C. elegans topoisomerase II regulates chromatin architecture and DNA damage for germline genome activation

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content

C. elegans topoisomerase II regulates chromatin architecture and DNA damage

for germline genome activation

by

Matthew Martin Wong

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Molecular Biology)

August 2018

ii
Table of Contents

LIST OF FIGURES v

LIST OF TABLES vi

ACKNOWLEDGEMENTS vii

ABSTRACT x

INTRODUCTION 1

A PERSPECTIVE 1

THE GERMLINE LINEAGE IN C. ELEGANS 2

TRANSCRIPTIONAL QUIESCENCE IN THE GERMLINE 3
FROM AN EARLY EMBRYO TO A STARVED L1

Z2/Z3 ZYGOTIC GENOME ACTIVATION AND CELL CYCLE REENTRY 7

CHECKPOINT SIGNALING AND DNA DAMAGE DURING ZGA 8

KNOWN ROLES AND MECHANISMS OF TOPOISOMERASE II 10

DNA BREAKS AND CHROMATIN DECOMPACTION AS MODES OF 11
GENE ACTIVATION IN OTHER SYSTEMS

RUVB-LIKE ATPase DECOMPACTION FACTORS 13

CHAPTER I: CHROMATIN DECOMPACTION VIA DNA BREAKS AND RUVB
PROTEINS ACTIVATES THE GERMLINE GENOME 15

INTRODUCTION 16

RESULTS 18

IDENTIFICATION OF GENES UPREGULATED DURING GERMLINE ZGA 18

TOP-2 IS REQUIRED FOR GERMLINE ZGA 20

RANDOMLY PLACED DNA BREAKS BYPASS THE REQUIREMENT 21
FOR TOP-2 IN ZGA

TOP-2 AND RUVB PROTEINS PROMOTE CHROMATIN DECOMPACTION 24

iii
CHROMATIN DECOMPACTION IS REQUIRED FOR GERMLINE ZGA 27

TOP-2 RECRUITS RUVB PROTEINS TO TRANSCRIPTIONALLY 31
ACTIVE CHROMOSOMES DURING GERMLINE ZGA

THE ROLE OF DNA BREAKS DURING GERMLINE ZGA IS TO 34
ACTIVATE RUVB PROTEINS FOR CHROMATIN DECOMPACTION

DISCUSSION 39

SIGNAL- AND DNA BREAK-MEDIATED GENOME DECOMPACTION 39
ALLOWS ZGA AND CELL CYCLE RE-ENTRY

SITE SPECIFIC BREAKS AT SPACER REGIONS MAY ALLOW 40
DECOMPACTION AND ACTIVATION

TRANSCRIPTION AND DNA DAMAGE: THE CHICKEN AND THE EGG 41

AN ‘EXPANDING’ ROLE FOR RUVB PROTEINS 42

TOP-2 DNA BREAKS FOR GENOME ACTIVATION: 43
EFFICIENT, EFFECTIVE, AND EVOLVED

MATERIALS AND METHODS 45

CHAPTER II: THE ROLE OF TOP-2 AND CONDENSIN II IN Z2/Z3 DURING
EMBRYOGENESIS 61

INTRODUCTION 62

RESULTS 67

EARLY TOP-2 INACTIVATION DURING EMBRYOGENESIS CAUSES 67
HYPO-CONDENSED CHROMATIN IN THE EMBRYO

TOP-2 SUPPRESSES PREMATURE GENE EXPRESSION IN THE 70
EMBRYO THROUGH CHROMATIN COMPACTION

TOP-2 MEDIATED CHROMATIN COMPACTION SUPPRESSES 74
EXPRESSION OF GERMLINE GENES

DISCUSSION 82

A NOVEL MECHANISM OF TRANSCRIPTIONAL REPRESSION 82
BY TOP-2/CONDENSIN II

4
THE POTENTIAL INTERACTIONS OF TOP-2/CONDENSIN II AND THE 83
GERMLINE GENOME DURING EMBRYOGENESIS

HETEROCHROMATIN CHARACTERIZES THE ENTIRE GERMLINE 85
GENOME IN THE LATE EMBRYO

MATERIALS AND METHODS 86

CHAPTER III: THE ROLE OF THE G2/M DNA DAMAGE CHECKPOINT
IN THE CELL CYCLE EXIT OF EMBRYONIC Z2/Z3 95

INTRODUCTION 96

RESULTS 103

DNA DAMAGE CHECKPOINT FAILS TO STAY ACTIVATED IN 103
Z2/Z3 IN THE LATE EMBRYO

CDK-1 INACTIVATION DOES NOT OCCUR IN EMBRYONIC Z2/Z3 105

CYB-3 IS MAINTAINED FOLLOWING P4 DIVISION BUT FAILS TO 107
LOCALIZE ON Z2/Z3 CENTROSOMES

Z2/Z3 CDK-1 INHIBITORY PHOSPHORYLATEION IS ACTIVATED 111
WITH FEEDING IN L1S

DISCUSSION 114

THE G2/M DNA DAMAGE CHECKPOINT DOES NOT MEDIATE 114
CELL CYCLE EXIT IN EMBRYONIC Z2/Z3

CELL CYCLE ARREST BY FAILURE TO ACTIVATE CYCLIN B-CDK-1 115

MATERIALS AND METHODS 118

REFERENCES 123

INTRODUCTION REFERENCES 123

CHAPTER I REFERENCES 127

CHAPTER II REFERENCES 130

CHAPTER III REFERENCES 135
v
LIST OF FIGURES

CHAPTER I

Figure 1.1. Genes encoding germ cell specific P-granule components are 19
activated during ZGA in the Z2/Z3 PGCs.

Figure 1.2. TOP-2 mediated DNA breaks trigger germline ZGA. 23

Figure 1.3. Factor requirements for Z2/Z3 chromatin decompaction 26
during ZGA.

Figure 1.4. Chromatin decompaction is required for ZGA. 28

Figure 1.5. TOP-2 is upstream of RUVB proteins in the ZGA pathway. 32

Figure 1.6. DNA breaks activate RUVB proteins for chromatin decompaction. 36

Figure S1.1. MabH5 reactivity is lost after taf-10 RNAi in feeding L1s and 55
rare instances of RAD-51 and MabH5 foci overlap during ZGA.

Figure S1.2. Method for quantification of chromatin decompaction. 56

Figure S1.3. Early inactivation of TOP-2 produces hypo-compacted 57
chromatin in starved L1s and allows ZGA.

CHAPTER II

Figure 2.1. Chromatin structures of Z2/Z3 during embryogenesis. 68

Figure 2.2. RNAPII elongation in top-2(it7ts) inactivated embryonic germ cells. 71

Figure 2.3. RNAPII elongation in capg-2(RNAi) embryonic germ cells. 73

Figure 2.4. PIE-1 in top-2(RNAi) and capg-2(RNAi) embryonic germ cells. 75

Figure 2.5. Hybridized chain reaction (HCR) in situ of ppw-2. 78

Figure 2.6. H3K9me3 staining in embryonic P lineage cells. 80

Figure S2.1. Validation of Ser2-P antibody (ab5095) for immunofluorescence 92
in C. elegans.

Figure S2.2. Gene expression in top-2(RNAi) depleted embryonic germ cells. 93

vi
Figure S2.3. ppw-2 expression using chromogenic in situ hybridization (CISH). 94

CHAPTER III

Figure 3.1. Detection of activated phospho-CHK-1 in PGCs during 104
early and late embryogenesis.

Figure 3.2. Detection of phosphor-CDK-1 in PGCs during early and late 106
embryogenesis.

Figure 3.3. Detection of CYB-3 in PGCs during early and late embryogenesis. 109

Figure 3.4. Centrosomal co-localization of CYB-3 and γ–tubulin in P1-P4 110
and Z2/Z3.

Figure 3.5. P-CDK-1 in Z2/Z3 in L1s during ZGA. 113

Figure S3.1. mRNA in situ hybridization (ISH) of cyb-3 during 122
embryogenesis.

LIST OF TABLES

Table 1.1. Summary of RNA-FISH analysis of ZGA gene expression. 59

Table 1.2. Summary of RAD-51 and MabH5 foci overlap. 60

vii
ACKNOWLEDGEMENTS

My success as a graduate student might best be summed up by Henry Ford’s
famous line, “I am not the smartest, but I surround myself by competent people,”.
Pursuing a doctorate in Molecular Biology was, at times, a grueling experience, but
the many individuals who offered their extraordinary friendship and support during
graduate school ultimately helped me succeed. I would first like to thank my
wonderful PhD advisor, Dr. W. Matthew Michael. Matt’s kindness and contagious
enthusiasm for science motivated me to work hard each day, and watching Matt
taught me how to ask insightful questions, be meticulous, interpret data, and even
communicate better. Perhaps most of all, however, I thank Matt for his vision and
support throughout my entire project; While Matt was incredibly sharp and could
logically approach almost any technical or scientific obstacle, it was his constant show
of confidence and optimism in both the science and my abilities that impacted me the
most. This energy he invested in me always pushed me further along whenever I felt
like my research had hit a wall, and without his guidance and mentorship, I would not
have been successful completing my doctorate. I could not have asked for a better
advisor and I am forever grateful.
I would also like to thank my PhD dissertation committee members Dr. Oscar
Aparicio, Dr. Carolyn Phillips, and Dr. Judd Rice for their support through the years. I
am incredibly grateful for the valuable feedback they provided me through their
expertise and the time they spent to facilitate my degree completion. Additionally, I
extend an acknowledgement to the labs of Dr. Irene Chiolo and Dr. Susan Forsburg,
viii
and again, Dr. Aparicio, for the many joint lab meetings surrounding our shared
interest in DNA damage. Next, I cannot express enough gratitude to Dr. Melina Butuči
for showing me the ropes of the Michael Lab, passing me along a fantastic project, and
continuing to offer insight long after she left. Melina was one of the most hard-
working and friendliest scientists I ever met, and she set a fantastic example for me
when I started. I also want to thank my lab members who helped work on my projects
like Mezmur Belew, Amanda Kwieraga, and again, Melina. Without their hard work
and effort, I certainly would not have been so fortunate in my research. I also
acknowledge all the other former members of the Michael Lab who I had the pleasure
of knowing and working with: Dr. Hovik Gasparyan, Dr. Julyana Acevedo, Quinn
Cowan, Ashley Kim, Holly Senebandith, and Dr. Frances Tran. Thank you also to my
closest friends, fellow graduate students, and post-docs friends in MCB and PIBBS for
sharing this journey with me and reminding me I wasn’t going it alone: TaeHyun Ryu,
Dr. Chris Caridi, Dr. Wilber Escorcia, Aaron Wolfe, Hans Dalton, and Jessica DeWitt.
Outside of the MCB program, I would like to thank Dr. Cory Nelson and Roger
Anderson at The USC Writing Center for giving me a part-time opportunity as a
Writing Consultant, and the team at Biotech Connection LA for a place as their Chapter
Correspondent. A special acknowledgement finally goes to my best friends Dr. Brian
Avanzino and Dr. Jeffrey West for the company, encouragement, and many laughs
during graduate school.
Thank you to my parents, Martin and Helen Wong, for raising and supporting
me in everything through the years, letting me choose my own path, and teaching me
hard work, discipline, and wisdom for 31 years. Thank you also to my extended family
ix
in Los Angeles who made my relocation to Southern California in 2012 a lot easier,
and especially my grandmother, Mary Wong, for her support in my personal life
during these years. Finally, I would like to thank my beautiful and loving wife, Vicky,
who was truly my greatest discovery during graduate school. I thank her for her
continued patience while I finished graduate school, always making me laugh, and
helping me when I fell short. I would not be where I am today without her
unconditional love and support and with her, I can’t wait for what the future holds.

x
ABSTRACT
The protection and transmission of an organism’s genome is arguably the
most important task of its lifetime. Central to this mission are faithful DNA damage
repair, gene expression, and proliferation of germline sex cells. Such demanding
activities have made germ cells excellent models over the years for identifying some
of the most robust molecular mechanisms in cellular biology. Still, we have only begun
to uncover the extraordinary lengths to which germ cells undergo to develop.
In C. elegans, the primordial germ cells Z2 and Z3 are born during early
embryogenesis and then held in a transcriptionally quiescent state where the genome
is highly compacted. When hatched L1s feed, the germline genome decompacts and
RNAPII is abruptly and globally activated. A previously documented yet unexplained
feature of germline genome activation in the worm is the appearance of numerous
DNA breaks coincident with RNAPII transcription. These studies highlighted that ZGA
generates genomic instability prior to cell cycle reentry, but critical questions
emerged, such as what is the enzyme responsible for the damage, and why would
Z2/Z3 risk destroying its genome during ZGA.
Here, we have found that those questions are answered by the DNA metabolic
enzyme topoisomerase II (TOP-2). Specifically, we show that DNA breaks are induced
by TOP-2, that they function to recruit the RUVB complex to chromosomes so that
RUVB can decompact the chromatin, and that DNA break- and RUVB-mediated
decompaction is required for zygotic genome activation. Following up on this work,
we have also started to show that embryonic genome compaction requires TOP-2 and
the condensin complex. Finally, we have also asked what cell cycle factors mediate
xi
the cell cycle arrest during embryogenesis. These works not only underscore the
unique nature of PGC development, but also highlight the importance of global
chromatin decompaction to the rapid induction of gene expression, and show that
one way cells achieve global decompaction is through programmed DNA breaks.

1
INTRODUCTION
A Perspective
In the mid-seventeenth century, Robert Hooke shaved cork from a plant stem
and excitedly described the first observed cell as perforated and porous (Hooke,
1665). He used one of the earliest microscopes about which he wrote, in Micrographia,
“promote the use of mechanical helps for the Senses, both in the surveying the already
visible World, and for the discovery of many others hitherto unknown” (Hooke, 1665).
To say that our knowledge of cells (and microscopes, for that matter) has grown
significantly since that initial discovery would be a gross understatement. Rather,
cells today are not only well-accepted as a basic unit of life, but also provide the
context for almost every discovery in biological science and medicine.
A watershed moment that pushed our understanding of cells began when
Sydney Brenner established Caenorhabditis elegans, the free-living nematode, as a
model organism in 1974, for their rapid life cycle, large brood size, and fixed cell
number (Brenner, 1974; reviewed in Ankeny, 2010). The ease with which C. elegans
permitted genetic analysis helped John Sulston in 1983 trace the complete genealogy
of every C. elegans cell from embryogenesis to adulthood (reviewed in Giurumescu
and Chisolm, 2011). It also helped Andrew Fire and Craig Mello inject C. elegans with
dsRNA, inadvertently discovering the application of RNAi (reviewed in Kurreck,
2009). And it helped several other scientists identify major mechanisms involved in
programmed cell death (Adams, 2003; Horvitz 2003; Danial and Korsmeyer 2004).
Germline studies is another area where C. elegans has been useful, particularly
in developmental biology. The idea that primordial germ cells (PGCs) carry the
2
immortal genome make PGCs compelling systems and over the years, research with
PGCs has shed light on epigenetic regulation, signal transduction, and gene
expression (Kimble and Hirsh, 1979; Fukuyama et al., 2006; Kelly et al., 2014; Ishi et
al., 2016). My own doctoral research, using advanced molecular imaging to study
PGCs, intended to answer how germ cells might develop and divide, but in doing so,
also revealed an unexpected yet incredible molecular pathway that only nature could
think to design. Thus, while we as scientists have come a long way since the first
observation of the cell, I have essentially come to the same basic conclusions as
Robert Hooke nearly 400 years later, and that is simply that microscopes are
awesome and that cells continue to surprise and amaze us.

The Germline Lineage in C. elegans
Germline specification is established during early embryonic development
and, to our advantage, has been well-characterized in C. elegans (Wang et al., 2013;
Strome, 2005). When sperm and egg meet during fertilization, the fusion of their
genomes results in a zygote known as P0. Within a few minutes, this one-cell embryo
cleaves asymmetrically into a larger somatic blastomere called AB and a smaller
germline blastomere called P1. Absent of gap phases, P1 proceeds through replication
and division to generate EMS, a somatic precursor for the endomesoderm and P2, the
next germline blastomere. These cleavages continue; In the embryonic 4-cell stage, P2
divides into C and P3 and at the 24-cell stage, P3 divides into D and P4. As described,
the germline is specified through the P-lineage and established in P4, as every
following cell division from P4 exclusively produces germ cells (Sulston et al., 1983).
3
At about the 100-cell stage, P4 divides into two primordial germ cells (PGCs)
called Z2 and Z3 (Z2/Z3). These PGCs finish one round of S-phase and then arrest
their cell cycle during late G2, after their chromosomes condense (Fukuyama et al.,
2006). Unlike P-cells that cleave rapidly with factors provided maternally,
continuation of the cell division in Z2/Z3 requires nutrients to proliferate (Butuci et
al., 2015a,b). In other words, while somatic cell division ensues, Z2/Z3 remain
arrested through the end of embryogenesis when larvae hatch. When worms in their
first larval stage (L1) begin feeding, Z2/Z3 finally re-enter the cell cycle (Butuci et al.,
2015a,b) PGCs continuously divide through L4 to produce about 2000 germ cells by
adulthood (Kimble and Hirsh, 1979). As young adults, germ cells enter the meiotic
pathway and complete maturation for the next generation (Crittenden et al., 2006;
Kimble and White, 1981).

Transcriptional Quiescence in the Germline from an Early Embryo to Starved
L1
P-cells (P1-P4) that eventually differentiate into germline PGCs are
programmed with unique properties enabling their specification. A major subcellular
structure exclusive to all germline cells throughout development are P-granules. P-
granules are dynamic organelles adjacent to the outside of the nuclear pore that
contain a heterogeneous mix of RNAs and proteins (Updike and Strome, 2010). After
the first blastomere cleaveage of P0, these maternally inherited factors segregate
asymmetrically to the posterior of the embryo to aid germ cell specification ((Strome
and Wood, 1982; Strome and Wood, 1983). Here, P-granules act as storage and
4
surveillance mechanisms in the cytoplasm for any maternal mRNAs that might
otherwise be inappropriately translated. (Sheth et al., 2010). When the germ cells
eventually perform transcription, P-granules regulate post-transcriptional
processing during the nuclear exit of nascent RNAs. Interestingly, ectopic expression
of P-granules in somatic cells does not induce germ specification, suggesting that P-
granules contribute to germ cell fate but not exclusively (reviewed in Voronina, 2013).
Nevertheless, the molecular nature of P-granules is consistent with the goal of the
germ cell to protect its identity in the early developmental stages of life.
P-granules contain over 40 proteins that either prevent premature
transcription or regulate translation of maternal transcripts in the embryonic
germline, and those with very central roles in this effort are PGL-1, OMA-1 and OMA-
2, and PIE-1 (Updike et al., 2011; Kawasaki et al., 2004; Detwiler et al., 2001; Shimada,
et al., 2006; Mello et al., 1996). PGL-1 is widely enriched in the cytoplasm at every
stage of development, and binds to RNA while maintaining the integrity of P-granules
(Kawasaki et al., 2004). OMA-1 and OMA-2, on the other hand, are multifunctional
CCCH-zinc-finger proteins produced exclusively in oocytes. They carry past
fertilization and are found in P0 and P1 but quickly degrade in P2 after the first cell
division. In addition to being required for oocyte maturation, OMA-1 and OMA-2, bind
to maternal mRNAs to facilitate germline specification (Detwiler et al., 2001) and
importantly, competitively bind to TAF-4 to downregulate the transcription complex
TFIID (Guven-Ozkan et al, 2008).
After OMA-1/2 degrade, the germline utilizes PIE-1, another maternally-
provided defense mechanism against premature transcription. PIE-1 is also a zinc-
5
finger protein that actively represses transcription. However, unlike OMA-1/2 that
acts on transcription factors outside of the nucleus, PIE-1 directly inhibits RNAPII by
binding to the Cyclin T1 subunit of the positive transcription elongation factor b (P-
TEFb) (Batchelder, et al., 1999). In doing so, PIE-1 competes with P-TEFb, which is
required to phosphorylate the C-terminal domain (CTD) of RNAPII for transcriptional
elongation (Zhang et al., 2003). The importance of PIE-1 in P2-P4 during
embryogenesis has been significantly underscored. For example, in pie-1 mutants,
transcription is prematurely activated in P2 and this inappropriately activates
somatic gene expression (Mello et al., 1996). Since other reports suggest that H3K4me,
a marker of transcriptionally active chromatin, is present throughout P1-P4 (Kelly,
2014), PIE-1 has a critical role to repress transcription where the chromatin state is
otherwise permissive to gene expression (Schaner et al., 2003). In summary, PIE-1
along with several other embryonic proteins coordinate the transcriptional
repression required to maintain the germline.
In addition to factors that promote transcriptional repression in germ cells
during embryogenesis, chromatin-based mechanisms via changes in histone
modifications also regulate the timely activation of germline genome. When Z2/Z3
are born, PIE-1 degrades and Z2/Z3 appear to briefly activate transcription, by
indication of the phosphor-Ser2 epitope (Seydoux and Dunn, 1997). Coincident with
the loss of PIE-1 in Z2/Z3, however, is the disappearance of H3K4me and H4K8Ac,
modifications associated with open chromatin. This sudden shift toward chromatin
repression is understood to maintain germline silencing for the remainder of
embryogenesis and L1 starvation, as H3K4me and H4K8Ac are not seen again until
6
L1s feed and activate the germline genome (Schaner et al., 2003). Germline genes are
also immediately silenced during embryogenesis with H3K27me, a repressive histone
modification (Strome, 2005). Ultimately, each of these histone modifications point to
a significant paradigm in germline gene expression that we will explore, and that is
the influence of chromatin relaxation on transcriptional activation.
Surrounding the unique transcriptional program during the birth of Z2/Z3 is
an equally compelling cell division cycle. In contrast to the early embryonic cell cycles
that lack detectable gap phases G1 and G2 (Edgar and McGhee, 1988), the Z2/Z3 cell
cycles occupy a greater temporal space. After their birth, Z2/Z3 replicate their DNA
and centrosomes in S phase, as indicated respectively by propidium-iodide staining
that suggests 4N DNA content in Z2/Z3 and anti-γ-tubulin staining that reveals two
centrosomes (Fukuyama et al., 2006). After replication, Z2/Z3 chromosomes
condense in late G2 and remain condensed starting when the embryo is comprised of
about 550-cells through L1 starvation (Fukuyama et al., 2006). Since Z2/Z3
centrosomes are not yet separated to the poles of the nucleus, cell cycle arrest is
considered to occur at early mitotic prophase (Fukuyama et al., 2006). While cell cycle
exit is not exclusive to the germ cells, the late G2/early prophase exit is specific to
Z2/Z3; In response to starvation that pauses L1 development, somatic cells arrest in
G1. Cell cycle arrest of both somatic cells and PGCs, however, is ultimately reversible
with feeding, however (Schaner et al., 2003; Butuci et al., 2015a,b, Seidel and Kimble
2015), and mechanisms by which cell cycle re-entry in Z2/Z3 occur were heretofore
still under investigation.

7
Z2/Z3 Zygotic Genome Activation and Cell Cycle Reentry
The global activation of RNAPII during early development is a critical event
observed in many model organisms that often marks the maternal-to-zygotic
transition. Here, maternally provided proteins and mRNAs degrade and cells must
rely exclusively on zygotic transcription for the first time (Tadros and Lipshitz, 2009).
This sudden onset of RNAPII activity is known as zygotic genome activation (ZGA)
and in C. elegans, Z2/Z3 remarkably display this feature when starved L1s begin
feeding (Butuci et al., 2015a,b).
Previous projects in our lab have sought to identify the timing and mechanism
of Z2/Z3 ZGA and cell-cycle reentry in response to nutrients. Since a complete,
proliferating germline requires cell-cycle reentry from arrested Z2/Z3, we first asked
how long it took for Z2/Z3 to divide after feeding. To count the division of PGCs as a
function of time, we utilized a C. elegans strain, SS747, that expressed GFP-tagged
PGL-1 so that germ cells could be quantified via immunofluorescence on a confocal
laser microscope. When starved L1s were synchronized and fed E. coli OP50 for up to
7 hours, we found that the first division of Z2 or Z3 occurred 4 to 5 hours after feeding
on average and that the emergence from cell cycle arrest was asynchronous (Butuci
et al., 2015 a,b).
In Drosophila, the shift to zygotic expression during the midblastula transition
is characterized by the degradation of maternal RNAs (Lee et al., 2014), and thus we
were also interested in the timing of zygotic transcription in Z2/Z3 following
maternal RNA degradation via PIE-1 loss during embryogenesis. To measure
transcription activity using immunofluorescence, we utilized a monoclonal antibody,
8
H5, that recognizes phospho-serine 2 on the tail of the carboxy-terminal domain of
RNA polymerase II (Patturajan et al., 1998). This phosphorylated residue specifically
corresponds to the elongating form of active RNAPII, and thus the H5 assay has been
a primary readout for transcription in both our previous and current studies.
Analyzing the appearance of H5 with respect to feeding revealed that Z2/Z3 activate
transcription as quickly as 1-hour after feeding and that the transcriptional load
increases with time (Butuci et al., 2015a,b). This demonstrated ZGA in Z2/Z3 is
nutrient-dependent.

Checkpoint Signaling and DNA Damage during ZGA
In addition to understanding when transcription was activated upon feeding,
we also sought what mechanisms controlled the timing of Z2/Z3 cell-cycle reentry.
An interesting observation that chk-1 (RNAi) embryos give rise to a terminal adult
lacking a germline (Brauchle et al., 2003), led us to investigate the germline
requirement of the checkpoint kinase, CHK-1, during early larval development. First,
chk-1 depleted L1s completed mitotic division earlier than controls. Furthermore,
immunofluorescent staining revealed that CHK-1 was phosphorylated and thus active,
up to 3-4 hours after feeding. From these efforts, we learned an active checkpoint
signal CHK-1 mediates the timing of Z2/Z3 division prior during ZGA.
We next assessed whether DNA damage is the source of CHK-1 activation in
Z2/Z3 during feeding. In response to DNA double strand breaks (DSB), RAD-51
localizes to the ends of RPA-coated single strands to facilitate homologous
recombination and repair (Haaf et al., 1995). In C. elegans germ cells, this exchange
9
takes place during meiotic recombination (Mets and Meyer, 2009). Thus, to assay for
DNA damage, we utilized an anti-RAD-51 antibody for immunofluorescence.
Surprisingly, signs of DNA damage repair were detected in Z2/Z3 after feeding. These
DNA DSBs were confirmed as new sites of damage generated by feeding. More
specifically, the damage foci were also observed only in transcriptionally active
autosomes and were excluded from the transcriptionally silent X chromosome. We
also observed that H5 RNAPII signals appeared earlier than RAD-51 foci, and when
transcription was chemically inhibited, RAD-51 foci failed to appear. These results
suggested that ZGA surprisingly triggers DNA damage during Z2/Z3 cell-cycle re-
entry (Butuci, et al., 2015 a,b).
To identify an enzyme responsible for DNA damage during ZGA, we pursued
the possibility that TOP-2 induced DNA DSBs because in meiosis, double-strand
breaks are made by the type II topoisomerase, SPO-11, and then repaired by
homologous recombination (Haffner et al., 2011). Analysis of PGC division and RAD-
51 foci in top-2 (RNAi)-treated larvae revealed that TOP-2 is required for DNA
damage and that its depletion accelerates Z2/Z3 division upon feeding (Butuci et al.,
2015). While these results showed a role for TOP-2 during ZGA, several open
questions about its precise nature and function remained. To start, while TOP-2 was
required for DNA DSBs, it was still unknown whether its enzymatic activity was
directly responsible for the induction of breaks in Z2/Z3. Additionally, we recognized
that endogenously breaking the genome in moments prior to Z2/Z3 cell division is
seemingly counterintuitive and threatening to the survival of the germline.
Considering this, we were compelled to determine why DNA DSBs are induced during
10
this critical period of development and to understand through what mechanism ZGA
and cell cycle reentry are ultimately achieved.

