Citation
Zygote arrest (Zar) protein-mediated translational control in Xenopus Laevis development

Material Information

Title:
Zygote arrest (Zar) protein-mediated translational control in Xenopus Laevis development requirement for the poly(a) tail and protein interactions
Creator:
Cook, Jonathan Michael ( author )
Place of Publication:
Denver, Colo.
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 electronic file (101 pages). : ;

Thesis/Dissertation Information

Degree:
Master's ( Master of Science)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Integrative Biology, CU Denver
Degree Disciplines:
Integrative Biology
Committee Chair:
Charlesworth, Amanda
Committee Members:
Johnson, Aaron
Stith, Bradley

Subjects

Subjects / Keywords:
Transfer RNA ( lcsh )
RNA ( lcsh )
Genetic translation ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Zygote arrest (Zar) proteins, Zar1 and Zar2, are required for successful fertilization and embryogenesis. Synthesis of developmentally important proteins is crucial to embryogenesis and is regulated by translation of mRNA. Zar proteins bind the mRNA of developmentally important proteins via a specific sequence found in the RNA called the Translational Control Sequence (TCS). It is already known that RNAs containing a TCS are translationally repressed in immature oocytes and activated in mature oocytes. It is also known that the N-termini of both Zar1 and Zar2 repress translation when tethered to reporter mRNA. The purpose of this study was to show Zar proteins are bona fide translation factors and are candidates for mediating translational regulation by the TCS. In a dual luciferase tethered assay, both Zar proteins repressed translation up to 50% in immature Xenopus oocytes and repression was relieved during oocyte maturation, consistent with translational regulation of developmentally important mRNAs. Interestingly, Zar1 required the reporter mRNA have a poly(A) tail to repress translation, whereas Zar2 did not. Furthermore, Zar1 and Zar2 interacted with overlapping but distinct sets of proteins. Proteins recovered from GST affinity purifications included many known translation factors, such as CPEB, 4E-T and embryonic poly(A)-binding protein. In particular, eukaryotic initiation factors were identified such as eIF4E-1b. These interactions changed during maturation, coincident with change in Zar function. Together, these data suggest Zar proteins do have roles as translation regulators and may mediate repression by the TCS. They also suggest mechanistic differences between Zar1 and Zar2.
Thesis:
Thesis (M.S.)--University of Colorado Denver. Integrative biology
Bibliography:
Includes bibliographic references.
System Details:
System requirements: Adobe Reader.
General Note:
Department of Integrative Biology
Statement of Responsibility:
by Jonathan Michael Cook.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
911184022 ( OCLC )
ocn911184022

Downloads

This item has the following downloads:


Full Text
ZYGOTE ARREST (ZAR) PROTEIN-MEDIATED TRANSLATIONAL CONTROL
IN XENOPUS LAEVIS DEVELOPMENT: REQUIREMENT FOR THE POLY(A)
TAIL AND PROTEIN INTERACTIONS
by
JONATHAN MICHAEL COOK
B.S., Metropolitan State University of Denver, 2008
B.A., Metropolitan State University of Denver, 2008
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Master of Science
Integrative Biology Program
2015


2015
JONATHAN MICHAEL COOK
ALL RIGHTS RESERVED


This thesis for the Master of Science degree by
Jonathan Michael Cook
has been approved for the
Integrative Biology Program
by
Amanda Charlesworth, Chair
Aaron Johnson
Bradley Stith
April 20, 2015


Cook, Jonathan Michael (M.S., Biology)
Zygote arrest (Zar) protein-mediated translational control in Xenopus development: a
requirement for the poly(A) tail and protein interactions
Thesis directed by Assistant Professor Amanda Charlesworth.
ABSTRACT
Zygote arrest (Zar) proteins, Zarl and Zar2, are required for successful
fertilization and embryogenesis. Synthesis of developmentally important proteins is
crucial to embryogenesis and is regulated by translation of mRNA. Zar proteins bind the
mRNA of developmentally important proteins via a specific sequence found in the RNA
called the Translational Control Sequence (TCS). It is already known that RNAs
containing a TCS are translationally repressed in immature oocytes and activated in
mature oocytes. It is also known that the N-termini of both Zarl and Zar2 repress
translation when tethered to reporter mRNA. The purpose of this study was to show Zar
proteins are bona fide translation factors and are candidates for mediating translational
regulation by the TCS.
In a dual luciferase tethered assay, both Zar proteins repressed translation up to
50% in immature Xenopus oocytes and repression was relieved during oocyte maturation,
consistent with translational regulation of developmentally important mRNAs.
Interestingly, Zarl required the reporter mRNA have a poly(A) tail to repress translation,
whereas Zar2 did not. Furthermore, Zarl and Zar2 interacted with overlapping but
distinct sets of proteins. Proteins recovered from GST affinity purifications included
many known translation factors, such as CPEB, 4E-T and embryonic poly(A)-binding
m


protein. In particular, eukaryotic initiation factors were identified such as eIF4E-lb.
These interactions changed during maturation, coincident with change in Zar function.
Together, these data suggest Zar proteins do have roles as translation regulators
and may mediate repression by the TCS. They also suggest mechanistic differences
between Zarl and Zar2.
The form and content of this abstract are approved. I recommend its publication.
Approved: Amanda Charlesworth
IV


DEDICATION
To
Jonathan David Lee Henley
I love you son.
I am always thinking of you, even when we are not together.
My parents
I treasure the pieces of me that come from each of you.
Kris, Liz, Katie, Randi, Steph and Ben
What an honor to be a member of a tribe
of such varied and amazing individuals.
Sean Bergman
You know more about poly(A) tails than you ever wanted to,
and I am better for sharing this part of the journey with you.
Thank you to all of my family and friends,
whom I love and who love me.
v


ACKNOWLEDGEMENTS
I would first like to acknowledge and thank my advisor and mentor
Amanda Charlesworth
who has taught me so much about science and what it means to be a scientist.
You have changed the way I think.
I would also like to acknowledge my other committee members,
Brad Stith and Aaron Johnson,
as well as other members of the University of Colorado Denver
biology and chemistry departments, who have contributed to my education and to the
development of this project.
Thank you
Kenneth Valles, Courtney Warren and Elana Costanza
for proof-reading and technical support.
Services were provided by Skaggs School of Pharmacy Mass Spectrometry Core Facility
and the University of Colorado Denver DNA Sequencing and Analysis Core.
Funding for this project has come from the American Cancer Society and from the
University of Colorado.
vi


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION...............................................................1
Zygote arrest (Zar) proteins and maternal control of development..............1
Zar proteins are essential for early vertebrate development................1
Gametogenesis, meiotic maturation and embryogenesis........................2
Maternal control and post-transcriptional regulation.......................4
Xenopus laevis as a model for translational regulation........................6
Maternal control of early development in Xenopus laevis....................6
Cytoplasmic polyadenylation and translational regulation...................9
Mechanisms of mRNA translational regulation...........................10
The closed-loop model of translation initiation..........................10
The polyadenylation hexanucleotide and the poly(A) tail............12
Specificity factors: c/.s-elements, /ra//.s-factors, combinatorial control.14
Zar proteins bind mRNA and regulate mRNA translation..........................21
Zar proteins bind the TCS of the maternal mRNAs mos and weel.......21
Both Zar and the TCS repress translation in immature Xenopus oocytes .22
II. HYPOTHESIS................................................................23
Zar proteins are bona fide translation factors and mediate the translation
regulation effects of the Translational Control Sequence (TCS) in Xenopus
laevis oocytes........................................................23
III. METHODS....................................................................24
Cloning and plasmid preparation.......................................24
vii


Oocyte isolation, culture and microinjection............................24
MS2-tethered assay......................................................25
mRNA stabilization assays...............................................27
mRNA polyadenylation assays.............................................28
Western blot............................................................29
GST affinity purification...............................................30
Immunoprecipitation.....................................................31
IV. EXPERIMENTS AM) RESULTS........................................................32
Aim 1. Show a larger effect for Zar-mediated translation repression.....32
Do Zar fusion proteins repress better when stabilized with inter-domain
linkers?.............................................................33
Do MS2-Zar fusion proteins repress better when tethered to reporter
mRNAs with 4 stem-loops rather than 2 stem-loops?....................41
Aim 2: Show the mechanism of Zar-mediated translational regulation is
consistent with that established for the TCS............................44
Do Zarl and Zar2 require a poly(A) tail to repress translation?......46
Do Zarl and Zar2 require the process of polyadenylation to activate
translation during maturation?.......................................50
Aim 3: Show that Zar proteins interact with other translation factors in the
oocyte..................................................................54
What proteins do Zarl and Zar2 interact with in Xenopus oocytes?.....54
Do Zarl and Zar2 interactions with translation factors change during
maturation?..........................................................60
viii


V. DISCUSSION..................................................................67
Major findings.........................................................67
Redesign of MS2-Zar fusion proteins improved stability and function ....68
Zarl required a poly(A) tail to repress translation; Zar2 did not..69
Translation regulation by Zar did not involve polyadenylation.......70
Zar proteins interacted with components of translational regulation
complexes....................................................71
Proposed Models of Zarl and Zar2-mediated translation regulation.72
Proposed models of Zar2-mediated translational regulation....72
Proposed models of Zarl-mediated translational regulation....73
REFERENCES..........................................................74
APPENDIX: Specific Plasmid Construction.............................83
IX


