic1 and Swi5, a transcription factor important for SIC1 expression [94, 115,
120]. Dephosphorylation of Cdh1 activates APCCdh1 which takes over for APCCdc20 in
targeting mitotic B-cyclins for destruction [94, 121]. Accordingly, over-expression of a
CLB2 allele with the APC targeting domain, the destruction box (db), removed (Clb2Δdb)
causes arrest in late mitosis [90]. Dephosphorylated Sic1 is stable and directly inhibits Bcyclin-CDK activity, while dephosphorylation of Swi5 allows nuclear import of the
transcription factor and activation of SIC1 transcription [96, 122-124]. Together, these
three Cdc14 substrates remove mitotic B-cyclin/CDK activity from the cell [121]. In
addition to inhibition of mitotic CDK activity, Cdc14 also reverses the phosphorylation of
many CDK targets which allows cytokinesis and return to a low-CDK, G1 state in both the
mother and daughter cell [reviewed in 125].
1.5 Checkpoints: mechanisms that ensure the order of cell cycle
events
Additional cell-cycle control mechanisms called checkpoints also exist in both
metazoans and yeast that ensure the correct order of essential cell cycle events.
19
Checkpoint mechanisms are defined as signaling pathways that delay cell-cycle
progression when important cell-cycle events are perturbed, making the delayed event
dependent on the perturbed event [126]. There are several checkpoint and checkpointlike mechanisms that have been defined in both yeast and metazoans; however this
discussion will be limited to the three classic checkpoints, the spindle checkpoint, the
DNA-replication checkpoint, and the DNA-damage checkpoint. The morphogenesis
checkpoint, which is yeast-specific, but utilizes a pathway common to both yeast and
metazoan checkpoints, is also discussed.
The spindle checkpoint ensures that cells do not enter anaphase until the spindle
has bipolar attachments to all sister chromatids by triggering a kinase cascade when
these conditions are not met. The result of the pathway is inhibition of APCCdc20 through
several mechanisms which include phosphorylation of Cdc20 by Bub1 and direct
inhibition of the APCCdc20 by Mad3 [127, 128, reviewed in 129]. Through this pathway,
spindle elongation is inhibited until biorientation is achieved ensuring equal segregation
of chromosomes.
Two checkpoints ensure that cells do not enter mitosis until the genome has
been faithfully replicated, the DNA-replication checkpoint and the DNA damage
checkpoint. These two checkpoints employ very similar mechanisms, but differ in the
conditions that trigger the checkpoint. The DNA -replication checkpoint detects stalled
replication forks, which are induced by depletion of nucleotides or inhibition of factors
20
involved in progression of DNA replication, while the DNA-damage checkpoint detects
DNA damage induced by irradiation, alkylation or DNA breaks [reviewed in 130]. In
addition, the DNA-damage checkpoint can delay the cell cycle at multiple points while
the lesion is repaired, while the DNA-replication checkpoint has only been shown to
delay mitotic progression. While many members of the signaling pathway involved in
the DNA-replication and DNA-damage checkpoints are conserved between yeast and
metazoans, the mechanism by which cell-cycle progression is stalled differs. In yeast,
both the DNA-replication and the DNA-damage checkpoint pathways can lead to
inhibition of the metaphase-anaphase transition through direct stabilization of Pds1 and
regulation of Cdc20 [131-134]. In metazoans, these checkpoints inhibit mitosis by
phosphorylating and inhibiting the Cdc25 phosphatase which removes inhibitory
phosphorylations from CDK thus, preventing cells from entering mitosis [reviewed in
135].
In yeast, inhibitory phosphorylations of CDK are important for stalling cell-cycle
progression in response to defects in bud formation through the morphogenesis
checkpoint. The morphogenesis checkpoint prevents yeast cells from entering mitosis
in the absence of bud formation, as progression through mitosis would result in a
binucleate cell [136, reviewed in 137]. This checkpoint also leads to accumulation of
CDK with inhibitory phosphorylations; however the kinase, Swe1, rather than the
phosphatase seems to be regulated. In the presence of a bud, Swe1 is degraded,
21
however, if bud formation is inhibited, Swe1 is stabilized and phosphorylated CDK
accumulates, blocking mitotic B-cyclin/CDK activity and mitotic entry [138, 139].
Through these and other mechanisms, checkpoints add an extra layer of
precision to the control of the cell cycle. They ensure that early events precede late
events, which ensures fidelity in the duplication and segregation of cellular contents.
1.6 Common threads in cell cycle regulation
I have described the regulation of the cell cycle in three different cell types,
metazoan early-embryonic cells, metazoan somatic cells, and yeast cells. There are
several important characteristics that these three different cell types share. First, cellcycle regulation in all three cell types promotes the faithful replication of cellular
contents including, but not limited to, the genome. Second, cell-cycle regulation
controls the segregation of its duplicated contents into two daughter cells. And finally,
cell-cycle regulation in all three cell types (and many other eukaryotic cell types that
have been investigated) involves low levels of CDK activity early in the cell cycle building
to high levels of B-cyclin-CDK activity at mitosis which promotes B-cyclin proteolysis at
the completion of mitosis (Figure 1.2). However, the early embryonic cell cycle differs
from both the somatic cell cycle and the yeast cell cycle in that it is not reliant on
transcription for accumulation of cyclin protein. These fundamental characteristics raise
several important questions.
22
First, in embryonic cells there is a constant input of cyclin allowing cyclin-CDK
oscillations to drive cell-cycle oscillations. However, as cyclin trancription is regulated in
both somatic cells and yeast, cyclin-CDK oscillations cannot form an autonomous
oscillator in these cells. Several models of cell-cycle oscillations in yeast and somatic
cells have been proposed that either ignore cyclin transcription altogether or include
both transcription and cyclin activation in a complicated yet incomplete manner [140142]. The question remains, what is the fundamental cell-cycle oscillator in yeast and
somatic cells, and how does this oscillator contribute to cell-cycle regulation?
Second, it is not entirely clear how are duplication and segregation events are
regulated to ensure faithful transmission of the genome from mother to daughter cell.
It is clear that cyclin-CDK activity has roles in promoting and inhibiting cell-cycle events,
and that checkpoints can ensure order of cell-cycle events. However, it is not clear how
these mechinisms ensure that duplicaiton and segregation events occur at the proper
time and occur only once per cell cycle.
1.7 What is the role of B-cyclins in the regulation of the yeast
cell cycle?
As I have described above, B-cyclins have been shown to have roles in both
activating and repressing cell-cycle events. In order to further elucidate the roles of
these important cell-cycle regulators, and answer fundamental questions about the
regulation of the cell cycle, several studies have analyzed cell-cycle events in yeast cells
23
lacking S-phase and mitotic B-cyclins (Δclb1-6 cells, or B-cyclin mutant cells). These
studies, detailed below, have interesting implications for understanding both what Bcyclins regulate and do not regulate during the cell cycle.
1.7.1 Independent oscillations of G1 events in the absence of B-cyclins
The gene CDC4 was initially isolated as a temperature sensitive mutation that
arrested at restricted temperature without initiating DNA replication [54]. Unlike other
mutants that arrest with unreplicated DNA, cdc4ts cells do not arrest with a single bud.
At the restrictive temperature these cells arrest the cell cycle prior to DNA replication,
however, bud initiation continues periodically at cell cycle intervals allowing cdc4 ts cells
to accumulate as many as five buds before arresting or lysing [54]. It was proposed that
the periodic bud initiation in cdc4ts mutants is driven by a cellular clock that keeps time
and controls bud initiation in the absence of cell cycle progression [54, 143]. Two
additional mutants, cdc34ts and cdc53ts, were later identified that are unable to replicate
DNA and produce multiple buds like cdc4 cells [144-146]. The gene products of these
mutants were shown to be involved in the SCFCdc4 ubiquitination of Sic1, suggesting that
the mutants are unable to initiate DNA replication due to inhibition of B-cyclin-CDK
activity by stabilized Sic1 [96]. The multi-budded phenotype of cdc4ts, cdc34ts and cdc5 ts
remained enigmatic.
Eighteen years after it was proposed that multiple bud formation in cdc4ts cells
is controlled by a cellular clock that runs independently of cell-cycle progression, this
24
idea was investigated further [68]. Like cdc4ts cells, cells lacking all six B-type cyclins
(Δclb1-6) and cells over-expressing a hyper-stable, non-phosphorylatable allele of SIC1
(Sic1Δ3P) were shown to periodically form new buds and express the G1 cyclin CLN2 at
cell cycle length intervals in the absence of DNA replication and mitosis [68]. These
experiments indicate that a B-cyclin independent oscillator (or cellular clock), can drive
G1 events in the absence of cell cycle progression.
Evidence also suggests that this B-cyclin-independent oscillator may be
independent of all cyclin-CDK activity. When haploid yeast cells are exposed to mating
pheromone from the opposite mating type they arrest the cell cycle prior to Start due to
the expression of the G1-cyclin specific CDK inhibitor Far1 [147]. While arrested with
mating pheromone, cells polarize to form mating projections or “shmoos”, which ideally
lead to contact between cells of opposing mating types and subsequent conjugation. If
cells do not conjugate, they will make another mating projection in a different direction
[148]. By treating synchronized mating type a cells lacking G1 cyclins with a uniform
concentration of pheromone from mating type α cells (α-factor), mating projection
formation was shown to occur periodically with a period similar to cell-cycle length in
cycling cells [68]. These results suggest that mating projection formation may be
entrained to the same cell-cycle oscillator that drives bud formation in the absence of Bcyclin activity, and that this oscillator may be independent of all CDK activity [68].
25
The relationship of the CDK-independent oscillator to checkpoints was also
investigated. When B-cyclin activity was inhibited by Sic1Δ3P over-expression in cells
arrested at the DNA-replication or the spindle-assembly checkpoint, the cells began to
rebud and express CLN2 [68]. These results suggest that B-cyclins either directly
prevent G1 events during checkpoint arrest, or that B-cyclins can restrain the activity of
the independent oscillator.
This study indicates that, although B-cyclin-CDK activity is essential for cell-cycle
progression, neither B-cyclins nor cell-cycle progression is essential for periodic cellcycle events. This observation raises several fundamental questions: what is the nature
of the B-cyclin-independent oscillator? Is the B-cyclin-independent oscillator a
fundamental cell-cycle regulator? How are the B-cyclin-independent oscillator and cellcycle events normally entrained to each other? In the next section I will describe a
study that begins to shed light on some of these questions.
1.7.2 Periodic transcription in the absence of cyclin-CDK activity
Upon discovery that cdc4ts cells continued to bud in the absence of cell-cycle
progression, it was proposed that the basis of the cellular clock that kept time in these
cells might be related to periodic transcription [54]. It has been shown that a large
number of genes are periodically expressed during the cycle and it is thought that cyclinCDK activity is important for regulation of periodic transcription [102, 149].
Alternatively, it has been suggested that a transcription-factor (TF) network, a network
26
with each node being a TF that is important for the transcriptional regulation of downstream TFs, could be responsible for the control of periodic transcription [80, 150]. To
test whether CDK activity or a TF network controls periodic transcription and to test the
hypothesis that periodic transcription may be involved in B-cyclin-independent
oscillations, global mRNA dynamics were measured in cells lacking all six B-type cyclins
(Δclb1-6) and compared to global mRNA dynamics in wild-type cells [99]. This
experiment showed that 70% of periodically expressed genes remain periodically
expressed in the absence of S-phase and mitotic cyclins as well as cell-cycle progression
[99]. This result indicates that cyclin-CDK activity is not solely responsible for the
regulation of periodic transcription during the cell cycle. Using known TF binding sites, a
TF network that has the ability to oscillate in silico when modeled within a Boolean
framework was constructed from TFs that remain periodic in cyclin-mutants cells [79,
99] (Figure 1.3). This result suggests that a TF network oscillator could be responsible
for maintaining oscillations in the absence of B-cyclin activity [99]. The proposal that a
B-cyclin-independent-TF-network oscillator maintains periodic transcription raises the
interesting possibility that normal cell-cycle oscillations are regulated by both TFnetwork oscillations and CDK-activity [99].
27
Figure 1.3: TF network oscillator proposed to maintain oscillations in the absence of Bcyclins. Green boxes indicate transcriptional activators, red boxes indicate
transcriptional repressors and the blue box indicates a cyclin. Edges from TFs represent
TF protein binding to the promoter of the downstream TF, while edges from the cyclin
indicate post-transcriptional activation. Adapted from [99].
This study suggests that B-cyclin-independent oscillations may be maintained by
a TF-network oscillator. In this dissertation, I will use this model of the B-cyclinindependent oscillator to ask further questions about the role of CDKs in cell-cycle
oscillations.
1.7.3 SPB reduplication in the absence of mitotic B-cyclins
The microtubule organizing center (the centrosome in metazoans and the SPB in
yeast) must duplicate once per cell cycle to form the bipolar spindle important for
segregation of the genome. Accumulation of additional centrosomes in metazoans has
been linked to genomic instability and cancer formation [151, 152]. CDK activity has
been implicated in the regulation of centrosome duplication and SPB duplication in
28
yeast [reviewed in 153, 154]. To elucidate the roles of B-cyclins in the regulation of the
SPB duplication cycle, the behavior of the SPB duplication cycle was examined in the
absence of B-cyclins.
The SPB is a multi-protein structure situated in the nuclear envelope throughout
the cell cycle and serving as the sole microtubule organizing center in yeast.
Observation of the SPB by fusion of GFP to the SPB protein Spc42 (Spc42-GFP) revealed
that cells lacking all B-type cyclins (Δclb1-6) arrest with a pair of duplicated SPBs, while
cells lacking only the mitotic B-cyclins (Δclb1-4) arrest with 3-4 SPBs indicating that
mitotic B-cyclins are important for the inhibition of SPB reduplication during the normal
cell cycle [67]. Interestingly, cells lacking 5 of the 6 B-type cyclins (Δclb1-5) undergo
several rounds of SPB re-duplication and accumulate many SPBs [67]. As S-phase cyclins
are necessary for SPB reduplication in the absence of mitotic cyclins, and CLB6 seems to
promote additional rounds of SPB reduplication in Δclb1-5 cells, S-phase cyclins are
thought to be important for a “maturation” step that is essential for SPB reduplication
and correlates with SPB separation. These data suggest new roles for B-cyclin-CDK
activity in promoting and restraining SPB duplication during the yeast cell cycle.
This study implicates B-cyclin-CDK activity in the regulation of SPB duplication
and raises interesting questions about the role of B-cyclins and the B-cyclin-independent
oscillator in promoting and restraining SPB duplication during the cell cycle.
29
1.7.4 Questions raised by studies of yeast cells lacking B-cyclin activity
Current studies of yeast cells lacking B-cyclin-CDK activity indicate that B-cyclins
have roles in both promoting and inhibiting cell-cycle events. A number of important
questions about the role of B-cyclins in the cell cycle as well as nature and role of the Bcyclin independent oscillator remain. Many of these questions will be investigated in
the course of this dissertation. Two key questions that fall out of previous studies of
cells lacking B-cyclin activity that will be addressed in depth are discussed in the sections
that follow.
1.8 What is the function of CDKs in cell cycle oscillations?
The finding that yeast cells maintain a majority of periodic transcription in the
absence of B-cyclin-CDK activity suggests that a TF network oscillator may be
responsible for maintaining oscillations in the absence of cell cycle progression. To
begin to explore the nature of this proposed oscillator, I will first consider the concept of
biological oscillators, in particular those thought to be involved in cell cycle oscillations.
1.8.1 Biological oscillators
Cyclical behaviors are ubiquitous in biology. Any repetitive behavior that occurs
regularly with respect to time can be described as an oscillation. The time it takes to
complete one round of behaviors is referred to as the period, while the magnitude of
the behavior is referred to as its amplitude. A biological oscillator is a pathway or
30
network of factors that are responsible for generating and maintaining the oscillatory
behavior of the biological system.
Recent interest in understanding and mathematically modeling simple genetic
circuits has led to a flurry of literature on biological oscillators. Mathematical models
have defined a series of conditions necessary to achieve robust, sustained oscillations in
silico, the most relevant of which is a requirement for delayed negative feedback
[reviewed in 155]. Negative feedback is thought to be important to bring a system back
to its original starting state, while the time delay is necessary to prevent the system
from reaching a steady state and often sets the period of the oscillation. These
principles have permitted the construction of synthetic biological oscillators, the first of
which is constructed of three heterologous TFs each repressing the transcription of the
next with the final TF repressing the transcription of the first [156]. This oscillator,
named the repressilator, creates cycles with a period of about 150 minutes in
Escherichia coli as observed by a GFP reporter [156]. The construction of the
repressilator represents a benchmark in understanding oscillatory networks, however,
there is still much to be learned about naturally occurring biological oscillators.
Perhaps the best-studied class of biological oscillators is circadian oscillators,
which are responsible for entraining living things to the light and dark cycles of the
rotating earth. Interestingly, most organisms continue to display daily behaviors in the
absence of light-dark cycles, suggesting that organisms have an internal clock that keeps
31
time. The first hints of the mechanism behind daily behavioral rhythms came from
forward genetic screens in Drosophilia melanogaster, Neurospora crassa and other
organisms [157, 158]. These screens identified mutants that either displayed periods
that dramatically differed from the normal twenty-four hour period, or that no longer
displayed periodic behaviors. Further investigations found several additional mutants
that affect circadian cycles [reviewed in 159]. Interestingly, these and other eukaryotes
studied have a similar mechanism for generating twenty-four hour oscillations. This
mechanism involves a simple negative feedback oscillator in which a conserved family of
TFs activates the transcription of genes whose product negatively regulates the TFs with
time delays sufficient for one complete cycle to take a full day [reviewed in 160]. The
genes regulated by the oscillator and the mechanisms by which light-dark cycles entrain
the oscillator differ among organisms. The structure of the oscillator, however, is
conserved, suggesting that this simple negative-feedback oscillator is a robust
mechanism for regulating daily behaviors.
1.8.2 Cell cycle oscillators
The cell-cycle is a series of events that occurs regularly with respect to time.
Thus, at its core, the cell-division cycle is a biological oscillator with complicated outputs
to regulate the essential processes described above. Self-propelled oscillations of cyclin
B in embryonic systems have suggested that cyclin-CDK activity composes the cell-cycle
oscillator in eukaryotes. The embryonic cyclin-CDK oscillator is also a relatively simple
32
negative feedback oscillator with cyclin B-Cdk1 activating the APC which promotes
degradation of cyclin B [16, 35]. This oscillator also includes a positive feedback loop
whereby cyclin B-Cdk1 activates the phosphatase Cdc25 and inhibits the kinase Wee1;
together, these events lead to dephosphorylated, active Cdk1 [161-165]. In addition,
this positive feedback loop has also been shown to be important for a robust transition
from interphase to mitosis [166]. Thus, the embryonic cyclin-CDK oscillator is important
for both regulation of cell-cycle transitions and the oscillatory behavior of the embryonic
cell-cycle [reviewed in 155].
An entirely different oscillator governing the cell-division cycle has been
discovered in the bacteria, Caulobacter crescentus [reviewed in 167]. Despite being a
prokaryote, Caulobacter displays many characteristics of eukaryotic cell cycles. It
initiates DNA replication during a discrete phase of its division cycle between a G1-like
pre-replication stage and a post-replication phase in which it undergoes cell division
[168, 169]. Interestingly, the Caulobacter cell cycle oscillator is made up of a network of
TFs. This network contains several sequentially activated TFs that both promote cellcycle-phase specific transcripts and the expression of other TFs in the network [170172]. This network possesses the essential components of an oscillator, including
delayed negative feedback, and has the capacity to regulate cell-cycle transitions and
serves as a model for the role of transcription in cell-cycle oscillations.
33
As discussed above, cell-cycle oscillations in yeast and somatic cells are distinct
from cell-cycle oscillations in embryonic systems, in that the expression of cyclins is
controlled transcriptionally. Thus the cell-cycle oscillator in yeast and somatic cells must
contain a transcriptional element in the regulation of cyclin-CDK activity. The finding
that B-type cyclins are not necessary for transcriptional oscillations in yeast suggests
that a transcriptional oscillator rather than a CDK oscillator may be at the core of yeast
cell cycle oscillations [68, 99]. As B-cyclin-CDK activity is essential for DNA synthesis and
mitosis each cell cycle, it was proposed that cell cycle oscillations in yeast are the result
of a TF network oscillator and a cyclin-CDK oscillator coupled to each other. In this
model, cyclin synthesis is regulated by transcription and in turn cyclin-CDK activity
regulates the activity of transcription factors through phosphorylation [173].
1.8.3 CDK-independent cell cycle oscillations?
The model that yeast cell-cycle oscillations are regulated by coupled TF network
and cyclin-CDK oscillators stems from the result that yeast lacking S-phase and mitotic
cyclins maintain a majority of periodic transcription [99]. The TF network proposed to
maintain oscillations in the absence of B-cyclins contains the cyclin, Cln3. However, it
was previously suggested that oscillations occurred in the absence of all CDK activity
[68]. So, are B-cyclin independent oscillations dependent on G1-cyclin-CDK activity? Do
CDK-independent oscillations exist? If so, what is the role of CDKs? What is the
relationship between CDK-independent oscillations and other known cell-cycle control
34
mechanisms? These questions will be considered and addressed in this dissertation,
and the resulting data will suggest new models for cell-cycle regulation.
1.9 How do B-cyclins ensure that SPB duplication occurs once
and only once per cell cycle?
Spindle formation and spindle elongation are essential cell-cycle events that
require exactly two SPBs, however, relatively little is known about how SPB duplication
is regulated. As cells lacking mitotic B-cyclins reduplicate SPBs, it is likely that mitotic Bcyclin-CDK activity inhibits SPB reduplication during the normal cell cycle. The
mechanisms by which B-cyclin-CDK activity restrains SPB duplication, however, are not
well understood. The mechanism preventing re-replication of the genome has been
studied extensively and will be considered as a model for the regulation of SPB
duplication.
1.9.1 Restriction of DNA replication to once per cell cycle
Initiation of DNA replication at the onset of S-phase requires assembly of a multiprotein complex, known as a pre-replicative complex (pre-RC) at origins of replication.
After formation of a pre-RC in early G1, Clb5/6-CDK activity triggers the initiation of DNA
replication at the beginning of S-phase. After initiation, the complex remaining at the
origin of replication, the post-replicative complex (post-RC), is no longer competent to
initiate DNA replication. DNA replication at each origin is thought to be limited to once
per cell cycle by B-cyclin-CDK phosphorylation of essential pre-RC proteins. Three
35
different targets of B-cyclin-CDK activity have been shown to be important for
preventing pre-RC assembly [174-176]. Mutations in all three of these B-cyclin-CDK
targets are necessary to allow DNA re-replication within a single cell cycle, indicating
that these mechanisms have redundant roles in preventing DNA re-replication [176].
Phosphorylation of these proteins, and probably others, helps prevent re-assembly of a
pre-RC during the remainder of the cell cycle. The post-RC persists until the completion
of mitosis, when B-cyclins have been destroyed, allowing the pre-RC to reassemble. The
assembly of pre-RCs is known as DNA replication licensing, as it makes origins of
replication competent to replicate again. By activating DNA replication and
subsequently inhibiting licensing, B-cyclin-CDK activity is able to ensure that DNA
replication occurs once, and only once, per cell cycle [reviewed in 177, 178].
1.9.2 Regulation of centrosome duplication
In metazoans, the microtubule organizing center is called the centrosome.
Despite a correlation between supernumerary centrosomes and cancer and implications
of accumulation of additional centrosomes in tumorigenesis [152, 179-184], little is
known about the centrosome or how its duplication is regulated. In this section, I will
discuss what is known about the centrosome and centrosome duplication.
The structure of the centrosome has been studied by electron microscopy (EM).
The centrosome consists of two centrioles arranged orthogonally within a relatively
dense mass of proteins called the pericentriolar material (PCM). The microtubule
36
nucleating protein, γ-tubulin, is one of the many proteins within the PCM. Microtubules
elongate from γ-tubulin to form the mitotic spindle and other cellular microtubules
[reviewed in 185].
At mitosis, there are two centrosomes that localize to opposite poles of the cell.
From this position, the centrosomes are able to produce the bipolar spindle that will
segregate the chromosomes into the two daughter cells. In order for successful
chromosome segregation, it is essential that there are only two centrosomes present in
the cell at the time of mitosis. If a cell has more than two centrosomes, a multi-polar
spindle may form, potentially causing unequal division of chromosomes and resulting in
mitotic catastrophe. Although such an event is unlikely to produce viable daughter cells,
an increase in centrosome number has been shown to cause chromosome instability
and produce anneuploid daughter cells [151, 152, 185, 186].
Centrosome duplication has best been characterized morphologically by electron
microscopy (EM) [reviewed in 185]. The centrosome duplication cycle is illustrated in
Figure 1.4A. The first event that takes place in between mitosis and G1 is centriole
disorientation, the centrioles move from their tight orthogonal configuration and
separate slightly, but still share a PCM [185, 187]. Next, a structure called a procentriole
assembles at the proximal end of each centriole. The new centrioles extend from the
procentriole during S-phase and G2, a process I will refer to as centriole duplication
[187, 188]. At mitosis the two pairs of centrioles separate to opposite poles of the cell,
37
Figure 1.4: Centrosome duplication cycles. (A)The metazoan centrosome. Step 1:
Procentriole formation. Step 2: Centriole duplication. Step 3: Spindle formation and
mitosis. Step 4: Centriole disorientation. (B) The yeast SPB. Step 1: Bridge elongation.
Step 2: Satellite deposition. Step 3: SPB duplication. Step 4: SPB separation. Step 5:
mitosis
38
forming the mitotic spindle [187, 188]. Each daughter cell will inherit one complete
centrosome and the cycle will repeat during the next cell cycle.
Numerous cell-cycle regulators have been implicated in the centrosome
duplication cycle. Centrosome reduplication can occur in some cell types during an
extended S-phase. This reduplication has been shown to be dependent on cyclin E-Cdk2
and cyclin A-Cdk2 activity and to be inhibited by cyclin B-Cdk1 activity [186, 189, 190].
These data suggest that cyclin E-Cdk2 and/or cyclin A-Cdk2 activity is important for
centriole duplication. In D. melanogaster, data suggest that cyclin A/B-Cdk1 may inhibit
centriole duplication and loss of Cdk1 leads to additional centriole formation [191, 192].
Additionally, it has been shown that centrosome disorientation is required for
subsequent centriole duplication and that centrosome disorientation is dependent on
separase [193, 194]. Taken together, these data suggest a mechanism in which different
CDK activities promote and inhibit centrosome duplication; however, as these studies
have taken place in a variety of metazoan systems, it is not yet clear how centrosome
duplication is regulated during the normal cell cycle in any given organism.
1.9.3 SPB duplication and SPB licensing
The SPB is the functional equivalent of the metazoan centrosome; both
organelles are essential for formation of a spindle at mitosis and segregation of sister
chromatids. While there are obvious morphological differences between the
centrosome and the SPB, there is conservation of function between the two organelles.
39
Furthermore, there are several parallels between SPB duplication and centrosome
duplication, making the yeast SPB a good model to study centrosome duplication [187].
In addition, the yeast SPB is an attractive model because it has been studied extensively
both genetically and structurally by EM. Evidence that CDK activity has roles in
promoting and inhibiting the duplication of both the SPB and the centrosome suggests
that regulatory mechanisms may be conserved [67, 87, 186, 189-192]. In this section, I
will summarize what is known about SPB duplication and pose questions that will be
addressed in this dissertation.
During the normal cell cycle, the cell inherits a single SPB during mitosis. EM
studies have revealed that the SPB is a cylindrical structure with at least three layers of
electron dense material called plaques [154, 195, reviewed by 196]. The outer plaque
faces the cytoplasm and is involved in the nucleation of cytoplasmic or astral
microtubules. The inner plaque faces the nucleus and nucleates the microtubules in the
nucleus which includes the mitotic spindle. The central plaque spans the nuclear
envelope and is associated with a lateral protrusion along the cytoplasmic face of the
nuclear envelope called the half-bridge [87, 88]. The process of SPB duplication during
the cell cycle is illustrated in Figure 1.4 B. Prior to Start, the bridge elongates and a
relatively small mass of SPB proteins accumulates on the distal end of the half bridge.
This structure is referred to as the satellite and, like the procentriole, is thought to be a
template for the construction of a new SPB [87, 88, 197]. Indeed, at Start, the SPB
40
duplicates at the distal end of the elongated half-bridge and is inserted into the nuclear
envelope [197]. The duplicated SPBs sit side-by-side in the nuclear envelope and are
linked by the bridge structure, now referred to as the full-bridge [87, 88, 197]. During Sphase, the bridge is compromised through an unknown mechanism, and the SPBs
separate to opposite sides of the nucleus forming a short spindle [87]. This process
requires not only B-cyclin activity, but microtubules and the microtubule motor proteins
Cin8 and Kip1 as well [65, 67, 198, 199]. At mitosis, the short spindle becomes attached
to chromosomes by nuclear microtubules such that when the spindle elongates at
anaphase the genome is divided equally between the mother and daughter cell.
Multiple genetic and proteomic screens have identified structural components of
the SPB as well as proteins important for the different steps in SPB duplication. These
proteins, as well as their localization and function, are listed in Table 1.2. As described
above, CDKs have been implicated in multiple steps of SPB duplication. G1-cyclin-CDK
activity is important for SPB duplication [87-89], while B-cyclins are thought to be
important for SPB separation [65, 66]. The finding that mitotic B-cyclins inhibit SPB
reduplication suggests that negative regulation of SPB duplication has an important role
during the cell cycle: SPB reduplication is prevented by a licensing mechanism
analogous to DNA replication licensing [67]. What is this licensing mechanism? How is
it regulated by B-cyclin-CDK activity? Are there additional mechanisms that regulate
41
Table 1.2: Genes important for SPB structure and duplication
42
gene
localization
essential?
functional role
references
TUB4
SPC98
SPC97
SPC72
NUD1
CNM67
SPC42
SPC29
CMD1
SPC110
NDC1
MPS2
BBP1
KAR1
MPS3
CDC31
SFI1
MPS1
γ-tubulin complex, inner and outer plaque
γ-tubulin complex, inner and outer plaque
γ-tubulin complex, inner and outer plaque
outer plaque, half bridge
outer plaque, satellite
outer plaque, satellite
central plaque, satellite
central plaque, satellite
central plaque, satellite
inner plaque
nuclear envelope
nuclear envelope
nuclear envelope
half-bridge
half-bridge
half-bridge
half-bridge
undefined
yes
yes
yes
no
yes
no
yes
yes
yes
yes
yes
no
yes
yes
yes
yes
yes
yes
microtubule nucleation
microtubule nucleation
microtubule nucleation
γ-tubulin binding
structural, MEN signaling
structural
structural, SPB duplication
structural, SPB duplication
structural
structural, γ-tubulin binding
SPB insertion
SPB insertion
SPB insertion
structural, SPB duplication
structural, SPB duplication
structural, SPB duplication
structural, SPB duplication
SPB duplication, spindle checkpoint
[197, 200]
[201, 202]
[202, 203]
[197, 204]
[197, 205]
[197, 206]
[197, 207]
[197, 208]
[209, 210]
[197, 211]
[212]
[213]
[214]
[215]
[216]
[217]
[218]
[213]
SPB duplication during the cell cycle? These questions will be investigated in the course
of this dissertation.
1.10 Concluding remarks
In a review published in Cell in 1988, William Dunphy and John Newport wrote,
“It is now taken for granted that these key events of the cell cycle are under the control
of a network of regulatory factors. The focus of cell cycle research has been the
identification of these factors and the elucidation of how they direct the transitions of
the cell cycle [219].” Over 20 years later, cell-cycle research still revolves around these
two directives, and both of these approaches will be utilized in this dissertation to
investigate the roles of CDKs in the cell cycle.
1.10.1 Dissertation outline
This dissertation is composed of 6 chapters. In chapter 2 I will investigate the
role of CDK in cell-cycle oscillations and present a new model for regulation of the cellcycle in which oscillations are independent of CDK activity. I will examine different
facets of this model in each chapter that follows. Chapter 3 details additional
experiments and ongoing studies into the nature of the CDK-independent oscillator and
its relationship to other cell-cycle regulators. In Chapter 4, I will examine CDKdependent and independent mechanisms regulating SPB duplication. Lastly, in chapter
5, I will investigate the role of periodic transcription in the regulation of SPB duplication.
In Chapter 6, I will summarize and discuss the findings presented in Chapters 2 though 5.
43
Chapter 2
The role of CDKs in cell-cycle oscillations
The cell cycle is a series of ordered events that culminates in a single cell dividing
into two daughter cells. Both the essential events of the cell cycle and cell-cycle
oscillations are thought to be regulated by cyclin-dependent kinase (CDK) activity
[reviewed in 220]. It has been proposed that a transcription-factor (TF) network
oscillator controls periodic events in the absence of S-phase and mitotic CDK activity and
may function as the underlying cell-cycle oscillator in wild-type cells [68, 99]. However,
G-cyclins remain active in these cells, and it remains possible that they have a critical
role in regulating the remaining periodic events. In this chapter, I will examine whether
G1-CDK activity is essential for cell-cycle oscillations. Additionally, I will present a new
model for the role of CDK activity in the regulation of these oscillations.
2.1 Introduction
In order to divide, cells must properly execute the sequence of duplication and
segregation events that make up the cell cycle. Repeated cycles of cell division can
generate exponential growth in cell number; a process essential for early embryogenesis
in multi-cellular organisms. The rapid cycles of cell division in early embryos are
believed to be dependent on oscillations in cyclin-CDK activity [16, 35, 36]. In these
44
cells, cyclin is constitutively synthesized from stores of maternal mRNA, allowing cyclinCDK activity to build throughout interphase [35]. When cyclin-CDK activity builds to
critical levels, it triggers the events of mitosis and the degradation of cyclin protein [36].
This cycle of cyclin synthesis and destruction composes a self-limiting biochemical
oscillator responsible for embryonic cell-cycle oscillations.
The discovery that CDKs trigger the bulk of cell-cycle events in yeast [21, 143],
suggested that oscillations in cyclin-CDK activity constitute the underlying cell-cycle
oscillator in all eukaryotic cells. However, this widely-accepted model has a major flaw.
In embryonic systems, cyclin is constantly synthesized from a pool of maternal mRNA,
however in yeast and somatic cells, cyclin synthesis is regulated transcriptionally
[reviewed in 78, 221]. Thus, in yeast and somatic cells, cyclin oscillations are not
autonomous. Although much is understood about transcriptional regulation of cyclins,
the role of transcription in cell-cycle oscillations remains unclear.