Known Roles and Mechanisms of Topoisomerase II
To provide answers for these questions, our work has very much centered on
the roles of TOP-2. Thus, a brief review of the mechanism and functional roles of
topoisomerase II in eukaryotes may help us understand how the C. elegans germline
genome might take advantage of TOP-2 during ZGA. In mammals, there are two TOP2
isoforms, a and b, however C. elegans possess only one topoisomerase II enzyme
(TOP-2), and that shares 52% amino acid sequence identity to human topoisomerase
IIa (Jaramillo-Lambert et al., 2016). Type II topoisomerases are most well-known for
making DNA double strand breaks as opposed to type I topoisomerses which induce
single strand DNA breaks. Structurally, topoisomerse II are homodimeric enzymes
where each subunit makes a single break on double stranded DNA using ATP
hydrolysis. Without dissociating from the DNA as to protect it from rearrangment,
recombination, or damage repair responses, topoisomerase II then passes the second
unbroken DNA strand through and quickly religates the broken strand. This DNA
unwinding strategy critically relieves the torsional stress induced by DNA replication,
transcription and chromosomal segregation (reviewed by Nitiss, 2009). While our
last finding that the appearance of DNA DSBs by RAD-51 requires TOP-2 seems to
contradict its well-established role in making transient breaks, others have reported
that Rad51 foci also increase in response to Top2-mediated DNA damage in human
cells (de Campos-Nebel et al., 2010). Furthermore, at least one other type II
11
topoisomerase, SPO-11, behaves similarly in activating a DNA damage response
(Haffner et al., 2011). Still, this paradox motivated further investigation as to the true
function of TOP-2 during ZGA.
In addition to its function inducing DNA DSBs, TOP-2 has a long-established
role in mitotic chromosome condensation. Previous studies have reported that
proper chromatin condensation requires the opposing activities of condensin, which
overwind DNA, and topoisomerase II, which relaxes DNA (Baxter and Aragón, 2012).
Specifically, work in budding yeast has showed that topoisomerase II relaxes the
supercoiling produced by condensin, and thus is required throughout mitosis to
maintain balance (Baxter et al., 2011). Top2 in yeast has also been shown to promote
chromosomal segregation via recruitment to condensin on the chromosomal arm
during anaphase (Leonard et al., 2015). These diverse roles, coupled with our last
finding requiring TOP-2 for DNA DSBs, potentially highlight a central role for TOP-2
in C. elegans germline development, which we were inclined to investigate.

DNA Breaks and Chromatin Decompaction as Modes of Gene Activation in
Other Systems
The hallmark characteristics of genome activation in many other systems
provide incredible insight as to how C. elegans germ cells might be programmed to
develop in response to nutrients. First, topoisomerase II-mediated double strand
breaks have recently been shown to have a critical role in activating transcription
(reviewed by Calderwood, 2016). For example, the electrical stimulation of mouse
neurons induces Topo II-mediated DNA breaks and this leads to the immediate
12
expression of early genes required for neuronal maturation (Kim et al., 2010;
Madabhushi et al., 2015). In addition to expression of development genes, TOP2 binds
to androgen to induce DNA double strand breaks on promoters of cancer specific
antigens, and this activation causes chromosomal translocations that can lead to
human cancers (Haffner et al., 2010). In these examples, induction of TOP2 breaks is
ultimately mediated by external signals, whether they be hormonal or neuronal, and
thus it stands that nutrients may also signal TOP-2 DNA breaks in the C. elegans
germline. Another common feature of the described examples of signal-mediated
break-induced transcription, is that TOP2 specifically was shown to make breaks
along the promoters of genes, versus elsewhere in the genome, and this is a
mechanism unbeknown in Z2/Z3 that we critically pursue.
Chromatin decompaction is another emerging feature of cells abruptly
activated for genome wide expression after relatively long periods of transcriptional
quiescence. Chromosome condensation is a natural barrier to transcription that must
eventually be overcome especially at the end of mitosis (Thadani et al., 2012).
Numerous reports have implicated signal-mediated chromatin decompaction as a
post-mitotic mechanism for chromatin accessibility and subsequently, large shifts in
gene expression. During B cell activation, for instance, chromatin decondenses before
RNA polymerase II (RNAPII) transcription surges 10-fold to enable a proper immune
response (Kieffer-Kwon et al., 2017). Similarly, activated T cell receptors induce
structural changes in chromatin resulting in a shift from highly condensed to
decondensed aiding in T cell proliferation (Rawlings et al., 2011). Together these
studies demonstrate that chromatin compaction acts as a gatekeeper of gene
13
expression and that signal-mediated chromatin decompaction promotes timely,
genome-wide transcription.

RUVB-Like ATPase Decondensation Factors
If hypo-compacted chromatin is indeed a requirement for transcriptional
activation in Z2/Z3, what then is the molecular mechanism of chromatin
decompaction? Currently, little is known about how extracellular signals recruit
enzymes that allow for decondensation and gene expression. However, previous
work with Xenopus egg extracts suggest that chromosome decondensation is an active
process driven by highly conserved RuvB-Like ATPases, RuvBL1 and RuvBL2
(Magalska et al., 2014). These proteins form a double hexameric ring complex and
localize to decondensing chromatin at the end of mitosis in preparation of interphase
(Jha and Dutta, 2009; Magalska et al., 2014). In C. elegans, its orthologs RUVB-1 and
RUVB-2 both localize in the cytoplasm and nucleus and form a complex with LINKIN
to provide cell adhesion at the plasma membrane (Kato et al., 2014). However, until
now, a role for chromatin decompaction in C. elegans had not been established.
Interestingly, the current knowledge that DNA damage can relax chromatin has lead
us to speculate that RUVB-1 and RUVB-2 might exploit already established damage
responses to activate the genome (Ziv et al., 2006; Luijsterburg et al., 2012). Thus, in
the context of signal-mediated DNA damage, chromatin decompaction by RUVB-1 and
RUVB-2 is a provocative mechanism for gene expression.
The possibility that the developing germline genome is subject to the
enzymatic activities of TOP-2 and the rearrangement of chromatin architecture has
14
left many open questions for how these mechanisms might interact to promote ZGA.
The following work addresses this puzzle in three distinct chapters that examine the
activation of the germline genome using programmed DNA damage and chromatin
decompaction, the compaction of the genome during embryogenesis, and the cell
cycle arrest of embryonic PGCs. While each of these topics are studied in the context
of C. elegans, they each have broader impacts on our understanding of molecular and
cellular biology. For example, elucidating a mechanism for a rapid shift in genome
activation is significant because it is as much a developmental hallmark for cells in
the brain or immune system. Understanding how germline chromatin is regulated in
C. elegans is also important as ectopic germline activation can drive tumorigenesis in
somatic cells (McFarlane et al., 2014; Whitehurst, 2014). Lastly, identifying the cell
cycle proteins mediating arrested development can improve how we test pathways
for cell cycle arrest and reentry in higher organisms like humans. Thus, while our
research addresses the C. elegans germline specifically, it ultimately expands our
knowledge of cellular development and the molecular events surrounding gene
repression and expression.

15

CHAPTER I

Chromatin Decompaction via DNA Breaks and RUVB Proteins
Activates the Germline Genome

Adapted from:

Wong et al., accepted

16
INTRODUCTION
The genomes of quiescent cells are often characterized by highly condensed
chromatin and low transcriptional output. Upon stimulation, the chromatin is
globally decompacted and transcriptional output can be amplified many fold. The
molecular pathways that govern signal-mediated genome decompaction are poorly
understood, yet are likely to be critical for efficient activation of resting cells in a wide
variety of contexts. Recent work on lymphocyte activation highlights the nature of
this problem. During T cell development, the ability of the STAT5 transcription factor
to access DNA binding sites in naïve cells is prevented via hyper-compaction of the
chromatin by condensin II (Rawlings et al., 2011). Upon stimulation, the chromatin
is rapidly decompacted and STAT5 target genes are induced. Importantly, the
chromatin of resting cells which lack the condensin II subunit kleisin-b is hypo-
compacted, and STAT5-dependent gene expression is observed in the absence of
stimulation. Similarly, during B cell activation, the genome transitions from a globally
compacted state to a more open state, allowing transcription factors to rapidly access
their target sites (Kieffer-Kwon et al., 2017). In this system, genome decompaction is
driven by a combination of histone acetylation and Myc- and ATP-dependent
disassembly of chromatin nanodomains. These studies on lymphocytes suggest that
genome compaction in resting cells functions as an important blockade to the
expression of genetic programs that are normally turned on only after stimulation.
Chromatin architecture also appears to govern gene expression in the
developing germline of the nematode C. elegans. The two primordial germ cells
(PGCs), Z2 and Z3 (Z2/Z3), are born during early embryogenesis and, after DNA
17
replication, they arrest at G2. Z2/Z3 then exist in a quiescent state that is hallmarked
by a low (undetectable) level of active RNA polymerase II (RNAPII), and a highly
compacted chromatin state (Seydoux and Dunn, 1997; Schaner et al., 2003; Fukuyama
et al., 2006; Furuhashi et al., 2010). Z2/Z3 remain in this condition through hatching,
and only activate transcription in earnest after L1 larvae have fed. Thus, nutrients
serve as a signal that triggers zygotic genome activation (ZGA) in these cells. Previous
work from our group has shown that, surprisingly, one manifestation of Z2/Z3 ZGA is
the appearance of DNA damage (Butuči et al., 2015a,b). Cytological markers for
damage are observed shortly after hatched L1s feed, coincident with the global
activation of RNAPII. The damage is specific for transcriptionally active autosomes,
and is not observed on the transcriptionally inert X chromosomes, suggesting that
RNAPII activation and the induction of DNA damage are intimately connected during
germline ZGA.
In this work we study the connection between genome decompaction, DNA
damage, and ZGA. We find that topoisomerase II-mediated DNA breaks trigger
decompaction, and that they do so through recruitment of the RUVB complex to DNA.
RUVB proteins form a double hexameric ring complex and participate in a wide
variety of cellular processes, including histone acetylation and chromosome
decompaction after mitosis (Jha and Dutta, 2009; Magalska et al., 2014; Antonin and
Neumann, 2016). RUVB-mediated chromatin decompaction is required for ZGA in
Z2/Z3, and thus our data reveal a new pathway whereby programmed DNA breaks
trigger global genome decompaction to enable efficient and rapid zygotic genome
activation.
18
RESULTS
Identification of genes upregulated during germline ZGA
Previous work from our laboratory showed that DNA breaks form in the Z2/Z3
genome concomitant with RNAPII activation and ZGA. The breaks are restricted to
transcriptionally active autosomes and can only be observed on the X chromosomes
if transcription is experimentally induced (Butuči et al., 2015a,b). Thus there is an
intimate relationship between the breaks and transcriptional activation, however the
molecular mechanism in play was not known. To continue our studies on germline
ZGA it was first important to identify specific genes that are upregulated by the ZGA
program. We reasoned that genes encoding components of the germ cell specific P-
granules would be good candidates, given that germline proliferation requires an
increase in production of new components, and thus new gene expression is likely to
occur at ZGA. We chose five genes, positioned on chromosomes I and III (Fig. 1.1A),
and assayed their expression in starved and fed L1s using a hybridization chain
reaction based RNA-FISH procedure with multiplex capacity (Choi et al, 2016). To
mark Z2/Z3 we also probed for xnd-1, a gene known to be expressed in embryonic
Z2/Z3 but not somatic cells (Mainpal et al., 2015). As shown in Fig. 1.1B and Table
1.1, all five candidate ZGA genes (vbh-1, ifet-1, car-1, wago-1, and cgh-1) displayed
very low or undetectable expression in starved samples, and were clearly
upregulated in fed samples. We conclude that these five genes are bona fide germline
ZGA genes.

19

Figure 1.1. Genes encoding germ cell specific P-granule components are activated
during ZGA in the Z2/Z3 PGCs. A. Relative chromosomal positions of candidate ZGA
genes. B. L1s were starved or fed for 3 hours and then processed for multiplex RNA-FISH
and probed for vbh-1 (white) and ifet-1 (red) transcripts in one sample set and car-1
(white) and wago-1 (red) in another. For all samples, DAPI was included to visualize the
DNA (blue) and the germline gene xnd- 1 (green) was used to mark Z2/Z3 nuclei in, as it is
expressed embryonically, prior to ZGA. Representative images are displayed and the
number of nuclei scoring positive for the presence of a given transcript is displayed in
Table 1.1. Scale bar, 1µm in this and all figures.

20
TOP-2 is required for germline ZGA
Previous work in mammalian cells has shown that topoisomerase II induces
site-specific DNA breaks at gene promoters as part of a poorly understood mode of
gene activation that we refer to as “promoter breakage” (reviewed in Calderwood,
2016). In addition, our previous work had revealed a connection between C. elegans
topoisomerase II (TOP-2) and the DNA damage that arises at ZGA (Butuči et al.,
2015a). Because of these links we asked if TOP-2 plays a role in germline ZGA in the
worm. We used a previously described temperature-sensitive (ts) mutant of top-2
(Jaramillo-Lambert et al., 2016), allowing us to inactivate the enzyme by shifting from
the permissive temperature of 15°C to the non-permissive temperature of 24°C.
Embryos were collected from adults raised at 15°C and allowed to hatch overnight in
minimal medium lacking nutrients, also at 15°C. Samples were then split in half, with
one half remaining at 15°C while the other was transferred to 24°C for 16 hours. The
animals were then fed for three hours and processed for RNA-FISH. As shown in Fig.
1.2A and Table 1.1, all five genes were expressed when the samples were incubated
at 15°C, as expected, however upshift to 24°C blocked expression in all five cases. The
lack of germline gene expression at the restrictive temperature was due to
inactivation of TOP-2, and not simply incubation at 24°C, as the wild type samples
shown in Fig. 1.1B were also grown at 24°C. We conclude that TOP-2 is required for
expression of germline genes at ZGA.

We next asked if TOP-2 is required for global RNAPII activation during ZGA,
and we did so by performing immunofluorescence (IF) studies with the monoclonal
21
antibody H5 (MabH5), which we and others have used previously to mark
transcriptionally active nuclei in C. elegans (Seydoux and Dunn, 1997; Furuhashi et
al., 2010; Butuči et al., 2015a). MabH5 recognizes a phospho-epitope, Ser2-PO4,
within the serine-rich heptide repeat of the C-terminal domain of RNAPII that
correlates with active and elongating RNAPII (Palancade and Bensuade, 2003). Our
previous work showed that little if any MabH5 reactivity is found in starved Z2/Z3
nuclei, and that signal intensity is greatly increased after feeding (Butuči et al., 2015a).
In the current study we found that MabH5 signal intensity in Z2/Z3 nuclei after
feeding is dependent on the basal transcription factor TAF-10 (Fig. S1.1), as it is in
early embryos (Walker et al., 2004), and thus we conclude that MabH5 reliably
reports on active RNAPII in the Z2/Z3 PGCs. We stained fed L1s that had been
processed at either 15°C or 24°C, as described above, and observed that TOP-2
inactivation prevents nutrient-dependent activation of RNAPII (Fig. 1.2B). Thus, as is
the case for individual gene expression, TOP-2 is required for global RNAPII
activation during ZGA.

Randomly placed DNA breaks bypass the requirement for TOP-2 in ZGA
Our previous and current data have shown that DNA breaks form during ZGA
(Butuči et al., 2015a) and that TOP-2, a DNA break-inducing enzyme, is required for
ZGA (Figs. 1.2A&B). To explore this further we next asked if exogenously applied, and
randomly placed, DNA breaks could bypass the requirement for TOP-2 in ZGA. For
this we used ionizing radiation (IR) to induce breaks. Previous work has assessed the
frequency of breaks produced by IR in C. elegans at 1 break/17 Mb DNA/10 Gy dose
22
of IR (Yokoo et al., 2012), and based on this we estimate that our dose, 75 Gys, makes
~40 breaks total or one break every 2.25 Mb of DNA. We incubated top-2
ts
animals at
the nonpermissive temperature, irradiated them, and then assessed ZGA gene
expression and global RNAPII activation after feeding. Strikingly, we found that IR
could efficiently rescue both individual ZGA gene expression - in all five cases - (Fig.
1.2A and Table 1.1) and global RNAPII activation (Fig. 1.2C) in the TOP-2 deficient
samples. These findings make two important points. First, they show that the major,
if not only, function of TOP-2 during ZGA is to induce DNA breaks. Second, the fact
that randomly placed breaks are sufficient to activate all five ZGA genes shows that
breaks need not be made in a site-specific manner at gene promoters. Promoters are
small in C. elegans, such that ~3kb of upstream sequences are typically included when
producing transcriptional reporter transgenes (Boulin et al., 2006), and thus the
chances that an IR-induced break falls within a 3kb window is 2.25 Mb/3kb = 1 in
750. This low value of 1/750 stands in striking contrast to the efficiency with which
IR rescues individual gene expression, defined as the ratio of FISH-positive nuclei in
the TOP-2
-
+IR condition over FISH-positive nuclei in the TOP-2
+
condition, as these
values ranged from 3/4 (cgh-1) to 12/13 (vbh-1). It is thus obvious that breaks falling
outside of promoter regions can nonetheless activate germline genes. This raises the
question of do the naturally occurring breaks during ZGA form in a distal or proximal
manner, relative to target genes? To explore this important point we used IF imaging
at early time points post-feeding to co-localize sites of damage (defined by RAD-51
foci) relative to sites of early transcription (defined by MabH5 foci). Two examples
are shown in Fig. S1.1B, where sites of damage and sites of transcription are, for the
23

Figure 1.2. TOP-2 mediated DNA breaks trigger germline ZGA. A. Starved top-2(it7) L1s
were either maintained at the permissive temperature of 15°C, or shifted to 24°C overnight and
optionally treated with 75 Gys of IR. In all conditions samples were then fed for 3 hours before
being processed for RNA-FISH as in Fig. 1.1. Representative images are shown and results are
quantified in Table 1.1. B. Starved top-2(it7) L1s were shifted to 24°C overnight and fed (3
hours) as in (A). Samples were fixed and stained for P-granules (white) and Mab H5 (green).
Representative IF images of Z2/Z3 are shown, and the frequency of nuclei with a positive H5
signal is shown below the images. A minimum of 20 animals were examined per condition in
each experiment. C. TOP-2 was inactivated in starved L1s by temperature upshift to 24°C and
then optionally exposed to 75 Gys of IR. L1s were fed (3 hours), fixed, and stained as in (B).
Representative IF images of Z2/Z3 are shown and graph depicts percentage of L1s rescued for
24
most part, spatially distinct from one another. All together we scored 66 nuclei across
four time points and examined 273 RAD-51 foci for overlap with 87 MabH5 foci
(Table 1.2). We found that RAD-51 foci overlapped with MabH5 foci just 9.5% of the
time, whereas MabH5 foci overlapped with RAD-51 foci about 30% of the time. Thus,
in both cases, the majority of the signals were clearly separated from one another.
These data show that sites of damage form distally to sites of new gene expression
during ZGA. Based on these findings, together with the ability of IR to activate ZGA
gene expression, we conclude that our DNA break-induced germline ZGA pathway
occurs via a mechanism distinct from the previously described promoter breakage
phenomenon.

TOP-2 and RUVB proteins promote chromatin decompaction
Having eliminated promoter breakage we considered alternative mechanisms.
One previously described feature of Z2/Z3 nuclei prior to ZGA is that the chromatin
is tightly compacted (Schaner et al., 2003), and thus it seems likely that decompaction
must also occur so that the DNA would be suitable as a template for transcription at
ZGA. Previous work has suggested a role for TOP-2 in sperm chromatin
decompaction in both Xenopus and mice (Takasuga et al., 1995; Bizzaro et al., 2000),
and thus we considered the possibility that TOP-2 and DNA breaks are required to
decompact Z2/Z3 chromatin. To get at this we performed live-cell imaging and
assessed the compaction state of Z2/Z3 chromatin under various conditions, using a
strain expressing mCherry-tagged histone H2B (McNally et al., 2006) for this purpose.
In starved wild type samples, the Z2/Z3 chromatin manifested as tightly compacted
25
bundles positioned at the nuclear periphery, with the center of the nucleus notably
devoid of signal (Fig. 1.3A, “starved”). Similar results were obtained when the DNA
was stained directly with Hoechst 33342 (Fig. 1.3B), showing that the mCherry signal
corresponds to chromatin and not free histone H2B, as expected. After three hours
of feeding, we found that the chromatin became decompacted and unbundled, and
more signal was now observed in the nuclear center (Fig. 1.3A, “fed”). Thus, as has
been observed for activated B cells (Kieffer-Kwon et al., 2017), activation of Z2/Z3
results in a spreading of the chromatin from the nuclear periphery to the center. To
quantify this phenomenon, we measured the chromatin signal within the inner one
third of the nucleus (Fig. S1.2), and this revealed a significant accumulation of
chromatin in the nuclear center in fed samples, relative to starved (Fig. 1.3A).

Having established a quantitative assay for chromatin decompaction we next
built a strain combining the top-2(it7) allele with mCherry-H2B and used it to explore
a role for TOP-2 in decompaction. As shown in Fig. 1.3C, starved L1s that had been
incubated at either the permissive or non-permissive temperatures displayed
compacted chromatin. Feeding triggered decompaction at the permissive
temperature but, importantly, not at the non-permissive temperature (Fig. 1.3C).
Strains that are wild type for top-2 decompact normally at 24°C (Fig. 1.3A), and
thus these data show clearly that TOP-2 is required for chromatin decompaction
during germline ZGA. To pursue these observations we searched for other factors
involved in the compaction state of Z2/Z3 chromatin. We used RNAi to deplete the
RUVB-1 and -2 proteins, as recent work in Xenopus has shown that these factors are
26

Figure 1.3. Factor requirements for Z2/Z3 chromatin decompaction during ZGA. A. Strain WMM1
(top-2
+
) L1s were optionally fed 3 hours and living samples were imaged for chromatin morphology by
virtue of mCherry-tagged histone H2B. Vertical axis of bar graph measures the percentage of pixels in
the inner one-third of the nucleus. Seven samples were measured for each condition and a Student’s t-
test was used to compare the experimental group to its control. Error bars reflect one standard deviation.
A full description of this methodology is detailed in Fig. S1.2. B. Representative live-cell image of Z2/Z3 in
starved L1 showing Hoechst 33342 dye (blue) counterstaining mCherry-tagged histone H2B (red). C.
Starved top-2(it7) L1s were optionally shifted to the non-permissive temperature 24°C to inactivate TOP-
2 prior to 3 hours of feeding, live-cell imaging, and quantification as described in (A). D. Same as (A) and
(C) except samples were fed control RNAi or ruvb-1/2 RNAi prior to feeding and live-cell imaging. E.
Same as (D) except with capg-2 RNAi and only starved samples were analyzed.

27
required for chromatin decompaction after mitosis (Magalska et al., 2014).
Importantly, depletion of RUVB-1/2 prevented decompaction after feeding (Fig.
1.3D). We also asked if the highly compacted state of the chromatin observed in
starved L1s was dependent on condensin II, and found that it was as depletion of the
condensin II subunit CAPG-2 gave rise to a more diffuse chromatin signal in starved
samples, with more chromatin in the nuclear center (Fig. 1.3E). We conclude that
CAPG-2/condensin II is required to establish the tightly compacted state of Z2/Z3
chromatin and that both TOP-2 and RUVB-1/2 are required to undo the compaction
at ZGA.

Chromatin decompaction is required for germline ZGA
Data presented thus far show that TOP-2 is required for both ZGA and
chromatin decompaction, suggesting that decompaction may be a crucial prerequisite
for ZGA. If so, then other decompaction factors would also be required for efficient
ZGA. To test this we determined the requirement for RUVB-1/2 in ZGA. As shown in
Figs. 1.4A&B, ruvb-1/2(RNAi) samples showed significantly less activated RNAPII
than did the control RNAi samples, with nearly half showing no sign of ZGA after three
hours of feeding, whereas 100% of the control samples had initiated ZGA. These data
show that RUVB-1/2 proteins play an important role in triggering germline ZGA. To
confirm this we examined Z2/Z3 cell division, as our previous work had shown that
blocking ZGA also blocked Z2/Z3 division (Butuči et al., 2015a). We found that
depletion of RUVB-1/2 caused a severe delay in cell division (Fig. 1.4C), as expected
if ZGA was occurring less efficiently. Previous work on the RUVB-1/2 complex has
28

Figure 1.4. Chromatin decompaction is required for ZGA. A. Control or ruvb-1/2(RNAi) L1s were
fed, fixed, and stained for MabH5 (green). Representative images of Z2/Z3 (white arrows) are
shown. B. L1s that had been treated with the indicated RNAi were assessed by IF staining for Mab H5
signal. P-value derived from the Fisher test. C. Graph depicting the percentage of L1s with 3 or more
PGCs after treatment with the indicated RNAi and 6 hours of feeding. P-value derived from the Fisher
test. D. L1s treated with the indicated RNAi were fixed and stained for RUVB-1 (red) and P-granules
(green). Representative images of Z2/Z3 are shown. E. top-2(ts) L1s were treated with the indicated
RNAi and then shifted to 24°C prior to feeding. Samples were then processed for IF; representative
images show MabH5 (green) and P-granule (white) staining. Graph depicts the percentage of L1s with
a MabH5 signal in Z2/Z3. P-value derived from the Fisher test.