LIST OF FIGURES
FIGURE
1. Representative cartoon of early Xenopus laevis development......................7
2. Cartoon of the closed-loop model of translation initiation......................11
3. Models of CPEB-mediated translational regulation...............................17
4. Cartoon of the 3UTRs of the mos and weel mRNAs................................19
5. Representative cartoon of Zarl and Zar2 protein sequences......................21
6. Cartoon of MS2-fusion proteins and reporter mRNAs used in tethered assay.......26
7. Schematic of GST affinity purification assay...................................30
8. Cartoon of MS2 fusion proteins: old and new constructs..........................34
9. Translation repression by old versus new MS2-Zar fusion proteins................37
10. Dose response of translation repression by MS2-Zar.............................38
11. Translation regulation by Zar during maturation................................39
12. Amount of MS2-Zar protein remaining after maturation...........................40
13. Effect of extra stem-loops on Zar-mediated translational regulation............43
14. Cartoon of luciferase reporter mRNAs of various poly(A) tail lengths...........45
15. Zar-mediated repression on reporters varying poly(A) tail length...............48
16. The effect of polyadenylation on Zar-mediated translational regulation.........53
17. GST-Zarl protein interactions in immature oocytes..............................55
18. GST-Zarl and GST-Zar2 affinity purification during maturation..................59
19. Cartoon of select translation initiation factors...............................61
20. Zar protein interactions with translation initiation factors..................63
21. Proposed models of Zar2-mediated translation regulation.......................72
x


LIST OF TABLES
TABLE
1. Mass spectrometry results from GST-Zarl affinity purification.......58
xi


LIST OF ABBREVIATIONS
4E Initiation factor 4E (eIF4E)
4G Initiation factor 4G (eIF4G)
4E-BP eIF4E-binding protein
4E-T eIF4E-transporter protein
AMP Adenosine monophosphate
AP Alanine/proline rich amino acid linker
AV Animal-vegetal
cAMP Cyclic AMP (adenosine monophosphate)
cdc Cell division cycle protein
cdk Cyclin-dependent kinase
Co-IP Co-immunoprecipitation
CPE Cytoplasmic polyadenylation element
CPEB Cytoplasmic polyadenylation element-binding
CPSF Cleavage and polyadenylation specificity factor
DAZL Deleted in azoospermia-like
DNA Deoxy-ribonucleic acid
elF Eukaryotic translation initiation factor
EK Glutamate/lysine rich amino acid linker
ePAB Embryonic poly(A) binding protein
Erk Extracellular signal-regulated kinase
flue Firefly luciferase
G2-phase 2nd gap phase (cell cycle)
Gd Growth and differentiation factor-9
Gld-2 Germline deficient-2
GS Glycine/serine rich amino acid linker
GST Glutathione S-transferase
GVBD Germinal vesicle breakdown
hex Polyadenylation hexanucleotide
IgG Immunoglobulin G
LDS Lithium dodecyl sulfate
ml GTP 7 methyl guanosine 5 triphosphate
MAPK Mitogen-activated protein kinase
Mater Maternal antigen that embryos require
MBE Musashi binding element
MBT Mid-blastula transition
MEK MAPK/Erk kinase
MPF Maturation promoting factor (M-phase promoting factor)
M-phase Mitotic phase (cell cycle)
miRNA Micro RNA (ribonucleic acid)
mRNA Messenger RNA
mRNP Message ribonucleoprotein complex
MZ1 MS2-Zarl
MZ2 MS2-Zar2
Npm2 Nucleoplasmin protein-2


PABP Poly(A)-binding protein
Paip Poly(A)-bind protein-interacting protein
PAP Poly(A) polymerase
PARN Poly(A)-specific ribonuclease
PAGE Polyacrylimide gel electrophoresis
Patl Topoisomerase Il-associated protein-1
PBE Pumilio binding element
PCR Polymerase chain reaction
PKA Protein kinase A
rluc Renilla luciferase
RNA Ribonucleic acid
RT-PCR Reverse transcription PCR (polymerase chain reaction)
SDS Sodium dodecyl sulfate
siRNA Short interfering RNA
SL Stem-loops
S-phase Synthesis phase (cell cycle)
TCS Translational control sequence
UTR Untranslated region
Vera vgLE-binding and ER association protein A
Xp54 Xenopus protein 54
Xpatla Xenopus protein Patl homolog a
Zar Zygote arrest
ZGA Zygotic genome activation


CHAPTER I
INTRODUCTION
Zygote arrest (Zar) proteins and maternal control of development
Zar proteins are essential for early vertebrate development
The recently described Zygote arrest (Zar) family of proteins comprises Zarl and
Zar2, that can be distinguished from each other based on sequence homology and gene
context (Yamamoto et al., 2013). Zar protein sequences have been found only in
vertebrate species, and are evolutionarily conserved in all vertebrate species where
sequence data is available (Wu et al., 2003b). Zarl and Zar2 are germ-cell specific:
mRNA and protein products are expressed primarily in ovaries, oocytes and eggs in
mammals ((Uzbekova et al., 2006); (Brevini et al., 2004)), birds ((Elis et al., 2008);
(Michailidis et al., 2010)) and fish (Bobe et al., 2008).
Female Zarl knockout mice are grossly normal, but infertile due to a complete
block to embryogenesis (Wu et al., 2003a). No Zar2 knockouts or Zarl/Zar2 double
knockouts have been described, but when dominant negative strategies have been used to
disrupt Zar2 function, embryogenesis is also blocked (Hu et al., 2010).
Zar protein sequences are found in all vertebrates examined and share striking
features of homology, including twelve cysteines whose location and spacing are
completely invariant across all identified Zar sequences ((Sangiorgio et al., 2008);
(Yamamoto et al., 2013)). Because Zar proteins share high levels of sequence homology
in vertebrate animals, it is likely that their function is conserved. While there are
fundamental differences between amphibian and mammalian oogenesis, it is tempting to
speculate that Zar proteins are essential for early development in vertebrates in general.
1


Zar proteins have been identified as markers of germ cells ((Pennetier et al., 2004); (Nath
et al., 2013)) and of developmental competence ((Chen et al., 2014); (Summers et al.,
2014)) in several livestock animals, suggesting they play a role in early development in
these species as well.
No Zar protein homologs have been identified in invertebrates to date, including
in non-vertebrate chordates. Although a couple of sequences have been identified in the
cephalochordate amphioxus and in the tunicate Ciona that show limited homology to Zar
protein sequences (NCBI accession numbers XP_002607077.1 and XP_002130560.1),
neither share all of the features characteristic of Zar sequences.
Gametogenesis, meiotic maturation and embryogenesis
Gametogenesis involves the differentiation and growth of an immature germ cell
followed by 1-2 rounds of meiotic division to yield functional gamete cells.
Spermatocytes and oocytes have distinct morphologies and play very different roles in
fertilization, and so their mechanisms of gametogenesis are quite different from one
another.
The primary role of the spermatocyte is to migrate to the egg and fuse with the
plasma membrane, delivering genetic information and triggering fertilization events in
the egg. The mature spermatocyte needs to be able to produce a large amount of energy
to power movement and needs modifications to protect and deliver its genetic cargo, but
does not need to maintain metabolism for an extended duration of time. The mature
spermatocyte contains: a single large, coiled mitochondria to power to the flagella, an
acrosome involved in plasma membrane fusion, a nucleus crystallized with protamines
2


and very little cytoplasm. Meiosis is symmetric and two rounds of meiotic division
produce four functional spermatozoa.
The egg (mature oocyte) does not have the same motility requirements as the
sperm, but must maintain its metabolism beyond fertilization and provide the material
and information necessary for the early stages of embryogenesis. The oocyte cytoplasm
is large and complex compared to that of the sperm, with yolk proteins, a complete set of
organelles to maintain the metabolism of the large cell and factors required for the
regulation of fertilization and early embryogenesis events. The immature oocyte arrests
in the early stages of meiosis I until the appropriate cue, and then must reenter the cell
cycle to become a fertilizable gamete in a process known as meiotic maturation. Meiosis
occurs asymmetrically, producing one large functional egg cell with cytoplasm and two
smaller polar bodies without cytoplasm that are discarded.
In most vertebrates the mature oocyte (egg) is arrested in meiosis II until
fertilization by the sperm, while in invertebrates, sperm entry occurs at other points
during meiotic maturation, such as the primary oocyte stage or after the completion of
meiosis II (Tadros and Lipshitz, 2009). Fertilization is the fusion of the sperm and the
egg plasma membranes and nuclei, resulting in the combination of the genetic material
and activation of the eggs developmental program. Upon fertilization the egg completes
meiosis II and the one-cell embryo undergoes the first mitotic division.
In the frog, the one-cell embryo, or zygote, enters several rounds of synchronous
cleavage (mitotic divisions and cytokinesis) to become an embryo composed of a ball of
cells called a blastula. When the embryo reaches a specific size and cell number (which
varies by species), the cell cycles become asynchronous and cell movement, called
3


gastrulation, occurs (Masui and Wang, 1998). This process results in the formation of the
gut and tissues within the embryo, which then undergoes neurulation and later stages of
embryogenesis.
Maternal control and post-transcriptional regulation
DNA transcription is largely inactivated during meiosis and does not resume until
after the two haploid pronuclei fuse followed by the activation of the new diploid genome
of the zygote. At this point, transcription increases and genes from both maternal and
paternal chromosomes are transcribed in a regulated fashion. Until zygotic genome
activation (ZGA), regulation of gene expression is under maternal control and mediated
by factors present in the oocyte before in a dormant state until the time that their activities
are required (Farley and Ryder, 2008).
These factors include maternal mRNAs that are transcribed from the genome of
the vertebrate oocyte well before fertilization and are post-transcriptionally regulated
throughout meiosis and fertilization. In some animals, particularly invertebrates, there is
significant maternal contribution from neighboring somatic cells. Maternal mRNAs
regulate cell-cycle and patterning in the absence of transcription, until the zygotic
genome is activated. Early embryogenesis, including early cleavages and the actual
activation of the zygotic genome, is under maternal control, as there are primarily
maternal transcripts present at this time. The timing, rate and localization of transcript
expression is regulated by other factors in the cell, such as protein translation factors
(Colegrove-Otero et al., 2005)
An example of translational regulation under maternal control comes from
anterior-posterior axis formation in Drosophila embryos, which depends on the initial
4