Previous studies have suggested that some CDK activities may not be essential
for cell cycle oscillations in budding yeast [68, 99]. In the absence of the yeast B-cyclin
homologues important for S-phase and mitosis (Δclb1-6), G1 events repeat at cell-cycle
intervals, even though cell-cycle progression arrests at the G1/S transition [68].
Recently, global transcriptional oscillations have been shown to occur in these cells,
suggesting that S-phase and mitotic cyclins are not required to maintain oscillations in
transcription [99]. It was proposed that a transcription factor (TF) network oscillator is
45
important for maintaining periodic transcription, and that during the normal cell cycle,
coupled cyclin-CDK and TF-network oscillators cooperate to produce robust cell-cycle
oscillations [99].
The TF-network oscillator proposed to maintain transcriptional oscillations in
cells lacking S-phase and mitotic cyclins contains eight well-characterized transcriptional
regulators, and the G1 cyclin, Cln3 [99]. The presence of Cln3 in the TF-network
oscillator suggests that G1 cyclin-CDK activity is important for oscillations in the absence
of S-phase and mitotic B-cyclins. As G1 cyclins have been shown to regulate both their
own transcription and their own proteolysis, they could, in principle, compose a selflimiting biochemical oscillator [83, 84, 222, 223]. It is possible that G1-cyclin oscillations
are important for cell-cycle oscillations in the absence of S-phase and mitotic cyclins.
To test the model that G1-cyclins are important for transcriptional oscillations,
and to examine the role of CDK activity in cell-cycle oscillations, we measured global
mRNA dynamics over time in two different conditions in which budding yeast lack all
cyclin-CDK activity. We find that a subset of periodically transcribed genes in wild-type
cells maintain periodicity in the absence of all CDK activity. In addition, we elaborate a
new model for cell-cycle regulation in which a TF network oscillator is important for
both maintaining cell-cycle oscillations and for promoting the expression of cyclins and
other periodically expressed genes. In turn, CDK activity has important roles in
promoting cell-cycle events and feeds back upon the TF-network oscillator.
46
2.2 Oscillations in G1-cyclin are not necessary for B-cyclin
independent oscillations.
In yeast cells lacking the six B-type cyclins essential for of S-phase and mitosis,
oscillations in both bud formation and periodic transcription occurs on schedule [68,
99]. While these cells lack the cyclin-CDK activity necessary for S-phase or mitosis, they
still express G1-cyclins. G1 cyclins are known to be important for both bud formation
and regulation of G1 transcription [62, 84, 223, 224]. In addition, G1 cyclins are also
involved in triggering their own proteolysis [222], suggesting that G1 cyclins could
produce oscillations in the absence of B-cyclins. To test this hypothesis, we examined Bcyclin independent oscillations in cells over-expressing a hyper-stable allele of CLN2.
This allele, CLN2-4t3s, lacks seven CDK phosphorylation sites. Phosphorylation of these
sites by G1-cyclin-CDK activity targets the Cln2 protein for ubiquitination by the SCF
ubiquitin ligase [222]. Thus, over-expression of this allele produces constant G1-cyclinCDK activity throughout the cell cycle [222].
Periodic bud formation was measured in a synchronized population of cells both
expressing Cln24t3s and lacking B-cyclin activity though over-expression the hyperstable B-cyclin-specific inhibitor, Sic1Δ3P, and compared to periodic bud formation in
cells expressing only Sic1Δ3P (Figure 2.1). Both cells expressing and not expressing
47
Figure 2.1: Oscillations in Cln2 are not essential for B-cyclin-independent oscillations.
Cells were syncronized in G1 and released in the presence of galactose to induce
expression of Sic1Δ3p and Cln2-4t3s from the GAL1 promoter. Cells expressing Cln24t3s, red; cells not expressing Cln24t3s, blue. (A and B) Bud formation was measured
over time and averaged over three experiments (A). (C and D) Budding curves were fit
with the CLOCCS mathematical model which learns both the average cell cycle time (C)
and average time to enter the cell cycle after syncroniztion (recovery time; D) for the
population [99, 225]. Values shown are the average of three experiments, and the error
bars represent the standard deviation across these experiments.
Cln2-4t3s periodically formed buds in the absence of B-cyclin activity [68] (Figure 2.1 A
and B). As expression of Cln2-4t3s does appear to affect budding cycles in these cells,
we fit population dynamics with the CLOCCS (Characterizing Loss of Cell-Cycle
48
Synchrony) model to learn the average cycle period length and recovery time, which is
the average amount of time it takes a cell in the population to enter the cell cycle after
synchronization [99, 225] (Figure 2.1 C and D). The average cell cycle time in cells overexpressing Cln2-4t3s is slightly longer than in control cells (Figure 2.1 C), however,
second bud formation in both populations occurs at a similar time (Figure 2.1 A),
suggesting that while over-expressing Cln2-4t3s can accelerate cell-cycle entry (Figure
2.1 A and D), it does not considerably affect B-cyclin-independent oscillations. These
results suggest that oscillations in G1-cyclin activity are not required for B-cyclinindependent oscillations.
2.3 Transcriptional dynamics in α-factor treated cells.
As oscillations in G1-cyclins are not required for B-cyclin-independent
oscillations, we wanted to ask whether G1-cyclin-CDK activity is necessary for
transcriptional oscillations. One condition in which yeast cells lack all CDK activity is
during cell-cycle arrest in response to α-factor. When mating-type a yeast cells are in
the presence of mating pheromone from mating-type α cells (α-factor), they arrest cellcycle progression in G1. Under these conditions, G1-cyclin-CDK activity is inhibited by
the G1-specific CDK-inhibitor, Far1 [147]. In the absence of G1-cyclin-CDK activity, the
B-cyclin-CDK inhibitor, Sic1 and the APCCdh1 should remain active preventing B-cyclinCDK activity [94-96].
49
Figure 2.2: Experimental controls for α-factor treatment experiments. Δcln1-3 cells
were synchronized in early G1 and treated with 100ng/ml α-factor. (A,B) Formation of
the first (solid blue line) and the second (broken blue line) mating projection were
monitored over time in α-factor treated cells. Representative images are shown (C).
Bud formation was also monitored in a reserved fraction of the population not treated
with α-factor (red line). Global mRNA dynamics were measured by oligonucleotide
array for two replicates (A and B). (D) Mean expression values were plotted as absolute
expression/1000 for replicate 1 and 2. Each point was colored according to the number
of points in a square with length 500 centered on that point. Data was fitted with a
linear model (grey line) and the associated r2 value is shown.
50
In the presence of α-factor, cells also polarize growth to form a mating
projection through a process that is similar to budding [226]. If cells do not successfully
conjugate with mating-type α cells, they will form additional mating projections over
time [148]. A population of mating-type a cells treated with α-factor will initiate rounds
of mating projection synchronously with a period similar to that of budding cycles in the
absence of B-cyclins [68]. Thus, periodic mating-projection formation was proposed to
be regulated by the same CDK-independent oscillator that has been proposed to drive
periodic events in the absence of B-cyclins and wild-type cells [68, 99].
If the CDK-independent oscillator is based on the dynamics of a TF-network, then
we should be able to observe transcriptional oscillations in these cells. To test this
hypothesis, we measured global mRNA dynamics over time in a synchronized population
of Δcln1-3 cells treated with α-factor. Cells lacking all three G1-cyclins were used to
ensure that all cyclin-CDK activity was absent during the experiment. Cells were
synchronized in early G1 by centrifugal elutriation and treated with 100ng/ml (59.4 nM)
α-factor. Aliquots were harvested at 10-20 minute intervals for 300 or more minutes.
Mating projection formation in α-factor treated cells and bud formation in untreated
cells was monitored over time (Figure 2.2, A-C). Global mRNA levels at each time-point
were measured with Affymetrix Yeast 2.0 oligonucleotide arrays for cells treated with αfactor. Results from two independent replicates were highly reproducible, with a r2 of
0.991 (Figure 2.2 D).
51
2.3.1 α-factor specific transcriptional regulation
Exposure of mating type a cells to α-factor triggers a mitogen-activated protein
kinase (MAPK) cascade that leads to cell-cycle arrest through the activation of the G1cyclin-CDK inhibitor, Far1, and activates an α-factor specific transcriptional program
thought to be mediated by the TF, Ste12 [147, 227, 228]. The affect of α-factor
treatment on global transcription had been measured previously using two-color
spotted arrays by Roberts et al. [229]. This study reports 133 genes that show two-fold
increases in expression and 103 genes that show two-fold decreases in expression in
asynchronous cells treated with 50nM α-factor for 30 min [229].
In our experiment, we measuring absolute expression in a synchronized
population of cells over time, which allows for observation of the dynamics of these
transcripts, providing additional information about the transcriptional response to αfactor treatment (Figure 2.3 A and B). To better compare our data to the data from
Roberts et al., we calculated fold-change between mRNA the average α-factor mRNA
level across the entire time course and compared it to average wild-type mRNA levels
[99]. About half of the genes previously found to change two-fold up or down are
reproduced in our dataset, although we find many more genes that are up- and downregulated in response to α-factor treatment (Figure 2.3 C and D). This difference
between our data and the data of Roberts et al., likely reflects the fact that our mean
expression levels reflect average expression over an extended period of α-factor
52
Figure 2.3: α-factor regulated expression. (A and B) The expression of genes over time
in α-factor treated cells is shown for genes reported to be two-fold over-expressed (A)
or two-fold under-expressed (B) in α-factor treated cells in Roberts et al. [229].
Transcript levels are expressed as log2-fold change versus mean expression and genes
are ordered by decreasing (A) or increasing (B) mean expression in our dataset. (C and
D) Fold-changes in mRNA levels were calculated for average expression across time in
α-factor treated cells versus wild-type cells [99]. Genes that are two-fold or more overexpressed (C) or under-expressed (D) were compared to the respective gene lists from
Roberts et al. [229] and Venn diagrams are shown.
53
treatment while measurements from Roberts et al. reflect a single, early time point. Of
the 101 genes reported to be down-regulated two-fold or more in Roberts et al., 91%
had been previously shown to be periodically expressed during the cell cycle [102, 229],
suggesting that α-factor arrest represses periodic transcription. In contrast, we find that
only 27% of the genes that are two-fold or more under-expressed in our α-factor
dataset were shown to be periodically expressed in cycling cells, using the periodic gene
list from Spellman et al. [102] used in Roberts et al. [229]. These results suggest that
periodic transcription may not be globally repressed in α-factor treated cells over longer
periods of time.
2.3.2 Periodic transcription in α-factor treated cells.
In order to ask if periodic gene expression is maintained in the absence of CDK
activity, mating projection formation was aligned to a cell-cycle timeline using an
adaptation of the CLOCCS model to allow for direct comparison to cycling cells [99, 225]
(see section 2.9.7). Of the 1271 periodic genes in wild-type, 196 show strong correlation
(Pearson coefficient ≥ 0.75) between the expression patterns in α-factor treated cells
and wild-type cells. These 196 genes maintain periodicity in α-factor treated cells and a
shown in Figure 2.4.
In Figure 2.4, we show evidence that periodic transcription is maintained in the
absence of CDK activity in cells treated with α-factor. Although periodic transcription in
54
Figure 2.4: Periodic expression persists in α-factor treated cells. Genes that remain
periodic in α-factor arrested cells were determined by aligning data to a cell-cycle
timeline using an adaptation of the CLOCCS model [99, 225]. Genes that correlate with
wild-type (Pearson coefficient ≥ 0.75) are shown in identical order in α-factor treated
cells (A) and wild-type cells (B). Shading of the cell-cycle timeline indicates lack of cellcycle progression through that phase. Transcript levels are expressed as log2-fold
change versus mean expression.
α-factor treated cells might be regulated by a TF-network oscillator, as we have seen in
other conditions [99], it is possible that the oscillations are governed by a extrinsic
oscillator. The MAPK activity involved in the α-factor signaling pathway has also been
shown to oscillate during α-factor treatment [230]. These oscillations have been
attributed to a biochemical oscillator summarized in Figure 2.5 A. Briefly, α-factor
triggers a MAPK cascade leading to the phosphorylation and activation of the Fus3
55
Figure 2.5: Oscillations in α-factor are not dependent on oscillations of the negative
regulators SST2 and MSG5. (A) Representation of the biochemical oscillator proposed to
be responsible for oscillations in α-factor arrested cells [230]. (C,E) Expression of SST2
and MSG5 are shown over time in α-factor arrested cells in both replicates (dark and
light blue). (A,D,F) Wild-type, Δsst2, and Δmsg5 cells were arrested with 100ng/ml αfactor for 4 hours, representative images are shown. Arrowheads indicate cells with
multiple mating projections.
56
MAPK. Active Fus3 activates the Ste12 transcription factor which drives expression of
many target genes. Among these genes are two negative regulators of the MAPK
cascade, SST2 and MSG5. Feedback from these two genes is suggested to reset the
system causing oscillations in Fus3 activity and mating projection formation [230].
One prediction of this model for oscillations during α-factor treatment is that
transcription of SST2 and MSG5 would oscillate over time. However, neither SST2 nor
MSG5 appear to oscillate in our α-factor experiments (Figure 2.5 C and E). Additionally,
although SST2 and MSG5 have been shown to be essential for oscillations in mating
projection formation [230], Δmsg5 cells form multiple mating projections similar to
those formed by wild-type cells when treated with α-factor (Figure 2.5 F). Δsst2 cells
have a defect in mating projection formation in response to α-factor, although Δsst2
cells that do form mating projections can form multiple mating projections (Figure 2.5
D).
The data in Figure 2.5 suggest that oscillations in MAPK activity may not drive
oscillations in α-factor arrested cells as previously reported [230], however, we cannot
completely rule out that another mechanism related to α-factor signaling causes the
transcriptional oscillations we observe. 2.4 Periodic transcription in cells lacking
functional CDK
57
2.4 Periodic transcription in cells lacking functional CDK
In order to test whether periodic transcription is independent of CDK activity in
the absence of pheromone treatment, we used a different strategy for elimination of all
CDK activities from cells. We measured periodic transcription in cells with a
temperature-sensitive mutation in the yeast CDK, cdc28-4 [56]. At the restrictive
temperature, cdc28-4 cells arrest cell-cycle progression in G1, with no evidence of
periodic events (Figure 2.6). In order to determine whether periodic transcription is
maintained in these cells, we measured global mRNA dynamics over time in cdc28-4
cells at the restrictive temperature. Cells were synchronized in early G1 by centrifugal
elutriation and shifted to 37°C. Aliquots were harvested at 20 minute intervals for 300
or more minutes. Bud formation was monitored over time for cells at both 37°C and
25°C (Figure 2.6, A-C). Global mRNA levels at each time-point from cells at the
restrictive temperature were measured with Affymetrix Yeast 2.0 oligonucleotide
arrays. Results from two independent replicates were highly reproducible, with a r2 of
0.995 (Figure 2.6 D).
Visual inspection of mRNA dynamics in cdc28-4 cells indicated that a subset of
wild-type periodic genes is still periodically transcribed. In order to determine which
genes oscillate in the cdc28-4 dataset, we needed a value for period length in these
experiments. As we could not measure an external periodic event (such as budding or
58
Figure 2.6: Experimental controls for cdc28-4 experiments. cdc28-4 cells were
synchronized in early G1 and released at the restrictive temperature (37°C). (A,B) Bud
formation was monitored in cdc28-4 cells at the restrictive temperature (blue line) and
the permissive temperature (redline). Representative images are shown (C). Global
mRNA dynamics were measured by oligonucleotide array for two replicates (A and B).
(D) Mean expression values were plotted as absolute expression/1000 for replicate 1
and 2. Each point was colored according to the number of points in a square with
length 500 centered on that point. The data was fitted with a linear model (grey line)
and the associated r2 value is shown.
59
mating projection formation) in cdc28-4 cells at the restrictive temperature, we used
mRNA dynamics from the cdc28-4 data of genes previously found to be periodically
transcribed in wild-type cells [99] to learn the period of transcriptional oscillations (see
section 2.9.6). We aligned mRNA dynamics over time in cdc28-4 cells to wild-type data
Figure 2.7: Periodic expression persists in cdc28-4 cells. The cdc28-4 data set was
aligned to a cell-cycle timeline based on periodic gene expression through an unbiased
clustering method (see section 4.9.6). Genes that remain periodic were determined by
a union of genes in the cluster and genes that correlate with wild-type (Pearson
coefficient ≥ 0.75). These genes are shown in identical order in cdc28-4 cells (A), wildtype cells (B) and B-cyclin mutants (C). Shading of the cell-cycle timeline indicates lack
of cell-cycle progression through that phase. Transcript levels are expressed as log2fold change versus mean expression.
60
from Orlando et al. [99] using a range of off-set values (µ0) and a range of period lengths
(λ) [225]. The correlation between each genes expression pattern in cdc28-4 cells and
wild-type was calculated for each pair of µ0 and λ values. We then used an unbiased
expectation maximization algorithm to cluster genes based on the values of µ0 and λ in
which they correlate with wild-type well. From this analysis, we derived values for µ0
and λ from the cluster centroid and also learned a set of genes that have a high
probability (0.6) of cluster membership. We combined that gene list with the genes that
correlate with wild-type (Pearson co-efficient 0.75) when aligned using the learned
values for µ0 and λ, and filtered the resulting gene list manually to remove visuallyidentifiable false-positives. The remaining 173 genes maintain periodicity in the
absence of CDK-activity and are shown in Figure 2.7.
2.5 TF-networks may maintain oscillations in the absence of
CDK activity.
As periodic transcription persists in both α-factor treated and cdc28-4 cells, the
most parsimonious model to explain these data is that a TF-network oscillator is
responsible for both driving periodic transcription and maintaining oscillations in these
cells. To support this model, we first constructed a TF network from periodic TFs in
wild-type cells (Figure 2.7). Nodes were selected as TFs that are periodically expressed
in wild-type cells with two confidence levels. The highest confidence nodes were above
61
the cut-off used in Orlando et al. to establish a list of periodic genes [99]. The lower
confidence nodes show periodic behavior in wild-type cells but were not identified as
periodic genes [99]. All nodes were placed on a cell-cycle time line based on their time
of peak expression in wild-type cells. Nodes were connected by edges representing
known interactions between a TF and the promoter of a downstream TF or known
regulatory physical interactions between TFs [74, 77, 79, 100, 149, 231-237].
Interactions defined by targeted studies or that were found through large scale analyses
with high significance (p≤0.001) are classified as high confidence edges, while
interactions defined by genetic studies or large scale analyses with lower significance
(0.051≥p≥0.001) are classified as lower confidence edges. The resulting TF-network is
densely populated with TFs throughout the cell cycle and is highly connected and
cyclical in nature (Figure 2.8 A).
To determine if a TF-network could maintain oscillations in the absence of CDK
activity, we constructed networks of TFs that maintain periodicity in α-factor treated
cells (Figure 2.8 B) or in cdc28-4 cells (Figure 2.8 C). For the α-factor network, high
confidence nodes were strongly correlated with wild-type (Pearson correlation
coefficient ≥ 0.75). Lower confidence nodes were less highly correlated with wild-type
(Pearson correlation coefficient ≥ 0.5). Finally, we also included TFs that appear to be
periodically expressed in α-factor treated cells although they are not included in the
62
Figure 2.8: Transcription factor networks may produce CDK-independent transcriptional
oscillations. Nodes were placed on a cell-cycle time-line based on peak expression in
wild-type cells; activators, red; repressors, green. Edges were drawn based on evidence
for TF-promoter interactions in the literature, black edges represent high-confidence
interactions, grey edges represent lower confidence edges, broken edges represent
protein-protein interactions. (A) TF-network in wild-type cells. Black-bordered nodes
were shown to be periodic in [99], grey-bordered nodes display periodic expression, but
were not included in the high-confidence list. (B) TF-network in α-factor-treated cells.
(C) TF-network in cdc28-4 cells. Black-bordered nodes are high confidence, dark greybordered nodes are lower confidence, and light grey-bordered nodes are the lowest
confidence nodes included.
63
Figure 2.8
64
statistically defined sets. We found evidence that 10 of the 21 TFs or TF complexes that
comprise the wild-type TF-network maintain periodicity in α-factor treated cells. For the
cdc28-4 network, high confidence nodes were strongly correlated with wild-type
(Pearson correlation coefficient ≥ 0.75) or had high probability of cluster membership in
our expectation maximization analysis (≥0.6). Lower confidence nodes were less highly
correlated with wild-type (Pearson correlation coefficient ≥ 0.5) or had a lower
probability of cluster membership in our expectation maximization analysis (≥0.2).
Finally, we also included TFs that appear to be periodically expressed in cdc28-4 cells
although they are not included in the statistically defined sets. We found evidence that
13 of the 21 TFs or TF complexes that comprise the wild-type TF-network maintain
periodicity in cdc28-4 cells.
The resulting TF-networks for α-factor treated and cdc28-4 cells are coherent
and cyclical in nature (Figure 2.8, B and C). Both TF-networks are sub-sets of the wildtype TF-network and they partially overlap. These new TF-networks are similar in
structure to the TF-network oscillator proposed to maintain oscillations in the absence
of B-cyclins [99]. Like the TF network from Orlando et al., the TF-networks shown in
Figure 2.8 are likely to behave as transmission oscillators which are formed through
stable transmission of a wave of activity around loop [99, 238]. We propose that
oscillations generated by the TF-networks in Figure 2.8 drive both oscillations in the
absence of CDK activity and during the wild-type cell cycle.
65
2.6 What is the role of CDKs in transcriptional oscillations?
We show that in two different conditions in which yeast cells lack CDK activity,
periodic transcription persists and propose that TF-networks maintain oscillations in
these cells. These data suggest that a TF-network oscillator rather than a CDK oscillator
constitutes the fundamental cell-cycle oscillator in yeast cells. As cells lacking CDK
activity do not engage in events necessary for cell division, CDKs clearly have an
important role in the cell-cycle. In order to ask whether CDKs also play a role in
controlling transcriptional oscillations, we compared oscillation data from experiments
with decreasing amounts of CDK activity (Figure 2.9). From a comparison of budding
curves from wild-type and B-cyclin mutants to mating projection formation curves in αfactor arrested cells, it is clear that the average period-length increases with decreasing
CDK activity (Figure 2.9 A and B). Closer inspection reveals that much of the change in
period is due to reduced population synchrony in cells lacking some or all CDK activity.
If we measure period length from the percent one budded (or mating projection) curve
to percent two budded curve over a range of percentages, we find that the period
increases approximately linearly with the percent at which these curves are compared
(Figure 2.9 C). The slope of these lines increases with decreasing CDK-activity (Figure 2.9
D), indicating that the less CDK-activity present in a population of cells, the more rapidly
they lose synchrony. Accordingly, the variance in velocity of cell-cycle progression
66
Figure 2.9: CDK activity adds robustness to cell-cycle oscillations. Wild-type, blue; Bcyclin mutants, red, α-factor, green and cdc28-4, purple. (A) Budding curves for wildtype cells and B-cyclin mutants as well as mating projection curves for α-factor arrested
cells were averaged between two replicates and aligned at along the left edge of the
curve such that the x-axis represents time after first bud or mating projection
emergence. Solid line, percent cells in the first budding/mating projection formation
cycle; broken line, percent cells in second budding/mating projection formation cycle.
For wild-type cells the first and second budding cycle were broken apart using mother
cell cohorts from the CLOCCS model [225],. (B) Average cell-cycle time learned by the
CLOCCS model for wild-type, B-cyclin mutant cells, and α-factor arrested cells [99, 225];
for cdc28-4 cells average cell cycle time is determined by cluster centroid in the
expectation maximization analysis. Each cycle time is an average of two replicates with
error bars representing standard deviation between the replicates. (C) Measurement of
cycle time at different y-values in (A). (D) Average slope for the best-fit line of the
measurements in (C) plotted on a log10 scale. (E) Variance in velocity as learned by the
CLOCCS model for wild-type cells, B-cyclin mutants and α-factor arrested cells shown as
an average of two replicates with error bars representing the standard deviation
between replicates. (F and G) gene expression curves in wild-type cells, B-cyclin
mutants, and cdc28-4 cells for GOT1 and POL32, respectively. Data was averaged
between two replicates and curves were aligned at along the left edge of the curve such
that the x-axis represents time after earliest expression for each condition.
67
Figure 2.9
68
within the populations (learned by the CLOCCS algorithm) also increases with decreased
CDK activity.
Although lack of periodic events prevents such analyses of cdc28-4 data, we do
see that the oscillatory period we learned from the mRNA data is longer than the period
learned through periodic cell-cycle events in other strains (Figure 2.9 B). In addition,
comparing gene expression curves for periodic genes suggests that the period is longer
in cdc28-4 cells than wild-type cells or B-cyclin mutants and shallowness of the cdc28-4
curves suggest that there may be more convolution due to population asynchrony in
cdc28-4 cells (Figure 2.9 F and G).
Together, we see evidence that loss of CDK activity leads to increased oscillatory
periods and that this is due, at least in part, to increased population asynchrony.
Greater population asynchrony indicates that individual cells are oscillating at different
rates, suggesting that there is considerable noise in CDK-independent oscillations.
Overall, we see an increase in noise with decreased CDK-activity indicating that CDKs
are, in fact, important for transcriptional oscillations, although they are not essential.
CDK activity appears to confer robustness and refinement to cell-cycle oscillations. It is
likely that CDK promotes robust oscillations by modulating of the TF-network through TF
phosphorylation.
These data suggest a new model to describe the roles of a TF-network oscillator
and CDK activity in promoting both cell-cycle events and cell-cycle oscillations during the
69
Figure 2.10: Model for cell-cycle regulation. Cell-cycle oscillations are maintained by a
TF-network oscillator that produces oscillations in CDK-activity by promoting periodic
transcription of cyclin genes. CDK-activity feeds back on the TF-network promoting
robustness in TF-network oscillations. CDK-activity is also important to regulate the
essential events of the cell-division cycle. The TF-network outputs the periodic
expression of many genes, while CDK-activity is essential for the periodic transcription
of other genes. The products of many periodic genes are important for cell-cycle
events.
normal cell cycle (Figure 2.10). In this model, oscillations are generated by a TF-network
oscillator. The TF-network also induces periodic transcription of at least 20% of the
yeast genome [99], including cyclin genes. Periodic expression of these cyclin genes
directs periodic waves of cyclin-CDK activity that is essential for cell-cycle events such as
budding, DNA-replication, spindle formation and mitosis [61, 63, 65, 66]. CDK-activity
also feeds back on the TF-network oscillator, modulating its activity to produce robust
cell-cycle oscillations and coordinate these oscillations with cell-cycle progression. In
70
addition, the products of several periodic genes are known to be important for cell-cycle
events. In Chapter 5, I will show evidence that periodic expression has a cyclinindependent role in the regulation of cell-cycle events.
2.7 Discussion
Self-sustaining oscillations in cyclin-CDK activity are known to drive cell-cycle
oscillations in early embryonic cells when cyclin is constitutively synthesized from
maternal stores of mRNA. Although cyclin expression is dependent on transcription in
somatic cells and yeast, it is still widely accepted that an autonomous cyclin-CDK
oscillator drives cell-cycle oscillations in these cells. In this chapter, I show that cellcycle oscillations still occur in the absence of CDK activity under two different
conditions. I propose that a TF-network oscillator rather than a cyclin-CDK oscillator is
the fundamental cell-cycle oscillator and that CDKs have important roles in enforcing
robustness upon TF-network oscillations and in regulating cell-cycle events.
2.7.1 A new model for cell-cycle oscillations
In this chapter I proposed a new model for cell-cycle oscillations in yeast. In this
model, CDKs do not form the fundamental cell-cycle oscillator, as oscillations persist in
the absence of all CDK activity. In Figures 2.4 and 2.7, I show that in both Δcln1-3 cells
treated with α-factor and in cdc28-4 cells, many genes continue to be properly
periodically transcribed. I propose that a TF-network made up of periodically
transcribed TFs maintains oscillations in the absence of CDK activity and that a TF
71
network may therefore be the fundamental oscillator that produces cell-cycle
oscillations.
This new model challenges the belief that CDKs form the core of the
fundamental cell-cycle oscillator, however, this model can be reconciled with previous
models of cell-cycle oscillations. In the model in Figure 2.10, CDKs do have an important
role in cell-cycle oscillations. I show in Figure 2.9 that loss of CDK activity leads to
increased noise in cell-cycle oscillations, suggesting that the role of CDKs in cell-cycle
oscillations is to confer robustness to the TF-network oscillator. It known that CDKs can
phosphorylate 13 of the 24 TFs in the TF network oscillator shown in Figure 2.8 A [73,
74, 149, 239, 240], indicating that CDKs do feedback on the TF-network oscillator. In
addition, as cells cannot trigger essential cell-cycle events such as budding, DNA
replication or mitosis in the absence of CDK activity, CDKs are essential for cell-cycle
events and cell-cycle progression [61, 63, 65, 66].
Data supporting the model in Figure 2.10 comes from experiments examining
cell-cycle events in the presence of different cyclin-CDK activities. In the absence of all
CDK activity, cells do not complete any cell cycle events [56, 241]. However, adding
back cyclins restores periodic cell-cycle events. In cells lacking S-phase and mitotic Bcyclins but containing G1 cyclins (Δclb1-6), cells bud at cell cycle intervals [68]. Adding
back the unstable S-phase cyclin Clb6 (Δclb1-5) [98], produces additional oscillations in
both spindle formation [67] and DNA replication [S. Haase, unpublished data]. These
72
data suggest that CDKs are effectors of the CDK-independent oscillator that is
responsible for the regulation of cell-cycle events, which is consistent with the model in
Figure 2.10.
2.7.2 Are cell-cycle events entrained to CDK-independent oscillations?
Circadian oscillators are responsible for generating the twenty-four hour
oscillations in the behavior of many organisms. Normally, this oscillator is entrained to
the rotation of the earth through the light-dark cycles and/or heat cycles experienced by
the organism. However, if organisms are maintained in an environment in the absence
of these environmental cues, the oscillator continues to regulate behavior; although
synchrony with earth’s rotation is lost. It has been shown that environmental signals
are able to feedback on components of the oscillator, entraining these inherent
oscillations [reviewed in 160].
If a CDK-independent oscillator is responsible for cell-cycle oscillations, how is it
entrained to cell-cycle progression? In normally cycling cells, CDK-activity is both an
effector of the TF-network oscillator and can feedback on the oscillator. This feedback
likely ensures that CDK-activity and transcriptional oscillations are synchronized.
However, TF-network oscillations can be uncoupled from cell-cycle progression either
genetically or by external perturbation [68, 99].
In the absence of S-phase and mitotic cyclins, oscillations persist in the absence
of cell-cycle progression [68, 99]. In these experiments, oscillations are genetically
73
uncoupled from cell cycle progression. If you restore B-cyclin activity in these cells by
promoting the expression of CLB1 from the GAL1 promoter, viability is low, however,
you get slight peaks in viability once each cell-cycle period [L. Simmons Kovacs, D.
Orlando and S. Haase, unpublished data]. This data suggests that once a cell-cycle,
when cell-cycle progression and network oscillations are aligned, cells can survive;
however, cells cannot entrain cell-cycle progression to CDK-independent oscillations
after they have been genetically uncoupled. However, it is unlikely that cells would
have naturally evolved a mechanism to entrain oscillations in this artificial situation. A
more interesting question concerns how oscillations are entrained to natural
purturbations.
In Figure 2.4, I show that transcriptional oscillations persist in α-factor treated
cells. However, it is routine laboratory procedure to arrest cell-cycle progression in
asynchronous cells by treating with α-factor and then releasing the G1-syncronized
population. These cells rapidly enter the cell-cycle and appear to behave normally. Our
data suggest that cell-cycle oscillations are not fully arrested in α-factor treated cells,
suggesting that a population of α-factor treated cells is not truly synchronized.
Nonetheless, synchronized populations of α-factor treated cells recommence cell-cycle
progression very synchronously [225].
This paradox suggests that an entrainment mechanism functions to promote
cell-cycle entry upon release from α-factor arrest. During α-factor treatments, G1-
74
cyclin-CDK activity is inhibited by the Far1 CDK-inhibitor [147]. Although Far1 inhibits
G1-cyclin-CDK activity, G1-cyclin-CDK activity can also phosphorylate Far1 and promote
its SCFCdc4 mediated ubiquitination [242]. During α-factor treatment FAR1 transcription
is highly up-regulated (four-fold over wild-type mean expression in our dataset) and
Far1 is activated by Fus3 phosphorylation [147]. As a result, Far1 levels dominate G1cyclin-CDK activity and G1-cyclin-CDK activity is repressed. However, upon release from
α-factor treatment, FAR1 expression is no longer induced and Far1 is no longer
phosphorylated. The balance then shifts towards G1-cyclin-CDK activity. This upregulation of G1-cyclin-CDK activity triggers a positive feedback loop that promotes cellcycle entry [83]. It is likely that this positive feedback loop and the inherent bistability
of Start can promote synchronous entry into the cell-cycle even if the TF-network
oscillator is out of phase.
Cell-cycle progression and CDK-activity is known to be coupled to environmental
conditions through checkpoint mechanisms. Evidence suggests that B-cyclin-CDK is
important to couple oscillations to cell-cycle progression [68]. When B-cyclin-CDK
activity is inhibited in DNA-replication or spindle checkpoint-arrested cells, oscillations in
budding and G1 transcription ensue [68]. These results suggest that stable B-cyclin-CDK
activity during checkpoint arrest maintains coordination of TF-network oscillations and
cell-cycle progression by inhibiting oscillations. Further experiments (detailed in
Chapter 3) will address this hypothesis.
75
Together, evidence suggests that CDK-activity not only provides robustness to
CDK-independent oscillations, but also serves to entrain cell-cycle progression to
oscillations. It is likely that the mechanism of this entrainment involves CDK-dependent
phosphorylation of TFs in the TF-network oscillator. It is also possible that constant
entrainment of the TF-network though phosphorylation of TFs is responsible for the
observed increase in oscillatory robustness conferred by CDKs.
2.8 Future Directions
In this chapter, I have proposed a new model for cell-cycle oscillations. This
model raises several fundamental questions about the roles of CDK activity and a TFnetwork oscillator in cell-cycle regulation. Many of these questions are being
addressed. Preliminary results are shown and discussed in Chapter 3. In this section I
will describe future directions that relate directly to the data included in this chapter.