29
shown that it performs a wide variety of functions, including chromatin
decompaction, chromatin remodeling, snRNP assembly, and, in C. elegans, cell
adhesion (Jha and Dutta, 2009; Kato et al., 2014). Thus a requirement for RUVB-1/2
in ZGA could be through any of these disparate functions. To determine if the RUVB-
1/2 role in chromatin decompaction explains its requirement for ZGA, we performed
co-depletion experiments with RUVB-1/2 and CAPG-2. The rationale is that depletion
of CAPG-2 produces hypo-compacted Z2/Z3 chromatin in starved L1s (Fig. 1.3E), and
thus if RUVB-1/2 trigger ZGA via chromatin decompaction, then co-depletion with
CAPG-2 should override this requirement and ZGA should occur efficiently despite
the loss of RUVB-1/2. This is indeed what was observed, as the number of nuclei with
active RNAPII was significantly increased by co-depletion of CAPG-2 and RUVB-1/2,
relative to the samples depleted of RUVB-1/2 alone (Fig. 1.4B). Antibody staining of
these samples confirmed that RUVB-1 was efficiently depleted in both the single- and
double-RNAi condition (Fig. 1.4D). These data show that co-depletion of CAPG-2
rescues the ZGA defect observed upon depletion of RUVB-1/2. To see if this also true
of TOP-2, we asked if capg-2 RNAi could ameliorate the ZGA defects observed when
the top-2(ts) allele is incubated at the non-permissive temperature, and found that it
could (Fig. 1.4E). These data show that the compacted state of Z2/Z3 chromatin is a
major barrier that must be overcome for efficient and timely ZGA.

Data presented thus far suggest a model whereby TOP-2 promotes ZGA
through chromatin decompaction. We note that in previous work we had used RNAi
to deplete top-2 and observed that the timing of Z2/Z3 cell division was accelerated
30
(Butuči et al., 2015a). This result is inconsistent with the requirement for TOP-2 in
ZGA defined here, given that ZGA is required for cell division. One major difference
between the two experiments is the time at which TOP-2 is inactivated. In the RNAi
experiments top-2 is silenced from early embryogenesis onward, whereas in the
current experiments top-2 is inactivated after embryogenesis is complete. To pursue
this apparent conundrum we performed an “early inactivation” using the ts strain,
whereby young adults were shifted to the non-permissive temperature and their
progeny were examined for defects in germline ZGA (Fig. S1.3A). Previous work has
shown that shifting young adults to the non-permissive temperature allows
production of viable embryos (Jaramillo-Lambert et al., 2016), and we also observed
a high rate of embryonic viability under this condition (see Materials Methods). We
scored for ZGA by MabH5 staining and observed no difference between samples
incubated at the permissive relative to non-permissive temperature (Fig. S1.3B),
consistent with our previous work showing that Z2/Z3 in top-2(RNAi) L1s readily
undergo cell division. We also assessed the chromatin compaction status of starved
L1s that had been derived from the permissive or non-permissive temperatures and
here we noticed a striking difference - chromatin from the samples grown at the
permissive temperature displayed the usual tightly compacted configuration
whereas samples where top-2 was inactivated early showed the more diffuse
chromatin pattern normally associated with fed samples (Fig. S1.3C). These data
show that loss of TOP-2 during embryogenesis produces a hypo-compacted
chromatin state, and this explains why ZGA readily occurs after feeding. Based on
these findings, we conclude that TOP-2 plays opposing roles in chromatin compaction
31
during development in C. elegans - in the embryo it is required to establish the highly
compacted state of Z2/Z3 chromatin whilst in L1s, at ZGA, it acts to reverse this
compaction to allow gene expression (Fig. S1.3Di). As such, early inactivation of TOP-
2 allows ZGA because the hyper-compacted chromatin state is never established (Fig.
S1.3Dii), whereas late TOP-2 inactivation blocks ZGA as the hyper-compacted state is
established but not reversed (Fig. S1.3Diii). We note that a conserved role for
topoisomerase II in promoting chromosome compaction has been known for some
time (Newport and Spann, 1987; Uemura et al., 1987).

TOP-2 recruits RUVB proteins to transcriptionally active chromosomes during
germline ZGA
Data presented thus far show that TOP-2 and condensin II establish the highly
compacted Z2/Z3 chromatin architecture found in starved L1s and then, at feeding,
TOP-2 and RUVB proteins act to decompact the chromatin to allow gene expression.
To pursue the roles of TOP-2 and RUVB in ZGA chromatin decompaction we assessed
localization of the proteins within Z2/Z3 nuclei. We used a strain where the top-2
gene is fused, at the endogenous locus, to sequences encoding the Flag tag (Jaramillo-
Lambert et al., 2016), and stained starved L1s for TOP-2 using Flag antibody. TOP-2
signal showed extensive overlap with DNA (Fig. 1.5A), revealing that TOP-2 is present
in starved Z2/Z3 nuclei. We next stained for the RUVB-1 protein, using a previously
published antibody (Kato et al., 2014), and found that while Z2/Z3 nuclei in starved
samples showed a faint RUVB-1 signal, the signal in fed animals was much more
robust (Fig. 1.5B). The strong RUVB-1 signal that accompanies feeding was found to
32

Figure 1.5. TOP-2 is upstream of RUVB proteins in the ZGA pathway. A.
Representative image of Z2/Z3 stained for TOP-2 (white) in starved L1s using strain
AG275. Flag antibody was used. B. Representative images of Z2/Z3 stained for RUVB-1
(red) and P-granules (green) after L1s were starved, or fed for 3 hours. Samples were
imaged using identical exposure conditions. At least 40 samples in two biological
replicates were quantified for strong and weak RUVB-1 signals. C. Representative
image of Z2/Z3 fed for 3 hours and co-stained for RUVB-1 (red), HIM- 8 (white), and P-
granules (green). Samples were quantified based on whether the HIM-8 focus
overlapped with RUVB-1 signal or not. D. Representative images depicting RUVB-1
(red) accumulation in Z2/Z3 in starved or fed top-2(it7) L1s at the permissive (15°C)
or non-permissive (24°C) temperatures. Samples were imaged and quantified as
described in (B).

33

Figure 1.5. (Continued). TOP-2 is upstream of RUVB proteins in the ZGA pathway.
E. Representative IF images of Z2/Z3 after TOP-2 inactivation at 24°C in starved L1s. Samples
were fed (3 hours), fixed, and stained for RAD-51 foci (red) and P-granules (white). Graph
shows percentage of L1s with RAD-51 foci. A minimum of 20 samples from each condition
were analyzed in both biological replicates. F. L1s were fed ruvb-1/2 RNAi and then fed (3
hours), fixed, and stained for RAD-51 (red). Samples were quantified as in (E).

34
cover much of the chromatin in Z2/Z3 nuclei, however we consistently observed
small portions of the nucleus that were positive for DNA yet lacked RUVB-1. We
considered the possibility that these regions correspond to the X chromosomes, as
our previous work has shown that the X is transcriptionally silent in Z2/Z3, even after
feeding (Butuči et al., 2015a). To explore this we co-stained for RUVB-1 and HIM-8,
which binds specifically to the X and manifests as a small focus (Phillips et al., 2005).
This analysis revealed that RUVB-1 and HIM-8 signals were anti-correlated the vast
majority (85%) of the time (Fig. 1.5C), which supports the notion that RUVB-1 is
recruited to transcriptionally active autosomes in Z2/Z3, and not to the
transcriptionally quiescent X chromosomes. We next asked if RUVB-1 recruitment is
dependent on TOP-2, and found that it is as late inactivation of TOP-2 prevented
RUVB-1 accumulation in fed L1s (Fig. 1.5D). These data allow a linear pathway to be
constructed whereby nutrients activate TOP-2 to produce DNA breaks and these
breaks, in turn, promote recruitment of RUVB to chromatin for decompaction. If so,
then we expect a requirement for TOP-2 but not in RUVB proteins in forming DNA
breaks during germline ZGA, and this is exactly what we observed (Figs. 1.5E&F).

The role of DNA breaks during germline ZGA is to activate RUVB proteins for
chromatin decompaction.
Data presented thus far show that TOP-2 acts upstream of RUVB proteins to
promote ZGA. Data also show that exogenous DNA breaks bypass the requirement
for TOP-2 in ZGA, and taken together the two findings suggest that the role of TOP-2
in ZGA is to produce the DNA breaks required by RUVB proteins to initiate chromatin
35
decompaction and to thereby allow gene expression. In a final set of experiments we
set out to test this prediction. We first determined if DNA breaks could indeed
activate RUVB proteins for chromatin decompaction. When starved L1s were
irradiated and then examined by live-cell imaging we observed that Z2/Z3 chromatin
became decompacted in a manner equivalent to what is typically observed during
ZGA (Fig. 1.6A). Furthermore, in starved Z2/Z3, RUVB proteins were recruited to the
DNA by IR (Fig. 1.6B) and are required for IR-induced decompaction (Fig. 1.6C).
These data show that RUVB proteins can be recruited to chromatin by DNA breaks to
promote decompaction - but is this happening at TOP-2 generated breaks during ZGA?
To address this important question we took advantage of the fact that when TOP-2
produces persistent DNA breaks the enzyme must be actively removed from DNA
ends by the 5'-tyrosyl-DNA phosphodiesterase activity of the TDTP-1 protein
(Pommier et al., 2014,2016). We reasoned that depletion of TDTP-1 would trap TOP-
2 on the DNA end, thereby inhibiting access of other cellular factors to the DNA. To
test this we determined if loss of TDTP-1 would suppress RAD-51 foci formation, as
predicted if the DNA end is blocked and inaccessible to the DNA repair machinery.
Control or tdtp-1(RNAi) L1s were fed for three hours and then assayed by IF for RAD-
51 foci. As shown in Fig. 1.6D, depletion of TDPT-1 attenuated RAD-51 foci formation,
showing that loss of TDPT-1 activity does indeed perturb the processing of DNA ends.
We next asked if loss of TDPT-1 would attenuate recruitment of RUVB proteins to
chromatin during ZGA, as expected if DNA ends serve as a portal for RUVB entry.
Control or tdtp-1(RNAi) L1s were processed for RUVB-1 IF after feeding, and while
strong signals were observed in control samples, the TDPT-1 depleted samples
36

Figure 1.6. DNA breaks activate RUVB proteins for chromatin decompaction. A. Starved
samples were optionally treated with 75 Gys of IR and living samples then imaged as in Fig.
1.3. Representative images of Z2/Z3 are shown with bar graph. B. Representative images of
Z2/Z3 in starved L1s after 75 Gys of IR and stained for P-granules (green) and RUVB-1 (red).
Quantification of strong and weak RUVB signals is shown below. C. Worms were treated with
control or ruvb-1/2 RNAi and L1 progeny were exposed to 75 Gys of IR prior to live-cell
imaging using mCherry H2B. Quantification as in (A).

37
showed weak signals, indicative of failed RUVB recruitment (Fig. 1.6E). If the reason
that RUVB recruitment fails after loss of TDTP-1 is due to persistence of TOP-2 on the
DNA end, then inducing DNA breaks in a TOP-2 independent manner should allow
RUVB recruitment despite inactivation of TDTP-1, and this is exactly what is observed
as IR-induced breaks readily recruit RUVB-1 to DNA in tdpt-1(RNAi) L1s (Fig. 1.6E).
Based on these data, we conclude that the role of TOP-2 induced DNA breaks during
germline ZGA is to recruit RUVB proteins to chromatin so that decompaction and gene
expression may ensue.

38

Figure 1.6 (Continued). DNA breaks activate RUVB proteins for chromatin
decompaction. D. Representative images of Z2/Z3 in samples fed tdpt-1 RNAi and then
fixed, stained, and quantified as in Fig. 1.5E. A minimum of 20 samples from each condition
were analyzed in both biological replicates. E. Control RNAi and tdpt-1(RNAi) L1s were fed
for 3 hours, fixed, and stained for RUVB-1 (red) and P-granules (green). tdpt-1(RNAi) L1s
were optionally exposed to 75 Gys gamma irradiation (IR) before feeding (far right panel).
Representative images of Z2/Z3 are shown and graph depicts percentage of L1s with a
strong RUVB-1 signal for each condition. F. A novel genome decompaction pathway that is
required for ZGA and the cell cycle re- entry of quiescent PGCs.

39
DISCUSSION

Signal- and DNA-break mediated genome decompaction allows ZGA and cell
cycle re-entry
The goal of this study was to determine how the previously documented
appearance of DNA damage is connected to RNAPII activation during germline ZGA
in the Z2/Z3 PGCs. Collectively, our results describe a novel pathway whereby
nutrients activate TOP-2 to produce persistent DNA breaks in the genome (Fig. 1.6F).
These breaks trigger TDPT-1 to remove TOP-2 cleavage complexes from the DNA, and
this, in turn, sets in motion two pathways. The first promotes RAD-51 filament
assembly and subsequent DNA repair by HR (Butuči et al., 2015a), while the second
promotes RUVB-dependent decompaction of Z2/Z3 chromatin (this study). The
latter pathway is required, ultimately, for RNAPII activation, germline gene
expression (ZGA), and cell cycle re-entry (Fig. 1.6F). In addition, HR-mediated repair
is also required for timely Z2/Z3 cell cycle re-entry (Butuči et al., 2015a). These
findings are, to our knowledge, the first report of signal- and DNA-break mediated
genome decompaction to allow global activation of RNAPII-dependent gene
expression. We note that while this DNA break-based decompaction pathway is
required for ZGA, it is not sufficient, as exogenously applied breaks fail to trigger ZGA
in starved animals (data not shown). We propose that, in addition to decompaction,
ZGA also depends on nutrient-activated TFs, operating in a parallel pathway, as
shown in Fig. 1.6F.

40
Site specific breaks at spacer regions may allow decompaction and activation
Although our studies successfully answered the question of how DNA damage
is connected to germline ZGA, they have also raised several new and important
questions that will guide our future research. One such question is: how do nutrients
activate TOP-2 to induce the DNA breaks? Over the years there have been several
reports of TOP-2 mediated DNA breaks made at gene promoters so that signal-
mediated gene expression may occur (promoter breakage, reviewed by Calderwood,
2016). One important difference between these previous studies and our current
study is that, for Z2/Z3 ZGA, TOP-2 is needed to globally activate transcription, and
this is unlikely to occur via DNA breaks made at individual promoters. Indeed, in the
Z2/Z3 system, randomly placed DNA breaks are sufficient for robust activation of
RNAPII and ZGA gene expression (Figs. 1.2A,C), and sites of damage and early
transcription do not show extensive overlap (Fig. S1.1B & Table 1.2), as the promoter
breakage model mandates. It thus appears topoisomerase II can control transcription
by at least two mechanisms - promoter breakage for individual genes and chromatin
decompaction for global activation events like ZGA. For the global pathway described
here a fascinating question becomes are the breaks made in a site-specific manner,
despite the lack of a stringent requirement for specificity? We note that recent work
on genome organization in C. elegans has identified chromatin state domains
containing clusters of genes with similar expression patterns (Evans et al., 2016).
Interestingly, the spacer regions between these domains are enriched for TF binding
sites, and thus one possibility is that TOP-2 is guided to these regions via interaction
with a nutrient-activated TF. TOP-2 would then produce a DNA break in the spacer
41
region, thereby allowing decompaction of nearby domain(s). Consistent with this
possibility is previous work showing physical association of TOP-2 with signal-
activated TFs, such as nuclear hormone receptors (Ju et al., 2006). In general, we
favor a site-specific mode for ZGA DNA breaks as this would prevent them from
forming in sensitive regions of the genome, for example within open reading frames.
Future work will seek to identify sites of ZGA DNA breaks genome wide so that this
important question may be answered.

Transcription and DNA damage: The chicken and the egg
A related issue to how TOP-2 is activated concerns our previous study
showing that RNAPII activity is required for DNA breaks to form during ZGA (Butuči
et al., 2015a). In these experiments, loss of the basal transcription factor TAF-10, or
treatment with the RNAPII elongation inhibitor a-amanitin, caused a significant
decrease in the number of RAD-51 foci that formed after feeding. Thus RNAPII
activity triggers TOP-2 dependent DNA breaks, which then globally activate RNAPII.
We envisage two possibilities for the relationship between transcription and DNA
breaks. One, it is possible that nutrients trigger expression of one or a few pioneer
factors that go on to activate TOP-2 and help it to induce the DNA breaks that are
required for global chromatin decompaction and genome activation. In this
scenario, a limited amount of early transcription occurs prior to global
decompaction, which could readily occur if small regions of the genome escape
decompaction, a possibility that is not inconceivable. Such a mechanism includes
the benefit of rendering DNA breaks absolutely dependent on nutrient-dependent
42
gene expression, as opposed to a condition where the break-inducing machinery is
present but inactive in starved Z2/Z3, as this latter scenario could lead to
inappropriate induction of DNA breaks prior to the acquisition of nutrients. A
second possibility is that the DNA break-inducing machinery senses RNAPII that has
been activated by nutrients yet stalled by compacted chromatin. Previous work has
suggested that RNAPII is prebound to promoters in starved Z2/Z3 (Furuhashi et al.,
2010), and thus it may be that nutrients promote RNAPII activation on DNA, leading
to stalled transcription complexes that are sensed in some manner by TOP-2 to
generate DNA breaks, trigger decompaction and thereby liberate RNAPII. Clearly,
further work on the very early events in germline ZGA is required to resolve this
important issue.

An ‘expanding’ role for RUVB proteins
Another interesting question raised by our findings concerns the relationship
between DNA breaks, the RUVB proteins, and chromatin decompaction. RUVB-1 and
-2 form a heterohexameric ATPase that has previously been shown to decondense
mitotic chromosomes (Magalska et al., 2014), however the mechanism for this is not
known. We have shown here that RUVB-1/2 is required for genome decompaction at
ZGA (Fig. 1.3D), and after DNA damage (Fig. 1.6E). It thus appears that, in C. elegans,
decompaction for ZGA and for DNA repair occur through a common, RUVB-mediated
mechanism. Importantly, previous work has shown that DNA breaks trigger ATP-
dependent chromatin decompaction and this response is conserved from yeast to
man (Kruhlak et al., 2006; Ziv et al., 2006; Falk et al., 2007; Amitai et al., 2017). It is
43
thus interesting to speculate that RUVB-mediated decompaction after DNA breaks is
an ancient module, conserved across divergent organisms, that has been co-opted
during nematode evolution to promote germline ZGA. It will thus be important to
determine if RUVB proteins play a role in break-triggered decompaction in human
cells. Although we do not yet understand why DNA breaks are required for RUVB-
mediated decompaction during ZGA, our data do show that breaks do not simply
relieve torsional stress along the chromosome to allow RUVB recruitment. We have
shown that both TOP-2 and TDPT-1 are required for RUVB recruitment, and the
TDPT-1 requirement shows that simply breaking the DNA is not sufficient, rather, the
end must be processed to allow RUVB binding. Thus another important avenue for
future work will be determining how RUVB-mediated decompaction occurs, and how
it is integrated with HR-based DNA repair.

TOP-2 DNA breaks for rapid genome activation: efficient, effective, and evolved
Another fascinating question is why evolution has produced a DNA break-
based mechanism for activation of the germline genome, as this strategy comes with
obvious risks to genome integrity and, therefore, to the very success of the species
itself. A potential explanation is found in the highly compacted state of the genome
prior to activation. Perhaps this highly compacted state protects the genome against
instability during embryogenesis and potentially long periods of starvation in
hatched L1s. Once nutrients are available, the protective role of compaction becomes
dispensable, and is now a barrier to gene expression. Thus a mode of rapid
decompaction may have been favored, to allow the organism to quickly respond to
44
the presence of nutrients. It, therefore, may be that DNA breaks serve a time-saving
function in genome decompaction and activation. The idea that highly compacted
genomes require rapid decompaction as cells transition from the resting to activated
state is not unique to Z2/Z3, as this has clearly been observed in human lymphocytes
(see Introduction).

45
MATERIALS AND METHODS

Animals
All C. elegans strains were obtained by the Caenorhabditis Genetics Center, provided
as gifts from other labs, or generated in the lab. Nematodes were plated on each
60mm petri dish containing standard nematode growth medium (NGM) seeded with
E. coli OP50. C. elegans were cultured and grown at 20°C or 24°C, except where
temperature sensitive mutants required an environment of 15°C. Strains were
maintained by chunking, picking, or bleaching. When nematodes were prepared for
experimentation, gravid adults were bleached with sodium hypochlorite to collect
fresh embryos using a method described below.

Microbe Strains
The primary C. elegans food source E. coli OP50 was cultured in lysogeny broth (LB)
at 37°C shaking overnight with 100 µg/mL streptomycin. 500 µL of OP50 was seeded
and dried on petri dishes containing NGM. We found that this amount of food was
sufficient to support the growth of ~500 worms from L1 to adulthood.

For RNAi knockdown experiments, bacteria were obtained from the Ahringer RNAi
feeding library or plasmids were generated in house. HT115(DE3) competent cells
transformed with RNAi plasmids of selected genes were cultured in LB at 37°C
shaking overnight with 100 µg/mL carbenicillin and 12.5 µg/mL tetracycline. A 1:100
dilution of the overnight stock solution was grown at 37°C in 2xYT broth to OD595 of
0.4, and then induced with 1mM IPTG for 4 hours at 37°C. The culture was
46
concentrated to a 10x stock and 400 µL of the RNAi feeding vector bacteria was
seeded onto NGM+IPTG plates. All prepared OP50 and RNAi plates were stored at 4°C
before further use.

Egg Preparation
Worms were grown to adulthood in petri dishes containing NGM E. coli OP50. Gravid
adults were collected in a 15 mL conical tube with 10 mL of M9 minimal medium
(22mM KH2PO4 , 22mM Na2HPO4, 85mM NaCl, 2mM MgSO4). A bleach solution (30%
NaOCl, 0.25N NaOH solution) was freshly prepared. Worms were spun down in a
centrifuge for 1 minute at 1900 RPM and M9 was aspirated. 5 mL of bleach was added
to each aliquot of adult worms. Worm bodies were dissolved in the conical tube by
vortexing for 30 seconds and inverting for 1 minute, and repeating this twice more.
The tube was vortexed for another 30 seconds, making the total time of vortexing and
inverting 5 minutes. Embryos and dissolved worm bodies were immediately spun
and the bleach solution was aspirated. Embryos were rinsed of bleach 3 times by
adding 10 mL of M9 to each tube, and spinning, and aspirating. Embryos were allowed
to hatch overnight on a shaker in 10 mL of M9.

Strain Construction
WMM1: OD56 males were crossed with SS747 hermaphrodites.
WMM2: WMM1 and KK381 strains were crossed to generate a top-2 temperature
sensitive mutant carrying mCherry-tagged histone H2B. P0 WMM1 young adult males
and P0 KK381 L4 hermaphrodites were picked and plated together on 12-well plates.
47
F1 progeny were generated by self-fertilization or mating, but F1 progeny were
screened in each well for ~50% males. From each well with a successful crossing, F1
hermaphrodites were individually picked and grown to adulthood. F2 worms were
singled by plating 1 worm per well in a 12-well plate and grown to adulthood. F3
progeny were scored in their wells for endogenous mCherry and embryonic lethality
at 24°C.

RNAi
E. coli HT115 transformed with pL4440 was used as a control RNAi in all experiments.
RNAi plasmids containing capg-2, top-2, ruvb-1, ruvb-2, and taf-10 were obtained from
the Ahringer RNAi feeding library. The RNAi plasmid for tdpt-1 was cloned in house.
All RNAi vectors were verified by Sanger sequencing, cultured, and seeded on
NGM+IPTG plates for induction by feeding. The following describes our analysis of
RNAi efficacy of each bacteria.

capg-2: P0 L1 worms were plated onto NGM plates and fed E. coli OP50 for 36 hours
at 24°C until L4 stage. L4 worms were then transferred to capg-2 RNAi plates for 16
hours at 24°C. Gravid adults were bleached and F1 embryos hatched overnight. Under
these conditions, only ~9.5% of our samples survived embryogenesis and hatched.
We note, however, that growth conditions variably affected embryonic lethality for
capg-2(RNAi). When L1s were grown to adulthood at 72 hours at room temperature,
and transferred to capg-2 RNAi 16 hours prior to egg extraction, embryonic lethality
was measured at ~33%. A similar phenotype was observed when L1s were grown to
48
adulthood at 15°C for 5 days and transferred to capg-2 RNAi as young adults for 24
hours.

ruvb-1 and ruvb-2: Previous work reported that RuvBL-1 and RuvBL-2 function
independently to decondense mitotic chromatin in Xenpous egg extracts (Megalska et
al., 2014). Therefore, we depleted both proteins of the hetero-dimeric complex by
seeding plates with cultures for both RNAis at a 1:1 ratio (henceforth referred to as
ruvb-1/2 RNAi). At 24°C, P0 L1 worms were grown on E. coli OP50 for 32 hours and
then transferred to ruvb-1/2 RNAi plates for 24 hours until gravid. Adults were
bleached and embryos were collected and allowed to hatch overnight. Embryonic
lethality was ~75% under these conditions. Additionally, when worms were grown
at room temperature and fed E. coli OP50 and ruvb-1/2 RNAi for 48 and 24 hours,
respectively, embryonic lethality was measured at ~76%.

taf-10: Worms were fed E. coli OP50 until early L4 stage, and were then transferred
to taf-10 RNAi plates for 12 hours. Gravid adults were bleached and experiments were
performed on F1 L1s. Embryonic lethality was measured at ~10%.

tdpt-1: Conditions were identical to ruvb-1/2 RNAi, as described above, where tdpt-
1 RNAi was administered by feeding for 24 hours prior to bleaching. Experiments
were performed on F1 L1s and embryonic lethality was consistently about ~10%.

49
Mutant Analysis
top-2: Strains KK381 top-2(it7ts) and WMM2 top-2(it7ts) were analyzed for TOP-2
function. Strain KK381 was a kind gift from Andy Golden (National Institutes of
Health). The allele top-2(it7ts) has previously been fully characterized (Jaramillo-
Lambert et al., 2016) but to confirm its behavior in our hands we observed that
animals shifted from a permissive temperature (15°C) to a non-permissive
temperature (24°C) at the L4 stage yielded an embryonic lethality of ~96%. For our
purposes, TOP-2 was inactivated at two different times during worm development.
To deplete TOP-2 in embryos, P0 worms were upshifted to 24°C as young adults.
Although this method yielded a lethality of ~70% in F1 embryos, we also note that
100% of embryos contained paternal pronuclei. To inactivate TOP-2 after
embryogenesis, P0 worms were grown at the permissive temperature and their
progeny, F1 L1s, were shifted to 24°C while starved in M9 media.

Antibodies and Dilutions
P-granules: mouse Mab OIC1D4, from the Developmental Studies Hybridoma Bank
(DSHB), was used at 1:10. Mouse Mab K76, from DSHB, was used at 1:5. RNAPII pSer2:
mouse Mab H5, from Biolegend, was used at 1:50. RAD-51: rabbit antibody
#2948.00.02, from SDIX, was used at 1:10,000. RUVB-1: Rabbit polyclonal antibody,
a gift of Paul W. Sternberg (CalTech), was used at 1:250. HIM-8: rat antibody, a gift of
Abby Dernburg (U.C. Berkeley) was used at 1:500. Secondary antibodies: Alexa Fluor
conjugated secondary antibodies were purchased from Invitrogen and used at 1:200.

