Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Distinct mechanisms of DDK recruitment to Fkh-activated and CEN-proximal origins control replication timing program in S. cerevisiae
(USC Thesis Other)
Distinct mechanisms of DDK recruitment to Fkh-activated and CEN-proximal origins control replication timing program in S. cerevisiae
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
i
Distinct mechanisms of DDK recruitment to Fkh-activated and CEN-proximal origins control
replication timing program in S. cerevisiae
by
Meghan Victoria Petrie
A Dissertation presented to the
FACULTY OF USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(MOLECULAR BIOLOGY)
August 2023
Copyright 2023 Meghan Victoria Petrie
ii
Acknowledgements
I would like to start by thanking everyone who has helped me in big and small ways
throughout my time here at USC. A PhD journey is never easy but with all the support from
colleagues, mentors, friends, and family along the way, it definitely made the tough times not so
tough and the great times even greater. Every PhD experience is unique and I want to thank Dr.
Oscar Aparicio for making mine something that I’ll be able to look back at fondly. He taught me
to think critically, independently, and was there to give me pushes in the right directions every
time I asked for guidance. I wouldn’t have had such a successful PhD without his continuous
support and I will always be thankful that he accepted me into his lab. Additionally, I want to thank
my committee, Dr. Irene Chiolo, Dr. Mathew Michael, Dr. Marc Vermulst, and Dr. Peter Calabrese
for helpful discussions and guidance throughout my graduate career.
Next, I have to thank my cohort: Mezmur Below, Caleb Ghione, Joseph Hale, Yiwei He,
Dan Ma, Joshua Park, and Nicole Stuhr. Our commitment to birthday dinners has been something
that has kept us all connected. I also have to thank my lab: Haiyang Zhang, Yiwie He, and Yan
Gan who have been a wonderful group to work beside. I would also like to thank MBGSA and
WiM; being a part of these groups was a wonderful way to interact with our department. I am so
appreciative of the events we put together that helped make the building into a community. I would
also like to thank all my many friends and colleagues around Ray Irani Hall for their support both
personally and professionally throughout the years.
Next, I would like to thank my family for their continuous support through the years. Mom
and Dad thank you for being there for me and always believing in me. You always pushed me to
do my best and stay focused on my goals. Katherine thanks for calling me almost every morning
and being someone that I can always rely on for a good laugh. Without my family, none of this
would have been possible. And finally, I want to thank Joey Mendez. You’ve been with me before
I started my PhD and I hope to be with you long after. I don’t think I’ll ever be able to fully express
how grateful I’ve been for your support in both the good times and bad. Thank you for being there
and I am looking forward to our future.
iii
Table of Contents
Acknowledgements ......................................................................................................................... ii
List of Figures ................................................................................................................................ vi
List of Supplemental Figures ........................................................................................................ vii
Abstract ........................................................................................................................................ viii
Chapter 1: Introduction ............................................................................................................... 1
Eukaryotic Replication Initiation and Timing ............................................................... 1
Effects of Chromatin Environment on Replication Initiation in Yeast ....................... 4
The Role of Forkhead Transcription Factors in Replication Timing .......................... 7
Recruitment of Limiting Factors Determines Origin Firing timing ............................ 9
Chapter 2: Dynamic relocalization of replication origins by Fkh1 requires execution of
DDK function and Cdc45 loading at origins ............................................................................ 12
Introduction ..................................................................................................................... 13
Materials and Methods ................................................................................................... 15
2.1 Plasmid constructions ................................................................................................. 15
2.2 Yeast strain constructions ........................................................................................... 17
2.3 Cell growth synchronization ....................................................................................... 18
2.4 Live-cell fluorescence microscopy and image analysis .............................................. 19
2.5 Quantitative BrdU Immunoprecipitation (QBU) ........................................................ 19
2.6 Chromatin immunoprecipitation analyzed by sequencing (ChIP) .............................. 19
2.7 Time-lapse video and MSD analysis .......................................................................... 19
Results .............................................................................................................................. 20
2.8 Fkh1-induced origin activation re-positions a subtelomeric origin in G1 phase ........ 20
2.9 Fkh1 globally regulates subnuclear positioning of early origins in G1 phase ............ 25
2.10 DDK- but not CDK-dependent step of replication initiation drives origin
relocalization ..................................................................................................................... 28
2.11 Origin mobility increases with origin relocalization ................................................ 34
Discussion......................................................................................................................... 36
Chapter 3: Dbf4 Zn-finger motif is specifically required for stimulation of Ctf19-
activated origins in Saccharomyces cerevisiae ......................................................................... 41
Abstract ............................................................................................................................ 41
Introduction ..................................................................................................................... 42
Materials and Methods ................................................................................................... 44
3.1 Plasmid and yeast strain construction. ........................................................................ 44
3.2 Other methods. ............................................................................................................ 45
3.3 Computation and statistics. ......................................................................................... 46
Results and Discussion .................................................................................................... 46
iv
3.4 Dbf4∆C is defective in essential Dbf4 function(s) beyond origin-targeting by
Fkh1 and Ctf19 ................................................................................................................. 46
3.5 dbf4∆C is defective in overall rate of genome replication while dbf4-Zn* and
ctf19∆ are not .................................................................................................................... 48
3.6 Dbf4-Zn* mutations specifically eliminate early activation of CEN-proximal
origins ............................................................................................................................... 51
3.7 CEN-proximal origins are differentially sensitive to loss of Ctf19 or Dbf4-Zn* ....... 56
3.8 Dbf4-Zn* retains Fkh1-dependent targeting to Fkh1-activated origins ..................... 59
3.9 Dbf4-Zn* is Defective in Its Recruitment to CENs .................................................... 59
Perspective ....................................................................................................................... 61
Chapter 4: FHA Domain of Fkh1 Is Required in Co-Recruitment of Fkh1 and Dbf4 for
Replication Origin Initiation ...................................................................................................... 62
Abstract ............................................................................................................................ 62
Introduction ..................................................................................................................... 63
Materials and Methods ................................................................................................... 66
4.1 Plasmid constructions ................................................................................................. 66
4.2 Yeast strain constructions ........................................................................................... 67
4.3 Other methods ............................................................................................................. 68
4.4 Computation and statistics .......................................................................................... 69
Results .............................................................................................................................. 70
4.5 Fkh1-FHA is required for replication origin stimulation ............................................ 70
4.6 Fkh1-FHA is dispensable for Orc1 and Mcm4 occupancies at origins ...................... 74
4.7 Fkh1-R80A is defective in Fkh1 recruitment to origins ............................................. 76
4.8 Loss of Fkh1 affects Dbf4 binding ............................................................................. 78
4.9 FKH DBD fusion to Dbf4 bypasses requirement for FHA domain............................ 79
4.10 Dbf4 is required for Fkh1 recruitment to origins ...................................................... 81
4.11 Fkh1 and Dbf4 physically interact ............................................................................ 82
Discussion......................................................................................................................... 84
Chapter 5: Future Directions ..................................................................................................... 91
Research Impacts of this Dissertation ........................................................................... 91
Future Experiments ........................................................................................................ 93
5.1 Defining the Dbf4 domain for Fkh-FHA interaction .................................................. 93
5.2 Global search for FKH1 interactors ............................................................................ 94
References .................................................................................................................................... 95
Chapter 1 References ...................................................................................................... 95
Chapter 2 References ...................................................................................................... 99
Chapter 3 References .................................................................................................... 105
Chapter 4 References .................................................................................................... 107
Chapter 5 References .................................................................................................... 110
v
Appendix A References ................................................................................................ 111
Appendix B References................................................................................................. 112
Appendices ................................................................................................................................. 113
Appendix A: Mating type switching in Saccharomyces cerevisiae............................ 113
A.1 The donor preference mechanism ............................................................................ 113
A.2 Does Dbf4 play a role in mating type switching? .................................................... 115
A.3 The search for the interaction between Fkh1 and HO ............................................. 117
A.4 Can a different endonuclease substitute for HO in donor preference? .................... 119
Appendix B: Plasmid Stability in Saccharomyces cerevisiae..................................... 121
B.1 CTF19 and the Dbf4 C-term reduced CEN proximal ARS activity ........................ 121
vi
List of Figures
Figure 1.1 Schematic of various short functional elements beyond the ACS……………………. 2
Figure 1.2 Overview of transition from prereplication complex to preinitiation complex………..3
Figure 1.3 Overview of variation in origin firing timing ………………………………………….5
Figure 1.4 Anatomy of Fkh-activated S. cerevisiae Replication Origins ………………………..10
Figure 2.1 Fkh1-induced origin activation re-positions a subtelomeric origin in G1 phase……..22
Figure 2.2 Normal dosage of Fkh1 is sufficient to relocalize ARS305
V-R
and advance its
firing time…………………………………………………………………………………...24
Figure 2.3 Fkh1 determines early origin positioning globally…………………………………...26
Figure 2.4 Fkh-binding sites and a functional ACS are required for Fkh1 to influence early
origin positioning globally ………………………………………………………………….27
Figure 2.5 Origin localization in G1 is DDK regulated ………………………………………….30
Figure 2.6 DDK regulation of origin localization reflects its phosphorylation of Mcm4 and
consequent Cdc45 loading ………………………………………………………………….33
Figure 2.7 Origin mobility increases with origin relocalization …………………………………35
Figure 2.8 Model of origin localization linked to initiation……………………………………...39
Figure 3.1 Dbf4∆C is defective in essential Dbf4 function(s) …………………………………...47
Figure 3.2 Dbf4∆C is defective in bulk genome replication, while Dbf4-Zn* mutants are not …49
Figure 3.3 Immunoblotting detects normal abundance of Dbf4-Zn* but reduced Dbf4∆C ……..51
Figure 3.4 Dbf4-Zn* is defective in CEN-proximal origin firing ……………………………….53
Figure 3.5 CEN-proximal origins are differentially sensitive to CTF19 deletion or dbf4-Zn*….57
Figure 3.6 Dbf4-Zn* retains Fkh1 origin stimulation……………………………………………58
Figure 3.7 Dbf4-Zn* is defective in its recruitment to CENs ……………………………………60
Figure 4.1 Fkh1-FHA regulates replication origin timing ……………………………………….71
Figure 4.2 Fkh1-R80A in the presence of Fkh2 regulate replication origin timing ……………..73
Figure 4.3 Fkh1-R80A has no effect on ORC1 or MCM4 binding ..…………………………….75
Figure 4.4 Fkh1-R80A has reduced binding at Fkh-activated origins……………………….. ….77
Figure 4.5 Dbf4 binds preferentially to Fkh-activated origins and the centromeres …………….78
Figure 4.6 FKH DNA-Binding Domain fusion to Dbf4 bypasses requirement for FHA………..80
Figure 4.7 dbf4∆C reduces Fkh1 binding to origins ……………………………………………..82
Figure 4.8 Dbf4 binds to Fkh1 is dependent on the Fkh1-FHA …………………………………83
Figure A.1 Donor preference during mating-type switching …………………………………...114
Figure A.2 PCR-based donor preference assay using Dbf4 mutants …………………………...116
Figure A.3 HO deletions and mutation fails to alter donor preference…………………………118
Figure A.4 CRISPR can substitute for HO mating type switching……………………………..120
Figure B.1 Plasmid loss rate per generation of the indicated strains …………………………...122
vii
List of Supplemental Figures
Figure S1 Supplement to figure 4.1 ……………………………………………………………...87
Figure S2 Supplement to figure 4.2 ……………………………………………………………...88
Figure S3 Supplement to figure 4.3 ……………………………………………………………...88
Figure S4 Supplement to figure 4.4 ……………………………………………………………...89
Figure S5 Supplement to figure 4.5 ……………………………………………………………...89
Figure S6 Supplement to figure 4.6 ……………………………………………………………...89
Figure S7 Supplement to figure 4.7 ……………………………………………………………...90
Figure S8 dbf4-10A does not alter orign replication patterns …………………………………...90
viii
Abstract
Organized and faithful DNA replication is required through any given mitotic cell division.
All life requires organized and faithful DNA replication each time it proceeds through a given cell
division. In eukaryotic genomes replication initiates at multiple sites called origins, dependent on
factors such as DNA sequence, chromatin environment, and sub-nuclear localization. In G1-phase
more origins than can be fired are licensed, so that as cells enter S-phase only a subset drives the
replication of entire genomes. This duplication occurs in a temporal manner due to limiting factors
allowing only select origins to activate and disruption of this process can result in genome
instability. In Saccharomyces cerevisiae, it has been observed that Forkhead transcription factors
bind and enable early firing of select origins named Fkh-activated origins. Our studies have further
elucidated the role of Fkh1 in driving sub-nuclear localization of these early firing origins. The
relationship between location and origin activation are linked with Fkh1 sequestering limiting
factors in G1-phase. In particular, we investigated recruitment mechanisms of Dbf4, a subunit of
DDK, and furthered the understanding of how it is directed to select origins. Although loss or
disruption of the C-term of Dbf4 greatly reduces CEN-proximal firing it does not fully eliminate
early firing of Fkh-activated origins nor Fkh1 interaction. We have found that the Fkh1-FHA plays
a crucial role in facilitating early firing of Fkh-activated origins and interacts with a yet unknown
region of Dbf4. Overall, our findings contribute additional details to established mechanisms of
origin firing, highlights the relationship between sub-nuclear localization and sequestering of
limiting factors to origin activation in early S-phase.
In chapter 1 I will summarize the known eukaryotic mechanisms that establish DNA
replication machinery and factors that influence activation along with key interactions that have
yet to be fully understood. Next, in chapter 2 I will present the relationship between sub-nuclear
ix
location of origins and firing timing, with focus on a Fkh1 dependent mechanism driving select
early origins away from the nuclear periphery. Then I will provide a detailed analysis on the role
of the Dbf4 C-terminus Zn-Finger in chapter 3 which we determine is requirement for Ctf19-
activated origin firing, but only partially explains the mechanism driving early Fkh-activated origin
firing. I will then conclude in chapter 4 with a deeper investigation and expansion of the previously
identified role of Fkh1-FHA domain, which we observed has a global role on replication timing
and its requirement in facilitating Dbf4 recruitment Chapter 4.
1
Chapter 1: Introduction
Eukaryotic Replication Initiation and Timing
One of the fundamental processes of life is DNA replication. This process must be tightly
controlled and result in a single duplication of the genome with minimal errors so that the genetic
material passed to every subsequent daughter cell remains almost, if not exactly identical. A
conserved mechanism to regulate DNA replication has evolved to ensure this process and it occurs
at specific sites called origins. In circular bacterial genomes such as E. coli there is a single origin
site where replication is initiated in opposite directions (Costa and Diffley 2022). This is feasible
for small prokaryotic genomes, but even lower eukaryotes have multiple linear chromosomes and
thus require multiple origins. In Saccharomyces cerevisiae, DNA replication relies on specific
sequences to define origin sites across its 16 chromosomes. Higher eukaryotes with even larger
genomes have more complex ways of establishing origins involving epigenetic modifications
along with DNA sequence which influence chromatin structure (Costa and Diffley 2022). Our
current understanding of DNA replication relies heavily on use of model organisms which retain
the conserved DNA replication machinery but occur at predictable locations. Thus, it is through
budding yeast that the community has gained some of the greatest insight into eukaryotic DNA
replication using their well-documented origins.
In S. cerevisiae, these specific DNA replication origins sites are named Autonomously
Replicating Sequences (ARS) and were initially identified as 100-200bp fragments discovered in
S. cerevisiae that conferred maintenance of an extrachromosomal plasmid (Stinchcomb et al.,
1979). The structure of an ARS includes several additional elements including an essential ARS
2
consensus sequence (ACS or A element) along with multiple auxiliary (B) elements. The ACS has
been shown to be a conserved, but degenerate A/T-rich sequence to which the Origin Replication
Complex (ORC) binds, fulfilling the first loading step of a potentially active origin. The
downstream B elements are less conserved across ARSs, with B1 playing a role in additional
assistance for ORC loading. The more distant B2 element makes up the DNA unwinding element
(DUE) and even further is the B3 element which is an Abf1 binding site. The additional B4 element
has also been identified although its exact function is not greatly understood (Lin and Kowalski
1997). Variation in how B elements appear, or not, seem to influence the identity of individual
replication origins (Rao and Stillman 1995; Eaton et al., 2010; Chang et al., 2011; Costa and
Diffley 2022).
Figure 1.1 Schematic of various short functional elements beyond the ACS. The ACS and B1 are known ORC binding
sites, B2 is usually defined as the DUE, B3 has been established as an Abf1 binding site, and B4 function has yet to
be fully described. Here ARS305 and ARS1 are shown with their various B elements which vary in size and distance
from the ACS. Adapted from Lin and Kowalski 1997.
While in G1 phase, the six-subunits of ORC binds to an ARS and then recruits Cdc6 and
Cdt1 to facilitate bidirectional loading of two copies of the mini-chromosomal maintenance
complex (MCM) (Coster and Diffley 2017). Together these proteins form a Pre-Replication
Complex (pre-RC), and the origin is referred to as being “licensed” as it has the ability to initiate
replication or “fire” once the cell transitions into S-phase (Bell and Kaguni 2013). The progression
from G1 into S-phase relies on the accumulating levels of kinases such as Dbf4-dependent-kinase
3
(DDK) and cyclin-dependent-kinase (CDK) (Leonard and Méchali 2013; Bertoli et al., 2013;
Duncker and Duncker 2016). The transition into S-phase begins with DDK phosphorylating the
MCM helicase complex allowing for the additional loading of Cdc45 and Sld3 to the replication
complex. Following this, CDK can then phosphorylate the newly recruited Cdc45 and Sld3 along
with the replication factor Sld2. With the addition of factors such as GINS and Dpb11 the pre-RC
has transformed into a pre-initiation complex. Finally, the CMG (Cdc45, MCM, and GINS) forms
the active helicase which initiates DNA unwinding and expansion of the replication bubble as
DNA polymerases are recruited (Zegerman 2015; Costa and Diffley 2022). Once a cell enters S-
phase, CDK has an additional role in degrading Cdc6 and thus inhibits pre-RC formation on the
nascent DNA effectively limiting genome duplication to once every cell cycle.
Figure 1.2 Overview of transition from prereplication complex to preinitiation complex. MCM is phosphorylated by
DDK. This leads to the loading of Sld3/7 along with Cdc45 onto MCM. Next CDK phosphorylates Sld3 and Sld2
which then recruits Dpb11 which binds Pol ε and GINS resulting in the formation of the preinitiation complex. Adapted
from Costa and Diffley 2022.
4
Effects of Chromatin Environment on Replication Initiation in Yeast
DNA replication occurs in reproducible temporal patterns. In G1 more pre-RCs are
licensed then can feasibly be activated as there are ~800 ARSs but limiting factors allow only
~200 to fire in any given S-phase cycle. This excess results in a selection process in which only
some pre-RCs become active replication complexes. Notably, DDK and Sld2 and Sld3 are in
limited abundance, resulting in a queued order that origin firing follows as pre-RCs wait on and
share these factors for initiation (Mantiero et al 2011). Mis-regulation of this program, such as
unregulated firing in which all origins are activated at similar times significantly slows replication
fork progression and activates checkpoints indicative of a replication stress response (Zhong et al.,
2013). This stress of excessive origin firing can lead to complications that threaten genome
stability. Our current understanding of how replication timing is controlled relies on analysis of
the transition between G1 and S phase, when pre-RCs are activated. As briefly outlined above pre-
RC assembly is understood in great mechanistic detail in S. cerevisiae. However, some origins are
efficient and usually early firing, while others are inefficient and fire later in S-phase (Aparicio,
2013). Further still, some origins rarely initiate and instead a nearby active replisome will passively
replicate over those origins that were later or less efficient than its neighboring origins. This has
left the question as to how certain origins are selectively activated or repressed with far fewer
origins firing than are licensed. One part of the answer seems to be the local chromatin
environment.
Chromatin structure can be broken into two main regions: heterochromatic and
euchromatic. Chromatin remodelers such as histone deacetylases and acetyltransferase play a
critical role in assembling the chromosomal landscape. A key role of differential chromatin
5
Figure 1.3 Overview of variation in origin firing timing. In G1-phase an excess of pre-RCs are established across the
genome. Once cells transition into S-phase origins fire at different times dependent on their access to low abundance
initiation factors. Here an example is shown from left to right of a late, early, and passively replicated origin.
geography is how it prevents or encourages certain protein-DNA interactions based on access. For
example, interrupting the compaction of chromatin by deleting histone deacetylases (HDACs)
such as Silent Information Regulator 2 (Sir2) and Reduced Potassium Dependency 3 (Rpd3) have
been shown to alter origin firing patterns (Vogelauer et al., 2002; Aparicio et al., 2004). Sir2, which
interacts with the Sir3 and Sir4, together serve to delay initiation of heterochromatic subtelomeric
origins and repetitive ribosomal DNA (rDNA) (Hoggard et al., 2020; Pasero et al., 2002). The
histone deacetylase Rpd3 influences both transcription and replication as it targets promoters
resulting in gene silencing and also delays origin firing. Specifically, when Rpd3 is deleted there
is a global change in origin firing patterns with over 100 Rpd3-regulated origins showing an
increase in firing timing (Knott et al., 2009). More recently Sir2 was shown to be epistatic to Rpd3
in controlling the rDNA and additive for other single-copy origins globally (He et al., 2022). In
direct contrast, if a histone acetyltransferase (HAT) such as General Control Nonderepressible 5
(Gcn5) is directed to a usually late firing origin in the subtelomeric region, that origin will be
activated earlier (Vogelauer et al., 2002). These studies reveal that the compact nature of
6
heterochromatin assembled by HDACs, delays origin firing within, reducing the competition with
euchromatin origins influenced by HATs to be more early firing and efficient.
CENs in larger eukaryotes are heterochromatic as they are large, highly repetitive regions,
but in budding yeast they are just ~125bp. Despite size discrepancy both contain histone H3
variants allowing for assembly of kinetochores and are well known for their early firing
pericentromeric origins (Lochmann and Ivanov 2012). Notably, the Chromosome Transmission
Fidelity 19 (Ctf19), which is part of the kinetochore complex, was identified as also playing a role
in facilitating early origin firing near CENs through recruitment of DDK in telophase (Natsume et
al., 2013). In contrast, subtelomeric regions which are usually associated with heterochromatin
contain late and often inefficient origins. In particular, the Rap1 interacting factor 1 (Rif1) has
been shown to block origin firing near telomeres as well as having a global role in regulating origin
firing (Peace et al., 2014; Yamazaki et al., 2012). Rif1 suppresses MCM helicase activation by
recruiting phosphatase Glc7, which antagonizes DDK activity. Interestingly, these Rif1-regulated
origins tend to localize near the nuclear periphery as Rif1 interacts with Rap1 which binds
telomeres that are anchored to the nuclear pore complex (Davé et al. 2014; Hiraga et al., 2014).
Due to this, these telomeric regions located along the nuclear periphery are usually passively
replicated by early firing neighboring origins. The relationship between origin firing timing and
subnuclear localization is not clearly defined but DNA confined to the periphery can obtain
chromatin marks that maintain later firing of these regions (Heun et al., 2001). This suggests that
the local chromatin environment is connected with origin firing timing, and interestingly also
linked to location within the nucleus.
The effect of environment and location on replication is unsurprising as transcription
machinery follows a similar pattern due to the accessibility of certain regions allowing for these
7
processes to take place. Both gene promoters and ARSs require nucleosome free regions in which
transcription factors or other DNA binding proteins can establish their respective machineries.
Indeed, it has been observed in higher eukaryotes that more actively transcribed regions of the
genome are highly correlated with early firing origins (Rhind and Gilbert, 2013). However,
accessibility alone cannot explain observed DNA replication patterns as previously mentioned an
excess of pre-RCs are established across the genome. This leaves the question as to what
mechanisms allow select origins to be consistently early firing beyond the indirect chromatin
environment. One answer to this was found to be the Forkhead (Fkh) transcription factors which
have been shown to advance the firing of roughly 100 origins termed Fkh-activated (Knott et al.,
2012).
The Role of Forkhead Transcription Factors in Replication Timing
The Fkh proteins are part of the large and well known Forkhead Box (Fox) family of
transcription factors involved in a wide variety of cellular functions across eukaryotes, ranging
from development, stress response, and cell cycle control (Murakami et al. 2010; Knott et al.,
2012). In budding yeast, there are four Fkh proteins: Fork Head-Like 1 (Fhl1), High-Copy
suppressor of Calmodulin 1 (Hcm1), Fkh1, and Fkh2. The most divergent of the set is Fhl1 as it
lacks the ability to even target Fkh binding sites (FBS) in the genome (Jin et al., 2020). Next is
Hcm1 which plays a role in advancing cell cycle progression through transcriptional regulation of
genes such as Fkh1/2 (Pramila et al., 2006). Fkh1 and its paralog Fkh2 are the most closely related
in protein sequence and DNA binding with both containing a Forkhead associated (FHA) domain,
a regulator of protein-protein interactions, and a similar FKH DNA binding domain (DBD) that
allows binding to FBS (Aparicio, 2013; Jin et al., 2020). The most obvious difference between the
two is that Fkh2 has an extended C-terminus aiding in an interaction with Nuclear Division
8
Defective 1 (Ndd1). Together Fkh1/2 are best-known for their role in the regulation of the ~30
CLB2 cluster genes, and are crucial for promoting cell cycle progression (Hollenhorst et al., 2000).
Although not lethal, the combined deletion of Fkh1/2 results in pseudohyphal growth due
to the deregulation of the G2 and M phase specific transcription of a group of genes known as the
CLB2 cluster (Zhu et al., 2000). However, single deletions of either leads to a more minor
transcriptional disruption as the remaining Fkh is sufficient to partially substitute for the other.
Specifically, Fkh2 has a longer C-terminus which represses CLB2 genes in G1 and through
interaction with Ndd1 activates the Mcm1-Fkh2 mediated transcription in S, G2, and M phase.
Ndd1 is an essential protein but its requirement for viability can be bypassed through deletion of
Fkh2 or its C-terminus as there would no longer be a need to lift the Fkh2 gene repression (Koranda
et al., 2000).
Fkh1 also has a unique function as it is involved in the regulation of mating type switching
in S. cerevisiae (Wu & Haber, 1995). Budding yeast has two mating types, either MATa or MATα,
depending on which gene is being expressed at the mating loci located on chromosome III. To
switch between these the HO endonuclease makes a double strand break (DSB) at the expressed
mating type gene, allowing for homologous recombination from either the silent mating type
cassette array on the left arm (HMLα) or the right arm (HMRa) on chromosome III (Wu & Haber,
1995). This switch is not random and there is a preference for a MATa cell to preferentially switch
to MATα through a mechanism involving the Fkh1-FHA recruiting as of yet unknown factors to
the recombination enhancer (RE) near the HMLα when the DSB is induced (Sun et al., 2002; Li
et al. 2012).
9
Analysis has shown that deletion of Fkh1/2 selectively delayed early firing origins which
contain an FBS (Knott et al., 2012; Ostrow et al., 2017; Peace et al., 2016). When only Fkh1 is
deleted there is a greater reduction in origin firing as compared to when Fkh2 alone is deleted,
which has no effect on its own, indicating that Fkh1 plays the predominant role in driving early
firing Fkh-activated origins. As Fkh1/2 are best known as transcription factors, gene expression
near origins were analyzed and revealed that the replication defect is not due to local changes in
transcription around origins (Knott et al., 2012). Furthermore, it has been shown that relocating an
FBS to late firing origins allows them to become early firing due to a Fkh1 dependent mechanism
(Reinapae et al 2017). In addition, the presence of Fkh1 has been observed to facilitate early
loading of Cdc45 to select pre-RCs, which as outlined previously is a downstream step after the
limiting factor DDK phosphorylates MCM (Knott et al., 2012). This suggests that regardless of
chromatin environment, the main role of Fkh1 may be in recruiting limiting factors allowing select
Fkh-activated origins to fire early in S-phase.
Recruitment of Limiting Factors Determines Origin Firing timing
Fkh1 has multiple cellular functions and therefore it is of little surprise that it has been
identified as a key player in influencing topologically associated domains (TADs) (Eser et al.
2017). The 3D-organization of genomes tends to result in TADs, which have similar histone
modifications, transcription levels, and replication timing. As Fkh1 mediates long range genome
interaction, it suggests that the TADs it influences become early replicating regions due to Fkh1
facilitated sequestering of S-phase factors such as the previously noted Cdc45 (Knott et al., 2012;
Ostrow et al., 2017). Interestingly deletion of Fkh1 has no effect on early CEN-proximal origins
but loss of Ctf19 does greatly delay their firing timing (Knot et al., 2012; Natsume et al. 2013).
This highlights a mechanism in which Ctf19 recruits DDK through interaction with the Dbf4
10
Figure 1.4 Anatomy of Fkh-activated S. cerevisiae Replication Origins. Budding yeast origins are well defined and
contain various domains including ACS and DUE. Select origins additionally contain an FBS, which can contribute
to early origin activity through recruitment of Fkh1/2. Without an FBS or the presence of Fkh1/2 these origins will
be late activating. Adapted from Aparicio, 2013.
subunit and the proximity by CEN-proximal origins. The more recent identification of Fkh1 as the
recruiter of Dbf4, which is upstream of Cdc45 recruitment to origins, to non-centromeric origins
was unsurprising but still leaves the mechanism of their interaction incomplete (Fang et al., 2018).