2.8.1 Damped oscillations in the absence of CDK activity?
As mentioned above, CDKs are known to phosphorylate many of the TFs in the
TF-network oscillator. Although we see that transcriptional oscillations can occur
independently of CDK activity, we see increased noise at the population level in cells
undergoing CDK-independent oscillations (Figure 2.9). These data suggest that CDKs
have a role in exerting robustness to TF-network oscillations; however, we do not fully
understand the relationship between oscillations in the presence and absence of CDKs.
76
Systems level analysis of another oscillating cell-cycle system shows that loss of
positive feedback produces damped oscillations, which are characteristic of the effect of
friction on a physically oscillating system [166]. CDK-activity is known to positively
feedback on the TF-network in several places including Start [73, 74] and entry into
mitosis [78, 104]. Thus, as we incrementally removed cyclin-CDK activity, we also
removed positive feedback on the system. Due to the fact that we collected data on a
population level, we cannot directly observe classic patterns of damped oscillations.
Two different approaches will be used to ascertain whether the TF-network produces
damped oscillations in the absence of CDK activity. First, mathematical models can be
used to determine whether a population of cells undergoing damped TF-network
oscillations could produce the patterns we see in Figure 2.9. Second, oscillations in
single transcripts can be measured on a single cell level using unstable fluorescent
proteins regulated transcriptionally by the TF-network oscillator. Together, these data
could determine whether CDK-independent oscillations are damped and if so, would
suggest that CDK-regulated positive feedback has a role in reducing frictional forces in
cell-cycle oscillations.
2.8.2 Roles of B-cyclins in enforcing robustness to CDK-independent
oscillations
Data in Figure 2.9 compares oscillations in cells with decreasing CDK-activity. As
deletion of multiple B-cyclins often produces inviable cells that form multiple buds
(Table 1.1), we can examine the contribution of each B-cyclin to robust CDK77
independent oscillations. Different B-cyclin mutants will be synchronized and budding
cycles will be measured over time. This data will be analyzed in a similar fashion as the
data in Figure 2.9. I expect we will find differences among B-cyclin-CDK activities in their
ability to add robustness to CDK-independent oscillations, but that the trend that more
CDK-activity produces more robust oscillations will remain.
2.8.3 Transcriptional response to α-factor treatment
As α-factor signaling utilizes pathways that are conserved in metazoans, there is
a considerable interest in understanding the cellular response to α-factor treatment
[reviewed in 243]. The α-factor treatment dataset presented in this chapter is the only
dataset, to my knowledge, to measure the transcriptional response to α-factor
treatment over time in cells synchronously arresting cell-cycle progression (see Figure
2.4). This data can be used to examine the temporal transcriptional response to αfactor treatments. Evidence suggests that a temporally activated TF-network is
activated in α-factor treated cells. In this network the TF, Ste12 activates numerous
target genes including the TF, Kar4, which has roles in activating additional target genes
[244]. Our data supports these conclusions; however, the complexity of the data
suggests that additional TFs may be involved in the transcriptional response. In
combination with TF-localization experiments in α-factor treated cells, we will dissect
the transcriptional response to α-factor treatments and construct a temporal TFnetwork that explains the observed patterns.
78
2.9 Experimental Details
2.9.1 Plasmid and strain construction
All strains used in this study are derivatives of BF264-15DU (MATa; ade1; his2;
leu2-3,112; trp1-1; ura3∆ns) [66]. Relevant genotypes of strains are detailed in Table
2.1. SBY1558 was made by crossing 15Da- to SBY348 [68] and sporulating. SBY1563 was
made by cutting pRS416-GAL-CLN2-4t3s with StuI and integrating into 15Da- at the
URA3 locus. SBY1565 and SBY1566 were made by transforming SBY1558 and SBY1563
with MscI-linearized pRS306-HIS2-SIC1Δ3P, which was made replacing a segment of the
URA3 gene in pRS306-SIC1Δ3P [97] from NcoI to NsiI with the HIS2 gene.
2.9.2 Cell growth and synchronization
Yeast cultures were grown in YEP medium (1% yeast extract, 2% peptone,
0.012% adenine, 0.006% uracil supplemented with 2% sugar (dextrose, sucrose or
galactose). For GAL1p-Sic1Δ3P experiments, cells were grown in YEP containing 2%
sucrose and 0.1% dextrose prior to centrifugal elutriation. The resulting population was
resuspended in YEPG. Δcln1-3; GAL-CLN3 cells were grown in YEP 4% galactose prior to
synchronization by centrifugal elutriation. The resulting, early G1 population, of cells
was resuspended in pre-warmed YEPD containing 100ng/ml α-factor (59.4 nM) or in
YEPD lacking α-factor to monitor Δcln1-3 arrest. Samples were taken at 10-20 minute
79
Table 2.1: Yeast strains used in Chapter 2
Strain
Relevant Genotype
DSY240
MATa; bar1; cln1; cln2x/s; cln3; GAL-CLN3-TRP1
SBY348
MATα; cdc4ts; CLN2-HA3-URA3
SBY371
MATa; cdc28-4
SBY1558
MATa; bar1; CLN2-HA3-URA3
SBY1563
MATa; bar1; GAL-CLN2-4t3s-HA3-URA3
SBY1565
MATa; bar1; GAL-CLN2-4t3s-HA3-URA3; GAL-SIC1Δ3P-HIS2
SBY1566
MATa; bar1; CLN2-HA3-URA3; GAL-SIC1Δ3P-HIS2
intervals until about 75% of cells had two or more mating projections. cdc28-4 cells
were grown at 25°C prior to synchronization by centrifugal elutriation. The resulting
early G1 population was resuspended in YEPD pre-warmed to 37°C. Samples were
taken every 20 minute for 5 or more hours.
2.9.3 Microscopy
All samples analyzed by microscopy were fixed in 2% paraformaldehyde.
Budding or mating projection formation was scored for 200 or more cells per sample.
For images shown, cells were viewed using a Zeiss Axio Imager widefield fluorescence
microscope, a 100x objective and DIC (differential interference contrast) filters. Images
were acquired with a Hamamatsu Orca ER monochrome cooled-CCD camera with IEEE
and captured using Metamorph 7.1 (Universal Imaging).
2.9.4 RNA isolation and microarray analysis
Total RNA was extracted from yeast as described previously [68] and purified
with the RNAeasy MinElute Cleanup Kit (Qiagen, Valencia, CA). cDNA synthesis and
80
fluorescent labeling was done with the GeneChip One cycle Labeling Kit (Affymetrix,
Santa Clara, CA) and hybridized to yeast 2.0 expression arrays (Affymetrix, Santa Clara,
CA). Labeling, hybridization and image collection was performed at the Duke Institute
for Genome Sciences and Policy DNA Microarray Core Facility
(www.genome.duke.edu/cores/microarray). CEL files from each oligonucleotide array
were normalized and summarized using the dChip method as implemented in the affy
package within Bioconductor (www.bioconductor.org) using default parameters [245,
246].
2.9.5 CLOCCS model fitting
Data was fit with the CLOCCS model of population distributions in synchrony
experiments as previously described [99, 225]. For mating projection formation data
from α-factor treatment, the modifications to the CLOCCS model described in [99] was
used. The value of the β, the fraction of the cell cycle before measured event, was set
to 0.075, a number determined by measuring the difference in time to initial mating
projection formation and budding in wild-type cells (data not shown). Lifeline points
were used for cell-cycle timeline mapping and to align experiments [225]. CLOCCS
parameters, especially the average cycle time (λ), recovery time (µ0), and variance in
velocity (σv), were used to compare synchrony experiments [225].
81
2.9.6 Parameter estimation by expectation maximization
Wild-type mRNA levels from Orlando et al. were aligned to a cell-cycle timeline
using CLOCCS, as previously described [99, 225]. For each of the 1274 previously
defined periodic genes, mRNA levels from cdc28-4 cells were aligned to a cell-cycle
timeline over a range of values for the period length (90 min ≤ λ ≤ 288 min) and a range
of values for recovery time (-175 min ≤ µ0 ≤ 38 min). Values of both parameters were in
increments of 3 min. Wild-type mRNA levels were averaged between replicates and
mRNA levels for both wild-type and cdc28-4 cells were fit with a cubic spine consisting
of 25 equally spaced points and log2 transformed. For each gene i, cdc28-4 mRNA levels
(Yi) were compared to wild-type (Xi) at all alignments within the defined range and
Pearson correlation coefficients were calculated for all alignments that overlapped by at
least 5 splines. All negative correlation coefficients were assigned a coefficient of 0.
Correlation values were converted to distances (D) by subtracting the correlation
coefficient from 1. The likelihood (L) of the gene behaving the same in both datasets is
calculated from equation 2.1.
Equation 2.1
An expectation maximization algorithm [247, reviewed in 248], was used to
assign genes to clusters, such that each cluster has a high probability of encompassing
the genes that have a high likelihood of behaving similarly in the two datasets (X, Y) at a
82
centroid (µ0, λ). The algorithm iterates between two steps: the E-step and the M-step.
The E-step estimates probability distributions over cluster assignments given current
values of α (the probability of a cluster), µ0, and λ. Cluster number (k) was set to two,
with one cluster representing genes that behave similarly and the other cluster
encompassing genes with low probability of belonging to any cluster (constant
probability mass corresponding to a gene distance of 0.5). The probability of
membership for gene i in cluster j (δ) is defined in equation 2.2.
Equation 2.2
The M-step maximizes the log-likelihood of the observed data, using the values
of µ0 and λ having the highest likelihood for each gene, µ0* and λ* (Figure 2.11 A). As
we are comparing wave-like data, we adjusted values of µ0* to µ0*-λ* if it falls within
the discretized range of µ0 values (Figure 2.11 B). At iteration t+1 of the algorithm, the
updated parameter values for cluster j are calculated according to Equations 2.3 and
2.4. The probability of each cluster is calculated using Equation 2.5.
Equation 2.3
83
Equation 2.4
Equation 2.5
In order to establish a probability cut-off for cluster membership, the EM
analysis was preformed for a comparison of the Δclb1-6 datasets from Orlando et al.
with wild-type [99]. Cluster membership in the Δclb1-6 dataset was compared to the
established list of genes that remain periodic in these cells from Orlando et al. [99]. At a
probability cut-off of 0.6 for both replicates, the overlap between gene sets was
maximized; however, even at low probabilities the entire gene set was not included in
the cluster (Figure 2.11 C and D).
2.9.7 Correlating expression data
In order to find which genes correlate with wild-type periodic genes in the αfactor and cdc28-4 experiments, each replicate was aligned with wild-type replicate 1
from Orlando et al. as described in sections 2.9.3 and 2.9.4 [99]. Each gene was fit with
a cubic spline and compared across the entire overlapping lifeline region for cdc28-4 or
84
Figure 2.11: Expectation maximization analysis. (A and B) Cluster membership
probability cut-offs were established by comparing cluster membership probability from
expectation maximization analysis of the Δclb1-6 datasets from Orlando et al [99] and
comparison to the list of genes previously shown to maintain periodicity in that dataset.
(C and D) Dot plots of periodic genes in the cdc28-4 replicate 1 data set. Each gene is
plotted at the values of µ0* and λ* before (C) or after (D) µ0*-λ* correction. Gene with
cluster membership probability ≥ 0.6 are shown in dark blue and gene with cluster
membership probability 0.2≥0.6 are shown in light blue. The cluster centroid is shown
in red.
from the beginning of the first cycle to the end of the second S-phase period for α-factor
data. Genes with Pearson correlation coefficients of 0.75 or greater and 0.5 or greater
from either replicate were considered.
85
2.10 Chapter 2 collaborator contributions
The α-factor data was collected in collaboration with Dr. David Orlando who was
invaluable in the analysis of both the α-factor and cdc28-4 datasets. The expectation
maximization algorithm used to align cdc28-4 data was written and implemented by
Michael Mayhew in collaboration with Dr. David Orlando and Dr. Sayan Mukherjee.
86
Chapter 3
Continuing studies into the nature of the CDKindependent transcriptional oscillator
In chapter 2, I presented evidence that a periodic transcription occurs in the
absence of all CDK activity and suggest that a transcription factor (TF) network rather
than a CDK oscillator is responsible for cell-cycle oscillations. That work, together with
the work of Orlando et al., represent a fundamental shift in the understanding of cellcycle regulation [99]. Accordingly, there are a number of interesting questions about
the nature and role of the TF-network oscillator that are beginning to be addressed, two
of which will be discussed in this chapter.
3.1 Is a TF-network oscillator responsible for CDK-independent
transcriptional oscillations?
In chapter 2, I showed that transcriptional oscillations persist in the absence of
all cyclin dependent kinase (CDK) activity. These results are consistent with the
hypothesis that a TF-network oscillator maintains oscillations in the absence of CDKactivity. In Orlando et al. [99], a TF-network was shown to oscillate in silico when
modeled with a synchronously updating Boolean framework, suggesting that oscillations
are an emergent property of the network. Recent modeling efforts suggest that all
networks with a similar structure to the network shown in Orlando et al. have this
87
oscillatory property [238]. However, the TF-network oscillator model for maintaining
transcriptional oscillations in the absence of CDK-activity has not been tested
biologically. In this section, I will describe experiments that begin to solidify the TFnetwork oscillator model.
3.1.1 Are mRNA levels a good proxy for TF function?
In the network models shown in Figures 1.3 and 2.8, nodes were identified as TFs
that are periodically expressed in the respective experiments and were placed along a
cell-cycle timeline based on peak mRNA levels [99]. When building these models, the
assumption was made that both the protein level and the activity of each TF mirrors its
mRNA level profile, increasing and falling with cell cycle progression. However, it is
possible that both post-transcriptional and post-translational regulation can affect the
periodic activity of each TF. In order to begin to test this assumption, TF proteins were
epitope tagged and protein levels through the wild-type cell cycle were measured by
quantitative western blots in synchrony experiments. Figure 3.1 shows the periodic
mRNA and protein levels of two TFs, Swi5 and Yox1 aligned on a cell-cycle timeline as
previously described[225]. The stability of all measured TFs is listed in Table 3.1.
Overall, we see that six of the ten TFs we measured are periodically present as
proteins (Figure 3.1), suggesting that these TFs are inherently unstable, as protein levels
rapidly fall after they reach peak levels. The finding that many TFs are present as
proteins periodically during the cell cycle suggests that TF activity is also short-lived
during the cell cycle.
88
Figure 3.1: TF protein dynamics. TF protein levels (red) were measured over time in
synchronized cells and aligned with wild-type transcript levels (grey) from [99]and
mapped to a cell-cycle time line as described in [225]. TF protein levels were quantified
and normalized to anti-PSTAIR. Both protein and mRNA levels are expressed as log2fold change relative to the mean. (A) Swi5-13myc. (B) Yox1-13myc.
Of the TFs that remain stable during the cell cycle, Swi4 and Mbp1 are effectively
unstable as they are active as heterodimers with the unstable protein, Swi6. The
remaining two proteins, Stb1 and Yhp1, have several mobility forms on a gel, suggesting
that these proteins are highly modified during the wild-type cell cycle. As these proteins
are not periodically expressed at the protein level, it is possible that these proteins do
not have periodic activity in the absence of CDK-activity and may not be part of the CDK-
89
Table 3.1: Transcription factor stability during the cell cycle
TF
Stb1
Swi4
Mbp1
Swi6
Yox1
Yhp1
Fkh1
Fkh2
Swi5
Mcm1
stable?
yes
yes
yes
no
no
yes
no
no
no
no
complex?
no
yes
yes
yes
no
no
yes
yes
no
yes and no
associates with
Swi6
Swi6
Swi4, Mbp1
Ndd1, Mcm1
Ndd1, Mcm1
Fkh1, Fkh2, Ndd1
independent TF-network oscillator. Accordingly, neither Yhp1 nor Stb1 appears to be
essential in TF-network models, with both having redundant edges (Figure 2.8).
It is unclear how TFs instability is regulated. Several TFs have been shown to be
ubiquitinated by mass spectrometry, suggesting that TF protein turnover is regulated by
the proteosome [249]. It is possible that they are intrinsically unstable proteins,
perhaps through the N-end rule pathway. This possibility can be tested by measuring TF
stability and B-cyclin-independent oscillations in the absence of UBR1, the E3 ubiquitin
ligase involved in the N-end rule pathway [250]. Another possibility is that they are
ubiquitinated by the SCF or APC ubiquitin ligase. However, the SCF is not essential for Bcyclin independent oscillations as cdc34and cdc53 mutants form multiple buds [144146], and the APC is only active between mitosis and the G1/S border [90, reviewed in
91]. Additionally, none of the unstable TFs in Table 3.1 have obvious APC targeting
domains in their protein sequence (destruction boxes or KEN boxes). A third possibility
90
is that TFs are made unstable by participating in transcription. Two different models,
called “timer” and “black widow”, suggest that TFs get modified and targeted for
proteosome-mediated destruction simply by participating in transcriptional activation
[reviewed in 251]. It is possible that several TFs in the network oscillator are regulated
by the same mechanism, or that each TF is independently regulated. Further
experiments will test these hypotheses in order to identify the mechanism regulating
the instability of TF proteins.
3.1.2 Are CDK-independent oscillations regulated by a TF network?
In the absence of CDK activity, transcriptional oscillations persist. It has been
proposed (in chapter 2 and in [99]), that a TF-network oscillator is responsible for CDKindependent oscillations. However, it remains possible that periodic transcription in the
absence of CDK activity is an output of an uncharacterized extrinsic oscillator. To
exclude this possibility, ongoing experiments are measuring oscillations in the absence
of B-cyclins when expression dynamics within the network are perturbed.
The TF-network oscillator model is composed of a cascade of transcriptional
regulators, in which each TF must be translated into an active protein before it can
influence the activity of downstream TFs. If the CDK-independent oscillator comprises a
TF-network oscillator, CDK-independent oscillations would be affected by protein
synthesis rates. To test this hypothesis, we treated synchronized cells over-expressing
the hyper-stable B-cyclin inhibitor (Sic1Δ3P) with sub-lethal doses of the translation
91
Figure 3.2: Protein synthetic rates affect B-cyclin-independent oscillations. Cells with
the hyper-stable B-cyclin-specific CDK inhibitor under the control of the GAL1 promoter
(Gal-Sic1Δ3P) were synchronized in G1 and released into the cell cycle in the presence
of galactose. Cells were treated with 1µg/ml CHX (A, green), 100ng/ml CHX (B, blue), or
untreated (C, red). (A-C)B-cyclin independent oscillations were measured by observing
budding cycles (1st bud, solid line; 2nd bud, broken line). Budding curves from B and C
were fit with the CLOCCS model [99, 225] and the average cycle length is shown in (D).
Error bars indicate 97.5% confidence interval from CLOCCS model.
inhibitor, cycloheximide (CHX) and measured B-cyclin independent oscillations (Figure
3.2). At a concentration of 1µg/ml, CHX completely blocks B-cyclin-independent
oscillations (Figure 3.2 A). At 100ng/ml CHX, B-cyclin independent oscillations persist;
92
however, the period of the oscillation is significantly slower than in untreated cells
(Figure 3.2 B-D). This experiment demonstrates that the CDK-independent oscillator is
sensitive to protein synthetic rates and these results are consistent with the hypothesis
that the CDK-independent oscillations are maintained by a TF-network oscillator.
We found that globally perturbing TF function by altering the protein synthetic
rate affects CDK-independent oscillations; however, inhibiting protein synthesis has
pleiotropic consequences. In order to more specifically perturb the TF network,
experiments are being done in which the expression of network TFs is perturbed. To
avoid potential confounding redundancy, dominant perturbations, specifically deletion
of transcriptional repressors or constitutive expression of transcriptional activators are
being analyzed. Preliminary experiments show that removal of transcriptional
repressors causes changes in the oscillatory period of the B-cyclin-independent
oscillator. For example, deletion of WHI5 causes extension of the oscillatory period in Bcyclin mutants, but not in the wild-type cell cycle. In contrast, deletion of MBP1 causes
a decrease in the length of the oscillatory period in B-cyclin mutants but not wild-type
cells. These and other similar experiments show that perturbing the TF-network can
cause changes in the period of B-cyclin-independent oscillations, suggesting that a TFnetwork oscillator rather than an extrinsic oscillator is responsible for CDK-independent
oscillations. Future experiments will determine how perturbing the TF-network affects
global transcriptional dynamics in the absence of B-cyclins.
93
3.1.3 Which transcriptional regulators compose the TF network oscillator?
In this section, I have described experiments that lend support to the model that
a TF-network oscillator maintains transcriptional oscillations in the absence of CDK
activity. However, it is still unknown which TFs are essential components of the network
oscillator. In order to begin to ask this question, it is first important to identify as many
potential members of the TF-network as possible, such that the level of redundancy
within the network is apparent. Then, both experimental approaches and mathematical
models can be used to determine which TFs are necessary or sufficient for robust
oscillations.
Two approaches will help identify potential components of the network. First, in
order to find additional TFs that are periodically transcribed, the global transcription
data in the absence of CDK activity will be processed to deconvolve true mRNA levels
during the cell cycle from noise due to population asynchrony. This approach has found
two-fold more periodic genes in wild-type cells than were previously identified [99,
252]. Second, experiments measuring TF localization in cells synchronously traversing
the cell-cycle by chromatin immuno-precipitation is likely to find TF binding sites that
were not identified in asynchronous cells because they are short-lived interactions.
Together, these approaches will likely uncover new TFs and connections in the TFnetwork oscillator, however, as it is impossible to say with any certainty that all
important nodes and edges are accounted for in the TF-network model, the search for
new TFs and interactions will be ongoing.
94
3.2 How does the TF-network oscillator interact with
checkpoints?
As I discussed in section 1.5, checkpoints are signaling pathways that delay cellcycle progression when important cell-cycle events are perturbed, making the delayed
event dependent on the perturbed event [126]. These pathways are known to arrest
cell-cycle progression directly or through inhibition of cyclin-CDK activity. Data in
Orlando et al., 2008 [99] and in Chapter 2 demonstrate that periodic transcription can
be uncoupled from both cell-cycle progression and CDK activity. As checkpoints are
important to maintain order in cell-cycle progression, it is likely that they regulate
periodic transcription during checkpoint arrest in order to maintain coordination of the
TF network and cell-cycle progression. This hypothesis raises the question: do
checkpoints affect periodic transcription? Ongoing experiments seek to understand the
relationship between checkpoint mechanisms and the TF-network oscillator.
To begin to understand if transcriptional dynamics are regulated by checkpoint
mechanisms, we measured global mRNA dynamics in synchronized cdc8ts cells over time
as they arrest at the DNA-replication checkpoint. Cells with a temperature sensitive
mutation in CDC8, which encodes thymidylate kinase a necessary enzyme for
deoxyribonucleotide synthesis, arrest as large budded cells with 1C DNA content [54,
253, 254]. This arrest is due to inhibition of the metaphase-to-anaphase transition
through Pds1 stabilization and APCCdc20 inhibition [132, 133]. We found that
transcriptional dynamics are significantly disrupted by checkpoint activation (Figure 3.2).
95
In checkpoint-arrested cells, many genes go through one cycle of transcription similar to
wild-type cells and then fail to accumulate in the next cycle, while many other genes are
activated and maintain high mRNA levels through the rest of the time series. Thus, it
appears that the DNA-replication checkpoint arrests the periodic transcription as well as
cell-cycle progression.
Figure 3.3: Periodic transcript dynamics during the DNA-replication checkpoint. Global
mRNA dynamics were measured over time in synchronized cdc8ts cells at the restrictive
temperature. Periodic genes from [99] are expressed as log2-fold change relative to
mean mRNA levels in rows that correspond to the same gene in (A) wild-type cells [99]
and (B) cdc8ts cells. Genes were grouped according fist cycle expression time in wildtype and each group was ordered according to expression pattern in cdc8ts cells.
Shading in (B) indicates that cdc8ts cells are arrested in S-phase by the DNA-replication
checkpoint.
96
3.2.1 What is the role of B-cyclin activity in arresting transcriptional
dynamics during checkpoint arrest?
During checkpoint arrest, cells arrest at the metaphase-to-anaphase transition
with high B-cyclin-CDK activity. A previous study showed that B-cyclins are important to
inhibit re-initiation of G1 events in checkpoint-arrested cells [68]. This data suggests
that persistent B-cyclin is responsible for inhibiting TF-network oscillator function.
There are several points in which B-cyclin-CDK activity is known to impinge on TFs in the
network oscillator. If high B-cyclin-CDK activity is involved in the inhibition of periodic
transcription, the transcription of genes known to be influenced by B-cyclin-CDK activity
should behave accordingly.
B-cyclin-CDK activity is known to repress the transcription of SBF target genes
late in the cell-cycle [69]. In the absence of B-cyclins, SBF-target genes are transcribed
at the appropriate time, however, unlike MBF-regulated G1 transcripts, mRNA levels of
SBF target genes fail to drop after peak expression is reached [99]. In DNA-replication
checkpoint arrested cells, SBF-target genes are expressed at the same time as wild-type
cells, however, the mRNA level of these genes drops after peak expression and fails to
increase again (Figure 3.4, A-D). The behavior of these genes is consistent with a role
for high B-cyclin CDK activity in preventing re-initiation of SBF-regulated transcripts
during the DNA-replication checkpoint.
Later in the cell cycle, B-cyclin-CDK activity is involved in enhancing the
transcription of the CLB2 cluster of periodic genes [102]. This positive feedback loop
97
Figure 3.4: Transcript during the replication checkpoint. mRNA levels were averaged
between replicates and are plotted for SBF target genes: CLN2 (A), CLN11 (B), PCL1 (C),
and TOS2 (D); CLB2 cluster genes: CLB2 (E), CDC20 (F), SWI5 (G), CLB1 (H), CDC5 (I) and
ACE2 (J); and Ace2/Swi5 target genes: SIC1 (K), NIS1 (L), DSE3 (M), PCL9 (N), ASH1 (O)
and PIR1 (P) in wild-type cells [99] (blue) and cdc8ts cells (red). Cell cycle time line is
based on wild-type data [99]
98
Figure 3.4
99
involves activating the SFF-subunit Ndd1 through B-cyclin-CDK phosphorylation
[reviewed in 78, 104]. Accordingly, in the absence of B-cyclins the expression of CLB2
cluster genes never reaches peak levels [99]. This mechanism would predict that high Bcyclin-CDK activity in DNA-checkpoint arrested cells would affect CLB2 cluster genes,
causing mRNA to accumulate to high levels that do not significantly drop over time.
While the mRNA levels of CLB2 cluster genes in checkpoint-arrested cells do not drop
after transcript accumulates, the peak mRNA level is much lower than in wild-type cells
(Figure 3.4 E-J).
B-cyclin-CDK activity also affects periodic expression by preventing the nuclear
import of the partially redundant transcription factors, Ace2 and Swi5 until mitotic exit
when they are dephosphorylated by Cdc14 [123, 255]. In the absence of B-cyclins, Ace2
and Swi5 target genes are expressed early, suggesting that Ace2 and Swi5 are not
delayed by nuclear exclusion [99]. In DNA-replication checkpoint-arrested cells, mRNA
of Ace2 and Swi5 target genes does not accumulate (Figure 3.4 K-P). This lack of
accumulation may be partially because SWI5 and ACE2 transcript does not reach wildtype levels (Figure 3.5 G and J). However, it is likely that persistent phosphorylation and
nuclear exclusion of Swi5 and Ace2 protein prevents transcription of their target genes.
The data suggests that B-cyclin-CDK activity affects periodic transcription at
multiple points in checkpoint-arrested cells, and that it prevents periodic transcription
from Ace2 and Swi5 target genes. It is possible that preventing periodic transcription of
Ace2 and Swi5 targets is the principal mechanism whereby B-cyclins prevents re100
initiation of G1 events in checkpoint arrested cells [68]. To test this hypothesis, G1
events will be measured in checkpoint-arrested cells expressing non-phosphorylatable
alleles of Swi5 and Ace2. Activation of a second round of G1 events in these cells would
suggest that Swi5 and Ace2 are the critical targets of B-cyclin-CDK activity that inhibit
TF-network oscillator function during checkpoint arrest. However, it is likely that
redundant mechanisms, such as B-cyclin-CDK inhibition of SBF will prevent oscillations in
these cells.
3.2.2 Are there B-cyclin independent mechanisms whereby checkpoints
affect global transcriptional dynamics?
The data in Figure 3.4 E-J suggests that high B-cyclin-CDK levels may affect CLB2
cluster transcription during the DNA-replication checkpoint. However, it is likely that
additional factors also affect expression of these genes, preventing them from reaching
peak levels. In fact, there are many changes in periodic gene expression that cannot be
explained by persistent B-cyclin-CDK activity in the cdc8ts data. It is possible that the
DNA-replication checkpoint pathway influences gene expression directly through
checkpoint-kinase phosphorylation of TFs, or indirectly through checkpoint-mediated
transcription. To tease out which changes in mRNA dynamics are due to B-cyclin-CDK
activity and which are due to checkpoint activation, future experiments will measure
global mRNA dynamics in cells arrested at the metaphase-to-anaphase transition with
high B-cyclin-CDK activity in the absence of checkpoint activation. Comparing these
101
results to mRNA dynamics in checkpoint-arrested cells should reveal which behaviors
are due to high B-cyclin-CDK activity and which behaviors are checkpoint-mediated.
3.2.3 A model for checkpoint regulation of periodic gene expression
Checkpoints are known to be important for maintaining the order of cell-cycle
events during the cell cycle. During DNA-replication stress, the DNA-replication
checkpoint arrests cell-cycle progression at the metaphase-to-anaphase transition to
allow completion of DNA replication before mitosis commences. As periodic
transcription can be uncoupled from periodic events and CDK activity, we asked if the
DNA-replication checkpoint also regulated periodic gene expression. We found that
during checkpoint arrest, periodic transcription was globally disrupted, suggesting that
checkpoint pathways arrest both periodic transcription and cell-cycle progression.
B-cyclin-CDK activity is known to be stabilized during the DNA-replication
checkpoint. Data suggests that this activity prevents re-initiation of G1 events during a
checkpoint arrest [68]. We find that some changes in periodic transcription can be
explained through persistent B-cyclin-CDK activity, and that persistent B-cyclin-CDK
activity may result in the arrest of periodic transcription. Other changes in periodic
gene expression cannot be explained by high B-cyclin-CDK activity. It is possible that
checkpoint pathways can influence periodic transcription through the activity of
checkpoint kinases or through checkpoint-mediated changes in transcription.
These observations lead to a new model for the DNA-replication checkpoint arrest,
shown in Figure 3.5. The DNA replication checkpoint arrests cell-cycle progression by
102
Figure 3.5: Model of the relationship between cell-cycle-control mechanisms. The DNA
replication checkpoint arrests cell cycle progression at the metaphase-to-anaphase
transition. As progression through anaphase leads to B-cyclin proteolysis, preventing
anaphase also stabilizes B-cyclin-CDK activity. Data suggests that B-cyclin-CDK activity
can inhibit the TF-network oscillator; however it is not yet clear whether checkpoint
pathways can directly inhibit the TF-network oscillator.
preventing progression into anaphase [132, 133], leading to the stabilization of Bcyclins. Persistent B-cyclin-CDK activity affects periodic transcription at several points in
the TF-network oscillator, and potentially arrests the oscillator by preventing nuclear
import of Swi5 and Ace2. The DNA-replication checkpoint pathway also appears to
directly affect periodic transcription; however, it is unclear if this influence has a
substantial effect on TF-network oscillator function. This new model unites checkpoint
mechanisms with the TF-network model of cell-cycle oscillations presented in Chapter 2,
and suggests that CDK-activity can entrain the TF-network to cell-cycle progression. CDK
103
entrainment of the TF-network oscillator during checkpoint arrest may be an important
means by which cells ensure fidelity during the cell cycle. Future studies will examine
the relationship between the TF-network and other checkpoints. I expect that every
checkpoint has a distinct but critical effect on periodic transcription and that modulation
of the TF-network oscillator is a common mechanism for ensuring a robust checkpoint
arrest.
3.3 Experimental details
3.3.1 Plasmid and strain construction
All strains used in this study are derivatives of BF264-15DU (MATa; ade1; his2;
leu2-3,112; trp1-1; ura3∆ns) [66]. Relevant genotypes of strains are detailed in Table
3.2. TFs (Swi4, Swi6, Yox1, Yhp1, Fkh1, Fkh2, Swi5 and Mcm1) were epitope tagged by
amplification of the 13-myc tag and the KanMX6 drug-resistance gene from pFA6a13myc-KanMX6 using gene-specific sequences as previously described [256]. Genomic
fusions were confirmed by western blot.
3.3.2 Cell growth and synchronization
Yeast cultures were grown in YEP medium (1% yeast extract, 2% peptone,
0.012% adenine, 0.006% uracil supplemented with 2% sugar (dextrose, sucrose or
galactose). cdc8ts cells were grown at 25°C and arrested at 30°C. All other yeast strains
were grown at 30°C. Mating pheromone arrest was accomplished by adding 30 to 50
ng/ml alpha-factor (α-factor) to the growth medium, cells were then released
104
Table 3.2: Yeast strains used in Chapter 3
Strain
SBY353
SBY1399
SBY1203
SBY1204
SBY1205
SBY1241
SBY1242
SBY1328
SBY1330
SBY1338
SBY1340
SBY1598
Relevant Genotype
MATa; cdc8ts
MATa; bar1; GAL1p-SIC1∆3p-URA3
MATa; bar1; SWI5-13myc-KanMx6
MATa; bar1; SWI6-13myc-KanMx6
MATa; bar1; YOX1-13myc-KanMx6
MATa; bar1; MBP1-6myc-URA3
MATa; bar1; STB1-13myc-URA3
MATa; bar1; SWI4-13myc-KanMx6
MATa; bar1; FKH2-13myc-KanMx6
MATa; bar1; MCM1-13myc-KanMx6
MATa; bar1; YHP1-13myc-KanMx6
MATa; bar1; FKH1-13myc-KanMx6
synchronously into cells lacking pheromone. cdc8ts cells were synchronized by
centrifugal elutriation and released into YEPD containing 1M sorbitol as previously
described [99]. For CHX experiments, cells were treated with CHX at a concentration of
1µg/ml or 100ng/ml or mock-treated with dimethyl sulfoxide (DMSO).
3.3.3 CLOCCS model fitting
Data was fit with the CLOCCS model of population distributions in synchrony
experiments as previously described [99, 225]. Lifeline points were used to align
experiments and for cell-cycle timeline mapping [225]. CLOCCS parameters, especially
the average cycle time (λ), were used to compare synchrony experiments [225].
105
3.3.4 Protein isolation and immunoblotting
Proteins were isolated from yeast cell extracts by TCA precipitation [257].