50
Gamma Irradiation
L1 worms were irradiated with 75 Grays (Gy) of gamma rays using a Precision X-RAD
iR160 X-ray source. Samples were allowed to recover for 1 hour before
experimentation.

Immunofluorescent Staining of L1 Worms
To stain L1 worms with Mab H5, worms were spun down in glass tubes and washed
three times with Milli-Q water. The L1 worms were then transferred to an Eppendorf
tube and 1 mL of 100% methanol at -20°C was added. Worms were spun down for
one minute and then immediately fixed at room temperature for 20 minutes (0.08M
Hepes pH 6.9, 1.6mM MgSO4, 0.8mM EGTA, 3.7% formaldehyde, 1X phosphate-
buffered saline). After fixation, worms were washed 3 times at room temperature for
10 minutes in TBS-T. A freshly prepared 100 µL aliquot of SDS-DTT (804 µL 0.31%
SDS, 196 µL 1M DTT) was then added to the tube after rinsing and worms were
shaken in a Thermomixer at 500 rpm at room temperature. Shaking times in SDS-DTT
varied between 10-30 minutes to ensure most of the worms were permeabilized, as
normally indicated by a bump in their cuticle. SDS-DTT was briskly rinsed 3 times in
TBS-T. To block, 50 µL of TNB blocking solution was added and tubes were incubated
in the Thermomixer at 500 rpm at room temperature for 2 hours. Worms were spun
and blocking solution was removed and replaced with a primary antibody solution of
50 µL. Worms were incubated for 16-hours overnight in the Thermomixer at 500 rpm
at 4°C. The next day, worms were washed 3 times for 10 minutes in TBS, and then
incubated with 50 µL of a secondary antibody solution for 2 hours in the
51
Thermomixer at 500 rpm at room temperature. After incubation, worms were
washed 3 more times in TBS and mounted on a glass slide coated with poly-L-lysine.
Mounted samples were counterstained with DAPI and coverslips were sealed with
Cytoseal XYL.

This same method was also used in staining experiments with RUVB-1, RAD-51, and
HIM-8 primary antibodies, with the exception that 1:10 normal goat serum (NGS) was
also added to TNB to make the blocking solutions.

RNA-FISH (Hybridized Chain Reaction in situ)
To detect mRNA transcription, a kit containing a DNA probe set, a DNA hybridized
chain reaction (HCR) amplifier, and hybridization, wash, and amplification buffers
was purchased from Molecular Instruments (molecularinstruments.org) for each
target mRNA. The xnd-1 probes initiate B5 (Alexa488) amplifiers, the car-1, cgh-1,
and vbh-1 probes initiate B3 (Alexa647) amplifiers, and the ifet-1 and wago-1 probes
initiate B4 (Alexa546) amplifiers. The experiments were performed exactly
according to the manufacturer’s instructions. We used the 3
rd
Generation system,
which is described in a bioRxiv preprint (https://doi.org/10.1101/285213).

L1 worms were cultured at 15°C or 24°C and optionally fed E. coli OP50, depending
on the experiment, before being permeabilized with 4% paraformaldehyde. All
samples were stored at -80°C for at least 1 day prior to performing HCR in situ.
Samples were thawed and treated with proteinase K (100 µg/mL) for 10 minutes at
52
37°C and washed twice with PBST. Samples were then treated with glycine (2
mg/mL) for 15 minutes on ice, and washed twice more with PBST. Probe
hybridization buffer (PHB) and PBST (1:1) was added to samples before pre-
hybridization in PHB for 30 minutes at 37°C. Probe mixture (2 pmol each) was
added to the PHB and samples overnight for 16 hours at 37°C. The next day, excess
probes were washed with probe wash buffer and 5x SSCT at 37°C and room
temperature, respectively. Samples were incubated in pre-amplification buffer for
30 minutes at room temperature. To amplify the HCR signal, fluorescent hairpins (3
µM) were snap cooled in amplification buffer before being added to samples, which
were incubated overnight for 16 hours at room temperature. The following day,
excess hairpins were washed with 5x SSCT and samples were mounted on slides
with DAPI.

Live-cell Imaging of Chromatin
Worm strains WMM1 and WMM2 carrying an mCherry-tagged histone H2B were live-
cell imaged in L1s to analyze states of chromosome compaction in cells Z2 and Z3. To
prepare L1s for live-cell imaging, adult worms were bleached on the previous day and
embryos hatched overnight. A 10 µL aliquot of L1s suspended in M9 was transferred
onto a glass coverslip. L1s were paralyzed with 10 µL of either 1mM of levamisole or
10mM sodium azide. Coverslips containing L1s in M9 were mounted on a 3% agarose
pad. Samples were immediately imaged using an Olympus Fluoview FV1000 confocal
microscope and analyzed in ImageJ with no additional processing software. In both
strains, we were readily able to identify Z2 and Z3 in starved L1s by ploidy; DNA was
53
previously replicated (2N) and thus a concomitant intensity of the mCherry signal
was observable by eye.

Hoechst Staining of Live Cells
A small aliquot of L1s was transferred with Milli-Q H2O to an Eppendorf tube and
spun at 1900 RPM for 1 minute. Worms were incubated with SDS-DTT (804 µL 0.31%
SDS, 196 µL 1M DTT) for 4 minutes to lightly soften their cuticles. Samples were then
incubated in Hoechst dye 33342 diluted at 1:5000 for 15 minutes. L1s were spun
down and mounted on a glass slide with an agar pad for live-cell imaging.

Immunofluorescent Imaging
All samples were imaged on an Olympus Fluoview FV1000 confocal microscope using
Fluoview Viewer software. To achieve consistent exposure levels, High Voltage lasers
were maintained at a tight range between samples and conditions in each experiment.

QUANTIFICATION AND STATISTICAL ANALYSIS
All figure data were obtained by independently performing two or three biological
replicates. In many figures, error bars were not used because the raw figures for any
given replicate may have varied due to day-to-day differences in timing. Importantly,
however, such variations had no overall impact on the trend of our results. Therefore,
54
in lieu of standard statistical measurements, we often show figures with data for both
experimental replicates side-by-side. For some quantifications, data were analyzed
using a Student’s t test or Fisher’s exact test and were considered significant with a
p-value <0.05.

DATA AND SOFTWARE AVAILABILITY
Software
The images captured on Olympus Fluoview FV1000 were processed through
Fluoview Viewer software without any further processing. For images captured of
live animals to quantify chromatin compaction, images were processed in ImageJ to
determine raw integrated densities of pixels. All images were saved and exported as
TIFF files.

55

Figure S1.1. MabH5 reactivity is lost after taf-10 RNAi in feeding L1s and rare
instances of RAD-51 and MabH5 foci overlap during ZGA. Related to Figure 1.2. A.
Control and taf-10(RNAi) L1s were fed for 3 hours and then fixed and stained for P-granules
(red) and Mab H5 (green). Representative IF images of Z2/Z3 are shown, and the frequency
of nuclei with a positive H5 signal is shown below the images. A minimum of 20 animals
were examined per condition in each experiment. Data reveal that loss of the basal
transcription factor TAF-10 prevents ZGA. B. L1s were fed for the indicated time and then
fixed and stained for RAD-51 (red), Mab H5 (green), and P-granules (white) to assess the
localization of sites of damage and early transcription. Representative images are shown, as
well as a graphic overlay (far right) with overlapping RAD-51 and H5 foci circled in white.
Quantification of RAD-51 and MabH5 foci overlap is shown in Table 1.2.

56

Figure S1.2. Method for quantification of chromatin decompaction.
An outer ellipse was drawn around the chromatin to count total pixel density. A
proportional inner ellipse was set to one-third of the area of the outer ellipse. The
inner ellipse was positioned to the darkest region in the center of the nucleus
using a Plot Profile. The pixel density for this area was captured, and the
percentage of pixels within the inner 1/3 of the nucleus was determined by
pixels inner circle/pixels outer circle.

57

Figure S1.3. Early inactivation of TOP-2 produces hypo-compacted chromatin in
starved L1s and allows ZGA. A. Method for early inactivation of TOP-2: young adults were
shifted from the permissive temperature (15°C) to the non-permissive temperature (24°C)
for 16 hours. Gravid adults were bleached and L1s hatched overnight in M9. L1 progeny
were fed E. coli OP50 for 3 hours, fixed, and stained for Mab H5 to assess ZGA. B. Samples
where TOP-2 had been inactivated early as depicted in (A) were fed for 3 hours and then
stained for P-granules (white) and Mab H5 (green). Representative IF images of Z2/Z3 are
shown, and the frequency of nuclei with a positive H5 signal is shown below the images. A
minimum of 20 animals was examined per condition in each experiment. Data reveal that
early inactivation of TOP-2 does not prevent ZGA. C. Strain WMM2 was used for live cell
imaging of starved samples after incubation at the permissive temperature (15°C) or after
early inactivation of TOP-2 (24°C Early). Data were quantified as in Fig. S1.2. P-value
determined by Student’s t-test, n = 7.
58

Figure S1.3 (Continued). Early inactivation of TOP-2 produces hypo-compacted
chromatin in starved L1s and allows ZGA. D. A model proposing that TOP-2 is
required for Z2/Z3 chromatin compaction in the embryo and also for chromatin
decompaction in feeding L1s.

59
Table 1.1. Summary of RNA-FISH analysis of ZGA gene
expression. Related to Figures 1 and 2.
FISH-positive
nuclei
Transcript Condition top-2
allele
Temperature Exp. 1 Exp. 2
starved wild type 24° 2/20 1/20
fed wild type 24° 13/20 11/20
vbh-1 fed it7 15° 11/20 14/20
fed it7 24° 4/20 3/20
fed+IR it7 24° 12/20 12/20
starved wild type 24° 2/10 2/20
fed wild type 24° 18/20 19/20
ifet-1 fed it7 15° 12/20 11/20
fed it7 24° 5/20 3/20
fed+IR it7 24° 16/20 13/20
starved wild type 24° 0/20 0/20
fed wild type 24° 12/20 14/20
car-1 fed it7 15° 14/20 17/20
fed it7 24° 2/20 4/20
fed+IR it7 24° 11/20 11/20
starved wild type 24° 0/20 5/20
fed wild type 24° 19/20 17/20
wago-1 fed it7 15° 19/20 20/20
fed it7 24° 4/20 6/20
fed+IR it7 24° 17/20 16/20
starved wild type 24° 3/20 1/20
fed wild type 24° 19/20 19/20
cgh-1 fed it7 15° 16/20 13/20
fed it7 24° 3/20 2/20
fed+IR it7 24° 16/25 13/25

60
Table 1.2. Summary of RAD-51 and MabH5 foci overlap. Related to Figure 2.
Time
post-
feeding
#
nuclei
scored
# RAD-
51 foci
scored
#
MabH5
foci
scored
# RAD-51
MabH5
overlap
% RAD-51
foci
overlap
with
MabH5
foci
% MabH5
foci
overlap
with RAD-
51 foci
15 min. 12 31 14 2 6.4 14.3
30 min. 15 58 16 3 5.2 18.7
45 min. 16 82 29 12 1.4 41.4
60 min. 23 102 28 9 8.8 32.1
Total 66 273 87 26 9.5 29.9

61

CHAPTER II

Chromatin Compaction via TOP-2 and Condensin II Mediates Transcriptional
Repression in Z2/Z3 during Embryogenesis

62
INTRODUCTION
The previous chapter showed that we identified TOP-2 induced DNA DSBs and
chromatin decompaction via RUVB proteins as a mechanism for zygotic genome
activation. These studies elucidate how cells rapidly reengage transcription and
reenter the cell-cycle after long periods of transcriptional arrest. The mechanism by
which these same cells prior exit the cell cycle and maintain transcriptional
quiescence, however, is equally compelling and worthy of investigation. To study this
paradigm and counterpart to cell cycle reentry, we once again employed C. elegans
germ cells, however this time, in the context of the early embryo. Importantly, these
cells are transcriptionally repressed during embryogenesis, and this essentially
protects the germline from ectopic gene expression and consequential germline loss.
The embryonic germline of C. elegans, as a tool to study transcriptional
repression, has broad implications in mammalian systems as well as several types of
human cancers (Kirienko et al., 2010). Recent studies have shown that the
misexpression of early germline genes can potentially drive tumorigenesis
(McFarlane et al., 2014; Janic et al., 2010; Akers et al., 2010). For example, some tumor
cells display protein antigens shared by the germline and expressed early in
developing germ cells (Akers et al., 2010). In a more direct example linking ectopic
germline transcription to cancer, researchers have found that human chorionic
gonadotropin is produced in germ cell tumors in the brain (Takami et al., 2015).
These studies present the potential for anti-tumor therapeutics that target germ cell-
expressed antigens (McFarlane et al., 2014). Therefore, we must continue to expand
63
our understanding of how inappropriate germline gene expression during early
development occurs.
Most of what is currently known about transcriptional regulation in the
developing germline is derived from studies in C. elegans (Reinke et al., 2013). As an
example, the discovery that active transcription is repressed in embryonic germ cells
is credited to Seydoux and Dunn, who used mAb H5 to detect active RNAPII in C.
elegans and Drosophila embryos (Palancade and Bensaude, 2003). The absence of H5
in germ nuclei and presence in somatic nuclei demonstrated that embryonic germ
cells are incapable of synthesizing new mRNAs (Seydoux and Dunn, 1997).
Interestingly, immunofluorescent signals for mAb H14, which detect the
phosphoserine 5 residue of the CTD of RNAPII and thus corresponds to transcription
initiation, were present in germ precursors P2-P4, suggesting that in these cells,
transcriptional elongation is blocked after initiation (Ghosh and Seydoux, 2008).
Since RNAPII is thereby effectively paused at initiation in embryonic germ cells,
RNAPII suppression is necessary to inhibit premature elongation (Bowman et al.,
2013).
At least two distinct mechanisms globally repress the transcription of P cells
at the level of RNAPII inhibition (Wang and Seydoux, 2013). First, in P0 and P1, OMA-
1 and OMA-2 bind to TAF-4 in the cytoplasm to prevent TAF-4 from assembling on
the TFIID complex for RNAPII preinitiation (Guven-Ozkan et al, 2008). In P2-P4,
however, PIE-1 stalls progression from transcriptional initiation to elongation by
competitively inhibiting phosphorylation of the Ser2 residue of the CTD of RNAPII
(Batchelder et al., 1999). Numerous studies using pie-1 mutants have demonstrated
64
its loss results in premature transcription and the production of only somatic
descendants (Mello et al, 1992). This consequence underscores the importance of
germline suppression, as a loss of the germline is, at least from a holistic perspective,
equivalent to extinction.
Despite these well-established mechanisms for transcriptional repression in
the germline during early embryogenesis, a direct mechanism for quiescence in
embryonic Z2/Z3 is largely unknown. Currently, it is postulated that after PIE-1
degrades, Z2/Z3 remain repressed due to the remodeling of chromatin via histone
modifications (Schaner and Kelly, 2006). Previous works have identified that there
are three specific histone modifications that change in Z2/Z3 at their birth (Schaner
et al., 2003). H3K4me2/3 and H4K8ac, markers generally associated with
transcriptional competence, are present in germline blastomeres P0-P4, but disappear
in Z2/Z3 after PIE-1 degradation (Schaner et al., 2003). These markers are distinct
from H3K27 (di- and trimethylation), which correlate with gene silencing on the X
chromosomes via the MES-2/3/6 complex and increase the transcriptionally silent X
chromosome (Strome, 2005). Where autosomal repression in Z2/Z3 is concerned,
however, the H3K36 methyltransferase MES-4 maintains decreased levels of
H3K4me2, thus contributing to transcriptional repression until hatched larvae begin
feeding and RNAPII is coincidently activated (Butuci et al., 2015a,b; Bender et al.,
2006; Furuhashi et al., 2010). The elaborate dynamics of these histone modifications
ultimately suggest that RNAPII silencing in Z2/Z3 is critically dependent on
chromatin architecture.
65
As previously described, transcription is directly repressed via OMA-1/2 and
PIE-1 in embryonic P-cells, and indirectly suppressed by histone modifications in
embryonic Z2/Z3. Still, little is known about how the latter mechanism maintains
RNAPII inactivity outside of changes in the histone code. Given our previous work
establishing a function of TOP-2 to activate transcription via DNA DSBs and
chromatin decompaction in feeding L1s, we began to pursue whether TOP-2 has a
role for transcriptional quiescence of embryonic germ cells. In that study, when TOP-
2 was inactivated early in the parental generation, signals for RNAPII H5 were
unaffected but RAD-51 foci were reduced in fed L1 progeny. However, when we
waited to inactivate TOP-2 until after the progeny hatched, both H5 and RAD-51
signals in these fed L1s were significantly attenuated. Thus, while we recognized TOP-
2 for its enzymatic ability to induce DNA DSBs for chromosome decompaction, a
second function of TOP-2 was necessary to rationalize these paradoxical phenotypes.
In this effort, we turned our attention towards the role of TOP-2 in chromatin
compaction.
TOP-2 has long been known to be required for chromosome condensation
during mitosis, however its exact relationship with the condensin complex was not
identified until more recently (Uemura et al., 1987, Leonard et al., 2015). The
condensin protein complex incudes both condensin I and condensin II which are
required for mitotic chromosome condensation (Csankovszki et al., 2009; Ford and
Schumacher, 2009). Condensin I localizes in the cytoplasm during interphase and
thus interacts with chromosomes only after nuclear envelope breakdown (Hirano,
2012). In contrast, condensin II associates with chromosomes in early mitotic
66
prophase to condense chromatin prior to division (Hirota et al., 2004), and thus
condensin II is the major contributor to the chromatin compaction of embryonic
Z2/Z3 prior to cell-cycle exit. Separately, the function of TOP-2 is to relieve the
entaglements and torsional stress created by DNA replication or transcription
(Haffner et al., 2011; Teves and Henikoff, 2014). Here, TOP-2 makes DNA double
strand breaks and passes the uncut strand through to resolve the catenases, and
immediately re-ligates the DNA (Furniss et al., 2013). In this regard, TOP-2 would
seemingly have an antagonistic relationship with condensin II; TOP-2 activity
generally facilitates DNA relaxation and condensin II essentially tightens DNA. Still,
however, proper mitotic chromatin condensation requires TOP-2, as TOP-2 responds
to condensin-driven DNA overwinding to remove entanglements from replicated
DNA (Baxter and Aragón, 2012). Thus, the molecular activities of condensin II and
TOP-2 are coordinated to fashion the condensation of chromosomes necessary for
mitosis (Leonard et al., 2015; Sutani et al., 2015). Given this relationship, the results
shown in this chapter investigate how TOP-2 and condensin might regulate Z2/Z3
chromatin for transcriptional repression during embryogenesis.

67
RESULTS

Early TOP-2 inactivation during embryogenesis causes hypo-condensed
chromatin in the embryo
Earlier reports on embryonic germ cell development identified that in
Drosophila, germline pole cells arrest in G2 (Su et al., 1998) and that similarly, Z2/Z3
arrest at G2 in C. elegans (Schaner et al., 2003; Fukuyama et al., 2006; Butuci et al.,
2015). Further analysis using a worm strain expressing a GFP::histone H2A fusion
protein has also characterized this G2 arrest to be associated with condensed
chromatin (Fukuyama et al., 2006). Consistent with these previous reports, we used
our strain WMM1 fused with mCherry-tagged histone H2B to examine the chromatin
status in embryonic Z2/Z3 and found the chromosomes condensed as shown in
Figure 2.1A. Like its morphology in starved L1s, we observed that the Z2/Z3
chromatin in the embryo also appeared segregated into distinct clusters, possibly
even representing the six C. elegans chromosomes themselves (Ankeny, 2001). Using
this structural patterning as a baseline, we next asked if we could visualize a
rearrangement of chromatin in the embryo if we knocked-down capg-2, an essential
subunit of condensin II (Lau et al., 2014, Csankovszki et al., 2009). As shown in Fig.
2.1B, when we fed L4s capg-2 RNAi and then subsequently harvested and live-cell
imaged capg-2(RNAi) embryos, chromatin in Z2/Z3 was significantly more diffuse
and lacked the neat chromosome clusters previously seen in condensed
chromosomes. This result indicated that we could indeed track the compaction state
of Z2/Z3 chromatin throughout embryogenesis but more importantly that condensin
68

Figure 2.1. Chromatin structures of Z2/Z3 during embryogenesis. A. Gravid
adults (strain WMM1) fed E. coli OP50 and transferred to control RNAi HT115
were hypochlorite-treated to collect embryos. Embryos were immediately live-cell
imaged using mCherry-tagged histone H2B to identify the structures of chromatin
(red) in Z2/Z3. Representative images are shown. B. Same as (A) except samples
were fed capg-2 RNAi prior to bleaching and imaging. C. Strain WMM2 L4s were
shifted to 24°C to inactivate TOP-2. Embryos were treated and imaged as in (A) and
(B) along with control samples grown at 15°C.

69
II is required for Z2/Z3 to maintain chromosome condensation throughout
embryogenesis.
The model (Fig S1.3D) identified in Chapter 1 where decompaction of
chromatin allows germline gene activation in larvae also supports germline gene
repression via chromatin compaction during embryogenesis. In other words, if TOP-
2 induced DNA DSBs are necessary for decompacting chromatin in feeding larvae, the
other TOP-2 function of chromatin condensation, may be required during
embryogenesis to compact chromatin. Our previous evidence that top-2 RNAi
advanced the appearance of RNAPII in fed L1s suggests that this function may be
indeed required. Given this result and our predicted model, we hypothesized that the
early loss of TOP-2 would cause hypo-condensed chromatin in the embryo. To test
this prediction, we used the TOP-2 temperature sensitive allele (it7ts) and shifted L4s
to the non-permissive temperature of 24°C to inactivate TOP-2, collected these TOP-
2 depleted embryos, and then live-cell imaged Z2/Z3. In the control group of adult
worms kept at the permissive temperature (15°C) the entire time, chromatin of Z2/Z3
appeared condensed (Fig 2.1C) in a similar manner to that of Fig 2.1A. However,
compared to the control, mCherry signals in TOP-2 depleted embryos were more
diffuse throughout the nucleus, indicating that chromatin was hypo-compacted after
TOP-2 inactivation (Fig 2.1D). Together the results of Figs. 2.1 indicate that both TOP-
2 and condensin II are required for the chromatin compaction associated with late G2
mitotic arrest of Z2/Z3 during embryogenesis, and that failure of either one of these
protein activities results in decondensed chromosomes.
70
TOP-2 suppresses premature gene expression in the embryo through
chromatin compaction
To track the transcriptional status of germ cells, we commonly employ
immunofluorescent staining of phosphorylated Serine 2 residues of the CTD of
RNAPII. We note that our studies use two different anti-RNAPII pSer2 antibodies,
clone H5 (Seydoux and Dunn, 1997) and Ab5095, the latter of which had not been
previously validated for immunofluorescence in C. elegans. Grown in rabbits, this
antibody gave us additional flexibility for co-staining experiments, and thus we
validated it as shown in Figure S2.1.
Given that we now know that chromosome condensation is a barrier to
transcription in larval Z2/Z3, and furthermore, that TOP-2 is required for
chromosome condensation in the embryonic Z2/Z3, we next asked whether TOP-2 is
required to prevent transcription in the germ blastomeres, P2-P4. To answer this
question, we used the strain carrying the top-2 (it7ts) ts allele and moved L4s from
the permissive temperature (15°C) to the non-permissive temperature (24°C) for 16
hours until adulthood to deplete their progeny of TOP-2. In this experiment, shifting
young adults to 24°C was an adequate procedure since we were immunostaining
embryos, and thus embryonic viability was of little concern. After we stained embryos
for Ser2-P, P2, P3, and P4 cells identified by P-granules were devoid of any Ser2-P
signal at the permissive temperature, as we expected. In top-2 inactivated embryos,
however, we detected positive signals for Ser2-P as shown in Figs. 2.2A&B. In contrast
to the robust H5 signals previously detected in feeding larvae, however, the Ser2-P

71

Figure 2.2. RNAPII elongation in top-2 (it7ts) inactivated embryonic
germ cells. A. Representative IF images of P4 in embryos grown at the
permissive temp. (15˚C) and non-permissive temp. (24˚C). Embryos were
stained for Ser2-P (red) and P-granules (green). B. Graph depicting
percentages of P2-P4 cells in embryos with a signal for Ser2-P.

72
signals in P cells appeared spottier, suggesting a low degree of transcription. We also
detected Ser2-P in P2-P4 with similar staining patterns by using top-2 RNAi (Figs.
S2.2A&B), showing that TOP-2 loss results in early gene expression in normally
transcriptionally repressed germ cells. Together, these findings suggest that TOP-2
functions in some manner to suppress expression during embryogenesis.
Since a major role of TOP-2 is to facilitate proper mitotic chromosome
condensation, we next predicted that inappropriate activation of RNAPII was due to
the failure of TOP-2 to compact chromatin. To test this possibility, we depleted CAPG-
2 of the condensin II complex using RNAi, and then stained for Ser2-P in control RNAi
and capg-2 (RNAi) embryos (Figure 2.3A). To avoid introducing chromosome
segregation defects, the capg-2 RNAi vector was fed to animals for only 16 hours.
After treatment, capg-2 depletion indeed resulted in a Ser2-P signal present between
roughly 30-60 percent of P2-P4 cells, and by contrast, control RNAi samples showed
little to no Ser2-P activity in P cells (Figure 2.3B). These results indicate that
chromosome condensation suppresses gene expression in germ cells and that
importantly, TOP-2 is required in this coordinated effort.
As previously mentioned, the global suppression of germline gene expression
in P0-P1 is controlled by OMA-1/2, and PIE-1 in P2-P4. Importantly, their mechanisms
prevent from germ cells from adopting a somatic fate as a result of premature
transcription. Therefore, it was formally possible that our positive staining for Ser2-
P in P2-P4 of top-2 and capg-2 depleted embryos was due to an effect of the RNAi on
PIE-1. To test this possibility, we immunostained mixed embryos with PIE-1 and
looked for presence of the protein in P2, P3, P4, and Z2/Z3 after feeding animals top-2
73

Figure 2.3. RNAPII elongation in capg-2 (RNAi) embryonic germ cells.
A. Immunofluorescent staining of P4 in control RNAi and capg-2 (RNAi)
embryos using antibodies against Ser2-P (red) and P-granules (green).
Representative images are shown. B. Graph depicting percentages of P2-P4
cells in embryos with a signal for Ser2-P. A minimum of 50 samples were
examined in each biological replicate.