Together these studies highlight distinct mechanisms in which Ctf19 recruits Dbf4 to CEN-
proximal origins, while Fkh1 recruits it to Fkh-activated origins (Natsume et al., 2013; Fang et al.,
2018). As previously described Fkh1 contains an FHA domain, which is a phosphothreonine
binder and facilitator of protein-protein interactions such as those required for donor preference
(Sun et al., 2002; Li et al. 2012). This makes the FHA domain an ideal candidate for additional
interaction during or after pre-RC formation, such as recruiting limiting factors. Thus far mounting
evidence continues to illustrate Fkh1 as a key influencer in early origin timing through recruiting
limiting factors as well as nuclear localization.
Recently we have identified a further role of Fkh1 as a driver of origin localization away
from the nuclear periphery, which correlates with earlier firing timing (Chapter 2). Fkh1/2 has
already been identified in the mechanism that selectively recruits Cdc45, which is loaded to pre-
11
RCs in a DDK dependent way during the G1 to S-phase transition (Knott et al., 2012). A recent
study identified Fkh1 as an upstream recruiter of DDK to origins providing clues into the molecular
mechanism for how these origins sequester initiating factors and thus can fire early. In particular,
the C-terminus of Dbf4, which contains a Zn-finger motif, was identified as a potential interactor
with Fkh1 (Fang et al., 2018). However, our study into the Dbf4 C-terminus indicates that Fkh1
still plays a role despite point mutation disruptions of the Zn-finger. In fact, it appears that
centromere proximal origins are primarily affected as was previously documented when a C-
terminus epitope tag was added to Dbf4 (Natsume et al., 2013; Chapter 3). Our recent investigation
into the role of Fkh1-FHA suggests its main function is in the recruitment of limiting factors such
as DDK not origin stabilization as previously indicated (Chapter 4). The major focus of this thesis
is to further elucidate key mechanisms driving early origin firing patterns which contribute to
consistent and reliable global replication patterns.
12
Chapter 2: Dynamic relocalization of replication origins by Fkh1
requires execution of DDK function and Cdc45 loading at origins
Adapted from:
Zhang H, Petrie MV, He Y, Peace JM, Chiolo IE, Aparicio OM. (2019) “Dynamic relocalization
of replication origins by Fkh1 requires execution of DDK function and Cdc45 loading at
origins”. Elife. 8:e45512. doi:10.7554/eLife.45512
My primary contribution to the following work included live-cell microscopy including
time lapse videos of the nuclear localization of replication origins (Figures 2.3, 2.4, 2.6, and 2.7).
Additionally, I contributed to strain construction, analysis and quantification of data, discussion,
and additional experiments for peer reviews.
Abstract
Chromosomal DNA elements are organized into spatial domains within the eukaryotic
nucleus. Sites undergoing DNA replication, high-level transcription, and repair of double-strand
breaks coalesce into foci, although the significance and mechanisms giving rise to these dynamic
structures are poorly understood. In S. cerevisiae, replication origins occupy characteristic
subnuclear localizations that anticipate their initiation timing during S phase. Here, we link
localization of replication origins in G1 phase with Fkh1 activity, which is required for their early
replication timing. Using a Fkh1-dependent origin relocalization assay, we determine that
execution of Dbf4-dependent kinase function, including Cdc45 loading, results in dynamic
relocalization of a replication origin from the nuclear periphery to the interior in G1 phase. Origin
13
mobility increases substantially with Fkh1-driven relocalization. These findings provide novel
molecular insight into the mechanisms that govern dynamics and spatial organization of DNA
replication origins and possibly other functional DNA elements.
Introduction
The spatial organization of chromosomal DNA elements within the nucleus is thought to
derive from and contribute to the regulation of their activity (reviewed in Shachar and Misteli,
2017). For example, euchromatin and heterochromatin represent distinct forms of chromatin that
are distinguished by their levels of transcriptional activity, replication timing, and subnuclear
localization (reviewed in Caridi et al., 2017). Chromosomes partition into subdomains ranging
from hundreds to thousands of kilobases in length that preferentially self-associate and are
consequently referred to as topologically associated domains (TADs) (reviewed in Zhao et al.,
2017). TAD boundaries correlate closely with replication timing domains, suggesting that
replication timing is determined or influenced by this domain structure and/or vice-versa.
In budding and fission yeast, specific mechanisms defining replication timing are linked
with chromosomal domain organization (reviewed in Aparicio, 2013; Yamazaki et al., 2013). Rif1,
which is highly enriched at telomeres, is globally responsible for delayed replication timing of
subtelomeric domains as well as internal late-replicating domains (Hafner et al., 2018; Hayano et
al., 2012; Peace et al., 2014; Tazumi et al., 2012). Rif1 acts by directly antagonizing replication
initiation triggered by Dbf4-dependent kinase (DDK) phosphorylation of MCM helicase proteins
(Davé et al., 2014; Hiraga et al., 2014; Mattarocci et al., 2014). Against this inhibitory backdrop,
specific origins are selected for early activation by mechanisms involving recruitment of Dbf4
(Dfp1 in fission yeast), which is one of several initiation proteins present in limited abundance and
14
thus rate-limiting for origin firing (Mantiero et al., 2011; Patel et al., 2008; Tanaka et al., 2011;
Wu and Nurse, 2009). In S. pombe, Dfp1 is recruited to kinetochores through heterochromatin
protein Swi6 (Hayashi et al., 2009), and in S. cerevisiae, kinetochore protein Ctf19 recruits Dbf4
to stimulate firing of origins within ~25 kb of the centromere (Natsume et al., 2013), thus ensuring
early centromere replication by distinct mechanisms regulating DDK activity. In S. cerevisiae,
Fkh1 and/or Fkh2 (Fkh1/2) recruits Dbf4 to many origins distributed throughout chromosome
arms, thereby ensuring earlier replication of many centromere-distal regions (Fang et al., 2017;
Knott et al., 2012; Lõoke et al., 2013; Ostrow et al., 2014).
Chromosome conformation capture experiments suggest that early-firing origins cluster
spatially in G1-phase prior to initiation and this clustering is dependent on Fkh1/2 (Duan et al.,
2010; Knott et al., 2012). These studies also indicated that early origins generally occupy a distinct
space than late origins. Further studies suggest that Fkh1/2 are enriched at TAD boundaries and
control contacts among origins within TADs (Eser et al., 2017). The distinct spatial distributions
suggested by these recent studies are in accord with earlier studies that examined the subnuclear
distribution of individual origins by fluorescence microscopy. These seminal studies from Heun
and Gasser showed that late-firing origins typically associate with the nuclear periphery during G1
phase whereas early-firing origins typically are found in the nuclear interior during G1 (Heun et
al., 2001a; Heun et al., 2001b). Despite the observed correlations between origin localization in
G1 and firing time in S, the main origin timing determinants mentioned above had not been
elucidated and have not been examined for their impact on subnuclear localization of replication
origin. In this study, we examined how origin stimulation by Fkh1 determines subnuclear origin
localization. Our results suggest that origin relocalized from the nuclear periphery upon execution
of the DDK-dependent step of replication initiation, which is stimulated by Fkh1. This may
15
represent the initial stages in the coalescence of replication origins into clusters that will become
replication factories.
Materials and Methods
2.1 Plasmid constructions
Plasmids are listed in Supplementary file 1. Plasmids were constructed using Gibson
Assembly kit (SGI cat#GA1200) unless otherwise indicated. Restriction enzymes, T4 DNA ligase,
and Klenow were from New England Biolabs and used according to their protocols. Mutagenesis
was carried out using QuikChange Lightning Multi kit (Agilent cat#210515); sequence changes
were confirmed by DNA sequencing (Retrogen Inc). STBLII cells were used for maintenance of
plasmids containing tandem repeats (Invitrogen cat#10268019). Primer sequences for plasmid
constructions are given in Supplementary file 2. NUP49-GFP was PCR-amplified from pUN100-
GFP*-Nup49 (from V Doye) using primers Nup49-GFP-F and Nup49-GFP-R and subcloned into
XhoI+SacI digested vectors pRS403 and pRS404 (Sikorski and Hieter, 1989) to yield p403-
Nup49-GFP and p404-Nup49-GFP, respectively. Primers ADE2-up-F and ADE2-int-R and
separately ADE2-farup-F and ADE2-up-R were used to amplify sequences of ADE2 for targeting
and as a selectable marked; these were inserted into pbluescriptKS+ to create pblueKS-
ADE2target. TetR-Tomato was PCR-amplified from plasmid p402-TetR-Tomato (from S
Sabatinos) using primers TetR-Tom-F and TetR-Tom-R and inserted into PacI-digested pblueKS-
ADE2target to generate pTetR-Tom-ADE2. 2.1kbp KpnI-SacI fragment containing LacI-GFP was
subcloned from pAFS135 (from J Bachant) into pRS404 digested with same enzymes to create
p404-LacI-GFP. Plasmids containing tetO (pGS004 from J Bachant) or lacO (pJBN164 from J
Bachant) arrays were modified by introduction of genomic sequences to target integration near
different origins. The following primer pairs were used to generate sequences adjacent to the
16
indicated origins (with the corresponding chromosomal coordinates given in parentheses): primers
ARS501-tetO-F and ARS501-tetO-R for ARS501 (V:547812–548329), primers ARS1103- tetO-
F and ARS1103-tetO-R for ARS1103 (XI:54673–54996) and primers ARS1303-tetO-F and
ARS1303-tetO-R for ARS1303 (XIII:31983–32247), and these were inserted into KpnI+ClaI
digested pGS004 yielding pARS501-tetO, pARS1103-tetO, and pARS1303-tetO, respectively.
Likewise, the following primer pairs were used to generate sequences adjacent to the indicated
origins: primers ARS710-lacO-F and ARS710-lacO-R for ARS710 (VII:204305–204831),
primers ARS718-lacO-F and ARS718-lacO-R for ARS718 (VII:422375–423281), and primers
ARS1018-lacO-F and ARS1018-lacO-R for ARS1018 (X:539662–540395), and these were
inserted into XhoI+KpnI digested pJBN164 yielding pARS710-lacO, pARS718-lacO, and
pARS1018-lacO, respectively. Two adjacent regions near ARS305 were PCR-amplified using
primer pair NotI-ARS305-5’ and XhoI-ARS305-5’ (III:37283–37778) and primer pair NotI-
ARS305-3’ and KpnI-ARS305-3’ (III:37779–38282) and digested with NotI and XhoI and NotI
and KpnI, respectively; these fragments were ligated into pRS404 digested with XhoI and KpnI.
The XhoI-KpnI fragment was subcloned by digestion and ligation into pJBN164 digested with
same enzymes to yield pARS305-lacO. Plasmids p501∆-ARS305-DACS and p501∆-ARS305-
D2BS were created by mutagenesis of p501∆-ARS305 (Peace et al., 2016) with primers ARS305-
∆ACS- mut1, ARS305-∆ACS-mut2 and ARS305-∆2BS-mut1, and ARS305-∆2BS-mut2,
respectively. Plasmid p404-ars305∆-BInc was constructed as described for p306-ars305∆-BrdU-
Inc (Zhong et al., 2013) except that p404-BrdU-Inc (Viggiani and Aparicio, 2006) was used instead
of p306-BrdU-Inc. p404- ars305∆-BInc was digested with PmlI and KpnI, blunted-ended with
Klenow, and ligated with T4 DNA ligase to remove the TRP1 selectable marker, yielding p400-
ars305∆-BInc. The 1.5 kb SalI-SpeI fragment containing the KanMx cassette from pFA6-KanMx
17
(Longtine et al., 1998) was ligated into SalI-SpeI-digested p400-ars305∆-BInc, creating pKanMx-
ars305∆-BInc. The cdc45-1 allele was PCR- amplified from strain YB298 (from B Stillman) with
primers Cdc45-F and Cdc45-R and inserted into SacI+KpnI digested pRS406 (Sikorski and Hieter,
1989) to create p406-cdc45-1.
2.2 Yeast strain constructions
All strains are congenic with SSy161, derived from W303-1a (RAD5) (Viggiani and
Aparicio, 2006); complete genotypes are given in Supplementary file 3. Strain constructions were
carried out by genetic crosses or lithium acetate transformations with linearized plasmids or PCR
products generated with hybrid oligonucleotide primers having homology to target loci (Ito et al.,
2001; Longtine et al., 1998); primer sequences for strain constructions are given in Supplementary
file 2. Genomic alterations were confirmed by PCR analysis or DNA sequence analysis as
appropriate.
FKH1 was deleted using primers Fkh1-up and Fkh1-down to amplify KanMx selectable
marker from pFA6-KanMx (Longtine et al., 1998). FKH2 was replaced by fkh2-dsm in two steps:
first, FKH2 was entirely replaced with URA3 (C. albicans) using pAG61 (Addgene), and the
resulting strain was transformed with fkh2-dsm DNA from p405-fkh2-dsm (Ostrow et al., 2017)
followed by selection on 5-FOA. GAL-FKH1 was introduced using p405-GAL-FKH1 and FKH1
was FLAG-tagged as described previously (Peace et al., 2016). ARS501 was replaced by ARS305
or mutant versions of ARS305 by transformation with p501∆-ARS305, p501∆-ARS305-∆ACS,
or p501∆-ARS305-∆2BS as described previously (Peace et al., 2016). BrdU incorporation cassette
was introduced, replacing ARS305, by transformation with BglII-digested p404-ARS305-
BrdUInc. The cdc28-as1 allele was introduced by pop-in/pop-out of plasmid pJUcdc28-as1
18
digested with HindIII. The cdc7-as3 allele was introduced as described previously (Zhong et al.,
2013); cdc7-4 was back-crossed from H7C4A1 (from L Hartwell) into the W303 background four
times, with the final cross to HYy151. MCM4-DD/E+DSP/Q (referred to in text as MCM4-14D)
was introduced by transformation with PacI-digested pJR179 (from SP Bell). The cdc45-1 allele
was introduced by crossing with strain YB298 (from B Stillman) or by pop-in/pop-out with BglII-
digested p406-cdc45-1. The dbf4∆C allele was constructed by insertion of a non-sense codon with
the KanMx cassette from pFA6-KanMx (Longtine et al., 1998) using primers Dbf4-up and Dbf4-
down. TetR-Tomato was introduced by transformation with PacI-digested pTetR-Tom-ADE2.
LacI-GFP was introduced by transformation with HindIII-digested p404-LacI-GFP. The tetO or
lacO arrays were introduced by transformation with pARS501-tetO, pARS1103-tetO, pARS1303-
tetO, pARS305-lacO, pARS710-lacO, pARS718- lacO and pARS1018-lacO digested with PacI,
PshAI, BlpI, NotI, PshAI, SnaBI, and BlpI, respectively.
2.3 Cell growth synchronization
Cells were grown at 25 ̊C unless otherwise indicated. For microscopy, cells were grown in
complete synthetic medium supplemented with 15 mg/mL adenine (CSM+ade) +2% dextrose,
unless otherwise indicated (raffinose or galactose); for QBU and ChIP-seq, cells were grown in
YEP +2% dextrose, unless otherwise indicated (raffinose or galactose). G1 arrest was achieved by
incubation with 2.5 nM (1x) a–factor (Sigma T6901); for most extended arrests, a fresh or
additional dose of a–factor was added at time of induction/non-induction or at time of temperature
shift as indicated in figure legend. PP1 (Tocris Biosciences) was added to 25 mM at the time of
initial a–factor incubation. Reagents are listed in Supplementary file 4.
19
2.4 Live-cell fluorescence microscopy and image analysis
~5x10
6
cells were harvested by centrifugation and spread on agarose pads made of
CSM+ade +4% dextrose. A DeltaVision wide-field deconvolution microscope was used to capture
28 Z-stacks in 0.25 mm increments for each image. SoftWorX software (Applied Precision/GE
Healthcare) was used for deconvolution and three-dimensional reconstruction of nuclei, and for
measuring the distance between replication origins and nuclear periphery. For experiments with
mutant strains having irregularly shaped nuclei (e.g.: fkh1∆ fkh2-dsm), measurements were made
in three-dimensions; other- wise, measurements were made in two dimensions using a few middle
sections as previously described (Ryu et al., 2015). A z-test was applied to compare the distribution
of measured distances. Images are max intensity projections of two to four middle Z-stacks.
2.5 Quantitative BrdU Immunoprecipitation (QBU)
QBU and analysis of sequencing reads was performed as described previously using KAPA
Hyper Prep Kit (KK8504) (Haye-Bertolozzi and Aparicio, 2018). Data analysis was performed
using 352 replication origins classified as Fkh-activated, Fkh-repressed, or Fkh-unregulated (Knott
et al., 2012).
2.6 Chromatin immunoprecipitation analyzed by sequencing (ChIP)
ChIP-seq and analysis of sequencing reads was performed as described previously using
KAPA Hyper Prep Kit (KK8504) (Ostrow et al., 2015). Data analysis was performed using 95
replication origins classified as Fkh-activated (Knott et al., 2012).
2.7 Time-lapse video and MSD analysis
20
A DeltaVision wide-field deconvolution microscope was used to capture 20 Z-stacks in
0.30 mm increments for each time point. GFP signals were imaged every 12 s for 5 min, with 0.1
s exposure for each Z-stack and 32% of transmitted light using an LED source. All time-lapse
movies were deconvolved using SoftWoRx. At least 20 individual cells with nearly stationary
nuclei were used to track the trajectory of origin focus for each strain using Imaris (Bitplane), and
MSD curves, Rc, and volumes were derived as previously described (Caridi et al., 2018); the error
bars represent standard error.
Results
2.8 Fkh1-induced origin activation re-positions a subtelomeric origin in G1 phase
The association between replication timing and subnuclear localization of replication
origins and the requirement of Fkh1/2 for the clustering of early-firing replication origins
according to chromosome conformation capture studies led us to examine whether Fkh1 has any
role in establishing the spatial positioning of origins within the nucleus. We adapted a system that
we recently engineered that restores origin timing by induction of FKH1 expression in G1-arrested
FKH1/2 mutant cells (Peace et al., 2016). This system has Fkh-activated origin ARS305 moved
into a well characterized, late-replicating, subtelomeric region of chromosome V-R, replacing the
endogenous, late-firing ARS501 (Figure 2.1A). In this context, we showed previously that
ARS305
V-R
fails to replicate early in fkh1∆ fkh2∆ cells. However, induction of FKH1 expression
in these cells in G1-phase results in early-firing of ARS305
V-R
in the ensuing S-phase. In the
current study, we used fkh1∆ fkh2-dsm instead of fkh1∆ fkh2∆ cells; fkh1∆ fkh2-dsm cells are
essentially null for replication timing control, but exhibit more normal growth and particularly,
more normal cell and nuclear morphologies favorable for cytological analysis (Ostrow et al.,
21
2017). To locate ARS305
V-R
(or ARS501) in vivo, we introduced tandem repeats of tetO binding
sites adjacent to the origin and expressed TetR-Tomato protein (Figure 2.1A); we also expressed
Nup49-GFP (Nup49 is a nuclear pore protein) to illuminate the nuclear envelope (Belgareh and
Doye, 1997).
Microscopic examination of cells showed a single Tomato focus per undivided nucleus
(Figure 2.1B). Images of cells from an unsynchronized population were sorted according to
budding morphology, which is reflective of cell cycle progression. The localization of the
ARS305
V-R
-Tomato focus correlates with cell cycle stage, showing primarily peripheral
localization in unbudded and small-budded cells and interior localization in larger-budded cells
(Figure 2.1B). This is consistent with previous studies showing peripheral localization of
subtelomeric/late-firing origins in G1 followed by relocalization to the interior during S phase
(Heun et al., 2001a). Because origin timing is normally established in G1, we focused further
analysis on origins in G1 phase cells.
In G1-arrested fkh1∆ fkh2-dsm cells, almost all cells exhibited peripheral localization of
ARS305
V-R
(Figure 2.1C, left panel). Induction of FKH1, however, resulted in an increase in the
proportion of cells with non-peripheral positioning of ARS305
V-R
(Figure 2.1C, right panel),
suggesting that origin relocalization is associated with initiation timing re-programming by FKH1.
We confirmed that relocalization is a direct result of FKH1 induction by demonstrating that neither
the induction scheme (raffinose -> galactose) with a strain lacking inducible FKH1 nor a non-
inducing change to a more favorable carbon source (raffinose -> dextrose) resulted in origin
relocalization (Figure 2.1). To confirm the change in origin localization resulting from FKH1
induction, we created three-dimensional image reconstructions from confocal z-stacks and
measured the shortest distance in three dimensions from the ARS305
V-R
focus to Nup49-GFP
22
Figure 2.1 Fkh1-induced origin activation re-positions a subtelomeric origin in G1 phase. (A) Schematic of
chromosome V-R showing tetO repeats inserted adjacent to the ARS501 locus, which has been replaced with ARS305
(designated ARS305
V-R
); TetR-Tomato binds to and illuminates the locus as a single focus. (B) Images of cells from
an unsynchronized culture of strain HYy132 (fkh1∆ fkh2-dsm ARS305
V-R
-Tomato NUP49-GFP GAL-FKH1) are
shown sorted according to cell morphology; all images are at the same magnification: scale bar = 0.5 µm. (C) FKH1
induction scheme: HYy132 cells grown at 25 ̊C in raffinose medium were arrested in G1 phase with 1x - factor 2.5
hr, incubated an additional 2 hr in raffinose (Non-induction) or galactose (Fkh1-induction) with 1.7x - factor, and
images of live cells captured, examples of which are shown; scale bar = 0.5 µm. (D) The shortest V-R distance from
the ARS305
V-R
-Tomato focus to the nuclear periphery (Nup49-GFP) in each cell was measured and plotted as quartile
boxplots (median shown as thick black segment) for non-induction and FKH1-induction; the result of a z-test
comparing the two distributions is given as P. (E) Cells of fkh1∆ fkh2-dsm GAL-FKH1 NUP49-GFP strains HYy119
(ars305-∆ACS -Tomato) and HYy120 (ars305-∆2BS -Tomato) were treated and analyzed as above.
23
signal in the nuclear envelope amongst populations of cells. Statistical analysis of these
measurements shows a significant increase in the distances associated with FKH1-induction versus
non-induction (Figure 2.1D).
We tested whether the function of ARS305 is required for relocalization by introducing
into the V-R locus ARS305 bearing a mutation of the ARS consensus sequence (ACS) (ars305-
∆ACS), which is essential for ORC binding and origin function. Disruption of ARS305 function
not only eliminated its relocalization in response to FKH1 induction but also resulted in an even
more peripheral distribution, suggesting that a functional origin is required for relocalization away
from the periphery (Figure 2.1E). We also tested ARS305 with mutations of two proximal Fkh1/2
binding sites (ars305-∆ 2BS), which retains origin function but is delayed in activation at its
normal locus (Knott et al., 2012); ars305-∆2BS
V-R
did not relocalize upon FKH1 induction,
confirming that Fkh1 acts through direct binding in cis to ARS305
V-R
(Figure 2.1E).
In the experiments above, relocalization of ARS305
V-R
involved induction of FKH1 from
the GAL1/10 promoter, which results in higher than normal levels of Fkh1 protein (Peace et al.,
2016). To determine whether this overabundance of Fkh1 was required for the origin
relocalization, we compared localization of ARS305
V-R
with ARS501 in cells with native FKH1
(and FKH2) expression (Figure 2.2A). The analysis showed that ARS305
V-R
was significantly
more distant from the nuclear periphery than ARS501 (Figure 2.2B). We also analyzed origin
timing of ARS305
V-R
in FKH1 (fkh2-dsm) versus fkh1∆ (fkh2-dsm) cells by quantitative BrdU
immunoprecipitation (QBU) of cells released from G1 phase into hydroxyurea (HU), in which
early but not late origins fire efficiently. We found that ARS305
V-R
fired efficiently in HU in FKH1
but not fkh1∆ cells (Figure 2.2C and Figure 2.2). Thus, normal Fkh1 levels are able to overcome
24
the effect that subtelomeric location has V-R on subnuclear localization and initiation timing of
ARS305.
Figure 2.2 Normal dosage of Fkh1 is sufficient to relocalize ARS305
V-R
and advance its firing time. (A) HYy160 (ARS501-Tomato
NUP49-GFP) and HYy157 (ARS305
V-R
-Tomato NUP49-GFP) cells were arrested in G1 phase at 25 ̊C with 1x - factor 2 hr and
images were collected; scale bar = 0.5 µm. (B) Distances from origin foci to nuclear periphery were determined, plotted as quartile
boxplots, and analyzed by a z-test. (C) Quantitative BrdU-IP-Seq (QBU) analysis was performed with ARS305
V-R
-bearing strains
HYy113 (fkh2-dsm) and HYy38 (fkh1∆ fkh2-dsm) after G1 block- and-release into hydroxyurea in the presence of BrdU; averaged
data from three experimental replicates was plotted for the V-R region with the positions of several replication origins indicated;
ARS305
V-R
resides at the ARS522 (aka: ARS501) locus.
25
2.9 Fkh1 globally regulates subnuclear positioning of early origins in G1 phase
We tested whether the Fkh1-dependent localization of ARS305
V-R
is also responsible for
ARS305 localization when residing at its native locus more distal from the telomere. We inserted
a lacO array near ARS305 and expressed LacI-GFP and Nup49-GFP; imaging showed that the
LacI-GFP focus is clearly distinguishable from the more diffuse Nup49-GFP signal (Figure 2.3A).
Consistent with previous analysis, ARS305 was non-peripheral in most G1-arrested WT cells
(Figure 2.3A) (Heun et al., 2001a), however, deletion of FKH1 significantly increased the
proportion of cells in which ARS305 was closer to the periphery (Figure 2.3A). Consistent with
this requirement for FKH1, elimination of the Fkh1/2 binding sites in ARS305 also resulted in
peripheral localization (Figure 2.4). Moreover, ars305-∆ACS, which lacks origin function (and
Fkh1/2 binding) exhibits an even more peripheral distribution (Figure 2.4A), suggesting that origin
function is required for interior localization in G1 and that Fkh1/2 stimulates this localization.
Previous analysis showed that ARS305 initiation timing was significantly delayed in the absence
of FKH1 or Fkh1/2 binding sites (Knott et al., 2012), so once again we observe a FKH1-dependent
relationship between subnuclear localization in G1 phase with replication initiation timing in S
phase.
To determine whether other Fkh-activated origins’ localizations are also determined by
FKH1, we performed similar tests by inserting a tetO array adjacent to a few additional Fkh-
activated origins, expressing TetR-Tomato and Nup49- GFP, and deleting FKH1. Like ARS305,
ARS1303 and ARS1103 were located closer to the periphery in G1-arrested fkh1D versus WT
cells (Figure 2.3B). ARS305, ARS1303 and ARS1103 are relatively telomere proximal, residing
39, 32, and 56 kb from the nearest telomere respectively (the lacO or tetO arrays add ~14 or 16
kb, respectively to these distances), which might constrain the extent to which FKH1 influences
26
Figure 2.3 Fkh1 determines early origin positioning globally. Diagrams of chromosomes with replication origins
labeled with lacO/LacI-GFP (green-filled segment) or tetO/TetR-Tomato (red-filled segment) are shown above the
corresponding images. Distances between origins (black-filled spheres) and telomeres, and in some cases centromeres
(ovals), are indicated and include ~14 kb or 16 kb added by lacO or tetO repeats, respectively; elements are not drawn
to scale. Cells of WT and fkh1∆ strains with ARS305-GFP (HYy151, HYy147) in (A), ARS1303-Tomato (HYy166,
HYy173) and ARS1103-Tomato (HYy165, HYy172) in (B), ARS710-GFP (MPy6, MPy10), ARS718-GFP (MPy20,
MPy21), and ARS1018-GFP (MPy19, MPy22) in (C), all expressing NUP49-GFP, were arrested in G1 phase at 25 ̊C
with 1x α-factor 2 hr and live images were captured; scale bar = 0.5 µm. Distances from origin foci to nuclear
periphery were determined, plotted as quartile boxplots, and analyzed by a z-test.
27
Figure 2.4 Fkh-binding sites and a functional ACS are required for Fkh1 to influence early origin positioning globally.
Cells with ARS305-GFP (HYy151), ARS305-∆2BS-GFP (MPy46), and ars305-∆ACS-GFP (MPy43) all expressing
NUP49-GFP, were treated and analyzed as in Figure 2.3 legend; scale bar = 0.5 μm. (B) Cycling cultures of WT and
fkh1∆ strains with ARS305-GFP (HYy151, HYy147) and ARS710-GFP (MPy6, MPy10) were imaged and G1 phase
(unbudded) cells were analyzed as described in Figure 2.3 legend. (C) Cells of WT and fkh1∆ strains with ARS501-
Tom (HYy198, HYy201) were arrested in G1 phase and analyzed as described in Figure 2.3 legend.
28
their positioning. To address this possibility, we tested localization of several additional Fkh-
activated origins that are more distal from telomeres, including: ARS710, ARS718, and ARS1018,
residing at 204, 421, and 205 kb from the nearest telomere. All of these origins show significant
reduction in distance from the nuclear periphery upon deletion of FKH1 (Figure 2.3C). We also
observed similar results with ARS305 and ARS710 in G1 cells from an unsynchronized population
(Figure 2.4B). In contrast, the peripheral localization in G1 of late origin ARS501, which is not
Fkh-activated, was not altered by deletion of FKH1 (Figure 2.4C). These results suggest that FKH1
plays an expansive role in relocalizing replication origins from the nuclear periphery to the nuclear
interior in G1 phase.