polarity of the oocyte that is established well before fertilization. Proper posterior and
anterior development require the transcription factors Caudal and Hunchback
respectively. While the caudal and hunchback mRNAs are generally distributed about
the oocyte, their protein products are expressed in gradients across the cell because of
localization of the maternal translation factors, bicoid and nanos. Establishment of these
gradients in the oocyte is essential for proper axis formation in the transcriptionally active
embryo (Lasko, 2011). It should be noted that bicoid also functions as a transcription
factor in Drosphila embryogenesis. Maternal control of development has been described
in vertebrate species as well, such as in zebrafish ((Bontems et al., 2009); (Lindeman and
Pelegri, 2010)) and in mouse ((Minami et al., 2007); (Kim and Lee, 2014)).
The zarl gene was first identified in a subtractive hybridization screen for
maternal effect genes using Gdf9 (growth and differentiation factor-9) knockout mice
(Wu et al., 2003a). Folliculogenesis is disrupted in these mice so that maternal effect
gene products accumulate above wild-type levels (Carabatsos et al., 1998). The zarl
gene and several other mammalian maternal effect genes showed altered expression in
this Gdf9 knockout (Matzuk, 2000).
When Zarl was knocked out, most of the zygotes arrested before the first cell
division in either S- or G2-phase and none of the embryos successfully progressed
beyond the 2-cell stage (Wu et al., 2003a). In another study, 1-cell mouse embryos were
injected with a dominant negative form of Zar2, a similar phenotype was observed, and
the cells arrest within one cleavage (Hu et al., 2010). In both of these cases, the zygotic
genome was not activated. Therefore, Zar is regulating maternal effects.
5


Xenoyus laevis as a model for translational regulation
Xenopus laevis, the African clawed frog, is a common model organism for
studying early development. The frogs are available in large numbers and are easy to
rear in the laboratory and produce large numbers of oocytes and eggs that can be
collected year-round. The mechanisms that regulate the involved processes are highly
conserved among animals, and the frogs oocytes and eggs are very large, making them
easy to manipulate. The oocytes are about a millimeter in diameter and a microliter in
volume so RNA and protein can be easily injected into the oocytes and large biochemical
samples can be collected. Due to its pseudotetraploid nature, Xenopus is not a good
model system for studying genetics because most genes have a redundant copy that
makes knockouts and transgenics difficult. However, it is a good model for studying the
cell cycle and early vertebrate development from a biochemical perspective.
Maternal control of early development in Xenopus laevis
During early oogenesis, maternal mRNAs, and the proteins that regulate them, are
synthesized in the growing oocyte, while transcription is active. Xenopus oogenesis is
divided into six stages in the Dumont classification system represented by Roman
numerals I-VI (Dumont, 1972). At all stages, the immature Xenopus oocyte is arrested in
prophase of meiosis I. A progesterone hormone signal induces the immature oocyte to
reenter the cell cycle to become a mature oocyte, capable of fertilization (fig. 1).
Immature Xenopus oocytes are darkly pigmented in the hemisphere of the animal
pole and more lightly colored in the vegetal hemisphere. As the oocyte reaches
metaphase I, the nucleus, or germinal vesicle, breaks down (referred to as GVBD) and
tthere is a redistribution of cortical granules, resulting in the formation of a white spot at
6


the animal pole. This white spot is a convenient marker for meiotic progression and
events that occur during meiotic maturation are often grouped as to whether they occur
before (early) or after (late) GVBD.
DNA transcription is largely shut down before meiotic maturation and doesnt
resume until the zygotic genome is activated in the blastula-stage embryo and so Xenopus
gene expression is under maternal control during this period. Zygotic genome activation
doesnt occur in Xenopus until after 12 rounds of mitosis, making this an ideal system for
studying the role of translational regulation during development.
Transcription Translation Only Transcription
Progesterone
jl II III IV V_______________W
Immature Oocyte
Meiosis
Metaphase I
m
GVBD
Fertilization
t
Metaphase II
Mature
Oocyte
MBT
Embryogenesis -l

2-Cell Stage 8
Embryo Blastula

Gastrula
Figure 1. Representative cartoon of early Xenopus laevis development.
The images show representative anatomy of the oocyte and embryo at different stages of
oogenesis, meiosis and embryogenesis. Anatomical stages are labeled below (Roman
numerals, Dumont stages of oogenesis; GVBD, germinal vesicle breakdown) and
relevant meiotic stages are labeled immediately above images. Major signaling events
are marked by red arrows (MBT, mid-blastula transition) and dominant form of gene
expression is indicated by upper arrow (light yellow, transcription; dark green,
translation).
Upon recognition of a progesterone signal, there is a decrease in cyclic adenosine
monophosphate (cAMP) levels and protein kinase A (PKA) signaling that ultimately
results in the activation of maturation promoting factor (MPF) (Ferrell, 1999). MPF is
composed in part of cyclin B and cdkl (cyclin-dependent kinase 1) and is responsible for
both the resumption of meiosis I and the arrest of the cell cycle at metaphase II (Masui,
7


2001). This activation of MPF is brought about in part through a mitogen-activated
protein kinase (MAPK) cascade (Mailer et al., 2002).
The mos mRNA is translationally activated so that the Mos protein, which is a
MAP kinase kinase kinase, is synthesized, initiating a phosphorylation cascade that leads
to the activation of cyclin B/cdkl and promotion of the cell cycle (Tunquist and Mailer,
2003). Mos mRNA is translationally activated early during maturation, before GVBD
(Charlesworth et al., 2002). Mos activity is further enhanced and stabilized by a MAPK-
dependent positive feedback loop ((Howard et al., 1999); (Nishizawa et al., 1992)). Mos
activity is also required after GVBD to maintain the metaphase II cell cycle arrest
((Yamamoto et al., 2005); (Dupre et al., 2002)).
Another example of a maternal mRNA that plays an important role in the cell
cycle is weel. Weel protein inhibits cdkl activity and extends the length of the first
mitotic M-phase to allow for pronuclear fusion and completion of fertilization in the first
cycle (Murakami et al., 1999). Weel also plays an important role in gastrulation, when
the cell-cycles of the embryo become asynchronous and mass cell movements begin,
extending the cell cycles of some cells so they divide slower than their neighbors, leading
to the necessary contortion of tissues (Murakami et al., 2004). Weel is translationally
activated late during meiotic maturation, after GVBD, but before fertilization
(Charlesworth et al., 2000).
The maternal proteins Mos and Weel play important roles in cell cycle regulation
across metazoa, in vertebrates such as Xenopus (Murakami et al., 2004) and mouse
((Zhao et al., 1991); (Tominaga et al., 2006)), as well as in invertebrates ((Amiel et al.,
2009); (Stumpff et al., 2005)).
8


Cytoplasmic polyadenylation and translational regulation
Because DNA transcription is largely shut down during meiosis and early
embryogenesis in Xenopus, proteins required to regulate developmental events are
translated from pre-existing maternal mRNAs that have been stored in a dormant state in
the egg. It has been noted that, at least in early development, the level of translation
activation from an mRNA is positively correlated to the length of the poly(A) tail of the
mRNA. RNAs with longer poly(A) tails are translated more robustly than RNAs with
shorter poly (A) tails.
Cytoplasmic polyadenylation is distinct from the nuclear polyadenylation that
occurs just after transcription. Immature mRNAs are polyadenylated in the nucleus, but
after processing and export, maternally regulated mRNAs undergo specific deadenylation
so that they have a short poly(A) tail and are translated at a low basal level. In many
maternally regulated mRNAs, this deadenylation is driven by a sequence within the 3-
UTR of the mRNA called the cytoplasmic polyadenylation element (CPE). Extension of
the poly(A) tail results in translation above basal levels. Cytoplasmic polyadenylation
plays a prominent role in translational regulation across metazoa and has been well
studied and characterized in Xenopus laevis, as it is a good biochemical model.
Both mos and weel mRNAs are polyadenylated coincident with their translational
activation ((Prasad et al., 2008); (Wang et al., 2008)). While mos is polyadenylated and
activated before GVBD, weel is polyadenylated and activated after GVBD. In Xenopus
the protein Mos contributes to the signaling that brings about GVBD, while the protein
Weel would inhibit GVBD if translated prior to the completion of meiosis I.
9


Mechanisms of mRNA translational regulation
The closed-loop model of translation initiation
Translation consists of three phases: initiation, elongation and termination. Any
of these phases can be regulated to control the rate of protein production, but most
mechanisms of translational regulation involve the disruption or facilitation of the
initiation step. Translation initiation consists of recruitment of the ribosome to the
mRNA, remodeling the 5-untranslated region (5UTR) of the mRNA and recognition of
an initiation codon. Recruitment of the ribosome to the mRNA is a multi-step process
that involves several proteins that cooperatively interact with the mRNA and the subunits
of the ribosome itself (Aitken and Lorsch, 2012). Asa result, this process of ribosome
recruitment offers many possibilities for translation control.
It is thought that ribosome recruitment is facilitated by a circularization of the
mRNA brought about by the interactions of a conserved family of eukaryotic translation
initiation factors (elFs) that interact with the poly(A) tail of the 3UTR and to the
methylated cap (m7 GTP) of the 5UTR. Evidence for this circularization comes from a
synergistic effect on translation activation when both the 5-cap and the 3-poly(A) tail
are present compared to when either one alone is present (Gallie, 1991). Further
evidence comes from electron micrographs of rough endoplasmic reticulum showing a
predominance of circular polysomes (Christensen et al., 1986).
The circularization is primarily mediated by poly(A)-binding protein (PABP) and
elFs. In the Xenopus oocyte, the predominate isoform of PABP is embryonic poly(A)-
binding protein (ePAB). ePAB binds to the polyadenosine tracts of mRNA and when
bound to the poly(A) tail, it has a strong affinity for eIF4G, a scaffolding protein that also
10