Protein preparations were subject to SDS-PAGE and immunoblotting using the following
antibodies: mouse anti-C Myc (9E10; Santa Cruz Biotechnology, Santa Cruz, CA), IRDye
800 conjugated goat anti-mouse (Li-Cor Biosciences, Lincoln, NE). Membranes were
analyzed with a Li-Cor Odyssey Infrared Imaging System (Li-Cor Biosciences, Lincoln, NE).
Signal was quantified using Image J 1.41o (National Institutes of Health, USA) and
normalized to anti-PSTAIR.
3.3.5 RNA isolation and microarray analysis.
Total RNA was extracted from yeast as described previously [68] and purified
with the RNAeasy MinElute Cleanup Kit (Qiagen, Valencia, CA). cDNA synthesis and
fluorescent labeling was done with the GeneChip One cycle Labeling Kit (Affymetrix,
Santa Clara, CA) and hybridized to yeast 2.0 expression arrays (Affymetrix, Santa Clara,
CA). Labeling, hybridization and image collection was performed at the Duke Institute
for Genome Sciences and Policy DNA Microarray Core Facility
(www.genome.duke.edu/cores/microarray). CEL files from each oligonucleotide array
were normalized and summarized using the dChip method as implemented in the affy
package within Bioconductor (www.bioconductor.org) using default parameters [245,
246].
106
3.4 Chapter 3 collaborator contributions
Many experiments in this chapter were done in collaboration with other
members of the Haase lab. The TF protein dynamics were done in collaboration with
Sara Bristow and Jianghai Ho. The experiments measuring B-cyclin-independent
oscillations in the absence transcriptional repressors were done by Yuanjie Jin and
Chenchen Huang. The DNA replication checkpoint data set was collected in
collaboration with Sara Bristow and analytical assistance was provided by Dr. David
Orlando.
107
Chapter 4
Intrinsic and CDK-dependent control of spindle pole
body duplication
Centrosome duplication must be tightly controlled such that occurs only once
each cell cycle. Accumulation of multiple centrosomes can result in the assembly of a
multi-polar spindle and lead to chromosome mis-segregation and genomic instability. In
this chapter, I explore two different mechanisms that prevent reduplication of the yeast
centrosome, the spindle pole body (SPB), during the cell cycle, one that is intrinsic to the
structure of the SPB itself, and another which is dependent on cyclin dependent kinase
(CDK) activity. This chapter is an adaptation with permission of a manuscript previously
published in Molecular Biology of the Cell (Copyright, The American Society for Cell
Biology, 2008) and is available at www.molbiolcell.org/cgi/content/full/19/8/3243 [258].
4.1 Introduction
Maintenance of genome stability is dependent on faithful segregation of
chromosomes at mitosis. The mitotic spindle mediates chromosome segregation
through partitioning sister chromosomes to opposite poles. In most cell types, the
mitotic spindle is organized by a pair of centrosomes present in the cell at mitosis.
However, accumulation of additional centrosomes has been observed in a variety of
human cancers [152, 259]. Supernumerary centrosomes have the potential to assemble
108
a multi-polar spindle which can lead to genomic instability and anneuploidy through
catastrophic errors in chromosome segregation [151]. In fact, several studies have
suggested that centrosome amplification can contribute to tumorigenesis [reviewed in
260].
Several circumstances could lead to generation of supernumerary centrosomes
including failed mitosis, cell fusion, de novo centrosome formation, and centrosome
reduplication during a single cell cycle [188]. Several studies demonstrate that
centrosomes can reduplicate during an extended S-phase [261-263] and this
reduplication has been shown to be dependent on cell-cycle regulatory proteins [186,
189, 190]. Additionally, mutations in a number of known cell-cycle regulators are
associated with centrosome amplification [reviewed in 188]. These lines of evidence
suggest that centrosome duplication is regulated during the cell cycle and that loss of
regulation can lead to centrosome amplification.
There is evidence to suggest that the structure of the duplicating centrosome
precludes reduplication. Cell fusion experiments patterned after the classic experiments
of Rao and Johnson [3] demonstrated that when cells with duplicated centrosomes
were fused with cells that had yet to duplicate their centrosomes, the duplicated
centrosomes did not reduplicate, even though they were placed in a cellular
environment that promotes centrosome duplication [194]. This observation suggests
that something about the constitution of the previously duplicated centrosome
prevents duplication. Recently, this centrosome-intrinsic block to centrosome
109
duplication was shown to be relieved by centriole disorientation (centriole
disengagement) [193]. Disorientation normally occurs during late mitosis or early G1
[264], and appears to be driven by separase [193]. These findings indicate that the
structure of engaged centrioles is not permissive for centriole duplication and that
disorientation exposes sites or structures necessary for centriole duplication after
mitosis.
Several studies suggest a role for cyclin dependent kinases (CDKs) in regulating
centrosome duplication during the cell cycle [reviewed in 153]. In experimental
systems that are permissive for centrosome reduplication, specific cyclin-CDK complexes
promote (cyclin A/E-Cdk2) [186, 189, 190]and restrain (cyclin B-Cdk1) [190] centrosome
duplication. In D. melanogaster, loss of Cdk1 leads to formation of additional centrioles,
whereas expression of hyper-stable cyclin A and cyclin B may inhibit centriole
duplication [191, 192]. These findings suggest that specific cyclin-CDK complexes are
important for both promoting centrosome duplication and inhibiting centrosome
reduplication during the normal cell cycle. However, it is unclear whether CDKs control
centrosome duplication directly or affect centrosome duplication indirectly by inhibiting
cell-cycle progression.
The centrosome analog in the budding yeast, S. cerevisiae, called the spindle
pole body (SPB), has been established as a model for many aspects of centrosome
biology [187]. Electron microscopy (EM) has been used to describe the structure of the
110
Figure 4.1: The SPB duplication cycle is coordinated with the cell cycle [reviewed in 196].
Steps in the SPB duplication cycle are indicated by number. Step 1, bridge elongation.
The half-bridge doubles in length. Step 2, satellite assembly. An electron dense mass of
protein, thought to be a template for the new SPB, is deposited at the distal end of the
half-bridge on the cytoplasmic face of the nuclear envelope. Step 3, SPB duplication.
The SPB is assembled in the cytoplasm through an intermediate called the duplication
plaque. The nascent SPB is then inserted into the nuclear envelope, resulting in two
side-by-side SPBs connected by a full bridge. Step 4, SPB separation. The two SPBs
move to opposite sides of the nucleus in a microtubule dependent manner, splitting the
full bridge into two half-bridges and creating a short spindle. Step 5, spindle elongation.
At mitosis the spindle elongates, segregating chromosomes into the mother and
daughter cells. Step 6, mitotic exit. At the completion of mitosis, the spindle
disassembles, the cells separate and each cell inherits a single SPB with a short halfbridge.
spindle pole body during the different stages of SPB duplication in great detail [87, 88,
197], as summarized in Figure 4.1. Subsequent genetic screens identified several genes
required for proper SPB duplication [reviewed in 154, 196] (see Table 1.2). By
examining the structure of the unduplicated SPB in mutant cells, a detailed order of
111
assembly for the SPB, and the activities required for each step has been established
[reviewed in 154, 196]. Nevertheless, relatively little is known about the regulation of
SPB duplication.
Several studies have demonstrated roles for CDK activity in the SPB duplication
cycle. Cln1/2-CDK activity has been shown to be important for SPB duplication in G1
(Figure 4.1, step 3) by directly phosphorylating important proteins in SPB duplication
[87, 89]. In addition, B-cyclins have been shown to be important for both promoting
SPB duplication and preventing SPB reduplication [67].
A role for CDK activity in the regulation of SPB duplication has been proposed in
which specific CDK complexes both promote SPB duplication and prevent SPB
reduplication [67]. In cells lacking the mitotic B-cyclins, SPBs will reduplicate, however,
in cells lacking all of the B-cyclin genes, SPBs duplicate but do not reduplicate. These
observations indicate that B-cyclin-CDK activity is required for SPB reduplication, and is
likely to promote a step(s) in the SPB duplication cycle. The B-cyclin-dependent step
appears to correlate with SPB separation, which is also known to require B-cyclin-CDK
activity [65, 67, 265]. In addition, mitotic cyclin-CDK complexes also function to inhibit
SPB reduplication, suggesting a mechanism reminiscent of the licensing model for the
control of DNA replication [67]. The evidence points to B-cyclin-CDK regulation at
multiple levels in the SPB duplication cycle, but it is not yet clear how cells prevent SPB
reduplication early in the cell cycle before B-cyclins are expressed, or how mitotic cyclinCDKs inhibit reduplication.
112
To address the mechanisms restraining SPB duplication to once per cell cycle, we
examined both CDK-dependent and -independent mechanisms that prevent SPB
reduplication. Our findings reveal a role for a SPB-intrinsic block to SPB reduplication,
similar to the centrosome-intrinsic block to centrosome duplication, which prevents SPB
reduplication early in the cell cycle. We also demonstrate that mitotic cyclin-CDKs but
not S-phase or G1 cyclin-CDKs can block SPB duplication when expressed early in G1,
and that mitotic cyclin-CDK activity can inhibit early steps in the SPB duplication cycle.
These results describe CDK-dependent and independent mechanisms that coordinate to
prevent SPB reduplication throughout the yeast cell cycle.
4.2 SPB separation is required for SPB reduplication.
Mitotic cyclin-CDKs inhibit SPB reduplication during the cell cycle [67], but
mitotic cyclins are not expressed until S phase and G2. Thus, SPB reduplication cannot
be inhibited by mitotic cyclins in G1 and early S-phase, a period when SPB components
are expressed and proteins known to be important for SPB duplication are active [89,
99]. Previous studies indicate that B-cyclin-CDK activity is required after SPB duplication
in order for SPBs to duplicate in the next cycle [67]. We investigated whether B-cyclinCDKs might function to relieve an inhibitory mechanism that normally prevents SPB
reduplication in late G1 and early S-phase.
In normal cell cycles, the SPBs separate to form a short spindle in late Sphase/G2 (Figure 4.1, step 4). In cells lacking mitotic B-cyclins, cells go on to reduplicate
113
SPBs after separation [67]. Because cells lacking all B-cyclins neither separate, nor
reduplicate their SPBs [67] and separation has been shown to be dependent on B-cyclinCDK activity (Figure 4.1, step 3) [65, 265], we hypothesized that SPB separation may be
required to relieve a block to SPB reduplication. To test this hypothesis, we took
advantage of the fact that SPB separation is dependent on the microtubule cytoskeleton
[198].
SPB separation was blocked by treating cells with the microtubule
depolymerizing drug, nocodazole, and the ability of cells to reduplicate SPBs under
conditions previously shown to promote SPB reduplication was determined [67]. Cells
bearing a hyper-stabilized allele of SIC1 (SIC1∆3P) controlled by the GAL1 promoter
were arrested in G1 with α-factor and then released in the presence or absence of
nocodazole in non-inducing medium. When cells growing in medium lacking nocodazole
had separated SPBs, galactose was added to both cell cultures to induce the expression
of SIC1∆3P, thereby inhibiting B-cyclin-CDK activity, and allowing SPB reduplication. SPB
separation and reduplication were monitored by fluorescence microscopy of Spc42-GFP.
In cells treated with nocodazole upon release from an α-factor arrest, SPBs
duplicated, but neither separated nor reduplicated, while untreated cells separated and
reduplicated SPBs (Figure 4.2, A and B). To control for unanticipated effects of
nocodazole treatment on SPB duplication, a third aliquot of cells was allowed to
separate SPBs and form a short spindle before the addition of nocodazole, which was
added concurrent with addition of galactose to stimulate expression of Sic1Δ3P. In
114
Figure 4.2: SPB separation is required for SPB duplication. Cells were synchronized in G1
and cell aliquots were either allowed to separate SPBs or prevented from separating
SPBs by addition of nocodazole or shifting cin8tskip1Δ cells to 38°C. Hyper-stable Sic1
expression was induced to promote SPB reduplication. SPBs were examined by
fluorescent microscopy of Spc42-GFP. (A) Percentages of separated and reduplicated
SPBs in cells either treated with nocodazole upon release from α-factor (broken line) or
untreated (solid line) over time. (B) Top; images showing separated and reduplicated
SPBs in untreated cells. Bottom; images showing unseparated and not reduplicated SPB
in cells treated with nocodazole upon release from α-factor. Bar, 50µm. (C) Images
showing collapsed and reduplicated SPBs in cells released from α-factor in the absence
of nocodazole to allow SPB separation and then treated with nocodazole to collapse
short spindles upon expression of Sic1Δ3P. Bar 50µm. (D) Percentages of separated
and reduplicated SPB in cin8tskip1Δ cells released at either 38°C (broken line) or 25°C
(solid line) over time.
115
Figure 4.2
116
these cells, the short spindle collapses [198], but some of the SPBs go on to reduplicate
despite their close proximity (Figure 4.2 C).
We also investigated the necessity of SPB separation prior to SPB reduplication
using mutations in the motor proteins Cin8 and Kip1. SPB separation is defective in cin8;
kip1 double mutants, and cells arrest with duplicated SPBs still attached by a full bridge
structure [199, 266]. cin8tskip1Δ cells were released from α-factor arrest at either
restrictive or non-restrictive temperature. Galactose was added to induce transcription
of SIC1∆3P from the GAL1 promoter in both cultures after the cells had been given
sufficient time to separate their SPBs at non-restrictive temperature (25°C). SPB
separation and reduplication were monitored by fluorescent microscopy of Spc42-GFP.
Many of the cin8tskip1Δ cells released at 25°C separated their SPBs and by 6 hours after
induction of SICΔ3P most of the cells with separated SPBs had reduplicated SPBs (Figure
4.2 D). As expected, few cin8tskip1Δ cells separated SPBs at 38°C, and only a small
fraction of cells released at restrictive temperature (38°C) reduplicated SPBs after 6
hours (Figure 4.2 D). The few cells at 38°C that went on to reduplicate SPBs had also
separated their SPBs. Taken together, these results demonstrate that SPB separation is
essential for SPB reduplication, and suggest that duplicated side-by-side SPBs are unable
to reduplicate until they have separated into two individual SPBs, each with its own
short half bridge (Figure 4.1, step 4).
117
4.3 A mitotic cyclin expressed in G1 can inhibit SPB duplication.
Following separation, it has been demonstrated that mitotic cyclin-CDKs prevent
SPB reduplication [67]. The structure of the SPB does not detectably change after
separation until the next G1 (Figure 4.1, steps 5, 6, and 1) [87, 88]. We posited that in
the absence of mitotic cyclins, the cell cycle arrests at metaphase, but the SPB
duplication cycle continues by progressing directly from the separated phase (Figure 4.1,
after step 4) through the early steps of the duplication cycle (Figure 4.1, steps 1-3). The
orientation of reduplicated SPBs (two sets of side-by-side SPBs separated by a short
spindle) (Figure 4.2 B) [67] is consistent with this hypothesis. In order to determine
whether mitotic cyclin-CDKs normally prevent SPB reduplication by preventing the
premature passage through these early steps in the SPB duplication cycle, we ectopically
expressed a destruction box mutant allele of the mitotic B-cyclin Clb2, Clb2Δdb, which is
stabile in G1, from the GAL1 promoter in early G1 cells.
Daughter cells in early G1 were collected from an asynchronous population of
cells by centrifugal elutriation 60 min after the induction of Clb2Δdb expression. SPB
duplication was then monitored over time by fluorescence microscopy (Figure 4.3, A
and C). Two hours after elutriation, cells not expressing Clb2Δdb had duplicated their
SPBs and initiated mitosis, however, about 70% of cells expressing Clb2Δdb failed to
duplicate their SPBs (Figure 4.3 A). Although early G1 cells expressing Clb2Δdb fail to
bud (data not shown) [62] they do rapidly replicate DNA and arrest with 2C DNA content
(Figure 4.3 B) [90], indicating that they progress through the G1/S transition.
118
Figure 4.3: Ectopic expression of Clb2 in early G1 inhibits SPB duplication. Clb2Δdb
expression was induced from the GAL1 promoter in asynchronous populations of cells,
and then small daughter cells were collected by elutriation. (A) SPB duplication and (B)
DNA content was monitored over time by fluorescence microscopy and flow cytometry
in cells expressing (red) or not expressing (blue) Clb2Δdb. (C) Western blot of Clb2
protein levels in untagged cells, cells synchronized in α-factor and released into
nocodazole for 1.5h, and in elutriated cells 30 minutes after elutriation. Cdk1 levels (αPSTAIR) are shown as a loading control. (D) Images of representative cells expressing
Clb2Δdb 120 minutes after elutriation. Bar, 50µm.
119
These results demonstrate that ectopic expression of Clb2 in early G1 can inhibit
SPB duplication, however, there is normally very little CDK activity during early G1 [31].
In order to ensure that the inhibition of SPB duplication is specific to mitotic B-cyclins
and not a simply a function of excess CDK activity in early G1, we also assayed the
effects of expressing the G1 cyclin, Cln2, and a destruction box mutant of the S-phase
cyclin, Clb5 (Clb5Δdb) [28], from the GAL1 promoter in early G1. Although Cln2 and
Clb5 promote SPB duplication later in G1, it is possible that a high level of Cln2-CDK or
Clb5-CDK activity before Start would have an inhibitory role similar to a high level of
Clb2-CDK activity. However, expression of either Cln2 or Clb5 in early G1 did not inhibit
SPB duplication, but rather accelerated the duplication process, either by direct
phosphorylation or indirectly by driving the cells more quickly through G1, as evidenced
by early bud emergence, in cells expressing either Cln2 or Clb5Δdb (Figure 4.4).
Additionally, cells expressing Clb5Δdb fail to exit mitosis (Figure 4.4, A and B) as
previously described [267]. Our findings indicate that the inhibition of SPB duplication
in early G1 is specific to mitotic cyclins and is not due to a general increase in CDK
activity during this time period.
4.3.1 Clb2 expression in early G1 inhibits satellite assembly.
Mitotic cyclin-CDK activity could block a number of early steps in the SPB
duplication cycle including bridge elongation, satellite assembly, or the assembly of a
new SPB (Figure 4.1, steps 1-3). In order to determine whether Clb2 inhibits SPB
120
Figure 4.4: Ectopic expression of G1 or S-phase cyclins in early G1 cannot inhibit SPB
duplication. (A-C) HA3-Clb5Δdb or (D-F) Cln2-HA3 were expressed from the GAL1
promoter in asynchronous populations of cells, and then small daughter cells were
collected by centrifugal elutriation. (A and D) SPB duplication and (B and E) bud
emergence were monitored over time by fluorescence microscopy in cells expressing
(red) or not expressing (blue) cyclin. Western blot of cyclin levels in cells prior to cyclin
induction (PI), prior to elutriation (PE), after elutriation (0), and 120 minutes after
elutriation (120). Cdk1 levels (α-PSTAIR) are shown as a loading control.
duplication after satellite assembly, we asked whether Clb2 could inhibit SPB duplication
in cells arrested in α-factor, when SPBs are known to have assembled a satellite [87, 88].
Cells were arrested in α-factor, and then released into media containing galactose to
induce expression of Clb2Δdb. Cells expressing Clb2Δdb and control cells not
expressing Clb2Δdb duplicated their SPBs at a similar rate (Figure 4.5 A). Furthermore,
121
SPB duplication was driven by the expression of Clb2Δdb in cells that were maintained
in α-factor (Figure 4.5 B), consistent with previous observations [90]. These findings
suggest that Clb2 inhibits early steps in the SPB duplication cycle, however, once a
satellite has formed, Clb2 no longer inhibits SPB duplication but can actually drive later
steps in the SPB duplication cycle.
Figure 4.5: Clb2 inhibits early steps in SPB duplication. (A) Asynchronous populations of
cells were arrested in α-factor and released into media inducing expression of Clb2Δdb.
(B) Cells were held in α-factor arrest as Clb2Δdb expression was induced. For both
experiments, SPB duplication was monitored by fluorescence microscopy of Spc42-GFP
over time for cells expressing (red) and not expressing (blue) Clb2Δdb. (C) Western blot
showing Clb2Δdb-HA3 levels in arrested cells (α-f), arrested cells 30 minutes after
Clb2Δdb induction (α-f + gal), and in cells released from α-factor and induced to express
Clb2Δdb for 30 minutes (cyc + gal).
122
To investigate which early steps in the SPB duplication cycle might be inhibited
by Clb2, we examined SPBs by EM in cells where SPB duplication was blocked by the
expression of Clb2 in early G1. We analyzed 50 SPBs from cells expressing Clb2Δdb 2
hours after elutriated cells were inoculated into fresh medium (Figure 4.6 A). Halfbridges were observed in 33/50 SPBs, however, we did not observe satellite structures
in any cells expressing Clb2. This result confirms that Clb2 inhibits SPB duplication prior
to satellite assembly, although we were unable to discern whether Clb2 inhibits bridge
elongation, satellite assembly or both. Interestingly, about one third of the SPBs were
observed to be larger than 100nm in diameter. Large SPB size has been previously
observed in cells with mutations that affect SPB duplication [197, 213, 268].
In order to control for our ability to identify satellite structures, we examined an
isogenic strain (not expressing Clb2Δdb) that was released into medium containing αfactor to arrest the cells at Start after elutriation. Cells arrested at Start should contain
satellite bearing SPBs as previously reported [87, 88, 197]. In the 29 control cells
examined, all SPBs observed were between 75nm and 100nm in diameter, a normal size
for haploid SPBs [87]. Eleven SPBs had observable half-bridges and nine of those SPBs
had satellite structures associated with the half-bridge (Figure 4.6 B).
Interestingly, some of the SPBs that we observed in cells expressing Clb2Δdb displayed
anomalous structures. Two half-bridges were observed emanating from opposite sides
of the SPB in 5 of the 50 SPBs analyzed (Figure 4.6 C). It is possible that additional SPBs
contain more than one half-bridge, but were not detected in the two-dimensional
123
Figure 4.6: Clb2 influences the structure of the SPB. (A, C, D) SPB structure was analyzed
by electron microscopy in elutriated cells expressing Clb2Δdb 2 hours after elutriation.
(B) SPB structure was analyzed by electron microscopy in cells arrested in α-factor after
elutriation. Bar, 100nm.
124
transmission EM images. One image of a SPB taken from the cytoplasmic face of the
structure suggests that there are indeed two separate half-bridges and not a ring of
bridge material around the SPB; however, it is difficult to distinguish whether the
electron dense material in the micrograph represents SPB material or nuclear pore
complexes adjacent to the SPB (Figure 4.6 D).
4.4 Role of B-cyclin localization in the regulation of SPB
duplication
Satellite assembly is spatially confined to the cytoplasmic face of the half-bridge
[197]. Clb2 has been shown to shuttle between the nucleus and cytoplasm [269], and to
localize to spindle poles [270]. Thus, it is possible that Clb2 may inhibit satellite
assembly by phosphorylating targets on the cytoplasmic face of the SPB. Therefore, we
examined whether Clb2 localization was important for its ability to block SPB duplication
in early G1 using CLB2 alleles that have mutations in either the nuclear localization
sequence (NLS, Δ183-200), or the nuclear export sequence (NES, L303A) [269]. Clb2,
Clb2ΔNES, and Clb2ΔNLS alleles with mutations in the destruction box were fused to
GFP, expressed under control of the GAL1 promoter, and integrated in cells with mRFPtagged Spc42 so that both Clb2 localization and SPB duplication could be monitored.
We ectopically expressed these Clb2 mutants in cells synchronized in early G1 by
elutriation as described above. As expected, Clb2 localized to both the nucleus and the
cytoplasm, Clb2ΔNES localized to the nucleus, and Clb2ΔNLS exhibited increased
cytoplasmic localization but was not excluded from the nucleus (Figure 4.7 C) [269]. SPB
125
Figure 4.7: Clb2 must be cytoplasmic in order to inhibit SPB duplication. (A-C)
Asynchronous populations of cells were induced to express GFP tagged Clb2Δdb (red),
Clb2ΔdbΔNES (blue), or Clb2ΔdbΔNLS (green) from the GAL1 promoter and small
daughter cells were then collected by centrifugal elutriation. (A) SPB duplication, and
(B) bud emergence was monitored over time by fluorescence microscopy. (C)
Representative images of cells at 120 min after elutriation. Bar, 50µm. In merged
image, GFP is green, mRFP is red and yellow represents co-localization. (D) Western blot
showing levels of Clb2 constructs in asynchronous cells grown in either dextrose “D” or
galactose “G”. Cdk1 levels (α-PSTAIR) are shown as a loading control.
126
duplication was inhibited in cells expressing either Clb2 or Clb2ΔNLS, but not in cells
expressing Clb2ΔNES (Figure 4.7 A), although all three constructs were expressed at
comparable levels (Figure 4.7 D). Additionally, while cells expressing Clb2 or Clb∆NLS
failed to bud, some, but not all cells expressing Clb2ΔNES budded over time (Figure
4.7B). These findings indicate that Clb2 must localize to the cytoplasm to inhibit SPB
duplication.
4.4.1 Cytoplasmic localization cannot confer the ability to inhibit SPB
duplication to Clb5
Previous studies demonstrate that the mitotic B-cyclins (Clb1, Clb2, Clb3, and
Clb4) can inhibit SPB reduplication, while the S-phase B-cyclins, Clb5 and Clb6, cannot
[67]. As we have shown that Clb2 must be cytoplasmic in order in inhibit SPB
duplication, it is possible that Clb5, a nuclear protein [267], cannot block SPB duplication
because it does not localize to the cytoplasm. Thus, we asked if Clb5 could block SPB
duplication if it was forced to localize to the cytoplasm. Two active leucine rich NES
sequences (NESA) or two inactive NES sequences that have key leucines mutated to
alanine (NESI) [271] were fused to mRFP-tagged Clb5Δdb [28], and expressed from the
GAL1 promoter in early G1 cells. As expected, Clb5-NESI localized to the nucleus (Figure
7D). Clb5-NESA was observed in the nucleus and cytoplasm and was observed to colocalize with Spc42-GFP (Figure 4.8 C), as has been shown previously for both Clb2 and
Clb4 [269, 270, 272]. However, localization to the cytoplasm (and the spindle pole) did
not confer the ability to inhibit SPB duplication on Clb5 (Figure 4.8 A). Thus, Clb5-CDK
127
Figure 4.8: Cytoplasmic Clb5Δdb cannot inhibit SPB duplication. Cells were induced to
express Clb5Δdb-NESA-mRFP (active NES) (red) or Clb5∆db-NESI-mRFP (inactive NES)
(blue) from the GAL1 promoter and small daughter cells were collected by centrifugal
elutriation. (A) SPB duplication and (B) bud emergence were monitored over time by
fluorescence microscopy. (C) Representative images showing Clb5Δdb-NESA-mRFP colocalization with Spc42-GFP as well as SPB duplication at 0 and 60 minutes after
elutriation. In merged image GFP is green, mRFP is red and yellow indicates colocalization. (D) Representative images showing Clb5Δdb-NESI-mRFP co-localization
with DNA as well as SPB duplication at 0 and 60 minutes after elutriation. In merged
image GFP is green, DAPI is blue and magenta represents co-localization. Bar, 50µm.
128
cannot phosphorylate the cytoplasmic Clb2-CDK targets important to inhibit SPB
reduplication.
4.5 Discussion
Ensuring that centrosomes are duplicated only once during each cell cycle is
critical for assembling a proper bipolar spindle, and for preventing mis-segregation
events during mitosis. In this chapter, I characterized both SPB-intrinsic and extrinsic
control mechanisms that prevent reduplication of the SPB during distinct cell cycle
intervals.
4.5.1 The SPB-intrinsic mechanism preventing SPB reduplication
Previous work established that SPBs reduplicate in the absence of mitotic cyclinCDKs [67]. However, we have now demonstrated that, in the absence of B-cyclin-CDK
activities, inhibition of SPB separation also prevents reduplication (Figure 4.2) prior to
mitotic cyclin expression. Both nocodazole treated and cin8tskip1Δ cells have been
shown to arrest with duplicated SPBs and an intact bridge [199, 266]. These findings
suggest that duplicated, side-by-side SPBs lack the proper structure to initiate a new
round of SPB duplication. The SPB duplication cycle normally begins with the elongation
of the half-bridge structure and the assembly of a satellite at the distal end of the half
bridge (Figure 4.1, steps 1 and 2) [87, 88]. It is likely that the proper cues for satellite
assembly are contained in the elongated half bridge, but not the full bridge that
connects side-by-side SPBs. Thus, SPB reduplication is inhibited by an SPB-intrinsic
129
mechanism, until SPB separation exposes new half bridge structures that can support
satellite assembly. This mechanism would prevent multiple rounds of SPB duplication
during G1 when B-cyclin-CDK activity is low and SPB components are expressed [99,
154].
Similar centrosome-intrinsic mechanisms have been proposed for preventing
centrosome reduplication in metazoans. Like side-by-side SPBs that are connected by a
full bridge, engaged centrioles cannot initiate a new round of centriole duplication
[193]. Centriole disorientation is essential for a new round of centriole duplication and
appears to be regulated by separase [193]. Despite the similarity in mechanism, it is
unlikely that separase plays a role in SPB separation, as separase mutants (esp1) do not
exhibit a defect in SPB separation [273]. It has been suggested that B-cyclin-CDKs
trigger SPB separation, although the mechanisms have yet to be elucidated [65, 265].
4.5.2 Inhibition of SPB licensing
A previous study demonstrated that extrinsic control by mitotic cyclin-CDK
activity also prevents SPB reduplication [67], however, the mechanisms by which mitotic
cyclins prevent SPB reduplication remained unexplored. By expressing a stabilized
version of the mitotic B-cyclin-CDK (Clb2Δdb) in at different points in G1, we established
that mitotic cyclin-CDKs are likely to prevent reduplication by inhibiting early events of
the SPB duplication cycle (Figures 4.1, 4.3, 4.5, and 4.6). The fact that Clb2 cannot
inhibit SPB duplication after α-factor arrest suggests that SPBs are already “licensed” to
duplicate at Start. Previous studies dissecting SPB structure during the duplication cycle
130
have shown that as cells transit from early G1 to Start, the SPB half bridge elongates,
and a satellite is assembled at the distal end of the bridge [67, 87, 88]. Thus mitotic
cyclin-CDKs may inhibit bridge elongation or satellite assembly. Indeed, when we
examined the SPB structure by EM in cells that expressed Clb2Δdb in early G1, we did
not observe satellites in any of the cells analyzed, although we could not determine
unambiguously if bridge elongation was inhibited. Taken together, these findings argue
that the satellite-bearing SPB may be a pre-duplicative structure, analogous to the prereplicative complex in DNA licensing models as previously hypothesized [67, 197], and
that one or more steps leading up to satellite assembly are inhibited by mitotic cyclinCDKs.
Mitotic cyclin-CDKs could inhibit satellite assembly indirectly or by direct
phosphorylation of satellite or half-bridge proteins. Interestingly, previous studies have
shown that mitotic cyclins Clb2 [270] and Clb4 [272] co-localize with the SPB, suggesting
that direct phosphorylation of satellite or half-bridge proteins could occur. The satellite
assembles on the cytoplasmic face of the SPB [197], suggesting that the sub-cellular
localization of Clb2 may be important for its ability to inhibit SPB licensing. Our finding
that Clb2 must be localized to the cytoplasm in order to inhibit licensing suggests that
the inhibitory targets of Clb2 may indeed be associated with the half bridge and/or
satellite.
131
4.5.3 Multiple half-bridges
Unexpectedly, we observed aberrant SPB structures by EM in some cells overexpressing Clb2 in early G1. In 10% of the SPBs examined, two separate half-bridges
were observable (Figure 4.6 C). As EM can capture only two-dimensional images, it is
possible that a larger percentage of cells had more than one half-bridge, or that the
SPBs observed with two half-bridges actually had several half-bridges or a ring of bridge
material around the perimeter of the SPB. An image taken from the top down suggests
that these SPBs have only two half-bridges (Figure 4.6 D), although this image alone
cannot rule out other possibilities. It is unclear why expressing Clb2 in early G1 may give
rise to additional half-bridges. It is possible that the cell can sense that the SPB has not
duplicated and assembles an additional half-bridge to attempt to remedy the problem.
It is also possible that the proteins that comprise the half-bridge accumulated during the
period that the cell was expressing Clb2. If Clb2 expression inhibits bridge elongation,
these proteins may undergo default assembly into a new half-bridge instead of an
elongated half-bridge. It is also possible that Clb2 directly promotes half-bridge
assembly through some unknown mechanism. The finding that multiple half-bridges
can assemble on the same SPB suggests that multiple sides of the SPB are competent to
support a half-bridge and raises interesting possibility that the SPB is not inherently
asymmetric.
132
4.5.4 Cyclin specificity in SPB licensing
Although Clb2 can inhibit SPB duplication if expressed in early G1, neither Clb5
nor Cln2 can inhibit SPB duplication when expressed similarly, indicating that inhibition
of SPB duplication in specific to Clb2 and not a function of elevated CDK activity in early
G1 (Figure 4.3 and 4.4). Furthermore, we have shown that cytoplasmic localization of
Clb5 is not sufficient for inhibiting SPB duplication (Figure 4.8), suggesting that the
targets of Clb2-CDK involved in inhibition of SPB duplication are not efficiently
phosphorylated by Clb5-CDK. This finding is consistent with in vitro studies indicating
that Clb5-CDK may have preferences for specific substrates, while Clb2-CDK can
phosphorylate a wider range of targets [27].
4.5.5 Model for restriction of SPB duplication to once per cell cycle
The results presented here suggest that at least two distinct mechanisms work
together to prevent SPB reduplication during the cell cycle. In early G1, when there is
low mitotic cyclin-CDK activity, the pre-duplicative structure, the satellite bearing SPB, is
allowed to assemble. Following Start, the activity of cyclin-CDK complexes (G1 cyclinCDK under normal circumstances, but mitotic cyclin-CDK will suffice) (Figure 4.5 B)
trigger the assembly of a new SPB resulting in a structure with two side-by-side SPBs
connected by a full bridge. This structure establishes a SPB-intrinsic block to SPB
reduplication during late G1 and early S-phase when the components of the SPB are
expressed [99] and the proper activating kinases are present [89]. SPB separation is
then triggered by the activation of mitotic B-cyclins [65, 265], or inefficiently by S-phase
133
cyclins [67], releasing the SPB-intrinsic block to SPB reduplication. After promoting SPB
separation, mitotic cyclin-CDKs inhibit SPB reduplication for the remainder of the cycle
by preventing the premature assembly of the pre-duplicative SPB complex. At the end
of mitosis, the inhibition and destruction of mitotic cyclins releases this inhibition, and
cells are able to reassemble the pre-duplicative SPB complex in the early G1. In this
model, CDKs have roles in both promoting and inhibiting SPB duplication during the cell
cycle.