2.10 DDK- but not CDK-dependent step of replication initiation drives origin relocalization
Because our previous study indicated that Fkh1/2 was required for origin recruitment of
Cdc45 in G1 phase (Knott et al., 2012), we tested the requirement for CDC7, which encodes the
catalytic subunit of DBF4-dependent kinase (DDK), and is required for Cdc45 origin-loading
(reviewed in Tanaka and Araki, 2013). We introduced the cdc7-as3 allele, the kinase activity of
which is inhibited by ATP analog PP1 (Wan et al., 2006), and tested whether FKH1- induced
origin relocalization occurs with inhibition of CDC7 function. Remarkably, ARS305
V-R
relocalization was eliminated by inhibition of Cdc7-as3 kinase with PP1 (Figure 2.5A, compare
with non-induction in Figure 2.1D). These results suggest that DDK activity is required for origin
relocalization in G1-arrested cells.
A role for DDK in G1 phase was unexpected as DDK activity has been reported to be low
in α-factor-arrested G1 cells due to instability of Dbf4 (Nougarède et al., 2000; Oshiro et al., 1999).
To provide further evidence for DDK’s role, we tested a native origin without FKH1
29
overexpression, and to inactivate CDC7, we chose the temperature-sensitive cdc7-4 allele
(Hereford and Hartwell, 1974). For this experiment, G1-arrested WT and cdc7-4 cells bearing a
lacO array inserted near ARS305 and expressing LacI-GFP were shifted to the non-permissive
temperature and ARS305 location was determined. Compared to WT cells, cdc7- 4 cells at the
non-permissive temperature showed a significant increase in the proportion of cells with ARS305
near the nuclear periphery (Figure 2.5B). This result supports the conclusion that DDK activity is
required for origin re-positioning in G1 phase cells. Fkh1-origin binding is cell cycle-regulated,
occurring in G1 and S phases (Ostrow et al., 2014), suggesting that the requirement for DDK
activity in Fkh1-stimulated origin relocalization might be due to dependence of Fkh1 origin-
binding on DDK. We tested this possibility by performing chromatin immunoprecipitation
analysis of Fkh1 comparing WT and cdc7-as3 cells. The results showed that binding of Fkh1 to
ARS305 and other Fkh-activated origins was largely unaffected by Cdc7 inhibition. Thus, Fkh1
origin binding appears to be independent of DDK activity, and, by inference, of the subcellular
change in localization resulting from DDK inhibition. Alternatively, the requirement for DDK
activity in Fkh1-stimulated origin relocalization may reflect Fkh1 acting upstream of DDK, which
would comport with a recent report that a critical role of Fkh1 in origin stimulation is DDK
recruitment through direct physical interaction with Dbf4 (Fang et al., 2017). We tested whether
the same mechanism is responsible for Fkh1-induced origin repositioning by testing the effect on
ARS305 positioning in cells expressing Dbf4 lacking its C-terminus (dbf4∆C), which is required
for interaction with Fkh1 (Fang et al., 2017). Deletion of DBF4’s C-terminus had a similar effect
on origin localization as FKH1 deletion, with greater enrichment of ARS305 near the nuclear
periphery (Figure 2.5C, compare with Figure 2.3A), consistent with Fkh1 and Dbf4 acting in the
same pathway.
30
Figure 2.5 Origin localization in G1 is DDK regulated. (A) HYy186 (fkh1∆ fkh2-dsm GAL-FKH1 ARS305
V-R
-Tomato
NUP49-GFP cdc7-as3) cells were subjected to FKH1-induction scheme as described in Figure 1C legend except that
PP1 or DMSO (vehicle) was included with α-factor, and images were captured. (B) Cells of ARS305-GFP NUP49-
GFP strains HYy151 (WT) and HYy191 (cdc7-4) were arrested in G1 with 1x α-factor 2 hr at 25 ̊C followed by 1 hr
incubation at 37 ̊C with 2x α-factor, and images were captured. (C) HYy181 (ARS305-GFP NUP49-GFP dbf4∆C)
cells were arrested in G1 phase with 1x α-factor 2 hr at 25 ̊C and live images were captured. The control experiment
with WT cells (HYy151) is shown in Figure 2.3A. (A–C) Scale bar = 0.5 µm. Distances from origin foci to nuclear
periphery were determined, plotted as quartile boxplots, and analyzed by a z-test.
31
The essential function of DDK in origin firing is phosphorylation of MCM helicase
subunits, particularly Mcm4, resulting in removal of auto-inhibition and enabling recruitment of
helicase accessory protein Cdc45 through its loading factor Sld3 (reviewed in Tanaka and Araki,
2013). To test whether the requirement for Cdc7 kinase activity in origin relocalization reflects its
function in Mcm4 helicase phosphorylation, we introduced into the cdc7-as3 strain an allele of
MCM4, MCM4-DD/E(7) +DSP/Q(7) abbreviated herein as MCM4-14D, which contains 14 S/T-
>D substitutions that mimic critical DDK-phosphorylated residues in Mcm4, and suppresses
reduced Cdc7 kinase activity (Randell et al., 2010). The presence of MCM4-14D restores
ARS305
V-R
relocalization upon FKH1 induction in the cdc7-as3 strain inhibited by PP1 (Figure
2.6A). This supports the conclusion that the function of Cdc7 kinase required for origin
relocalization is phosphorylation of Mcm4.
We tested whether completion of the DDK-dependent step, that is Sld3 and Cdc45 loading,
is required for origin relocalization by testing the effect of inactivation of CDC45 function. The
cold-sensitive cdc45-1 allele exhibits interdependence with heat-sensitive alleles cdc7-4 and dbf4-
1 in reciprocal temperature-shift experiments, tightly inhibits replication initiation, and reduces
Sld3-origin association in G1 phase (Aparicio et al., 1999; Kamimura et al., 2001; Owens et al.,
1997). We synchronized WT and cdc45-1 cells in G1 phase at the permissive temperature and
shifted the cultures to semi-permissive temperature while maintaining the G1 arrest. Analysis
showed that ARS305 was more peripherally localized in cdc45-1 cells at the semi-permissive
temperature in G1 phase (Figure 2.6B). As origin binding of Cdc45 and Sld3 is interdependent and
Cdc45-1 inactivation reduces Sld3-origin binding (Kamimura et al., 2001), these results suggest
that assembly of Sld3- Cdc45 onto origins is required for origin relocalization.
32
Cdc45 is incorporated into replisomes as a component of the active helicase complex
together with MCM and GINS. However, Cdc45 is present in low abundance and is likely limiting
for the total number of active replisomes that may be simultaneously active (Mantiero et al., 2011;
Tanaka et al., 2011). We noticed that the presence of Mcm4-14D was not sufficient to relocalize
ARS305
V-R
in fkh1∆ fkh2-dsm cells in the absence of Fkh1 induction (Figure 2.6C), which might
be contrary to expectations if the only function of Fkh1 is to physically recruit DDK, which has
been rendered dispensable by Mcm4-14D. We note, however, that MCM4-14D does not suppress
a deletion of CDC7 (S.P. Bell, personal communication) suggesting that residual DDK activity is
required for sufficient origin firing, and hence, Fkh1 may act to target this residual activity to
specific origins. Absent this targeting, we postulated that the limited abundance of Cdc45 would
be further diluted amongst all licensed origins due to potentiation by Mcm4-14D. To test this idea,
we introduced a high-copy plasmid expressing Cdc45 from its native promoter into the fkh1∆
fkh2-dsm MCM4-14D strain, and examined origin location. Consistent with the notion that Cdc45
is limiting for execution of the DDK-dependent step, expression of high copy Cdc45 significantly
increased the frequency of ARS305 more distal from the periphery (Figure 2.6C). This finding
supports the conclusion that full execution of the DDK-dependent step in the form of Cdc45
loading, as opposed to Mcm4 phosphorylation itself or phosphorylation of other targets is required
for origin relocalization.
As the interior localization of early origins occurs in α-factor-arrested, G1 phase cells,
cyclin- dependent kinase (CDK) activity would appear to be dispensable because G1 phase cells
have very low levels of S/G2/M-CDK activities, and G1-CDKs, which are required for passage
through Start, are inhibited by α-factor (reviewed in Mendenhall and Hodge, 1998). Nevertheless,
low levels of S/ G2/M-CDK activities in G1 phase cannot be ruled out, and indeed, it appears that
33
Figure 2.6 DDK regulation of origin localization reflects its phosphorylation of Mcm4 and consequent Cdc45 loading.
(A) Fkh1-induction scheme with PP1 as described in Figure 4A legend was carried out with fkh1∆ fkh2-dsm GAL-
FKH1 ARS305
V-R
-Tomato NUP49-GFP strains HYy186 (cdc7-as3) and HYy177 (cdc7-as3 MCM4-14D), and images
captured. (B) ARS305-GFP NUP49-GFP strains HYy151 (WT) and HYy184 (cdc45-1) were arrested in G1 with 1x
α-factor 1 hr at 30 ̊C followed by 2 hr incubation at 16 ̊C with 1x α-factor, and images were captured. (C) Cells of
strain HYy177 harboring no plasmid or high-copy plasmid expressing CDC45 were arrested in G1 with 1x α-factor
2 hr at 25 ̊C and images captured. (D) ARS305-GFP NUP49-GFP strain HYy197 (cdc28-as1) cells were arrested in
G1 phase with 0.5x α-factor 2 hr at 25 ̊C, PP1 or DMSO was added and incubated one additional hour with 0.5x α-
factor, and images were captured. (A–D) Scale bar = 0.5 µm. Distances from origin foci to nuclear periphery were
determined, plotted as quartile boxplots, and analyzed by a z-test.
34
low levels of DDK are involved. Thus, to address the possibility that CDK activity might be
contributing to G1 phase origin dynamics, we tested the requirement for CDC28, the Cdk1 kinase,
using analog-sensitive cdc28-as1 cells (Bishop et al., 2000). In G1-arrested cells, inhibition of
Cdc28-as1 activity with PP1 did not alter localization of ARS305 (Figure 2.6D), although budding
was inhibited upon release from α-factor arrest indicating effective inhibition of Cdc28-as1
(supplement online). Similarly, PP1 treatment of cycling cdc28-as1 cells did not alter distribution
of ARS305 in G1 phase cells (supplement online), while DNA content analysis showed delayed
entry of cells into S phase indicating effective inhibition of Cdc28-as1 (supplement online). Thus,
CDK activity appears to be dispensable for normal, Fkh1-dependent positioning of ARS305.
Overall, our findings indicate that DDK but not CDK activity stimulates origin relocalization in
G1 phase.
2.11 Origin mobility increases with origin relocalization
Fkh1 might facilitate origin relocalization by promoting origin mobilization (release from
the periphery or movement per se), or by increasing the stability of origin-origin interaction after
relocalization. In addition to changes in location, replication origins exhibit decreased rate of
mobility during progression into S phase (Heun et al., 2001b). We directly investigated how Fkh1
affects origin mobility by tracking the locations of ARS305 and ARS718 in individual WT and
fkh1∆ cells over time, and applying mean square displacement (MSD) analyses (Marshall et al.,
1997). The analysis shows significantly lower plateau of MSD curves in fkh1∆ cells (Figure 2.7A),
consistent with less nuclear space explored. Calculation of the radius of constraint (Rc) and the
corresponding volume of space explored reveals that ARS305 explores about 2.5-fold more
35
volume and ARS718 explores about 3.8- fold more volume in WT than fkh1∆ cells. Tracings of
the paths of origin foci show confinement proximal to the nuclear periphery in fkh1∆ cells (Figure
2.7B). Together, these data show that Fkh1 stimulates origin mobilization.
Figure 2.7 Origin mobility increases with origin relocalization(A) Mean-squared-displacement (MSD) analysis of
tracking data for ARS305-GFP strains HYy151 (WT) and HYy147 (fkh1∆) and ARS718-GFP strains MPy20 (WT)
and MPy21 (fkh1∆). Radius of constraint (Rc) and volume searched (V) are given, and statistical significance
comparing WT and fkh1∆ was estimated by two-tailed Mann-Whitney test. (B) Images (left) and 3D reconstructions
with Imaris (right) showing examples of tracks of origin focus over time (color corresponding to time progression);
scale bar = 0.4 µm. Movies of the individual ARS305 time-courses are available from eLife.
36
Discussion
This study reveals new links between key molecular interactions in replication initiation
and the localization and mobility of replication origins within the nucleus. In particular, we show
that early origin specification in G1 phase by Fkh1 induces a change from peripheral to interior
nuclear localization of Fkh1-activated origins. Quite remarkably, we find that origin relocalization
requires execution of the DDK-dependent step of origin firing that loads Cdc45. That the DDK
requirement reflects the key, recognized function of DDK in replication initiation, that is,
phosphorylation of MCM proteins leading to Sld3-Cdc45 origin-loading, is demonstrated by the
bypass of CDC7 requirement by phospho- mimetic mutations in MCM4-14D as well as the
dependence on CDC45 function. This early execution of the DDK step was unexpected because
DDK levels have been reported to be very low in α- factor-arrested G1 cells due to Dbf4 instability
(Nougarède et al., 2000; Oshiro et al., 1999). Our findings provide direct evidence that DDK is
active in G1 phase and has already established origin timing by late G1 phase in α-factor arrest.
This finding explains previous observations that Sld3 and Cdc45 associate with early replication
origins in G1 phase (Aparicio et al., 1999; Kamimura et al., 2001). Our findings are also consistent
with a more recent study showing that Sld3- and Cdc45-origin association in G1 phase is DDK-
dependent and CDK-independent, as well as the conclusion that DDK acts prior to and
independently of S-CDK (Heller et al., 2011; Yeeles et al., 2015), the latter of which is dispensable
for the observed origin relocalization.
The finding that Fkh1 and DDK are required for origin relocalization fits well with the
recent finding that Fkh1 acts to stimulate origin firing by directly recruiting Dbf4 through physical
interaction (Fang et al., 2017), and extends our understanding of the significance of this interaction
to replication initiation via nuclear positioning of replication sites. As predicted by this interaction
37
model, inactivation of DDK activity should phenocopy deletion of FKH1, as we have herein
demonstrated with depletion of CDC7 function. Moreover, specific deletion of Dbf4’s C-terminus,
which is required for interaction with Fkh1, also phenocopies deletion of FKH1. Furthermore, the
absence of FKH1 and FKH2 function is bypassed by the MCM4-14D allele in the presence of
increased levels of Cdc45. Together, these results support a mechanism involving Fkh1
recruitment of DDK activity to load Cdc45 at a subset of origins in G1, corresponding with a
change in subnuclear positioning of these origins, and early firing in the subsequent S phase.
Chromosome conformation capture (Hi-C) experiments have indicated that early firing
replication origins preferentially interact with each other, or ‘cluster’ in G1 phase (Duan et al.,
2010). Related studies have shown that Fkh1/2 is required for these spatial interactions amongst
early origins (Eser et al., 2017; Knott et al., 2012; Ostrow et al., 2017). We propose that the origin
clustering interactions revealed by Hi-C experiments directly reflect origin localization to distinct
nuclear territories as observed microscopically. Thus, localization to the nuclear interior might
increase the likelihood for physical interaction amongst this subset of origins. Such interactions
may be driven by cooperative interactions between Fkh1-bound origins recruiting limiting
initiation factors such as Dbf4, Sld3 and Cdc45. This aggregation of origins selected for
early/efficient activation has the inevitable consequence that replication initiation transforms these
origin clusters into replication foci, which have been observed as concentrations of DNA synthesis
and replication factors (Berezney et al., 2000; Frouin et al., 2003; Hozák et al., 1994; Kitamura et
al., 2006; Nakamura et al., 1986; Newport and Yan, 1996). These assemblages may contribute to
efficient chromosomal replication initiation and elongation in multiple ways, such as accretion of
activities and co-factors directly required for DNA synthesis (e.g.: dNTP production), and
38
scaffolding to co-localize and coordinate replication with related activities like chromatin
assembly, cohesion establishment, topological resolution, and DNA repair.
It remains to be determined exactly what maintains either the peripheral or interior
localization of origins or what drives relocalization between different subnuclear zones. While
telomere tethering to the nuclear envelope has been assumed to cause the peripheral localization
of telomere-proximal origins, we find that early origins distal from telomeres that are normally
enriched in the nuclear interior, are closer to the nuclear periphery in cells lacking Fkh1, suggesting
that perinuclear localization represents a default state for most origins irrespective of telomere
tethering (Figure 2.8A). It is unclear what promotes this origin localization. Complete elimination
of origin function results in even more peripheral distribution of the locus suggesting that
peripheral localization is independent of an origin tethering mechanism, and that interior
localization is linked to activation of origin function, which may occur, with less efficiency, in the
absence of Fkh1/2. There may be passive exclusion from the interior where other activities like
transcription may predominate in early G1, or there may be a dedicated tethering mechanism,
though origin association with the periphery does not appear to be as stringently localized or as
stable as that of telomeres (Hediger et al., 2002; Heun et al., 2001a; Heun et al., 2001b).
In addition to subtelomeric origins, Rif1 regulates and associates with origins distal from
telomeres and with the nuclear envelope, and therefore could potentially tether origins to the
periphery (Figure 2.8A) (Hafner et al., 2018; Park et al., 2011; Peace et al., 2014). Rif1 interacts
with Dbf4 and with the counteracting PP1 phosphatase, suggesting that the Rif1-origin interaction
may be down- regulated by DDK-dependent phosphorylation of MCM proteins and/or Rif1 (Dave ́
et al., 2014; Hiraga et al., 2014; Mattarocci et al., 2014). Thus, Fkh1-mediated, origin-specific
recruitment of DDK may overwhelm Rif1-mediated PP1 inhibition locally and thereby release the
39
origin from peripheral tethering (Figure 2.8B). Consequent Cdc45 loading might effectively
prevent reversal of MCM phosphorylation and fully disrupt interaction with Rif1-PP1.
Alternatively, MCM phosphorylation and/or Cdc45 loading might change the licensed origin’s
biophysical properties, thereby forcing the origin to occupy and search different space and/or
capture scaffolding factors, which may themselves be localized to the interior and thus stabilize
interior localization. Future studies aimed at more detailed examination of how individual factors
affect origin mobility should provide further insights.
Figure 2.8 Model of origin localization linked to initiation. (A) Absent Fkh1, most replication origins are enriched at
nuclear periphery, however, Fkh1 binding to a subset of origins allows execution of the DDK-dependent step of
initiation, resulting in release from the nuclear periphery and/or capture in the nuclear interior to form early origin
clusters. (B) Hypothetical mechanism for origin tethering to the nuclear periphery regulated by Rif1-PP1 versus Fkh1-
DDK-Cdc45 activities. Rif1 associates with inner nuclear membrane and with licensed replication origins, while
associated PP1 antagonizes execution of the DDK-dependent step. Fkh1-dependent recruitment of DDK results in
phosphorylation of MCM, Cdc45 loading and local release from Rif1 and PP1. See text for further discussion.
40
Previous studies have concluded that peripheral localization of origins is neither necessary
nor sufficient to regulate initiation timing. In one study, subtelomeric origin ARS501 remained
late firing following excision (in α-factor-arrested G1 cells) from the chromosome, which allowed
its diffusion away from the nuclear periphery, leading the authors to suggest that peripheral
localization might promote a chromatin mark that maintains late timing (Heun et al., 2001a).
However, we have shown that induction of Fkh1 (in α-factor -arrested cells) can reprogram timing
of a Fkh-activated origin inserted into the ARS501 locus (Peace et al., 2016). We propose that the
relevant chromatin mark is MCM phosphorylation, removal of which is promoted by peripheral
localization and addition by DDK recruitment. Thus, excised ARS501 remains late despite its
mobilization because DDK is limiting and already recruited by Ctf19 and Fkh1/2 to other origins.
Other studies have shown that tethering of early origins to the nuclear periphery does not delay
their activation (Ebrahimi et al., 2010; Zappulla et al., 2002). However, both origins in these
previous studies, ARS305 and ARS607, are Fkh-activated origins that we have shown can
overcome the replication initiation delay associated with peripheral localization. Overall, these
previous findings fit neatly into our model, which suggests that interior origin localization is a
consequence rather than a cause of early timing.
41
Chapter 3: Dbf4 Zn-finger motif is specifically required for stimulation of
Ctf19-activated origins in Saccharomyces cerevisiae
Adapted from:
Petrie MV, Zhang H, Arnold EM, Gan Y, Aparicio OM. (2022) “Dbf4 Zn-Finger Motif Is
Specifically Required for Stimulation of Ctf19-Activated Origins in Saccharomyces cerevisiae”.
Genes (Basel). 13(12):2202 doi:10.3390/genes13122202
Abstract
Eukaryotic genomes are replicated in spatiotemporal patterns that are stereotypical for
individual genomes and developmental profiles. In the model system Saccharomyces cerevisiae,
two primary mechanisms determine the preferential activation of replication origins during early
S phase, thereby largely defining the consequent replication profiles of these cells. Both
mechanisms are thought to act through specific recruitment of a rate-limiting initiation factor,
Dbf4-dependent kinase (DDK), to a subset of licensed replication origins. Fkh1/2 is responsible
for stimulation of most early-firing origins, except for CEN-proximal origins that recruit DDK via
the kinetochore protein Ctf19, which is required for their early-firing. The C-terminus of Dbf4
has been implicated in its recruitment to origins via both the Fkh1/2 and Ctf19 mechanisms. Here,
we show that the Zn-finger motif within the C-terminus is specifically required for Dbf4 to
functionally target CEN-proximal/Ctf19-dependent origins, whereas targeting to origins via the
Fkh1/2 pathway remains largely intact. These findings re-open the question of exactly how Fkh1/2
and DDK act together to stimulate replication origin initiation.
42
Introduction
Entry into S phase and initiation of replication of eukaryotic chromosomes is regulated by
multiple cell cycle kinases, including essential activators Cyclin-Dependent Kinase (CDK) and
Dbf4-Dependent Kinase (DDK). These major regulators are characterized by a kinase subunit
activated and targeted by a protein that exhibits cell cycle regulation of its expression and stability.
Dbf4 is the cell cycle-regulated, activating subunit of the Cdc7 kinase that together comprise DDK,
an essential DNA replication kinase (reviewed in (Gillespie and Blow 2022). As such, DDK is a
node of regulation linking checkpoint signaling and establishment of cohesion during the
replication process (Natsume et al. 2013). DDK’s essential function in replication initiation is to
phosphorylate N-terminal domains of Mcm4 and Mcm6 at licensed origins to relieve intrinsic
repression of MCM complex helicase activity (Sheu and Stillman 2010).
DDK is a rate-limiting activator of replication origins (Mantiero et al. 2011; Patel et al.
2008; Tanaka et al. 2011). As a result of DDK and other initiation proteins being present in limiting
quantities in relation to the number of licensed origins, only a subset of origins can initiate
replication simultaneously at the beginning of S phase in response to DDK and CDK activities.
Mechanisms that specifically target DDK to subsets of origins have been described, resulting in
strongly preferential activities of these origins (reviewed in (Aparicio 2013). For example, Fkh1
and/or Fkh2 (Fkh1/2), are responsible for the preferential activation of most early-firing origins in
the S. cerevisiae genome with the exception of CEN-proximal origins, which remain early firing
in fkh1∆ fkh2∆ cells (Knott et al. 2012). CEN-proximal origins are instead dependent on Ctf19, a
kinetochore protein, for their early efficient activities (Natsume et al. 2013). Through apparently
independent mechanisms, Fkh1/2 or Ctf19 is thought to recruit Dbf4, and hence DDK, to the
selected origins, resulting in their preferential early initiation (Fang et al. 2017; Knott et al. 2012;
43
Natsume et al. 2013; Zhang et al. 2019). Fkh1/2 binding sequences are intimately associated with
Fkh1/2-activated origins (Knott et al. 2012; Looke et al. 2012; Ostrow et al. 2014; Reinapae et al.
2017), whereas the Ctf19-kinetochore complex appears to recruit Dbf4 to the kinetochore itself,
resulting in stimulation of the nearest flanking origin(s), up to ~25kb distant (Natsume et al. 2013).
Despite the independent DDK-origin recruitment mechanisms, the C-terminal region of
Dbf4 has been implicated as a shared feature, required for both (Fang et al. 2017). The C-terminus
of Dbf4 contains a Zn-finger motif (CCHH), which was characterized previously, and suggested
to play a direct role in replication through interaction with Mcm2, and a role in replication stress
response (Jones et al. 2010). A subsequent study strongly suggested that the C-terminal domain
functions in replication of CEN-proximal origins, as simply appending one of several epitope-tags
on the C-terminus is sufficient to specifically ablate early firing of CEN-proximal origins; this was
further shown to phenocopy ctf19∆ regarding CEN-proximal origin firing and physical interaction
(Natsume et al. 2013). More recently, the C-terminus of Dbf4 was implicated in the Fkh1/2-
dependent stimulation of origins, suggesting direct physical interaction between the C-terminal
Zn-finger region and Fkh1 (Fang et al. 2017). These findings prompted us to follow up and
examine this proposed mechanism in greater depth.
In this work, we have carefully examined and compared the cellular growth and viability,
the rate of genome duplication, and the specific effects on individual origins genome-wide in
DBF4 mutants lacking the 50-amino acid Zn-finger domain or bearing point mutations precisely
targeting the Zn-finger motif. We compare these results with strains bearing mutations that
eliminate one or both origin-stimulating pathways, Fkh1/2 or Ctf19. Our findings show that
deletion of the entire C-terminal domain causes a severe defect in replication and viability. In
contrast, mutation of the Zn finger motif has specific effects on targeting to CEN-proximal versus
44
Fkh-activated origins. These results significantly revise the possible mechanisms involved in
origins stimulation, particularly of Fkh-activated origins.
Materials and Methods
3.1 Plasmid and yeast strain construction.
Oligonucleotide primer sequences for plasmid and strain constructions are given. Plasmid
pMP35 was constructed for pop-in/pop-out of dbf4-Zn* mutant alleles to create precise, unmarked
replacements of DBF4. Primers ∆N-dbf4_F and ∆N-dbf4_R were used to PCR-amplify an ∆N-
dbf4 fragment with homology tails to pRS406 (Sikorski and Hieter 1989). The fragment and
pRS406 were digested with KpnI-HF and SacI-HF and ligated (enzymes from New England
BioLabs). Zinc finger mutations were created in pMP35 with Quick-Change Lightning Multi Site-
directed mutagenesis kit (Agilent) using primers dbf4-Zn-aaHH and dbf4-Zn-CCHc to create
plasmids pMP36 and pMP38, respectively. Desired sequence changes were confirmed by DNA
sequencing. Plasmid pHY62 was constructed with primers TB_FLAG-swap and swap-FLAG-
Dbf4 replacing the 6xHA epitope in KpnI-NaeI-digested plasmid pT1892 (from T. Tanaka) with
3xFLAG epitope from plasmid p2L-3FLAG-TRP1 (from T. Tsukiyama) using the In-Fusion HD
Cloning Kit (Clontech).
Strain constructions were carried out by genetic crosses or lithium acetate transformations
with DNA from linearized plasmids or PCR products generated with hybrid oligonucleotide
primers having homology to target loci (Goldstein, Pan, and McCusker 1999; Ito et al. 1983;
Longtine et al. 1998); Genomic alterations were confirmed by PCR analysis and/or DNA sequence
analysis as appropriate. Yeast strain genotypes are given. All strains are congenic with W303
background, and most are derived from BrdU-incorporating strains CVy63 and CVy70, which are
derived from SSy161 and SSy162, respectively (Viggiani and Aparicio 2006). Primers dbf4∆C_F
45
and dbf4∆C_R were used to create a construct truncating DBF4 with selection
(dbf4∆C::HIS3MX6) and introduced into diploid of CVy63 x CVy70. The diploid was sporulated
and spores dissected and germinated at 23 °C; haploid spores were genotyped yielding strains
JOSHy1 and JOSHy2. JOSHy1 was transformed with KANMX module to replace HIS3MX
module to create HYy176. Deletions of FKH1, FKH2, and replacement of FKH2 with fkh2-dsm
have been described previously (He et al. 2022); HYy143 and OAy1123 were similarly
constructed. HYy215, HYy217 and HYy218 were obtained as segregants from a cross of HYy143
and JOSHy2. CTF19 was deleted using primers CTF19∆_F and CTF19∆_R and plasmid p2L-
3FLAG-TRP1(K. lactis) as template for the selectable marker in diploid of HYy143 x JOSHy2;
sporulation and dissection yielded strains HYy207, HYy210 and HYy211. MPy74 and MPy76
were constructed by transformation of CVy63 with pMP36 and pMP38 digested with EcoNI.
Following selection for URA3 and confirmation of proper integration (pop-in), non-selective
growth allowed for isolation of 5-FOA-resistant clones (pop-out); clones were sequenced to
confirm replacement of DBF4 by dbf4-Zn*1 or dbf4-Zn*2. MPy86 and MPy90 were derived from
MPy74 and MPy76, respectively, through a cross with OAy1123. N-terminal 3xFLAG-tagging of
DBF4 to construct MPy125 was accomplished by transformation of MPy35 with XhoI-digested
pHY62; construction was confirmed by PCR and immunoblotting. EAy1 and EAy2 are sister
isolates from pop-in/pop-out of pMP36 into strain MPy125 to introduce dbf4-Zn*1; constructions
were confirmed by sequencing.