binds to the protein eIF4E (fig. 2B). eIF4E binds both to eIF4G and to the 5-cap of the
mRNA.
m7 GTP
3
4E
Translationally Inactive
Translationally Active
Figure 2. Cartoon of the closed-loop model of translation initiation.
The commonly accepted view of translation initiation involves the binding of several
canonical translation factors to circularize the mRNA. A. The translationally inactive
mRNA has a short poly(A) tail (each A represents approximately 5 adenylate residues)
and recruits relatively few ePAB molecules to the mRNA (4e-BP, eIF4E-binding protein;
4E, eIF4E; m7 GTP, methyl-guanosine cap; AUG, translation initiation start codon; stop,
stop codon; hexagon, polyadenylation hexanucleotide; A-tract, 3 poly(A) tail; ePAB,
embryonic poly(A)-binding protein). B. Translationally active mRNAs generally have
longer poly(A) tails and bind many ePAB molecules, recruiting other translation factors
(labeled 3, 4A, 4E and 4G) and creating a protein bridge between the 5 and 3 ends of
the mRNA, facilitating the binding of the 40S ribosome.
The interaction between ePAB and eIF4G strengthens the interactions between
eIF4G and eIF4E, as well as the interaction between eIF4E and the 5-cap of the mRNA.
This link of RNA:protein and protein:protein interactions results in the 5- and 3- ends of
the mRNA being brought together in a closed loop (Kahvejian et al., 2001).
In addition to binding ePAB and the cap-binding protein eIF4E, eIF4G also binds
eIF4A and eIF3. eIF4A is thought to be involved in the remodeling of the 5UTR that is
necessary for 40S ribosome loading and may also assist the ribosomal subunit along the
mRNA while it scans for the initiation codon. eIF3 interacts directly with subunits of the
40S ribosome and facilitates ribosomal loading. Once the 40S ribosome recognizes the
initiation codon, most of the elFs are released, the 60S ribosomal subunit joins and
translation begins.
11


One notable example of modulators of translation initiation are the eIF4E-binding
proteins, or 4E-BPs (fig. 2A), which bind to eIF4E and compete for binding to eIF4G,
thereby preventing the formation of the closed-loop translation initiation complex and
preventing translation (de Moor et al., 2005). 4E-BPs can be phosphorylated, changing
their interactions with eIF4E, and coupling translation regulation to signaling pathways
that regulate early development. Disruption of any of these elFs or their protein
interactions can prevent translation and most known translation factors modulate these
interactions and the formation of a closed loop structure. Translation activators
generally promote this closed-loop formation and translation repressors generally inhibit
its formation.
The polyadenylation hexanucleotide and the poly(A) tail
This closed-loop model of translation proposes that mRNAs that are
translationally silenced have short poly(A) tails and are not circularized, while mRNAs
that are translationally active have longer poly(A) tails and are circularized. This
suggests that the length of the poly(A) tail is correlated to the amount of protein produced
from the mRNA and that controlling the size of the poly(A) tail is a potential mechanism
for regulating translation initiation. There is strong evidence for a correlation between
poly(A) tail length and amount of translation repression or activation in Xenopus oocytes
(Gray et al., 2000).
After an immature mRNA is transcribed, it is capped and polyadenylated before
being exported from the nucleus as a mature mRNA. In the 3UTR of the mRNA is the
polyadenylation hexanucleotide (hex), with a consensus sequence of AAUAAA. The
cleavage and polyadenylation specificity factor (CPSF) complex binds to the hex
12


sequence in the mRNA and contains an endonuclease that cleaves the mRNA 3 of the
hex site. CPSF also recruits nuclear poly(A) polymerases (PAPs) to the mRNA and the
poly(A) tail is greatly extended to about 250 adenosine residues immediately following
the polyadenylation hexanucleotide sequence in a process called nuclear polyadenylation
(Fox et al., 1989). This is the approximate length of the poly(A) tail for most mature
mRNAs exported from the nucleus.
Approximately one ePAB molecule can bind per 25 adenosine residues. This
means that the longer the poly(A) tail, the more ePAB molecules can bind the mRNA. It
is thought that the larger the number of ePAB molecules bound to the mRNA, the greater
the enhancement of elF interactions and the greater the level of translation. The poly(A)
tract of a typical mRNA exported from the nucleus is long enough to bind several ePAB
molecules and the mRNA is translationally active. It is thought that the default state for
mRNAs exiting the nucleus is to be translated and mRNAs that are not yet to be
translated are silenced through deadenylation. This allows for decoupling of the timing
of transcription and translation and allows for the coordinated regulation of gene
expression in an environment where transcription is not active, such as in the maturing
oocyte. Deadenylation of maternally regulated mRNAs and their transient translation
silencing after nuclear export has been observed for mouse oocytes (Huarte et al., 1992)
and like in Xenopus (McGrew et al., 1989), translational activation is correlated with
polyadenylation.
Deadenylation of mRNAs to be silenced is brought about by deadenylases, such
as the poly(A)-specific ribonuclease (PARN) (Copeland and Wormington, 2001). When
PARN is bound to the mRNA, the poly(A) tail is shortened to about 30 adenosines or less
13


in many repressed mRNAs. This means that between zero and a couple of ePAB
molecules can bind to shortened poly(A) tail, which is not enough to stimulate translation
initiation. When the mRNA is to be translationally activated, the poly(A) tail is extended
through interactions between components of the same CPSF complex and the AAUAAA
hex sequence that were involved in nuclear polyadenylation. CPSF binds to the hex
sequence and recruits the protein Germ line deficient-2 (Gld-2), which is a poly(A)
polymerase that is required for cytoplasmic polyadenylation in oocytes.
When the polyadenylation hexanucleotide sequence is mutated, CPSF does not
bind to the mRNA, polyadenylation does not occur and translation of the mRNA is
generally reduced. This suggests that regulation of the poly(A) tail length can regulate
the level of translation and that the AAUAAA hex sequence is necessary for regulation of
poly(A) tail length ((Fox et al., 1989); (Paris and Richter, 1990)).
There is also an antagonist relationship between poly(A) polymerases such as
Gld-2 and deadenylases such as PARN. When Gld-2 function is disrupted, the poly(A)
tail remains short and the mRNA is poorly translated and when PARN function is
disrupted, the poly(A) tail remains long and the mRNA is translated more robustly.
Modulation of the interactions between CPSF and Gld-2 and those between CPSF and
PARN is an important mechanism for regulating translation initiation.
Specificity factors: cis-elements, trans-factors, combinatorial control
The polyadenylation hexanucleotide (AAUAAA) that recruits CPSF to the
mRNA is present in all mRNAs except for those of histones, but the mRNAs that are
present in the cell have very different poly(A) tail lengths that are specific to mRNA
species. This means that regulation of the binding of CPSF to the hex sequence cannot
14


be the only mechanism of regulating the length of the poly(A) tail, and that there must be
other cytoplasmic polyadenylation specificity factors that modulate the activity of CPSF.
Several such factors have been identified that change the activity of CPSF, ePAB,
and elFs to regulate the levels of polyadenylation and translational activation of maternal
mRNAs in the developing oocyte. These translation factors will be grouped into two
categories for the purpose of this discussion: c/.s-elements and trans-factors. Sequences
within the mRNA itself, such as the AAUAAA hex sequence that regulate translation, are
c/.s-elements (as they are in the same molecule that is being regulated) that recruit other
molecules that can regulate translation, such as proteins or small RNAs known as trans-
factors (as they are regulating a different molecule). While micro-RNAs (miRNAs) and
short interfering RNAs (siRNAs) have been shown to play a role in translation regulation,
this discussion will focus exclusively on protein trans-factors. The c/.s-element mRNA
sequence and its associated trans-factor binding protein work together to regulate the
translation of the mRNA.
Several c/.s-elements have been identified in maternal mRNAs that are
developmental^ regulated, most being located in the 3UTR within 100-200 nucleotides
of the hex sequence and therefore, the site of polyadenylation. These include: the
cytoplasmic polyadenylation element (CPE), the Musashi binding element (MBE), the
Pumilio binding element (PBE) and the translational control sequence (TCS). The CPE
recruits CPE-binding protein (CPEB), the MBE recruits Musashi, the PBE recruits
Pumilio and the TCS recruits Zar.
15


CPEB: the first 'and only' specificity factor
The CPE and its RNA-binding protein CPEB are the most well characterized of
the cytoplasmic polyadenylation specificity factors. For a long time, it was thought that
CPEB was the only translation factor that played a role in cytoplasmic polyadenylation in
oocytes. While this is no longer considered to be true, CPEB remains the most well
understood specificity factor in oocytes and is involved in many mechanisms of
translation regulation.
CPEB binds to the CPE sequence (consensus sequence U4-6A1-2U1-2), which
imparts deadenylation and translation repression in immature oocytes and
polyadenylation and translation activation during meiotic maturation ((Hake and Richter,
1994); (Charlesworth et al., 2004)). CPEB is involved in the regulation of many mRNAs
and has been implicated in several distinct mechanisms of translation repression during
early development. CPE-mediated translational regulation mechanisms have also been
shown to play a role in synaptic plasticity and cellular senescence in somatic cells
(Richter, 2007). As a result, there are several proposed models that explain CPEB
function, in terms of the translation factors effect on cytoplasmic polyadenylation and
disruption of elF interactions. Two of these models include: the Maskin model and the
opposing polymerase-deadenylase model.
16


Translationally Inactive Translationally Active
Figure 3. Models of CPEB-mediated translational regulation.
CPEB has been implicated in several mechanisms of translational regulation, including
the Maskin model and the opposing deadenylase-polymerase model. A. In some
translation regulating complexes, CPEB recruits Maskin, preventing 4E:4G interactions
in immature oocytes. B. In mature oocytes, CPEB is phosphorylated leading to
conformational changes in Maskin causing it to release 4E. C. In some complexes CPEB
recruits the deadenylase PARN in immature oocytes, preventing elongation of the
poly(A) tail and minimizing the number of ePAB molecules recruited to the mRNA. D.
In mature oocytes, the Gld-2 that is recruited by CPEB leads to polyadenylation of the
mRNA.
In the Maskin model (fig. 3 A-B), CPEB recruits the scaffold protein Symplekin
and the eIF4E-binding protein Maskin to the mRNA. Symplekin acts as a scaffold and
allows other translation factors to dock once recruited to CPE-containing mRNAs.
Maskin acts like a 4E-BP and prevents eIF4E from interacting with 4G, thereby
preventing translation initiation. During meiotic maturation CPEB protein is
phosphorylated, which results in a conformational change in Maskin that causes it to
release eIF4E and leads to translational activation in the maturing oocyte (Stebbins-Boaz
et al., 1999).
17