74
RNAi, capg-2 RNAi and control RNAi. As shown in Figure 2.4A&B, PIE-1 was present
in the cytoplasm of germ cells P2-P4 under all RNAi conditions, but absent in Z2/Z3
which was expected given that PIE-1 degrades after P4 divides. The cytoplasmic
localization of PIE-1 in the P-cells also agreed with PIE-1 immunostaining done
previously under normal conditions (Seydoux and Dunn, 1997; Daniels et al., 2012).
Therefore since PIE-1 appears to be functional, these results suggest that the ectopic
transcription associated with top-2(RNAi) (Fig. 2.2) and capg-2(RNAi) (Fig. 2.3) are
not a result of a PIE-1 defect. More importantly, these results underline that TOP-2
and condensin II suppress germline gene expression independently of PIE-1 and
ultimately through chromatin compaction.

TOP-2 mediated chromatin compaction suppresses expression of germline
genes
Out of our finding that transcription was prematurely activated upon TOP-2
and CAPG-2 depletion, we next set out to determine the gene types being
inappropriately expressed. There are at least two classes of genes that take the
spotlight during germline specification. The first are genes that are normally only
expressed in germ cells called germline genes, and the second class are somatic genes
that are normally only expressed in somatic cells. In a developing C. elegans
embryonic germ cells, both classes of genes are normally suppressed. However,
premature expression of either of these gene types may have different consequences
for the cell. For example, somatic genes are inappropriately expressed in pie-1
mutants, causing loss of the germline (Mello et al, 1992). In our studies, however, it
75

Figure 2.4. PIE-1 in top-2 (RNAi) and capg-2 (RNAi) embryonic germ cells.
A. Representative images of PIE-1 (green) and P-granules (red) in embryo after
capg-2 and top-2 RNAi treatment. B. Combined graph for two biological
replicates depicting percentages of P2-P4 cells in embryos with PIE-1. A
minimum of 50 embryos were scored in each replicate.

76
was unknown what form of premature gene expression TOP-2 and condensin II
protect against. Therefore, we sought to understand what type of genes chromatin
compaction is responsible for suppressing.
To pursue this answer, we looked specifically at the expression patterns of
germline genes in capg-2(RNAi) and top-2(it7ts) backgrounds. As opposed to PIE-1
which represses somatic genes, we predicted that TOP-2 and condensin II suppress
germline genes, since we observed that top-2 and capg-2(RNAi) germline blastomeres
gastrulated and Z2/Z3 developed normally. To test this hypothesis with a germline
gene, we profiled thousands of C. elegans genes based on their expression patterns in
embryos and larvae, sourced from at least two different databases (Tintori et al., 2016;
Shin-i and Kohara, 2005). Having generated a list of genes, we identified ppw-2 as an
ideal candidate germline gene to study. PPW-2 is an argonaute protein known for
transposon silencing (Vastenhouw et al., 2003) and importantly, expressed solely in
Z2/Z3 during embryogenesis and larval development (Spencer et al., 2011; Shin-i and
Kohara, 2005; Tintori et al., 2016).
The exclusivity of ppw-2 expression in primordial germ cells made it possible
to target using in situ hybridization, and to further test for expression in P-cells under
TOP-2 depleted conditions. We first employed chromogenic in situ hybridization
(CISH) where antisense digoxigenin-labeled RNA probes were used to bind to
endogenous RNA transcripts. These probes were then detected by an anti-
digoxigenin antibody bound to alkaline phosphatase, which was developed using
color substrates. Indeed, our CISH assays show that under control RNAi conditions,
ppw-2 transcripts are only enriched in Z2/Z3 of older embryos compared to P2 in a 4-
77
cell embryo (Fig. S2.3A). Comparatively, however, after top-2 and capg-2 were
knocked-down via RNAi, P2 stained darker via in situ analysis suggesting that ppw-2
was enriched in P2 as seen in Fig. S2.3B. As a result of the early activation of ppw-2 in
these conditions, we surmised that condensed chromatin may repress RNAPII
activation of germline genes.
Throughout the course of our in situ analysis, we recognized certain
limitations with traditional chromogenic staining, however. For example, while CISH
is useful in histology and pathology, it offers little information regarding the
transcript copy number and spatial localization of targeted genes (Jensen, 2014; Choi
et al., 2014; Gong et al., 2005). Therefore, we continued to carry out our analysis of
ppw-2 using a newer, higher resolution method called hybridized chain reaction in
situ (HCR ISH). In comparison, HCR ISH amplifies mRNA targets with fluorophores
chained to DNA probes (Choi et al., 2016). Expecting ppw-2 enrichment in Z2/Z3, we
first tested this assay in late embryos (~100 cell stage) in animals fed control RNAi
and indeed a few, punctate fluorescent signals were observable around Z2/Z3 nuclei
(Figure 2.5A). To answer whether these signals were indications of bona fide ppw-2
transcripts, we next examined ppw-2(RNAi) embryos at the same stage and after
doing so, observed no fluorescent signals (Figure 2.5B). These assays demonstrated
the feasibility with which we could use HCR ISH in C. elegans embryos.
To confirm the CISH results of Figure S2.3, HCR ISH targeting ppw-2 was also
performed on 4-cell top-2(it7ts) and capg-2(RNAi) embryos. Results in Figure 2.5A&C
show that compared to control RNAi, capg-2(RNAi) had a higher frequency of a ppw-
2 HCR signals in both biological replicates. Furthermore, these signals, which were
78

Figure 2.5. Hybridized chain reaction in situ (HCR) of ppw-2. A. Representative
images of late embryos and 4-cell fed control RNAi, ppw-2 RNAi, and capg-2 RNAi.
Embryos were assayed for ppw-2 mRNA signals (white) using HCR and P-granules
(green). B. Graph depicting frequency of Z2/Z3 in late embryos with a ppw-2 HCR signal
in control and ppw-2 RNAi-treated samples. C. Graph depicting frequency of P2 in 4-cell
embryos with a ppw-2 HCR signal in control and capg-2 RNAi-treated samples. A
minimum of 20 samples were analyzed in each condition for both experiments. D.
Representative images of 4-cell embryos assayed for ppw-2 mRNA signals using HCR
after inactivating TOP-2 at 24˚C. E. Same as Fig 2.6C but using top-2 (it7ts) cultured at
15˚C and 24˚C.

79
cytoplasmic yet distinct from P-granules, resembled those of Z2/Z3. When the top-2
T.S. allele was used, ppw-2 HCR signals were absent in 4-cell embryos at the
permissive temperature, but present at least 25% of the time in P2 at the TOP-2 non-
permissive temperature (Figs. 2.5D&E). Together, the results of these in situ
experiments suggest that chromosome condensation via TOP-2 and condensin II help
repress expression of germline genes. However, it is formally possible that somatic
genes are equally suppressed through this mechanism and therefore further analysis
on the expression of gene transcripts is required.
The work thus far has substantiated a model whereby TOP-2 and condensin II
function in parallel to compact chromatin and maintain transcriptional arrest
through the end of embryogenesis and L1 starvation. To truly assess the state of the
chromatin in Z2/Z3, however, we next asked whether regions of the genome
heterochromatinized to achieve transcriptional repression. To answer this question
we performed protein antibody staining against H3K9me3, a mark of
heterochromatin that has been specifically identified in condensed, transcriptionally
inactive cells and functions by binding to heterochromatin protein 1 (HP1) (Machida
et al., 2018). A temporal analysis of H3K9me3 across the P lineage cells and early and
late Z2/Z3 revealed that this mark for heterochromatin was present at each stage of
development. As shown in Figure 2.6, from P2 to early Z2/Z3, we found no difference
in neither localization or intensity of H3K9me3 between the germline and somatic
cells. Among later stage embryos, however, late Z2/Z3 nuclei consistently showed a
hyperaccumulation of H3K9me3 relative to all other somatic cells. The delay in
H3K9me3 hyperaccumulation of newly divided Z2/Z3 is consistent with the
80

Figure 2.6. H3K9me3 staining in embryonic P lineage cells.
Representative images of H3K9me3 (red) in P2-P4, early Z2/Z3, and late
Z2/Z3. Two biological replicates were performed.

81
previously reported transient upshifts in H5 transcription seen after their birth
(Seydoux and Dunn, 1997; Schaner et al., 2003). Importantly, this data suggests that
Z2/Z3 germline gene expression is suppressed in late embryogenesis by constitutive
heterochromatin, which is distinct from the levels of H3K27me2/3 seen in previous
work (Bender et al., 2004; Schaner et al., 2003).
In summary, these results substantiate and resolve a model suggested in the
previous chapter, where TOP-2 must create DNA double strand breaks to ultimately
activate gene expression. We now know this is because TOP-2 and condensin II co-
function in Z2/Z3 during embryogenesis to facilitate chromatin compaction and
suppress gene expression (See Chapter 1; Fig. S1.3Di-iii). This chromatin compaction
is a barrier to ZGA that TOP-2 must overcome, or otherwise be unable to activate the
germline genome during L1 feeding. Our results have furthermore showed that when
either of these functions are depleted, transcription is ectopically activated in P cells,
and this is also consistent with an advanced activation of transcriptional elongation
in feeding L1s (See Chapter 1; Fig. S1.3Dii).
82
DISCUSSION

A novel mechanism of transcriptional repression by TOP-2/condensin II
We sought to determine the role of TOP-2 and condensin II for transcriptional
repression in C. elegans embryonic germ cells. Our results show that indeed TOP-2
and CAPG-2 are both required compact chromatin, and that this repression is also
likely aided by the genome becoming entirely heterochromatin in embryonic Z2/Z3.
This chromatin compaction is not only required in Z2/Z3, but also maintains
transcriptional repression in P cells of early embryos, because loss of these functions
lead to premature transcription in P cells. We also suggest that more specifically,
germline gene expression, such as that of ppw-2, is suppressed by TOP-2 and CAPG-2.
Additionally, we find that the function of TOP-2 and condensin II to maintain
transcriptional quiescence through chromatin compaction occurs not because PIE-1
function is impaired. These findings support that chromatin compaction is a natural,
protective barrier against ectopic transcription that regulates the entire genome, as
opposed to other proteins like PIE-1 that targets RNAPII against elongation of somatic
transcripts specifically. We also note that even pie-1 mutants with precocious gene
expression fail to induce transcription prematurely in P2, whereas knockdown of the
TOP-2/condensin II interaction inappropriately activates transcription in all
germline blastomeres. To this end, we introduce genome-wide chromosome
compaction as a novel mechanism of transcriptional repression in the C. elegans
germline that is independent from the mechanism of PIE-1 repression.

83
The potential interactions of TOP-2/condensin II and the germline genome
during embryogenesis
While our work elucidates an important role for the chromosome
condensation machinery in repressing RNAPII activation in embryonic PGCs, the
mechanism by which condensin achieves suppression with TOP-2 in PGCs is still
unknown. A major related question is: where does condensin physically bind to on
chromatin to repress transcription? Given that we see germline gene expression in
PGCs when the condensation machinery is impaired, condensin II may preferentially
bind to germline genes (Kranz et al., 2013). This would further be supported by the
ectopic expression of somatic transcripts in pie-1 mutants despite having functional
chromatin compaction machinery (Mello et al., 1992). Of additional interest, it is
unknown whether TOP-2 or condensin II binds specifically at promoters of such
germline genes or rather at gene clusters. Our previous data suggests that germline
gene promoters themselves do not require direct breaks to initiate ZGA, and therefore
it is possible that condensins also need not bind directly on promoters to repress
transcription (Kranz et al, 2013; Evans et al., 2016). Still, the mechanism for germline
gene repression during embryogenesis may be so critical that it is possible that
condensin binds at promoters anyway as an evolutionarily conserved feature (Kranz
et al, 2013).
At least one potential insight into the mechanistic basis for transcriptional
repression during embryogenesis could be found in X chromosome dosage
compensation, which targets both X chromosomes of hermaphrodites to reduce their
transcript levels by half (Ercan, 2015). This dosage compensation complex (DCC)
84
regulates gene expression by attracting condensin-like proteins that are structurally
similar to mitotic condensin II proteins (Ercan, 2015). Importantly, however, the
condensins of the DCC are notably different from mitotic condensins. First DCC
condensins lack the ancillary CAPG-2 subunit of condensin II (Lau et al., 2014), which
our work shows to have repressor function for germline expression. Secondly, the
DCC recruits many other factors such as several dumpy genes to specifically repress
X chromosome gene expression, and complexes such as these remain assembled on
chromatin throughout the entire cell cycle (Meyer, 2010). Therefore, though the
mechanisms of DCC and mitotic condensins are distinct, it is possible that features of
the DCC might also govern embryonic germline repression. If so, this paradigm would
provide insight into how condensins are recruited to DNA targets and the subunits
that function as repressors during germline development.
Despite the unknown mechanisms for compaction, our observations that
Z2/Z3 seem to gastrulate normally and that even Z2/Z3 ZGA and division is advanced
when TOP-2 is depleted early then implore us to ask what is the actual purpose of
germline gene repression. The impact of chromosome condensation on the
development of other organisms sheds light on this question, as transcriptional
quiescence during prophase arrest is not limited to C. elegans germ cells. For example,
oocytes undergoing meiosis I in Drosophila also feature epigenetic modifications to
chromatin that control development and regulate gene expression. Interestingly, the
remodeling of meiotic chromosomes corresponded to premature RNAPII activation,
and consequently loss of fertilization due to meiotic abnormalities (Novarro-Costa et
al., 2016). Thus, although we observe gastrulation despite premature transcription,
85
the importance of repression in other systems highlights how chromatin
condensation may have co-evolved as a protectionary measure against inappropriate
germline activation.

Heterochromatin characterizes the entire germline genome in the late embryo
Another significance of this work is that it identifies global, constitutive
heterochromatin displayed exclusively in the late Z2/Z3 genome, and one question is
how the Z2/Z3 DNA containing H3K9me3 is maintained throughout embryogenesis
and at such high levels compared to somatic nuclei. Previous reports would suggest
that this is accomplished by HP1 (Machida et al., 2018), and if so, identifying an
interaction between TOP-2/condensin II and HP1 would shed further insight into the
mechanism by which TOP-2 aids in chromatin compaction. One other intriguing
possibility for the strong presence of H3K9me3 in Z2/Z3 is that the germline genome
uses H3K9me3 to prepare for the induction of DNA DSBs during ZGA. Here, the
histone acetyltransferase TIP60 would be activated and accumulate readily on
H3K9me3 for activating ATM (Ayrapetov et al., 2014; Sun et al., 2010), and efficient
DNA repair by this very signaling pathway is critical in Z2/Z3 during ZGA (Butuci et
al., 2015a,b). Thus, a germline genome replete of H3K9me3 might serve a two-fold
function; first, to maintain compaction with heterochromatin by HP1 and second by
efficiently facilitating DSB repair signaling.

86
MATERIALS AND METHODS

Strains
The strains used in this study were N2; Bristol wild-type, SS747; bnIs1[pie-
1::GFP::pgl-1 + unc-119(+)], KK381; unc-4(e120) top-2(it7ts) II, WMM1; bnIs1[pie-
1::gfp::pgl-1 + unc-119(+)]; ltls37 [(pAA64) pie-1p::mCherry::his-58 +unc-119(+)] IV,
and WMM2; ltls37 [(pAA64) pie-1p::mCherry::his-58 +unc-119(+)] IV; unc-4(e120)
top-2(it7ts) II. Strains were maintained on NGM E. coli OP50 plates using standard
culture techniques.

Antibodies and dilutions
P-granules: mouse mAb K76, from DSHB, was used at 1:5. RNAPII pSer2: rabbit
polyclonal antibody Ab5095, from Abcam, was used at 1:50. PIE-1: goat polyclonal
antibody sc-9245, from Santa Cruz Biotechnology, was used at 1:50. H3K9me3: rabbit
monoclonal antibody Ab176916, from Abcam, was used at 1:2000. Secondary
antibodies: Alexa Fluor conjugated secondary antibodies were purchased from
Invitrogen and used at 1:200.

Immunofluorescent Staining of C. elegans embryos
Embryos were washed 3 times with Milli-Q water, immediately mounted on a poly-L-
lysine glass slide, and then covered with a coverslip. Embryos were freeze-cracked by
placing the slide on a dry ice block for 10 minutes and flicking off the coverslip. Slides
were then immersed in 100% methanol at -20°C for 10 seconds and incubated in a
87
fixing solution (0.08M Hepes pH 6.9, 1.6mM MgSO4, 0.8mM EGTA, 3.7%
formaldehyde, 1X phosphate-buffered saline) for 10 minutes. Slides were washed in
Coplin jars with 50 mL of TBS-T three times for 10 minutes each. After rinsing, 50 µL
TNB was added directly to the slides for blocking for 2 hours. A primary antibody
solution was then directly applied and slides were incubated overnight at 4°C in a
humidity chamber. The primary antibody solution was washed 3 times for 10 minutes
in Coplin jars and a secondary antibody solution was added directly to each slide to
incubate for 2 hours. The secondary antibody was removed with three washes of TBS
for 10 minutes. Slides were mounted with DAPI counterstain and sealed with a
coverslip. We found that this freeze-cracking technique could also be utilized with L1s
without adversely affecting worm morphology.

Live-cell imaging of C. elegans embryos
Gravid adult hermaphrodites were suspended in M9, spun down, and 10 µL was
transferred onto a glass coverslip. Embryos were dissected from hermaphrodites by
using a 25-gauge syringe needle to make a cut in the worm near the spermatheca.
Multiple adults were dissected until a sufficient quantity of embryos were released
onto the coverslip. Coverslips containing embryos were mounted on a 3% agarose
pad. Samples were immediately imaged using an Olympus Fluoview FV1000 confocal
microscope and analyzed in ImageJ with no additional processing software.

88
Chromogenic in situ hybridization (CISH)
A method of CISH was adapted from Broitman-Maduro and Maduro’s “In situ
Hybridization of Embryos with Antisense RNA Probes.” RNA antisense probes were
generated by ordering DNA oligonucleotides with a T7 RNA polymerase recognition
sequence (IDT DNA) and then by in vitro transcription using a DIG RNA labeling kit
(Roche #1175025). To prepare samples, embryos were collected from gravid adults
via hypochlorite treatment and then permeabilized using standard freeze-cracking
techniques. Samples underwent a hydration series of methanol for 5 minutes each at
room temperature (100%, 90%, 70%, 50% methanol and DEPC-ddH2O) before
incubation in a non-toxic fixative (3% bronopol, 3% diazolidinyl urea, 1.2% zinc
sulfate heptahydrate, 0.3% sodium citrate) for 1 hour at 37°C. Samples were rinsed
with DEPC-ddH2O and 2xSSC twice each for 5 minutes and prehybridized in
prehybridization buffer (PB) (1xSCC, 10% dextran sulfate, 50% formamide, 0.4%
0.5M EDTA, 2% Denhardt’s Solution, 10% salmon sperm DNA) at 42°C for 1 hour.
RNA probes were diluted 1:500 in PB and heated to 65°C for 5 minutes. Diluted
probes were added to the slides and incubated overnight at 42°C. The next day,
samples were rinsed with 2xSSC and formamide buffer twice each for 5 minutes at
42°C, followed by two rinses each at room temperature in 2xSSC and Tris-NaCl.
Samples were blocked in blocking buffer (BB) at 37°C for 30 minutes before anti-
digoxigenin AP Fab fragments in BB was added to the slides for 3 hours at 37°C.
Samples were rinsed three times in Tris-NaCl for 10 minutes each, and then incubated
overnight in a developer solution (TNM, PVA, NBT, BCIP, levamisole) at room
89
temperature in the dark. Samples were rinsed the next day twice with TN-EDTA
before mounting with DAPI counterstain.

Hybridized chain reaction in situ hybridization (HCR ISH)
HCR ISH was adapted from the C. elegans larvae protocol (Choi et al., 2016) for use in
C. elegans embryos. Embryos were collected by hypochlorite treating gravid and were
permeabilized using standard freeze-cracking. Slides were washed 3x in PBS-T and
incubated in 50% PBS-T and 50% probe hybridization buffer for 5 minutes at room
temperature. Samples were then covered with 300µL probe hybridization buffer at
37°C for 1 hour. Probes ordered from IDT DNA were prepared in hybridization buffer
pre-heated to 37°C at 1 pmol. 200µL of probe solution was added to the samples and
allowed to incubate overnight at 37°C. Excess probes were removed the following day
using a probe wash buffer at 37°C and then using 5xSCCT. Embryos were pre-
amplified on slides with 150µL of amplification buffer for 30 minutes at room
temperature. 15pmol of fluorescent hairpins supplied by Molecular Instruments, Inc.
were snap-cooled by heating to 95°C for 90 seconds before cooling at room
temperature for 30 minutes. Hairpins were added to 100µL of amplification buffer
and the solution was added to the slides for an overnight incubation at room
temperature. Excess hairpins were removed by washing with 5xSSCT at room
temperature. Amplified hairpins were fixed with 4% PFA and then washed with PBS
and mounted using DAPI counterstain.

90
RNAi
E. coli HT115 transformed with pL4440 was used as a control RNAi in all experiments.
RNAi plasmids containing capg-2 and top-2 were obtained from the Ahringer RNAi
feeding library. RNAi vectors were verified by Sanger sequencing, cultured, and
seeded on NGM+IPTG plates for induction by feeding. The following describes our
analysis of RNAi efficacy of each bacteria.
capg-2: P0 L1s were plated onto NGM plates and fed E. coli OP50 for 36 hours at 24°C
until L4 stage. L4s were then transferred to IPTG plates containing capg-2 RNAi for
16 hours at 24°C to complete adulthood. Gravid adults were bleached and F1 embryos
were used for experimentation. Embryonic viability was measured by counting
hatched L1s the following day, and only ~9.5% of samples hatched under these
conditions.
top-2: L1s were plated onto NGM plates and fed E. coli OP50 for 48 hours at 24°C until
worms were young adults. Worms were then transferred to IPTG plates containing
top-2 RNAi for 9 hours at 24°C until gravid. Under these conditions, embryonic
viability was typically measured at 10%.

Mutant analysis
top-2: Strain KK381 top-2(it7ts) was analyzed for TOP-2 function. Prior to
embryogenesis, L4s were shifted from a permissive temperature (15°C) to a non-
permissive temperature (24°C) inactivating TOP-2. These animals yielded an
embryonic lethality of ~96%, however, only embryos were examined in these
91
experiments. This TOP-2 allele had also been previously fully characterized with
similar results (Jaramillo-Lambert et al., 2016).

Imaging
All samples were imaged on an Olympus Fluoview FV1000 confocal microscope using
Fluoview Viewer software. To achieve consistent exposure levels, High Voltage lasers
were maintained at a tight range between samples and conditions in each experiment.

Statistical analysis and presentation of figures
All figure data were obtained by independently performing two biological replicates.
Due to day-to-day differences in a feeding time-course, the results of both replicates
were shown in lieu of error bars, and thus we were able to show that trends were
similar between biological replicates. The images presented in the figures were
captured on an Olympus Fluoview FV1000 and were processed through Fluoview
Viewer software without any further processing. All images were saved and exported
as TIFF files.

92

Figure S2.1. Validation of Ser2-P antibody (ab5095) for
immunofluorescence in C. elegans. IF images of mixed embryos stained for
Ser2-P (red) and P-granules (green) under various conditions and treatments.
L1s were fed control RNAi (left panel) or ama-1 RNAi (middle panel) to deplete
the large subunit of RNAPII. Control RNAi worms were also treated with calf
intestinal phosphatase (right panel) to dephosphorylate RNAPII. All samples
were stained with Ser2-P antibody from Abcam (ab5095) to detect serine 2
phosphorylation.

93
Figure S2.2. Gene expression in top-2 (RNAi) depleted embryonic germ cells.
A. Representative images of P3 in early embryos immunostained for Ser2-P (red)
and P-granules (green) after treatment with control RNAi and top-2 (RNAi). B.
Graph depicting percentages of P2-P4 cells in embryos with a signal for Ser2-P in
control RNAi and top-2 (RNAi) samples. C. Representative image of P2 in a top-2
(RNAi) depleted embryo assayed for ppw-2 mRNA (white)using HCR ISH and P-
granules (green). For control RNAi sample see Fig. 2.5A. D. Graph depicting
percentage of P2 and Z2/Z3 with ppw-2 HCR ISH signal in control RNAi and top-2
(RNAi) embryos.

94

Figure S2.3. ppw-2 expression using chromogenic in situ hybridization (CISH).
A. Representative images of control RNAi embryos at ~100-cell stage (Z2/Z3) and
4-cell stage (P2) after fixation and detection of ppw-2 mRNA using antisense RNA
probes. Arrows point to germ cells. B. Same as (A) except adults were fed top-2
RNAi and capg-2 RNAi. Only 4-cell embryos are shown. Arrows point to P2.