4.5.6 Implications for the centrosome duplication cycle
It remains unclear whether centrosome duplication in metazoans is governed by
the same mechanisms we have identified in yeast, but remarkable similarities between
the duplication cycles [187] suggest that analogous regulatory themes may exist.
Centrosome-intrinsic mechanisms clearly prevent centrosome reduplication, and
centriole disorientation may be an analogous process to SPB separation. However,
centriole disorientation in metazoans appears to be regulated by separase [193], while
in budding yeast separation is regulated by B-cyclin-CDKs [65, 265]. Furthermore,
centriole disorientation does not occur until mitosis, so the centrosome-intrinsic
mechanism prevents reduplication for the bulk of the cell cycle. The question then
arises; are additional mechanisms required to prevent centrosome reduplication during
the interval between centriole disorientation in mitosis and cytokinesis, and if so, do
CDKs inhibit the assembly of a pre-duplicative centrosome complex, perhaps by
inhibiting pro-centriole assembly? It is also possible that in certain conditions or in
134
certain cell types there is a delay between centriole disorientation and cell-division.
Under such conditions, a CDK-dependent mechanism to restrict centrosome duplication
may be important. Accordingly, several studies have suggested that cyclin B-Cdk1
inhibits centrosome duplication [190-192, 274], however, it is not clear whether cyclin
B-Cdk1 inhibits centrosome duplication directly, if it inhibits separase activation [275],
or if it inhibits progression into cell cycle phases permissive for centrosome duplication.
4.6 Future Directions
In this chapter I have defined specific roles for B-cyclin-CDK activity in the
regulation of the SPB-duplication cycle, further experiments proposed in this section will
focus on the identification of the targets of CDK phosphorylation that are involved in
this process.
4.6.1 What CDK targets are important for SPB separation?
In this chapter, I defined a SPB-intrinsic block to SPB duplication in which SPB
separation is necessary for SPB reduplication. CDK activity is known to be essential for
SPB separation [65, 67, 265], however, it is unclear which CDK-targets are involved in
the regulation of SPB separation.
It has been suggested that CDK-dependent regulation of the microtubule motor
proteins Cin8 and Kip1 is involved in SPB separation. Data suggests that under low Bcyclin-CDK conditions, the APC activating protein Cdh1 is stabilized such that APCCdh1
remains active and targets Cin8 and Kip1 for proteolysis preventing SPB separation
135
[265]. However, more recently it has been shown that Cin8 and Kip1 are stable in Bcyclin mutant cells, and that direct phosphorylation of the motor proteins may be
involved in the regulation of SPB separation [276]. Further experiments will determine
how direct phosphorylation of these motor proteins affects their function and whether
phosphorylation of Cin8 and/or Kip1 is the only target of CDK-activity important for
regulation of SPB separation.
A second possible target of B-cyclin-CDK activity important for SPB separation is
the half-bridge protein Sfi1. Mutations have been found in Sfi1 that block SPB
separation [277]. Both of these mutations fall in a consensus CDK phosphorylation site,
and Sfi1 has been shown to be highly phosphorylated by Clb2-CDK in vitro [240, 277].
Additionally, these mutations are located in the C-terminus of Sfi1, where Sfi1 is
proposed to bind the C-termini of other Sfi1 molecules forming the structure of the
bridge [278]. Thus it is possible that phosphorylation of Sfi1 by mitotic cyclin-CDKs may
contribute to SPB separation by destabilization of the bridge. Analysis of Sfi1
phosphorylation during the cell cycle and functional studies of Sfi1 phosphorylation
using site-directed mutants could reveal whether Sfi1 is an important CDK-target
involved in SPB separation.
4.6.2 What are the CDK targets involved in SPB licensing?
As has been shown for DNA replication, we suspect that mitotic cyclin-CDKs
phosphorylate multiple protein involved in SPB duplication, and that any of those
phosphorylations could block SPB reduplication. However, data in this chapter suggests
136
that the important targets are localized to the cytoplasm and at least a subset of these
proteins may be structural components of the satellite or bridge structure. Of the eight
know satellite and bridge components, six are known to be essential for SPB duplication
and four of these contain minimal CDK phosphorylation sites. These four proteins,
Spc42, Spc29, Kar1 and Sfi1 are good candidate CDK-targets involved in inhibition of SPB
licensing.
Spc42 and Spc29 are components of the SPB central plaque that have been
shown to localize to the satellite [197], and are known to be phosphorylated by CDK [89,
207, 240]. Cln1/2-CDK phosphorylations have been mapped to 2 of the 8 minimal CDK
consensus sites within the Spc42. These phosphorylations likely contribute to SPB
assembly [89], however, it remains possible that phosphorylation of additional residues
in Spc42 by B-cyclin/CDK could be important in the inhibition of SPB duplication later in
the cell cycle. Recently, a role for Mps1 phosphorylation of Spc29 has been implicated
in SPB assembly, but no role for CDK phosphorylation of Spc29 has been uncovered
[279]. It is possible that phosphorylation of these proteins by mitotic-B-cyclin-CDK could
prevent Spc42 and/or Spc29 from localizing to the satellite, thus preventing SPB
licensing by inhibiting satellite deposition. Future studies could test this hypothesis
through a combination of mutational analysis of consensus CDK sites and immuno-EM
localization of Spc42 and Spc29 under different conditions.
Sfi1 and Kar1 are both structural components of the bridge that interact with the
centrin, Cdc31, which is homologous to a metazoan centrosome component [217, 218].
137
Kar1 has roles in karyogamy and SPB duplication, but has not yet been shown to be
phosphorylated by CDK. Sfi1, as mentioned above, is a half-bridge protein that has been
shown to be phosphorylated by CDK [218, 240]. Sfi1 filaments make up the structure of
the bridge, with the N-terminus of Sfi1 associating with the mother SPB and the Cterminus distal to the SPB [278]. After SPB duplication, the full-bridge is made up of two
anti-parallel Sfi1 lattices associated with each other at the C-termini [278], which
suggests that association of Sfi1 at its C-terminus is critical for bridge elongation (Figure
4.1, step 1). Sfi1 contains 7 minimal CDK phosphorylation sites in its C-terminus, and
mutations in 2 of these sites blocks SPB separation [277]. These data suggest a model in
which phosphorylation by mitotic-B-cyclin-CDK activity destabilizes the anti-parallel
association of Sfi1 molecules at the C-terminus, thus preventing SPB licensing though
inhibition of bridge elongation, and promoting SPB separation. This model suggests that
the same phosphorylation events could both relieve the intrinsic block to SPB
duplication and inhibit SPB licensing, which would be a robust mechanism to prevent
SPB reduplication. This model could be tested through a biochemical study of Cterminal binding in the presence or absence of phosphorylation. Mutational analysis of
the Sfi1 C-terminus could also provide support for this model, in fact, mutations in Sfi1
phosphorylation sites have already been shown to inhibit SPB separation [277].
However, as I expect there to be multiple CDK-targets involved in the inhibition of SPB
licensing, it is unlikely that non-phosphorylatable alleles of any potential CDK-target will
allow SPB reduplication within a single cell cycle.
138
4.7 Experimental details
4.7.1 Plasmid and strain construction.
All strains used in this study are derivatives of BF264-15DU (MATa; ade1; his2;
leu2-3,112; trp1-1; ura3Δns) [66]. Relevant genotypes of strains are detailed in Table
4.1. SBY408 was constructed by cutting plasmid pZSPC42-GFP [67] with BglII and
integrating at the TRP1 locus. SBY844 was constructed by PCR amplification of the
mRFP-KanMX6 tag from the genome of ATCC201389: SPC42-mRFP-KanMX6 [280] using
oligonucleotides FW (5’ATGGCCTCCTCCGAGGACGT3’) and RV
(5’GTTAGTATCGAATCGACAGC3’) and cloning into pDrive (Qiagen, Valencia, CA). The
sequence was re-amplified using oligonucleotides FW
(5’TATGTCAGAAACATTCGCAACTCCCACTCCCAATAATCGAGGAATGGCCTCCTCCGAGGACG
T3’) and RV
(5’TATAAAAGGCCTTTACGTTTCCGGCTTCTGTTGGAAAATAGTTAGTATCGAATCGAATCGACA
GC3’) and integrated at the SPC42 locus via homologous recombination. SBY533 was
constructed by cutting plasmid YIpG2-CLB2Δdb (a gift from Steve Reed) with Kpn1 and
integrating at the LEU2 locus. SBY520 was made by cutting plasmid pGAL1-SIC1Δ3P [97]
with EcoRV and integrating into SBY408 at the URA3 locus. SBY1353 was made by PCR
139
Table 4.1: Yeast strains used in Chapter 4.
Strain
Relevant Genotype
SBY408
MATa; bar1; SPC42-GFP-TRP1-ZeoR
SBY520
MATa; bar1; SPC42-GFP-TRP1-ZeoR; GAL-SIC1Δ3P-URA3
SBY533
MATa; bar1; SPC42-GFP-TRP1-ZeoR; GAL-CLB2Δdb-LEU2
SBY844
MATa; bar1; SPC42-mRFP-KanMX6
SBY916
MATa; bar1; SPC42-mRFP-KanMX6; GAL-CLB2Δdb-GFP-TRP1
SBY918
MATa; bar1; SPC42-mRFP-KanMx6; GAL-CLB2ΔdbΔNES-GFP-TRP1
SBY920
MATa; bar1; SPC42-mRFP-KanMx6; GAL-CLB2ΔdbΔNLS-GFP-TRP1
SBY1099
MATa bar1 SPC42-mRFP-KanMX6; GAL-CLN2-HA3-TRP1
SBY1157
MATa bar1 SPC42-GFP-TRP1-ZeoR; GAL-HA3-CLB5Δdb-URA3
SBY1167
MATa bar1; CLB2-HA3-KanR; SWE1-myc-HIS2
SBY1168
MATa; bar1; SPC42-GFP-TRP1-ZeoR; GAL-CLB2Δdb-HA3-LEU2-KanR
SBY1171
MATa; bar1; SPC42-GFP-TRP1-ZeoR; GAL-HA3-CLB5Δdb-NES-I-mRFP-URA3
SBY1173
MATa; bar1; SPC42-GFP-TRP1-ZeoR; GAL-HA3-CLB5Δdb-NES-A-mRFP-URA3
amplification of CIN8 from the genome followed by ligation into pDrive (Qiagen,
Valencia, CA). The cin8ts allele was made by site directed mutagenesis using primers
FW(5' GATAAAAGCGGCCATATACCTGCACGTGAATCGAAATTGACC 3') and RV
(GGTCAATTTCGATTCACGTGCAGGTATATGGCCGCTTTTATC 3') to generate the
temperature sensitive F429A mutation [281] and a unique Pml1 site. A Kpn1/Xba1
fragment containing cin8ts allele was subcloned into pRS306 [282]. The resulting
plasmid was cut with BglII and transformed into a version of SBY408 carrying
kip1∆::KanMX4 at the CIN8 locus. CIN8 allele replacement was achieved by growing
transformants in the presence of 5’-FOA, and confirmed by PCR followed by Pml1
digestion. The resulting strain was transformed with pGAL-Sic1Δ3P [97] cut with EcoRI
140
for integration at the URA3 locus. SBY1168 was made by amplification of the HA3 tag
and KanR from plasmid pKHA3 using primers FW
(5’GGTTAGAAAAAACGGCTATGATATAATGACCTTGCATGAACGCATCTTTTACCCATACG3’)
and RV
(5’CATACATTTTATATGGACATTTATCGATTATCGTTTTAGACATGCCATCCGTAAGATGC3’) and
integrating into SBY533 at both CLB2Δdb and CLB2. SBY1099 was made by cutting
CWB194 (pRS414-GAL1-CLN2-HA3) [222] with MfeI and integrating into SBY844 at the
TRP1 locus. SBY1157 was made by cutting plasmid DBRI (pGAL1-HA3-CLB5Δdb) [28]
with EcoRV and integrating into SBY408 at the URA3 locus.
The pGAL1-CLB2∆db∆NES-GFP and pGAL1-CLB2∆db∆NLS-GFP plasmids were
made by swapping a BbvCI/ApaI fragment from the C-terminal half of CLB2 in pPS2191
(2µ GAL1-CLB2Δdb-GFP) with a BbvC1/Apa1 fragment from either pPS2190 (2µ GAL1CLB2(L303A)-GFP) to make 2µ GAL1-CLB2ΔdbΔNES-GFP or pPS2192 (2µ GAL1CLB2(Δ183-200)-GFP) to make 2µ GAL1-CLB2ΔdbΔNLS-GFP [269]. pRS304 GAL1CLB2Δdb-GFP (and ΔNES and ΔNLS derivatives) were made by cutting 2µ plasmids
containing the construct with EcoRI/NsiI and cloning fragments into pRS304 [282] cut at
EcoRI/PstI. SBY916, SBY918, and SBY920 were made by cutting pRS304 GAL1-CLB2ΔdbGFP constructs with Bsu36I and integrating at the TRP1 locus of SBY844.
The pGAL1-HA3-CLB5Δdb-NES-mRFP constructs were made as follows. CUP1CLB5 was amplified from pKCUP1-CLB5-HA [67] using oligonucleotides FW
141
(5’CTTGCATGCCTGCAGGTC3’) and RV
(5’GTCGACCGCGGCCGCACTTAAGATTAAATAGATTTTGAAAGTTGCTATGCATTTC3’). The
resulting product was cloned into pDrive (Qiagen, Valencia, CA). A SalI fragment of
pDrive-CUP1-CLB5 was cloned into SalI digested pDrive-mRFP-KanMX6 to make pDriveCUP1-CLB5-mRFP-KanMX6 of which a SacI/BglII fragment containing CUP1-CLB5-mRFP
was subcloned into YCplac33 [283] digested with SacI/BamHI. The active (NES-A) and
inactive (NES-I) cassettes were amplified from p306-NES-A and p306-NES-I plasmids
respectively [271], using oligonucleotides FW (5’CGAA
ATGCATAGCAACTTTCAAAATCTATTTAATCTTAAGGGTTTAGCACTTAAATTAGC3’) and
either NES-A RV
(5’CGTCCTCGGAGGAGGCCATAATCTGAATTCGTCGACAAGCACTACCGATATCTAAACCTG3’)
or NES-I RV
(5’CGTCCTCGGAGGAGGCCATAATCTGAATTCGTCGACAAGCACTACCGATATCAGCACCTG3’)
and cloned into SacII digested YCplac33-CUP1-CLB5-mRFP via gap repair. GAL1-HA3CLB5∆db was subcloned from DBRI [28] into pRS306 [282] on EcoRI ends. A NotI/BspEI
fragment containing the C-terminus of CLB5 was replaced by NotI/BspEI fragments from
YCplac33-CUP1-CLB5-NES-A-mRFP or YCplac33-CUP1-CLB5-NES-I-mRFP to make pRS306GAL1-HA3-CLB5Δdb-NES-A-mRFP and pRS306-GAL1-HA3-CLB5Δdb-NES-I-mRFP. These
constructs were digested with EcoRV and integrated into the URA3 locus of SBY408 to
make SBY1173 (NES-I) and SBY1171 (NES-A).
142
4.7.2 Cell growth and synchronization
Yeast cultures were grown in YEP medium (1% yeast extract, 2% peptone,
0.012% adenine, 0.006% uracil supplemented with 2% sugar (dextrose, sucrose or
galactose). Cells were grown at 30°C. Mating pheromone arrest was accomplished by
adding 30 to 60 ng/ml alpha-factor (α-factor) to the growth medium. For
synchronization experiments, cells were then released into growth medium without αfactor. For mitotic arrest, cells were treated with 15 µg/ml nocodazole.
SPB reduplication experiments were done as previously described [67]. For
nocodazole experiments, hydroxyurea was used at 100mM to slow progression through
S-phase, and nocodazole was used at 15 µg/ml to destabilize microtubules either at
release from α-factor or at the time of induction of GAL1-SIC1Δ3P. For cin8tskip1Δ
experiments cells were incubated at either 25°C or 38°C upon release from α-factor and
galactose was added to both cultures to induce GAL1-SIC1∆3P after the 25°C culture had
become 80% budded.
Elutriation experiments were performed as follows. Cells were grown to 1.5*107
– 2.5*107 cells/ml in YEP-sucrose. The GAL1 promoter was induced by addition of 2%
galactose into the media and incubated for 45 min (for CLN2 or CLB5 constructs) or 60
min (for CLB2 constructs). Cells were then put on ice and subjected to centrifugal
elutriation and small daughter cells were collected. Cells synchronized in early G1 were
released into YEP-galactose at t=0 min.
143
4.7.3 Microscopy
All samples analyzed by fluorescence microscopy were fixed in 2%
paraformaldehyde for 15 – 30 min, were then washed with PBS and stored in 30%
glycerol. DNA staining with 4’,6-Diamidine-2-phenylindole dihydrochloride (DAPI; Roche
Diagnostics, Indianapolis, IN) was done at 1µg/ml to visualize nuclei. Cells were viewed
using a Zeiss Axio Imager widefield fluorescence microscope, a 100x objective and
standard filter sets. Foci of Spc42-GFP or Spc42-mRFP were counted for a minimum of
200 cells per sample to determine the state of SPB duplication in the sample. Images
were acquired with a Hamamatsu Orca ER monochrome cooled-CCD camera with IEEE
and captured using Metamorph 7.1 (Universal Imaging). Images were merged digitally
using Photoshop 7.0 (Adobe Systems, Inc.).
Samples were prepared for electron microscopy as previously described [284].
Samples were viewed using Philips CM 12 transmission electron microscope (FEI Co.,
Hillsboro, OR) and images were captured using an AMT XR100 Digital Camera (Advanced
Microscopy Techniques, Danvers, MA)
4.7.4 Immunoblotting
Cell lysates were subjected to SDS-PAGE and immunoblotting using the following
antibodies: mouse anti-HA (Roche Diagnostics, Indianapolis, IN), mouse anti-GFP
(Covance, Berkley, CA), mouse anti-PSTAIR (Abcam, Inc., Cambridge, MA), and HRP
conjugated goat anti-mouse (Pierce Biotechnology, Rockford, IL).
144
4.7.5 Flow Cytometry
Preparation of cells and flow cytometric analysis of DNA content was carried out
as described previously [285].
4.8 Chapter 4 collaborator contributions
The idea that an intrinsic mechanism regulates SPB duplication is based on
observations by Christine Nelson who also performed the experiment presented in
Figure 4A.
145
Chapter 5
Periodic transcription regulates spindle pole body
duplication
In chapter 2, I presented a model in which periodic genes are important
controlling for cell-cycle events. It is unclear, however, if periodic expression of genes
other than cyclins contributes to successful completion of cell-cycle events. In this
chapter, I will explore whether cell-cycle regulated transcription of spindle pole body
(SPB) component genes helps ensure that SPB duplication occurs only once per cell
cycle.
5.1 Introduction
In a variety of organisms, assembly of multi-protein structures is regulated
through temporal control of component protein expression at the level of transcription.
This method of regulation is very common in prokaryotes, with the assembly of many
structures, such as the bacterial flagella being elegantly regulated by transcriptional
mechanisms [reviewed in 286, 287]. In eukaryotes, however, protein abundance and
function is known to be regulated by a number of additional mechanisms including
mRNA stability, translation control, and post-translational regulation. Each step towards
expressing a functional protein requires energy input from the cell, thus it is
energetically favorable to regulate gene expression at an early stage. Indeed, despite all
146
the other regulatory mechanisms present in eukaryotes, regulation of mRNA expression
remains an important method to regulate gene expression.
As discussed in chapter 1.9 and chapter 4, the budding yeast SPB is a complex
multi-protein structure imbedded in the nuclear envelope where it is important for
organizing the mitotic spindle and other cellular microtubules. It is essential that the
SPB duplicates once and only once per cell cycle such that there are two SPBs in the cell
at mitosis to form a bipolar spindle. In chapter 4, I described two mechanisms
preventing SPB reduplication in the cell cycle. One mechanism that is intrinsic to the
structure of the SPB and another that is dependent on CDK activity. In this chapter I will
propose a third mechanism that prevents SPB reduplication late in the cell cycle:
regulation of SPB component proteins by transcriptional activation during a discrete
portion of the cell cycle.
5.1.1 Expression of SPB components during the cell cycle.
It is thought that SPB duplication takes place in G1 of the yeast cell cycle after
Start. Concurrently, the cell is expressing the G1 transcriptional program, a wave of
transcription including over 100 genes [102]. The activation of these genes is largely
attributed to the activity of two partially redundant heterodimeric transcriptional
activators, SBF (composed of Swi4 and Swi6) and MBF (composed of Mbp1 and Swi6)
[70, 71]. Interestingly, several SPB component genes have MBF binding sites located in
their promoter [154, 207, 288], and of the 18 SPB component genes listed in Table 1.2,
147
eleven of them were identified as periodic transcripts expressed early in the cell cycle
[99], however, most of these genes display some periodic behavior (Figure 5.1).
It is possible that SPB duplication occurs after Start because the SPB preduplicative state (see chapter 4) requires new SPB component proteins for SPB
duplication. Alternatively, there is evidence that that the wave of CDK activity at Start is
important for the activation of existing proteins involved in SPB duplication [87-89].
Evidence that transcription may regulate SPB duplication comes from the observation
that SPB size is correlates with gene dosage. The SPB of a diploid is approximately twice
Figure 5.1: Dynamics of SPB component genes in wild-type cells. Heat map depicting
mRNA levels of SPB component genes from global microarray analysis of wild-type cells
as the traverse the cell cycle [99]. Genes found to be periodically expressed are
indicated in red. mRNA levels were mapped to a cell-cycle timeline displayed across the
X-axis [225]. Transcript levels are displayed as log2-fold change relative to the mean
expression of that gene.
148
the size of a haploid and the SPB of a tetraploid yeast is even bigger [88]. Additionally
over-expression of the SPB component Spc42 causes aberrant growth of the SPB [207].
These data suggest that new transcription of SPB component genes may be important
to build a new SPB and to determine the size of the SPB. Interestingly, centrosome
duplication in higher eukaryotes requires activation of G1 transcription by E2F,
suggesting that a requirement for transcription in centrosome duplication may be
conserved [186].
If transcription is important for SPB duplication, restriction of SPB component
gene expression to G1 may contribute to SPB duplication occurring only in G1, and thus
only once per cell cycle. In this chapter, I investigate the role of transcription in SPB
duplication and reduplication. Experiments described in this chapter support the
hypothesis that just-in-time transcription is necessary for SPB duplication and that
expression of SPB component genes outside of G1 may contribute to SPB reduplication.
5.2 SPB duplication requires new protein synthesis
To test whether de novo synthesis of SPB components is required for SPB
duplication, it would be ideal to engineer a strain in which we could directly control the
transcription of all SPB components. As this is technically difficult, we tested the
importance of transcriptional activation in SPB duplication indirectly by observing SPB
duplication in the presence of cycloheximide (CHX) an inhibitor of eukaryotic
translation. In order to insure that CHX did not block the CDK activity, we expressed a
149
hyper-stable allele of CLB2, Clb2Δdb from the GAL1 promoter to drive SPB duplication.
To prevent activity of Clb2-CDK before CHX treatment, cells used for these experiments
carry the cdc28-as1 allele, such that we can inhibit CDK activity by treating cells with the
ATP analogue 1-NM-PP1 [241]. Cells were arrested in α-factor and then released into
media containing galactose and 1-NM-PP1 to induce the expression of Clb2Δdb, but
inhibit the cdc28-as1 activity. Cells were then released from 1-NM-PP1 inhibition into
media containing glucose to terminate the expression of Clb2Δdb. The culture was split
and half of the cells were treated with CHX. Almost all untreated cells duplicated their
SPBs while less than 25% of CHX-treated cells duplicated their SPBs [Figure 5.2 A]. This
experiment shows that new protein synthesis in G1 is important for SPB duplication,
implying that transcription during this time is important for SPB duplication.
Clb2Δdb has been shown to drive SPB duplication while cells are arrested in αfactor, a condition in which SPB components are under-expressed [90, 258] (Figure 4.5).
It is possible that Clb2-CDK activity may be promoting SPB duplication by up-regulating
G1 transcription in these cells. To test if Clb2Δdb-driven SPB duplication in α-factor is
dependent on new protein synthesis, an experiment similar to the one above was
carried out. Cells arrested with α-factor were induced to express Clb2Δdb in the
presence of 1-NM-PP1. Cells were released from 1-NM-PP1 into media containing
glucose to stop expression of Clb2∆db while maintaining α-factor arrest, and half of the
cells were treated with CHX. Few cells duplicated their SPBs in the absence of CHX,
however, a barely detectable number of cells duplicated their SPBs in CHX [Figure 5.2B].
150
Figure 5.2 New protein synthesis is important for SPB duplication. In all experiments
Clb2Δdb was expressed from the GAL1 inducible promoter, with cdc28-as1 activity
attenuated by the addition of 1-NM-PP1. ClbΔdb expression was then shut off and cells
were either treated with CHX (red) or mock-treated (blue). SPB duplication was
monitored by Spc42-GFP fluorescence. (A) Cells were synchronized in α-factor and
released into CHX at t=0 hours. (B) Cells were arrested in alpha factor and treated with
CHX at t=0. (C) Western analysis of the stability of the Clb2Δdb protein in an
experiment similar to B. α-factor was added to the population at t=0.
151
The low level of SPB duplication in the control culture can be explained by analysis of
protein levels [Figure 5.2 C], Clb2Δdb is not stable for the extended periods of time
necessary to get robust SPB duplication in α-factor [Figure 5.2]. Despite the low level of
SPB duplication in control cells, these data suggest that the ability of Clb2Δdb-CDK to
drive SPB duplication in α-factor requires new protein synthesis, suggesting that Clb2CDK activity may promote G1 transcription in these cells.
5.3 The role of MBF in SPB duplication.
It is well established that Clb2-CDK activity normally inhibits SBF function late in
the cell cycle [69], suggesting that MBF alone could be driving G1 transcription in
Clb2Δdb-expressing, α-factor-arrested cells. Accordingly, many SPB genes have MBF
binding sites in their promoters [154, 207, 288]. The DNA binding component of the
MBF transcription factor Mbp1 has been shown to have both transcriptional activation
and repression activities [100]. At Start, MBF is involved in the activation of a large
number of transcripts including the MBF co-repressor, NRM1 [100]. The Nrm1 protein
then binds to Mbp1 and the Nrm1-bound MBF complex then functions as a
transcriptional repressor for MBF target genes. Indeed, in Δmbp1 cells, MBF targets are
expressed constitutively, although at low levels compared to their peak expression in
MBP1 cells [81, 82, 100]. Thus, we might expect that in Δmbp1 cells, SPB component
genes would be over-expressed outside of late G1, including during α-factor treatment.
152
Figure 5.3: Clb2Δdb driven SPB duplication in α-factor is more efficient in Δmbp1 cells.
Clb2Δdb was induced in α-factor arrested cells and SPB duplication was monitored by
Spc42-GFP fluorescence. MBP1 cells expressing Clb2Δdb (blue, solid line), MBP1 cells
not expressing Clb2Δdb (blue, broken line), Δmbp1 cells expressing Clb2Δdb (red, solid
line), Δmbp1 cells not expressing Clb2Δdb (red, broken line).
To examine the role of MBF transcription in SPB duplication, SPB duplication was
observed in Δmbp1 cells over-expressing Clb2Δdb while arrested in α-factor. While
Δmbp1 cells did not duplicate their SPBs in α-factor arrested cells, Δmbp1 cells overexpressing Clb2Δdb duplicated their SPBs much more rapidly than MBP1 cells
expressing Clb2Δdb (Figure 5.3). This result suggests that MBF transcription is involved,
but not sufficient for SPB duplication, and that Clb2-CDK activity may activate G1
transcription (see section 5.6.2).
5.4 G1 transcription and inhibition of SPB reduplication
It has been shown that in the absence of B-cyclin-CDK activity, periodic
transcription and downstream events such as budding occur, and oscillate with normal
periodicity, even in the absence of cell division [68, 99]. This activity has been
153
Figure 5.4: SPB reduplication is concurrent with re-budding and transcription of SPB
component genes. (A)Cells in S-phase were induced to express Sic1Δ3P. SPB
reduplication (red) was monitored by Spc42-GFP fluorescence. Emergence of the
second bud (blue) was also monitored. (B) Dynamics of SPB component genes in Δclb1-6
cells. Heat map depicting mRNA levels of SPB component genes from global microarray
analysis of Δclb1-6 cells as the traverse the cell cycle [99, 225]. Genes found to be
periodically expressed in both wild-type and Δclb1-6 cells are indicated in red. mRNA
levels were mapped to a cell-cycle timeline displayed across the X-axis [225]. The grey
bar indicates that cells are arrested by conventional cell-cycle measures [99]. Transcript
levels are displayed as log2-fold change relative to the mean expression of that gene.
154
attributed to a transcription factor (TF) network oscillator[99] (see chapters 2 and 3). If
transcription is important for SPB duplication, then, one might expect that SPB reduplication in the absence of mitotic-B-cyclin-CDK activity should be entrained to this
oscillator and occur with the same periodicity as other G1 transcription-activated events
such as budding [67]. To test this hypothesis, cells were synchronized by α-factor and
then released into low levels of hydroxyurea (HU) to slow S-phase and obtain a
population of cells with separated SPBs. Once the population reached 60% budded, the
cells were washed and released into media containing galactose to drive expression of
the hyper-stable B-cyclin-specific CDK-inhibitor, SIC1Δ3P [97], from the GAL1 promoter.
Cells were scored for SPB reduplication and the emergence of a second bud. Although it
is difficult to correlate events late in the time course due to synchrony loss, duplication
of SPBs occurs around the time as second bud formation, suggesting that SPB
duplication may be entrained to the B-cyclin-independent oscillator, and that G1
transcription may trigger SPB reduplication (Figure 5.4 A). The result of this experiment
leads to an interesting new hypothesis; if G1 transcription is necessary for SPB
duplication then SPBs are restricted from duplicating during other phases of the cell
cycle. In the absence of all six yeast B-cyclins (Δclb1-6), about 70% of periodic genes
maintain periodic expression; these genes include all eleven periodic SPB components
[99] (Figure 5.4 B). Expression of these SPB component genes may contribute to SPB
duplication, however, we cannot rule out the possibility that the expression of any
number G1 transcripts may be important for SPB reduplication.
155
Figure 5.5: SPB reduplication in Δclb1-3; Δmbp1 cells. Cells were arrested in α-factor
and released into the cell cycle in the absence of the indicated B-cyclins. SPB
duplication was monitored by Spc42-GFP fluorescence. (A) Percent cells that
reduplicated SPBs in Δclb1-3 cells (grey), Δclb1-4 cells (blue) and Δclb1-3; Δmbp1 cells.
(B) Micrographs showing representative cells of each genotype at 6 hours after α-factor
release.
156
Figure 5.5
157
As SPB reduplication occurs concurrently with budding and expression of key SPB
component genes in the absence of B-cyclins, we propose that expression of G1
transcripts is involved in SPB reduplication in the absence of B-cyclins. To test this
hypothesis, we tested whether over-expression of MBF-component genes could drive
SPB duplication in cells with impaired SPB licensing (Chapter 4). We examined SPB
reduplication in cells lacking three of the four mitotic B-cyclins (Δclb1; clb2; clb3). In
these cells SPBs do not reduplicate, presumably because Clb4-CDK activity is sufficient
to inhibit SPB licensing, as SPBs do reduplicate in cells lacking all four mitotic B-cyclins
(Δclb1; clb2; clb3; clb4) [67]. We found that in Δclb1-3; Δmbp1 cells, 27% of cells
reduplicate SPBs, as compared to 56% of Δclb1-4 cells and 10% of Δclb1-3 cells (Figure
5.5). Thus, over-expression of MBF target genes, which includes SPB component genes,
can contribute to SPB reduplication if other mechanisms restraining SPB duplication are
impaired. However, as it is likely that the G1 transcriptional program oscillates in Δclb13 cells, it is unclear why these cells do not reduplicate SPBs.
5.5 Discussion
It has been observed that several SPB component genes are expressed
periodically and contain MBF binding sites in their promoters [99, 154, 207, 288],
however, no functional role for transcriptional regulation in SPB assembly or function
has been described. In section 5.2, I have shown that SPB duplication requires new
protein synthesis, and that over-expression of MBF transcripts accelerates SPB
158
duplication. I have also shown that SPB reduplication is concurrent with expression of
SPB component genes in the absence of B-cyclins and that over-expression of MBF
transcription allows SPB duplication in conditions that are otherwise restrictive for SPB
duplication. These results have several interesting implications for understanding the
SPB duplication cycle.
5.5.1 Just-in-time transcription of SPB component genes
In many organisms, regulated assembly of large multi-protein structures involves
transcriptional activation of the component proteins at the appropriate time.
Expression of many SPB component genes occurs in G1 at the time when SPB
duplication takes place (Figure 5.1). In addition, SPB duplication requires new proteins
synthesis, suggesting that expression of SPB component genes is important for SPB
duplication. These data suggest that yeast use just-in-time transcription to regulate the
assembly of a new SPB. As over-expression of the SPB component Spc42 causes large,
aberrant structures to form at the SPB, it may be critical that SPB component genes are
expressed at the right time and in the right ratio [197, 207]. Measuring transcription of
SPB component genes with greater temporal resolution may reveal that these genes are
expressed in a tight order that mirrors models of SPB assembly [289, 290].
5.5.2 Cyclin specificity in G1 events
It has been shown that SPBs will duplicate in α-factor arrested cells expressing
Clb2Δdb [90, 258] (see Figure 4.5 B). DNA also replicates under these conditions,
159
however, cells fail to bud and the SBF-regulated gene HCS26 is not expressed [90]. In αfactor treated cells, G1 cyclins, which are thought to be essential for SPB duplication are
specifically inhibited by Far1 [291], although I cannot rule out the possibility that G1cyclins are active when Clb2Δdb is expressed. Thus, is possible that stabilized Clb2-CDK
can substitute for G1-cyclin-CDK activity during SPB duplication under these conditions,
as previously suggested [90]. Previous studies imply that Clb2-CDK activity cannot
substitute for G1-cyclin-CDK activity in SBF mediated G1 transcription or budding, in
fact, B-cyclin-CDK activity inhibits these activities [62, 69].