In the opposing polymerase-deadenylase model, both the poly(A) polymerase
Gld-2 and the deadenylase PARN are localized to CPE-containing mRNAs via CPEB and
Symplekin. While Gld-2 functions to extend the poly(A) tail ((Benoit et al., 2008);
(Kwak et al., 2004)), PARN functions to remove the poly(A) tail (Kim and Richter,
2006). By localizing both enzymes to the same mRNA, they antagonize each other and
the poly(A) tail remains short enough to favor translation repression in immature oocytes.
During meiotic maturation though, the phosphorylation and degradation of CPEB result
in the ejection of PARN from the complex and Gld-2 is allowed to extend the length of
the poly(A) tail and translation is activated (fig. 3C-D).
The MBE, PBE and TCS: the introduction of combinatorial control
The maternal protein Musashi binds the MBE sequence ((G/A)Ui-3AGU) and
represses translation in immature oocytes. Musashi and the MBE both activate
cytoplasmic polyadenylation and translation early during meiotic maturation. An
important maternal mRNA that is regulated in part by Musashi and the MBE is mos
(Charlesworth et al., 2006). It should be noted that there appear to be fundamental
differences in the way that Mos protein expression is regulated among different model
organisms, such as between frog and mouse (Prasad et al., 2008).
Another maternal trans-factor that regulates cytoplasmic polyadenylation and
translation is Pumilio protein, which binds to the PBE (UGUAU(A/U)UAU) and
contributes to CPE-mediated translation repression (Pique et al., 2008).
The Translational Control Sequence (TCS) is a c/.s-element found in the 3UTRs
of some developmental^ important mRNAs. The TCS has been shown to confer
repression of mRNA in immature oocytes and activation of translation in mature eggs
18


(Wang et al., 2008). In isolation, the TCS confers early polyadenylation to reporter
mRNAs and early translational activation that is before GVBD.
A
MBE CPE hex TCS
mos 3' UTR
CPE TCS TCS CPE hex CPE
weel 3' UTR
Figure 4. Cartoon of the 3UTRs of the mos and weel mRNAs.
The diagrams show the location of known c/.s-elements relative to the polyadenylation
hexanucleotide (oval, MBE; circle, CPE; hexagon, polyadenylation hexanucleotide;
square, TCS). Combinations of these multiple c/.s-elements contribute to the timing and
extent of protein translation from these mRNAs. A. The mos mRNA includes an MBE, a
CPE and a TCS. B. The weel mRNA includes three CPEs and two TCSs.
The discovery of multiple /ra//.s-factors mediating the effects of multiple cis-
elements in developmental^ regulated mRNAs introduced the notion of combinatorial
code as an important component of the coordination of timing and extent of mRNA
translation (Pique et al., 2008). In isolation, each cA-element has certain properties in
terms of the extent to which it represses or activates translation and in terms of when
during development its function changes. Most maternal mRNAs have several different
c/.s-elements though, with different numbers of each element and different arrangements
of elements within the UTRs of the mRNAs (fig. 4). This allows for intricate translation
regulation, with built-in redundancies and versatility.
19


Xpatla, Xp54, DAZL, and Vera: a growing list of translation regulators
In addition to the polyadenylation specificity factors just described, there are
many other proteins that contribute to the regulation of translation after being recruited to
specific mRNAs without changing the length of the poly(A) tail. The translation
regulators Xpatla and Xp54 work together, along with Rap55, to regulate translation in
Xenopus oocytes (Marnef et al., 2009). Xpatla is a scaffold protein that is recruited to
the mRNA by an unknown specificity factor, which binds to and recruits Xp54 to the
RNA. Xp54 binds eIF4E in immature oocytes (Minshall and Standart, 2004), and it is
possible that Xp54 disrupts the eIF4E:5-cap interaction. During meiotic maturation
Xpatla is degraded (Nakamura et al., 2010), which could release Xp54 and allowing
eIF4e to bind the 5-cap.
The DAZ family of proteins are necessary for germ cell development (Xu et al.,
2001) and regulate translation (Smith et al., 2011). DAZL proteins bind mRNA via GUU
triplets (Brook et al., 2009) and also bind 4G, strengthening the interaction with 4E and
enhancing translation. This has a similar overall effect to that of ePAB.
In addition to regulating the extent and timing of mRNA translation, the location
of mRNA translation is tightly regulated. Vera is an RNA-binding protein that interacts
with kinesins and the microtubule network to localize mRNAs to specific locations in the
oocyte (Choo et al., 2005). Vera-localized mRNAs include vegT (Kwon et al., 2002) and
vgl (Deshler et al., 1997) in Xenopus and cyclinB (Takahashi et al., 2014) in zebrafish.
Because Vera localizes mRNAs that it recognizes, it indirectly localizes all of the
proteins and translation machinery (Arthur et al., 2009). This allows the extent of mRNA
translation to be regulated in both time and space.
20


Zar proteins bind mRNA and regulate mRNA translation
Zar proteins bind the TCS of the maternal mRNAs mos and weel
In a yeast 3-hybrid screen to identify proteins that bind to the 3UTR of the mos
mRNA, two protein sequences were identified: 1) Musashi, the binding partner of the
MBE, a c/.s-element found in the 3UTR of the mos mRNA and 2) the C-terminal of Zar2
(Charlesworth et al., 2006). Zar2 was then shown to immunoprecipitate both mos and
weel mRNAs inXenopus oocytes (Charlesworth et al., 2012).
Both Zarl and Zar2 bind the Translational Control Sequence (TCS) in the 3UTR
of the mos and weel mRNAs in vitro ((Charlesworth et al., 2012); (Yamamoto et al.,
2013)). Additionally, Zar proteins bind to the TCS when inserted in a (3-globin 3UTR,
which does not contain any known trans-factor binding c/.s-elements.
Figure 5. Representative cartoon of Zarl and Zar2 protein sequences.
Zar primary protein structure shows areas of high sequence homology (gray shaded
regions) that are proposed to be involved in translation regulation (N-terminus) and
RNA-binding (C-terminus) functions and can be generalized to both Zarl and Zar2.
Invariant cysteines thought to be involved in RNA binding are highlighted in yellow.
Zar proteins contain two regions that are well conserved across vertebrate species,
one region in the N-terminal half of the protein and one in the C-terminal half of the
protein (fig. 5). The N-terminus of the protein represses translation in the tethered assay
as does full-length Zar protein and the C-terminus binds mRNA whereas the N-terminus
does not ((Charlesworth et al., 2012); (Yamamoto et al., 2013)).
N-
-C
Translational
Control
Domain
mRNA
binding
domain
21


Both Zar and the TCS repress translation in immature Xenopus oocytes
Zar proteins have previously been thought to be transcriptional regulators.
However, Zar proteins bind mRNA, both in vitro and in vivo, and there is evidence that
they regulate translation of reporter mRNAs, suggesting they may have a role as
translational regulators. When MS2-Zar fusion proteins are tethered to a reporter RNA,
N-terminals of both Zarl and Zar2 reduce the amount of protein produced without
affecting the stability of the RNA in immature oocytes ((Charlesworth et al., 2012);
(Yamamoto et al., 2013)). They repress translation of reporter mRNAs in immature
oocytes and repression is relieved during meiotic maturation. This is consistent with
what is known about TCS- mediated translation regulation, both in terms of repression
and a change in function during meiotic maturation.
While Zar proteins have been shown to repress translation when tethered to
reporter mRNAs in Xenopus oocytes, the effect demonstrated has been small, with Zar
proteins repressing translation by only about 30%. While this effect is reproducible and
dose-responsive, it would provide greater support for Zar being a translation repressor if
the effect was more robust, equal in magnitude as that seen for the TCS (about 50%
repression when compared to a reporter without a TCS) and similar to other translation
repressors such as Xp54.
22


CHAPTER II
HYPOTHESIS
Zar proteins are bona fide translation factors and mediate the translation regulation
effects of the Translational Control Sequence (TCS) in Xenopus laevis oocytes
The following aims were established to test this hypothesis:
Aim 1. Show a larger effect for Zar-mediated translation repression.
Do Zar fusion proteins repress better when stabilized with inter-domain linkers?
Do MS2-Zar fusion proteins repress better when tethered to reporter mRNAs with
4 stem-loops rather than 2 stem-loops?
Aim 2: Show the mechanism of Zar-mediated translational regulation is consistent with
that established for the TCS.
Do Zarl and Zar2 require a poly(A) tail to repress translation?
Do Zarl and Zar2 require the process of polyadenylation to activate translation
during maturation?
Aim 3: Show that Zar proteins interact with other translation factors in the oocyte.
What proteins do Zarl and Zar2 interact with in Xenopus oocytes?
Do Zarl and Zar2 interactions with translation factors change during maturation?
23