3.2 Other methods.
Except as otherwise noted in Figure 1, cultures were grown at 23 °C and G1-
synchronization was performed as described previously using 5ng/mL α-factor for bar1∆ strains,
and hydroxyurea (Sigma) at 200mM (Haye-Bertolozzi and Aparicio 2018). DNA content analysis
46
by flow cytometry (FACScan) has been described previously (Aparicio et al. 2004). Preparation
of protein extracts, PAGE, and immunoblotting were performed essentially as described
previously (Ostrow et al. 2017), with the following specifics: 10% gel, semi-dry transfer, blot was
incubated with polyclonal anti-Dbf4 serum (from B. Stillman) (1:1000) overnight at 4 °C, and
detected with anti-rabbit secondary (1:5000, Sigma GENA934). Quantitative BrdU-IP-seq (QBU)
analysis was performed as described (Haye-Bertolozzi and Aparicio 2018). ChIP-seq was
performed as described (Ostrow et al. 2015), using anti-FLAG M2 (Sigma F1804) at 1:200. QBU
and ChIP-seq libraries were constructed using KAPA Hyper Prep Kit (KK8504). High-throughput
DNA sequencing was performed by the USC Genome Core or Novogene. Sequencing data is
available at GEO (GSE215190).
3.3 Computation and statistics.
All sequencing data were binned (50bp) and median-smoothed over a 1kb window. The
list of origins is from OriDB “confirmed” set (n=410) (Nieduszynski et al. 2007). The FKH-
activated list is from (Knott et al. 2012), and the CEN-proximal origin list is from (He et al. 2022).
Matlab was used for generation of most data displays and analyses.
Results and Discussion
3.4 Dbf4∆C is defective in essential Dbf4 function(s) beyond origin-targeting by Fkh1 and Ctf19
To investigate the function in DNA replication of Dbf4′s C-terminal Zn finger domain,
with a focus on its targeting of origins, we examined previously described DBF4 alleles either
lacking the C-terminal 50-amino acids (dbf4∆C), or containing point mutations precisely targeting
the Zn finger motif (dbf4-AAHH (Zn*1) and dbf4-CCHC (Zn*2)), which we will collectively refer
to as dbf4-Zn* (Fang et al. 2017; Jones et al. 2010) (Figure 3.1A). During construction of the
47
Figure 3.1 Dbf4∆C is defective in essential Dbf4 function(s). (A) Schematic representation of Dbf4 protein domains with sequence
detail of the Zn-finger domain; amino acid changes in the Zn* alleles are indicated in lower case, red font. (B) Strains CVy63
(WT), HYy176 (dbf4ΔC), MPy74 (dbf4-Zn*1), MPy76 (dbf4-Zn*2), MPy86 (dbf4-Zn*1 fkh1∆), MPy90 (dbf4-Zn*2 fkh1∆), HYy210
(cft19∆), HYy211 (cft19∆ dbf4∆C), HYy207 (cft19∆ fkh1Δ fkh2-dsm), and HYy218 (fkh1Δ fkh2-dsm) were grown to mid-log phase,
diluted, and plated onto rich media, then incubated at the indicated temperatures and imaged after 2-3 days.
48
dbf4∆C strain, we observed that it was temperature sensitive at 30 °C (Figure 3.1B), which had
been previously noted for a 45 amino acid deletion at 37 °C (Jones et al. 2010). In contrast, the
dbf4-Zn* strains are not temperature sensitive at 30 °C, though they exhibit some sensitivity to
growth at 37 °C (Figure 3.1B, 1st panel). The overall poor growth plus temperature sensitivity of
dbf4∆C cells suggests a substantial defect in Dbf4∆C’s essential function to stimulate individual
replication origins. By comparison, elimination of the known Dbf4 origin-targeting pathways
individually or in combination in fkh1∆ fkh2-dsm (equivalent to fkh1∆ fkh2∆ for origin regulation
but lacking additional pleiotropic defects (Ostrow et al. 2017)), ctf19∆C, and fkh1∆ fkh2-dsm
ctf19∆C cells, respectively, does not phenocopy the temperature sensitivity (Figure 3.1B, 2nd
panel). Moreover, we have recently discovered that dbf4∆C fkh1∆ cells are inviable at the
permissive temperature for dbf4∆C (He et al. 2022), reinforcing the idea that Dbf4∆C is defective
in function(s) in addition to its recruitment via Fkh1, while also suggesting that recruitment to
origins by Fkh1 ameliorates the defect(s) of Dbf4∆C. In contrast, deletion of CTF19 is viable in
combination with dbf4∆C and does not enhance the dbf4∆C phenotype at higher temperatures,
suggesting Ctf19 is not supporting Dbf4∆C function (Figure 3.1B, 3rd panel). Additionally, in
contrast to the lethality of FKH1 deletion in dbf4∆C cells, FKH1 deletion is viable in dbf4-Zn*
cells, clearly indicating that Dbf4∆C lacks function(s) retained in the specific dbf4-Zn* alleles
(Figure 3.1B, 4th panel). It is also notable that deletion of FKH1 enhances the temperature
sensitivity of dbf4-Zn* alleles at 37 °C, suggesting that FKH1 function remains important in the
dbf4-Zn* mutant cells.
3.5 dbf4∆C is defective in overall rate of genome replication while dbf4-Zn* and ctf19∆ are not
To gain more insight into the anticipated replication defects, we examined bulk genome
replication by flow cytometry in the dbf4 mutant strains. Cells were synchronized in G1 phase and
49
released into S phase. In WT cells, DNA content mostly doubled between 30 and 60 min (Figure
3.2, left panel), though it is not possible to precisely define beginning and endpoints of replication
by this type of analysis. The dbf4-Zn* mutants showed similar timing of progression through and
completion of S phase (Figure 3.2, left panel). However, dbf4∆C cells exhibited substantial delay
in progressing through and completing S phase (Figure 3.2, left panel), consistent with a more
severe defect than the Zn* mutants. Cells lacking Ctf19 (ctf19∆) exhibited no defect in S phase
progression, whereas cells lacking both FKH1 and FKH2 origin stimulation function (fkh1∆ fkh2-
Figure 3.2 Dbf4∆C is defective in bulk genome replication, while Dbf4-Zn* mutants are not. Strains described in
Figure legend 1 plus HYy217 (fkh1∆) and HYy215 (dbf4∆C—instead of HYy176) were synchronized in G1 phase and
released into S phase for DNA content analysis. Note the different time intervals of the different panels, and the unique
time intervals for dbf4∆C (inset). Vertical red lines indicate 1C (un-replicated) and 2C (replicated) chromosomal
DNA contents, respectively.
50
dsm) exhibited a modest delay in S phase completion (Figure 3.2, center panel), perhaps of slightly
greater magnitude than previously reported (Knott et al. 2012). This is not surprising given the
large numbers of origins dependent on Fkh1/2 for full activity (Knott et al. 2012). Additional
elimination of the Ctf19-dependent targeting pathway (ctf19∆ fkh1∆ fkh2-dsm), however, did not
appear to enhance the replication delay of fkh1∆ fkh2-dsm cells (Figure 3.2, center panel). Cells
lacking Fkh1 (fkh1∆) exhibited no defect in S phase progression, whereas deletion of FKH1 in
dbf4-Zn* mutants caused a replication delay (Figure 3.2, right panel), indicating that Fkh1
contributes to function of Dbf4-Zn* as inferred earlier from the cell growth analysis. The
replication delay of fkh1∆ dbf4-Zn* is slightly greater than that of cells lacking both targeting
pathways (ctf19∆ fkh1∆ fkh2-dsm), suggesting that Dbf4-Zn* has defects beyond the Ctf19 and
Fkh1/2 targeting pathways. The data also indicate that while elimination of the Ctf19 targeting
pathway has little if any effect on the rate of bulk genome replication, the Fkh1/2 pathway is
needed for normal replication rate and is critical to sustain replication in dbf4-Zn* cells.
Given the sensitivity of the Dbf4 C-terminus to epitope tagging, we wondered whether the
deletion of the C-terminal domain resulted in destabilization of Dbf4. To examine protein levels
of the different mutants, we performed immunoblots of crude extracts from G1-synchronized cells
using a native anti-Dbf4 antibody (Figure 3.3). Dbf4 and Dbf4-Zn* were readily identified near
their predicted size (78kDa), and the band shifts upward in a strain expressing N-terminally
epitope-tagged DBF4 (Figure 3.3). Dbf4 and Dbf4-Zn* were present in similar abundances;
however, Dbf4∆C (~73kDa) was not clear above the background, suggesting a reduction in its
steady-state level and/or stability (Figure 3.3). Similar results were reported by Jones et al., though
the proteins were expressed from heterologous promoters and epitope tagged for detection (Jones
et al. 2010). A reduction in quantity of a protein considered to normally be present in limiting
51
quantity would seemingly explain Dbf4∆C’s overall replication defects described thus far by
acting as a general hypomorph, in addition to defects in possible specific functions such as
targeting.
Figure 3.3 Immunoblotting detects normal abundance of Dbf4-Zn* but reduced Dbf4∆C. Immunoblot with anti-Dbf4
antibody and Ponceau stain are shown for replicate cultures of each strain.
3.6 Dbf4-Zn* mutations specifically eliminate early activation of CEN-proximal origins
To delve more deeply into the specific replication defects, we used quantitative BrdU
incorporation (QBU) in G1-synchronized cells released into S phase in the presence of
hydroxyurea (HU) to generate genome-wide, early S-phase origin firing profiles. QBU data from
replicates were examined for correlation, averaged, scale-normalized, and plotted for comparisons
between experimental strains. A representative plot chromosome XV in WT cells yielded the
expected early S-phase replication profile showing robust peaks of QBU signal at early/efficient
origins (e.g., Fkh-activated, a few of which are identified, and CEN-proximal) and smaller QBU
peaks at later/less-efficient origins (Figure 3.4A). The data for dbf4∆C cells showed profound
defects in BrdU incorporation at most origins, including CEN-proximal origins, while a few
relatively active origins dominate the early replication landscape (Figure 3.4A). Two-dimensional
52
scatter plots allow global comparison of the mutant data in comparison to WT (Figure 3.4B). In
the dbf4∆C strain, CEN-proximal origins exhibit severely reduced replication, as previously
reported, while most of the remaining active origins are Fkh-activated origins (Figure 3.4A,B).
This result is inconsistent with the C-terminus of Dbf4 being required for Fkh1-mediated
recruitment.
The dbf4-Zn* mutant strains exhibited similar chromosome XV replication profiles to each
other but quite distinct from WT and dbf4∆C (Figure 3.4A). The dbf4-Zn* replication profiles
generally show robust activation of a subset of Fkh-activated origins, along with notably weak
activation of CEN-proximal origins, and reduced activation of many other origins (Figure 3.4A,B).
These results are consistent with the C-terminal Zn-finger motif having a crucial function in
stimulation of CEN-proximal origins. The results further suggest that the C-terminal Zn-finger
domain is not required for stimulation of origins via Fkh1; indeed, a subset of Fkh-activated origins
appeared to dominate the replication landscape.
To examine the distribution of origin firing amongst major origin categories, we generated
distribution boxplots for 410 previously confirmed replication origins sequences, 95 Fkh-activated
origins, and 32 CEN-proximal origins, defined as the closest origin on each side of each
centromere (Figure 3.4C). These distribution plots reinforce the conclusion that dbf4∆C has a
profound effect on firing of all origins, including CEN-proximal, but a smaller reduction in Fkh-
activated origins (Figure 3.4C). The dbf4-Zn* mutants show similar but less severe reductions as
dbf4∆C; however, CEN-proximal origins remain more comparatively reduced than Fkh-activated
origins (Figure 3.4C, Table 3.1).
53
Figure 3.4 Dbf4-Zn* is defective in CEN-proximal origin firing. Strains described in Figures 3.1 and 3.2 legends were synchronized
in G1 phase, released into S phase with HU for 60 min (90 min for dbf4∆C), and harvested for QBU analysis. (A) QBU values
averaged for two replicates and scale-normalized across strains are shown for representative chromosome XV; origins and origin
sub-groups are indicated with color-coded circles below the x-axis. (B) Two-dimensional scatter plots comparing QBU signals
averaged across 500bp regions centered on 410 confirmed origins; origins and sub-groups are color-coded as indicated. (C)
Boxplot distributions of averaged QBU counts across 500bp regions aligned on origins of the indicated groups; the number of
origins in each group is indicated within parentheses; results of statistical analyses are given in Table 3.3. (D) Two-dimensional
scatter plot comparing the differences in QBU values for ctf19∆ (WT-ctf19∆) versus dbf4-Zn* (WT-dbf4-Zn*) for Fkh-activated
and CEN-proximal origins.
54
Table 3.1 Results of statistical tests applied to origin groups. Two-tailed t-tests were executed on all strain
combinations.
55
A typical feature sometimes distinguishing replication profiles is the peak width, which is
inversely correlated with the overall level of origins firing, reflecting greater fork progression
enabled by reduced numbers of total forks initiated (Zhong et al. 2013). In the dbf4-Zn* mutants,
the dominant peaks were substantially wider than in WT cells, while much less QBU signal
originated from other loci (Figure 3.4A,B). This result suggests that the relative rate of origin firing
has tilted further in favor of a smaller number of origins, most of which are Fkh-activated origins.
Given that both dbf4∆C and dbf4-Zn* mutant cells show preferential firing of Fkh-activated
origins, and that deletion of FKH1 is lethal in combination with dbf4∆C, we conclude that the C-
terminal Zn finger of Dbf4 is required for its recruitment to CEN-proximal origins but much less
so for its recruitment through Fkh1/2. Still, there appears to be a defect in Dbf4-Zn* recruitment
to other origins lacking either the Cft19 or Fkh1/2 mechanisms. In support of this, the dbf4-Zn*
mutants show a more severe genome-wide replication defect than ctf19∆ cells, which are like WT
cells in QBU profiles except for strong reduction in CEN-proximal origin firing (Figure 3.4A–C).
Direct comparison of the changes in firing levels relative to WT (∆QBU) in ctf19∆ versus dbf4-
Zn* strains illustrates their similar, stronger effects on CEN-proximal versus Fkh-activated origins
(Figure 3.4D). As noted above, the dbf4-Zn* mutants also show a degree of temperature sensitivity
not shown by ctf19∆ fkh1∆ fkh2-dsm cells, which lack both recruitment pathways and
consequently show a highly disrupted pattern of origin initiation (Figure 3.4A–C). Still, these
targeting-defective cells exhibit relatively robust origin firing densities compared with the dbf4-
Zn* mutant strains (Figure 3.4A–C). Given the sensitivity of the C-terminal domain of Dbf4 to
structural perturbation, we think the likeliest explanation is that the Zn-finger point mutations
partly disrupt an aspect of Dbf4 structure/function leading to a mild, generalized replication defect
56
(perhaps interaction with Mcm2) in addition to the fully penetrant defect in CEN-proximal origin
targeting, presumably via the Ctf19 pathway.
3.7 CEN-proximal origins are differentially sensitive to loss of Ctf19 or Dbf4-Zn*
An interesting feature we noted differentiating chromosome replication profiles was the
extent to which CTF19 deletion, as well as the dbf4-Zn* alleles, affected one versus both CEN-
flanking origins. For example, from the scatter plots, it is apparent that a few CEN-proximal
origins are minimally affected by CTF19 deletion or dbf4-Zn* (Fig. 3.4B). A plot of chromosome
X shows an example where one of the two CEN-flanking origin remains robustly active in ctf19∆
and dbf4-Zn* cells (Fig. 3.5A). This origin, ARS1014, remains a rather robust origin, even in
dbf4∆C cells, as well as several other mutant combinations (Fig. 3.5A). However, this origin is
strongly diminished by elimination of both Fkh1 and Fkh2 (Fig. 3.5A). We also observe, as
previously, that deletion of FKH1 often elevates the activity of other origins, particularly CEN-
proximal origins, likely reflecting competition for rate-limiting replication factors (Knott et al.
2012). These observations suggest that some origins retain other features that influence their level
of function, at least in the absence of the recognized, dominant Fkh1/2 and Ctf19 mechanisms.
We also noticed that some origins greater than 25kbp distal to a CEN showed diminished activity
in ctf19∆ cells (Fig. 3.5B), indicating a somewhat more extensive domain over which Ctf19 can
stimulate origin firing than previously reported (Natsume et al. 2013).
57
Figure 3.5 CEN-proximal origins are differentially sensitive to CTF19 deletion or dbf4-Zn*. (A) QBU values for experiments
described in Figures 3 and 4 legends are plotted for chromosome X. (B) The difference in QBU values between WT and ctf19∆
(WT-ctf19∆) is plotted as a function of distance of each origin (n = 410) to its centromere. The solid line indicates line of best fit
for data beyond 25kb; broken lines indicated the 95% confidence interval.
58
Figure 3.6 Dbf4-Zn* retains Fkh1 origin stimulation. Strains described in Figures 3.1 and 3.2 legends were synchronized in G1
phase, released into S phase with HU for 60 min, and harvested for QBU analysis. (A) QBU values averaged for two replicates
and scale-normalized across strains are shown for representative chromosome XV; origins and origin sub-groups are indicated
with color-coded circles below the x-axis. (B) Two-dimensional scatter plots comparing QBU signals averaged across 500bp
regions centered on 410 confirmed origins; origins and sub-groups are color-coded as indicated. (C) Boxplot distributions of
averaged QBU counts across 500bp regions aligned on origins of the indicated groups; the number of origins in each group is
indicated within parentheses; results of statistical analyses are given in Table 3.1.
59
3.8 Dbf4-Zn* retains Fkh1-dependent targeting to Fkh1-activated origins
To directly test the hypothesis that Dbf4-Zn* continues to be directly recruited to origins
by Fkh1/2, we performed QBU in dbf4-Zn* mutants lacking FKH1. Deletion of FKH1 in WT cells
resulted in significant reduction in QBU signal of Fkh1-activated origins (n=35, origins sensitive
to the loss of only FKH1 in WT cells (Knott et al. 2012)), such as ARS1509, ARS1513.5 and
ARS1529.5, as well as other Fkh-activated origins (Fig. 3.6A-C, Table 3.1). Deletion of FKH1 in
dbf4-Zn* cells also resulted in a significant decrease in QBU signal for Fkh1-activated origins
(Fig. 3.6A-C, Table 3.1), supporting the conclusion that dbf4-Zn* mutants continue to be recruited
to origins via Fkh1. These results further support the conclusion that the Zn-finger motif of Dbf4
is specifically involved with targeting to CEN-proximal origins but depends on a different motif
or domain for recruitment to origins via Fkh1.
3.9 Dbf4-Zn* is Defective in Its Recruitment to CENs
To directly test whether the recruitment of Dbf4 to replication origins is affected by Dbf4-
Zn*, we performed ChIP-seq of Dbf4 and Dbf4-Zn*1, each bearing a triple-FLAG epitope at the
N-terminus. Previous ChIP analysis of N-terminally tagged Dbf4 detected enrichment at several
replication origins, and strong enrichment at CENs (Natsume et al. 2013). Our analysis of Dbf4
yielded similar results with robust enrichment at CENs and less, but detectable enrichment at CEN-
proximal origins (Figure 3.7). However, we detected little to no enrichment at most other
replication origins (Figure 3.7). Dbf4-Zn*1 showed significantly reduced enrichment at CENs, as
well as reduced enrichment at CEN-proximal origins although the difference at the origins was not
significant (Figure 3.7). Minor enrichment of Dbf4-Zn*1 was detected at other replication origins;
however, the levels were not significantly different than for Dbf4 (Figure 3.7). These results
strongly support the function of the Zn-finger motif in Dbf4′s targeting to CENs and consequent
60
Figure 3.7 Dbf4-Zn* is defective in its recruitment to CENs. Strains EAy1 (3xFLAG-DBF4) and EAy2 (3xFLAG-
dbf4-Zn*1) were synchronized in G1 phase and subjected to ChIP-seq analysis. (A) Heatmaps of averaged ChIP-seq
values across 10kbp regions centered on origins (n = 408 because two rDNA origins removed), origin sub-groups, or
CENs. (B) Boxplot distributions of ChIP-seq values for 500bp windows centered on the indicated features, subjected
to two-sided t-tests. (C) Two-dimensional scatter plots of ChIP-seq values for 500bp windows centered on the
indicated features, showing lines of best fit for origin sub-groups and CENs.
stimulation of CEN-proximal origins. However, these data do not allow us to conclude whether
the Zn-finger motif plays a role in recruitment to other replication origins such as Fkh-activated
61
origins. Nevertheless, the QBU data indicate that the Zn-finger motif is dispensable for most Fkh-
dependent origin targeting.
Perspective
In retrospect, it may not be surprising that the C-terminal domain of Dbf4 is not the critical
feature for physical interaction with Fkh1. It is anticipated that Fkh1 interacts with a phospho-
regulated partner through its FHA domain, which is expected to function as a phosphothreonine
binding module. However, no threonine resides amongst the C-terminal 50 amino acids of Dbf4,
undermining the notion that this domain mediates the critical interaction with Fkh1-FHA. Further
work will be required to define the exact nature of the interaction between Fkh1 and Dbf4, and
with licensed origins.
62
Chapter 4: FHA Domain of Fkh1 Is Required in Co-Recruitment of Fkh1
and Dbf4 for Replication Origin Initiation
Adapted from:
Petrie MV, He Y, Villwock SK, Ghoshal K, Zhang H, Gan Y, and Aparicio OM. (In prep)
This project was started by Sandra, before I joined the lab and she constructed and carried
out Figure 4.1. Haiyang constructed the strains and Yiwei carried out the experiment shown in
Figure 4.6. The experiments done in Figure 4.8 were carried out by Yan. My primary contribution
to the following work includes strain creation, QBU, ChIP-seq, analysis and quantification of the
data, additional writing and development of an early draft of the main text (Figures 4.2-4.5, 4.6,
and S1-8).
Abstract
Budding yeast Fkh1 and Fkh2 proteins control gene expression, mating-type switching,
and DNA replication through sometimes distinct, sometimes complementary functions. In
regulation of replication origins, Fkh1 plays the primary role while Fkh2 plays a secondary and/or
backup role. Fkh1 and Fkh2 each contain a ForkHead-Associated (FHA) domain, which typically
mediates interaction with a phosphorylated target protein to execute a function. Here, we
examined whether the Fkh1 FHA is required for its function in origin regulation. Analysis of
replication by BrdU immunoprecipitation analysis demonstrated a defect of origin regulation in
cells with a mutation in Fkh1-FHA (R80A) predicted to eliminate phospho-threonine binding and
that fully eliminates donor preference. Fkh1 lacking the entire FHA domain was fully defective
63
in origin regulation, suggesting phosphorylation-dependent and -independent interactions may
contribute to origin regulation by Fkh1. Current data suggest that Dbf4 is a likely target of Fkh1-
FHA. In support of this conclusion, we show that fusion of Dbf4 with the Fkh1 DNA-Binding
Domain (DBD) bypasses the requirement for the FHA domain. This result also implies that Dbf4
recruitment is the exclusive function of the FHA domain in origin firing.
Introduction
Forkhead Box (Fox) proteins comprise a large eukaryotic protein family that shares the
Forkhead “winged helix” DNA binding domain (DBD) (reviewed in Lalmansingh, 2012). Fox
proteins have been widely implicated as transcription factors that regulate cell growth and division,
cellular differentiation and development, and organismal homeostasis and stress responses, with
consequences for cancer and aging (reviewed in Murakami, 2010). In S. cerevisiae, four proteins
contain identifiable Forkhead (FKH) domains: Fkh1, Fkh2, Hcm1, and Fhl1, the last being highly
diverged and functionally unrelated (Murakami, 2010). Fkh1 and Fkh2 are most similar, each
having an N-terminal ForkHead-Associated (FHA) domain, which is apparently absent in Hcm1.
FHA domains usually function as phospho-threonine binding modules, thereby mediating
phosphorylation-dependent interactions (reviewed in Durocher, 2002).
Fkh1, Fkh2, and Hcm1 have been implicated in transcriptional regulation of dozens of cell
cycle genes, with Hcm1 being responsible for the periodic expression of chromosome segregation
genes during S phase, and Fkh1 and Fkh2 being responsible for the periodic expression of the
“CLB2 cluster” of cell division genes during G2-M, as well as a set of G1 phase genes
(Hollenhorst, 2000; Koranda, 2000; Kumar, 2000; Pic, 2000; Pramila, 2006; Voth, 2007; Zhu,
2000). Analysis of CLB2 cluster gene expression in fkh1∆ versus fkh2∆ cells indicates that Fkh1
64
is required for appropriate temporal repression and Fkh2 for appropriate temporal activation of
these genes (Hollenhorst, 2000; Kumar, 2000; Pic, 2000). Consistent with Fkh1 and Fkh2 having
at least a subset of distinct functions, Fkh1 and Fkh2 contribute additively to CLB2 cluster gene
regulation as deletion of both genes causes more severe deregulation than single deletions,
resulting in pseudohyphal and invasive growth phenotypes (Hollenhorst, 2000; Koranda, 2000;
Kumar, 2000; Pic, 2000; Zhu, 2000).
The precise functions of Fkh1 and Fkh2 in transcriptional regulation have mainly focused
on Fkh2, which unlike Fkh1, binds to promoters cooperatively with Mcm1 to regulate target genes
(Boros, 2003; Hollenhorst, 2001; Koranda, 2000; Kumar, 2000; Pic, 2000; Zhu, 2000). Fkh2 also
has an extended C-terminus containing several CDK-dependent phosphorylation sites that
regulates interaction with the transcriptional co-activator Ndd1 (Pic-Taylor, 2004; Reynolds,
2003). The Fkh2 FHA domain is also involved in mediating phospho-threonine-specific
interaction with CDK-phosphorylated Ndd1 (Darieva, 2003; Pic-Taylor, 2004; Reynolds, 2003).
Whereas Fkh1 FHA is likely able to recruit Ndd1 for co-activation (particularly in the absence of
FKH2), only deletion of FKH2 but not FKH1 suppresses the lethality of NDD1 deletion,
suggesting that Fkh2 strictly depends on Ndd1 for CLB2 cluster gene activation and strongly
represses in its absence (Koranda, 2000). In fkh2∆ ndd1∆ cells, Fkh1 was hypothesized to activate
CLB2 cluster genes through FHA-mediated recruitment of an unknown co-activator (Reynolds,
2003).
Yet another function in the Fkh1 and Fkh2 repertoire is replication origin timing regulation
(reviewed in Aparicio, 2013). Fkh1 and Fkh2 are required for the early activation of most early-
firing origins throughout the genome (Knott, 2012). Here again, genetic analyses have shown that
Fkh1 and Fkh2 are not functionally equivalent in replication origin regulation. Deletion of FKH2
65
has no effect, while deletion of FKH1 causes delayed activation of many replication origins.
Double deletion of FKH1 and FKH2 causes more profound deregulation in both the number of
affected origins and degree of delayed activation. These results suggest that Fkh1 plays the
primary role at origins with Fkh2 acting at a subset, at least in the absence of Fkh1. This is
consistent with genome-wide chromatin immunoprecipitation (ChIP) analysis of Fkh1 and Fkh2
chromatin binding that showed association of either or both proteins with most Fkh1/2-activated
replication origins, however, Fkh1 was more frequently detected than Fkh2 (Ostrow, 2014).
Origin stimulation by Fkh1 (and Fkh2) appears to involve direct recruitment of Dbf4,
which is a rate-limiting factor for origin initiation (Petrie et al., 2022) Nevertheless, the nature of
the interaction between Dbf4 and Fkh1 requires further investigation (Fang et al., 2018). It would
be anticipated that the FHA domain of Fkh1 acts in replication origin control, however this has not
been directly tested. Here, we show that the phosphothreonine-binding motif of the FHA domain
is required for full origin stimulation. Unexpectedly, however, we find that this motif regulates
Fkh1 binding to origins, which is sufficient to explain its defect in origin stimulation. Fkh1 and
Dbf4 exhibit interdependence in origin binding suggesting that Fkh1 and Dbf4 (and DDK partner
Cdc7) form a complex mediated by FHA to bind origins. We show that fusing the FKH domain
to the C-terminus of Dbf4 fully substitutes for wild-type Fkh1, strongly suggesting that recruitment
of Dbf4 to origins is the only essential function of the FHA domain in origin control. Finally, we
show that there is a physical interaction between Dbf4 and Fkh1 with dependency on a functional
Fkh1-FHA domain.
66
Materials and Methods
4.1 Plasmid constructions
Restriction endonucleases were purchased from New England Biolabs and used according
to their protocols. pRS406-Fkh1 was constructed by subcloning an XhoI-NotI fragment containing
FKH1 with upstream and downstream sequences from pCF480 (Hollenhorst et al., 2000) into
pRS406 digested with the same enzymes (Sikorski et al., 1989). pRS406-Fkh1 was subjected to
site-directed mutagenesis according to manufacturer’s protocol (Agilent Quickchange Multi) with
primer Fkh1-R80A to create pRS406-Fkh1-R80A. pBlue-Fkh1-fha∆ was constructed by Gibson
assembly (SGI #GA1200) with primers Fkh1-up-F, Fkh1-int55-R, Fkh1-int202-F, and Fkh1-
down-R to generate overlapping 5′ and 3′ FKH1 gene fragments that fuse together to delete amino
acids 56-201 and overlap with SacI- and KpnI-digested pBluescript. pET28a-FKH1 was
constructed by PCR amplification of FKH1 using primers Fkh1-EcoRI-HIS-F and Fkh1-HindIII-
HIS-R and template p405-FKH1 (Ostrow et al., 2017) the PCR product was digested with HindIII
and EcoRI and inserted into HindIII+EcoRI-digested pET28a. pET28a-FKH1-fha∆ was
constructed by Gibson assembly with primers FKH1-fha∆-HIS-F and FKH1-fha∆-HIS-R to PCR
amplify FKH1-fha∆ from pBlue-Fkh1-fha∆ and insert into NdeI-digested pET28a. pCD43-DBF4-
FKH484 was constructed by Gibson assembly with primer sets GAL1p-EcoRI-DBF4-F and
DBF4-704-fuse-FKH-231-R, and DBF4-704-fuse-FKH-231-F and GAL1p-SalI-FKH-484-R
using genomic DNA as template to generate full-length DBF4 and FKH1 FKH domain (amino
acids 231-484) fragments inserted into EcoRI+SalI-digested pCD43 (Peace et al., 2016). A
fragment containing DBF4 sequence fused with FKH(231-484) was PCR amplified with primers
∆N-DBF4-FKH484-F and p306-R from pCD43-DBF4-FKH484, digested with NotI and KpnI, and
ligated into NotI+KpnI-digested pRS306 to yield p306-∆Ndbf4-FKH484. FKH1-MYC9 was
67
PCR-amplified from p403-Fkh1-Myc9 (from Ostrow et al., 2017 REF) using primers Fkh1-MYC-
F and Fkh1-MYC-R and subcloned into EcoRI+MscI-digested p306-ΔNdbf4-FKH484 to yield
p306-ΔNdbf4-FKH484-MYC9. Sequences were confirmed by DNA sequencing (Retrogen Inc.).