95

CHAPTER III

The Role of the G2/M DNA Damage Checkpoint in the Cell Cycle Exit of
Embryonic Z2/Z3

96
INTRODUCTION
In many developing organisms, proper oocyte maturation is hallmarked by
timely germline arrest. Mammalian oocytes, for example, exit the cell cycle in late G2,
and can remain arrested for several years in meiotic prophase in order to maintain
genome integrity (Lincoln et al., 2002; Solc et al., 2010; Lesch and Page, 2012). These
studies reveal that cell cycle checkpoints, such as that of the G2/M transition, control
germline development. Thus, while we have discussed in detail how C. elegans
germline proliferation is regulated by TOP-2 and chromatin compaction and
decompaction in both embryos and larvae, it is also possible that Z2/Z3 development
is controlled through the molecular processes of the cell cycle. In general, the cell
cycle and its regulatory proteins are well-established across multiple model
organisms. Still, a collective model for how primordial germ cells arrest and re-enter
the cell cycle has yet to be fully established, and we believe understanding Z2/Z3 cell
cycle exit would provide a paradigm that could be swiftly tested in other systems.
The G2/M DNA damage checkpoint is a compelling mechanism for C. elgeans
Z2/Z3 arrest during embryogenesis for at least two reasons. First, Z2/Z3 exit G2
shortly after S-phase, so the checkpoint kinase, CHK-1, may be activated. Second,
when the maternal regulator PIE-1 is lost after P4 divides, Z2/Z3 experience a brief
period transcriptional elongation (Mello et al., 1996; Seydoux and Dunn 1997;
Schaner et al., 2003). We now know that a sudden shift from transcriptional silence
to transcriptional activity has the capacity to generate damage and activate the DNA
damage checkpoint, even in the germline (Tadros and Lipshitz, 2009; Butuci et al.,
2015).
97
Our previous work supporting an upregulation of both CHK-1 and cdc-25.1
mRNA during feeding demonstrate that Z2/Z3 ZGA utilizes cell cycle proteins and
DNA damage for re-entry (Butuci et al., 2015). Thus in this study, we hypothesized
that similar mechanisms, albeit with the opposite goal of cell cycle exit, may be at play
during embryogenesis. CHK-1 is a master regulator of the DNA damage checkpoint
that controls entry into mitosis during the G2/M transition (Smits and Gillespie,
2015). In response to damage, most often induced by replication stress, C. elegans
ATL-1 phosphorylates CHK-1 on Serine344 (Aoki et al., 2000; Lee et al., 2010) and
phospho-CHK-1 (P-CHK-1) blocks cell cycle progression occur in multiple fashions.
For example, P-CHK-1 phosphorylates CDC-25.1 to prevent dephosphorylation of
CDK-1 (Donzelli and Draetta, 2003; Kalogeropoulos et al., 2004;). Additionally, P-
CHK-1 is well-known activator of WEE-1, which opposes CDC-25.1 activity (Lee et al.,
2001). A brief but detailed review of this protein network that controls cell division
timing and direction is important for understanding where Z2/Z3 may be arrested in
the cell cycle.
A central regulatory mechanism preventing uncontrolled cell division is the
activity of two primary classes of molecules called cyclins and cyclin-dependent
kinases (CDKs). Separately, these molecules are inactive. However, when cyclins and
CDKs form a heterodimer, the kinase function of CDK phosphorylates downstream
targets to initiate the next phase of the cell cycle (reviewed by Malumbres, 2014). To
this end, the activation or inactivation of CDKs determines advancement through the
cell cycle. The genes for these regulatory proteins are almost all conserved among
eukaryotes, and therefore the C. elegans homologs have been useful for studying cell
98
cycle control (van den Heuvel, 2005). In the somatic cell cycle, CDK-4 and Cyclin D1
(CYD-1) promote progression through the first gap phase, G1, whereas binding
between CDK-1 and the Cyclin A and B subfamilies are required for the transition
from G2 to mitosis (Koreth and van den Heuvel, 2005; Van der voet et al., 2009). As is
the case in many developing metazoans, the earliest cellular divisions in the C. elegans
embryo, including those of P-cells, lack G1 and G2 gap phases to rapidly proliferate
(Kermi et al., 2017). The gap phases become more ubiquitous in the later embryo as
cell cycles lengthen, though the precise timing of G1 and G2 establishment in each
lineage remains unknown (Edgar and McGhee, 1988). What is known, however, is
that in the germline, Z2/Z3 undergo replication, and then condense their
chromosomes before exiting the cell cycle presumably in G2 (Fukuyama et al., 2006;
Furuhashi et al., 2010). That is why this chapter focuses primarily on the activities of
the maturation promoting-factor (MPF) at the G2/M transition.
A close look at the well-established mechanisms of CDK-1 and cyclins is
necessary to address the source of cell cycle arrest and re-entry in Z2/Z3. The gene
cdk-1 was discovered with cell division cycle cdc-2 in fission yeast, and encodes a
highly conserved 34kDa protein (Nurse, 1990). Previous work in C. elegans has found
that CDK-1 is necessary for mitosis in all cells, as larval cells fail to divide in cdk-1
mutants, and cdk-1 knockdown via RNAi prevents germline cell division even in a
one-cell embryo (Boxem et al., 1999). In the early embryo, cdk-1 is maternally
deposited (van den Heuvel, 2005) and during the G2/M transition of older cells,
activated CDK-1 translocates from the cytoplasm to nucleus to phosphorylate its
targets (Jackman et al., 2003). However, while CDK-1 is essential and ubiquitous, it
99
does not function independently; it also requires cyclin B accumulation, other CDK-
activating kinases (CAKs), and the removal of inhibitory signals. These modifications
ultimately allow time for the cell to reach its critical size or respond to DNA damage
(Boxem, 2006).
During embryogenesis in C. elegans, only the B-type cyclins CYB-1 and CYB-3
bind to CDK-1 to promote mitosis (Van der voet et al., 2009). Both CYB-1 and CYB-3
are expressed throughout the cytoplasm and nucleus and, like all B-type cyclins,
accumulate in G2 (Bailly et. al., 1992; Gavet and Pines, 2010). Binding between Cyclin
B and CDK-1 occurs with equimolar amounts of protein, is ATP-dependent, and may
require T161 phosphorylation between CDK-1 and cyclin B (Desai et al., 1992). Once
dimerized, active sites of CDK-1 are altered to facilitate substrate binding. In addition
to controlling CDK-1 activation, B type cyclins also impart specificity (Van der voet et
al., 2009). For example, though both CYB-1 and CYB-3 are required for mitotic entry,
only CYB-3 facilitates mitotic chromosome condensation during interphase in the
early embryo and promotes anaphase by disengaging the spindle assembly
checkpoint (Deyter et al., 2010). Further work has shown that CYB-3 is required for
both replication and mitosis in the early embryo to control cell cycle timing (Michael,
2016). Thus, given its multiple roles during embryonic development, it is possible
CYB-3 expression and distribution might also be significant during PGC arrest.
The precise mechanism of activation between CDK-1 and cyclin B upon
association remains elusive. However, we do know from prior in vivo studies in both
Xenopus laevis and C. elegans that pre-formed cyclin B-CDK-1 complexes involve
additional regulation (Boxem, 2006). To this end, the activity of cyclin B-CDK-1
100
complexes are precisely controlled. A central mechanism to this idea are the
antagonistic phosphorylation and de-phosphorylation activities of WEE-1 and CDC-
25, respectively. WEE-1 is a negative mitotic regulator that blocks prophase by
phosphorylating Thr-14 and Tyr-15 of CDK-1 and is expressed widely in both the
embryo and germline (Burrows et al., 2006; Lamintina and L’hernault, 2002).
However, of the C. elegans WEE-1 kinase family comprising Myt1/Wee1 homologs
wee-1.1, wee-1.2, and wee-1.3, only wee-1.3(RNAi) causes a defect in cell division and
counteracts a loss of the germline in a cdc-25.1 null mutant (Detwiler et al., 2001;
Yoon et al., 2012). These findings suggest that WEE-1.3 specifically helps govern
germline proliferation at the G2/M transition.
Given that pre-formed complexes of cyclin B-CDK-1 are inactivated by WEE-
1.3 phosphorylation in C. elegans, CDC-25 dephosphorylation is certainly a
requirement for the cell to complete prophase (Burrows et al., 2006). The C. elegans
CDC-25 family is comprised of four homologs, cdc-25.1, cdc-25.2, cdc-25.3, and cdc-
25.4 and it is hypothesized that each function during a different developmental stage
(Ashcroft et al., 1999). Despite the variability, all cdc-25 genes are catalytically
identical and because CDC-25 dephosphorylates Thr14 and Tyr15, it is known as a
dual specificity phosphatase (Kristjánsdóttir and Rudolph, 2004). Several findings
identify cdc-25.1 to be involved in both embryonic development and germline
proliferation. First, while expression of cdc-25.1 is high in the adult germline, the
protein initially appears maternally in the embryo and its depletion by RNAi induces
embryonic arrest (Ashcroft et al., 1999). Second, during post-embryonic development,
cdc-25.1 loss-of-function mutants suffer an inability to divide in the germline thus
101
causing sterility (Ashcroft and Golden, 2002). Additionally, CDC-25.1 counteracts and
targets WEE-1.3 phosphorylation specifically, as depletion of both genes restores PGC
division (Yoon et al., 2012). Ultimately, these works suggest that cdc-25.1 is required
for germline proliferation and maintenance in both early embryonic and post-
embryonic cell cycles.
Another conserved feature of cyclin B-CDK-1 cell cycle regulation in
metazoans is the CDK-activating kinase (CAK), Cdk7. In Xenopus and Drosophila, Cdk7
promotes cell cycle progression passed G2 by phosphorylating Thr161 in the T-loop
of Cdk1 (Liu and Kipreos, 2000). Additionally, Cdk7 is a component of the
transcription factor TFIIH, which phosphorylates Ser5 in the CTD of RNAPII to
activate transcriptional initiation (Boxem, 2006; Glover-Cutter et al., 2009). C. elegans
possess a Cdk7 homolog, cdk-7, whose RNAi has previously been shown to block both
transcriptional initiation and elongation, as well as block all embryonic cell divisions
at the one-cell stage (Wallenfang and Seydoux, 2002). Additionally, cdk-7 mutants
have lengthened cell cycle times of interphase and mitosis, ultimately suggesting that
cdk-7 in C. elegans activates CDK-1 in similar fashion as other systems (Wallenfang
and Seydoux, 2002). To date, however, little is known about the role of cdk-7 in
germline transcription during ZGA and almost nothing is known about how cdk-7
could impact PGC arrest at G2/M.
To understand how each of these proteins might govern Z2/Z3 cell cycle exit
at the G2/M checkpoint, we must look at processes downstream of cyclin B-CDK-1.
The current literature identifies that this complex targets at least 75 substrates for
phosphorylation in S. cerevisiae at various levels of chromosome condensation,
102
cohesion, and nuclear envelope breakdown (Enserink and Kolodner, 2010; Blethrow
et al., 2008). For example, through a network of kinases, Cdk1 phosphorylates Plk1
and Aurora A, which critically associate with the mitotic spindle (Rhind and Russel,
2012). Additionally, Cyclin B-CDK-1 is known to phosphorylate subunits of the
anaphase promoting complex (APC), which allows separation of sister chromatids
prior to division, as well as providing feedback responses for Cyclin B degradation
(Kraft et al., 2003; Rhind and Russel, 2012). Despite its conserved roles, an exhaustive
list of specific cyclin B-CDK-1 phospho-targets in C. elegans has yet to be established,
although in vitro and in vivo studies suggest that CDK-1 activates PLK-1, required for
nuclear envelope breakdown, through phosphorylation of SPAT-1 (Thomas et al.,
2016). These findings, combined with cyb-3(RNAi) delaying mitotic spindle
separation, highlight only a fraction of the roles cyclin B-CDK-1 phosphorylation have
on mitotic progression in C. elegans (Deyter et al., 2010; Rhind and Russel, 2012). To
date, however, it remains unknown which of these mechanisms, if any, precisely
underlie the Z2/Z3 cell cycle exit.

103
RESULTS

DNA damage checkpoint fails to stay activated in Z2/Z3 in the late embryo
As previously mentioned, the status of the mitotic checkpoint in Z2/Z3 after
their birth can provide further insight into the mechanism of Z2/Z3 cell cycle arrest
during embryogenesis. To begin this investigation, we sought whether CHK-1 was
activated via phosphorylation in response to DNA damage. To test this, we used
immunofluorescence by staining embryos with an antibody against active CHK-1
phosphorylated on Ser344 (Lee et al., 2010). Given our previous work that showed P-
CHK-1 not present in starved L1s prior to ZGA (Butuci et al., 2015a,b) it was unlikely
that a chronic cell cycle checkpoint maintained Z2/Z3 arrest through embryogenesis.
On the other hand, however, we also hypothesized that P-CHK-1 may be activated at
the birth of Z2/Z3, since those cells would have recently completed S-phase before
cell cycle exit. Indeed, this is precisely what we observed. In early PGCs, P1-P4, a P-
CHK-1 signal was frequently present. As shown in Fig. 3.1A, a strong P-CHK-1 signal
appeared around a perinuclear region in P4 consistent with previous reports of
activated P-CHK-1 (Butuci, et al. 2015a,b; Jaramillo-Lambert et al., 2010). However,
among Z2/Z3, a P-CHK-1 signal was less frequent, appearing in only 16 percent of the
total Z2/Z3 samples (Fig. 3.1B). To address whether those Z2/Z3 with activated P-
CHK-1 were related to damage via DNA replication, we next scored for P-CHK-1 using
morphology to distinguish younger embryos, of which Z2/Z3 were recently born,
from older embryos. Here, younger embryos with less than 100 cells are referred to
as “early Z2/Z3,” and older embryos with more than 100 cells are referred to as “late
104

Figure 3.1. Detection of activated Phospho-CHK-1 in PGCs during early and late
embryogenesis. A. Representative IF images of P4 and Z2/Z3 in early (<100 cells)
and late embryos (>100 cells) stained for P-CHK-1 (red) and P-granules (green). B.
Graph depicting percentages of P-CHK-1 present between P4 cells and Z2/Z3. A
minimum of 50 samples were examined in each biological replicate. C. Graph
depicting percentages of P-CHK-1 in Z2/Z3 between embryos with early and late
Z2/Z3. A minimum of 40 samples were examined in each biological replicate.

105
Z2/Z3.” Using this analysis, we consistently observed that 48% of early Z2/Z3
samples had a P-CHK-1 signal, whereas late Z2/Z3 samples failed to show any P-CHK-
1 (Fig. 3.1A&C). Here, the drastic reduction of P-CHK-1 in late Z2/Z3 is likely because
the checkpoint is no longer necessary as the genome compacts in late G2. However, it
is also formally possible that chk-1 mRNA fails to be maintained when PIE-1 degrades
after P4 division (Tenenhaus et al., 2001). Ultimately, these results indicate that
another mechanism besides that of P-CHK-1 maintains Z2/Z3 cell cycle arrest
through embryogenesis.

CDK-1 inactivation does not occur in embryonic Z2/Z3
Despite a lack of persistent P-CHK-1 in embryonic Z2/Z3, the possibility
remained that Z2/Z3 mitotic progression was halted by an inactivated MPF. A
hallmark of prophase arrest during the meiotic development of mammalian oocytes
is cyclin B-CDK-1 inactivation by WEE-1 phosphorylation (Han et al., 2005; Carroll
and Marangos, 2014). In this system, WEE-1 phosphorylates CDK-1 on its tyrosine
residue (Tyr 15) of the ATP binding site (Carroll and Marangos, 2014; Potapova et al.,
2009), and thus we hypothesized that Z2/Z3 would show constitutive
phosphorylation of CDK-1. To answer this, we again turned to immunofluorescence,
this time to detect phospho-CDK-1 (P-CDK-1) in the PGCs of the embryo. Using a p-
Cdc2 antibody recognizing Tyr 15 phosphorylation, we stained embryos and
observed that the P-CDK-1 signal frequently appeared in P1-P4 nuclei, but was
noticeably absent in both early and late Z2/Z3 (Fig. 3.2A). Specifically, P-CDK-1 was
observed in P1-P4 approximately 60-80 of the time, compared to zero percent in
106

Figure 3.2. Detection of Phospho-CDK-1 in PGCs during early and late
embryogenesis. A. Representative IF images of P1-P4, young Z2/Z3 (<100
cells), and late Z2/Z3 (>100 cells) stained for P-CDK-1 (red) and P-granules
(green). B. Graph depicting percentages of P-CDK-1 present between PGCs. A
minimum of 25 samples were examined in each biological replicate.

107
Z2/Z3 (Fig. 3.2B). While a P-CDK-1 signal in P1-P4 was expectedly normal following
replication (Norbury et al., 1991), the rationale for no P-CDK-1 after P4 division was
more perplexing. We note that a similar experiment was recently published by the
Yanowitz lab after our attempts, and those results agreed with ours (Mainpal et al.,
2015). There are at least two distinct possibilities for the lack of P-CDK-1, despite that
Z2/Z3 arrest at G2. It is either possible that a cyclin B-CDK-1 complex never formed
in Z2/Z3, or CDK-1 was never phosphorylated due to the loss of WEE-1. Regardless,
Z2/Z3 arrest does not require CDK-1 inhibitory phosphorylation normally seen
during G2 exit.

CYB-3 is maintained following P4 division but fails to localize on Z2/Z3
centrosomes
Since mitotic entry requires both the nuclear accumulation of Cyclin B and
removal of inhibitory phosphorylation from CDK-1, we next postulated that CYB-3
fails to accumulate in Z2/Z3 nuclei during G2. Here, we suspected that loss of CYB-3
would serve as a backup mechanism for Z2/Z3 cell cycle arrest, since CDK-1 did not
appear to be held inactive by tyrosine phosphorylation. One potential reason CYB-3
might never accumulate in Z2/Z3 is because PIE-1, which maintains maternal mRNA
in the germline, abruptly degrades following P4 division to Z2/Z3 (Tenenhaus et al.,
2001). Under this circumstance, the maternal messages of cyb-3 would be equally
destroyed. To test this possibility, we used mRNA in situ hybridization targeting cyb-
3 during embryogenesis in PGCs. Our preliminary attempts to analyze cyb-3
expression showed no discernable difference between cyb-3 signals in P1-P4 and
108
Z2/Z3 (Fig. S3.1), indicating that cyb-3 mRNA may have been slow to degrade after
PIE-1 loss.
To confirm these in situ results, we next asked whether CYB-3 proteins were
also present during each stage of embryogenesis using a Cyclin B monoclonal
antibody derived from Drosophila (clone F2F4). More recent work from our lab
identified F2F4 to recognize CYB-3 specifically in C. elegans (Michael, 2016). As
shown in Fig. 3.3A&B, strikingly almost all PGCs showed a nuclear accumulation of
CYB-3, even early and late Z2/Z3 nuclei. Specifically, we noticed a decrease in CYB-3
accumulation from P1 to P3, presumably due to increasing cell cycle lengths, but then
a spike in CYB-3 in early and late Z2/Z3 samples. These results ultimately suggest that
Z2/Z3 do not arrest by downregulating CYB-3.
While we did not see CYB-3 degradation after Z2/Z3 birth, we did observe one
noticeable difference between CYB-3 accumulation in P1-P4 and Z2/Z3 during our
analysis. Occasionally, in P1-P4 cells, CYB-3 localized to one or two foci at the nuclear
periphery in what appeared to be the mitotic spindle, as seen in the panel ‘P2’ of Fig.
3.3A. In Z2/Z3, however, CYB-3 never localized to these foci. Previous work
completed in Drosophila has demonstrated that Cyclin B associates with the
centrosomes during mitosis (Debec and Montmory, 1992), and so we predicted that
centrosomes were the structures that CYB-3 bound to in P1-P4. More interestingly,
however, other studies have shown that Cyclin B is initially phosphorylated in the
cytoplasm on centrosomes, and thus active cyclin B-CDK-1 first appears on
centrosomes during prophase (Jackman et al., 2003). Given that Z2/Z3 exit the cell
cycle in early prophase, we also predicted that Z2/Z3 centrosomes might lack
109

Figure 3.3. Detection of CYB-3 in PGCs during early and late embryogenesis.
A. Representative IF images of P1-P4, young Z2/Z3 (<100 cells), and late Z2/Z3
(>100 cells) stained for CYB-3 (red) and P-granules (green). B. Graph depicting
percentages of CYB-3 present in PGCs at various stages taken from three
biological replicates.

110

Figure 3.4. Centrosomal co-localization of CYB-3 and γ-tubulin in P1-P4 and Z2/Z3.
A. Representative IF images of P1, P2 and Z2/Z3 co-stained for gamma tubulin (red), CYB-3
(white), and P-granules (green). Images were merged (bottom row panels) by reducing
DAPI overlay to 30% in post-processing. B. Graph depicting percentages of PGCs that
showed co-localization of CYB-3 and gamma tubulin.

111
localized CYB-3. Using an antibody against gamma tubulin as a marker for
centrosomes, this is indeed what we observed. In P1-P4, 34 percent of samples had
CYB-3 overlap with gamma tubulin. However, in Z2/Z3 that number was zero percent
(Fig 3.4A&B). These results suggest that the lack of centrosomal activation of CYB-3
in Z2/Z3 may be a critical requirement preventing Z2/Z3 from mitotic entry.

Z2/Z3 CDK-1 inhibitory phosphorylation is activated with feeding in L1s
While our data so far has failed to elucidate a clear mechanism by which Z2/Z3
exit the cell cycle at the G2/M DNA damage checkpoint, one early question we had
was whether that mechanism maintains PGC arrest in L1s during starvation. Thus, in
our most initial work, we approached this question and asked if CDK-1 was
phosphorylated in Z2/Z3 of starved L1s. Here we hypothesized that P-CDK-1 would
increase with feeding for cells to repair the ZGA-associated DNA damage (Butuci et
al., 2015), and indeed when we stained Z2/Z3 in L1s for P-CDK-1 after 1 to 4 hours of
feeding, we observed that frequencies of P-CDK-1 signals increased with each feeding
timepoint (Fig. 3.4B). However, P-CDK-1 signals were never detected at the 1 hour
timepoint, suggesting that P-CDK-1 was not responsible for maintaining arrest during
starvation. Additionally, besides the frequency of P-CDK-1 signals increasing at each
timepoint, so did their intensity. For example, at 2 hours post-feeding, P-CDK-1
signals appeared as punctate foci, whereas at 4 hours post-feeding, P-CDK-1 signals
covered most, if not all, of the nucleus (Fig. 3.4A). Therefore, it appears that CDK-1 is
phosphorylated in Z2/Z3 during L1 feeding in response to the damage induced at ZGA.
112
However, P-CDK-1 is not required during L1 starvation, and we now know that this
is consistent with it also not being required during cell cycle exit in the embryo.

113

Figure 3.5. P-CDK-1 in Z2/Z3 in L1s during ZGA. A. Representative IF images
of Z2/Z3 in L1s that were fixed and stained for P-CDK-1 (red) and P-granules
(green) after 2-, 3-, and 4-hrs of feeding on E. coli OP50. B. Graph depicting
percentages of Z2/Z3 in L1s with P-CDK-1 signal after1-4 hours of feeding.

114
DISCUSSION

The G2/M DNA damage checkpoint does not mediate cell cycle exit in
embryonic Z2/Z3
The essential question surrounding the development of embryonic PGCs is
how do Z2/Z3 anticipate starvation and prepare their genomes for a long period of
quiescence before their next meal. For much of our early studies presented in this
chapter, we focused on the possibility that this could be controlled by molecular
events normally occurring at the G2/M DNA damage checkpoint, by virtue of some
transcriptional activation caused by PIE-1 degradation (Mello et al., 1996). To this
end, we sought the localizations of P-CHK-1, P-CDK-1, and CYB-3 in Z2/Z3.
Immunofluorescent staining for these proteins presented no surprises in the
youngest PGCs, P1-P4, however failed to explain how Z2/Z3 exit in G2. Specifically, we
observed a lack of both P-CHK-1 and P-CDK-1, but an accumulation of CYB-3, in
embryonic Z2/Z3. Our evidence here provides no support of a persistent DNA damage
checkpoint blocking Z2/Z3 division. Since the time when these studies were
conceptualized, however, we have learned that cell cycle exit is maintained through
chromatin compaction, and therefore perhaps genome structure plays a larger role in
G2 arrest than originally thought. Even still, it is intriguing to consider how an
unknown mechanism might arrest the Z2/Z3 cell cycle in advance of genome
compaction.
Despite that we were unable to connect the G2/M checkpoint response to
Z2/Z3 cell cycle exit, this work has raised some interesting questions. The first
115
question is, what mRNA messages are still being expressed after Z2/Z3 is born and
are their proteins required for arrest? As an example, the surprising persistence of
cyb-3 mRNA in Z2/Z3 (Fig. S3.1), could be offset with a lack of expression of other
factors like wee-1.3 or cdc-25.1. Since CDK-1 fails to show inhibitory phosphorylation
in Z2/Z3, it is possible that wee-1.3 is lost in Z2/Z3. Additionally, because we now
know that cdc-25.1 transcripts are absent in starved L1s and only generated during
feeding, it is also likely that cdc-25.1 mRNA is degraded along with PIE-1 (Seydoux
and Dunn 1997). Thus, further in situ experiments would help determine whether
wee-1.3 and cdc-25.1 transcripts were destroyed after P4 division, or whether they
were just never activated.

Cell cycle arrest by failure to activate Cyclin-B-CDK-1
Another interesting point is our observation that CYB-3 fails to appear on
Z2/Z3 centrosomes (Fig 3.4), and this leads us to ask if this a functionally relevant
phenotype. A brief review of cyclin B-CDK-1 activation reveals that during late G2,
Cyclin B concentrations increase on centrosomes (Lindqvist et al, 2009) and that in
the subsequent prophase, cyclin B-CDK-1 is first activated on centrosomes before
translocating to the nucleus during late prophase (Jackman et al., 2003). In this
context, centrosomes are a critical center of mitotic regulation required for mitotic
entry (Hachet et al, 2007). We speculate that a loss of any of these processes could
disrupt Z2/Z3 mitotic progression, and thus there are at least two possibilities as to
how failure of CYB-3 localization on centrosomes could lead to a late G2 arrest. First,
cyclin B-CDK-1 may not be activated if CYB-3 never concentrates on centrosomes.
116
This would be consistent with our other observation that CYB-3 in Z2/Z3 was entirely
nuclear, as previous reports suggest that cyclin B is first exported from the nucleus to
the cytoplasm during interphase (Toyoshima et al., 1998). An alternative explanation
for the loss of CYB-3 on centrosomes, is that centrosome function is downregulated
during Z2/Z3 arrest, but how that might occur is entirely undetermined. Currently,
we know that Z2/Z3 centrosomes duplicate in S-phase and fail to separate
(Fukuyama et al., 2006), but beyond this, little is known about their exact function in
Z2/Z3. If centrosomes do play a significant role in delaying Z2/Z3 mitotic entry, the
next question is whether a chronically active spindle assembly checkpoint (SAC) is
also induced. Studies in budding yeast and Drosophila identify a mitotic checkpoint
complex (MMC) containing BubR1, Mad2, Bub3, and Cdc20, which, when activated,
causes mitotic arrest at the G2/M transition to allow cyclin B accumulation
(Malmanche et al., 2006). Testing whether C. elegans homologs like bub-1, are
normally upregulated in Z2/Z3, or whether their loss initiates precocious mitotic
entry, can be a focus of further studies in the context of Z2/Z3 cell cycle exit.
Combining our results with the current knowledge, we might consider a
pathway where CYB-3 remains trapped in Z2/Z3 nuclei upon their birth, which would
prevent CYB-3 from being exported to the cytoplasm for cyclin B-CDK-1 activation on
centrosomes. In its inactive state, cyclin B-CDK-1 would fail to phosphorylate
downstream substrates like the APC, required for mitotic progression (Castro et al.,
2005; Suryadinata et al., 2010). While this model is purely speculative, it does present
the possibility that unconventional cellular mechanisms may initiate and maintain
cell cycle exit in the germline. To that end, this project ultimately highlights the
117
complexity with which Z2 and Z3 prepare for the long trip of genomic arrest through
the remainder of embryogenesis and L1 starvation.

118
MATERIALS AND METHODS

Strains
The strains used in this study were N2; Bristol wild-type and SS747; bnIs1[pie-
1::GFP::pgl-1 + unc-119(+)]. Strains were maintained on NGM E. coli OP50 plates using
standard culture techniques.

Antibodies and Dilutions
P-granules: mouse Mab OIC1D4, from the Developmental Studies Hybridoma Bank
(DSHB), was used at 1:10. Mouse Mab K76, from DSHB, was used at 1:5. GFP tag
polyclonal antibody (Alexa Fluor 555) was used at 1:200. Phospho-CHK-1: rabbit Mab
#2348, from Cell Signaling Technology, was used at 1:50. Phospho-CDK-1: goat
polyclonal antibody sc-7989, from Santa Cruz Biotechnology, was used at 1:50. CYB-
3: mouse Mab F2F4, from DSHB, was used at 1:50. Secondary antibodies: Alexa Fluor
conjugated secondary antibodies were purchased from Invitrogen and used at 1:200.