CHAPTER III
METHODS
Cloning and plasmid preparation
All restriction enzymes, Klenow fragment and Quick Ligase were obtained from
New England Biolabs. DNA oligonucleotides were synthesized by Integrated DNA
Technologies and Life Technologies. Due to the long list of plasmids constructed for this
study, descriptions of specific plasmids are in APPENDIX. Competent bacteria were
prepared using chemical methods based on (Hanahan et al., 1991) using OneShot ToplO
E. coli (Life Technologies) and transformed according to manufacturers instructions.
All plasmids were sequenced to verify integrity using the University of Colorado
Denver DNA Sequencing and Analysis Core. For in vitro transcription, all plasmids
were linearized with PstI unless otherwise noted. 5' capped RNA was synthesized in vitro
with SP6 mMessage mMachine transcription kit (Life Technologies). RNA quality was
assessed using gel electrophoresis. Nucleotide and amino acid sequence alignments were
performed with Mac Vector 11.1.2.
Oocyte isolation, culture and microiniection
Adult female Xenopus laevis (Nasco) were housed and sacrificed according to
internationally recognized guidelines and with the approval of the University of Colorado
Denver Institutional Animal Care and Use Committee. Oocytes were isolated and
cultured as has been described (Machaca and Haun, 2002). All incubations were carried
out in 0.5X L-15 (MediaTech, Inc.) with penicillin (100 mg/ml) and streptomycin (50
mg/ml). Dumont stage VI (Dumont, 1972) oocytes were selected and injected with 23 nl
24


of the appropriate mRNA using a Drummond Nanolnject II microinjector. Oocytes were
incubated to allow for in vivo protein translation. Appropriate samples were then induced
to mature with 2 \\M progesterone and immature samples were time-matched to
progesterone-treated samples (Charlesworth et al., 2012).
MS2-tethered assay
To test whether Zar proteins regulate translation, an MS2-tethered assay was used
(fig. 6). The viral protein tag MS2 binds with high affinity to a specific primary RNA
sequence that forms a secondary stem-loop structure (Bardwell and Wickens, 1990).
This stem-loop sequence was inserted into a reporter mRNA and the MS2 protein was
fused to Zar, thereby tethering Zar to the reporter mRNA (Coller and Wickens, 2002).
The reporter mRNA used codes for luciferase, an enzyme that produces light with its
substrate, directly proportional to the amount of luciferase protein present. The luciferase
coding region resides in a (3-globin UTR, which is not known to be translationally
regulated during development so that any translational regulation effects on the reporter
mRNA should be due solely to the tethered MS2 fusion protein.
There were two luciferase coding mRNAs used in this assay: one codes for firefly
luciferase and contains the MS2-binding stem-loops and the other codes Renilla
luciferase and does not contain the stem-loops (fig 6B). Because firefly and Renilla
luciferases react with distinct substrates, the mRNAs were co-injected into the same
oocyte and the amounts of protein translated from each was measured separately. The
stem-loop containing, firefly luciferase-coding mRNA was affected by the tethered
protein, while the Renilla luciferase-coding mRNA, without stem-loops, was the loading
control and should not have been affected by the fusion protein.
25


B
MS2 translation ? MS2 Zar
Neaative contr al for translation reaulation m GTP firefly A33C17
MS2 Xp54
3ositive control for translation
MS2 Zarl

MS2 Zar2
Experimental treatments
fluc-2SL-A33C17 (Reporter mRNA)
m7GTP-
Renilla
rluc-A33C17 (Loading control mRNA)
- A33Ci-
1
p-globin luciferase p-globin poly(A)
5' UTR coding 3' UTR tail
c
MS2 fusion
mRNA
______________i
Progesterone
Samples collected
Luciferase
mRNA
i
Matured oocyte (P)
Immature oocyte (I)
MS2 protein expression
~16 hours
Luciferase expression
~11 hours
Figure 6. Cartoon of MS2-fusion proteins and reporter mRNAs used in tethered assay.
This list of fusion proteins and luciferase-coding mRNAs is representative of the
constructs used in this study to investigate Zar-mediated translation regulation. Modified
constructs are described in more detail where appropriate. A. MS2-fusion proteins
contain an MS2 stem-loop binding domain attached to the protein to be tethered to the
reporter mRNA to evaluate translational regulation function (Gray shading, MS2 RNA-
binding domain; white shading, protein domain to be evaluated for translation regulation
activity). MS2 alone is a negative control and MS2-Xp54 is a positive control for
translation repression. B. Luciferase coding mRNAs include either firefly or Renilla
luciferase coding region in a P-globin UTR and either contain MS2-binding stem-loops
or do not. Firefly luciferase with stem-loops tethers MS2 fusion proteins and acts a
reporter for translation regulation. Luciferase-coding mRNAs without stem-loops are
used as specificity (firefly luciferase-coding) and loading (Renilla luciferase-coding)
controls. C. Schematic of the method of the tethered assay.
MS2 alone was a negative control for translation regulation as it bound to the
reporter RNA but should not have affected translation (fig 6A). Xp54, a known
translation repressor (Minshall et al., 2009), was a positive control for translation
repression. Zar was fused to MS2 to evaluate its effect on translation.
26


MS2 mRNA was in vitro transcribed from pXen C-MS2, MS2-Xp54 was
transcribed from pXen MS2-Xp54 and Zar-MS2 and MS2-Zar mRNAs were transcribed
from various pXen Zar-MS2 constructs (see APPENDIX), using mMessage mMachine
(Life Technologies). Oocytes were injected with mRNA for the fusion protein (1 ng 50
ng) and incubated at 18C for 16-24 hours to allow protein expression (fig 6C). The
mRNA coding for the fusion protein was injected rather than the protein itself because
mRNA can be rapidly and easily purified and is translated efficiently in Xenopus oocytes.
Oocytes were then injected with both firefly (100 pg) and Renilla (5 pg) coding
mRNA and incubated at 18-24C to allow for translation of luciferase. In some
experiments, half the oocytes were administered progesterone to induce maturation and
both immature and mature oocytes were collected at the same time, at least 3 hours after
progesterone treated oocytes reached GVBD. Each sample consisted of duplicates of 5
oocytes each and were lysed in 250 pL passive lysis buffer (Promega). 10 pL of cleared
lysate was used in the Dual Luciferase Reporter Assay (Promega) according to
manufacturers instructions and samples were analyzed using a Synergy HT plate reader
(BioTek).
mRNA stabilization assays
The relative amount of mRNA was determined by semi-quantitative PCR.
cDNA was synthesized from 1-5 oocyte equivalent lysate, but from pools of 5-20 oocytes
with iScript (Bio-Rad) using random hexamer primers and PCR performed with Taq
polymerase (Life Technologies). The cycle number in which the PCR was in a linear
range was determined by running 1/100 and 1/1000 oocyte equivalent of cDNA from
27


total RNA and making sure there was a difference in the amount of PCR product between
these two amounts of template cDNA.
mRNA polyadenylation assays
Following the indicated treatments, total RNA was extracted from pools of 5-10
oocytes using TRI reagent (Life Technologies) or STAT-60 (TelTest) using manufacturer
instructions followed by an additional phenol/chloroform extraction. To obtain high-
quality RNA, samples were re-precipitated with 8 M LiCl (Charlesworth et al., 2000).
RNA samples were re-suspended in water. RNA ligation-coupled RT-PCR is a modified
version of the technique described previously ((Rassa et al., 2000); (Charlesworth et al.,
2004)).
To assess the length of the poly(A) tail of reporter mRNAs extracted from
Xenopus oocytes: 1) a primer (PI) was ligated to the 3 end of the mRNA, 2) the reverse
complement of the ligated primer (PL) was used reverse transcribe cDNA and 3) PCR
was performed using gene specific primers and primer PL to amplify the 3 end of the
mRNA, including the poly(A) tail. As shorter PCR fragments are generated from
mRNAs with shorter poly(A) tails and longer PCR fragments are generated from mRNAs
with longer poly(A) tails, changes in the size of the PCR product represented changes in
poly(A) tail length.
4 pg of total oocyte RNA, from pools of 5 or 6 oocytes, was ligated to 0.4 pg of
PI anchor primer, in a 10-pl reaction using T4 RNA ligase (New England Biolabs)
according to the manufacturers instructions. The whole 10-pl RNA ligation reaction was
used in a 50-pl reverse transcription reaction using Superscript III (Life Technologies),
according to manufacturers instructions using 0.4 pg of Pl as the reverse primer. 1 pi of
28


this cDNA preparation was used in each 50-pl PCR using Platinum Taq (Life
Technologies), according to the manufacturers instructions. PCR was performed for 40
cycles, using a 56 C annealing temperature and 1.5 pM final concentration of Mg2+,
except where parameters needed to be changed (Charlesworth et al., 2004).
Western blot
Pools of 5-10 oocytes were collected from each sample and lysed in 10 pl/oocyte
of NP-40 lysis buffer (1% Igepal CA-630, 20 mM Tris pH 8.0, 137 mM NaCl, 10%
glycerol, 2 mM EDTA) supplemented with protease and phosphatase inhibitors (HALT,
Pierce). 5 or 10 pi of cleared lysate (0.5 or 1 oocyte equivalent respectively) was loaded
onto a NuPage Bis-Tris polyacrylamide gel (Life Technologies). Electrophoresis was
performed using MOPS-SDS running buffer (Life Technologies), then transferred to 0.45
pm Immobilon-FL PVDF membrane (Millipore) using an XCell II Blot Module (Life
Technologies) according to NuPage technical guide protocol.
Membranes were probed with antibodies against: MS2 (TetraCore), GST (Santa
Cruz Biotechnology), P-Tubulin (Sigma), eIF4G (Cell Signaling Technology), 4E-T (Cell
Signaling Technology), eIF4E (GeneTex), eIF4E-lb (Minshall et al., 2007) (kindly
provided by Nancy Standard, Univeristy of Cambridge, UK), ePAB (USBiological), Zarl
(Charlesworth et al., 2012) or Zar2 (Yamamoto et al., 2013). Secondary antibodies were
1:20,000 goat anti-rabbit IR Dye 800 CW and goat anti-mouse IR Dye 680LT (LiCor).
Membranes were imaged on an Odyssey infrared imager and data was analyzed using
Odyssey 2.1 software (LiCor).
29