4.2 Yeast strain constructions
Plasmids and integrating constructs were introduced into yeast by lithium acetate
transformation with appropriate selection (Gietz, 2007). Gene deletions were constructed by PCR
amplification of selectable markers with hybrid oligonucleotides having homology flanking the
gene of interest (Longtine, 1998). Genomic alterations were confirmed by PCR analysis and/or
DNA sequence analysis as appropriate.
All strains are congenic with W303 background, and most are derived from BrdU-
incorporating strains CVy63 and CVy70, which are derived from SSy161 and SSy162,
respectively (Viggiani and Aparicio 2006). JPy32 was constructed by deletion of FKH2 in CVy68
followed by replacement of FKH1 with fkh1-R80A through pop-in/pop-out using p306-Fkh1-
R80A. JPy44 is a MATa derivative of JPy32 created by HO-induced mating-type switch. YHy1
was constructed by transformation of strain JPy44 using pTOPO-Fkh1-MYC. OAy1106 was
constructed by deletion of FKH1 in CVy68. OAy1112 was constructed by replacing fkh1∆::URA3
in OAy1106 with fkh1-fha∆ by transformation with SacI- and KpnI-digested pBlue-Fkh1-fha∆ and
selection on 5-FOA. SVy62 and SVy64 are haploid segregants from a cross of strains SVy55 and
JPy32. SVy65 is a haploid segregant from a cross of strains SVy55 and OAy1112. SVy66 is a
haploid segregant from a cross of strains SVy55 and OAy1106. OAy1146 and OAy1147 were
constructed by transformation of strains CVy68 and OAy1112 with PCR product of primers 2xL-
3xFLAG-FKH1-F and 2xL-3xFLAG-FKH1-R using plasmid p2xL-3xFLAG (TRP1) as template.
68
MPy210 is a haploid segregant from a cross of strains JPy32 and OAy1106. MPy181 and MPy187
are haploid segregants from a cross of strains MPy102 and OAy1106. MPy181 was transformed
with a PCR of fkh1-R80A to give MPy195. MPy187 was crossed with OAy503 resulting in the
haploid segregant MPy204. Both OAy1184 and OAy1185 were constructed by transformation of
strain MPy105 with PCR products of primers Fkh1-up400bp and Fkh1-down275bp using ZOy48
and YHy1 respectively. MPy126 is a haploid segregant from a cross of strains MPy125 and
OAy1106. HYy236 was constructed by transformation of HYy196 with EcoNI-digested p306-
∆Ndbf4-FKH484-MYC9. Expression of Dbf4-FKH484-MYC9 was confirmed by
immunoblotting. HYy203 is a haploid segregant from a cross of HYy180 and HYy236. MPy108
is a haploid segregant from a cross of strains MPy100 and ZOy3. OAy1189 was derived from
JOSHy1 and was crossed with MPy108 resulting in the haploid segregant was MPy189.
4.3 Other methods
Cultures were grown in YEPD logarithmically at 23 °C and G1-synchronization was
performed as described previously using 5ng/mL α-factor for bar1∆ strains, and hydroxyurea
(Sigma) at 200mM (Haye-Bertolozzi and Aparicio 2018). DNA content analysis by flow
cytometry (FACScan) has been described previously (Aparicio et al. 2004). Preparation of protein
extracts, PAGE, and immunoblotting were performed essentially as described previously (Ostrow
et al. 2017), with the following specifics: 10% gel, semi-dry transfer, blot was incubated with anti
MYC 9E10 (Invitrogen 13-2500) (1:2000) or anti-FLAG M2 (Sigma F1804) (1:1000) as noted
and incubated overnight at 4 °C, and detected with anti-mouse (Sigma GENA931) (1:5000).
EMSA was performed as follows: On ice, 10ng of DNA probe (see Table 4.1 for sequences
of oligonucleotide pairs annealed together to create double-stranded probes) was combined with
69
protein (amount indicated in figure legend) in 20μL total volume in 20mM Tris-HCl pH 7.9, 50mM
KCl, 5mM MgCl2, 3mM DTT, 0.1mg/ml BSA, 10% (v/v) Glycerol. Binding reactions were
incubated on ice for 15 min followed by incubation at room temperature (22°C) for 15 min.
Samples were loaded onto a 10% (w/v) (21:1, acrylamide:bis-acrylamide) polyacrylamide gel and
separated in 0.5x TBE buffer at 120V for 90 min in a 4°C environmental chamber. The gel was
stained in 0.5xTBE, 0.2nM SYBR Green I (Molecular Probes) 10 min at 22°C with gentle mixing.
The gel was de-stained by incubating in H2O 10 min at 22°C with gentle mixing, and repeating.
The gel image was captured on a BioRad FX scanner.
Quantitative BrdU-IP-seq (QBU) analysis was performed as described (Haye-Bertolozzi
and Aparicio 2018). ChIP-seq was performed in triplicate as described (Ostrow et al. 2015), using
equal combination of anti-HA 12CA5 (Invitrogen MA1-12429) and 16B12 (Invitrogen A-21287)
at (1:200), or anti-MYC 9E10 (Invitrogen 13-2500) at 1:200, or anti-FLAG M2 (Sigma F1804) at
1:200 as noted. QBU and ChIP-seq libraries were constructed using KAPA Hyper Prep Kit
(KK8504). High-throughput DNA sequencing was performed by the USC Genome Core or
Novogene. Sequencing data is available upon request.
4.4 Computation and statistics
All sequencing data were binned (50bp) and median-smoothed over a 1kb window. The
list of origins is from OriDB “confirmed” set (n=410) (Nieduszynski et al. 2007). The FKH-
activated and -repressed lists are from (Knott et al. 2012), and the CEN-proximal origin list is from
(He et al. 2022). Matlab was used for generation of most data displays and analyses.
70
Results
4.5 Fkh1-FHA is required for replication origin stimulation
To test the function of the Fkh1-FHA domain in regulation of replication origin timing, we
evaluated FKH1 alleles in fkh2∆ strains, as FKH2 can partially compensate for deletion of FKH1
in origin regulation. To evaluate FHA function, we used the R80A allele to disrupt putative
phospho-threonine binding, and we constructed fkh1-fha∆, a deletion of the FHA domain (amino
acids 56-201) and confirmed that Fkh1-fha∆ binds DNA similarly to Fkh1 by electrophoretic
mobility shift analysis (EMSA) of E. coli-expressed and purified proteins (Figure S1A). Both
alleles were introduced to precisely replace the wild-type (WT) allele; however, our strain
validation approaches revealed that Fkh1-fha∆ was present at significantly lower level than Fkh1
or Fkh1-R80A (Figure S1B,C), so we dropped further use of this allele. Cultures were
synchronized in G1 phase and released into S phase in the presence of 5-bromo-2′-deoxyuridine
(BrdU) to label replicating DNA and hydroxyurea (HU) to arrest replication in early S phase and
reveal early origin firing profiles; bulk DNA content analysis shows that the fkh1 mutant strains
entered and completed S phase in the absence of HU with similar timing as wild-type cells (Figure
S1D). We analyzed BrdU incorporation using a quantitative BrdU immunoprecipitation and
analysis protocol we developed called qBrdU-seq (or QBU) (Peace et al., 2016). Two-dimensional
scatter plots of experimental replicates demonstrate the high reproducibility of the data (Figure
S1E). QBU profiles of representative chromosomes VI and IX show the expected wild-type
replication profiles in cells with FKH1 (fkh2∆), and the expected deregulated replication profiles
of fkh1∆ (fkh2∆) cells, characterized by diminished activation of Fkh-activated origins and
increased activation of Fkh-repressed origins (Fig. 1A) (Knott et al., 2012; Ostrow et al., 2017).The
fkh1-R80A (fkh2∆) cells also exhibited origin deregulation, but to a lesser extent than the deletion,
71
Figure 4.1 Fkh1-FHA regulates replication origin timing. Strains SVy62 (fkh2Δ), SVy66 (fkh1Δ fkh2Δ), SVy64 (fkh1-R80A fkh2Δ),
and SVy65 (fkh1-fha∆ fkh2Δ) were analyzed by QBU after G1 phase block-and-release into S phase for 60 min in the presence of
HU. (A) Average BrdU incorporation profiles are shown for representative chromosomes VI and IX; Fkh-activated, -repressed,
and other origins are indicated with blue, red, and gray circles, respectively, below the x-axis. (B) Heat maps of QBU counts for
10kb regions are aligned on origins of the indicated classes. (C) Boxplot distributions of averaged QBU counts across 500bp
regions aligned on origins of the indicated groups; the number of origins in each group is indicated within parentheses. Two-sided
t tests were performed on all pairs of strains, and results are indicated as *P < 0.05, **P < 0.01, ***P < 0.001, and ****P <
0.0001. (D) Two-dimensional scatter plot comparing the differences in QBU values for each fkh1 mutant versus FKH1 wild-type.
Origins are color-coded as indicated. A linear regression deriving the best-fit line is shown for the sub-group
72
indicating residual function of Fkh1-R80A (Figure 4.1A). Two-way correlation analyses to
quantitatively compare the genome-wide replication profiles objectively support the conclusion
that fkh1-R80A exhibits less than a null phenotype, though still more like null than wild-type
(Figure S1F).
To examine the different effects of the FKH1 alleles on specific origin groups, we
generated heat maps of averaged QBU signal comprising the previously defined Fkh-activated
(n=95) and Fkh-repressed (n=80) origin sets (Knott, 2012) (Figure 4.1B). We also generated
boxplots of the distributions of individual origins in those sets and applied statistical analysis to
each group. These data show that fkh1-R80A affects these origins like fkh1∆, but to an intermediate
degree (Figure 4.1C). We also generated two-dimensional scatter plots to examine the effects on
individual origins in these different origin sets amongst all origins. These data show the profound
deregulation of origin firing across the genome in the absence of Fkh1 (and Fkh2) function. The
more robust effect of fkh1∆ versus fkh1-R80A is supported by the line of best fit (Figure 4.1D).
We also conducted QBU to examine fkh1-R80A in the presence of FKH2. We
hypothesized that fkh1-R80A might exhibit a more severe defect than fkh1∆ in the presence of
FKH2 due to potential substitution by Fkh2 being blocked by resident Fkh1-R80A. Two-
dimensional scatter plots of experimental replicates demonstrate the high reproducibility of the
data (Figure S2A). Analysis of QBU profiles of WT, fkh1∆, and fkh1-R80A cells for representative
chromosomes VI and IX yielded qualitatively similar results as in the absence of FKH2 (Figure
4.2A). For this analysis, we also examined the set of Fkh1-activated origins, which were defined
as significantly impacted in cells with only FKH1 deletion in addition to the Fkh-activated and
Fkh-repressed origin sets (Knott et al., 2012). Though, in examining the effect on different origin
73
groups through heat maps of averaged QBU signal the number of affected origins and degree of
effect was reduced
Figure 4.2 Fkh1-R80A in the presence of Fkh2 regulate replication origin timing. Strains CVy63 (WT) and MPy210 (fkh1-R80A)
were analyzed by QBU after G1 phase block-and-release into S phase for 60 min in the presence of HU. (B) Heat maps of QBU
counts for 10kb regions are aligned on origins of the indicated classes. (C) Boxplot distributions of averaged QBU counts across
500bp regions aligned on origins of the indicated groups; the number of origins in each group is indicated within parentheses.
Two-sided t tests were performed on all pairs of strains, and results are indicated as *P < 0.05 , **P < 0.01, ***P < 0.001, and
****P < 0.0001. (D) Two-dimensional scatter plot comparing the differences in QBU values (500-bp regions) for fkh1-R80A
versus wild-type. Origins are color-coded as indicated. A linear regression deriving the best-fit line is shown for the sub-group
74
as expected (Figure 4.2B). In addition, we generated boxplots of the distributions of individual
origins in those sets and applied statistical analysis to each group. However, fkh1-R80A does not
show evidence of a more severe phenotype amongst Fkh-activated origins than fkh1∆ that might
arise from Fkh1-R80A blocking functional substitution by Fkh2 (Figure 4.2C). We also generated
two-dimensional scatter plots to examine the effects on individual origins in these different origin
sets amongst all origins. These origins are significantly affected by the fkh1-R80A allele, like the
deletion (Figure 4.2D). This difference can still be seen though to a lesser extent when considering
all the Fkh-activated origins (Figure S2B). Together, these results demonstrate that Fkh1-R80A is
defective in origin stimulation, suggesting that phospho-threonine binding by FHA is important
for Fkh1’s function in origin firing.
4.6 Fkh1-FHA is dispensable for Orc1 and Mcm4 occupancies at origins
Whereas the above data are consistent with a role for the FHA domain in recruitment of
downstream effector as anticipated, we considered the possibility that FHA plays some role in
origin and/or chromatin binding by Fkh1, perhaps precluding interpretation of any downstream
function of FHA in effector recruitment. Indeed, we previously addressed this idea by performing
ChIP-chip to examine ORC and MCM occupancies at origins in fkh1∆ fkh2∆ cells, and concluded
that Fkh1/2 do not act in origin regulation through modulation of ORC and MCM origin
associations (Knott et al., 2012). However, a recent study has reported that Fkh1-FHA functions
to stabilize ORC binding, and hence, origin function (Hoggard et al., 2021). To address these
conflicting results, we performed ChIP-seq of Orc1-3xHA in G1-synchronized WT and fkh1-R80A
cells. Two-dimensional scatter plots of sample replicates show excellent reproducibility of the
data (Fig. S3A). Chromosome plots of Orc1 enrichment show robust signal peaks at most known
origins in WT, and fkh1-R80A, cells (Figure 3A). Further comparisons of WT and mutant data by
75
two-dimensional scatter plots show robust correlation of signals across the genome and at
replication origins, including selected subsets of origins such as Fkh1-activated origins and
Figure 4.3 Fkh1-R80A has no effect on ORC1 or MCM4 binding. Strains MPy199 (ORC1-3xHA) and MPy204 (fkh1-R80A, ORC1-
3xHA) were synchronized in G1 phase and strains MPy184 (MCM4-3xHA) and MPy195 (fkh1-R80A, MCM4-3xHA) were subjected
to ChIP-seq. (A) Average ChIP-seq profiles are shown for representative chromosomes IX and XII; origins and select features are
color coded as indicated above plots (B) Two-dimensional scatter plots comparing ORC1 ChIP-seq signals averaged across 500bp
sliding window of the genome and regions centered on 410 confirmed origins, Fkh-activated, and positive chromosome sub-group
are color-coded as indicated. A linear regression deriving the best-fit line is shown for the sub-group. (C) Box plots distributions
of ChIP-seq enrichments for 500bp regions aligned on origins of the indicated sub-groups; the number of origins in each group is
indicated within parentheses. Two-sided t tests were performed on all pairs of strains. * indicates a negative T-value.
positive-Chromatin origins (Fox), which are highlighted (Figure 3B). Distribution boxplots and
statistical analysis of Orc1 enrichment show similar distributions in WT versus fkh1-R80A
76
suggesting that Fkh1 is dispensable for Orc1’s association with origins in G1 phase. In fact, there
was a slight increase in Orc1 binding to origins when the FHA was disrupted. In addition to these
origins we also examined origins defined by Hoggard 2021 which were identified as specifically
influenced by loss of the fkh1-fha. (Figure 3C)
We also tested whether Fkh1 influences MCM occupancies at origins by performing ChIP-
seq of Mcm4-3xHA WT and fkh1-R80A cells (Figure 3A). As above, replicate datasets show high
reproducibility and Mcm4 enrichment is robust at most known origins (Figure S3B). Box plots of
Mcm4 enrichment show similar distributions in WT versus fkh1-R80A suggesting that Fkh1 is
dispensable for Mcm4’s association with origins (Figure 3C). Because Mcm4 association with
origins depends on ORC, these results support the conclusion that Fkh1 does not significantly
control Orc1 and Mcm4 origin occupancies.
4.7 Fkh1-R80A is defective in Fkh1 recruitment to origins
To determine whether the FHA domain has a function in Fkh1 origin-binding, we
performed ChIP-Seq of Fkh1-Myc9 and Fkh1-R80A-Myc9 in unsynchronized and G1-
synchronized cultures. In unsynchronized cells, Fkh1 and Fkh1-R80A binding is observed at
known Fkh1 binding sites including the recombination enhancer (RE) and CLB2 cluster genes, but
less so at replication origins as seen across chromosome plots (Figure 4.1A). In G1-synchronized
cells, Fkh1 binding increases across the genome with the notable exception of the RE, and many
replication origins are bound, especially Fkh-activated origins. In contrast, Fkh1-R80A exhibits
similar or greater enrichment at CLB2 cluster genes, and similar enrichment at RE, but less
enrichment at Fkh-activated origins as show by direct ChIP signal or heatmaps (Figure 4.4A,B) In
fact, in G1 there is significantly less binding at origins and in particular Fkh-activated origins.
77
(Figure 4.4C). These results suggest that the phospho-threonine binding motif of FHA is required
for Fkh1’s interaction with replication origins. We (and others) have previously shown that Fkh1
binds origins in a
Figure 4.4 Fkh1-R80A has reduced binding at Fkh-activated origins. Strains OAy1184 (FKH1-MYC) and OAy1185 (fkh1-R80A-
MYC) were either asynchronous or synchronized in G1 phase and subjected to ChIP-seq. (A) Average ChIP-seq profiles are shown
for representative chromosomes III and IX; origins and select features are color coded as indicated above plots. (B) Heatmaps of
ChIP-seq enrichment at selected origin and Fkh1 binding loci; n of each group is indicated within parentheses. (C) Box plots
distributions of ChIP-seq enrichments for 500bp regions aligned on indicated groups; the number of origins in each group is
indicated within parentheses. Two-sided t tests were performed on all pairs of strains.
sequence-dependent manner at Fkh1 consensus binding sequences associated with origins, with
binding dependent on the FKH domain (Knott et al., 2012; Ostrow et al., 2014; Hoggard et al.,
2021). The dependence on FHA suggests that cooperativity with its target phospho-protein, likely
Dbf4, determines the cell cycle-regulated binding of Fkh1 to origins.
78
4.8 Loss of Fkh1 affects Dbf4 binding
Previous studies have shown Fkh1/2-dependent association of Dbf4 and Dbf4-dependent
replication proteins Cdc45 and Sld3 with Fkh-activated origins in G1 (Knott et al., 2012). We
performed ChIP-seq with N-terminally triple-FLAG-tagged Dbf4 (3xFLAG-Dbf4) expressed from
its endogenous locus in WT and fkh1∆ cells synchronized in G1. Consistent with previous studies,
robust enrichment of Dbf4 was detected at centromeres (CENs) (Fig 4.8A). Heat maps show local,
though comparatively minor enrichment of Dbf4 at origins in WT cells that is reduced in fkh1∆
cells (Fig. 4.8A). Boxplots and statistical analysis show significant reduction of Dbf4 at Fkh-
activated origins in fkh1∆ cells (Fig. 4.8B). This result indicates that FKH1 is mediating this
binding and thus we make the hypothesis that it is the Fkh1-FHA that is required for origin
recruitment of Dbf4.
Figure 4.5 Dbf4 binds preferentially to Fkh-activated origins and the centromeres. Strains MPy123 (DBF4-3xFLAG) and
MPy129 (fkh1∆, DBF4-3xFLAG) were synchronized in G1 phase and subjected to ChIP-seq. (A) Heatmaps of ChIP-seq
enrichment at selected origin and centromeres; n of each group is indicated within parentheses. (B) Box plots distributions of
ChIP-seq enrichments for 500bp regions aligned on indicated groups; the number of origins in each group is indicated within
parentheses. Two-sided t tests were performed on all pairs of strains.
79
4.9 FKH DBD fusion to Dbf4 bypasses requirement for FHA domain
The requirement of Fkh1’s FHA domain for origin stimulation has been proposed to
involve direct interaction with Dbf4 (Fang et al. 2018). The finding that fkh1∆ is defective in Dbf4
recruitment to origins is also consistent with this idea. We performed an experiment previously
reported by Fang et al in which the FKH DNA-binding domain (DBD) of Fkh1 was fused to the
C-terminus of Dbf4 and the ability to complement the loss of FKH1 was tested. We wished to
apply QBU to be able to draw more quantitative conclusions about the extent of complementation,
so we subjected WT, fkh1∆, and fkh1∆ Dbf4-FKH cells to the experimental scheme outlined earlier
to capture early S-phase replication profiles. The results show that Dbf4-FKH fully complements
the origin-firing deficiencies observed in the fkh1∆ strain (Fig. 4.6A). The heatmaps show that the
average signal at Fkh-activated origins is equal or higher in fkh1∆ dbf4-FKH than in WT cells, and
the distribution of signals at individual origins is indistinguishable between the strains. The notable
exception is the CEN-proximal origins, which are significantly decreased in activity, as expected,
due to the C-terminal fusion to DBF4 that disrupts recruitment to CENs via Ctf19 kinetochore
complex interaction (Petrie et al., 2022) (Fig. 4.6B).
We also generated boxplots of the distributions of individual origins in these sets and
applied statistical analysis to each group. These data show that fkh1∆ dbf4-FKH rescues the fkh1∆
phenotype to WT origin firing levels (Figure 4.1C). We also generated two-dimensional scatter
plots to examine the effects on individual origins in these different origin sets amongst all origins.
These data show the return of Fkh-activated origins to WT levels in the presence of the dbf4-FKH
fusion despite the deletion of FKH1. The return to a WT origin firing pattern save for the robust
loss of CEN-proximal origins is supported by the line of best fit (Figure 4.1D). These results
demonstrate that the FHA domain is dispensable for Fkh-mediated origin activation, in so far as
80
Figure 4.6 FKH DNA-Binding Domain fusion to Dbf4 bypasses requirement for FHA. Strains CVy63 (WT), JPy84 (fkh1∆), and
HYy203 (fkh1∆, dbf4-FKH) were analyzed by QBU as described in Figure 1 legend. (A) Average BrdU incorporation profiles are
shown for representative chromosomes VI and IX; Fkh-activated, -repressed, and other origins are indicated with blue, red, and
gray circles, respectively, below the x-axis. (B) Heat maps of QBU counts for 10kb regions are aligned on origins of the indicated
classes. (C) Boxplot distributions of averaged QBU counts across 500bp regions aligned on origins of the indicated groups; the
number of origins in each group is indicated within parentheses. Two-sided t tests were performed on all pairs of strains, and
results are indicated as *P < 0.05 , **P < 0.01, ***P < 0.001, and ****P < 0.0001. (D) Two-dimensional scatter plot comparing
the differences in QBU values (500-bp regions) for each fkh1 mutant versus wild-type. Origins are color-coded as indicated.
Dbf4 is recruited to origins through a fusion to the Fkh1 DBD. Our results here are consistent with
those of Fang et al., and the quantitative analysis here showing that Dbf4-FKH fully complements
fkh1∆ in origin firing seemingly solidifies the conclusion that the only consequential replication
81
function of Fkh1’s FHA is to recruit Dbf4 to origins. However, we note that this experiment also
bypasses any dependence on binding to Dbf4 that Fkh1 might have.
4.10 Dbf4 is required for Fkh1 recruitment to origins
The finding that FHA is normally required for Fkh1-origin association while being
dispensable for origin recruitment of and stimulation by Dbf4-FKH suggests that Fkh1 origin
binding requires its association with Dbf4, likely mediated by FHA, for docking with origin-loaded
ORC and/or MCM complex(es). Moreover, robust complementation by the Dbf4-FKH fusion
protein implies that fusion of Dbf4 with the Fkh1 FKH domain enhances binding of the Fkh1 FKH
domain to origins, because we showed the Fkh1-R80A is defective in origin binding, suggesting
that Fkh1-FKH should also be defective as it lacks the entire FHA domain. To test the idea that
Dbf4 helps recruit Fkh1 to origins, we tested Fkh1 origin binding in cells harboring temperature-
sensitive, hypomorphic allele dbf4∆C (Jones et al., 2010; Petrie et al., 2022). G1-synchronized
WT and dbf4∆C cells were grown at a permissive temperature and analyzed by ChIP-seq of Fkh1-
Myc9. Two-dimensional scatter plots of sample replicates show excellent reproducibility of the
data (Fig. S7). This data shows that there is a reduction in Fkh1 binding across the genome (4.7A).
Particularly, comparisons of WT and dbf4∆C data by two-dimensional scatter plots show robust
correlation of signals across the genome except at replication origins, including Fkh1-activated
origins which are highlighted (Figure 4.7B). Further boxplot and statistical analysis indicated that
there is a significant loss of Fkh1 binding at Fkh-activated origins, CLB2 genes, but not CEN-
proximal origins which are not usually bound by Fkh1 (4.7C). This disruption of Dbf4 function
reduces or arguably eliminates Fkh1 binding, strongly suggesting a cooperativity that enables these
proteins to bind at origins.
82
Figure 4.7 dbf4∆C reduces Fkh1 binding to origins. Strains MPy108 (FKH1-MYC) and MPy189 (dbf4∆C, FKH1-MYC) were
synchronized in G1 phase then subjected to ChIP-seq. (A) Average ChIP-seq profiles are shown for representative chromosomes
III and IX; origins and select features are color coded as indicated above plots. (B) Two-dimensional scatter plots comparing
ChIP-seq signals averaged across 500bp regions centered on 410 confirmed origins; origins and Fkh1 binding loci groups are
color-coded as indicated. A linear regression deriving the best-fit line is shown for the groups. (C) Box plots distributions of ChIP-
seq enrichments for 500bp regions aligned on indicated groups; the number of features in each group is indicated within
parentheses. Two-sided t tests were performed on all pairs of strains.
4.11 Fkh1 and Dbf4 physically interact
The idea that Fkh1 and Dbf4 are interdependent in origin DNA binding suggests that these
proteins should be physically associated with each other. We tested for physical interaction by co-
immunoprecipitation (co-IP) using extracts from G1-synchronized cells expressing 3xFLAG-
DBF4 and expressing FKH1-9xMYC from the GAL promoter, allowing repression in raffinose (R)
and over-expression in galactose (G); this approach was used because previous attempts with
native expression of FKH1-9xMYC were unsuccessful. Whole cell extracts (WCE) and IPs
against the epitope-tagged proteins were analyzed by immunoblotting against the epitope-tagged
proteins, and a polyclonal antibody against Dbf4 was used to verify identification of 3xFLAG-
DBF4 alongside strains expressing native DBF4, which differ by ~3kDa. IP with anti-MYC
antibody against FKH1-9xMYC resulted in co-IP of 3xFLAG-Dbf4 only in strains containing
83
Figure 4.8 Dbf4 binds to Fkh1 is dependent on the Fkh1-FHA. Strains SSy161 harboring episomal plasmid expressing GAL-FKH1-
MYC and MPy138 3xFLAG-Dbf4 strains harboring an episomal plasmid expressing either GAL-FKH1-MYC or GAL-fkh1-R80A-
MYC. Cells were grown in raffinose (R) overnight then as indicated shifted to galactose (G) for 1 hr before harvesting unless
otherwise indicated. (A) Cells containing the indicated tagged proteins were grown as indicated and then subjected to a MYC-IP
or just WCE. (B) Cells were subjected to FLAG-IP followed by treatment indicated levels of EtBr.(C) Cells containing the indicated
proteins were grown and subjected to MYC-IP or FLAG-IP or just WCE with or without Benzonase as indicated.
3xFLAG-DBF4 and galactose, but not when the carbon source was raffinose or glucose (Figure
4.8A). Thus, further experiments were performed similarly as the overexpression of Fkh1-9xMYC
appears to be required for co-IP of 3xFLAG-Dbf4. The reciprocal IP with anti-FLAG antibody
against DBF4-3xFLAG resulted in co-IP of Fkh1-9xMYC. However, the co-IP was affected by
incubating the IP with increasing levels of ethidium bromide (EtBr), known to dissociate proteins
from DNA suggesting that Dbf4 and FKH1 interact in a DNA dependent manner (Figure 4.8B).