Immunofluorescent Staining of C. elegans Embryos
Embryos were washed 3 times with Milli-Q water, immediately mounted on a poly-L-
lysine glass slide, and then covered with a coverslip. Embryos were freeze-cracked by
placing the slide on a dry ice block for 10 minutes and flicking off the coverslip. Slides
were then immersed in 100% methanol at -20°C for 10 seconds and incubated in a
fixing solution (0.08M Hepes pH 6.9, 1.6mM MgSO4, 0.8mM EGTA, 3.7%
formaldehyde, 1X phosphate-buffered saline) for 10 minutes. Slides were washed in
119
Coplin jars with 50 mL of TBS-T three times for 10 minutes each. After rinsing, 50 µL
TNB+NGS was added directly to the slides for blocking for 2 hours. A primary
antibody solution was then directly applied and slides were incubated overnight at
4°C in a humidity chamber. The primary antibody solution was washed 3 times for 10
minutes in Coplin jars and a secondary antibody solution was added directly to each
slide to incubate for 2 hours. The secondary antibody was removed with three washes
of TBS for 10 minutes. Slides were mounted with DAPI counterstain and sealed with
a coverslip. We found that this freeze-cracking technique could also be utilized with
L1s without adversely affecting worm morphology.

Immunofluorescent Staining of L1 Worms
To stain L1 worms for P-CDK-1, worms were spun down in glass tubes and washed
three times with Milli-Q water. The L1 worms were then transferred to an Eppendorf
tube and 1 mL of 100% methanol at -20°C was added. Worms were spun down for
one minute and then immediately fixed at room temperature for 20 minutes in a
fixing solution (0.08M Hepes pH 6.9, 1.6mM MgSO4, 0.8mM EGTA, 3.7%
formaldehyde, 1X phosphate-buffered saline). After fixation, L1 worms were washed
3 times at room temperature for 10 minutes in TBS-T. A freshly prepared 100 μL
aliquot of SDS-DTT (804 μL 0.31% SDS, 196 μL 1M DTT) was then added to the tube
after rinsing and worms were shaken in a Thermomixer at 500 rpm at room
temperature. Shaking times in SDS-DTT varied between 10-30 minutes to ensure
most of the worms were permeabilized, as normally indicated by a bump in their
cuticle. SDS-DTT was briskly rinsed 3 times in TBS-T. To block, 50 μL of TNB+NGS
120
blocking solution was added and tubes were incubated in the Thermomixer at 500
rpm at room temperature for 2 hours. Worms were spun and blocking solution was
removed and replaced with a primary antibody solution of 50 μL. Worms were
incubated for 16-hours overnight in the Thermomixer at 500 rpm at 4°C. The next
day, worms were washed 3 times for 10 minutes in TBS, and then incubated with 50
μL of a secondary antibody solution for 2 hours in the Thermomixer at 500 rpm at
room temperature. After incubation, worms were washed 3 more times in TBS and
mounted on a glass slide coated with poly-L-lysine. Mounted samples were
counterstained with DAPI and coverslips were sealed with Cytoseal.

DIG In situ hybridization

RNA antisense probes were generated by ordering DNA oligonucleotides with a T7
RNA polymerase recognition sequence (IDT DNA) and then by in vitro transcription
using a DIG RNA labeling kit (Roche #1175025). To prepare samples, embryos were
collected from gravid adults via hypochlorite treatment and then permeabilized using
standard freeze-cracking techniques. Samples underwent a hydration series of
methanol for 5 minutes each at room temperature (100%, 90%, 70%, 50% methanol
and DEPC-ddH2O) before incubation in a non-toxic fixative (3% bronopol, 3%
diazolidinyl urea, 1.2% zinc sulfate heptahydrate, 0.3% sodium citrate) for 1 hour at
37°C. Samples were rinsed with DEPC-ddH2O and 2xSSC twice each for 5 minutes and
prehybridized in prehybridization buffer (PB) (1xSCC, 10% dextran sulfate, 50%
formamide, 0.4% 0.5M EDTA, 2% Denhardt’s Solution, 10% salmon sperm DNA) at
42°C for 1 hour. RNA probes were diluted 1:500 in PB and heated to 65°C for 5
121
minutes. Diluted probes were added to the slides and incubated overnight at 42°C.
The next day, samples were rinsed with 2xSSC and formamide buffer twice each for 5
minutes at 42°C, followed by two rinses each at room temperature in 2xSSC and Tris-
NaCl. Samples were blocked in blocking buffer (BB) at 37°C for 30 minutes before
incubating samples overnight at 4°C in a 1:200 dilution of anti-GFP tag polyclonal
antibody (Alexa Fluor 555) and anti-Digoxigenin antibody (FITC). These antibodies
were used to label P-granules expressing GFP and DIG-labeled RNA, respectively.
Samples were rinsed the next day twice with TN-EDTA before mounting with DAPI
counterstain.

Imaging, Statistical Analysis, and Presentation of Figures
All samples were imaged on an Olympus Fluoview FV1000 confocal microscope using
Fluoview Viewer software. To achieve consistent exposure levels, High Voltage lasers
were maintained at a tight range between samples and conditions in each experiment.
All figure data were obtained by independently performing two or three biological
replicates and images were saved and exported as TIFF files.
122

Figure S3.1. mRNA in situ hybridization (ISH) of cyb-3 during
embryogenesis. A. ISH images targeting cyb-3 mRNA (green) and P-granules
(red) in embryos. Representative images of a early embryo (P3) and slightly
older embryo (Z2/Z3) are shown. B. Positive control for ISH experiment
using, hlh-1 a gene normally expressed in the body wall muscles of C. elegans
as shown. See Materials and Methods for in situ protocol.

123
INTRODUCTION REFERENCES

Adams J.M. (2003). Ways of dying: multiple pathways to apoptosis. Genes Dev. 17,
2481–2495.

Ankeny, R.A. (2001). The natural history of Caenorhabditis elegans research. Nat Rev
Genet. 2, 474-479.

Batchelder, C., Dunn, M.A., Choy, B., Suh, Y., Cassie, C., Shim, E.Y., Shin, T.H., Mello, C.,
Seydoux, G., and Blackwell TK. (1999). Transcriptional repression by the
Caenorhabditis elegans germ-line protein PIE-1. Genes Dev. 13, 202-212.

Baxter, J. and Aragón, L. (2012). A model for chromosome condensation based on the
interplay between condensin and topoisomerase II. Trends Genet. 28, 110-117.

Baxter, J., Sen, N., Martínez, V.L., De Carandini, M.E., Schvartzman, J.B., Diffley, J.F., and
Aragón, L. (2011). Positive supercoiling of mitotic DNA drives decatenation by
topoisomerase II in eukaryotes. Science 331, 1328-1332.

Brenner S. (1974). The genetics of Caenorhabditis elegans. Genetics 77, 71-94.

Butuči, M., Williams, A.B., Wong, M.M., Kramer, B., and Michael, W.M. (2015). Zygotic
genome activation triggers chromosome damage and checkpoint signaling in C.
elegans primordial germ cells. Dev. Cell 34, 85-95.

Butuči, M., Wong, M.M., and Michael, W.M. (2015). Trouble in transitioning: activation
of zygotic transcription can lead to DNA breakage and genome instability. Worm 4,
e1115946.

Calderwood, S.K. (2016). A critical role for topoisomerase IIb and DNA double strand
breaks in transcription. Transcription 7, 75-83.

Crittenden, S.L., Leonhard, K.A., Byrd, D.T., and Kimble, J. (2006). Cellular analyses of
the mitotic region in the Caenorhabditis elegans adult germ line. Mol Biol Cell 17,
3051–3061.

Danial, N.N. and Korsmeyer, S.J. (2004). Cell death: critical control points. Cell 116,
205–219.

De Campos-Nebel, M., Larripa, I., and González-Cid, M. (2010). Topoisomerase II-
mediated DNA damage is differently repaired during the cell cycle by non-
homologous end joining and homologous recombination. PLoS ONE, 5.

124
Detwiler, M.R., Reuben, M., Li, X., Rogers, E., and Lin, R. (2001). Two zinc finger
proteins, OMA-1 and OMA-2, are redundantly required for oocyte maturation in C.
elegans. Dev Cell. 1, 187-199.

Edgar, B., Kiehle, C., and Schubiger, G. (1986). Cell cycle control by the nucleo-
cytoplasmic ratio in early Drosophila development. Cell 44, 365–372.

Edgar, L.G. and McGhee, J.D. (1988). DNA synthesis and the control of embryonic gene
expression in C. elegans. Cell 53, 589-599.

Fukuyama, M., Rougvie, A.E., and Rothman, J.H. (2006). C. elegans DAF-18/PTEN
mediates nutrient- dependent arrest of cell cycle and growth in the germline. Curr.
Biol. 16, 773-779.

Giurumescu, C.A. and Chisholm, A.D. (2011). Cell identification and cell lineage
analysis. Methods Cell Biol. 106, 325-341.

Guven-Ozkan, T., Nishi, Y., Robertson, S.M., and Lin R. (2008). Global transcriptional
repression in C. elegans germline precursors by regulated sequestration of TAF-4.
Cell 135, 149-160.

Haaf, T., Golub, E.I., Reddy, G., Radding, C.M., and Ward, D.C. (1995). Nuclear foci of
mammalian Rad51 recombination protein in somatic cells after DNA damage and its
localization in synaptonemal complexes. Proc. Natl. Acad. Sci. USA 92, 2298–2302.

Haffner, M.C., De Marzo, A.M., Meeker, A.K., Nelson, W.G., and Yegnasubramanian, S.
(2011). Transcription-induced DNA double strand breaks: both oncogenic force and
potential therapeutic target? Clin. Cancer Res. 17, 3858–3864.

Haffner, M.C., Aryee, M.J., Toubaji, A., Esopi, D.M., Albadine, R., Gurel, B., Isaacs, W.B.,
Bova, G.S., Liu, W., Xu, J., Meeker, A.K., Netto, G., De Marzo, A.M., Nelson, W.G., and
Yegnasubramanian, S. (2010). Androgen-induced TOP2B-mediated double- strand
breaks and prostate cancer gene rearrangements. Nat Genet. 42, 668-675.

Hooke, R. (1665). Micrographia, or, Some physiological descriptions of minute bodies
made by magnifying glasses: With observations and inquiries thereupon. London:
Printed by J. Martyn and J. Allestry.

Horvitz, H.R. (2003). Worms, life, and death (Nobel lecture). ChemBioChem 4, 697–
711.

Ishii, T., Funato, Y., Hashizume, O., Yamazaki, D., Hirata, Y., Nishiwaki, K., Nozomu, K.,
Arai, H., and Miki, H. (2016). Mg2+ extrusion from intestinal epithelia by CNNM
proteins is essential for gonadogenesis via AMPK-TORC1 signaling in Caenorhabditis
elegans. PLoS Genet. 12, e1006276.

125
Jha, S., and Dutta, A. (2009). RVB1/RVB2: running rings around molecular biology.
Mol. Cell 34, 521–533.

Kato, M., Chou, T.F., Yu, C.Z., Demodena, J., and Sternberg, P.W. (2014). LINKIN, a new
transmembrane protein necessary for cell adhesion. Elife 3, e04449.

Kawasaki, I., Amiri, A., Fan, Y., Meyer, N., Dunkelbarger, S., Motohashi, T., Karashima,
T., Bossinger, O., and Strome, S. (2004). The PGL family proteins associate with germ
granules and function redundantly in Caenorhabditis elegans germline development.
Genetics 167, 645–661.

Kelly, W.G. (2014). Transgenerational epigenetics in the germline cycle of
Caenorhabditis elegans. Epigenetics Chromatin 7, 6.

Kieffer-Kwon, K., Nimura, K., Rao, S., Xu, J., Jung, S., Pekowska, A., Dose, M., Stevens, E.,
Mathe, E., Dong, P., et al. (2017). Myc regulates chromatin decompaction and nuclear
architecture during B cell activation. Mol Cell 67, 566-578.

Kim, T.K., Hemberg, M., Gray, J.M., Costa, A.M., Bear, D.M., Wu, J. Harmin, D.A.,
Laptewicz, M., Barabara-Haley, K., Kuersten, S., et al. (2010). Widespread
transcription at neuronal activity-regulated enhancers. Nature 465, 182-187.

Kimble, J. and Hirsh, D. (1979). The postembryonic cell lineages of the hermaphrodite
and male gonads in Caenorhabditis elegans. Dev. Biol. 70, 396–417.

Kimble, J.E. and White, J.G. (1981). On the control of germ cell development in
Caenorhabditis elegans. Dev Biol. 81, 208–219.

Kurreck, J. (2009). RNA interference: from basic research to therapeutic applications.
Angew Chem. Int. Ed. Engl. 48, 1378-1398.

Leonard, J., Sen, N., Torres, R., Sutani, T., Jarmuz, A., Shirahige, K., and Aragón, L.
(2015). Condensin Relocalization from Centromeres to Chromosome Arms Promotes
Top2 Recruitment during Anaphase. Cell Rep. 13, 2336-2344.

Luijsterburg, M.S., Lindh, M., Acs, K., Vrouwe, M.G., Pines, A., van Attikum H.,
Mullenders, L.H., and Dantuma, N.P. (2012). DDB2 promotes chromatin
decondensation at UV-induced DNA damage. J Cell Biol. 197, 267-281.

Madabhushi, R., Gao, F., Pfenning, A.R., Pan, L., Yamakawa, S., Seo, J., Rueda, R., Phan,
T.X., Yamakawa, H., Pao, P.C., Stott, R.T., Gjoneska, E., Nott, A., Cho, S., Kellis, M., and
Tsai, L.H. (2015). Activity-Induced DNA Breaks Govern the Expression of Neuronal
Early-Response Genes. Cell. 161, 1592-1605.

126
Magalska, A., Schellhaus, A.K., Moreno-andrés, D., Zanini, F., Schooley, A., Sachdev, R.,
Schwarz, H., Madlung, J., and Antonin, W. (2014). RuvB-like ATPases function in
chromatin decondensation at the end of mitosis. Dev. Cell. 31, 305-318.

Mcfarlane, R.J., Feichtinger, J., and Larcombe L. (2014). Cancer germline gene
activation: friend or foe? Cell Cycle 13, 2151-2152.

Mello, C.C., Schubert, C., Draper, B., Zhang, W., Lobel, R., Priess, J.R. (1996). The PIE-1
protein and germline specification in C. elegans embryos. Nature 382, 710-712.

Mets, D.G., and Meyer, B.J. (2009). Condensins regulate meiotic DNA break
distribution, thus crossover frequency, by controlling chromosome structure. Cell
139, 73–86.

Patturajan, M., Schulte, R.J., Sefton, B.M., Berezney, R., Vincent, M., Bensaude, O.,
Warren S.L., and Corden, J.L. (1998). Growth-related changes in phosphorylation of
yeast RNA polymerase II. J. Biol. Chem. 273, 4689–4694.

Rawlings, J.S., Gatzka, M., Thomas, P.G., and Ihle, J.N. (2011). Chromatin condensation
via the condensin II complex is required for peripheral T-cell quiescence. EMBO J 30,
263-276.

Sheth, U., Pitt, J., Dennis, S., and Priess, J.R. (2010). Perinuclear P granules are the
principal sites of mRNA export in adult C. elegans germ cells. Development 137, 1305-
1314.

Seidel, H.S. and Kimble, J. (2015). Cell-cycle quiescence maintains Caenorhabditis
elegans germline stem cells independent of GLP-1/Notch. Elife 4.

Shimada, M., Yokosawa, H., and Kawahara, H. (2006). OMA-1 is a P granules-
associated protein that is required for germline specification in Caenorhabditis
elegans embryos. Genes Cells 11, 383-396.

Strome, S. (2005). Specification of the germ line. WormBook,
doi/10.1895/wormbook.1.9.1.

Strome, S. and Wood, W.B. (1982). Immunofluorescence visualization of germ-line-
specific cytoplasmic granules in embryos, larvae, and adults of C. elegans. Proc. Natl
Acad. Sci. USA 79, 1558-1562.

Strome, S. and Wood, W.B. (1983). Generation of asymmetry and segregation of germ-
line granules in early C. elegans embryos. Cell 35, 15-25.

Sulston, J.E., Schierenberg, E., White, J.G., and Thomson, J.N. (1983). The embryonic
cell lineage of the nematode Caenorhabditis elegans. Dev. Biol. 100, 64–119.
127

Tadros, W. and Lipshitz, H.D. (2009). The maternal-to-zygotic transition: a play in two
acts. Development 136, 3033–3042.

Thadani, R., Uhlmann, F., and Heeger, S. (2012). Condensin, chromatin crossbarring
and chromosome condensation. Curr Biol. 22, 1012-1021.

Updike, D.L., Hachey, S.J., Kreher, J., and Strome, S. (2011). P granules extend the
nuclear pore complex environment in the C. elegans germ line. J Cell Biol, 192, 939-
948.

Voronina, E. (2013). The diverse functions of germline P-granules in Caenorhabditis
elegans. Mol. Reprod. Dev. 80, 624-631.

Wang, J.T. and Seydoux, G. (2013). Germ cell specification. Adv Exp Med Biol. 757, 17-
39.

Whitehurst, A.W. (2014). Cause and consequence of cancer/testis antigen activation
in cancer. Annu Rev Pharmacol Toxicol. 54, 251-272.

Zhang, F., Barboric, M., Blackwell, T.K., and Peterlin, B.M. (2003). A model of
repression: CTD analogs and PIE-1 inhibit transcriptional elongation by P-TEFb.
Genes Dev. 17, 748-758.

Ziv, Y., Bielopolski, D., Galanty, Y., Lukas, C., Taya, Y., Schultz, D.C., Lukas, J., Bekker-
Jensen, S., Bartek, J., and Shiloh, Y. (2006). Chromatin relaxation in response to DNA
double-strand breaks is modulated by a novel ATM- and KAP-1 dependent pathway.
Nat Cell Biol. 8, 870-876.

CHAPTER I REFERENCES
Amitai, A., Seeber, A., Gasser, S.M., and Holcman, D. (2017). Visualization of Chromatin
Decompaction and Break Site Extrusion as Predicted by Statistical Polymer Modeling
of Single-Locus Trajectories. Cell Rep. 18, 1200-1214.

Antonin, W. and Neumann, H. (2016). Chromosome condensation and
decondensation during mitosis. Curr Opin Cell Biol. 40, 15-22.

Bizzaro, D., Manicardi, G., Bianchi, P.G., and Sakkas, D. (2000). Sperm decondensation
during fertilisation in the mouse: presence of DNase I hypersensitive sites in situ and
a putative role for topoisomerase II. Zygote. 8, 197-202.

Boulin, T., Etchberger, J.F., and Hobert, O. (2006) Reporter gene fusions. WormBook.
April 5, 1-23.

128
Butuči, M., Williams, A.B., Wong, M.M., Kramer, B., and Michael, W.M. (2015a). Zygotic
Genome Activation Triggers Chromosome Damage and Checkpoint Signaling in C.
elegans Primordial Germ Cells. Dev Cell. 34, 85-95.

Butuči, M., Wong, M.M., and Michael, W.M. (2015b). Trouble in transitioning:
activation of zygotic transcription can lead to DNA breakage and genome instability.
Worm 4, e1115946.

Calderwood, S.K. (2016). A critical role for topoisomerase IIb and DNA double strand
breaks in transcription. Transcription 7, 75-83.

Choi, H.M., Calvert, C.R., Husain, N., Huss, D., Barsi, J.C., Deverman, B.E., Hunter, R.C.,
Kato, M. Lee, S.M., Abelin, A.C., Rosenthal, A.Z., Akbari, O.S., Li, Y., Hay B.A., Sternberg,
P.W., Patterson, P.H., Davidson, E.H., Mazmanian, S.K., Prober, D.A., van de Rjin, M.,
Leadbetter, J.R., Newman, D.K., Readhead, C., Bronner, M.E., Wold, B., Lansford, R.,
Sauka-Spengler, T., Fraser, S.E. and Pierce, N.A. (2016). Mapping a multiplexed zoo of
mRNA expression. Development 143, 3632-3637.

Evans, K.J., Huang, N., Stempor, P., Chesney, M.A., Down, T.A., and Ahringer, J. (2016).
Stable Caenorhabditis elegans chromatin domains separate broadly expressed and
developmentally regulated genes. Proc Natl Acad Sci USA.

Falk, M., Lukasova, E., Gabrielova, B., Ondrej, V., and Kozubek, S. (2007). Chromatin
dynamics during DSB repair. Biochim Biophys Acta. 1773 ,1534-1545.

Fukuyama, M., Rougvie, A.E., and Rothman, J.H. (2006). C. elegans DAF-18/PTEN
mediates nutrient- dependent arrest of cell cycle and growth in the germline. Curr.
Biol. 16, 773-779.

Furuhashi, H., Takasaki, T., Rechtsteiner, A., Li, T., Kimura, H., Checchi, P.M., Strome, S.,
and Kelly, W.G. (2010). Trans-generational epigenetic regulation of C. elegans
primordial germ cells. Epigenetics Chromatin 3, 15.

Jaramillo-Lambert, A., Fabritius, A.S., Hansen, T.J., Smith, H.E., and Golden, A. (2016).
The identification of a novel mutant allele of topoisomerase II in Caenorhabditis
elegans reveals a unique role in chromosome segregation during spermatogenesis.
Genetics 204, 1407-1422.

Jha, S. and Dutta, A. (2009). RVB1/RVB2: running rings around molecular biology. Mol.
Cell 34, 521–533.

Ju, B.G., Lunyak, V.V., Perissi, V. Garcia-Bassets, I., Rose, D.W., Glass, C.K., and Rosenfeld,
M.G. (2006). A topoisomerase IIbeta-mediated dsDNA break required for regulated
transcription. Science 312, 1798-1802.

129
Kato, M., Chou, T.F., Yu, C.Z., Demodena, J., and Sternberg, P.W. (2014). LINKIN, a new
transmembrane protein necessary for cell adhesion. Elife 3, e04449.

Kruhlak, M.J., Celeste, A., Dellaire G., Fernandez-Capetillo, O., Müller, W.G., McNally,
J.G., Bazett-Jones, D.P., and Nussenzweig, A. (2006). Changes in chromatin structure
and mobility in living cells at sites of DNA double-strand breaks. J Cell Biol. 172, 823-
834.

Kieffer-Kwon, K., Nimura, K., Rao, S., Xu, J., Jung, S., Pekowska, A., Dose, M., Stevens, E.,
Mathe, E., Dong, P., et al. (2017). Myc regulates chromatin decompaction and nuclear
architecture during B cell activation. Mol Cell, in press

Magalska, A., Schellhaus, A.K., Moreno-Andrés, D., Zanini, F., Schooley, A., Sachdev, R.,
Schwarz, H., Madlung, J., and Antonin, W. (2014). RuvB-like ATPases function in
chromatin decondensation at the end of mitosis. Dev. Cell. 31, 305-318.

Mainpal, R., Nance, J., and Yanowitz, JL. (2015). A germ cell determinant reveals
parallel pathways for germ line development in Caenorhabditis elegans.
Development 142, 3571-3582.

McNally, K., Audhya, A., Oegema, K., and McNally, F.J. (2006). Katanin controls mitotic
and meiotic spindle length. J Cell Biol. 175, 881-891.

Newport, J. and Spann, T. (1987). Disassembly of the nucleus in mitotic extracts:
membrane vesicularization, lamin disassembly, and chromosome condensation are
independent processes. Cell 48, 219-230.

Palancade, B. and Bensaude, O. (2003). Investigating RNA polymerase II carboxyl-
terminal domain (CTD) phosphorylation. Eur J Biochem. 270, 3859-3870.

Phillips, C.M., Wong, C., Bhalla, N., Carlton, P.M., Weiser, P., Meneely, P.M, and
Dernburg, A.F. (2005). HIM-8 binds to the X chromosome pairing center and mediates
chromosome-specific meiotic synapsis. Cell 123, 1051-1063.

Pommier, Y., Huang, S.Y., Gao, R., Das, B.B., Murai, J., and Marchand, C. (2014). Tyrosyl-
DNA-phosphodiesterases (TDP1 and TDP2). DNA Repair (Amst) 19, 114-129.

Pommier, Y., Sun, Y., Huang, S.N., and Nitiss, J.L. (2016). Roles of eukaryotic
topoisomerases in transcription, replication and genomic stability. Nat Rev Mol Cell
Biol. 17, 703-721.

Rawlings, J.S., Gatzka, M., Thomas, P.G., and Ihle, J.N. (2011). Chromatin condensation
via the condensin II complex is required for peripheral T-cell quiescence. EMBO J 30,
263-276.

130
Takasuga, Y., Andoh, T., Yamashita, J., and Yagura, T. (1995). ICRF-193, an inhibitor of
topoisomerase II, demonstrates that DNA replication in sperm nuclei reconstituted in
Xenopus egg extracts does not require chromatin decondensation. Exp Cell Res. 217,
378-384.

Uemura, T., Ohkura, H., Adachi, Y., Morino, K., Shiozaki, K., and Yanagida, M. (1987).
DNA topoisomerase II is required for condensation and separation of mitotic
chromosomes in S. pombe. Cell 50, 917–925.

Schaner, C.E., Deshpande, G., Schedl, P.D., and Kelly, W.G. (2003). A conserved
chromatin architecture marks and maintains the restricted germ cell lineage in
worms and flies. Dev. Cell 5, 747-757.

Seydoux, G. and Dunn, M.A. (1997). Transcriptionally repressed germ cells lack a
subpopulation of phosphorylated RNA polymerase II in early embryos of
Caenorhabditis elegans and Drosophila melanogaster. Development 124, 2191–2201.

Walker, A.K., Shi, Y., and Blackwell TK. (2004). An extensive requirement for
transcription factor IID-specific TAF-1 in Caenorhabditis elegans embryonic
transcription. J Biol Chem. 279, 15339-15347.

Yokoo, R., Zawadzki, K.A., Nabeshima, K., Drake, M., Arur, S., and Villeneuve A.M.
(2012). COSA-1 reveals robust homeostasis and separable licensing and
reinforcement steps governing meiotic crossovers. Cell. 149, 75-87.

Yoon, S., Kawasaki, I., and Shim, Y.H. (2012). CDC-25.1 controls the rate of germline
mitotic cell cycle by counteracting WEE-1.3 and by positively regulating CDK-1 in
Caenorhabditis elegans. Cell Cycle 11, 1354-1363.

Ziv, Y., Bielopolski, D., Galanty, Y., Lukas, C., Taya, Y., Schultz, D.C., Lukas, J., Bekker-
Jensen, S., Bartek, J. and Shiloh, Y. (2006). Chromatin relaxation in response to DNA
double-strand breaks is modulated by a novel ATM- and KAP-1 dependent pathway.
Nat Cell Biol. 8, 870-876.