GST affinity purification
In order to identify interacting proteins, Zar was fused to the tag GST, which
binds glutathione-sepharose beads, allowing for purification of GST-Zar fusion protein
along with co-purification of proteins that interact with GST-Zar from treated oocyte
lysates (fig. 7). Isolated proteins were separated by electrophoresis and identified by
mass spectrometry.
Inject oocytes with GST fusion
Lyse oocytes
Glutathione sepharose beads
I
Elute with glutathione
*
GST GST-Zar
Western
Figure 7. Schematic of GST affinity purification assay.
Oocytes were injected with mRNA encoding the GST fusion, incubated to allow for
protein expression and then lysed. The lysate was applied to agarose beads (white circle)
covalently attached to glutathione (white hexagons), which binds GST (grey rectangle).
After elution of the beads with reduced glutathione, the GST protein, along with proteins
bound to GST (grey trapezoid), were isolated from the lysate, then separated by
electrophoresis and visualized by Coomassie or western blot.
30


GST alone was used as a negative control and only bands that appeared more
strongly in the GST-Zar lane compared to GST alone were cut out for identification.
GST mRNA was in vitro transcribed using mMessage mMachine (Life Technologies)
from pXenl (MacNicol et al., 1997) and GST-Zar fusion protein coding mRNA was
transcribed by pXen GST-Zar plasmids described in APPENDIX. Oocytes were injected
with 3.7 ng of GST coding mRNA or 20 ng of GST-Zar coding mRNA and allowed to
incubate at 18C for 16-24 hours to allow expression of the fusion protein. Oocytes were
then collected and lysed in 5 uL per oocyte of either NEMO lysis buffer or PBS-triton and
lysates purified on glutathione sepharose beads followed by elution with 10 mM reduced
glutathione. Purified proteins were separated by electrophoresis and visualized by
Coomassie or by western blot as appropriate.
Immunoprecipitation
Pools of 100 oocytes were lysed in 500 pi PBS-Triton homogenization buffer
supplemented with HALT protease and phosphatase inhibitors, RNAse and 5 pM DTT.
Lysates were clarified by centrifugation 2x5 min, 12,000 x g, 4 C. 4 pg of Zar2
antibody was added and incubated for 6 h, 4 C. 20 pi of protein A DynaBeads (Life
Technologies) were washed 3 x with PBS triton buffer and added to samples then rotated
for 1 hour 4 C. Beads were again washed 3 x with PBS-Triton buffer and then extracted
with 50 uL IX LDS sample buffer. Purified proteins were separated by electrophoresis
and visualized by Coomassie or by western blot as appropriate.
31


CHAPTER IV
EXPERIMENTS AND RESULTS
Aim 1. Show a larger effect for Zar-mediated translation repression.
While both Zarl and Zar2 have been shown to repress translation of a reporter
mRNA using the MS2-tethered assay approach, this repression has not been robust.
Repression of up to about 30-40%, relative to when Zar is not tethered to the reporter
mRNA (MS2 alone), has been observed for the N-terminus of Zar2 and only about 30%
has been observed for the N-terminus of Zarl ((Charlesworth et al., 2012); (Yamamoto et
al., 2013)). This is inconsistent with the level of translation repression observed for TCS-
containing reporter mRNA, with 50% repression compared to control in immature
Xenopus oocytes (Wang et al., 2008). Zar-MS2 fusion proteins have not expressed
particularly well and have required 20-100 ng of fusion protein-coding mRNA to see
significant effects in tethered assay experiments and injection of more than 100 ng was
toxic to oocytes (not shown).
Poor expression has been particularly problematic for full-length Zarl-MS2,
which has not expressed well in Xenopus oocytes and relatively little fusion protein
accumulation was detected at the end of experiments. Even when the amount of MS2-
fusion RNA was injected at high levels, very little protein was produced (not shown).
While the N-Zarl-MS2 fusion protein was expressed to levels high enough to show effect
in the tethered assay, there were similar expression problems with it and much higher
levels of RNA had to be injected to get the same level of protein expression and
repression effect as that seen for N-Zar2-MS2: about lOOng of N-Zarl-MS2 was required
to see the same effect as about 20ng ofN-Zar2-MS2 (Yamamoto et al., 2013).
32


If MS2-Zar fusion proteins could be made to express better and/or be more stable
in oocytes, then perhaps they could repress translation of reporter mRNAs to a greater
extent. Translation repression of about 50% would be considered successful in this aim,
as this is the amount of translation repression observed for the TCS and other translation
repressors, such as Xp54. This evidence would provide support for the notion that Zar
proteins regulate translation and mediate the effects of the TCS. It would also be
technically advantageous if less MS2-Zar RNA was required to detect the same level of
translation repression by minimizing non-specific artifacts that may be introduced. In
order to get better repression, changes were made to the design of both the MS2-Zar
fusion proteins and the firefly luciferase reporters used in the MS2-tethered assay.
Do Zar fusion proteins repress better when stabilized with inter-domain linkers?
Rationale:
To improve the function of Zar-MS2 fusion proteins in the tethered assay, longer
inter-domain linkers were inserted between the MS2 and Zar domains with the goal of
improving protein stability and preventing unwanted interactions between the domains
(Minshall et al., 2010). The old Zar-MS2 fusion proteins consisted of the two domains, a
Zar protein or Zar protein truncation and an MS2 tag, fused adjacent to each other with a
very short linker region (fig. 8A). The Zar domain was in the N-terminal position of the
fusion and MS2 was C-terminal. The reasoning behind choosing to not use a linker was
to minimize any extra sequence that may result in undesired artifact and to take
advantage of available restriction sites. The Zar domain was fused N-terminal of MS2
because the N-terminal of Zar was shown to be the translational domain (Charlesworth et
al., 2012) and this arrangement allows the N-terminus of Zar to be unhindered, with hope
33


of promoting a more natural folding of the domain. However, observations from other
labs suggest thatat least in some casesthe N-terminal MS2-tagged fusion proteins
were more stable than the C-terminal tagged fusions. So it was decided to reverse the
orientation of the fusions as another measure to increase protein stability.
A Old constructs
B New constructs
Zar MS2 MS2 Zar
Full-length
Zar protein
A(GGGGS)3A
Flexible
Random coil
N-Zar MS2 MS2 H* Zar
Proposed translation AfFAAAin a Rigid
regulation domain ^ '3 a-helix
MS2
Zar
(AP)SA
Rigid
Proline-based
Figure 8. Cartoon of MS2 fusion proteins: old and new constructs.
Longer inter-domain linkers were added between MS2 and Zar. A. The MS2 fusion
proteins that were used to show moderate repression in previous publications included
very short linkers between MS2 RNA-binding domains and Zar protein domains. Full-
length Zar and N-Zar were arranged N-terminal of MS2 so that the N-terminus of Zar
was free, as would be in the endogenous protein. B. In the new MS2 fusion protein
constructs, a longer linker was inserted and all Zar protein domains, full-length or
truncated, were placed C-terminal of MS2. Three linkers were evaluated, one flexible
and two rigid, referred to as GS (upper), EK (middle) and AP (lower). The amino
acid sequence of the linker is labeled.
Linkers are small amino acid sequences inserted in between domains of a fusion
protein to provide physical space between the domains and have been shown to stabilize
and to improve the performance of some fusion proteins (Lee et al., 2013). The use of
inter-domain linkers to improve fusion protein function comes from observations that
many natural multi-domain proteins contain linkers of characteristic lengths and chemical
34


properties that are believed to contribute to effective functioning of flanking domains
(George and Heringa, 2002). Additionally, the use of synthetic linkers similar to those
found in endogenous proteins have been used to improve stability and function of fusion
proteins in a variety of experimental systems (Chen et al., 2013). While there is evidence
that particular linker sequences have improved the function of specific fusion proteins,
this does not mean that all linkers will improve Zar-MS2 fusions and the appropriateness
of a linker must ultimately be empirically determined.
The new MS2-Zar fusion proteins contained one of three inter-domain linkers: a
flexible glycine/serine rich linker (GS), a rigid alanine/proline linker (AP) or a rigid
gluatamate/lysine rich linker (EK) (fig. 8B). Each linker consisted of a repeating amino
acid sequence, 17 amino acids in length, so that each linker fusion was the same size as
its counterparts. The first linker chosen was A(GGGGS)3A and forms a flexible linker
with little secondary structure due to the small hydrophilic residues. The second linker
was A(EAAAK)3A and due to the hydrophilic interactions between the acidic glutamate
residues and the basic lysine residues, forms a rigid alpha helix that provides separation
of the domains. The final linker chosen was (AP)gA, also a rigid linker that provides
fixed separation of domains, but forms a rigid structure due to the molecular geometry of
the proline residue. The three linkers are hereafter referred to as GS, EK, and AP
respectively (example: MS2-AP-Zarl).
It should be noted that the orientation of the domains with respect to each other
was reversed. In the new constructs the MS2 domain is N-terminal of the Zar domain,
whereas in the old constructs, the tag was C-terminal of Zar. This orientation is similar to
other MS2 fusion controls used in the study, such as MS2-Xp54. Both the inclusion of
35


linker and inversion of orientation may contribute to a difference in function of the new
constructs compared to the old.
In addition to the issue of protein stability in immature oocytes, progesterone
dependent protein degradation has contributed to difficulties in probing the mechanism of
Zarl-mediated translation regulation during meiotic maturation using the tethered assay.
Endogenous Zarl levels are stable through maturation while Zar2 levels are reduced
dramatically in progesterone-matured oocytes compared to immature oocytes.
Surprisingly, fusion protein levels for both old Zarl-MS2 and old Zar2-MS2 were
reduced in response to progesterone though, meaning the relief of repression observed for
Zarl-MS2 from immature to mature oocytes in the tethered assay could be due solely to
the degradation of the fusion protein and not represent a change in function of
endogenous Zarl in the mature Xenopus oocyte. In fact, this is why the observation that
Zarl changes function during maturation has not yet been published.
Results:
MS2-tethered assays were performed with the old constructs (fig. 9A)and the new
constructs side-by-side to see if any or all of the new constructs repressed translation to a
greater extent (fig 9B). MS2 alone did not repress translation of reporter mRNA
compared to no fusion protein injected. MS2-Xp54, a positive control for translation
repression, repressed translation by 60% compared to no fusion protein or to MS2 alone.
Zarl-MS2 and NZarl-MS2, the old fusion protein constructs, repressed translation by
about 20-30% when 50ng of fusion mRNA was injected. The new MS2-Zarl fusion
proteins, with longer inter-domain linkers, repressed translation by about 50% when the
same amount of fusion mRNA was injected. The old Zar2-MS2 and N-Zar2-MS2 fusion
36


proteins repressed translation by about 30%, whereas the newly designed MS2-Zar2
constructs repressed translation by about 40-50%, similar to Xp54. Therefore, new MS2-
Zar fusion proteins (with longer linkers) repressed translation to a greater extent than the
old Zar-MS2 fusion proteins with a minimal linker region and all new constructs
repressed to similar levels, regardless of which linker was used.
A
N-Zar-MS2
N-Zar MS2
Zar-MS2
Zar
MS2
MS2-AP-Zar
MS2 Zar
MS2-EK-Zar
MS2 H Zar
MS2-GS-Zar
MS2 Zar