84
To test whether the Fkh1-FHA domain plays a role in this interaction the R80A point
mutation was similarly expressed under the GAL promoter. IP of anti-FLAG against 3xFLAG-
DBF4 followed by incubation with nuclease resulted in a strong reduction of the co-IP of both
FKH1-9xMYC and fkh1-R80A-9xMYC. The band intensities make distinguishing differences
between the co-IP unclear. The reciprocal IP of anti-MYC against either FKH1-9xMYC or fkh1-
R80A-9xMYC and then incubation with nuclease resulted in co-IP of 3xFLAG-Dbf4 only for
FKH1-9xMYC. This result implicates the Fkh1-FHA domain to be critical in facilitating the
physical interaction between Fkh1 and Dbf4. Interestingly the interaction does not seem to be
DNA dependent when the IP is done for Fkh1 first conflicting with what was implied when the IP
is done for Dbf4. Overall, these results suggest that Fkh1 and Dbf4 physically associate in vivo
and in a Fkh1-FHA and potentially DNA dependent manner.
Discussion
The above results elucidate a key molecular interaction between Fkh1 and Dbf4 which is
dependent on the Fkh1-FHA and crucial for effective origin binding. The Fkh1-FHA domain is
responsible for mediating phosphorylation-dependent interactions through binding phospho-
threonines. Loss of the FHA domain function through the point mutation R80A prevented protein-
protein interactions and resulted in reduced origin firing of Fkh-activated origins across the
genome. This result was apparent without and with Fkh2, which our past studies have suggested
can substitute Fkh1 loss (Knott et al., 2012). In contrast, to a recent study by Hoggard et al. we did
not observe that the disruption of the Fkh1-FHA affects ORC stabilization or MCM4 binding to
origins. However, we found that rather surprisingly the disruption of the Fkh1-FHA function
reduced its own binding. Specifically, there was a reduction in Fkh1 binding to origins, but not at
other known sites such as the CLB2 cluster genes or the RE element. Previous investigations have
85
noted Fkh1’s separation of function as a transcription factor and as a regulator of origin firing
(Knott et al., 2012). In addition, the non-dominance by fkh1-R80A in origin activation is consistent
with its defect in chromatin binding allowing substitution by Fkh2. Fkh1/2 both contain a Forkhead
associated (FHA) domain through which they regulate transcription of the CB2 cluster genes, with
Fkh1 acting as a temporal repressor and Fkh2 a temporal activator Hollenhorst, 2000; Kumar,
2000; Pic, 2000; Zhu et al., 2000). The combined deletion is pseudohyphal but not lethal and single
deletions of either leads to a more minor transcriptional disruption as the remaining Fkh is
sufficient to partially substitute for the other. Additional studies have indicated that specifically
the Fkh1-FHA domain plays a role in yeast mating type switching when bound to the RE
facilitating donor preference (Li, 2012). Further investigation revealed two Fkh1 FHA-interacting
proteins, Mph1 and Fdo1, as partially required for donor preference, suggesting there are
additional targets that remain to be identified (Dummer et al., 2016). Together this indicates that
the FHA domain is an important effector of Fkh1 function encompassing donor preference,
transcription, and origin activation.
With our identification of the Fkh1-FHA as the responsible domain for the binding and
early firing of Fkh-activated origins this fits well into investigating the potential phosphothreonine
binding target Dbf4 part of the DDK complex. We found that when Fkh1 is deleted Dbf4 also has
reduced binding to origins implicating a Fkh1 directed binding of Dbf4 to origins. Therefore, we
investigated how the fusion construct DBF4-FKH in the absence of FKH1 affected global
replication (Fang et al., 2018). This construct was able to fully rescue the Fkh-activated origin
firing. This supports previous findings that the C-term of DBF4, which is disrupted by the FKH
tag, prevents recruitment to CENs via Ctf19 kinetochore complex interaction (Natsume et al.,
86
2013). In addition, this finding implies that the sole function of the Fkh1-FHA in origin activation
is to bind DBF4.
To further investigate the potential interdependency of Fkh1 and Dbf4 for origin binding
we utilized the dbf4∆C hypomorphic allele. We found that similar to the fkh1-R80A there is also
reduced binding of Fkh1 at origins. Our chromatin binding and replication firing patterns together
with past findings have indicated that Fkh1 and Dbf4 physically interact (Fang et al., 2018). As
predicted we demonstrated with co-IP that not only do Fkh1 and Dbf4 physically bind, this is also
dependent on the Fkh1-FHA. As this time, it remains unclear as to which phosphothreonine(s) in
Dbf4 are responsible for the interaction with the Fkh1-FHA. Thus far mutation of 10 potential
phosphor-threonines in Dbf4 have yet to explain the observed interaction with Fkh1 (Figure S8).
Clarifying the mechanism of interaction of Dbf4 and Fkh1 would solidify the proposed model that
Fkh1-FHA binds with Dbf4 to cooperatively bind to Fkh-activated origins and thus cause their
selective early firing.
87
Figure S1 Supplement to figure 4.1. (A) Fkh1 and Fkh1-fha∆ expressed in and purified from E. coli as described previously
(Ostrow et al., 2017) was subjected to EMSA. Lanes labeled 1-8 contain 0, 3.1, 6.2, 12.5, 25, 50, 100, 200ng Fkh1 or Fkh1-fha∆,
respectively. (B) Strains OAy1146 (FKH1-3xFLAG), OAy1147 (fkh1-fha∆-3xFLAG), OAy1112 (fkh1-fha∆-9xMYC), ZOy48
(FKH1-9xMYC), YHy1(fkh1-R80A-9xMYC), and MPy105 (MOP) were harvested and subjected to western blot using anti-FLAG
or anti-MYC and indicates the deletion of the FHA domain reduces Fkh1 expression but the R80A mutation does not. (C) Strains
described in Figure 4.1 legend were analyzed by FACScan after G1 phase block-and-release into S phase in the absence of HU.
(D) Two-dimensional scatter plots analyzing correlation of QBU replicates of strains (E) Heat maps of correlation coefficients of
QBU counts for 5kb regions surrounding 409 origins.
88
Figure S2 Supplement to figure 4.2. (A) Two-dimensional scatter plots analyzing correlation of QBU replicates of strains
described in Figure 4.2 (B) Heat maps of correlation coefficients of QBU counts for 5kb regions surrounding 410 origins. (C)
Two-dimensional scatter plot comparing the differences in QBU values (500-bp regions) for fkh1-R80A versus wild-type. Origins
are color-coded as indicated.
Figure S3 Supplement to figure 4.3. (A) Two-dimensional scatter plots analyzing correlation of G1 arrested ChIP-seq replicates
of ORC1 strains described in Figure 3 (B) Two-dimensional scatter plots analyzing correlation of ChIP-seq replicates of MCM4
strains described in Figure 4.3
89
Figure S4 Supplement to figure 4.4. (A) Two-dimensional scatter plots analyzing correlation of ChIP-seq replicates of strains
described in Figure 4.4 (B) Two-dimensional scatter plots analyzing correlation of G1 arrested ChIP-seq replicates of strains
described in Figure 4.4
Figure S5 Supplement to figure 4.5. Two-dimensional scatter plots analyzing correlation of G1 arrested ChIP-seq replicates of
strains described in Figure 4.5
Figure S6 Supplement to figure 4.6. Two-dimensional scatter plots analyzing correlation of QBU replicates of strains described
in Figure 4.6
R
2
=0.992 R
2
=0.987 R
2
=0.995
0
10
5
15
15 10 5 0
Rep1
Rep2
0
10
5
15
15 10 5 0
Rep1
Rep3
0
10
5
15
15 10 5 0
Rep2
Rep3
3xFLAG-DBF4
R
2
=0.988 R
2
=0.979 R
2
=0.993
0
10
5
15
15 10 5 0
Rep1
Rep2
0
10
5
15
15 10 5 0
Rep1
Rep3
0
10
5
15
15 10 5 0
Rep2
Rep3
f k h 1 ∆ 3x FLAG- DBF4
90
Figure S7 Supplement to figure 4.7. Two-dimensional scatter plots analyzing correlation of G1 arrested and shifted to 30˚C
ChIP-seq replicates of strains described in Figure 4.7
Figure S8 dbf4-10A does not alter origin replication patterns. Strains CVy63 (WT) and MPy206 (dbf4-10A) were analyzed by QBU
as described in Figure 1 legend. (A) Average BrdU incorporation profiles are shown for representative chromosomes VI and IX;
origins and Fkh-activated are indicated with gray and blue circles, respectively. (B) Heat maps of QBU counts for 10kb regions
are aligned on origins of the indicated classes. (C) Two-dimensional scatter plots analyzing correlation of QBU replicates.
91
Chapter 5: Future Directions
Research Impacts of this Dissertation
Chapter 2 of my dissertation expanded upon the knowledge that Fkh1 drives early
replication firing and outlines a relationship between sub-nuclear localization and origin activation
timing. Although our proposed model in chapter 2 indicates that mechanisms that allow origins to
move away from or remain tethered to the nuclear periphery remains unclear, we still demonstrated
that Fkh1 increases movement and overall a significant relocalization of Fkh-activated origins to
the interior of the nucleus as compared to when it is absent. Understanding 3D-organization of the
genome continues to be an ongoing question. Long distance regions of the genome can be seen to
influence each other through the formation of TADs, some of which have been proposed to be
established by Fkh1/2 (Eser et al. 2017). We found that FBSs are required for both Fkh1 driven
early origin firing and mediated relocalization away from the nuclear periphery (Knott et al. 2012
and section 2.8 and 2.9). Lastly, we found that Fkh-activated origins had significantly more
dynamic movement as compared to the reduced and constricted localization to the periphery when
Fkh1 was deleted (section 2.11). This outlines a mechanism in which Fkh1 influences origin firing
through localization of Fkh-activated origins away from nuclear periphery and in proximity to
early firing origin TADs.
In chapter 3 we demonstrated that there are distinct mechanisms of recruitment of DDK to
CEN-proximal and Fkh-activated origins. Specifically, we found that the C-terminus of Dbf4 is
crucial for Ctf19 recruitment but only partially affects Fkh1 recruitment (see section 3.7 and 3.8
further discussion). In our case we asked whether the disruption of the Dbf4 Zn-finger through
point mutations affected both CEN-proximal and Fkh-activated origins similarly using QBU (see
92
section 3.6). Our results showed that the dbf4-Zn* that we constructed phenocopied Ctf19 deletion
in which only the CEN-proximal origins were significantly delayed. This research extends other
studies showing that Ctf19 is the main recruiter for Dbf4 to CEN-proximal origins and that deletion
of the Dbf4 C-terminus leads to a general replication defect seen through its temperature sensitivity
potentially due to protein misfolding (Natsume et al. 2013; Jones et al. 2010). We therefore bridge
these observations and were able to conclude that the Dbf4-Zn finger is not temperature sensitive
and is required for Ctf19 recruitment (see chapter 3). However, the recent study indicating the C-
term as the key recruiter of Fkh1 was not supported by our results (Fang et al. 2018). Instead we
determined that there is still an additional region of Dbf4 that is interacting with Fkh1.
Lastly in chapter 4 we discovered that it is the Fkh1-FHA domain that regulates its
interaction with Dbf4. We additionally found that when the Fkh1-FHA is deleted or disrupted
through the R80A point mutation, thus preventing protein-protein interactions through
phosphothreonine binding, there is a significant decrease in Fkh-activated origin firing (see 4.5)
This points to a mechanism in which the role of Fkh1-FHA domain it to bind another protein
through phosho-theronine(s) enabling this observed early firing. A recent study proposed that the
FHA domain of Fkh1 plays a role in the stabilization of ORC binding (Hoggard et al., 2021). This
was investigated due to the close or overlapping FBS at select ACSs, which when present allows
these origins to be more efficient and earlier firing (Hoggard et al., 2021; Aparicio et al., 2013).
Unlike Hoggard et al 2021, we were unable to uncover any sign of FHA requirement for pre-RC
stabilization (Knott et al., 2012, see section 4.6). Our data suggests that the FHA domain is
primarily responsible for both origin activation and binding to origins (see section 4.5 and 4.7).
When analyzing Dbf4 binding we determined that it preferentially binds Fkh-activated origins and
Centromeres. This supports that Ctf19 recruits Dbf4 to the kinetochore complex (Natsume et al.
93
2013). In addition, analysis has shown a dependency of the Fkh1-FHA for Dbf4 interaction and
vice versa as when this domain is disrupted or FKH1 is absent their binding to origins is reduced
(see section 4.8 and 4.10). Overall, we determined that the Fkh1-FHA influences origin firing
through a physical recruitment of Dbf4 (see section 4.9 and 4.11). However, it remains unclear
what region of Dbf4 the Fkh1-FHA binds to and whether the Fkh1-Dbf4 interaction occurs before
or after binding to the pre-RC.
Future Experiments
Understanding the finer details of how DNA replication machinery is established and
initiation timing is organized remains an ongoing area of exploration. The fact that research is still
being done to determine how MCM, a critical factor of the pre-RC is loaded at origins highlights
the continuing details in which the field strives to define (Coster and Diffley 2017). Thus, the
defining of pre-IC composition and the mechanism of recruitment will help to further
understanding of this essential process. These future experiments will further elucidate the role of
Fkh1 in origin firing timing and 3D-genome organization.
5.1 Defining the Dbf4 domain for Fkh-FHA interaction
In chapter 4 we demonstrated through QBU and ChIP-seq that Fkh1 and Dbf4 appear to
function in the same pathway to bind and enable early firing of select origins and co-IP analysis
implies there is a physical interaction between Fkh1 and Dbf4. Yet as we argued in chapter 3 (see
sections 3.1 and 3.2), one of the difficulties is that Dbf4 is essential and disrupting the protein can
quickly lead to temperature sensitivity or inviability making it a delicate process. Due to our
interest in an interaction between Dbf4 and the phosphor-threonine binding FHA domain of Fkh1
we focused previous analysis on mutating 10 threonines identified as being phosphorylation
94
targets, but thus far this has yet to explain the observed interaction (Figure S8). Other studies had
to mutate 19 sites in Dbf4 that were a combination of potential phospho-serines and phospho-
theronies to disrupt the interaction between Dbf4 and Rad53 (Zegerman and Diffley 2010).
Therefore, we propose to create and test Dbf4 mutants in which all 63 threonines are changed to
alanines or serines. However, as again Dbf4 is essential there is reason to suspect that these many
mutations may result in inviability. As an alternative we will also create mutants in which there is
a reduction of mutations for example with only roughly half the theronines mutated at either the N
or C terminus split after the defined M domain. An additional avenue would be to further reduce
the number of mutations to 15A, which comprises the 10A already investigated along with 5 more
potential phosphor-threonine sites. Understanding the regions of Dbf4 that interact with Fkh1
would solidify the molecular interactions and strengthen the model as to how these select origins
fire early.
5.2 Global search for FKH1 interactors
In chapters 2-4 we described a number of pathways involving Fkh1. There are still further
interactors we have not identified. So, we propose to shift from directed interaction analysis such
as those with Dbf4 to look more globally. As there is the distinct possibility that Fkh1 is interacting
with Dbf4 in a complex such as with its Cdc7 binding partner, which together comprises DDK.
This will be done by purifying Fkh1 and fkh1-R80A and subjecting it to Mass Spectrometry. This
will not only open possibilities of discovering further interactors, but will hopefully serve to
reinforce past interaction networks such as Fkh1 binding with Fkh2 and Cdc28 (Ho et al., 2002).
95
References
Chapter 1 References
Aparicio JG, Viggiani CJ, Gibson DG, Aparicio OM. (2004). The Rpd3-Sin3 histone deacetylase
regulates replication timing and enables intra-S origin control in Saccharomyces
cerevisiae. Molecular and cellular biology, 24(11), 4769–4780.
https://doi.org/10.1128/MCB.24.11.4769-4780.2004
Aparicio OM. (2013). Location, location, location: it's all in the timing for replication origins.
Genes Dev. 27(2):117-28. doi: 10.1101/gad.209999.
Bell SP and Kaguni JM. (2013). Helicase loading at chromosomal origins of replication. Cold
Spring Harbor perspectives in biology, 5(6), a010124. doi: 10.1101/cshperspect.a010124
Chang F, May CD, Hoggard T, Miller J, Fox CA, Weinreich M. (2011). High-resolution analysis
of four efficient yeast replication origins reveals new insights into the ORC and putative
MCM binding elements. Nucleic Acids Res. 39(15):6523-6535. doi:10.1093/nar/gkr301
Costa A and Diffley JFX. (2022). The Initiation of Eukaryotic DNA Replication. Annual review
of biochemistry, 91, 107–131. doi: 10.1146/annurev-biochem-072321-110228
Coster G and Diffley JFX. (2017). Bidirectional eukaryotic DNA replication is established by
quasi-symmetrical helicase loading. Science 357(6348): p314-318. doi:
10.1126/science.aan0063.
Davé A, Cooley C, Garg M, Bianchi A. (2014). Protein phosphatase 1 recruitment
by Rif1 regulates DNA replication origin firing by counteracting DDK activity. Cell Rep.
7(1):53-61. doi: 10.1016/j.celrep.2014.02.019.
Ding Q and MacAlpine DM. (2011). Defining the replication program through the chromatin
landscape. Critical reviews in biochemistry and molecular biology, 46(2), 165–179.
https://doi.org/10.3109/10409238.2011.560139
Duncker L and Duncker BP. (2016). Mechanisms Governing DDK Regulation of the Initiation of
DNA Replication. Genes 8(1), 3; https://doi.org/10.3390/genes8010003
Eaton ML, Galani K, Kang S, Bell SP, MacAlpine DM. (2010). Conserved nucleosome
positioning defines replication origins. Genes Dev. 24(8):748-53. doi:
10.1101/gad.1913210.
Ebrahimi H, Robertson ED, Taddei A, Gasser SM, Donaldson AD, Hiraga S. (2010). Early
initiation of a replication origin tethered at the nuclear periphery. Journal of Cell
Science 123:1015–1019. doi: 10.1242/jcs.060392
96
Fang D, Lengronne A, Shi D, Forey R, Skrzypczak M, Ginalski K, Yan C, Wang X, Cao Q,
Pasero P, Lou H. (2018). Dbf4 recruitment by forkhead transcription factors defines an
upstream rate-limiting step in determining origin firing timing. Genes & Dev 31:2405–
2415. DOI: https://doi.org/10.1101/gad.306571.117
He Y, Petrie MV, Zhang H, Peace JM, Aparicio OM. (2022). Rpd3 regulates single-copy origins
independently of the rDNA array by opposing Fkh1-mediated origin
stimulation. PNAS, 119(40), e2212134119. doi: /10.1073/pnas.2212134119
Heun P, Laroche T, Raghuraman MK, Gasser SM. (2001). The positioning and dynamics of
origins of replication in the budding yeast nucleus. J. Cell Bio 152:385–400.
Hiraga S, Alvino GM, Chang F, Lian HY, Sridhar A, Kubota T, Brewer BJ, Weinreich M,
Raghuraman MK, Donaldson AD. (2014). Rif1 controls DNA replication by directing
Protein Phosphatase 1 to reverse Cdc7-mediated phosphorylation of the MCM complex.
Genes Dev. 28(4):372-83. doi: 10.1101/gad.231258.113.
Hoggard T, Müller CA, Nieduszynski CA, Weinreich M, Fox CA.(2020). Sir2 mitigates an
intrinsic imbalance in origin licensing efficiency between early- and late-replicating
euchromatin. PNAS. 117(25):14314-14321. doi:10.1073/pnas.2004664117
Hoggard T, Hollatz AJ, Cherney RE, Seman MR, Fox CA. (2021). The Fkh1 Forkhead
associated domain promotes ORC binding to a subset of DNA replication origins in
budding yeast. Nucleic Acids Res.49(18):10207-10220. doi:10.1093/nar/gkab450
Hollenhorst PC, Bose ME, Mielke MR, Müller U, Fox CA. (2000). Forkhead genes in
transcriptional silencing, cell morphology and the cell cycle. Overlapping and distinct
functions for FKH1 and FKH2 in Saccharomyces cerevisiae. Genetics. 2000
Apr;154(4):1533-48.
Jin Y, Liang Z, Lou H. The Emerging Roles of Fox Family Transcription Factors in
Chromosome Replication, Organization, and Genome Stability. (2020). Cells. 9(1):258.
https://doi.org/10.3390/cells9010258
Jones DR, Prasad AA, Chan PK, Duncker BP. (2010). The Dbf4 motif C zinc finger promotes
DNA replication and mediates resistance to genotoxic stress. Cell Cycle 9(10):2018-26.
Knott SR, Peace JM, Ostrow AZ, Gan Y, Rex AE, Viggiani CJ, Tavaré S, Aparicio OM. (2012).
Forkhead Transcription Factors Establish Origin Timing and Long-Range Clustering in
S. cerevisiae. Cell 148(1-2): 99-111. https://doi.org/10.1016/j.cell.2011.12.012.
Knott SR, Viggiani CJ, Tavaré S, Aparicio OM. (2009). Genome-wide replication profiles
indicate an expansive role for Rpd3L in regulating replication initiation timing or
efficiency, and reveal genomic loci of Rpd3 function in Saccharomyces cerevisiae. Genes
Dev. 23(9):1077-1090. doi:10.1101/gad.1784309
97
Koranda M, Schleiffer A, Endler L, Ammerer G. (2000). Forkhead-like transcription factors
recruit Ndd1 to the chromatin of G2/M-specific promoters. Nature. 406(6791):94-98.
doi:10.1038/35017589
Kumar A, des Etages SA, Coelho PS, Roeder GS, Snyder M. (2000). High-throughput methods
for the large-scale analysis of gene function by transposon tagging. Methods Enzymol.
328:550-574. doi:10.1016/s0076-6879(00)28418-8
Leonard AC and Méchali M. (2013). DNA replication origins. Cold Spring Harb Perspect Biol.
5(10):a010116. doi: 10.1101/cshperspect.a010116.
Li J, Coïc E, Lee K, Lee CS, Kim JA, Wu Q, Haber JE. (2012). Regulation of budding yeast
mating-type switching donor preference by the FHA domain of Fkh1. PLoS Genet.
8(4):e1002630. doi:10.1371/journal.pgen.1002630
Lochmann B and Ivanov D. (2012). Histone H3 localizes to the centromeric DNA in budding
yeast. PLoS genetics, 8(5), e1002739. https://doi.org/10.1371/journal.pgen.1002739
Mantiero D, Mackenzie A, Donaldson A, Zegerman P. (1996). Limiting replication initiation
factors execute the temporal programme of origin firing in budding yeast. Nature.
381(6579):251-3.
Matthews LA, and Guarné A. (2013). Dbf4: the whole is greater than the sum of its parts. Cell
Cycle. 12(8):1180-8. doi: 10.4161/cc.24416.
Matthews LA, Selvaratnam R, Jones DR, Akimoto M, McConkey BJ, Melacini G, Duncker BP,
Guarné A. (2014). Novel Non-canonical Forkhead-associated (FHA) Domain-binding
Interface Mediates the Interaction between Rad53 and Dbf4 Proteins. JCB 289(5):
p2589-99. doi: 10.1074/jbc.M113.517060.
Mondeel TDGA, Holland P, Nielsen J, Barberis M. (2019). ChIP-exo analysis highlights Fkh1
and Fkh2 transcription factors as hubs that integrate multi-scale networks in budding
yeast. Nucleic Acids Research 47(15):7825–7841
Natsume T, Müller CA, Katou Y, Retkute R, Gierliński M, Araki H, Blow JJ, Shirahige K,
Nieduszynski CA, Tanaka TU. (2013). Kinetochores coordinate pericentromeric
cohesion and early DNA replication by Cdc7-Dbf4 kinase recruitment. Mol Cell 50(5):
p661-74. doi: 10.1016/j.molcel.2013.05.011.
Natsume T and Tanaka TU. (2013). Spatial regulation and organization of DNA replication
within the nucleus. Chromosome Res. 18(1):7-17. doi: 10.1007/s10577-009-9088-0.
Ostrow AZ, Kalhor R, Gan Y, Villwock SK, Linke C, Barberis M, Chen L, Aparicio OM.
(2017). Conserved forkhead dimerization motif controls DNA replication timing and
spatial organization of chromosomes in S. cerevisiae. PNAS 114(12) E2411-
E2419. https://doi.org/10.1073/pnas.1612422114
98
Pasero P, Bensimon A, Schwob E. (2002). Single-molecule analysis reveals clustering and
epigenetic regulation of replication origins at the yeast rDNA locus. Genes Dev. 16,
2479–2484. doi: 10.1101/gad.232902.
Peace JM, Ter-Zakarian A, Aparicio OM. (2014). Rif1 Regulates Initiation Timing of Late
Replication Origins throughout the S. cerevisiae Genome. PLOS ONE 9(5):
e98501. https://doi.org/10.1371/journal.pone.0098501
Peace JM, Villwock SK, Zeytounian JL, Gan Y, Aparicio OM. (2016). Quantitative BrdU
immunoprecipitation method demonstrates that Fkh1 and Fkh2 are rate-limiting
activators of replication origins that reprogram replication timing in G1 phase. Genome
Res 2016. 26:365-375. doi: 10.1101/gr.196857.115
Pramila T, Wu W, Miles S, Noble WS, Breeden, LL. (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 Dev. 20, 2266–2278
Rao H and Stillman B. (1995). The origin recognition complex interacts with a bipartite DNA
binding site within yeast replicators. PNAS. 92(6):2224-8
Rhind N and Gilbert DM. (2013). DNA replication timing. Cold Spring Harb Perspect Biol.
5(8):a010132. doi:10.1101/cshperspect.a010132
Shi BJ. (2016). Decoding common and divergent cellular functions of the domains of forkhead
transcription factors Fkh1 and Fkh2. Biochem J. 2016 Nov 1;473(21):3855-3869.
Stinchcomb DT, Struhl K, Davis RW. (1979). Isolation and characterisation of a yeast
chromosomal replicator. Nature. 282(5734):39-43. doi:10.1038/282039a0
Sun K, Coïc E, Zhou Z, Durrens P, Haber JE. (2002). Saccharomyces forkhead protein Fkh1
regulates donor preference during mating-type switching through the recombination
enhancer. Genes Dev. 16(16):2085-2096. doi:10.1101/gad.994902
Varrin AE, Prasad AA, Scholz RP, Ramer MD, Duncker BP. (2005). A mutation in Dbf4 motif M
impairs interactions with DNA replication factors and confers increased resistance to
genotoxic agents. Mol Cell Biol 25(17): p7494-504.
Vogelauer M, Rubbi L, Lucas I, Brewer BJ, Grunstein M. (2002). Histone acetylation regulates
the time of replication origin firing. Mol. Cell 10, 1223–1233
Wu X and Haber JE. (1995). MATa donor preference in yeast mating-type switching: activation
of a large chromosomal region for recombination. Genes Dev. 9(15):1922-1932.
doi:10.1101/gad.9.15.1922
Xu W, Aparicio JG, Aparicio OM, Tavaré S. (2006). Genome-wide mapping of ORC and Mcm2p
binding sites on tiling arrays and identification of essential ARS consensus sequences
in S. cerevisiae. BMC Genomics 7(276) doi:10.1186/1471-2164-7-276
99
Yamazaki S, Ishii A, Kanoh Y, Oda M, Nishito Y, Masai H. (2012). Rif1 regulates the
replication timing domains on the human genome. EMBO J. 31(18):3667-77. doi:
10.1038/emboj.2012.180.
Zegerman P. (2015). Evolutionary conservation of the CDK targets in eukaryotic DNA
replication initiation. Chromosoma 124(3):309-21
Zhang H, Petrie MV, He Y, Peace JM, Chiolo IE, Aparicio OM. (2019). Dynamic relocalization
of replication origins by Fkh1 requires execution of DDK function and Cdc45 loading at
origins. eLife 2019;8:e45512. https://doi.org/10.7554/eLife.45512
Zhong Y, Nellimoottil T, Peace JM, Knott SR, Villwock SK, Yee JM, Jancuska JM, Rege S,
Tecklenburg M, Sclafani RA, Tavaré S, Aparicio OM. (2013). The level of origin firing
inversely affects the rate of replication fork progression. The Journal of Cell
Biology 201(3):373-383; doi: 10.1083/jcb.201208060
Zhu G, Spellman PT, Volpe T, Brown PO, Botstein D, Davis TN, Futcher B.(2000). Two yeast
forkhead genes regulate the cell cycle and pseudohyphal growth. Nature. 406(6791):90-
94. doi:10.1038/35017581
Chapter 2 References
Aparicio OM, Stout AM, Bell SP. (1999). Differential assembly of Cdc45p and DNA
polymerases at early and late origins of DNA replication. PNAS 96:9130–9135. DOI:
https://doi.org/10.1073/pnas.96.16.9130, PMID: 10430907
Aparicio OM. (2013). Location, location, location: it’s all in the timing for replication origins.