In Figure 5.2, I show that SPB duplication requires new protein synthesis,
suggesting that transcriptional activation is important for SPB duplication. As Clb2-CDK
inhibits SBF-regulated transcription [69, 90], it is possible that MBF-regulated G1
transcription is important for SPB duplication. Indeed, many SPB component genes
have cis-regulatory sequences suggesting that they are regulated by MBF [154, 207,
211]. Additionally, over-expression of MBF-target genes in Δmbp1 cells accelerates SPB
duplication in α-factor (Figure 5.3) and allows SPB duplication in cells that otherwise do
not reduplicate SPBs (Figure 5.5). However, it is unclear how B-cyclin-CDK activity might
activate MBF transcription allowing for both SPB duplication and DNA replication in the
absence of G1-cyclins.
One possible mechanism by which Clb2-CDK activity could specifically activate
MBF transcription involves the repressor of G1 transcription Whi5. Whi5 binds and
inhibits SBF until Start when it is phosphorylated by G1-cyclin-CDK activity which
160
promotes the nuclear export of Whi5 and activation of SBF-dependent transcription [73,
74]. Whi5 has also been shown to bind and inhibit MBF, although this is a topic of
debate in the field [73, 74]. Whi5 can also be phosphorylated by Clb2-CDK in vitro,
suggesting that Clb2-CDK could substitute for G1-cyclin-CDK in activation of G1
transcription through this mechanism, however it remains unclear whether MBF is
regulated by Whi5 [240].
It has been previously shown that CDK activity is essential for SPB duplication
and that phosphorylation of Spc42 by G1-cyclin-CDK activity is important for SPB
duplication [87-89]. Accordingly, Δmbp1 cells cannot duplicate SPBs during α-factor
treatment, although Δmbp1 cells over-expressing Clb2Δdb duplicate SPBs in α-factor
more rapidly than MBP1 cells over-expressing Clb2Δdb. These data confirm that CDK
activity is essential for SPB duplication and suggests that Clb2-CDK can substitute for G1cyclin-CDK activity in SPB duplication.
In all, experiments looking at G1 functions in α-factor treated cells overexpressing hyper-stable Clb2 suggest that Clb2-CDK can substitute for G1-cyclin-CDK
activity in some, but not all essential G1 functions. While Clb2-CDK cannot initiate SBF
transcription or budding, it is able to drive SPB duplication and DNA replication, and may
also activate MBF transcription [90]. Interestingly, these events fit into the functional
categories traditionally assigned to SBF and MBF target genes; SBF targets are important
for budding and include G1-cyclins, while MBF targets are important for SPB duplication
161
and DNA replication and include the S-phase B-cyclins, CLB5 and CLB6 [77, reviewed in
78].
5.5.3 Three mechanisms for inhibition of SPB duplication
In Chapter 2, I demonstrated that two different mechanisms inhibit SPB
duplication throughout the cell cycle. In late G1 and early S-phase, the newly-duplicated
SPBs are joined together by the full bridge; this structure prevents SPB reduplication
until SPB separation is promoted by B-cyclin-CDK activity. After separation, mitotic Bcyclins inhibit SPB reduplication by preventing early SPB duplication events [67, 258]. In
this chapter, I suggest that a third mechanism inhibits SPB reduplication during the cell
cycle.
If SPB duplication requires expression of SPB component genes, lack of G1
transcription during G2 and mitosis in wild-type cells could contribute the restriction of
SPB duplication to G1. In Figure 5.3, I show that SPB reduplication coincides with rebudding and SPB gene expression in the absence of B-cyclins. The loss of B-cyclins in
these cells eliminates both SPB licensing and restriction of SPB component expression to
G1, thus allowing SPB reduplication. The observation that SPB reduplication does not
occur immediately after B-cyclin activity is inhibited, but is delayed for a period of time,
perhaps until G1 transcripts are expressed, supports this hypothesis. Additionally, more
cells reduplicate SPBs in Δclb1-3; Δmbp1 cells than Δclb1-3 cells, suggesting that deregulated expression of MBF transcripts, including SPB component genes, can
contribute to SPB reduplication. In the normal cell cycle, mitotic-B-cyclin-CDK inhibition
162
of SPB licensing may be sufficient to inhibit SPB reduplication; however, the
requirement for G1 transcription may be important in stress situations. For example,
during the morphogenesis checkpoint, SPBs separate; however, mitotic-B-cyclin-CDK
activity is inhibited by phosphorylation [138]. Under these conditions, lack of G1
transcription may prevent SPB reduplication.
5.5.4 Periodic transcription regulates cell-cycle events
In Figure 2.10, I present a model in which a TF-network oscillator is responsible
for generating oscillations as well as regulating periodic gene expression. One class of
periodic genes, cyclins, is known to be important for cell-cycle events. In Chapter 4, I
presented a mechanism through which oscillations CDK activity regulate SPB
duplication, an important cell-cycle event. In this chapter, I have presented evidence
that periodic transcription of SPB components plays a role in the regulation of SPB
duplication. This evidence suggests that periodic expression of genes other than cyclins
have a role in the regulation of cell-cycle events.
5.6 Future Directions
In this chapter, I have proposed a role for G1 transcription in SPB duplication and
restriction of SPB duplication to the G1 phase of the cell cycle. Further experiments
proposed in this section will confirm that expression of SPB components is concurrent
with SPB duplication in the experiments within this chapter and begin to explore how
Clb2Δdb can drive SPB duplication in α-factor arrested cells.
163
5.6.1 Does SPB component genes expression correlate with SPB
duplication?
In this chapter, I show that SPB duplication requires new protein synthesis and
argue that under different conditions, changes in SPB duplication are related to changes
in the expression of SPB component genes. In order to strengthen the conclusions
drawn in this chapter, expression of SPB component genes could be demonstrated in
the various conditions analyzed in these experiments. Using primer sets for the eleven
periodic SPB components and control sets for the SBF-regulated gene CLN2 and the
MBF-regulated gene, RNR1, gene expression could be measured relative to control
conditions by qPCR. These experiments would provide more direct evidence that
expression of SPB component genes is important for SPB duplication. Further support
could come from experiments in which the expression of periodic SPB components is
regulated by heterologous promoters.
5.6.2 How does Clb2Δdb drive SPB duplication in α-factor arrested cells?
It has been shown that expression of Clb2Δdb in α-factor treated cells can drive
SPB duplication as well as DNA replication [90, 258] (see Figure 2.4). In this chapter, I
suggest that one function of Clb2Δdb-CDK activity under these conditions is activation
of MBF transcription. Experiments in section 5.6.1, above, will confirm that MBF
transcription is, in fact, active under these conditions. In section 5.5.2, I propose that
Clb2Δdb-CDK may promote MBF transcription through phosphorylation of Whi5, which
is known to repress G1 transcription in the hypo-phosphorylated state [73, 74]. In order
164
to test this hypothesis, SPB duplication in α-factor arrested cells could be assessed in
cells bearing a non-phosphorylatable allele of Whi5 [292]. Loss of SPB duplication would
suggest that Whi5 is a target of Clb2Δdb-CDK activity under these conditions. The
opposing experiment, looking at SPB duplication in Δwhi5 cells might also suggest that
Whi5 phosphorylation is involved in SPB duplication in α-factor. There are several other
Clb2-CDK targets known to be involved in G1 transcription such as Stb1, Yox1, Plm2 and
Tos4 [232, 234, 240, 293]. Potential roles for these proteins in SPB duplication in αfactor could be assessed in a similar way as Whi5.
5.7 Experimental Details
5.7.1 Plasmid and strain construction
All strains used in this study are derivatives of BF264-15DU (MATa; ade1; his2;
leu2-3,112; trp1-1; ura3Δns) [66] unless otherwise noted. Relevant genotypes of strains
are detailed in Table 5.1. The cdc28-as1 allele [241] was moved to the 15D background
by amplifying the allele from SBY632 [98] using oligonucleotides FW
(5’GGTTGGTCTAACTCTTTGGC3’) and RV (5’GCAATGAATTTGCGGAGGTC3’), cloned into
pDrive (Qiagen) and sequenced. The URA3 gene was moved from pUCH-URA3 into
pDrive-cdc28-as1 on BamH1 ends. Finally, the cdc28-as1-URA3 construct was amplified
from pDrive-cdc28-as1-URA3 with oligonucleotides FW
(5’TAATAAAAATCGTGTTGCGATGTAAATAATGAAGAAATACGCATGCTGCAACGCGTTAC3’)
165
Table 5.1: Yeast strains used in Chapter 5.
Strain
Relevant Genotype
SBY632
W303; MATa; bar1; Δcdc28::cdc28-as1
SBY746
MATa; Δcdc28::cdc28-as1-URA3
SBY760
MATa; bar1; SPC42-GFP-TRP1-ZeoR; GAL-CLB2Δdb-LEU2; Δcdc28::cdc28-as1-URA3
SBY533
LSY050
MATa; bar1; SPC42-GFP-TRP1-ZeoR; GAL-CLB2Δdb-LEU2
MATa; bar1; SPC42-GFP-TRP1-ZeoR; GAL-CLB2Δdb-HA3-LEU2-KanR;
Δcdc28::cdc28-as1-URA3
MATa; bar1; SPC42-GFP-TRP1-ZeoR; Δmbp1::URA3
LSY052
MATa; bar1; SPC42-GFP-TRP1-ZeoR;GAL-CLB2Δdb-LEU2; Δmbp1::URA3
SBY520
MATa; bar1; SPC42-GFP-TRP1-ZeoR; GAL-SIC1Δ3p-URA3
SBY417
MATa; SPC42-GFP-TRP1-ZeoR; clb1::URA3; clb2::LEU2; clb3::TRP1; GAL-CLB1-LEU2
MATa; SPC42-GFP-TRP1-ZeoR; clb1::URA3; clb2::LEU2; clb3::TRP1; clb4::HIS2;
GAL-CLB1-LEU2
MATa; SPC42-GFP-TRP1-ZeoR; clb1::URA3; clb2::LEU2; clb3::TRP1; GAL-CLB1-LEU2;
Δmbp1::HphMx4
SBY774
SBY419
SBY1628
and RV (5’GAGAGTGAAGACAGAGGTGTTCC3’) and transformed into 15Dα, resulting in
SBY746. Gene replacement was confirmed by sequencing. SBY760 was isolated by
crossing SBY746 with SBY533 [258] and sporulating. SBY774 was made by amplification
of the HA3 tag and KanR from plasmid pKHA3 using primers FW
(5’GGTTAGAAAAAACGGCTATGATATAATGACCTTGCATGAACGCATCTTTTACCCATACG3’)
and RV
(5’CATACATTTTATATGGACATTTATCGATTATCGTTTTAGACATGCCATCCGTAAGATGC3’) and
integrating into SBY760 at CLB2Δdb. LSY050 and LSY052 were made by amplification of
URA3 from YIPlac211 [283] with oligonucleotides FW
(5’CTTAACATTCCGAGACACAACGTAAATCCCAGAAACACAAGATAAGCTCTCAAACATGAG3’)
and RV
166
(5’GAGGCCCTTTCGTCTTCAAGCTTCGCCGTTTTCATCTTTACTGCTCTCTTTCAATACCTC3’) and
transformation into SBY408 and SBY533 [258]. Gene replacement was confirmed by
PCR using oligonucleotides FW (5’GCGATTGGTCTGCAAAATCG3’) and RV
(5’CCGTTTGTATCGTATGAGCC3’). SBY1628 was made by amplification of the HphMX4
gene from pAG32 [294] using oligonucleotides FW
(5’TAACATTCCGAGACACAACGTAAATCCCAGAAACACAAGCCGGATCCCCGGGTTAATTAA3’)
and RV
(5’GTATATGGATACATGTAAAGTTCCTCTATTTATGTATATTGAATTCCGAGCTCGTTTAAAC3’)
and transformation into SBY417. Gene replacement was checked by PCR using
oligonucleotides FW (5’GCGATTGGTCTGCAAAATCG3’) and RV
(5’GCACTGCTTACTGTTATGTC3’).
5.7.2 Cell growth and synchronization
Yeast cultures were grown in YEP medium (1% yeast extract, 2% peptone,
0.012% adenine, 0.006% uracil supplemented with 2% sugar (dextrose, sucrose or
galactose). Cells were grown at 30°C. Mating pheromone arrest was accomplished by
adding 50 ng/ml α-factor to the growth medium for bar1 cells, or 2mg/ml alpha-factor
for BAR1 cells.
For CHX experiments, cells were grown in YEP-sucrose and arrested with α-factor
for 1.5 hours. 2% galactose and 500nM 1-NM-PP1 were added and cells were incubated
for 1h. Cells were isolated by centrifugation and resuspended in YEP-dextrose with or
167
without α-factor and cells were either treated with 2.5 µg/ml CHX or mock-treated with
100% ethanol.
SPB reduplication experiments in GAL-Sic1Δ3P cells and B-cyclin-mutant cells
were carried out as previously described [67, 258].
5.7.3 Microscopy
All samples analyzed by fluorescence microscopy were fixed in 2%
paraformaldehyde for 15 – 30 min, were then washed with PBS and stored in 30%
glycerol. DNA staining with 4’,6-Diamidine-2-phenylindole dihydrochloride (DAPI; Roche
Diagnostics, Indianapolis, IN) was done at 1µg/ml to visualize nuclei. Cells were viewed
using a Zeiss Axio Imager widefield fluorescence microscope, a 100x objective and
standard filter sets. Foci of Spc42-GFP were counted for a minimum of 200 cells per
sample to determine the state of SPB duplication in the sample. Images were acquired
with a Hamamatsu Orca ER monochrome cooled-CCD camera with IEEE and captured
using Metamorph 7.1 (Universal Imaging). Images were merged digitally using Image J
1.41o (National Institutes of Health, USA).
5.7.4 Immunoblotting
Cell lysates were subjected to SDS-PAGE and immunoblotting using the following
antibodies: mouse anti-HA (Roche Diagnostics, Indianapolis, IN), (mouse anti-PSTAIR
(Abcam, Inc., Cambridge, MA), and HRP conjugated goat anti-mouse (Pierce
Biotechnology, Rockford, IL).
168
Chapter 6
Discussion: new perspectives on the role of CDKs in
the regulation of the cell-cycle
“The next ten years will reveal whether we have commitment for the hard
experiments that will be needed to challenge current dogma, overturn it when
necessary, and move on to a deeper understanding of the cell cycle.”
- Andrew Murray, 2004 [220]
In this dissertation, I have presented data that explores the roles of cyclin
dependent kinase (CDK) activity during the yeast cell cycle. In chapter 2, I present a new
model for regulation of the cell-cycle in which a transcription factor (TF)-network
oscillator forms the underlying cell-cycle oscillator, a role previously attributed to CDK
activity [reviewed in 37, 38, 220] (Figure 2.10 and 6.1). In this model, oscillations in CDK
activity are entrained to the autonomous TF-network oscillator. Cyclin-CDK activity is
essential to promote cell-cycle events and feeds back on the TF-network oscillator. In
addition, the TF-network oscillator promotes the periodic transcription of many other
genes whose products are important for cell-cycle progression. The data presented in
Chapters 3, 4, and 5, support this model and further characterize the roles of the TFnetwork oscillator, CDK activity, and periodic transcription during the cell-cycle. In this
chapter, I will discuss this new model of cell-cycle regulation, focusing on roles of CDKdependent and –independent mechanisms.
169
6.1 CDK-independent mechanisms regulating the cell-cycle.
Studies in early embryonic cells suggested that cyclin-CDK activity is the masterregulator of the cell cycle in eukaryotes. In embryonic cells, cyclin is transcribed from
maternal stores of mRNA and both forms the underlying oscillator and serves as an
effector of the oscillator, promoting cell-cycle events by phosphorylation of target
genes. However, in somatic cells and yeast, cyclins must be transcribed each cycle,
suggesting that cyclin-CDK activity cannot autonomously oscillate in all eukaryotic cells.
Recently, it has been proposed that transcriptional oscillations in the absence of S-phase
and mitotic cyclins are maintained by a TF-network oscillator, suggesting that periodic
transcription has a fundamental role in the regulation of cell-cycle oscillations [99].
In chapter 2, I present data indicating that transcriptional oscillations occur with
a period similar to normal cell-cycle oscillations in the absence of all CDK activity
(Figures 2.4 and 2.7). These data indicate that CDKs do not form the underlying cellcycle oscillator. I suggest that a TF-network oscillator generates both CDK-independent
oscillations and is the fundamental mechanism producing oscillations in normally cycling
cells (Figure 2.8, 6.1 A).
Several studies have proposed that a serial network of transcriptional regulators
could generate periodic transcription [80, 149, 150], and it has been suggested that
coupled CDK and TF-network oscillators produce cell-cycle oscillations [99]. However,
the data in chapter 2 represent the first evidence that a TF-network oscillator could
170
Figure 6.1: Model for cell-cycle regulation. (A) a TF-network generates cell-cycle
oscillations and promotes periodic transcription (C), including periodic expression of
cyclins (B). Periodic gene expression contributes to the regulation of cell-cycle events
(D). CDK-activity is essential for cell-cycle events (G) and also feeds back on the TFnetwork (E) enforcing robust oscillations, and contributes to some periodic gene
expression (F).
function independently of all CDK activity. Data discussed in Chapter 3 indicates that
perturbing the TF-network changes the period of B-cyclin-independent oscillations,
providing evidence that a TF-network is, in fact, responsible for CDK- independent
oscillations.
I propose that the TF-network oscillator not only produces cell-cycle oscillations,
it also outputs periodic expression of about 20% of the yeast genome, including cyclin
genes (Figure 6.1 B and C) [99]. Thus, cyclin-CDK oscillations are generated by periodic
171
transcription; this periodic cyclin-CDK activity is necessary for the regulation of cell-cycle
events [60, 61, 63, 66]. As CDK activity is also known to influence periodic transcription,
it is not surprising that the independence of transcriptional oscillations had not
previously been appreciated.
Evidence presented in chapter 5 suggests that cyclin genes are not the only
output of the TF-network oscillator important for the regulation of cell-cycle events
(Figure 6.1 D). Many of the proteins that make up the yeast centrosome, the spindle
pole body (SPB), are periodically transcribed in G1, when SPB duplication takes place. I
show evidence that suggests G1 transcription is essential for SPB duplication (Figure 3.2
and 3.3). In addition, I show that restriction of G1 transcription to G1 is important to
prevent reduplication of the yeast centrosome, the SPB (Figure 5.5). These data suggest
that periodic transcription of SPB component genes is involved in the regulation of SPB
duplication, which provides evidence that cyclins are not the only genes whose periodic
transcription is important for the regulation of cell-cycle events.
In chapter 4, I show that cell-cycle events can be regulated through mechanisms
independent of both CDK-activity and periodic transcription. I demonstrate that SPB
reduplication is prevented early in the cell-cycle through a mechanism intrinsic to the
structure of duplicated SPBs (Figure 4.2). This mechanism prevents SPB reduplication by
ensuring that sites necessary for SPB duplication are not accessible until CDK-dependent
mechanisms able to prevent SPB duplication are present.
172
173
Figure 6.2: Regulatory network regulating the yeast cell-cycle. Transcriptional activators, green, and transcriptional
repressors, red, regulate each other’s expression (black edges) and the expression of cyclins (blue nodes and edges).
CDKs feedback on the TF-network oscillator (blue edges) and regulate cell-cycle events (yellow nodes and edges).
In this dissertation, I presented evidence that CDK-independent mechanisms
regulate many facets of the cell-cycle. Together, these data suggest that the cell-cycle is
not regulated by a “master regulator” alone, but instead is regulated by a larger
network of regulatory factors that collaborate to regulate cell-cycle events (Figure 6.2).
6.2 CDK-dependent mechanisms regulating the cell-cycle.
Although CDK-activity is not essential for cell-cycle oscillations, CDKs do have
crucial roles during the cell cycle. In chapter 2, I show that removal of CDK activity leads
to increased noise in TF-network oscillations (Figure 2.9). These data suggest that CDKactivity feeds back on the TF-network enhancing the robustness of TF-network
oscillations (Figure 6.1 E). This feedback is likely a result of CDK-dependent
phosphorylation of many TFs within the network. During the wild-type cell-cycle CDK
activity is known to phosphorylate 13 of the 24 TFs in the TF-network [73, 74, 149, 239,
240] (Figure 6.2). This post-transcriptional regulation of TFs likely fine-tunes the
inherent oscillatory behavior of the TF network, and contributes to the regulation of
periodic transcription that fails to occur in the absence of CDK-activity (Figure 6.1 F). In
addition, I suggest in Chapter 2 that CDK-activity also serves to entrain TF-network
oscillations to cell-cycle progression. Data supporting this hypothesis can be found in
Chapter 3. I show that when a cell-cycle event is perturbed, checkpoint mechanisms not
only arrest cell-cycle progression, but also arrest TF-network oscillations (Figure 3.5). It
is likely that the arrest of transcriptional oscillations is dependent, at least in part, on B174
cyclin-CDK activity [68] and that phosphorylation of TFs in the TF-network oscillator may
be involved (Figure 3.6).
As CDKs promote cell-cycle events, oscillations in CDK-activity help ensure that
cell-cycle events occur only once per cell-cycle (Figure 6.2 G) [67, 153, reviewed in 177,
178]. In chapter 4, I explore the mechanism by which mitotic-B-cyclin-CDK activity
regulates SPB duplication during the cell cycle. I show that mitotic-B-cyclin-CDK activity
prevents SPB reduplication during the normal cell-cycle by inhibiting early steps in the
SPB duplication process (Figures 4.3 and 4.6). As mitotic-B-cyclin-CDK activity oscillates
during the cell-cycle, SPB duplication also oscillates, occurring only when mitotic-Bcyclin-CDK activity is low. This licensing mechanism, and a similar licensing mechanism
for DNA replication [177, 178], indicate that oscillations in CDK activity are important for
ensuring that cell-cycle events occur only once per cell-cycle.
In this dissertation, I have defined new roles for CDK activity in the regulation of
the cell-cycle and have explored previously defined roles in more detail. Although I have
found that CDKs do not form the underlying cell-cycle oscillator, they are critical
effectors of the TF-network oscillator. CDKs both contribute to robust network
oscillations and coordinate cell-cycle events with these oscillations. Together, these
data indicate that CDKs have key roles in cell-cycle regulation, and suggest that CDKs are
an integral part of the network of factors that regulate cell-cycle progression (Figure
6.2).
175
6.3 The evolution of cell-cycle control.
In Figures 6.1 and 6.2, I show models for cell-cycle regulation that involve a
network of intra-connected factors. In order to understand why the cell-cycle may be
regulated in such a way, it is important to consider how the eukaryotic cell-cycle
regulatory network evolved.
The cell-cycle in prokaryotes differs from the cell-cycle in eukaryotes in many
ways. Most fundamentally, the circular prokaryotic chromosome is structured
differently than eukaryotic chromosomes and is therefore replicated and segregated
through different mechanisms than eukaryotic chromosomes. The bacteria,
Caulobacter crescentus, differs from most bacteria in that DNA replication is initiated
only once per cell-cycle in a distinct period [reviewed in 169]. This parallel to the
eukaryotic cell-cycle has made Caulobacter an important bacterial model for cell-cycle
regulation.
The Caulobacter cell-cycle has been shown to be regulated by a TF-network
composed of at least 4 transcriptional regulators (Figure 6.3) [reviewed in 167, 170,
295]. The most well-studied of these TFs, CtrA, promotes its own transcription as well
as the transcription of the methylase CcrM before DNA replication [296]. CcrM then
promotes transcription of DnaA by methylating its promoter [297]. DnaA then activates
transcription of GcrA [171], which completes the circuit by activating CtrA [170]. CtrA
then negatively feeds back on GcrA, repressing its expression [298]. Interestingly, this
176
Figure 6.3: The cell-cycle in Caulobacter is regulated by a TF-network. TFs (green)
regulate the expression of each other and directly regulate DNA replication (yellow).
TF-network also directly regulates initiation of DNA replication. The TFs, CtrA and DnaA
both bind to the Caulobacter origin of replication inhibiting or activating replication
initiation respectively (Figure 6.3) [299, 300]. Thus, in Caulobacter, a TF-network can
regulate both cell-cycle oscillations and directly regulate cell-cycle events. As both yeast
and Caulobacter utilize a TF-network oscillator, it is likely that their common ancestor
also regulated its cell-cycle through a TF-network oscillator [173].
Other features of Caulobacter cell-cycle regulation are shared by eukaryotes.
The CtrA protein’s activity is regulated by phosphorylation and the stability of all four
transcriptional regulators (CtrA, CcrM, DnaA and GcrA) are regulated by temporally
controlled proteolysis [170, 300-302]. Regulation through these conserved mechanisms
suggests that phosphorylation and proteolysis were also important cell-cycle regulatory
mechanisms in the common ancestor of Caulobacter and eukaryotes.
Phylogenetic analyses of eukaryotic kinases indicate that CDKs evolved recently
as compared to other cell-cycle kinases, suggesting that primordial eukaryotic cell-cycle
177
regulation was CDK-independent [303]. How CDKs became an essential cell-cycle
regulator in the common eukaryotic ancestor is unclear, however, it has been suggested
that the primordial CDK was involved in transcriptional regulation [304]. If so, it is
possible that the initial role for CDK was to feedback on the TF-network oscillator. It is
also unclear how acquisition of eukaryotic-specific features involved in the cell-cycle
such as nuclei, microtubule organizing centers, and linear chromosomes are related to
the incorporation of CDK into the cell-cycle regulatory network [304]. We know that
CDKs, ubiquitin-dependent proteolysis and several CDK-targets are shared features of
cell-cycle regulation in all eukaryotes, suggesting that establishment of CDKs in the
network of cell-cycle regulators preceded eukaryotic expansion. Is there a relationship
between the evolution of CDKs and the success of eukaryotes?
6.3.1 Conservation and variation among eukaryotic cell-cycle networks
I have suggested that primordial eukaryotes used a TF-network to regulate both
cell-cycle oscillations and cell-cycle events, and that the last common ancestor of all
eukaryotes had integrated CDKs into their cell-cycle regulatory network. This cell-cycle
regulatory network is likely similar in structure to the yeast cell-cycle regulatory network
shown in Figure 6.2, which is generalized in Figure 6.4. Why might evolution of a cellcycle network with this structure have been advantageous for the last common ancestor
of all eukaryotes?
In order to begin to answer that question, we will need to examine in more
detail the entire cell-cycle regulatory network in a variety of species. Do other
178
Figure 6.4: Generalized structure of a eukaryotic cell-cycle regulatory network. A TFnetwork oscillator (green) generates oscillations. CDKs (blue) are effectors of the
oscillator that both feedback on the TF-network and regulate cell-cycle events by
phosphorylating target proteins (purple). All potential interaction types are represented
by edges.
eukaryotes have similar cell-cycle networks? Certainly, they all have TFs and CDKs that
promote cell-cycle events through phosphorylation of a number of targets; however, it
is not yet clear whether the relationships between these factors are conserved.
Interestingly, evidence suggests that evolution has not affected the entire
network uniformly. If we compare nodes in the yeast cell-cycle regulatory network with
their homologues in metazoans, we find that cyclins and CDKs are highly conserved. In
fact, human cyclins can complement yeast cyclin mutants [305]. Conversely, many TFs
and CDK targets are not conserved. In the networks in Figure 6.2 and 6.4, CDKs are
located in the center of the network and are highly connected. Several studies in a
179
variety of networks have shown that highly connected nodes, called hubs, are often
essential genes and tend to be conserved [reviewed in 306]. In contrast, a comparison
of inferred TF networks across bacterial species found that target genes are much more
highly conserved than their regulatory TFs, and a comparison of CDK-targets across
yeast species found that CDK-phosphorylation sites are not well conserved [307, 308].
These findings support the observations that both TFs and CDK targets are less
conserved than cyclins and CDKs, suggesting that variation among cell-cycle networks
may be at the TF-network oscillator and CDK-target level. Perhaps the conservation of
CDKs allows flexibility in the rest of the network, including in the TF-network oscillator.
Changes in the oscillator could give rise to variation in period length, potentially
allowing the eukaryotic cell-cycle network to adapt to different lifestyles. In chapter 2, I
show that CDKs confer robustness to cell-cycle oscillations. Could CDKs also have a role
in making the structure of the primordial eukaryotic cell-cycle network evolutionarily
robust?
6.4 The role of CDKs in the metazoan cell-cycle
In this dissertation, I have shown that CDKs are not essential for cell-cycle
oscillations and have suggested that a TF-network oscillator is the fundamental cellcycle oscillator in the yeast, Saccharomyces cerevisiae, a single-cellular fungi. However,
it is unknown whether CDK-independent oscillations exist in multi-cellular organisms.
180
The cell-cycle of multi-cellular organisms is fundamentally different than that of
free-living single-cellular organisms, such as yeast, in several ways. Perhaps the most
fundamental difference is that all yeast cells divide in order to reproduce, while cell
specialization in multi-cellular organisms requires that some cells divide while others are
quiescent. This systemic regulation of cell-division in multi-cellular organisms is likely
responsible for the vast differences among well-studied models for cell-cycle regulation:
yeast, metazoan embryonic and metazoan somatic cells.
Metazoan early embryonic cells are unique among cell-types in that they are
fueled by maternal stores of proteins and mRNAs. Thus, it is likely that the transcriptionindependent cyclin-CDK oscillator found in these cells is unique to embryonic systems,
although its core components are common to all eukaryotic cells. In somatic cells,
however, cyclin expression is dependent on transcription, suggesting that a TF-network
may be involved in cell-cycle oscillations in somatic cells [reviewed in 173].
Cell-division must be tightly regulated in multi-cellular organisms to ensure
coordination with organismal development. As a result, the cell-cycle does not oscillate
in most somatic cells, as they spend much of their life in a non-proliferative G0 state. To
exit G0 and re-enter the cell cycle, somatic cells require the presence of growth factors.
One major pathway by which growth factors induce cell-cycle entry is through the Ras
small GTPase, which triggers a MAPK cascade that ultimately activates the myc
transcription factor. Myc transcriptionally activates cyclin D as well as many other genes
important in cell-cycle entry [reviewed in 309]. Thus, in metazoans, cell-cycle entry is
181
ultimately a transcriptional response, suggesting that cyclin-CDK oscillations alone
cannot regulate the somatic cell-cycle.
It is possible that periodic transcription continues to oscillate during G0, although
cell-cycle entry is inhibited. This state would be similar to α-factor treated yeast cells
(see Chapter 2). As somatic cells release fairly synchronously from G0 arrest, induction
of cyclins upon growth-factor treatment may entrain cell-cycle progression and
transcriptional oscillations. Alternatively, during G0, like during checkpoint arrest (see
Chapter 3), periodic transcription may arrest and await growth factor signals to
continue.
Normally quiescent somatic cells do undergo continuous cell-cycle oscillations if
they are treated with constant growth factors, or if they have developed growth-factor
independent cell-cycle entry, through oncogenic Ras or myc, for example. What is the
underlying oscillator in these cells? It is possible that a TF-network is the fundamental
oscillator in these cells, or that growth factor signaling, itself, is responsible for cell-cycle
oscillations. Intriguingly, oscillations in different cell-types may be entrained to different
oscillators. Likewise, cell-cycle oscillations in individual cancers could be driven by
unique oscillators. These different oscillators may be more or less dependent on CDKactivity. Variation in CDK-dependency would suggest that the cell-cycle regulatory
network in metazoans is very flexible, allowing for several distinct oscillatory states.
182
6.5 Concluding Remarks
In this dissertation, I have examined CDK-dependent and –independent
processes involved in regulation of the cell cycle. I have found evidence that supports
previous models in which CDK-activity has a central role in regulating the cell-cycle.
However, perhaps surprisingly, I uncover fundamental roles for CDK-independent
mechanisms in cell-cycle control. While data presented within this dissertation conflicts
with previous notions that CDK-activity is the “master regulator” of the cell-cycle, these
data contribute to a new model for cell-cycle regulation in which a large network of
regulatory factors are responsible for cell-cycle regulation. Although this model raises
as many questions as it answers, it will hopefully contribute to a new approach to
understanding the fundamentals of cell division.
183
References
1.
Mazia, D. (1974). The cell cycle. Scientific American 230, pages 53-64.
2.
Howard, A., and Pelc, S.R. (1951). Nuclear incorporation of P32 as demonstrated
by autoradiographies. Experimental Cell Research 2, pages 178-187.
3.
Rao, P.N., and Johnson, R.T. (1970). Mammalian cell fusion: studies on the
regulation of DNA synthesis and mitosis. Nature 225, pages 159-164.
4.
Masui, Y., and Markert, C.L. (1971). Cytoplasmic control of nuclear behavior
during meiotic maturation of frog oocytes. Journal of Experimental Zoology 177,
pages 129-145.
5.
Smith, L.D., and Ecker, R.E. (1971). The interaction of steroids with Rana pipiens
Oocytes in the induction of maturation. Developmental Biology 25, pages 232247.
6.
Masui, Y. (2001). From oocyte maturation to the in vitro cell cycle: the history of
discoveries of Maturation-Promoting Factor (MPF) and Cytostatic Factor (CSF).
Differentiation 69, pages 1-17.
7.
Guerrier, P., Moreau, M., and Doree, M. (1977). Hormonal control of meiosis in
starfish: stimulation of protein phosphorylation induced by 1-methyladenine.
Molecular and Cellular Endocrinology 7, pages 137-150.
8.
Sunkara, P.S., Wright, D.A., and Rao, P.N. (1979). Mitotic factors from
mammalian cells: a preliminary characterization. Journal of Supramolecular
Structure 11, pages 189-195.
9.
Sunkara, P.S., Wright, D.A., and Rao, P.N. (1979). Mitotic factors from
mammalian cells induce germinal vesicle breakdown and chromosome
condensation in amphibian oocytes. Proceedings of the National Academy of
Sciences U S A 76, pages 2799-2802.
10.
Weintraub, H., Buscaglia, M., Ferrez, M., Weiller, S., Boulet, A., Fabre, F., and
Baulieu, E.E. (1982). [Demonstration of maturation promoting factor activity in
Saccharomyces cerevisiae]. Comptes rendus des séances de l'Académie des
sciences. Série III, Sciences de la vie 295, pages 787-790.
184
11.
Masui, Y., and Clarke, H.J. (1979). Oocyte maturation. International Review of
Cytology 57, pages 185-282.
12.
Wasserman, W.J., and Smith, L.D. (1978). The cyclic behavior of a cytoplasmic
factor controlling nuclear membrane breakdown. Journal of Cell Biology 78,
pages R15-22.
13.
Wasserman, W.J., and Masui, Y. (1975). Effects of cyclohexamide on a
cytoplasmic factor initiating meiotic naturation in Xenopus oocytes. Experimental
Cell Research 91, pages 381-388.