CHAPTER II REFERENCES

Akers, S.N., Odunsi, K., and Karpf, A.R. (2010). Regulation of cancer germline antigen
gene expression: implications for cancer immunotherapy. Future Oncol. 6, 717-732.

Ankeny, R.A. (2001). The natural history of Caenorhabditis elegans research. Nat Rev
Genet. 2, 474-479.

131
Ayrapetov, M.K., Gursoy-Yuzugullu, O., Xu, C., Xu, Y., and Price, B.D. (2014). DNA
double-strand breaks promote methylation of histone H3 on lysine 9 and transient
formation of repressive chromatin. Proc Natl Acad Sci USA. 111, 9169-9174.

Batchelder, C., Dunn, M.A., Choy, B., Suh, Y., Cassie, C., Shim, E.Y., Mello, C., Seydoux, G.,
and Blackwell, T.K. (1999). Transcriptional repression by the Caenorhabditis elegans
germ-line protein PIE-1. Genes Dev. 13, 202-212.

Baxter, J. and Aragón, L. (2012). A model for chromosome condensation based on the
interplay between condensin and topoisomerase II. Trends Genet. 28, 110-117.

Bender, L.B., Cao, R., Zhang, Y., and Strome, S. (2004). The MES-2/MES-3/MES-6
Complex and Regulation of Histone H3 Methylation in C. elegans. Curr. Biol. 14, 1639–
1643.

Bender, L.B., Suh, J., Carroll, C.R., Fong, Y., Fingerman, I.M., Briggs, S.D., Cao, R., Zhang,
Y., Reinke, V., and Srome, S. (2006). MES-4: an autosome-associated histone
methyltransferase that participates in silencing the X chromosomes in the C. elegans
germ line. Development 133, 3907-3917.

Bowman, E.A., Bowman, C.R., Ahn, J.H., and Kelly, W.G. (2013). Phosphorylation of
RNA polymerase II is independent of P-TEFb in the C. elegans germline. Development
140, 3703-3713.

Butuči, M., Williams, A.B., Wong, M.M., Kramer, B., and Michael, W.M. (2015). Zygotic
Genome Activation Triggers Chromosome Damage and Checkpoint Signaling in C.
elegans Primordial Germ Cells. Dev Cell. 34, 85-95.

Butuči, M., Wong, M.M., and Michael, W.M. (2015). Trouble in transitioning: activation
of zygotic transcription can lead to DNA breakage and genome instability. Worm 4,
e1115946.

Choi, H.M., Beck, V.A., and Pierce, N.A. (2014). Next-generation in situ hybridization
chain reaction: higher gain, lower cost, greater durability. ACS Nano 8, 4284-4294.

Choi, H.M., Calvert, C.R., Husain, N., Huss, D., Barsi, J.C., Deverman, B.E., Hunter, R.C.,
Kato, M., Lee, S.M., Abelin, A.C., Rosenthal, A.Z., Akbari, O.S., Li, Y., Hay, B.A., Sternberg,
P.W., Patterson, P.H., Davidson, E.H., Mazmanian, S.K., Prober, D.A., van de Rijn, M.,
Leadbetter, J.R., Newman, D.K., Readhead, C., Bronner, M.E., Wold, B., Lansford, R.,
132
Sauka-Spengler, T., Fraser, S.E., and Pierce, N.A. (2016). Mapping a multiplexed zoo of
mRNA expression. Development 143, 3632-3637.

Csankovszki, G., Collette, K., Spahl, K., Carey, J., Snyder, M., Petty, E., Patel, U., Tabuchi,
T., Liu, H., McLeod, I., Thompson, J., Sarkeshik, A., Yates, J., Meyer, B.J., and Hagtrom, K.
(2009). Three distinct condensin complexes control C. elegans chromosome
dynamics. Curr Biol. 19, 9-19.

Daniels, B.R., Perkins, E.M., Dobrowsky, T.M., Sun, S.X., and Wirtz, D. (2012).
Asymmetric enrichment of PIE-1 in the Caenorhabditis elegans zygote mediated by
binary counterdiffusion. J Cell Biol. 184, 473-479.

Ercan, S. (2015). Mechanisms of x chromosome dosage compensation. J Genomics 3,
1-19.

Evans, K.J., Huang, N., Stempor, P., Chesney, M.A., Down, T.A., and Ahringer, J. (2016).
Stable Caenorhabditis elegans chromatin domains separate broadly expressed and
developmentally regulated genes. Proc Natl Acad Sci USA.

Ford, J.R. and Schumacher, J.M. (2009). Chromosome dynamics: the case of the
missing condensin. Curr Biol. 19, 127-129.

Fukuyama, M., Rougvie, A.E., and Rothman, J.H. (2006). C. elegans DAF-18/PTEN
mediates nutrient- dependent arrest of cell cycle and growth in the germline. Curr.
Biol. 16, 773-779.

Furniss, K.L., Tsai, H.J., Byl, J.A., Lane, A.B., Vas, A.C., Hsu, W.S., Osheroff, N., and Clarke,
D.J. (2013). Direct monitoring of the strand passage reaction of DNA topoisomerase
II triggers checkpoint activation. PLoS Genet. 9, e1003832.

Furuhashi, H., Takasaki, T., Rechtsteiner, A., Li, T., Kimura, H., Checchi, P.M., Strome, S.,
and Kelly, W.G. (2010). Trans-generational epigenetic regulation of C. elegans
primordial germ cells. Epigenetics Chromatin 3, 15.

Ghosh, D. and Seydoux, G. (2008). Inhibition of transcription by the Caenorhabditis
elegans germline protein PIE-1: genetic evidence for distinct mechanisms targeting
initiation and elongation. Genetics 178, 235-243.

Gong, Y., Gilcrease, M., and Sneige, N. (2005). Reliability of chromogenic in situ
hybridization for detecting HER-2 gene status in breast cancer: comparison with
133
fluorescence in situ hybridization and assessment of interobserver reproducibility.
Mod Pathol 18, 1015-1021.

Guven-Ozkan, T., Nishi, Y., Robertson, S.M., and Lin, R. (2008). Global transcriptional
repression in C. elegans germline precursors by regulated sequestration of TAF-4.
Cell 135, 149-160.

Haffner, M.C., De Marzo, A.M., Meeker, A.K., Nelson, W.G., and Yegnasubramanian, S.
(2011). Transcription-induced DNA double strand breaks: both oncogenic force and
potential therapeutic target. Clin Cancer Res. 17, 3858-3864.

Hirano T. (2012). Condensins: universal organizers of chromosomes with diverse
functions. Genes Dev. 26, 1659-1678.

Hirota, T., Gerlich, D., Koch, B., Ellenberg, J., and Peters, J.M. (2004). Distinct functions
of condensin I and II in mitotic chromosome assembly. J Cell Sci. 117, 6435-6445.

Janic, A., Mendizabal, L., Llamazares, S., Rossell, D., and Gonzalez, C. (2010). Ectopic
expression of germline genes drives malignant brain tumor growth in Drosophila.
Science 330, 1824-1827.

Jensen, E. (2014). Technical review: In situ hybridization. Anat Rec (Hoboken) 297,
1349-1353.

Kirienko, N.V., Mani, K., and Fay, D.S. (2010). Cancer models in Caenorhabditis elegans.
Dev Dyn. 239, 1413-1448.

Kranz, A.L., Jiao, C.Y., Winterkorn, L.H., Albritton, S.E., Kramer, M., and Ercan S. (2013).
Genome-wide analysis of condensin binding in Caenorhabditis elegans. Genome Biol.
14, R112.

Lau, A.C. and Csankovszki, G. (2014). Condensin-mediated chromosome organization
and gene regulation. Front Genet. 5, 473.

Lau, A.C., Nabeshima, K., and Csankovszki, G. (2014). The C. elegans dosage
compensation complex mediates interphase X chromosome compaction. Epigenetics
Chromatin. 7, 31.

134
Leonard, J., Sen, N., Torres, R., Sutani, T., Jarmuz, A., Shirahige, K., and Aragón, L.
(2015). Condensin Relocalization from Centromeres to Chromosome Arms Promotes
Top2 Recruitment during Anaphase. Cell Rep. 13, 2336-2344.

Machida, S., Takizawa, Y., Ishimaru, M., Sugita, Y., Sekine, S., Nakayama, J.I., Wolf, M.,
and Kurumizaka, H. (2018). Structural Basis of Heterochromatin Formation by
Human HP1. Mol Cell 69, 385-397.

Mcfarlane, R.J., Feichtinger, J., and Larcombe L. (2014). Cancer germline gene
activation: friend or foe? Cell Cycle 13, 2151-2152.

Mello, C.C., Draper, B.W., Krause, M., Weintraub, H., Priess, J.R. (1992). The pie-1 and
mex-1 genes and maternal control of blastomere identity in early C. elegans embryos.
Cell 70, 163-176.

Meyer, BJ. (2010). Targeting X chromosomes for repression. Curr Opin Genet Dev. 20,
179-189.

Palancade, B and Bensaude, O. (2003). Investigating RNA polymerase II carboxyl-
terminal domain (CTD) phosphorylation. Eur J Biochem. 270, 3859-3870.

Reinke, V., Krause, M., and Okkema, P. (2013) Transcriptional regulation of gene
expression in C. elegans. WormBook, doi/10.1895/wormbook.1.45.2

Schaner, C.E. and Kelly, W.G. (2006). Germline chromatin. WormBook,
doi/10.1895/wormbook.1.73.1

Schaner, C.E., Deshpande, G., Schedl, P.D., Kelly, W.G. (2003). A conserved chromatin
architecture marks and maintains the restricted germ cell lineage in worms and flies.
Dev Cell 5, 747-757.

Seydoux, G. and Dunn, M.A. (1997). Transcriptionally repressed germ cells lack a
subpopulation of phosphorylated RNA polymerase II in early embryos of
Caenorhabditis elegans and Drosophila melanogaster. Development. 124, 2191-2201.

Shin-i, T. and Kohara, Y. (2005). NEXTDB (The nematode expression pattern
database). Mishima, Japan: National Institute of Genetics.

Strome, S. (2005). Specification of the germ line. WormBook,
doi/10.1895/wormbook.1.9.1
135

Sun ,Y., Jiang, X., and Price, B.D. (2010). Tip60: connecting chromatin to DNA damage
signaling. Cell Cycle 9, 930-936.

Sutani, T., Sakata, T., Nakato, R., Masuda, K., Ishibashi, M., Yamashita, D., Suzuki, Y.,
Hirano, T., Bando, M., and Shirahige, K. (2015). Condensin targets and reduces
unwound DNA structures associated with transcription in mitotic chromosome
condensation. Nat Commun. 6, 7815.

Takami, H., Fukushima, S., Fukuoka, K., Suzuki, T., Yanagisawa, T., Matsushita, Y.,
Nakamura, T., Arita, H., Mukasa, A., Saito, N., Kanamori, M., Kmabe, T., Tominaga, T.,
Kobaashi, K., Nagane, M., Iuchi, T., Tamura, K., Maehara, T., Sugiyama, K., Nakada, M.,
Kanemura, Y., Nonaka, M., Yokogami, K., Takeshima, H., Narita, Y., Shibui, S., Nakazato,
Y., Nishikawa, R., Ichimura, K., and Matsutani, M. (2015). Human chorionic
gonadotropin is expressed virtually in all intracranial germ cell tumors. J Neurooncol.
124, 23-32.

Teves, S.S. and Henikoff, S. (2014). DNA torsion as a feedback mediator of
transcription and chromatin dynamics. Nucleus 5, 211-218.

Tintori, S.C., Osborne Nishimura, E., Golden, P., Lieb, J.D., and Goldstein, B. (2016). A
Transcriptional Lineage of the Early C. elegans Embryo. Dev Cell 38, 430-444.

Uemura, T., Ohkura, H., Adachi, Y., Morino, K., Shiozaki, K., and Yanagida, M. (1987).
DNA topoisomerase II is required for condensation and separation of mitotic
chromosomes in S. pombe. Cell 50, 917–925.

Vastenhouw, N.L., Fischer, S.E., Robert, V.J., Thijssen, K.L., Fraser, A.G., Kamath, R.S.,
Ahringer, J., and Plasterk, R.H. (2003. A genome-wide screen identifies 27 genes
involved in transposon silencing in C. elegans. Curr Biol. 13, 1311-1316.

Wang, J.T. and Seydoux, G. (2013). Germ cell specification. Adv Exp Med Biol. 757, 17-
39.

CHAPTER III REFERENCES

Aoki, H., Sato, S., Takanami, T., Ishihara, T., Katsura, I., Takahashi, H., and Higashitani,
A. (2000). Characterization of Ce-atl-1, an ATM-like gene from Caenorhabditis elegans.
Mol Gen Genet. 264, 119-126.

Ashcroft, N. and Golden, A. (2002). CDC-25.1 regulates germline proliferation in
Caenorhabditis elegans. Genesis. 33, 1-7.
136

Ashcroft, N.R., Srayko, M., Kosinski, M.E., Mains, P.E., and Golden A. (1999). RNA-
Mediated interference of a cdc25 homolog in Caenorhabditis elegans results in
defects in the embryonic cortical membrane, meiosis, and mitosis. Dev Biol. 206, 15-
32.

Bailly, E., Pines, J., Hunter, T., and Bornens, M. (1992). Cytoplasmic accumulation of
cyclin B1 in human cells: association with a detergent-resistant compartment and
with the centrosome. J Cell Sci. 101, 529-545.

Blethrow, J.D., Glavy, J.S., Morgan, D.O., and Shokat, K.M. (2008). Covalent capture of
kinase-specific phosphopeptides reveals Cdk1-cyclin B substrates. Proc Natl Acad Sci
USA 105, 1442-1447.

Boxem, M., Srinivasan, D.G., and van den Heuvel, S. (1999) The Caenorhabditis elegans
gene ncc-1 encodes a cdc2-related kinase required for M phase in meiotic and mitotic
cell divisions, but not for S phase. Development 126, 2227-2239.

Boxem, M. (2006). Cyclin-dependent kinases in C. elegans. Cell Div. 1.

Burrows, A.E., Sceurman, B.K., Kosinski, M.E., Richie, C.T., Sadler, P.L., Schumacher,
J.M., and Golden, A. (2006). The C. elegans Myt1 ortholog is required for the proper
timing of oocyte maturation. Development 133, 697-709.

Carroll, J. and Marangos, P. (2013). The DNA damage response in mammalian oocytes.
Front Genet. 4, 117.

Castro, A., Bernis, C., Vigneron, S., Labbé, J.C., and Lorca, T. (2005). The anaphase-
promoting complex: a key factor in the regulation of cell cycle. Oncogene 24, 314-325.

Debec, A. and Montmory, C. (1992). Cyclin B is associated with centrosomes in
Drosophila mitotic cells. Biol Cell 75, 121-126.

Desai, D. Gu, Y., and Morgan, D.O. (1992). Activation of human cyclin-dependent
kinases in vitro. Mol Biol Cell. 3, 571-582.
Detwiler, M.R., Reuben, M., Li, X., Rogers, E., and Lin, R. (2001). Two zinc finger
proteins, OMA-1 and OMA-2, are redundantly required for oocyte maturation in C.
elegans. Dev Cell. 1, 187-199.

Deyter, G.M., Furuta, T., Kurasawa, Y., and Schumacher, J.M. (2010). Caenorhabditis
elegans cyclin B3 is required for multiple mitotic processes including alleviation of a
spindle checkpoint-dependent block in anaphase chromosome segregation. PLoS
Genet. 6: e1001218.

Donzelli, M. and Draetta, G.F. (2003). Regulating mammalian checkpoints through
Cdc25 inactivation. EMBO Rep. 4, 671-677.
137

Enserink, J.M. and Kolodner, R.D. (2015). An overview of Cdk1-controlled targets and
processes. Cell Div. 5, 11.

Fukuyama, M., Rougvie, A.E., and Rothman, J.H. (2006). C. elegans DAF-18/PTEN
Mediates Nutrient-Dependent Arrest of Cell Cycle and Growth in the Germline.
Current Biology 16, 773-779.

Furuhashi, H., Takasaki, T., Rechtsteiner, A., Li, T., Kimura, H., Checchi, P.M., Strome, S.,
and Kelly, W.G. (2010). Trans-generational epigenetic regulation of C. elegans
primordial germ cells. Epigenetics Chromatin 3, 15.

Gavet, O. and Pines, J. (2010). Activation of cyclin B1-Cdk1 synchronizes events in the
nucleus and the cytoplasm at mitosis. J Cell Biol. 189, 247-259.

Glover-Cutter, K., Larochelle, S., Erickson, B., Zhang, C., Shokat, K., Fisher, R.P., and
Bentley, D.L. (2009). TFIIH-associated Cdk7 kinase functions in phosphorylation of C-
terminal domain Ser7 residues, promoter-proximal pausing, and termination by RNA
polymerase II. Mol Cell Biol. 29, 5455-5464.

Han, S.J., Chen, R., Paronetto. M.P., and Conti, M. (2005). Wee1B is an oocyte-specific
kinase involved in the control of meiotic arrest in the mouse. Curr Biol. 15, 1670-1676.

Jackman, M., Lindon, C., Nigg, E.A., and Pines, J. (2003). Active cyclin B1-Cdk1 first
appears on centrosomes in prophase. Nat Cell Biol. 5, 143-148.

Jaramillo-Lambert, A., Harigaya, Y., Vitt, J., Villeneuve, A., and Engebrecht, J. (2010).
Meiotic errors activate checkpoints that improve gamete quality without triggering
apoptosis in male germ cells. Curr Biol. 20, 2078-2089.

Kalogeropoulos, N., Christoforou, C., Green, A.J., Gill, S., and Ashcroft, N.R. (2004). chk-
1 is an essential gene and is required for an S-M checkpoint during early
embryogenesis. Cell Cycle 3, 1196-1200.

Kermi, C., Lo Furno, E., and Maiorano, (2017). D. Regulation of DNA Replication in
Early Embryonic Cleavages. Genes 8.

Koreth, J. and Van den heuvel, S. (2005). Cell-cycle control in Caenorhabditis elegans:
how the worm moves from G1 to S. Oncogene 24, 2756-2764.

Kraft, C., Herzog, F., Gieffers, C., Mechtler, K., Hagting, A., Pines, J., and Peters, J.M.
(2003). Mitotic regulation of the human anaphase-promoting complex by
phosphorylation. EMBO J 22, 6598-6609.

Kristjánsdóttir, K. and Rudolph, J. (2004). Cdc25 phosphatases and cancer. Chem Biol.
11, 1043-1051.
138

Lamitina, S.T. and L'hernault, S.W. (2002). Dominant mutations in the Caenorhabditis
elegans Myt1 ortholog wee-1.3 reveal a novel domain that controls M-phase entry
during spermatogenesis. Development 129, 5009-5018.

Lee, J., Kumagai, A., and Dunphy, W.G. (2001). Positive regulation of Wee1 by Chk1
and 14-3-3 proteins. Mol. Biol. Cell 12, 551-563.

Lee, S.J., Gartner, A., Hyun, M., Ahn, B., and Koo, H.S. (2010). The Caenorhabditis
elegans Werner syndrome protein functions upstream of ATR and ATM in response
to DNA replication inhibition and double-strand DNA breaks. PLoS Genet. 6,
e1000801.

Lesch, B.J. and Page, D.C. (2012). Genetics of germ cell development. Nat Rev Genet.
13, 781-794.

Lincoln, A.J., Wickramasinghe, D., Stein, P., Schultz, R.M., Palko, M.E., De Miguel, M.P.,
Tessarollo, L., and Donovan, P.J. (2002). Cdc25b phosphatase is required for
resumption of meiosis during oocyte maturation. Nat. Genet 30, 446-449.

Liu J. and Kipreos, E.T. (2000). Evolution of cyclin-dependent kinases (CDKs) and
CDK-activating kinases (CAKs): differential conservation of CAKs in yeast and
metazoa. Mol. Biol. Evol. 17, 1061-1074.

Mainpal, R., Nance, J., and Yanowitz, JL. (2015). A germ cell determinant reveals
parallel pathways for germ line development in Caenorhabditis elegans.
Development 142, 3571- 3582.

Malumbres, M. (2014). Cyclin-dependent kinases. Genome Biol. 15, 122.

Mello, C.C., Schubert, C., Draper, B., Zhang, W., Lobel, R., and Priess, J.R. (1996). The
PIE-1 protein and germline specification in C. elegans embryos. Nature 382, 710–712.

Michael, W.M. (2016). Cyclin CYB-3 controls both S-phase and mitosis and is
asymmetrically distributed in the early C. elegans embryo. Development 143, 3119-
3127.

Norbury, C., Blow, J. and Nurse, P. (1991). Regulatory phosphorylation of the p34cdc2
protein kinase in vertebrates. The EMBO journal 10, 3321-3329.

Nurse, P. (1990). Universal control mechanism regulating onset of M-phase. Nature
344, 503-508.

Potapova, T.A., Daum, J.R., Byrd, K.S., and Gorbsky, G.J. (2009). Fine tuning the cell
cycle: activation of the Cdk1 inhibitory phosphorylation pathway during mitotic exit.
Mol. Biol. Cell 20, 1737-1748.
139

Rhind, N. and Russell. P. (2012). Signaling pathways that regulate cell division. Cold
Spring Harb. Perspect Biol. 4.

Seydoux, G. and Dunn, M.A. (1997). Transcriptionally repressed germ cells lack a
subpopulation of phosphorylated RNA polymerase II in early embryos of
Caenorhabditis elegans and Drosophila melanogaster. Development 124, 2191–2201.

Smits, V.A. and Gillespie DA. (2015). DNA damage control: regulation and functions of
checkpoint kinase 1. FEBS J 282, 3681-3692.

Solc, P., Schultz. R.M., and Motlik, J. (2010). Prophase I arrest and progression to
metaphase I in mouse oocytes: comparison of resumption of meiosis and recovery
from G2-arrest in somatic cells. Mol. Hum. Reprod. 16, 654-664.

Suryadinata, R., Sadowski, M., and Sarcevic, B. (2010). Control of cell cycle
progression by phosphorylation of cyclin-dependent kinase (CDK) substrates. Biosci.
Rep. 30, 243-255.

Tadros, W., and Lipshitz, H.D. (2009). The maternal-to-zygotic transition: a play in
two acts. Development 136, 3033–3042.

Tenenhaus, C., Subramaniam, K., Dunn, M.A., and Seydoux, G. (2001). PIE-1 is a
bifunctional protein that regulates maternal and zygotic gene expression in the
embryonic germ line of Caenorhabditis elegans. Genes Dev. 15, 1031-1040.

Thomas, Y., Cirillo, L., Panbianco, C., Martino, L., Tavernier, N., Schwager, F., Van Hove,
L., Joly, N., Santamaria, A., Pintard, L., and Gotta, M. (2016). Cdk1 Phosphorylates
SPAT-1/Bora to Promote Plk1 Activation in C. elegans and Human Cells. Cell Rep. 15,
510-518.

Van den Heuvel, S. (2005). Cell Cycle Regulation. Wormbook,
doi/10.1895/wormbook.1.28.1

Van der voet, M., Lorson, M.A., Srinivasan, D.G., Bennett, K.L., and Van den heuvel, S.
(2009). C. elegans mitotic cyclins have distinct as well as overlapping functions in
chromosome segregation. Cell Cycle 8, 4091-4102.

Wallenfang, M.R. and Seydoux, G. (2002). cdk-7 Is required for mRNA transcription
and cell cycle progression in Caenorhabditis elegans embryos. Proc. Natl. Acad. Sci.
USA. 99, 5527-5532.

Wang, X., Liu, M., Li, W., Suh, C.D., Zhu, Z., Jin, Y., and Fan, Q. (2009). The function of a
spindle checkpoint gene bub-1 in C. elegans development. PLoS One 4, 6.

140
Yoon, S., Kawasaki, I., and Shim, Y.H. (2012). CDC-25.1 controls the rate of germline
mitotic cell cycle by counteracting WEE-1.3 and by positively regulating CDK-1 in
Caenorhabditis elegans. Cell Cycle 11, 1354-1363.

Asset Metadata

Creator Wong, Matthew Martin (author)

Core Title C. elegans topoisomerase II regulates chromatin architecture and DNA damage for germline genome activation

Contributor Electronically uploaded by the author (provenance)

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Molecular Biology

Publication Date 07/20/2018

Defense Date 06/19/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag C. elegans,chromatin decompaction,DNA double-strand breaks,germline,nuclear architecture,oai:digitallibrary.usc.edu:usctheses,OAI-PMH Harvest,primordial germ cells,RUVB,topoisomerase II,zygotic genome activation

Format application/pdf (imt)

Language English

Advisor Michael, Matthew (committee chair), Aparicio, Oscar (committee member), Phillips, Carolyn (committee member), Rice, Judd (committee member)

Creator Email thechosenwong@gmail.com,wong472@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-20838

Unique identifier UC11668943

Identifier etd-WongMatthe-6431.pdf (filename),usctheses-c89-20838 (legacy record id)

Legacy Identifier etd-WongMatthe-6431.pdf

Dmrecord 20838

Document Type Dissertation

Format application/pdf (imt)

Rights Wong, Matthew Martin

Type texts

Source 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 a...

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

Abstract (if available)

Abstract The protection and transmission of an organism’s genome is arguably the most important task of its lifetime. Central to this mission are faithful DNA damage repair, gene expression, and proliferation of germline sex cells. Such demanding activities have made germ cells excellent models over the years for identifying some of the most robust molecular mechanisms in cellular biology. Still, we have only begun to uncover the extraordinary lengths to which germ cells undergo to develop. In C. elegans, the primordial germ cells Z2 and Z3 are born during early embryogenesis and then held in a transcriptionally quiescent state where the genome is highly compacted. When hatched L1s feed, the germline genome decompacts and RNAPII is abruptly and globally activated. A previously documented yet unexplained feature of germline genome activation in the worm is the appearance of numerous DNA breaks coincident with RNAPII transcription. These studies highlighted that ZGA generates genomic instability prior to cell cycle reentry, but critical questions emerged, such as what is the enzyme responsible for the damage, and why would Z2/Z3 risk destroying its genome during ZGA. Here, we have found that those questions are answered by the DNA metabolic enzyme topoisomerase II (TOP-2). Specifically, we show that DNA breaks are induced by TOP-2, that they function to recruit the RUVB complex to chromosomes so that RUVB can decompact the chromatin, and that DNA break- and RUVB-mediated decompaction is required for zygotic genome activation. Following up on this work, we have also started to show that embryonic genome compaction requires TOP-2 and the condensin complex. Finally, we have also asked what cell cycle factors mediate the cell cycle arrest during embryogenesis. These works not only underscore the unique nature of PGC development, but also highlight the importance of global chromatin decompaction to the rapid induction of gene expression, and show that one way cells achieve global decompaction is through programmed DNA breaks.