Genes & Development 27:117–128. DOI: https://doi.org/10.1101/gad.209999.112,
PMID: 23348837
Belgareh N and Doye V. (1997). Dynamics of nuclear pore distribution in nucleoporin mutant
yeast cells. The Journal of Cell Biology 136:747–759. DOI:
https://doi.org/10.1083/jcb.136.4.747, PMID: 9049242
Berezney R, Dubey DD, Huberman JA. (2000). Heterogeneity of eukaryotic replicons, replicon
clusters, and replication foci. Chromosoma 108:471–484. DOI:
https://doi.org/10.1007/s004120050399, PMID: 10794569
Bishop AC, Ubersax JA, Petsch DT, Matheos DP, Gray NS, Blethrow J, Shimizu E, Tsien JZ,
Schultz PG, Rose MD, Wood JL, Morgan DO, Shokat KM. (2000). A chemical switch
for inhibitor-sensitive alleles of any protein kinase. Nature 407:395–401. DOI:
https://doi.org/10.1038/35030148, PMID: 11014197
100
Caridi PC, Delabaere L, Zapotoczny G, Chiolo I. (2017). And yet, it moves: nuclear and
chromatin dynamics of a heterochromatic double-strand break. Philosophical
Transactions of the Royal Society B: Biological Sciences 372:20160291. DOI:
https://doi.org/10.1098/rstb.2016.0291, PMID: 28847828
Caridi CP, Delabaere L, Tjong H, Hopp H, Das D, Alber F, Chiolo I. (2018). Quantitative
methods to investigate the 4D dynamics of heterochromatic repair sites in Drosophila
cells. Methods in Enzymology 601:359–389. DOI:
https://doi.org/10.1016/bs.mie.2017.11.033, PMID: 29523239
Davé A, Cooley C, Garg M, Bianchi A. (2014). Protein phosphatase 1 recruitment by Rif1
regulates DNA replication origin firing by counteracting DDK activity. Cell Reports
7:53–61. DOI: https://doi.org/10.1016/j. celrep.2014.02.019, PMID: 24656819
Duan Z, Andronescu M, Schutz K, McIlwain S, Kim YJ, Lee C, Shendure J, Fields S, Blau CA,
Noble WS. (2010). A three-dimensional model of the yeast genome. Nature 465:363–
367. DOI: https://doi.org/10.1038/ nature08973, PMID: 20436457
Ebrahimi H, Robertson ED, Taddei A, Gasser SM, Donaldson AD, Hiraga S. (2010). Early
initiation of a replication origin tethered at the nuclear periphery. Journal of Cell Science
123:1015–1019. DOI: https://doi.org/10.1242/ jcs.060392, PMID: 20197407
Eser U, Chandler-Brown D, Ay F, Straight AF, Duan Z, Noble WS, Skotheim JM. (2017). Form
and function of topologically associating genomic domains in budding yeast. PNAS
114:E3061–E3070. DOI: https://doi.org/10. 1073/pnas.1612256114, PMID: 28348222
Fang D, Lengronne A, Shi D, Forey R, Skrzypczak M, Ginalski K, Yan C, Wang X, Cao Q,
Pasero P, Lou H. (2017). Dbf4 recruitment by forkhead transcription factors defines an
upstream rate-limiting step in determining origin firing timing. Genes & Development
31:2405–2415. DOI: https://doi.org/10.1101/gad.306571.117, PMID: 2 9330352
Frouin I, Montecucco A, Spadari S, Maga G. (2003). DNA replication: a complex matter. EMBO
Reports 4:666– 670. DOI: https://doi.org/10.1038/sj.embor.embor886, PMID: 12835753
Hafner L, Lezaja A, Zhang X, Lemmens L, Shyian M, Albert B, Follonier C, Nunes JM, Lopes
M, Shore D, Mattarocci S. (2018). Rif1 binding and control of Chromosome-Internal
DNA replication origins is limited by telomere sequestration. Cell Reports 23:983–992.
DOI: https://doi.org/10.1016/j.celrep.2018.03.113, PMID: 2 9694906
Hayano M, Kanoh Y, Matsumoto S, Renard-Guillet C, Shirahige K, Masai H. (2012). Rif1 is a
global regulator of timing of replication origin firing in fission yeast. Genes &
Development 26:137–150. DOI: https://doi.org/10. 1101/gad.178491.111, PMID:
22279046
Hayashi MT, Takahashi TS, Nakagawa T, Nakayama J, Masukata H. (2009). The
heterochromatin protein Swi6/HP1 activates replication origins at the pericentromeric
region and silent mating-type locus. Nature Cell Biology 11: 357–362. DOI:
https://doi.org/10.1038/ncb1845, PMID: 19182789
101
Haye-Bertolozzi JE, Aparicio OM. (2018). Quantitative bromodeoxyuridine
immunoprecipitation analyzed by High-Throughput sequencing (qBrdU-Seq or QBU).
Methods in Molecular Biology 1672:209–225. DOI: https:// doi.org/10.1007/978-1-4939-
7306-4_16, PMID: 29043627
Hediger F, Neumann FR, Van Houwe G, Dubrana K, Gasser SM. (2002). Live imaging of
telomeres: yku and sir proteins define redundant telomere-anchoring pathways in yeast.
Current Biology : CB 12:2076–2089. PMID: 12498682
Heller RC, Kang S, Lam WM, Chen S, Chan CS, Bell SP. (2011). Eukaryotic origin-dependent
DNA replication in vitro reveals sequential action of DDK and S-CDK kinases. Cell
146:80–91. DOI: https://doi.org/10.1016/j.cell. 2011.06.012, PMID: 21729781
Hereford LM, Hartwell LH. (1974). Sequential gene function in the initiation of Saccharomyces
cerevisiae DNA synthesis. Journal of Molecular Biology 84:445–461. DOI:
https://doi.org/10.1016/0022-2836(74)90451-3, PMID: 4618856
Heun P, Laroche T, Raghuraman MK, Gasser SM. (2001a). The positioning and dynamics of
origins of replication in the budding yeast nucleus. The Journal of Cell Biology 152:385–
400. DOI: https://doi.org/10.1083/jcb.152.2. 385, PMID: 11266454
Heun P, Laroche T, Shimada K, Furrer P, Gasser SM. (2001b). Chromosome dynamics in the
yeast interphase nucleus. Science 294:2181–2186. DOI:
https://doi.org/10.1126/science.1065366, PMID: 11739961
Hiraga S, Alvino GM, Chang F, Lian HY, Sridhar A, Kubota T, Brewer BJ, Weinreich M,
Raghuraman MK, Donaldson AD. (2014). Rif1 controls DNA replication by directing
Protein Phosphatase 1 to reverse Cdc7- mediated phosphorylation of the MCM complex.
Genes & Development 28:372–383. DOI: https://doi.org/10. 1101/gad.231258.113,
PMID: 24532715
Hozák P, Jackson DA, Cook PR. (1994). Replication factories and nuclear bodies: the
ultrastructural characterization of replication sites during the cell cycle. Journal of Cell
Science 107:2191–2202. PMID: 7 983177
Ito T, Chiba T, Ozawa R, Yoshida M, Hattori M, Sakaki Y. (2001). A comprehensive two-hybrid
analysis to explore the yeast protein interactome. PNAS 98:4569–4574. DOI:
https://doi.org/10.1073/pnas.061034498, PMID: 112 83351
Kamimura Y, Tak YS, Sugino A, Araki H. (2001). Sld3, which interacts with Cdc45 (Sld4),
functions for chromosomal DNA replication in Saccharomyces cerevisiae. The EMBO
Journal 20:2097–2107. DOI: https:// doi.org/10.1093/emboj/20.8.2097, PMID: 11296242
Kitamura E, Blow JJ, Tanaka TU. (2006). Live-cell imaging reveals replication of individual
replicons in eukaryotic replication factories. Cell 125:1297–1308. DOI:
https://doi.org/10.1016/j.cell.2006.04.041, PMID: 16814716
102
Knott SR, Peace JM, Ostrow AZ, Gan Y, Rex AE, Viggiani CJ, Tavaré S, Aparicio OM. (2012).
Forkhead transcription factors establish origin timing and long-range clustering in S.
cerevisiae. Cell 148:99–111. DOI: https://doi.org/10.1016/j.cell.2011.12.012, PMID:
22265405
Longtine MS, McKenzie A, Demarini DJ, Shah NG, Wach A, Brachat A, Philippsen P, Pringle
JR. (1998). Additional modules for versatile and economical PCR-based gene deletion
and modification in Saccharomyces cerevisiae. Yeast 14:953–961. DOI:
https://doi.org/10.1002/(SICI)1097-0061(199807)14:10<953::AID-YEA293>3.0.CO;2-
U, PMID: 9717241
Lõoke M, Kristjuhan K, Värv S, Kristjuhan A. (2013). Chromatin-dependent and -independent
regulation of DNA replication origin activation in budding yeast. EMBO Reports
14:191–198. DOI: https://doi.org/10.1038/embor. 2012.196, PMID: 23222539
Mantiero D, Mackenzie A, Donaldson A, Zegerman P. (2011). Limiting replication initiation
factors execute the temporal programme of origin firing in budding yeast. The EMBO
Journal 30:4805–4814. DOI: https://doi.org/ 10.1038/emboj.2011.404, PMID: 22081107
Marshall WF, Straight A, Marko JF, Swedlow J, Dernburg A, Belmont A, Murray AW, Agard
DA, Sedat JW. (1997). Interphase chromosomes undergo constrained diffusional motion
in living cells. Current Biology 7:930– 939. DOI: https://doi.org/10.1016/S0960-
9822(06)00412-X, PMID: 9382846
Mattarocci S, Shyian M, Lemmens L, Damay P, Altintas DM, Shi T, Bartholomew CR, Thomä
NH, Hardy CF, Shore D. (2014). Rif1 controls DNA replication timing in yeast through
the PP1 phosphatase Glc7. Cell Reports 7:62– 69. DOI:
https://doi.org/10.1016/j.celrep.2014.03.010, PMID: 24685139
Mendenhall MD and Hodge AE. (1998). Regulation of Cdc28 cyclin-dependent protein kinase
activity during the cell cycle of the yeast saccharomyces cerevisiae. Microbiology and
Molecular Biology Reviews : MMBR 62:1191– 1243. PMID: 9841670
Nakamura H, Morita T, Sato C. 1986. Structural organizations of replicon domains during DNA
synthetic phase in the mammalian nucleus. Experimental Cell Research 165:291–297.
DOI: https://doi.org/10.1016/0014-4827(86) 90583-5, PMID: 3720850
Natsume T, Müller CA, Katou Y, Retkute R, Gierliński M, Araki H, Blow JJ, Shirahige K,
Nieduszynski CA, Tanaka TU. (2013). Kinetochores coordinate pericentromeric cohesion
and early DNA replication by Cdc7-Dbf4 kinase recruitment. Molecular Cell 50:661–
674. DOI: https://doi.org/10.1016/j.molcel.2013.05.011, PMID: 23746350
Newport J and Yan H. (1996). Organization of DNA into foci during replication. Current
Opinion in Cell Biology 8: 365–368. DOI: https://doi.org/10.1016/S0955-
0674(96)80011-1, PMID: 8743888
103
Nougarè de R, Della Seta F, Zarzov P, Schwob E. (2000). Hierarchy of S-phase-promoting
factors: yeast Dbf4-Cdc7 kinase requires prior S-phase cyclin-dependent kinase
activation. Molecular and Cellular Biology 20:3795–3806. DOI:
https://doi.org/10.1128/MCB.20.11.3795-3806.2000, PMID: 10805723
Oshiro G, Owens JC, Shellman Y, Sclafani RA, Li JJ. (1999). Cell cycle control of Cdc7p kinase
activity through regulation of Dbf4p stability. Molecular and Cellular Biology 19:4888–
4896. DOI: https://doi.org/10.1128/MCB. 19.7.4888, PMID: 10373538
Ostrow AZ, Nellimoottil T, Knott SR, Fox CA, Tavaré S, Aparicio OM. (2014). Fkh1 and Fkh2
bind multiple chromosomal elements in the S. cerevisiae genome with distinct
specificities and cell cycle dynamics. PLOS ONE 9:e87647. DOI:
https://doi.org/10.1371/journal.pone.0087647, PMID: 24504085
Ostrow AZ, Viggiani CJ, Aparicio JG, Aparicio OM. (2015). ChIP-Seq to analyze the binding of
replication proteins to chromatin. Methods in Molecular Biology 1300:155–168. DOI:
https://doi.org/10.1007/978-1-4939- 2596-4_11, PMID: 25916712
Ostrow AZ, Kalhor R, Gan Y, Villwock SK, Linke C, Barberis M, Chen L, Aparicio OM.
(2017). Conserved forkhead dimerization motif controls DNA replication timing and
spatial organization of chromosomes in S. cerevisiae. PNAS 114:E2411–E2419. DOI:
https://doi.org/10.1073/pnas.1612422114, PMID: 28265091
Owens JC, Detweiler CS, Li JJ. (1997). CDC45 is required in conjunction with CDC7/DBF4 to
trigger the initiation of DNA replication. PNAS 94:12521–12526. DOI:
https://doi.org/10.1073/pnas.94.23.12521, PMID: 9356482
Park S, Patterson EE, Cobb J, Audhya A, Gartenberg MR, Fox CA. (2011). Palmitoylation
controls the dynamics of budding-yeast heterochromatin via the telomere-binding protein
Rif1. PNAS 108:14572–14577. DOI: https://doi.org/10.1073/pnas.1105262108, PMID:
21844336
Patel PK, Kommajosyula N, Rosebrock A, Bensimon A, Leatherwood J, Bechhoefer J, Rhind N.
(2008). The Hsk1 (Cdc7) replication kinase regulates origin efficiency. Molecular
Biology of the Cell 19:5550–5558. DOI: https:// doi.org/10.1091/mbc.e08-06-0645,
PMID: 18799612
Peace JM, Ter-Zakarian A, Aparicio OM. (2014). Rif1 regulates initiation timing of late
replication origins throughout the S. cerevisiae genome. PLOS ONE 9:e98501. DOI:
https://doi.org/10.1371/journal.pone.0098501, PMID: 24879017
Peace JM, Villwock SK, Zeytounian JL, Gan Y, Aparicio OM. (2016). Quantitative BrdU
immunoprecipitation method demonstrates that Fkh1 and Fkh2 are rate-limiting
activators of replication origins that reprogram replication timing in G1 phase. Genome
Research 26:365–375. DOI: https://doi.org/10.1101/gr.196857.115, PMID: 26728715
104
Randell JC, Fan A, Chan C, Francis LI, Heller RC, Galani K, Bell SP. (2010). Mec1 is one of
multiple kinases that prime the Mcm2-7 helicase for phosphorylation by Cdc7. Molecular
Cell 40:353–363. DOI: https://doi.org/10. 1016/j.molcel.2010.10.017, PMID: 21070963
Ryu T, Spatola B, Delabaere L, Bowlin K, Hopp H, Kunitake R, Karpen GH, Chiolo I. (2015).
Heterochromatic breaks move to the nuclear periphery to continue recombinational
repair. Nature Cell Biology 17:1401–1411. DOI: https://doi.org/10.1038/ncb3258, PMID:
26502056
Shachar S and Misteli T. (2017). Causes and consequences of nuclear gene positioning. Journal
of Cell Science 130: 1501–1508. DOI: https://doi.org/10.1242/jcs.199786, PMID:
28404786
Sikorski RS and Hieter P. (1989). A system of shuttle vectors and yeast host strains designed for
efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics 122:19–27. PMID:
2659436
Tanaka S, Nakato R, Katou Y, Shirahige K, Araki H. (2011). Origin association of Sld3, Sld7,
and Cdc45 proteins is a key step for determination of origin-firing timing. Current
Biology 21:2055–2063. DOI: https://doi.org/10. 1016/j.cub.2011.11.038, PMID:
22169533
Tanaka S, Araki H. (2013). Helicase activation and establishment of replication forks at
chromosomal origins of replication. Cold Spring Harbor Perspectives in Biology
5:a010371. DOI: https://doi.org/10.1101/cshperspect. a010371, PMID: 23881938
Tazumi A, Fukuura M, Nakato R, Kishimoto A, Takenaka T, Ogawa S, Song JH, Takahashi TS,
Nakagawa T, Shirahige K, Masukata H. (2012). Telomere-binding protein Taz1 controls
global replication timing through its localization near late replication origins in fission
yeast. Genes & Development 26:2050–2062. DOI: https://doi.
org/10.1101/gad.194282.112, PMID: 22987637
Viggiani CJ and Aparicio OM. (2006). New vectors for simplified construction of BrdU-
Incorporating strains of Saccharomyces cerevisiae. Yeast 23:1045–1051. DOI:
https://doi.org/10.1002/yea.1406, PMID: 17083135
Wan L, Zhang C, Shokat KM, Hollingsworth NM. (2006). Chemical inactivation of cdc7 kinase
in budding yeast results in a reversible arrest that allows efficient cell synchronization
prior to meiotic recombination. Genetics 174:1767–1774. DOI:
https://doi.org/10.1534/genetics.106.064303, PMID: 17057233
Wu PY, Nurse P. (2009). Establishing the program of origin firing during S phase in fission
Yeast. Cell 136:852– 864. DOI: https://doi.org/10.1016/j.cell.2009.01.017, PMID:
19269364
Yamazaki S, Hayano M, Masai H. (2013). Replication timing regulation of eukaryotic replicons:
Rif1 as a global regulator of replication timing. Trends in Genetics 29:449–460. DOI:
https://doi.org/10.1016/j.tig.2013.05.001, PMID: 23809990
105
Yeeles JT, Deegan TD, Janska A, Early A, Diffley JF. (2015). Regulated eukaryotic DNA
replication origin firing with purified proteins. Nature 519:431–435. DOI:
https://doi.org/10.1038/nature14285, PMID: 25739503
Zappulla DC, Sternglanz R, Leatherwood J. (2002). Control of replication timing by a
transcriptional silencer. Current Biology 12:869–875. DOI:
https://doi.org/10.1016/S0960-9822(02)00871-0, PMID: 12062049
Zhao PA, Rivera-Mulia JC, Gilbert DM. (2017). Replication domains: genome
compartmentalization into functional replication units. Advances in Experimental
Medicine and Biology 1042:229–257. DOI: https://doi.org/10.1007/978-981-10-6955-
0_11, PMID: 29357061
Zhong Y, Nellimoottil T, Peace JM, Knott SR, Villwock SK, Yee JM, Jancuska JM, Rege S,
Tecklenburg M, Sclafani RA, Tavaré S, Aparicio OM. (2013). The level of origin firing
inversely affects the rate of replication fork progression. The Journal of Cell Biology
201:373–383. DOI: https://doi.org/10.1083/jcb.201208060, PMID: 23629964
Chapter 3 References
Aparicio JG, Viggiani CJ, Gibson DG, and Aparicio OM. (2004). “The Rpd3-Sin3 histone
deacetylase regulates replication timing and enables intra-S origin control in
Saccharomyces cerevisiae”, Mol Cell Biol, 24: 4769-80.
Aparicio OM. (2013). “Location, location, location: it”s all in the timing for replication origins”,
Genes Dev, 27: 117-28.
Fang D, Lengronne A, Shi D, Forey R, Skrzypczak M, Ginalski K, Yan C, Wang X, Cao Q,
Pasero P, Lou H. (2017). Dbf4 recruitment by forkhead transcription factors defines an
upstream rate-limiting step in determining origin firing timing. Genes & Development
31:2405–2415. DOI: https://doi.org/10.1101/gad.306571.117, PMID: 2 9330352
Gillespie PJ, and Blow JJ. (2022). “DDK: The Outsourced Kinase of Chromosome
Maintenance”, Biology (Basel), 11.
Goldstein AL, Pan X, and McCusker JH. (1999). “Heterologous URA3MX cassettes for gene
replacement in Saccharomyces cerevisiae”, Yeast, 15: 507-11.
Haye-Bertolozzi JE, and Aparicio OM. (2017). “Quantitative Bromodeoxyuridine
Immunoprecipitation Analyzed by High-Throughput Sequencing (qBrdU-seq or QBU)”,
Methods Mol Biol-Genome Instability: Methods and Protocols.
He Y, Petrie MP, Zhang H, Peace JM, and Aparicio OM. (2022). “Rpd3 regulates single-copy
origins independently of the rDNA array by opposing Fkh1-mediated origin stimulation”,
Proc Natl Acad Sci U S A.
Ito H, Fukuda Y, Murata K, and Kimura A. (1983). “Transformation of intact yeast cells treated
with alkali cations”, J Bacteriol, 153: 163-8.
106
Jones DR, Prasad AA, Chan PK, and Duncker BP. (2010). “The Dbf4 motif C zinc finger
promotes DNA replication and mediates resistance to genotoxic stress”, Cell Cycle, 9:
2018-26.
Knott SR, Peace JM, Ostrow AZ, Gan Y, Rex AE, Viggiani CJ, Tavaré S, Aparicio OM. (2012).
“Forkhead transcription factors establish origin timing and long-range clustering in S.
cerevisiae”, Cell, 148: 99-111.
Longtine MS, McKenzie A 3rd, Demarini DJ, Shah NJ, Wach A, Brachat A, Philippsen P,
Pringle JR. (1998). “Additional modules for versatile and economical PCR-based gene
deletion and modification in Saccharomyces cerevisiae”, Yeast, 14: 953-61.
Looke M, Kristjuhan K, Varv S, and Kristjuhan A. (2012). “Chromatin-dependent and -
independent regulation of DNA replication origin activation in budding yeast”, EMBO
Rep, 14: 191-8.
Mantiero D, Mackenzie A, Donaldson A, Zegerman P. (2011). “Limiting replication initiation
factors execute the temporal programme of origin firing in budding yeast”, EMBO J, 30:
4805-14.
Natsume T, Muller CA, Katou Y, Retkute R, Gierlinski M, Araki H, Blow JJ, Shirahige K,
Nieduszynski CA, and Tanaka, TU. (2013). “Kinetochores coordinate pericentromeric
cohesion and early DNA replication by Cdc7-Dbf4 kinase recruitment”, Mol Cell, 50:
661-74.
Nieduszynski CA, Hiraga S, Ak P, Benham CJ, Donaldson AD. (2007). “OriDB: a DNA
replication origin database”, Nucleic Acids Res, 35: D40-6.
Ostrow AZ, Nellimoottil T, Knott SR, Fox CA, Tavaré S, Aparicio OM. (2014). Fkh1 and Fkh2
bind multiple chromosomal elements in the S. cerevisiae genome with distinct
specificities and cell cycle dynamics. PLOS ONE 9:e87647. DOI:
https://doi.org/10.1371/journal.pone.0087647, PMID: 24504085
Ostrow AZ, Viggiani CJ, Aparicio JG, Aparicio OM. (2015). ChIP-Seq to analyze the binding of
replication proteins to chromatin. Methods in Molecular Biology 1300:155–168. DOI:
https://doi.org/10.1007/978-1-4939- 2596-4_11, PMID: 25916712
Ostrow AZ, Kalhor R, Gan Y, Villwock SK, Linke C, Barberis M, Chen L, Aparicio OM.
(2017). Conserved forkhead dimerization motif controls DNA replication timing and
spatial organization of chromosomes in S. cerevisiae. PNAS 114:E2411–E2419. DOI:
https://doi.org/10.1073/pnas.1612422114, PMID: 28265091
Patel PK, Kommajosyula N, Rosebrock A, Bensimon A, Leatherwood J, Bechhoefer J, Rhind N.
(2008). “The Hsk1(Cdc7) replication kinase regulates origin efficiency”, Mol Biol Cell,
19: 5550-8.
107
Reinapae A, Jalakas K, Avvakumov N, Looke M, Kristjuhan K, and Kristjuhan A. (2017).
“Recruitment of Fkh1 to replication origins requires precisely positioned Fkh1/2 binding
sites and concurrent assembly of the pre-replicative complex”, PLoS Genet, 13:
e1006588.
Sheu YJ and Stillman B. (2010). “The Dbf4-Cdc7 kinase promotes S phase by alleviating an
inhibitory activity in Mcm4”, Nature, 463: 113-7.
Sikorski RS and Hieter P. (1989). “A system of shuttle vectors and yeast host strains designed
for efficient manipulation of DNA in Saccharomyces cerevisiae”, Genetics, 122: 19-27.
Tanaka S, Nakato R, Katou Y, Shirahige K, and Araki H. (2011). “Origin association of Sld3,
Sld7, and Cdc45 proteins is a key step for determination of origin-firing timing”, Curr
Biol, 21: 2055-63.
Viggiani CJ and Aparicio OM. (2006). “New vectors for simplified construction of BrdU-
Incorporating strains of Saccharomyces cerevisiae”, Yeast, 23: 1045-51.
Zhang H, Petrie MV, He Y, Peace JM, Chiolo IE, Aparicio OM. (2019). “Dynamic relocalization
of replication origins by Fkh1 requires execution of DDK function and Cdc45 loading at
origins”, Elife, 8.
Zhong Y, Nellimoottil T, Peace JM, Knott SR, Villwock SK, Yee JM, Jancuska JM, Rege S,
Tecklenburg M, Sclafani RA, Tavaré S, Aparicio OM. (2013). The level of origin firing
inversely affects the rate of replication fork progression. The Journal of Cell Biology
201:373–383. DOI: https://doi.org/10.1083/jcb.201208060, PMID: 23629964
Chapter 4 References
Aparicio JG, Viggiani CJ, Gibson DG, Aparicio OM. (2004). “The Rpd3-Sin3 histone
deacetylase regulates replication timing and enables intra-S origin control in
Saccharomyces cerevisiae”, Mol Cell Biol, 24: 4769-80.
Boros J, Lim FL, Darieva Z, Pic-Taylor A, Harman R, Morgan BA, Sharrocks AD. (2003).
Molecular determinants of the cell-cycle regulated Mcm1p-Fkh2p transcription factor
complex. Nucleic acids research, 31(9), 2279–2288. https://doi.org/10.1093/nar/gkg347
Darieva Z, Pic-Taylor A, Boros J, Spanos A, Geymonat M, Reece RJ, Sedgwick SG, Sharrocks
AD, Morgan BA. (2003). Cell cycle-regulated transcription through the FHA domain of
Fkh2p and the coactivator Ndd1p. Current biology : CB, 13(19), 1740–1745.
https://doi.org/10.1016/j.cub.2003.08.053
Dummer AM, Su Z, Cherney R, Choi K, Denu J, Zhao X, Fox CA. (2016). Binding of the Fkh1
Forkhead Associated Domain to a Phosphopeptide within the Mph1 DNA Helicase
Regulates Mating-Type Switching in Budding Yeast. PLoS genetics, 12(6), e1006094.
https://doi.org/10.1371/journal.pgen.1006094
108
Durocher D and Jackson SP. (2002). The FHA domain, FEBS Letters, 513, doi: 10.1016/S0014-
5793(01)03294-X
Fang D, Lengronne A, Shi D, Forey R, Skrzypczak M, Ginalski K, Yan C, Wang X, Cao Q,
Pasero P, Lou H. (2017). Dbf4 recruitment by forkhead transcription factors defines an
upstream rate-limiting step in determining origin firing timing. Genes & Development
31:2405–2415. DOI: https://doi.org/10.1101/gad.306571.117, PMID: 2 9330352
Gietz RD and Schiestl RH. (2007). High-efficiency yeast transformation using the LiAc/SS
carrier DNA/PEG method. Nature protocols, 2(1), 31–34.
https://doi.org/10.1038/nprot.2007.13
Haye-Bertolozzi JE and Aparicio OM. (2018). Quantitative bromodeoxyuridine
immunoprecipitation analyzed by High-Throughput sequencing (qBrdU-Seq or QBU).
Methods in Molecular Biology 1672:209–225. DOI: https:// doi.org/10.1007/978-1-4939-
7306-4_16, PMID: 29043627
Hoggard T, Hollatz AJ, Cherney RE, Seman MR, Fox CA. (2021). The Fkh1 Forkhead
associated domain promotes ORC binding to a subset of DNA replication origins in
budding yeast. Nucleic Acids Res.49(18):10207-10220. doi:10.1093/nar/gkab450
Hollenhorst PC, Bose ME, Mielke MR, Müller U, Fox CA. (2000). Forkhead genes in
transcriptional silencing, cell morphology and the cell cycle. Overlapping and distinct
functions for FKH1 and FKH2 in Saccharomyces cerevisiae. Genetics. 2000
Apr;154(4):1533-48.
Jones DR, Prasad AA, Chan PK, Duncker BP. (2010). “The Dbf4 motif C zinc finger promotes
DNA replication and mediates resistance to genotoxic stress”, Cell Cycle, 9: 2018-26.
Knott SR, Peace JM, Ostrow AZ, Gan Y, Rex AE, Viggiani CJ, Tavaré S, Aparicio OM. (2012).
Forkhead transcription factors establish origin timing and long-range clustering in S.
cerevisiae. Cell 148:99–111. DOI: https://doi.org/10.1016/j.cell.2011.12.012, PMID:
22265405
Koranda M, Schleiffer A, Endler L, Ammerer G. (2000). Forkhead-like transcription factors
recruit Ndd1 to the chromatin of G2/M-specific promoters. Nature. 406(6791):94-98.
doi:10.1038/35017589
Kumar A, des Etages SA, Coelho PS, Roeder GS, Snyder M. (2000). High-throughput methods
for the large-scale analysis of gene function by transposon tagging. Methods Enzymol.
328:550-574. doi:10.1016/s0076-6879(00)28418-8
Lalmansingh AS, Karmakar S, Jin Y, Nagaich AK (2012). Multiple modes of chromatin
remodeling by Forkhead box proteins. Biochimica et biophysica acta, 1819(7), 707–715.
https://doi.org/10.1016/j.bbagrm.2012.02.018
109
Li J, Coïc E, Lee K, Lee CS, Kim JA, Wu Q, Haber JE. (2012). Regulation of budding yeast
mating-type switching donor preference by the FHA domain of Fkh1. PLoS Genet.