14.
Maller, J., Wu, M., and Gerhart, J.C. (1977). Changes in protein phosphorylation
accompanying maturation of Xenopus laevis oocytes. Developmental Biology 58,
pages 295-312.
15.
Doree, M. (1982). Protein synthesis is not involved in initiation or amplification
of the maturation-promoting factor (MPF) in starfish oocytes. Experimental Cell
Research 139, pages 127-133.
16.
Evans, T., Rosenthal, E.T., Youngblom, J., Distel, D., and Hunt, T. (1983). Cyclin: a
protein specified by maternal mRNA in sea urchin eggs that is destroyed at each
cleavage division. Cell 33, pages 389-396.
17.
Arion, D., Meijer, L., Brizuela, L., and Beach, D. (1988). cdc2 is a component of
the M phase-specific histone H1 kinase: evidence for identity with MPF. Cell 55,
pages 371-378.
18.
Gautier, J., Norbury, C., Lohka, M., Nurse, P., and Maller, J. (1988). Purified
maturation-promoting factor contains the product of a Xenopus homolog of the
fission yeast cell cycle control gene cdc2+. Cell 54, pages 433-439.
19.
Labbe, J.C., Lee, M.G., Nurse, P., Picard, A., and Doree, M. (1988). Activation at
M-phase of a protein kinase encoded by a starfish homologue of the cell cycle
control gene cdc2+. Nature 335, pages 251-254.
20.
Lohka, M.J., Hayes, M.K., and Maller, J.L. (1988). Purification of maturationpromoting factor, an intracellular regulator of early mitotic events. Proceedings
of the National Academy of Sciences U S A 85, pages 3009-3013.
185
21.
Nurse, P., Thuriaux, P., and Nasmyth, K. (1976). Genetic control of the cell
division cycle in the fission yeast Schizosaccharomyces pombe. Molecular
General Genetics 146, pages 167-178.
22.
Simanis, V., and Nurse, P. (1986). The cell cycle control gene cdc2+ of fission
yeast encodes a protein kinase potentially regulated by phosphorylation. Cell 45,
pages 261-268.
23.
Labbe, J.C., Capony, J.P., Caput, D., Cavadore, J.C., Derancourt, J., Kaghad, M.,
Lelias, J.M., Picard, A., and Doree, M. (1989). MPF from starfish oocytes at first
meiotic metaphase is a heterodimer containing one molecule of cdc2 and one
molecule of cyclin B. EMBO Journal 8, pages 3053-3058.
24.
Draetta, G., Luca, F., Westendorf, J., Brizuela, L., Ruderman, J., and Beach, D.
(1989). Cdc2 protein kinase is complexed with both cyclin A and B: evidence for
proteolytic inactivation of MPF. Cell 56, pages 829-838.
25.
Morgan, D.O. (1995). Principles of CDK regulation. Nature 374, pages 131-134.
26.
Morgan, D.O. (1997). Cyclin-dependent kinases: engines, clocks, and
microprocessors. Annual Review of Cell and Developmental Biology 13, pages
261-291.
27.
Loog, M., and Morgan, D.O. (2005). Cyclin specificity in the phosphorylation of
cyclin-dependent kinase substrates. Nature 434, pages 104-108.
28.
Cross, F.R., Yuste-Rojas, M., Gray, S., and Jacobson, M.D. (1999). Specialization
and targeting of B-type cyclins. Molecular Cell 4, pages 11-19.
29.
Roberts, J.M. (1999). Evolving ideas about cyclins. Cell 98, pages 129-132.
30.
Bloom, J., and Cross, F.R. (2007). Multiple levels of cyclin specificity in cell-cycle
control. Nature Reviews Molecular Cell Biology 8, pages 149-160.
31.
Miller, M.E., and Cross, F.R. (2001). Cyclin specificity: how many wheels do you
need on a unicycle? Journal of Cell Science 114, pages 1811-1820.
32.
Songyang, Z., Blechner, S., Hoagland, N., Hoekstra, M.F., Piwnica-Worms, H., and
Cantley, L.C. (1994). Use of an oriented peptide library to determine the optimal
substrates of protein kinases. Current Biology 4, pages 973-982.
186
33.
Holmes, J.K., and Solomon, M.J. (1996). A predictive scale for evaluating cyclindependent kinase substrates. A comparison of p34cdc2 and p33cdk2. Journal of
Biological Chemistry 271, pages 25240-25246.
34.
Nigg, E.A. (1991). The substrates of the cdc2 kinase. Seminars in Cell Biology 2,
pages 261-270.
35.
Murray, A.W., and Kirschner, M.W. (1989). Cyclin synthesis drives the early
embryonic cell cycle. Nature 339, pages 275-280.
36.
Murray, A.W., Solomon, M.J., and Kirschner, M.W. (1989). The role of cyclin
synthesis and degradation in the control of maturation promoting factor activity.
Nature 339, pages 280-286.
37.
Murray, A.W., and Kirschner, M.W. (1989). Dominoes and clocks: the union of
two views of the cell cycle. Science 246, pages 614-621.
38.
Murray, A.W. (1989). Cell biology: the cell cycle as a cdc2 cycle. Nature 342,
pages 14-15.
39.
Lu, X., Li, J.M., Elemento, O., Tavazoie, S., and Wieschaus, E.F. (2009). Coupling of
zygotic transcription to mitotic control at the Drosophila mid-blastula transition.
Development 136, pages 2101-2110.
40.
Newport, J., and Kirschner, M. (1982). A major developmental transition in early
Xenopus embryos: I. characterization and timing of cellular changes at the
midblastula stage. Cell 30, pages 675-686.
41.
Etkin, L.D. (1988). Regulation of the mid-blastula transition in amphibians.
Developmental Biology (N Y 1985) 5, pages 209-225.
42.
Ekholm, S.V., and Reed, S.I. (2000). Regulation of G(1) cyclin-dependent kinases
in the mammalian cell cycle. Current Opinion in Cell Biology 12, pages 676-684.
43.
Dyson, N. (1998). The regulation of E2F by pRB-family proteins. Genes and
Development 12, pages 2245-2262.
44.
Helin, K. (1998). Regulation of cell proliferation by the E2F transcription factors.
Current Opinion in Genetics and Development 8, pages 28-35.
187
45.
Yam, C.H., Fung, T.K., and Poon, R.Y. (2002). Cyclin A in cell cycle control and
cancer. Cellular and Molecular Life Sciences 59, pages 1317-1326.
46.
Strohmaier, H., Spruck, C.H., Kaiser, P., Won, K.A., Sangfelt, O., and Reed, S.I.
(2001). Human F-box protein hCdc4 targets cyclin E for proteolysis and is
mutated in a breast cancer cell line. Nature 413, pages 316-322.
47.
Furuno, N., den Elzen, N., and Pines, J. (1999). Human cyclin A is required for
mitosis until mid prophase. Journal of Cell Biology 147, pages 295-306.
48.
Gerace, L., and Foisner, R. (1994). Integral membrane proteins and dynamic
organization of the nuclear envelope. Trends in Cell Biology 4, pages 127-131.
49.
Vandre, D.D., Davis, F.M., Rao, P.N., and Borisy, G.G. (1984). Phosphoproteins
are components of mitotic microtubule organizing centers. Proceedings of the
National Academy of Sciences U S A 81, pages 4439-4443.
50.
Bailly, E., Pines, J., Hunter, T., and Bornens, M. (1992). Cytoplasmic accumulation
of cyclin B1 in human cells: association with a detergent-resistant compartment
and with the centrosome. Journal of Cell Science 101 ( Pt 3), pages 529-545.
51.
Peters, J.M. (2002). The anaphase-promoting complex: proteolysis in mitosis and
beyond. Molecular Cell 9, pages 931-943.
52.
Culotti, J., and Hartwell, L.H. (1971). Genetic control of the cell division cycle in
yeast. 3. Seven genes controlling nuclear division. Experimental Cell Research 67,
pages 389-401.
53.
Hartwell, L.H. (1971). Genetic control of the cell division cycle in yeast. IV. Genes
controlling bud emergence and cytokinesis. Experimental Cell Research 69, pages
265-276.
54.
Hartwell, L.H. (1971). Genetic control of the cell division cycle in yeast. II. Genes
controlling DNA replication and its initiation. Journal of Molecular Biology 59,
pages 183-194.
55.
Hartwell, L.H., Culotti, J., and Reid, B. (1970). Genetic control of the cell-division
cycle in yeast. I. Detection of mutants. Proceedings of the National Academy of
Sciences U S A 66, pages 352-359.
188
56.
Hartwell, L.H., Mortimer, R.K., Culotti, J., and Culotti, M. (1973). Genetic Control
of the Cell Division Cycle in Yeast: V. Genetic Analysis of cdc Mutants. Genetics
74, pages 267-286.
57.
Cross, F.R. (1988). DAF1, a mutant gene affecting size control, pheromone arrest,
and cell cycle kinetics of Saccharomyces cerevisiae. Molecular and Cellular
Biology 8, pages 4675-4684.
58.
Nash, R., Tokiwa, G., Anand, S., Erickson, K., and Futcher, A.B. (1988). The WHI1+
gene of Saccharomyces cerevisiae tethers cell division to cell size and is a cyclin
homolog. EMBO Journal 7, pages 4335-4346.
59.
Hadwiger, J.A., Wittenberg, C., Richardson, H.E., de Barros Lopes, M., and Reed,
S.I. (1989). A family of cyclin homologs that control the G1 phase in yeast.
Proceedings of the National Academy of Sciences U S A 86, pages 6255-6259.
60.
Cross, F.R. (1990). Cell cycle arrest caused by CLN gene deficiency in
Saccharomyces cerevisiae resembles START-I arrest and is independent of the
mating-pheromone signalling pathway. Molecular and Cellular Biology 10, pages
6482-6490.
61.
Richardson, H.E., Wittenberg, C., Cross, F., and Reed, S.I. (1989). An essential G1
function for cyclin-like proteins in yeast. Cell 59, pages 1127-1133.
62.
Lew, D.J., and Reed, S.I. (1993). Morphogenesis in the yeast cell cycle: regulation
by Cdc28 and cyclins. Journal of Cell Biology 120, pages 1305-1320.
63.
Schwob, E., and Nasmyth, K. (1993). CLB5 and CLB6, a new pair of B cyclins
involved in DNA replication in Saccharomyces cerevisiae. Genes and
Development 7, pages 1160-1175.
64.
Epstein, C.B., and Cross, F.R. (1992). CLB5: a novel B cyclin from budding yeast
with a role in S phase. Genes and Development 6, pages 1695-1706.
65.
Fitch, I., Dahmann, C., Surana, U., Amon, A., Nasmyth, K., Goetsch, L., Byers, B.,
and Futcher, B. (1992). Characterization of four B-type cyclin genes of the
budding yeast Saccharomyces cerevisiae. Molecular Biology of the Cell 3, pages
805-818.
189
66.
Richardson, H., Lew, D.J., Henze, M., Sugimoto, K., and Reed, S.I. (1992). Cyclin-B
homologs in Saccharomyces cerevisiae function in S phase and in G2. Genes and
Development 6, pages 2021-2034.
67.
Haase, S.B., Winey, M., and Reed, S.I. (2001). Multi-step control of spindle pole
body duplication by cyclin-dependent kinase. Nature Cell Biology 3, pages 38-42.
68.
Haase, S.B., and Reed, S.I. (1999). Evidence that a free-running oscillator drives
G1 events in the budding yeast cell cycle. Nature 401, pages 394-397.
69.
Amon, A., Tyers, M., Futcher, B., and Nasmyth, K. (1993). Mechanisms that help
the yeast cell cycle clock tick: G2 cyclins transcriptionally activate G2 cyclins and
repress G1 cyclins. Cell 74, pages 993-1007.
70.
Dirick, L., Moll, T., Auer, H., and Nasmyth, K. (1992). A central role for SWI6 in
modulating cell cycle Start-specific transcription in yeast. Nature 357, pages 508513.
71.
Primig, M., Sockanathan, S., Auer, H., and Nasmyth, K. (1992). Anatomy of a
transcription factor important for the start of the cell cycle in Saccharomyces
cerevisiae. Nature 358, pages 593-597.
72.
Andrews, B.J., and Herskowitz, I. (1990). Regulation of cell cycle-dependent gene
expression in yeast. Journal of Biological Chemistry 265, pages 14057-14060.
73.
Costanzo, M., Nishikawa, J.L., Tang, X., Millman, J.S., Schub, O., Breitkreuz, K.,
Dewar, D., Rupes, I., Andrews, B., and Tyers, M. (2004). CDK activity antagonizes
Whi5, an inhibitor of G1/S transcription in yeast. Cell 117, pages 899-913.
74.
de Bruin, R.A., McDonald, W.H., Kalashnikova, T.I., Yates, J., 3rd, and Wittenberg,
C. (2004). Cln3 activates G1-specific transcription via phosphorylation of the SBF
bound repressor Whi5. Cell 117, pages 887-898.
75.
Wang, H., Carey, L.B., Cai, Y., Wijnen, H., and Futcher, B. (2009). Recruitment of
Cln3 Cyclin to Promoters Controls Cell Cycle Entry via Histone Deacetylase and
Other Targets. PLoS Biology 7, pages e1000189.
76.
Ogas, J., Andrews, B.J., and Herskowitz, I. (1991). Transcriptional activation of
CLN1, CLN2, and a putative new G1 cyclin (HCS26) by SWI4, a positive regulator
of G1-specific transcription. Cell 66, pages 1015-1026.
190
77.
Iyer, V.R., Horak, C.E., Scafe, C.S., Botstein, D., Snyder, M., and Brown, P.O.
(2001). Genomic binding sites of the yeast cell-cycle transcription factors SBF and
MBF. Nature 409, pages 533-538.
78.
Wittenberg, C., and Reed, S.I. (2005). Cell cycle-dependent transcription in yeast:
promoters, transcription factors, and transcriptomes. Oncogene 24, pages 27462755.
79.
Harbison, C.T., Gordon, D.B., Lee, T.I., Rinaldi, N.J., Macisaac, K.D., Danford, T.W.,
Hannett, N.M., Tagne, J.B., Reynolds, D.B., Yoo, J., et al. (2004). Transcriptional
regulatory code of a eukaryotic genome. Nature 431, pages 99-104.
80.
Simon, I., Barnett, J., Hannett, N., Harbison, C.T., Rinaldi, N.J., Volkert, T.L.,
Wyrick, J.J., Zeitlinger, J., Gifford, D.K., Jaakkola, T.S., and Young, R.A. (2001).
Serial regulation of transcriptional regulators in the yeast cell cycle. Cell 106,
pages 697-708.
81.
Bean, J.M., Siggia, E.D., and Cross, F.R. (2005). High functional overlap between
MluI cell-cycle box binding factor and Swi4/6 cell-cycle box binding factor in the
G1/S transcriptional program in Saccharomyces cerevisiae. Genetics 171, pages
49-61.
82.
Koch, C., Moll, T., Neuberg, M., Ahorn, H., and Nasmyth, K. (1993). A role for the
transcription factors Mbp1 and Swi4 in progression from G1 to S phase. Science
261, pages 1551-1557.
83.
Skotheim, J.M., Di Talia, S., Siggia, E.D., and Cross, F.R. (2008). Positive feedback
of G1 cyclins ensures coherent cell cycle entry. Nature 454, pages 291-296.
84.
Dirick, L., and Nasmyth, K. (1991). Positive feedback in the activation of G1
cyclins in yeast. Nature 351, pages 754-757.
85.
Stuart, D., and Wittenberg, C. (1995). CLN3, not positive feedback, determines
the timing of CLN2 transcription in cycling cells. Genes and Development 9, pages
2780-2794.
86.
Carey, L.B., Leatherwood, J.K., and Futcher, B. (2008). Huxley's revenge: cell-cycle
entry, positive feedback, and the G1 cyclins. Molecular Cell 31, pages 307-308.
191
87.
Byers, B., and Goetsch, L. (1974). Duplication of spindle plaques and integration
of the yeast cell cycle. Cold Spring Harbor Symposium Quantitative Biology 38,
pages 123-131.
88.
Byers, B., and Goetsch, L. (1975). Behavior of spindles and spindle plaques in the
cell cycle and conjugation of Saccharomyces cerevisiae. Journal of Bacteriology
124, pages 511-523.
89.
Jaspersen, S.L., Huneycutt, B.J., Giddings, T.H., Jr., Resing, K.A., Ahn, N.G., and
Winey, M. (2004). Cdc28/Cdk1 regulates spindle pole body duplication through
phosphorylation of Spc42 and Mps1. Developmental Cell 7, pages 263-274.
90.
Amon, A., Irniger, S., and Nasmyth, K. (1994). Closing the cell cycle circle in yeast:
G2 cyclin proteolysis initiated at mitosis persists until the activation of G1 cyclins
in the next cycle. Cell 77, pages 1037-1050.
91.
Zachariae, W., and Nasmyth, K. (1999). Whose end is destruction: cell division
and the anaphase-promoting complex. Genes and Development 13, pages 20392058.
92.
Mendenhall, M.D. (1993). An inhibitor of p34CDC28 protein kinase activity from
Saccharomyces cerevisiae. Science 259, pages 216-219.
93.
Schwob, E., Bohm, T., Mendenhall, M.D., and Nasmyth, K. (1994). The B-type
cyclin kinase inhibitor p40SIC1 controls the G1 to S transition in S. cerevisiae. Cell
79, pages 233-244.
94.
Zachariae, W., Schwab, M., Nasmyth, K., and Seufert, W. (1998). Control of cyclin
ubiquitination by CDK-regulated binding of Hct1 to the anaphase promoting
complex. Science 282, pages 1721-1724.
95.
Nash, P., Tang, X., Orlicky, S., Chen, Q., Gertler, F.B., Mendenhall, M.D., Sicheri,
F., Pawson, T., and Tyers, M. (2001). Multisite phosphorylation of a CDK inhibitor
sets a threshold for the onset of DNA replication. Nature 414, pages 514-521.
96.
Feldman, R.M., Correll, C.C., Kaplan, K.B., and Deshaies, R.J. (1997). A complex of
Cdc4p, Skp1p, and Cdc53p/cullin catalyzes ubiquitination of the phosphorylated
CDK inhibitor Sic1p. Cell 91, pages 221-230.
192
97.
Verma, R., Annan, R.S., Huddleston, M.J., Carr, S.A., Reynard, G., and Deshaies,
R.J. (1997). Phosphorylation of Sic1p by G1 Cdk required for its degradation and
entry into S phase. Science 278, pages 455-460.
98.
Jackson, L.P., Reed, S.I., and Haase, S.B. (2006). Distinct mechanisms control the
stability of the related S-phase cyclins Clb5 and Clb6. Molecular and Cellular
Biology 26, pages 2456-2466.
99.
Orlando, D.A., Lin, C.Y., Bernard, A., Wang, J.Y., Socolar, J.E., Iversen, E.S.,
Hartemink, A.J., and Haase, S.B. (2008). Global control of cell-cycle transcription
by coupled CDK and network oscillators. Nature 453, pages 944-947.
100.
de Bruin, R.A., Kalashnikova, T.I., Chahwan, C., McDonald, W.H., Wohlschlegel, J.,
Yates, J., 3rd, Russell, P., and Wittenberg, C. (2006). Constraining G1-specific
transcription to late G1 phase: the MBF-associated corepressor Nrm1 acts via
negative feedback. Molecular Cell 23, pages 483-496.
101.
Cho, R.J., Campbell, M.J., Winzeler, E.A., Steinmetz, L., Conway, A., Wodicka, L.,
Wolfsberg, T.G., Gabrielian, A.E., Landsman, D., Lockhart, D.J., and Davis, R.W.
(1998). A genome-wide transcriptional analysis of the mitotic cell cycle.
Molecular Cell 2, pages 65-73.
102.
Spellman, P.T., Sherlock, G., Zhang, M.Q., Iyer, V.R., Anders, K., Eisen, M.B.,
Brown, P.O., Botstein, D., and Futcher, B. (1998). Comprehensive identification
of cell cycle-regulated genes of the yeast Saccharomyces cerevisiae by
microarray hybridization. Molecular Biology of the Cell 9, pages 3273-3297.
103.
Lydall, D., Ammerer, G., and Nasmyth, K. (1991). A new role for MCM1 in yeast:
cell cycle regulation of SW15 transcription. Genes and Development 5, pages
2405-2419.
104.
Reynolds, D., Shi, B.J., McLean, C., Katsis, F., Kemp, B., and Dalton, S. (2003).
Recruitment of Thr 319-phosphorylated Ndd1p to the FHA domain of Fkh2p
requires Clb kinase activity: a mechanism for CLB cluster gene activation. Genes
and Development 17, pages 1789-1802.
105.
Koranda, M., Schleiffer, A., Endler, L., and Ammerer, G. (2000). Forkhead-like
transcription factors recruit Ndd1 to the chromatin of G2/M-specific promoters.
Nature 406, pages 94-98.
193
106.
Loy, C.J., Lydall, D., and Surana, U. (1999). NDD1, a high-dosage suppressor of
cdc28-1N, is essential for expression of a subset of late-S-phase-specific genes in
Saccharomyces cerevisiae. Molecular and Cellular Biology 19, pages 3312-3327.
107.
Rudner, A.D., Hardwick, K.G., and Murray, A.W. (2000). Cdc28 activates exit from
mitosis in budding yeast. Journal of Cell Biology 149, pages 1361-1376.
108.
Rudner, A.D., and Murray, A.W. (2000). Phosphorylation by Cdc28 activates the
Cdc20-dependent activity of the anaphase-promoting complex. Journal of Cell
Biology 149, pages 1377-1390.
109.
Cohen-Fix, O., Peters, J.M., Kirschner, M.W., and Koshland, D. (1996). Anaphase
initiation in Saccharomyces cerevisiae is controlled by the APC-dependent
degradation of the anaphase inhibitor Pds1p. Genes and Development 10, pages
3081-3093.
110.
Ciosk, R., Zachariae, W., Michaelis, C., Shevchenko, A., Mann, M., and Nasmyth,
K. (1998). An ESP1/PDS1 complex regulates loss of sister chromatid cohesion at
the metaphase to anaphase transition in yeast. Cell 93, pages 1067-1076.
111.
Uhlmann, F. (2000). Chromosome cohesion: a polymerase for chromosome
bridges. Current Biology 10, pages R698-700.
112.
Lim, H.H., Goh, P.Y., and Surana, U. (1998). Cdc20 is essential for the cyclosomemediated proteolysis of both Pds1 and Clb2 during M phase in budding yeast.
Current Biology 8, pages 231-234.
113.
Yeong, F.M., Lim, H.H., Padmashree, C.G., and Surana, U. (2000). Exit from
mitosis in budding yeast: biphasic inactivation of the Cdc28-Clb2 mitotic kinase
and the role of Cdc20. Molecular Cell 5, pages 501-511.
114.
Stegmeier, F., and Amon, A. (2004). Closing mitosis: the functions of the Cdc14
phosphatase and its regulation. Annual Reviews in Genetics 38, pages 203-232.
115.
Visintin, R., Craig, K., Hwang, E.S., Prinz, S., Tyers, M., and Amon, A. (1998). The
phosphatase Cdc14 triggers mitotic exit by reversal of Cdk-dependent
phosphorylation. Molecular Cell 2, pages 709-718.
116.
Traverso, E.E., Baskerville, C., Liu, Y., Shou, W., James, P., Deshaies, R.J., and
Charbonneau, H. (2001). Characterization of the Net1 cell cycle-dependent
194
regulator of the Cdc14 phosphatase from budding yeast. Journal of Biological
Chemistry 276, pages 21924-21931.
117.
Visintin, R., Hwang, E.S., and Amon, A. (1999). Cfi1 prevents premature exit from
mitosis by anchoring Cdc14 phosphatase in the nucleolus. Nature 398, pages
818-823.
118.
Stegmeier, F., Visintin, R., and Amon, A. (2002). Separase, polo kinase, the
kinetochore protein Slk19, and Spo12 function in a network that controls Cdc14
localization during early anaphase. Cell 108, pages 207-220.
119.
Bardin, A.J., and Amon, A. (2001). Men and sin: what's the difference? Nature
Reviews Molecular Cell Biology 2, pages 815-826.
120.
Jaspersen, S.L., Charles, J.F., and Morgan, D.O. (1999). Inhibitory phosphorylation
of the APC regulator Hct1 is controlled by the kinase Cdc28 and the phosphatase
Cdc14. Current Biology 9, pages 227-236.
121.
Wasch, R., and Cross, F.R. (2002). APC-dependent proteolysis of the mitotic
cyclin Clb2 is essential for mitotic exit. Nature 418, pages 556-562.
122.
Knapp, D., Bhoite, L., Stillman, D.J., and Nasmyth, K. (1996). The transcription
factor Swi5 regulates expression of the cyclin kinase inhibitor p40SIC1. Molecular
Cell Biology 16, pages 5701-5707.
123.
Moll, T., Tebb, G., Surana, U., Robitsch, H., and Nasmyth, K. (1991). The role of
phosphorylation and the CDC28 protein kinase in cell cycle-regulated nuclear
import of the S. cerevisiae transcription factor SWI5. Cell 66, pages 743-758.
124.
Toyn, J.H., Johnson, A.L., Donovan, J.D., Toone, W.M., and Johnston, L.H. (1997).
The Swi5 transcription factor of Saccharomyces cerevisiae has a role in exit from
mitosis through induction of the cdk-inhibitor Sic1 in telophase. Genetics 145,
pages 85-96.
125.
Queralt, E., and Uhlmann, F. (2008). Cdk-counteracting phosphatases unlock
mitotic exit. Current Opinion in Cell Biology 20, pages 661-668.
126.
Hartwell, L.H., and Weinert, T.A. (1989). Checkpoints: controls that ensure the
order of cell cycle events. Science 246, pages 629-634.
195
127.
Tang, Z., Shu, H., Oncel, D., Chen, S., and Yu, H. (2004). Phosphorylation of Cdc20
by Bub1 provides a catalytic mechanism for APC/C inhibition by the spindle
checkpoint. Molecular Cell 16, pages 387-397.
128.
Burton, J.L., and Solomon, M.J. (2007). Mad3p, a pseudosubstrate inhibitor of
APCCdc20 in the spindle assembly checkpoint. Genes and Development 21, pages
655-667.
129.
Yu, H. (2002). Regulation of APC-Cdc20 by the spindle checkpoint. Current
Opinion in Cell Biology 14, pages 706-714.
130.
Elledge, S.J. (1996). Cell cycle checkpoints: preventing an identity crisis. Science
274, pages 1664-1672.
131.
Wang, H., Liu, D., Wang, Y., Qin, J., and Elledge, S.J. (2001). Pds1 phosphorylation
in response to DNA damage is essential for its DNA damage checkpoint function.
Genes and Development 15, pages 1361-1372.
132.
Clarke, D.J., Segal, M., Andrews, C.A., Rudyak, S.G., Jensen, S., Smith, K., and
Reed, S.I. (2003). S-phase checkpoint controls mitosis via an APC-independent
Cdc20p function. Nature Cell Biology 5, pages 928-935.
133.
Clarke, D.J., Segal, M., Jensen, S., and Reed, S.I. (2001). Mec1p regulates Pds1p
levels in S phase: complex coordination of DNA replication and mitosis. Nature
Cell Biology 3, pages 619-627.
134.
Searle, J.S., Schollaert, K.L., Wilkins, B.J., and Sanchez, Y. (2004). The DNA
damage checkpoint and PKA pathways converge on APC substrates and Cdc20 to
regulate mitotic progression. Nature Cell Biology 6, pages 138-145.
135.
McGowan, C.H., and Russell, P. (2004). The DNA damage response: sensing and
signaling. Current Opinion in Cell Biology 16, pages 629-633.
136.
Lew, D.J., and Reed, S.I. (1995). A cell cycle checkpoint monitors cell
morphogenesis in budding yeast. Journal of Cell Biology 129, pages 739-749.
137.
Lew, D.J. (2003). The morphogenesis checkpoint: how yeast cells watch their
figures. Current Opinion in Cell Biology 15, pages 648-653.
138.
Keaton, M.A., Bardes, E.S., Marquitz, A.R., Freel, C.D., Zyla, T.R., Rudolph, J., and
Lew, D.J. (2007). Differential susceptibility of yeast S and M phase CDK
196
complexes to inhibitory tyrosine phosphorylation. Current Biology 17, pages
1181-1189.
139.
Sia, R.A., Bardes, E.S., and Lew, D.J. (1998). Control of Swe1p degradation by the
morphogenesis checkpoint. EMBO Journal 17, pages 6678-6688.
140.
Chen, K.C., Calzone, L., Csikasz-Nagy, A., Cross, F.R., Novak, B., and Tyson, J.J.
(2004). Integrative analysis of cell cycle control in budding yeast. Molecular
Biology of the Cell 15, pages 3841-3862.
141.
Cross, F.R. (2003). Two redundant oscillatory mechanisms in the yeast cell cycle.
Developmental Cell 4, pages 741-752.
142.
Tyson, J.J., Chen, K.C., and Novak, B. (2003). Sniffers, buzzers, toggles and
blinkers: dynamics of regulatory and signaling pathways in the cell. Current
Opinion in Cell Biology 15, pages 221-231.
143.
Hartwell, L.H., Culotti, J., Pringle, J.R., and Reid, B.J. (1974). Genetic control of the
cell division cycle in yeast. Science 183, pages 46-51.
144.
Dumas, L.B., Lussky, J.P., McFarland, E.J., and Shampay, J. (1982). New
temperature-sensitive mutants of Saccharomyces cerevisiae affecting DNA
replication. Molecular General Genetics 187, pages 42-46.
145.
Goebl, M.G., Yochem, J., Jentsch, S., McGrath, J.P., Varshavsky, A., and Byers, B.
(1988). The yeast cell cycle gene CDC34 encodes a ubiquitin-conjugating enzyme.
Science 241, pages 1331-1335.
146.
Mathias, N., Johnson, S.L., Winey, M., Adams, A.E., Goetsch, L., Pringle, J.R.,
Byers, B., and Goebl, M.G. (1996). Cdc53p acts in concert with Cdc4p and Cdc34p
to control the G1-to-S-phase transition and identifies a conserved family of
proteins. Molecular Cell Biology 16, pages 6634-6643.
147.
Peter, M., Gartner, A., Horecka, J., Ammerer, G., and Herskowitz, I. (1993). FAR1
links the signal transduction pathway to the cell cycle machinery in yeast. Cell 73,
pages 747-760.
148.
Bucking-Throm, E., Duntze, W., Hartwell, L.H., and Manney, T.R. (1973).
Reversible arrest of haploid yeast cells in the initiation of DNA synthesis by a
diffusible sex factor. Experimental Cell Research 76, pages 99-110.
197
149.
Pramila, T., Wu, W., Miles, S., Noble, W.S., and Breeden, L.L. (2006). The
Forkhead transcription factor Hcm1 regulates chromosome segregation genes
and fills the S-phase gap in the transcriptional circuitry of the cell cycle. Genes
and Development 20, pages 2266-2278.
150.
Lee, T.I., Rinaldi, N.J., Robert, F., Odom, D.T., Bar-Joseph, Z., Gerber, G.K.,
Hannett, N.M., Harbison, C.T., Thompson, C.M., Simon, I., et al. (2002).
Transcriptional regulatory networks in Saccharomyces cerevisiae. Science 298,
pages 799-804.
151.
Ganem, N.J., Godinho, S.A., and Pellman, D. (2009). A mechanism linking extra
centrosomes to chromosomal instability. Nature 460, pages 278-282.
152.
Lingle, W.L., Barrett, S.L., Negron, V.C., D'Assoro, A.B., Boeneman, K., Liu, W.,
Whitehead, C.M., Reynolds, C., and Salisbury, J.L. (2002). Centrosome
amplification drives chromosomal instability in breast tumor development.
Proceedings of the National Academy of Sciences U S A 99, pages 1978-1983.
153.
Delattre, M., and Gonczy, P. (2004). The arithmetic of centrosome biogenesis.
Journal Cell Science 117, pages 1619-1630.
154.
Jaspersen, S.L., and Winey, M. (2004). The budding yeast spindle pole body:
structure, duplication, and function. Annual Review of Cell and Developmental
Biology 20, pages 1-28.
155.
Novak, B., and Tyson, J.J. (2008). Design principles of biochemical oscillators.
Nature Reviews Molecular and Cellular Biology 9, pages 981-991.
156.
Elowitz, M.B., and Leibler, S. (2000). A synthetic oscillatory network of
transcriptional regulators. Nature 403, pages 335-338.
157.
Feldman, J.F., and Hoyle, M.N. (1973). Isolation of circadian clock mutants of
Neurospora crassa. Genetics 75, pages 605-613.
158.
Konopka, R.J., and Benzer, S. (1971). Clock mutants of Drosophila melanogaster.
Proceedings of the National Academy of Sciences U S A 68, pages 2112-2116.
159.
Dunlap, J.C. (1993). Genetic analysis of circadian clocks. Annual Reviews in
Physiology 55, pages 683-728.
198
160.
Dunlap, J.C. (1998). Common threads in eukaryotic circadian systems. Current
Opinion in Genetics and Development 8, pages 400-406.
161.
Hoffmann, I., Clarke, P.R., Marcote, M.J., Karsenti, E., and Draetta, G. (1993).
Phosphorylation and activation of human cdc25-C by cdc2--cyclin B and its
involvement in the self-amplification of MPF at mitosis. EMBO Journal 12, pages
53-63.
162.
Izumi, T., Walker, D.H., and Maller, J.L. (1992). Periodic changes in
phosphorylation of the Xenopus cdc25 phosphatase regulate its activity.
Molecular Biology of the Cell 3, pages 927-939.
163.
Kumagai, A., and Dunphy, W.G. (1992). Regulation of the cdc25 protein during
the cell cycle in Xenopus extracts. Cell 70, pages 139-151.
164.
Mueller, P.R., Coleman, T.R., and Dunphy, W.G. (1995). Cell cycle regulation of a
Xenopus Wee1-like kinase. Molecular Biology of the Cell 6, pages 119-134.
165.
Tang, Z., Coleman, T.R., and Dunphy, W.G. (1993). Two distinct mechanisms for
negative regulation of the Wee1 protein kinase. EMBO Journal 12, pages 34273436.
166.
Pomerening, J.R., Sontag, E.D., and Ferrell, J.E., Jr. (2003). Building a cell cycle
oscillator: hysteresis and bistability in the activation of Cdc2. Nature Cell Biology
5, pages 346-351.
167.