8(4):e1002630. doi:10.1371/journal.pgen.1002630
Longtine MS, McKenzie A, Demarini DJ, Shah NG, Wach A, Brachat A, Philippsen P, Pringle
JR. (1998). Additional modules for versatile and economical PCR-based gene deletion
and modification in Saccharomyces cerevisiae. Yeast 14:953–961. DOI:
https://doi.org/10.1002/(SICI)1097-0061(199807)14:10<953::AID-YEA293>3.0.CO;2-
U, PMID: 9717241
Murakami H, Aiba H, Nakanishi, M, Murakami-Tonami Y. (2010). Regulation of yeast forkhead
transcription factors and FoxM1 by cyclin-dependent and polo-like kinases. Cell cycle
(Georgetown, Tex.), 9(16), 3233–3242. https://doi.org/10.4161/cc.9.16.12599
Nieduszynski CA, Hiraga S, Ak P, Benham CJ, Donaldson AD. (2007). “OriDB: a DNA
replication origin database”, Nucleic Acids Res, 35: D40-6.
Ostrow AZ, Nellimoottil T, Knott SR, Fox CA, Tavaré S, Aparicio OM. (2014). Fkh1 and Fkh2
bind multiple chromosomal elements in the S. cerevisiae genome with distinct
specificities and cell cycle dynamics. PLOS ONE 9:e87647. DOI:
https://doi.org/10.1371/journal.pone.0087647, PMID: 24504085
Ostrow AZ, Kalhor R, Gan Y, Villwock SK, Linke C, Barberis M, Chen L, Aparicio OM.
(2017). Conserved forkhead dimerization motif controls DNA replication timing and
spatial organization of chromosomes in S. cerevisiae. PNAS 114:E2411–E2419. DOI:
https://doi.org/10.1073/pnas.1612422114, PMID: 28265091
Peace JM, Villwock SK, Zeytounian JL, Gan Y, Aparicio OM. (2016). Quantitative BrdU
immunoprecipitation method demonstrates that Fkh1 and Fkh2 are rate-limiting
activators of replication origins that reprogram replication timing in G1 phase. Genome
Research 26:365–375. DOI: https://doi.org/10.1101/gr.196857.115, PMID: 26728715
Petrie MV, Zhang H, Arnold EM, Gan Y, Aparicio OM. (2022). “Dbf4 Zn-Finger Motif Is
Specifically Required for Stimulation of Ctf19-Activated Origins in Saccharomyces
cerevisiae”. Genes (Basel). 13(12):2202 doi:10.3390/genes13122202
Pic A, Lim FL, Ross SJ, Veal EA, Johnson AL, Sultan MR, West AG, Johnston LH, Sharrocks
AD, Morgan, B. A. (2000). The forkhead protein Fkh2 is a component of the yeast cell
cycle transcription factor SFF. The EMBO journal, 19(14), 3750–3761.
https://doi.org/10.1093/emboj/19.14.3750
Pic-Taylor A, Darieva Z, Morgan BA, Sharrocks AD. (2004). Regulation of cell cycle-specific
gene expression through cyclin-dependent kinase-mediated phosphorylation of the
forkhead transcription factor Fkh2p. Molecular and cellular biology, 24(22), 10036–
10046. https://doi.org/10.1128/MCB.24.22.10036-10046.2004
110
Pramila T, Wu W, Miles S, Noble WS, Breeden, LL. (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 Dev. 20, 2266–2278
Reynolds D, Shi BJ, McLean C, Katsis F, Kemp B, 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 & development, 17(14), 1789–1802.
https://doi.org/10.1101/gad.1074103
Voth WP, Yu Y, Takahata S, Kretschmann KL, Lieb JD, Parker RL, Milash B, Stillman DJ.
(2007). Forkhead proteins control the outcome of transcription factor binding by
antiactivation. The EMBO journal, 26(20), 4324–4334.
https://doi.org/10.1038/sj.emboj.7601859
Zhu G, Spellman PT, Volpe T, Brown PO, Botstein D, Davis TN, Futcher B. (2000). Two yeast
forkhead genes regulate the cell cycle and pseudohyphal growth. Nature. 406(6791):90-
94. doi:10.1038/35017581
Chapter 5 References
Aparicio OM. (2013). “Location, location, location: it”s all in the timing for replication origins”,
Genes Dev, 27: 117-28.
Costa A and Diffley JFX. (2022). The Initiation of Eukaryotic DNA Replication. Annual review
of biochemistry, 91, 107–131. doi: 10.1146/annurev-biochem-072321-110228
Coster G and Diffley JFX. (2017). Bidirectional eukaryotic DNA replication is established by
quasi-symmetrical helicase loading. Science 357(6348): p314-318. doi:
10.1126/science.aan0063.
Dummer AM, Su Z, Cherney R, Choi K, Denu J, Zhao X, Fox CA. (2016). Binding of the Fkh1
Forkhead Associated Domain to a Phosphopeptide within the Mph1 DNA Helicase
Regulates Mating-Type Switching in Budding Yeast. PLoS genetics, 12(6), e1006094.
https://doi.org/10.1371/journal.pgen.1006094
Eser U, Chandler-Brown D, Ay F, Straight AF, Duan Z, Noble WS, Skotheim JM. (2017). Form
and function of topologically associating genomic domains in budding yeast. PNAS
114:E3061–E3070. DOI: https://doi.org/10. 1073/pnas.1612256114, PMID: 28348222
Fang D, Lengronne A, Shi D, Forey R, Skrzypczak M, Ginalski K, Yan C, Wang X, Cao Q,
Pasero P, Lou H. (2017). Dbf4 recruitment by forkhead transcription factors defines an
upstream rate-limiting step in determining origin firing timing. Genes & Development
31:2405–2415. DOI: https://doi.org/10.1101/gad.306571.117, PMID: 2 9330352
111
Heun P, Laroche T, Raghuraman MK, Gasser SM. (2001). The positioning and dynamics of
origins of replication in the budding yeast nucleus. The Journal of Cell Biology 152:385–
400. DOI: https://doi.org/10.1083/jcb.152.2. 385, PMID: 11266454
Ho Y, Gruhler A, Heilbut A, Bader GD, Moore L, Adams SL, Millar A, Taylor P, Bennett K,
Boutilier K, Yang L, Wolting C, Donaldson I, Schandorff S, Shewnarane J, Vo M,
Taggart J, Goudreault M, Muskat B, Alfarano C, Dewar D, Lin Z, Michalickova K,
Willems AR, Sassi H, Nielsen PA, Rasmussen KJ, Andersen JR, Johansen LE, Hansen
LH, Jespersen H, Podtelejnikov A, Nielsen E, Crawford J, Poulsen V, Sɇrensen BD,
Matthiesen J, Hendrickson RC, Gleeson F, Pawson T, Moran MF, Durocher D, Mann M,
Hogue CWV, Figeys D Tyers, M. (2002). Systematic identification of protein complexes
in Saccharomyces cerevisiae by mass spectrometry. Nature, 415(6868), 180–183.
https://doi.org/10.1038/415180a
Hoggard T, Hollatz AJ, Cherney RE, Seman MR, Fox CA. (2021). The Fkh1 Forkhead
associated domain promotes ORC binding to a subset of DNA replication origins in
budding yeast. Nucleic Acids Res.49(18):10207-10220. doi:10.1093/nar/gkab450
Jones DR, Prasad AA, Chan PK, Duncker BP. (2010). “The Dbf4 motif C zinc finger promotes
DNA replication and mediates resistance to genotoxic stress”, Cell Cycle, 9: 2018-26.
Knott SR, Peace JM, Ostrow AZ, Gan Y, Rex AE, Viggiani CJ, Tavaré S, Aparicio OM. (2012).
Forkhead transcription factors establish origin timing and long-range clustering in S.
cerevisiae. Cell 148:99–111. DOI: https://doi.org/10.1016/j.cell.2011.12.012, PMID:
22265405
Natsume T, Muller CA, Katou Y, Retkute R, Gierlinski M, Araki H, Blow JJ, Shirahige K,
Nieduszynski CA, Tanaka TU. (2013). “Kinetochores coordinate pericentromeric
cohesion and early DNA replication by Cdc7-Dbf4 kinase recruitment”, Mol Cell, 50:
661-74.
Zegerman P and Diffley JF. (2010). Checkpoint-dependent inhibition of DNA replication
initiation by Sld3 and Dbf4 phosphorylation. Nature, 467(7314), 474–478.
https://doi.org/10.1038/nature09373
Appendix A References
Haber JE. (2012). Mating-type genes and MAT switching in Saccharomyces cerevisiae.
Genetics, 191(1), 33–64. https://doi.org/10.1534/genetics.111.134577
Li J, Coïc E, Lee K, Lee CS, Kim JA, Wu Q, Haber JE. (2012). Regulation of budding yeast
mating-type switching donor preference by the FHA domain of Fkh1. PLoS Genet.
8(4):e1002630. doi:10.1371/journal.pgen.1002630
112
Sun K, Coïc E, Zhou Z, Durrens P, Haber JE. (2002). Saccharomyces forkhead protein Fkh1
regulates donor preference during mating-type switching through the recombination
enhancer. Genes Dev. 16(16):2085-2096. doi:10.1101/gad.994902
Thon G, Maki T, Haber JE, Iwasaki H. (2019). Mating-type switching by homology-directed
recombinational repair: a matter of choice. Current genetics, 65(2), 351–362.
doi:10.1007/s00294-018-0900-2
Petrie MV, Zhang H, Arnold EM, Gan Y, Aparicio OM. (2022). “Dbf4 Zn-Finger Motif Is
Specifically Required for Stimulation of Ctf19-Activated Origins in Saccharomyces
cerevisiae”. Genes (Basel). 13(12):2202 doi:10.3390/genes13122202
Wu X, Haber JE. (1995). MATa donor preference in yeast mating-type switching: activation of a
large chromosomal region for recombination. Genes Dev. 9(15):1922-1932.
doi:10.1101/gad.9.15.1922
Wu X, Haber JE. (1996). A 700 bp cis-acting region controls mating-type dependent
recombination along the entire left arm of yeast chromosome III. Cell, 87(2), 277–285.
doi: 10.1016/s0092-8674(00)81345-8
Xie ZX, Mitchell LA, Liu HM, Li BZ, Liu D, Agmon N, Wu Y, Li X, Zhou X, Li B, Xiao WH,
Ding MZ, Wang Y, Yuan YJ, Boeke JD. (2018). Rapid and Efficient CRISPR/Cas9-
Based Mating-Type Switching of Saccharomyces cerevisiae. G3 (Bethesda, Md.), 8(1),
173–183. https://doi.org/10.1534/g3.117.300347
Zhang H, Petrie MV, He Y, Peace JM, Chiolo IE, Aparicio OM. (2019). “Dynamic relocalization
of replication origins by Fkh1 requires execution of DDK function and Cdc45 loading at
origins”, Elife, 8.
Appendix B References
Jones DR, Prasad AA, Chan PK, Duncker BP. (2010). “The Dbf4 motif C zinc finger promotes
DNA replication and mediates resistance to genotoxic stress”, Cell Cycle, 9: 2018-26.
Knott SR, Viggiani CJ, Tavaré S, Aparicio OM. (2009). Genome-wide replication profiles
indicate an expansive role for Rpd3L in regulating replication initiation timing or
efficiency, and reveal genomic loci of Rpd3 function in Saccharomyces cerevisiae. Genes
Dev. 23(9):1077-1090. doi:10.1101/gad.1784309
Natsume T, Muller CA, Katou Y, Retkute R, Gierlinski M, Araki H, Blow JJ, Shirahige K,
Nieduszynski CA, Tanaka TU. (2013). “Kinetochores coordinate pericentromeric
cohesion and early DNA replication by Cdc7-Dbf4 kinase recruitment”, Mol Cell, 50:
661-74.
Petrie MV, Zhang H, Arnold EM, Gan Y, Aparicio OM. (2022). “Dbf4 Zn-Finger Motif Is
Specifically Required for Stimulation of Ctf19-Activated Origins in Saccharomyces
cerevisiae”. Genes (Basel). 13(12):2202 doi:10.3390/genes13122202
113
Appendices
Appendix A: Mating type switching in Saccharomyces cerevisiae
A.1 The donor preference mechanism
Budding yeast has two distinct mating types a and alpha which can be switched between
depending on the gene being expressed at the mating loci on chromosome III. On either arm of
chromosome III there are silent mating loci in which a (HMRa) and alpha (HML ) are present
but not expressed at this location. To make this switch require HO endonuclease which cuts and
then can be repaired through homologous recombination from either of the chromosome arms.
Specifically, it was found that a 700-1275bp region known as the recombination element (RE)
drives the preferred usage of HML to switch MATa to MAT (Wu & Harber, 1996; Li et al.,
2012). When the RE is deleted this preference is switched and HMRa is favored as the repair
template leaving the cell to remain as MATa. The mechanism for this donor preference
presumably involves looping of either chromosomal arm to juxtapose the donor and recipient
loci (Fig A.1). Remarkably, in MATa cells, HML is preferentially selected as the donor 85-90%
of the time, which ensures efficient mating-type switching. In contrast, the lack of RE results in
90-95% of the time the cell remains MATa highlighting the importance of this region (Li et al.,
2012). The need for the RE in order to favor one mating type template over the other also applies
to Schizosaccharomyces pombe or fission yeast, which has two RE elements (Thon et al., 2019).
This preferred use of one chromosome as the donor for repair over the other is useful as it allows
budding and fission yeast to undergo self-diploidization. Yeast can exist as either a haploid or
diploid, with the diploid state providing a survival advantage as in nutritionally limiting
conditions they can undergo meiosis and sporulation (Haber, 2012).
114
Figure A.1 Donor preference during mating-type switching. On chromosome 3 the MAT locus is located centrally
with HML and HMRa to the left and right respectively. When RE is present, MATa cells preferentially use HML
resulting in mating type switch from a to . In contrast, HML usage reduces when RE is deleted resulting in very
little mating type switching. Adapted from Wu & Haber, 1995.
In addition to the RE it has also been discovered that Fkh1, but not Fkh2, is required for
donor preference in budding yeast (Sun et al., 2002). Our findings indicate Fkh1 as a central hub
for interactions in S. cerevisiae with specifically it playing a role in mediate long-range origin
interactions, which could also allow it to mediate a similar interaction between MATa and the
RE to specify the recombination between MATa and HML (Ostrow et al., 2017). Specifically,
a past study revealed that the FHA domain of Fkh1 is sufficient for proper donor preference. By
fusing LexA-Fkh1-FHA and replacing the RE with the LexA binding site they observed the same
preference for recombination between MATa and HML . Then by introducing an R80A
mutation in the LexA-Fkh1-FHA, donor preference was lost. Thus, their proposed model for
donor preference regulation involves the FHA domain of Fkh1 interacting with phosphorylated
115
threonine residues bound near the DSB. But, it has yet to be determined which phosphorylated
protein FHA interacts with and which protein kinase phosphorylates this target protein.
A.2 Does Dbf4 play a role in mating type switching?
When investigating the roles of Fkh1 and Dbf4 in terms of firing select origins in early S-
phase we concluded that there is an inter-dependency required for origin localization and firing
(Zhang et al., 2019; Petrie et al., 2022). With the prior knowledge that Fkh1 has a known role in
mating type switching we tested whether Dbf4 may also affect donor preference. To
quantitatively measure donor preference, we took advantage of the previously established PCR-
based donor preference assay developed by the Haber group. The assay uses a MATa strain in
which HMRa is replaced by an α sequence containing a unique BamHI restriction site (HMRα-
B). Then by using an episomal galactose-inducible HO endonuclease DSB can be induced and
the MATa will always result in a MATα cell (α or α-B, depending on whether HML or HMR
respectively was used). Donor preference can then be determined through PCR amplification of
MATα (1.5kb), subsequent digestion with BamHI (1kb and 0.5kb), and comparing relative
abundance of the different sized products (Li et al. 2012).
First, we tested the effect of dbf4∆C, although this strain was difficult to work with as it
suffers from temperature sensitivity and a severely disrupted replication pattern (Petrie et al.,
2022). Based on the donor preference assay there was no difference between WT and dbf4∆C
strains indicating that despite the effect on origin firing, a Dbf4-C term deletion does not
influence the mating-type switching mechanism (fig A.2A). A limitation to this investigation was
that we have already established that our observed Fkh1-Dbf4 interaction was not limited to the
Dbf4 C-term. So instead to directly address Dbf4 and Fkh1 interaction as well as the role of the
Fkh1 FHA domain we utilized a Dbf4-FKH fusion construct which contains only the DNA
116
binding domain of FKH (Fang et al. 2018). This construct has been shown to rescue DNA
replication patterns to WT, except for the CEN-proximal origins (data not shown). Using the
donor preference assay on fkh1∆ strains our results indicate that the overexpression of Dbf4-
FKH fusion fails to rescue mating type switching as shown by overexpressed Fkh1 (fig A.2B).
Unfortunately, this appears to rule out Dbf4 as the target of the Fkh1 FHA domain and thus its
involvement in donor preference. Again, this still leaves the question as to what the Fkh1 FHA
domain is binding to facilitate the observed donor preference. One possible target that may have
been in front of us all along is HO.
Figure A.2 PCR-based donor preference assay using Dbf4 mutants. Cultures were grown in YEP+ 2% galactose for
1 hour to induce mating type switching. The resulting MATα/MATα-B cassettes were amplified, subsequently
digested with restriction enzyme BamHI, and run on an agarose gel. (A) Asynchronous cultures of WT (XW431),
dbf4∆C (MPy63), and Δfkh1 (SVy6) containing GAL-HO (pJH132). (B) Asynchronous cultures of Δfkh1, GAL-
empty (MPy16), Δfkh1,GAL-Fkh1 (MPy17), and Δfkh1, GAL-Dbf4-FKH (MPy24) containing GAL-HO
(pLAY555).
117
A.3 The search for the interaction between Fkh1 and HO
Mating-type switching in budding yeast is initiated in G1 when HO makes a double
strand break (DSB) at the MAT locus. Once the DSB is formed at MAT, resection occurs
leaving a 3’-ended ssDNA tail that is bound by the ssDNA-binding protein complex RPA and
the Rad51 recombinase protein. Rad51 then facilitates the use of either HMLα or HMRa which
has homologous sequence to the MAT locus. The resulting strand invasion creates a displacement
loop so that the ssDNA is now paired with complementary and homologous sequence. Synthesis
begins and once this process creates a strand with homology to the other end of the DSB, the
MAT locus is replaced with the donor sequence. Depending on which sequence is used will
result in either a switch or not of mating type (Li et al., 2012). Due to the requirement of HO in
initiating mating type switching it seemed an obvious candidate to test for interaction with Fkh1
using the established Yeast-Two-Hybrid (2H) complementation assay. This system utilizes a
split transcription factor in which the proteins of interest can be fused to either a plasmid
containing an activating or binding domain (pGAD or pGBK respectively). If the proteins
interact, they will bring together the split transcription factor allowing it to bind an upstream
promoter sequence and activate the downstream reporter gene.
To begin we created fragments of HO to test in the 2H along with a fkh1-fbm (Forkhead
binding mutant) which disrupted the DBD and would prevent intrinsic Fkh1 DNA binding (fig
A.3A). Then we inserted these constructs into the pGBK and pGAD backbones respectively and
tested for expression before proceeding with the 2H (data not shown). We found that there is an
interaction between HO and Fkh1 specifically in the N-term of HO (fig A.3B). Next to test the
effect on donor preference we again used the PCR-based donor preference assay. Unfortunately,
all the HO N-term deletions and potentially phosphorylated threonine were identical to WT
118
(figA.3C). One caveat to these preliminary experiments is that the N-term deletions resulted in a
noticeable decrease in amplification of the MATα cassette. Thus far we cannot rule out that these
HO deletions are less functional than the WT and therefore we cannot conclusively determine
donor preference.
Figure A.3 HO deletions and mutation fails to alter donor preference. (A) A schematic of HO and the mutation and
deletions constructed. (B) The Yeast 2-Hybrid assay using the combinations of indicated pGAD and pGBK plasmids
and grown on selective media to allow growth when strains are harboring both plasmids and additionally expressing
the reporter gene. (C) WT (XW431) cultures containing the indicated GAL-HO constructs were grown in YEP+ 2%
galactose for 1 hour to induce mating type switching. The resulting MATα/MATα-B cassettes were amplified,
subsequently digested with restriction enzyme BamHI, and run on an agarose gel.
119
A.4 Can a different endonuclease substitute for HO in donor preference?
Although we have preliminary results that Fkh1 is interacting with the N-term of HO we
have yet to determine the mechanism of this interaction. So, in parallel we began to investigate if
HO is required for donor preference by utilizing a different endonuclease to cut at the MAT
locus. A recent study published the use of a CRISPR system in which the Cas9 is expressed on
one plasmid and the guide RNA (gRNA) is expressed on another (Xie et al., 2018). After
transforming both these plasmids step wise (first the gRNA and subsequently the Cas9) we
screened 48 individual colonies to check for the MAT / -B product (fig A.4A). Of these only
26 had clear products which we then subjected to a BamHI digest. Of these only 2 appeared to
have utilized the MAT -B (fig A.4B). Thus, even without HO there still appears to be a
preference for MATa to switch to MAT , indicating HO is not need in the mechanism that
drives this preference. These results were not necessarily surprising as the lab that constructed
these plasmids previously reported that their CRISPR system resulted in efficient switching of
mating type. However, a caveat to this system is that the Cas9 will continue cutting at the MAT
locus regardless of cell cycle phase and until the gRNA directed site is mutated or the MAT site
has switch. This then allows us to conclude that a different endonuclease other than HO can
allow for mating type switching to occur. But it is unclear if the Cas9 system is just persistent
which results in the donor preference or if there is no interaction requirement of HO.
120
Figure A.4 CRISPR can substitute for HO mating type switching. WT (XW431) cultures containing Cas9
(pNA0306) and gRNA for MATa (pXZX501) were grown on selective media plates (A) 48 individual colonies were
selected and their MAT / -B locus was amplified and run on an agarose gel. The isolates with clear products are
circled. (B) The isolates that successfully amplified the MAT / -B were digested with BamHI and then run on an
agarose gel. Those that had clear MAT -B digest products are boxed in red.
121
Appendix B: Plasmid Stability in Saccharomyces cerevisiae
B.1 CTF19 and the Dbf4 C-term reduced CEN proximal ARS activity
Yeast Centromere plasmids are a convenient way to introduce expression of a protein
without worrying about integration into the genome. They usually have a low loss rate and are
replicated as if they were another chromosome. CEN-proximal origins are among the early firing
origins and specifically CTF19 has been identified as a key player in enabling this early firing
through a mechanism involving the directed recruitment of DDK (Natsume et al., 2013). And in
fact, we have seen similar results as deletion of CTF19 resulted in reduced Cen-proximal origin
activity (Petrie et al., 2022). To further investigate the role of CTF19 in replication of CEN-
proximal origins we analyzed ARS activity through retention of an episomal plasmid. Plasmid
loss assays were performed by transforming an episomal plasmid into desired strain backgrounds
and then maintaining selection through drop-out media. Then a known amount of colony forming
units (CFU) were allowed to grow non-selectively overnight (~16hrs). Following this the
cultures were grown on non-selective media to determine the number of generations and
selectively to determine the percentage of cells still harboring the plasmid.
When CTF19 was deleted there was a nearly 3-fold increase in the rate of plasmid loss
per generation compared to WT (fig B.1). As the role of CTF19 appears to be linked to
maintaining early firing of CEN-proximal origins through DDK we also tested a dbf4 C-term
ZN-finger mutant (dbf4-ZN*). It was previously seen that disruption of the C-term Zn-finger in
Dbf4 reduced MCM2 interaction and S-phase entry (Jones et al., 2010). In addition to this our
previous result indicated that there are global changes in origin firing and specifically CEN-
proximal origins are less active in dbf4 Zn-finger mutants (Petrie et al., 2022). As CTF19 is
linked to DDK recruitment to origins we likewise tested the effect of the dbf4 C-term ZN-finger
122
mutant on CEN-proximal ARS activity. And similar to the CTF19 deletion, the dbf4-Zn*
showed a 4.5-fold increase in plasmid loss per generation as compared to WT (fig B.1). Overall
this suggests that the mechanism allowing CEN-proximal origins to be early firing is dependent
on both CTF19 recruitment of DDK as well as a functional DBF4 C-term.
Further planned investigation is to exchange the current CEN-proximal origins on the
episomal plasmid, which is crucial for its replication and thus retention in daughter cells, for an
alternative origin. In particular we plan to insert a Fkh-activated origin as defined previously and
examine if this will rescue plasmid stability to WT levels (Knott et al., 2012). Our hypothesis is
that Fkh1 will preferentially fire a Fkh-activated origin early in S-phase and will increase the
number of daughter cells that maintain the episomal plasmid. This would argue that any early
origin would be sufficient to maintain plasmid stability not just the designated CEN-proximal
origins.
Figure B.1 Plasmid loss rate per generation of the indicated strains. Cultures of dbf4-Zn*(MPy76), WT(CVy63), and
ctf19∆(HYy210) containing the plasmid pRS412 were grown non-selectively overnight and then plated on non-
selective and selective media plates. The resulting plasmid loss was determined with error bars indicating standard
deviation.
Abstract (if available)
Abstract
Organized and faithful DNA replication is required through any given mitotic cell division. All life requires organized and faithful DNA replication each time it proceeds through a given cell division. In eukaryotic genomes replication initiates at multiple sites called origins, dependent on factors such as DNA sequence, chromatin environment, and sub-nuclear localization. In G1-phase more origins than can be fired are licensed, so that as cells enter S-phase only a subset drives the replication of entire genomes. This duplication occurs in a temporal manner due to limiting factors allowing only select origins to activate and disruption of this process can result in genome instability. In Saccharomyces cerevisiae, it has been observed that Forkhead transcription factors bind and enable early firing of select origins named Fkh-activated origins. Our studies have further elucidated the role of Fkh1 in driving sub-nuclear localization of these early firing origins. The relationship between location and origin activation are linked with Fkh1 sequestering limiting factors in G1-phase. In particular, we investigated recruitment mechanisms of Dbf4, a subunit of DDK, and furthered the understanding of how it is directed to select origins. Although loss or disruption of the C-term of Dbf4 greatly reduces CEN-proximal firing it does not fully eliminate early firing of Fkh-activated origins nor Fkh1 interaction. We have found that the Fkh1-FHA plays a crucial role in facilitating early firing of Fkh-activated origins and interacts with a yet unknown region of Dbf4. Overall, our findings contribute additional details to established mechanisms of origin firing, highlights the relationship between sub-nuclear localization and sequestering of limiting factors to origin activation in early S-phase.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Subnuclear localization of replication origins is controlled by Fkh1-dependent recruitment of DDK to origins in S. cerevisiae
PDF
Forkhead transcription factors control genome wide dynamics of the S. cerevisiae replication timing program
PDF
The function of Rpd3 in balancing the replicaton initiation of different genomic regions
PDF
Forkhead transcription factors regulate replication origin firing through dimerization and cell cycle-dependent chromatin binding in S. cerevisiae
PDF
Measuring, modeling and identifying factors that influence eukaryotic DNA replication
PDF
The intra-S phase checkpoint and its effect on replication fork dynamics in saccharomyces cerevisiae
PDF
New tools for whole-genome analysis of DNA replication timing and fork elongation in saccharomyces cerevisiae
PDF
Computational analysis of DNA replication timing and fork dynamics in S. cerevisiae
PDF
Exploring three-dimensional organization of the genome by mapping chromatin contacts and population modeling
PDF
Biochemical mechanism of TopBP1 recruitment to sites of DNA damage
PDF
The role of Cdc7 in replication fork progression in response to DNA damage
PDF
Maintenance of genome stability at fragile sites in Schizosaccharomyces pombe
PDF
Zygotic genome activation triggers chromosome damage and checkpoint signaling in Caenorhabditis elegans primordial germ cells
PDF
C. elegans topoisomerase II regulates chromatin architecture and DNA damage for germline genome activation
PDF
Dps contributes to typical growth, survival, and genome organization in E. coli
PDF
Cisplatin activates mitochondrial oxphos leading to acute treatment resistance in bladder cancer
PDF
Decoding the embryo: on spatial and genomic tools to characterize gene regulatory networks in development
PDF
Molecular mechanism of the recruitment of SWI/SNF chromatin remodeling complex and histone acetyltransferase to estrogen-responsive promoters
PDF
Detecting joint interactions between sets of variables in the context of studies with a dichotomous phenotype, with applications to asthma susceptibility involving epigenetics and epistasis
PDF
Temporal and spatial characterization of cisplatin treatment and emerging acute resistance in bladder cancer cells
Asset Metadata
Creator
Petrie, Meghan Victoria
(author)
Core Title
Distinct mechanisms of DDK recruitment to Fkh-activated and CEN-proximal origins control replication timing program in S. cerevisiae
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Molecular Biology
Degree Conferral Date
2023-08
Publication Date
06/07/2023
Defense Date
05/02/2023
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
chromosome domains,Ctf19,Dbf4-dependent-kinase,DNA-binding protein,Fkh1,genome,OAI-PMH Harvest,replication origins
Format
theses
(aat)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Aparicio, Oscar (
committee chair
), Calabrese, Peter (
committee member
), Chiolo, Irene (
committee member
), Michael, Matthew (
committee member
), Vermulst, Marc (
committee member
)
Creator Email
mpetrie@usc.edu,mvpetrie1@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC113169222
Unique identifier
UC113169222
Identifier
etd-PetrieMegh-11931.pdf (filename)
Legacy Identifier
etd-PetrieMegh-11931
Document Type
Dissertation
Format
theses (aat)
Rights
Petrie, Meghan Victoria
Internet Media Type
application/pdf
Type
texts
Source
20230609-usctheses-batch-1053
(batch),
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Repository Email
cisadmin@lib.usc.edu
Tags
chromosome domains
Ctf19
Dbf4-dependent-kinase
DNA-binding protein
Fkh1
genome
replication origins