Holtzendorff, J., Reinhardt, J., and Viollier, P.H. (2006). Cell cycle control by
oscillating regulatory proteins in Caulobacter crescentus. Bioessays 28, pages
355-361.
168.
Marczynski, G.T. (1999). Chromosome methylation and measurement of faithful,
once and only once per cell cycle chromosome replication in Caulobacter
crescentus. Journal of Bacteriology 181, pages 1984-1993.
169.
Shapiro, L. (1976). Differentiation in the Caulobacter cell cycle. Annual Reviews in
Microbiology 30, pages 377-407.
170.
Holtzendorff, J., Hung, D., Brende, P., Reisenauer, A., Viollier, P.H., McAdams,
H.H., and Shapiro, L. (2004). Oscillating global regulators control the genetic
circuit driving a bacterial cell cycle. Science 304, pages 983-987.
199
171.
Hottes, A.K., Shapiro, L., and McAdams, H.H. (2005). DnaA coordinates
replication initiation and cell cycle transcription in Caulobacter crescentus.
Molecular Microbiology 58, pages 1340-1353.
172.
Zweiger, G., and Shapiro, L. (1994). Expression of Caulobacter dnaA as a function
of the cell cycle. Journal of Bacteriology 176, pages 401-408.
173.
Simmons Kovacs, L.A., Orlando, D.A., and Haase, S.B. (2008). Transcription
networks and cyclin/CDKs: the yin and yang of cell cycle oscillators. Cell Cycle 7,
pages 2626-2629.
174.
Drury, L.S., Perkins, G., and Diffley, J.F. (1997). The Cdc4/34/53 pathway targets
Cdc6p for proteolysis in budding yeast. EMBO Journal 16, pages 5966-5976.
175.
Nguyen, V.Q., Co, C., Irie, K., and Li, J.J. (2000). Clb/Cdc28 kinases promote
nuclear export of the replication initiator proteins Mcm2-7. Current Biology 10,
pages 195-205.
176.
Nguyen, V.Q., Co, C., and Li, J.J. (2001). Cyclin-dependent kinases prevent DNA
re-replication through multiple mechanisms. Nature 411, pages 1068-1073.
177.
Botchan, M. (1996). Coordinating DNA replication with cell division: current
status of the licensing concept. Proceedings of the National Academy of Sciences
U S A 93, pages 9997-10000.
178.
Chevalier, S., and Blow, J.J. (1996). Cell cycle control of replication initiation in
eukaryotes. Current Opinion in Cell Biology 8, pages 815-821.
179.
Ghadimi, B.M., Sackett, D.L., Difilippantonio, M.J., Schrock, E., Neumann, T.,
Jauho, A., Auer, G., and Ried, T. (2000). Centrosome amplification and instability
occurs exclusively in anneuploid, but not in diploid colorectal cancer cell lines,
and correlates with numerical chromosomal aberrations. Genes Chromosomes
and Cancer 27, pages 183-190.
180.
Giehl, M., Fabarius, A., Frank, O., Hochhaus, A., Hafner, M., Hehlmann, R., and
Seifarth, W. (2005). Centrosome aberrations in chronic myeloid leukemia
correlate with stage of disease and chromosomal instability. Leukemia 19, pages
1192-1197.
200
181.
Hsu, L.C., Kapali, M., DeLoia, J.A., and Gallion, H.H. (2005). Centrosome
abnormalities in ovarian cancer. International Journal of Cancer 113, pages 746751.
182.
Ventura, R.A., Martin-Subero, J.I., Knippschild, U., Gascoyne, R.D., Delsol, G.,
Mason, D.Y., and Siebert, R. (2004). Centrosome abnormalities in ALK-positive
anaplastic large-cell lymphoma. Leukemia 18, pages 1910-1911.
183.
Weber, R.G., Bridger, J.M., Benner, A., Weisenberger, D., Ehemann, V.,
Reifenberger, G., and Lichter, P. (1998). Centrosome amplification as a possible
mechanism for numerical chromosome aberrations in cerebral primitive
neuroectodermal tumors with TP53 mutations. Cytogenetics and Cell Genetics
83, pages 266-269.
184.
Yamamoto, Y., Matsuyama, H., Furuya, T., Oga, A., Yoshihiro, S., Okuda, M.,
Kawauchi, S., Sasaki, K., and Naito, K. (2004). Centrosome hyperamplification
predicts progression and tumor recurrence in bladder cancer. Clinical Cancer
Research 10, pages 6449-6455.
185.
Wang, Q., Hirohashi, Y., Furuuchi, K., Zhao, H., Liu, Q., Zhang, H., Murali, R.,
Berezov, A., Du, X., Li, B., and Greene, M.I. (2004). The centrosome in normal and
transformed cells. DNA and Cell Biology 23, pages 475-489.
186.
Meraldi, P., Lukas, J., Fry, A.M., Bartek, J., and Nigg, E.A. (1999). Centrosome
duplication in mammalian somatic cells requires E2F and Cdk2-cyclin A. Nature
Cell Biology 1, pages 88-93.
187.
Adams, I.R., and Kilmartin, J.V. (2000). Spindle pole body duplication: a model for
centrosome duplication? Trends in Cell Biology 10, pages 329-335.
188.
Nigg, E.A. (2002). Centrosome aberrations: cause or consequence of cancer
progression? Nature Reviews Cancer 2, pages 815-825.
189.
Hinchcliffe, E.H., Li, C., Thompson, E.A., Maller, J.L., and Sluder, G. (1999).
Requirement of Cdk2-cyclin E activity for repeated centrosome reproduction in
Xenopus egg extracts. Science 283, pages 851-854.
190.
Lacey, K.R., Jackson, P.K., and Stearns, T. (1999). Cyclin-dependent kinase control
of centrosome duplication. Proceedings of the National Academy of Sciences U S
A 96, pages 2817-2822.
201
191.
Vidwans, S.J., Wong, M.L., and O'Farrell, P.H. (1999). Mitotic regulators govern
progress through steps in the centrosome duplication cycle. Journal of Cell
Biology 147, pages 1371-1378.
192.
Vidwans, S.J., Wong, M.L., and O'Farrell, P.H. (2003). Anomalous centriole
configurations are detected in Drosophila wing disc cells upon Cdk1 inactivation.
Journal of Cell Science 116, pages 137-143.
193.
Tsou, M.F., and Stearns, T. (2006). Mechanism limiting centrosome duplication
to once per cell cycle. Nature 442, pages 947-951.
194.
Wong, C., and Stearns, T. (2003). Centrosome number is controlled by a
centrosome-intrinsic block to reduplication. Nature Cell Biology 5, pages 539544.
195.
Moens, P.B., and Rapport, E. (1971). Spindles, spindle plaques, and meiosis in the
yeast Saccharomyces cerevisiae (Hansen). Journal of Cell Biology 50, pages 344361.
196.
Chial, H.J., and Winey, M. (1999). Mechanisms of genetic instability revealed by
analysis of yeast spindle pole body duplication. Biology of the Cell 91, pages 439450.
197.
Adams, I.R., and Kilmartin, J.V. (1999). Localization of core spindle pole body
(SPB) components during SPB duplication in Saccharomyces cerevisiae. Journal of
Cell Biology 145, pages 809-823.
198.
Jacobs, C.W., Adams, A.E., Szaniszlo, P.J., and Pringle, J.R. (1988). Functions of
microtubules in the Saccharomyces cerevisiae cell cycle. Journal of Cell Biology
107, pages 1409-1426.
199.
Roof, D.M., Meluh, P.B., and Rose, M.D. (1992). Kinesin-related proteins required
for assembly of the mitotic spindle. Journal of Cell Biology 118, pages 95-108.
200.
Sobel, S.G., and Snyder, M. (1995). A highly divergent gamma-tubulin gene is
essential for cell growth and proper microtubule organization in Saccharomyces
cerevisiae. Journal of Cell Biology 131, pages 1775-1788.
201.
Geissler, S., Pereira, G., Spang, A., Knop, M., Soues, S., Kilmartin, J., and Schiebel,
E. (1996). The spindle pole body component Spc98p interacts with the gamma202
tubulin-like Tub4p of Saccharomyces cerevisiae at the sites of microtubule
attachment. EMBO Journal 15, pages 3899-3911.
202.
Knop, M., and Schiebel, E. (1997). Spc98p and Spc97p of the yeast gammatubulin complex mediate binding to the spindle pole body via their interaction
with Spc110p. EMBO Journal 16, pages 6985-6995.
203.
Knop, M., Pereira, G., Geissler, S., Grein, K., and Schiebel, E. (1997). The spindle
pole body component Spc97p interacts with the gamma-tubulin of
Saccharomyces cerevisiae and functions in microtubule organization and spindle
pole body duplication. EMBO Journal 16, pages 1550-1564.
204.
Hoepfner, D., Schaerer, F., Brachat, A., Wach, A., and Philippsen, P. (2002).
Reorientation of mispositioned spindles in short astral microtubule mutant
spc72Δ is dependent on spindle pole body outer plaque and Kar3 motor protein.
Molecular Biology of the Cell 13, pages 1366-1380.
205.
Gruneberg, U., Campbell, K., Simpson, C., Grindlay, J., and Schiebel, E. (2000).
Nud1p links astral microtubule organization and the control of exit from mitosis.
EMBO Journal 19, pages 6475-6488.
206.
Brachat, A., Kilmartin, J.V., Wach, A., and Philippsen, P. (1998). Saccharomyces
cerevisiae cells with defective spindle pole body outer plaques accomplish
nuclear migration via half-bridge-organized microtubules. Molecular Biology of
the Cell 9, pages 977-991.
207.
Donaldson, A.D., and Kilmartin, J.V. (1996). Spc42p: a phosphorylated
component of the S. cerevisiae spindle pole body (SPB) with an essential function
during SPB duplication. Journal of Cell Biology 132, pages 887-901.
208.
Elliott, S., Knop, M., Schlenstedt, G., and Schiebel, E. (1999). Spc29p is a
component of the Spc110p subcomplex and is essential for spindle pole body
duplication. Proceedings of the National Academy of Sciences U S A 96, pages
6205-6210.
209.
Cyert, M.S. (2001). Genetic analysis of calmodulin and its targets in
Saccharomyces cerevisiae. Annual Reviews in Genetics 35, pages 647-672.
210.
Sun, G.H., Hirata, A., Ohya, Y., and Anraku, Y. (1992). Mutations in yeast
calmodulin cause defects in spindle pole body functions and nuclear integrity.
Journal of Cell Biology 119, pages 1625-1639.
203
211.
Kilmartin, J.V., and Goh, P.Y. (1996). Spc110p: assembly properties and role in
the connection of nuclear microtubules to the yeast spindle pole body. EMBO
Journal 15, pages 4592-4602.
212.
Winey, M., Hoyt, M.A., Chan, C., Goetsch, L., Botstein, D., and Byers, B. (1993).
NDC1: a nuclear periphery component required for yeast spindle pole body
duplication. Journal of Cell Biology 122, pages 743-751.
213.
Winey, M., Goetsch, L., Baum, P., and Byers, B. (1991). MPS1 and MPS2: novel
yeast genes defining distinct steps of spindle pole body duplication. Journal of
Cell Biology 114, pages 745-754.
214.
Schramm, C., Elliott, S., Shevchenko, A., and Schiebel, E. (2000). The Bbp1pMps2p complex connects the SPB to the nuclear envelope and is essential for
SPB duplication. EMBO Journal 19, pages 421-433.
215.
Rose, M.D., and Fink, G.R. (1987). KAR1, a gene required for function of both
intranuclear and extranuclear microtubules in yeast. Cell 48, pages 1047-1060.
216.
Jaspersen, S.L., Giddings, T.H., Jr., and Winey, M. (2002). Mps3p is a novel
component of the yeast spindle pole body that interacts with the yeast centrin
homologue Cdc31p. Journal of Cell Biology 159, pages 945-956.
217.
Spang, A., Courtney, I., Grein, K., Matzner, M., and Schiebel, E. (1995). The
Cdc31p-binding protein Kar1p is a component of the half bridge of the yeast
spindle pole body. Journal of Cell Biology 128, pages 863-877.
218.
Kilmartin, J.V. (2003). Sfi1p has conserved centrin-binding sites and an essential
function in budding yeast spindle pole body duplication. Journal of Cell Biology
162, pages 1211-1221.
219.
Dunphy, W.G., and Newport, J.W. (1988). Unraveling of mitotic control
mechanisms. Cell 55, pages 925-928.
220.
Murray, A.W. (2004). Recycling the cell cycle: cyclins revisited. Cell 116, pages
221-234.
221.
Fung, T.K., and Poon, R.Y. (2005). A roller coaster ride with the mitotic cyclins.
Seminars in Cell and Developmental Biology 16, pages 335-342.
204
222.
Lanker, S., Valdivieso, M.H., and Wittenberg, C. (1996). Rapid degradation of the
G1 cyclin Cln2 induced by CDK-dependent phosphorylation. Science 271, pages
1597-1601.
223.
Cross, F.R., and Tinkelenberg, A.H. (1991). A potential positive feedback loop
controlling CLN1 and CLN2 gene expression at the start of the yeast cell cycle.
Cell 65, pages 875-883.
224.
Cvrckova, F., and Nasmyth, K. (1993). Yeast G1 cyclins CLN1 and CLN2 and a GAPlike protein have a role in bud formation. EMBO Journal 12, pages 5277-5286.
225.
Orlando, D.A., Lin, C.Y., Bernard, A., Iversen, E.S., Hartemink, A.J., and Haase, S.B.
(2007). A probabilistic model for cell cycle distributions in synchrony
experiments. Cell Cycle 6, pages 478-488.
226.
Chenevert, J., Valtz, N., and Herskowitz, I. (1994). Identification of genes required
for normal pheromone-induced cell polarization in Saccharomyces cerevisiae.
Genetics 136, pages 1287-1296.
227.
Elion, E.A., Satterberg, B., and Kranz, J.E. (1993). FUS3 phosphorylates multiple
components of the mating signal transduction cascade: evidence for STE12 and
FAR1. Molecular Biology of the Cell 4, pages 495-510.
228.
Fields, S., and Herskowitz, I. (1985). The yeast STE12 product is required for
expression of two sets of cell-type specific genes. Cell 42, pages 923-930.
229.
Roberts, C.J., Nelson, B., Marton, M.J., Stoughton, R., Meyer, M.R., Bennett, H.A.,
He, Y.D., Dai, H., Walker, W.L., Hughes, T.R., Tyers, M., Boone, C., and Friend,
S.H. (2000). Signaling and circuitry of multiple MAPK pathways revealed by a
matrix of global gene expression profiles. Science 287, pages 873-880.
230.
Hilioti, Z., Sabbagh, W., Jr., Paliwal, S., Bergmann, A., Goncalves, M.D., Bardwell,
L., and Levchenko, A. (2008). Oscillatory phosphorylation of yeast Fus3 MAP
kinase controls periodic gene expression and morphogenesis. Current Biology 18,
pages 1700-1706.
231.
Chua, G., Morris, Q.D., Sopko, R., Robinson, M.D., Ryan, O., Chan, E.T., Frey, B.J.,
Andrews, B.J., Boone, C., and Hughes, T.R. (2006). Identifying transcription factor
functions and targets by phenotypic activation. Proceedings of the National
Academy of Sciences U S A 103, pages 12045-12050.
205
232.
Horak, C.E., Luscombe, N.M., Qian, J., Bertone, P., Piccirrillo, S., Gerstein, M., and
Snyder, M. (2002). Complex transcriptional circuitry at the G1/S transition in
Saccharomyces cerevisiae. Genes and Development 16, pages 3017-3033.
233.
McBride, H.J., Yu, Y., and Stillman, D.J. (1999). Distinct regions of the Swi5 and
Ace2 transcription factors are required for specific gene activation. Journal of
Biological Chemistry 274, pages 21029-21036.
234.
Pramila, T., Miles, S., GuhaThakurta, D., Jemiolo, D., and Breeden, L.L. (2002).
Conserved homeodomain proteins interact with MADS box protein Mcm1 to
restrict ECB-dependent transcription to the M/G1 phase of the cell cycle. Genes
and Development 16, pages 3034-3045.
235.
Tuch, B.B., Galgoczy, D.J., Hernday, A.D., Li, H., and Johnson, A.D. (2008). The
evolution of combinatorial gene regulation in fungi. PLoS Biology 6, pages e38.
236.
Voth, W.P., Yu, Y., Takahata, S., Kretschmann, K.L., Lieb, J.D., Parker, R.L., Milash,
B., and Stillman, D.J. (2007). Forkhead proteins control the outcome of
transcription factor binding by antiactivation. EMBO Journal 26, pages 43244334.
237.
Workman, C.T., Mak, H.C., McCuine, S., Tagne, J.B., Agarwal, M., Ozier, O.,
Begley, T.J., Samson, L.D., and Ideker, T. (2006). A systems approach to mapping
DNA damage response pathways. Science 312, pages 1054-1059.
238.
Sevim, V., Gong, X., and Socolar, J.E. Reliability of transcriptional cycles and the
yeast cell-cycle oscillator. manuscript in preparation.
239.
de Bruin, R.A., Kalashnikova, T.I., and Wittenberg, C. (2008). Stb1 collaborates
with other regulators to modulate the G1-specific transcriptional circuit.
Molecular and Cellular Biology 28, pages 6919-6928.
240.
Ubersax, J.A., Woodbury, E.L., Quang, P.N., Paraz, M., Blethrow, J.D., Shah, K.,
Shokat, K.M., and Morgan, D.O. (2003). Targets of the cyclin-dependent kinase
Cdk1. Nature 425, pages 859-864.
241.
Bishop, A.C., Ubersax, J.A., Petsch, D.T., Matheos, D.P., Gray, N.S., Blethrow, J.,
Shimizu, E., Tsien, J.Z., Schultz, P.G., Rose, M.D., Wood, J.L., Morgan, D.O., and
Shokat, K.M. (2000). A chemical switch for inhibitor-sensitive alleles of any
protein kinase. Nature 407, pages 395-401.
206
242.
Henchoz, S., Chi, Y., Catarin, B., Herskowitz, I., Deshaies, R.J., and Peter, M.
(1997). Phosphorylation- and ubiquitin-dependent degradation of the cyclindependent kinase inhibitor Far1p in budding yeast. Genes and Development 11,
pages 3046-3060.
243.
Bardwell, L. (2005). A walk-through of the yeast mating pheromone response
pathway. Peptides 26, pages 339-350.
244.
Lahav, R., Gammie, A., Tavazoie, S., and Rose, M.D. (2007). Role of transcription
factor Kar4 in regulating downstream events in the Saccharomyces cerevisiae
pheromone response pathway. Molecular and Cellular Biology 27, pages 818829.
245.
Gautier, L., Cope, L., Bolstad, B.M., and Irizarry, R.A. (2004). affy--analysis of
Affymetrix GeneChip data at the probe level. Bioinformatics 20, pages 307-315.
246.
Li, C., and Hung Wong, W. (2001). Model-based analysis of oligonucleotide
arrays: model validation, design issues and standard error application. Genome
Biology 2, pages RESEARCH0032.
247.
Dempster, A.P., Liard, N.M., and Rubin, D.B. (1977). Maximum likelihood from
incomplete data via the EM algorithm. Journal of the Royal Statistical Society 39,
pages 1-3.
248.
Do, C.B., and Batzoglou, S. (2008). What is the expectation maximization
algorithm? Nature Biotechnology 26, pages 897-899.
249.
Radivojac, P., Vacic, V., Haynes, C., Cocklin, R.R., Mohan, A., Heyen, J.W., Goebl,
M.G., and Iakoucheva, L.M. (2009). Identification, analysis, and prediction of
protein ubiquitination sites. Proteins.
250.
Bartel, B., Wunning, I., and Varshavsky, A. (1990). The recognition component of
the N-end rule pathway. EMBO Journal 9, pages 3179-3189.
251.
Kodadek, T., Sikder, D., and Nalley, K. (2006). Keeping transcriptional activators
under control. Cell 127, pages 261-264.
252.
Guo, X., Bernard, A., Orlando, D.A., Hartemink, A.J., and Haase, S.B. manuscript
in preparation for Proceedings of the National Academy of Sciences, USA.
207
253.
Jong, A.Y., Kuo, C.L., and Campbell, J.L. (1984). The CDC8 gene of yeast encodes
thymidylate kinase. Journal of Biological Chemistry 259, pages 11052-11059.
254.
Sclafani, R.A., and Fangman, W.L. (1984). Yeast gene CDC8 encodes thymidylate
kinase and is complemented by herpes thymidine kinase gene TK. Procedings of
the National Academy of Sciences U S A 81, pages 5821-5825.
255.
O'Conallain, C., Doolin, M.T., Taggart, C., Thornton, F., and Butler, G. (1999).
Regulated nuclear localisation of the yeast transcription factor Ace2p controls
expression of chitinase (CTS1) in Saccharomyces cerevisiae. Molecular and
General Genetics 262, pages 275-282.
256.
Longtine, M.S., McKenzie, A., 3rd, Demarini, D.J., Shah, N.G., Wach, A., Brachat,
A., Philippsen, P., and Pringle, J.R. (1998). Additional modules for versatile and
economical PCR-based gene deletion and modification in Saccharomyces
cerevisiae. Yeast 14, pages 953-961.
257.
Keaton, M.A., Szkotnicki, L., Marquitz, A.R., Harrison, J., Zyla, T.R., and Lew, D.J.
(2008). Nucleocytoplasmic trafficking of G2/M regulators in yeast. Molecular
Biology of the Cell 19, pages 4006-4018.
258.
Simmons Kovacs, L.A., Nelson, C.L., and Haase, S.B. (2008). Intrinsic and cyclindependent kinase-dependent control of spindle pole body duplication in budding
yeast. Molecular Biology of the Cell 19, pages 3243-3253.
259.
Pihan, G.A., Purohit, A., Wallace, J., Knecht, H., Woda, B., Quesenberry, P., and
Doxsey, S.J. (1998). Centrosome defects and genetic instability in malignant
tumors. Cancer Research 58, pages 3974-3985.
260.
D'Assoro, A.B., Lingle, W.L., and Salisbury, J.L. (2002). Centrosome amplification
and the development of cancer. Oncogene 21, pages 6146-6153.
261.
Balczon, R., Bao, L., Zimmer, W.E., Brown, K., Zinkowski, R.P., and Brinkley, B.R.
(1995). Dissociation of centrosome replication events from cycles of DNA
synthesis and mitotic division in hydroxyurea-arrested Chinese hamster ovary
cells. Journal of Cell Biology 130, pages 105-115.
262.
Gard, D.L., Hafezi, S., Zhang, T., and Doxsey, S.J. (1990). Centrosome duplication
continues in cycloheximide-treated Xenopus blastulae in the absence of a
detectable cell cycle. Journal of Cell Biology 110, pages 2033-2042.
208
263.
Sluder, G., and Lewis, K. (1987). Relationship between nuclear DNA synthesis and
centrosome reproduction in sea urchin eggs. Journal of Experimental Zoology
244, pages 89-100.
264.
Kuriyama, R., and Borisy, G.G. (1981). Centriole cycle in Chinese hamster ovary
cells as determined by whole-mount electron microscopy. Journal of Cell Biology
91, pages 814-821.
265.
Crasta, K., Huang, P., Morgan, G., Winey, M., and Surana, U. (2006). Cdk1
regulates centrosome separation by restraining proteolysis of microtubuleassociated proteins. EMBO Journal 25, pages 2551-2563.
266.
Hoyt, M.A., He, L., Loo, K.K., and Saunders, W.S. (1992). Two Saccharomyces
cerevisiae kinesin-related gene products required for mitotic spindle assembly.
Journal of Cell Biology 118, pages 109-120.
267.
Jacobson, M.D., Gray, S., Yuste-Rojas, M., and Cross, F.R. (2000). Testing cyclin
specificity in the exit from mitosis. Molecular and Cellular Biology 20, pages
4483-4493.
268.
Byers, B. (1981). Multiple roles of the spindle pole bodies in the life cycle of
Saccharomyces cerevisiae. In Molecular Genetics in Yeast, Volume 16, M.K.-B. A.
Stenderup, J. Friis, ed. (Munksgaard, Copenhagen), pp. 119-133.
269.
Hood, J.K., Hwang, W.W., and Silver, P.A. (2001). The Saccharomyces cerevisiae
cyclin Clb2p is targeted to multiple subcellular locations by cis- and trans-acting
determinants. Journal of Cell Science 114, pages 589-597.
270.
Bailly, E., Cabantous, S., Sondaz, D., Bernadac, A., and Simon, M.N. (2003).
Differential cellular localization among mitotic cyclins from Saccharomyces
cerevisiae: a new role for the axial budding protein Bud3 in targeting Clb2 to the
mother-bud neck. Journal of Cell Science 116, pages 4119-4130.
271.
Edgington, N.P., and Futcher, B. (2001). Relationship between the function and
the location of G1 cyclins in S. cerevisiae. Journal of Cell Science 114, pages 45994611.
272.
Maekawa, H., and Schiebel, E. (2004). Cdk1-Clb4 controls the interaction of astral
microtubule plus ends with subdomains of the daughter cell cortex. Genes and
Development 18, pages 1709-1724.
209
273.
Baum, P., Yip, C., Goetsch, L., and Byers, B. (1988). A yeast gene essential for
regulation of spindle pole duplication. Molecular and Cellular Biology 8, pages
5386-5397.
274.
Hinchcliffe, E.H., Cassels, G.O., Rieder, C.L., and Sluder, G. (1998). The
coordination of centrosome reproduction with nuclear events of the cell cycle in
the sea urchin zygote. Journal of Cell Biology 140, pages 1417-1426.
275.
Stemmann, O., Zou, H., Gerber, S.A., Gygi, S.P., and Kirschner, M.W. (2001). Dual
inhibition of sister chromatid separation at metaphase. Cell 107, pages 715-726.
276.
Chee, M.K., and Haase, S.B. B-cyclin/CDKs regulate mitotic spindle assembly by
phosphorylating kinesins-5 in budding yeast. submitted to PLoS Genetics.
277.
Anderson, V.E., Prudden, J., Prochnik, S., Giddings, T.H., Jr., and Hardwick, K.G.
(2007). Novel sfi1 alleles uncover additional functions for Sfi1p in bipolar spindle
assembly and function. Molecular Biology of the Cell 18, pages 2047-2056.
278.
Li, S., Sandercock, A.M., Conduit, P., Robinson, C.V., Williams, R.L., and Kilmartin,
J.V. (2006). Structural role of Sfi1p-centrin filaments in budding yeast spindle
pole body duplication. Journal of Cell Biology 173, pages 867-877.
279.
Holinger, E.P., Old, W.M., Giddings, T.H., Jr., Wong, C., Yates, J.R., 3rd, and
Winey, M. (2009). Budding yeast centrosome duplication requires stabilization of
Spc29 via Mps1-mediated phosphorylation. Journal of Biological Chemistry 284,
pages 12949-12955.
280.
Huh, W.K., Falvo, J.V., Gerke, L.C., Carroll, A.S., Howson, R.W., Weissman, J.S.,
and O'Shea, E.K. (2003). Global analysis of protein localization in budding yeast.
Nature 425, pages 686-691.
281.
Gheber, L., Kuo, S.C., and Hoyt, M.A. (1999). Motile properties of the kinesinrelated Cin8p spindle motor extracted from Saccharomyces cerevisiae cells.
Journal of Biological Chemistry 274, pages 9564-9572.
282.
Sikorski, R.S., and Hieter, P. (1989). A system of shuttle vectors and yeast host
strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae.
Genetics 122, pages 19-27.
210
283.
Gietz, R.D., and Sugino, A. (1988). New yeast-Escherichia coli shuttle vectors
constructed with in vitro mutagenized yeast genes lacking six-base pair
restriction sites. Gene 74, pages 527-534.
284.
Byers, B., and Goetsch, L. (1991). Preparation of yeast cells for thin-section
electron microscopy. Methods in Enzymology 194, pages 602-608.
285.
Haase, S.B., and Reed, S.I. (2002). Improved flow cytometric analysis of the
budding yeast cell cycle. Cell Cycle 1, pages 132-136.
286.
Macnab, R.M. (1992). Genetics and biogenesis of bacterial flagella. Annual
Reviews in Genetics 26, pages 131-158.
287.
Shapiro, L. (1995). The bacterial flagellum: from genetic network to complex
architecture. Cell 80, pages 525-527.
288.
Kilmartin, J.V., Dyos, S.L., Kershaw, D., and Finch, J.T. (1993). A spacer protein in
the Saccharomyces cerevisiae spindle poly body whose transcript is cell cycleregulated. Journal of Cell Biology 123, pages 1175-1184.
289.
Bullitt, E., Rout, M.P., Kilmartin, J.V., and Akey, C.W. (1997). The yeast spindle
pole body is assembled around a central crystal of Spc42p. Cell 89, pages 10771086.
290.
Muller, E.G., Snydsman, B.E., Novik, I., Hailey, D.W., Gestaut, D.R., Niemann, C.A.,
O'Toole, E.T., Giddings, T.H., Jr., Sundin, B.A., and Davis, T.N. (2005). The
organization of the core proteins of the yeast spindle pole body. Molecular
Biology of the Cell 16, pages 3341-3352.
291.
Peter, M., and Herskowitz, I. (1994). Direct inhibition of the yeast cyclindependent kinase Cdc28-Cln by Far1. Science 265, pages 1228-1231.
292.
Wagner, M.V., Smolka, M.B., de Bruin, R.A., Zhou, H., Wittenberg, C., and Dowdy,
S.F. (2009). Whi5 regulation by site specific CDK-phosphorylation in
Saccharomyces cerevisiae. PLoS One 4, pages e4300.
293.
Ho, Y., Costanzo, M., Moore, L., Kobayashi, R., and Andrews, B.J. (1999).
Regulation of transcription at the Saccharomyces cerevisiae start transition by
Stb1, a Swi6-binding protein. Molecular and Cellular Biology 19, pages 52675278.
211
294.
Goldstein, A.L., and McCusker, J.H. (1999). Three new dominant drug resistance
cassettes for gene disruption in Saccharomyces cerevisiae. Yeast 15, pages 15411553.
295.
Bowers, L.M., Shapland, E.B., and Ryan, K.R. (2008). Who's in charge here?
Regulating cell cycle regulators. Current Opinion in Microbiology 11, pages 547552.
296.
Domian, I.J., Reisenauer, A., and Shapiro, L. (1999). Feedback control of a master
bacterial cell-cycle regulator. Proceedings of the National Academy of Sciences U
S A 96, pages 6648-6653.
297.
Collier, J., McAdams, H.H., and Shapiro, L. (2007). A DNA methylation ratchet
governs progression through a bacterial cell cycle. Proceedings of the National
Academy of Sciences U S A 104, pages 17111-17116.
298.
Collier, J., Murray, S.R., and Shapiro, L. (2006). DnaA couples DNA replication and
the expression of two cell cycle master regulators. EMBO Journal 25, pages 346356.
299.
Quon, K.C., Yang, B., Domian, I.J., Shapiro, L., and Marczynski, G.T. (1998).
Negative control of bacterial DNA replication by a cell cycle regulatory protein
that binds at the chromosome origin. Proceedings of the National Academy
Sciences U S A 95, pages 120-125.
300.
Gorbatyuk, B., and Marczynski, G.T. (2001). Physiological consequences of
blocked Caulobacter crescentus dnaA expression, an essential DNA replication
gene. Molecular Microbiology 40, pages 485-497.
301.
Domian, I.J., Quon, K.C., and Shapiro, L. (1997). Cell type-specific
phosphorylation and proteolysis of a transcriptional regulator controls the G1-toS transition in a bacterial cell cycle. Cell 90, pages 415-424.
302.
Wright, R., Stephens, C., Zweiger, G., Shapiro, L., and Alley, M.R. (1996).
Caulobacter Lon protease has a critical role in cell-cycle control of DNA
methylation. Genes and Development 10, pages 1532-1542.
303.
Krylov, D.M., Nasmyth, K., and Koonin, E.V. (2003). Evolution of eukaryotic cell
cycle regulation: stepwise addition of regulatory kinases and late advent of the
CDKs. Current Biology 13, pages 173-177.
212
304.
Nasmyth, K. (1995). Evolution of the cell cycle. Philosophical Transactions of the
Royal Society of London: Biological Sciences 349, pages 271-281.
305.
Lew, D.J., Dulic, V., and Reed, S.I. (1991). Isolation of three novel human cyclins
by rescue of G1 cyclin (Cln) function in yeast. Cell 66, pages 1197-1206.
306.
Madan Babu, M., Balaji, S., and Aravind, L. (2007). General trends in the
evolution of prokaryotic transcriptional regulatory networks. Genome Dynamics
3, pages 66-80.
307.
Holt, L.J., Tuch, B.B., Villen, J., Johnson, A.D., Gygi, S.P., and Morgan, D.O. (2009).
Global analysis of Cdk1 substrate phosphorylation sites provides insights into
evolution. Science 325, pages 1682-1686.
308.
Madan Babu, M., Teichmann, S.A., and Aravind, L. (2006). Evolutionary dynamics
of prokaryotic transcriptional regulatory networks. Journal of Molecular Biology
358, pages 614-633.
309.
Coqueret, O. (2002). Linking cyclins to transcriptional control. Gene 299, pages
35-55.
213
Biography
Laura Anne Simmons Kovacs was born July 12, 1981 in Phoenix, Arizona. As a
child she lived in Chandler, Arizona and Redding, California before attending high school
in Albuquerque, New Mexico. Laura attended college at Colgate University in Hamilton,
New York. At Colgate, she performed undergraduate research in the labs of Dr. Barbara
Hoopes and Dr. Kenneth Belanger, where she developed a love for “the awesome power
of yeast genetics”. She received her Bachelor of Arts degree in May 2003, graduating
magna cum laude with High Honors in Molecular Biology and Distinction in the Liberal
Arts.
In August, 2003, Laura matriculated into the Cell and Molecular Biology training
program at Duke University, in Durham, North Carolina. She joined the laboratory of Dr.
Steven Haase, and the University Program in Genetics and Genomics for her dissertation
research, which focuses on cell-cycle regulation in the yeast, Saccharomyces cerevisiae.
During graduate school, she published a paper in Molecular Biology of the Cell in 2008
entitled, “Intrinsic and cyclin-dependent kinase-dependent control of spindle pole body
duplication in budding yeast” [258]. She also authored a review published in Cell Cycle
in 2008 entitled, “Transcription networks and cyclin/CDKs: The yin and yang of cell-cycle
oscillators” [173]. After completing her Ph.D., she intends to continue her research
training as a post-doctoral fellow.
214