<\
>o /V
^ .< .< O' J__'
&


* 4?4?4? *
/v ZV- ZV
Figure 9. Translation repression by old versus new MS2-Zar fusion proteins.
Increased translation repression was observed from MS2-Zar fusion proteins redesigned
with longer inter-domain linkers. A. Cartoon of MS2-Zar fusion proteins used in tethered
assay. The newly constructed fusion proteins (lower three) contain longer inter-domain
linkers and an N-terminal orientation with respect to the MS2 tag. B. Bar chart of
relative luciferase activity of reporter mRNAs when tethered to MS2 fusion proteins
showing that all new MS2-Zar fusion proteins (with longer inter-domain linkers)
repressed translation to a greater extent than old fusion proteins, regardless of which
linker (AP, EK or GS) was inserted. Results were normalized to MS2 alone; MS2-Xp54
was a positive control for translation repression (n = 3-5).
Next, a dose response was performed to determine the optimal conditions for the
new fusion proteins (fig 10). The AP and EK linkers were chosen for closer analysis, as it
was thought that the rigid linkers are most likely to encourage functional separation of the
MS2 and Zar domains. MS2-GS-Zar was not used in subsequent experiments (although
there was no apparent difference in function among the various linkers).
37


1.2
MS2 MS2-AP-Zarl MS2-EK-Zarl MS2-AP-Zar2 MS2-EK-Zar2
Figure 10. Dose response of translation repression by MS2-Zar.
Bar chart showing MS2-AP-Zar and MS2-EK-Zar repressed translation in a dose-
responsive manner and to similar levels relative to the amount of fusion protein-coding
mRNA injected (1 ng 50 ng), regardless of whether the linker was AP (alanine/proline)
or EK (glutamate/lysine) ,or whether the Zar domain was Zarl or Zar2 (n = 1-5, no error
bar indicates n = 1).
When 1-3 ng of fusion mRNA was injected, little or no repression was observed
for any of the fusion proteins tested. For all four fusion proteins, significant repression
was observed when 10 ng, 30 ng or 50 ng was injected. About 20-30% repression was
observed when 10 ng was injected, about 30-40% repression was observed when 30 ng
was injected and about 40-50% repression was observed when 50 ng was injected for all
four fusion proteins examined. MS2-AP-Zar and MS2-EK-Zar repressed translation to
similar levels with respect to each other at all doses examined, and so choice of linker did
not affect the level of repression. Most of the experiments in this study were done with
MS2-AP-Zar injected at 30 ng of coding mRNA, with 30 ng chosen to get maximal
repression while minimizing non-specific effects from overexpression.
38


^ 1.2
Figure 11. Translation regulation by Zar during maturation.
Bar chart of relative luciferase activity (relarive activity) when mRNA reporters were
tethered to MS2-Zar fusion proteins in immature (I, black) and progesterone-matured (P,
white) oocytes, showing that MS2-AP-Zar repressed to a greater extent in immature
oocytes than Zar-MS2, but progesterone-dependent relief of repression was still
observed, (n = 1 3, no error bar indicates n = 1).
To test whether Zar-mediated repression of new constructs changed during oocyte
maturation, fusion proteins were expressed in oocytes, then reporters were injected and
half the oocytes given progesterone (fig. 11). When Zarl-MS2 was tethered, about 20%
translation repression was observed in immature oocytes and about 10% repression
observed in mature oocytes. When MS2-Zar2 was tethered, about 30% repression was
observed in immature and 10% in mature oocytes. When MS2-AP-Zarl and MS2-AP-
39


Zar2 were tethered, about 40-50% translation repression was observed in immature
oocytes and about 30% repression observed in mature.
Although the absolute value of translational activity in mature oocytes was
different for Zar2 fusion proteins, the amount of translational activity relative to activity
in immature oocytes was similar, about a 25-30% increase. These results show that when
MS2-AP-Zar2 was tethered in progesterone-matured oocytes, repression was relieved
relative to translation repression in immature oocytes.
u
to
LL
MS2
western
I P
I P I P I P I P
_______________11________________I I________________11_______________
A
& jy
s /V ^
* #


*

Figure 12. Amount of MS2-Zar protein remaining after maturation.
Lower, representative MS2 western blot from immature (I) and progesterone-matured (P)
oocytes, showing new MS2-AP-Zarl fusion protein expressed to higher levels in
immature oocytes than old Zarl-MS2 and more fusion protein remained after maturation.
Upper, bar chart showing quantification of the fraction of Zar fusion protein that
remained after progesterone treatment compared to immature samples across multiple
experiments (n = 3 5). (MS2-AP-Zar2 western not representative of all experiments.)
To determine why the new proteins were better at repressing translation, the
amount of fusion protein expressed was measured (fig. 12, lower). New MS2-AP-Zarl
40


fusion proteins expressed to greater levels than old Zarl-MS2 in immature oocytes and
this effect was observed over several experiments.
MS2-AP-Zar2 was expressed to only modestly higher levels than Zar2-MS2 in
immature oocytes. Both MS2-AP-Zarl and MS2-AP-Zar2 fusions produce fewer
degradation products than the old Zar2-MS2 constructs, suggesting the new fusion
proteins are more stable in immature oocytes.
To determine whether new MS2-Zar fusion proteins were more stable during
maturation, the amount of protein present in mature oocytes was compared to the amount
in immature oocytes (fig 12, upper). MS2-AP-Zarl protein levels in mature oocytes were
about 90% of the levels observed in immature oocytes over 3-5 experiments, whereas
Zarl-MS2 proteins levels in mature oocytes were only about 30% relative to levels in
immature oocytes. Proteins levels for both MS2-AP-Zar2 and Zar2-MS2 in mature
oocytes were about 60% of levels in immature oocytes. The MS2-AP-Zar protein was
more stable during maturation than the old Zarl-MS2 and this effect was observed over
several experiments. While a stabilization effect was observed for MS2-AP-Zar2 in
some experiments, this trend was not consistent across multiple experiments.
Do MS2-Zar fusion proteins repress better when tethered to reporter mRNAs with 4
stem-loops rather than 2 stem-loops?
Rationale:
The MS2-Zar fusion protein is tethered to the reporter RNA by high affinity
interactions between the MS2 domain of the fusion protein and an MS2-binding stem-
loop sequence inserted into the 3UTR of the luciferase reporter RNA (Coller and
Wickens, 2007). There are two tandem copies of the stem-loop sequence in the firefly
41


reporters (fluc-2SL-A33Ci7) that have been used to show repression by N-Zar-MS2
fusions previously. The decision to insert two stem-loops rather than one was made
because RNA containing two MS2-binding stem-loops binds better to MS2 protein-
coated beads than RNA containing only one stem-loop (Bardwell and Wickens, 1990),
possibly due to a more stable interaction of the MS2 dimer (Keryer-Bibens et al., 2008).
The use of multiple MS2-binding stem-loops in an MS2-tethered assay has been
evaluated by other labs, with 3 stem-loops increasing the effect seen for the MS2 fusion
protein evaluated compared to the presence of only one stem-loop. The inclusion of 9
stem-loops, however, did not further enhance the effect of the fusion proteins used in the
assay compared to 3 stem-loops (Minshall et al., 2010). As 2 stem-loops versus 3 stem-
loops has not been tested, we were not able to reconcile whether our results were
equivalent to their 3 stem-loop results. Therefore, firefly luciferase reporters were made
with either two stem-loops (fluc-2SL-A30) or four stem-loops (fluc-4SL-A30) (fig 13 A).
Results:
To test whether an increase in the number of MS2-binding stem-loops increased
MS2-Zar fusion protein-mediated translation repression, the levels of translation
repression by MS2-Zarl and MS2-Zar2 on fluc-2SL-A3o and fluc-4SL-A30 were measured
and compared to each other (fig. 13B). In immature oocytes, MS2-Zarl repressed
translation of fluc-2SL-A3o (black bars) by about 30-40%, whereas translation of fluc-
4SL-A30 (dark grey bars) was repressed by about 20-30%. In immature oocytes MS2-
Zar2 repressed translation of fluc-2 SL-A30 by about 50% and of fluc-4 SL-A30 reporter by
about 40%. Therefore, there was no significant increase in the amount of translation
repression observed when more stem-loops were present.
42


A
m7GTP
firefly
fluc-2SL-A30
| Immature (I)
30
Mature (P)
m7GTP-----
fluc-45L-A30
firefly
30
Immature (I) Q Mature (P)
B
1.2
I P I P I P
l______________I i_____________i l______________i
No fusion MS2-Zarl MS2-Zar2
Figure 13. Effect of extra stem-loops on Zar-mediated translational regulation.
A. Cartoon of the 2 stem-loop and 4 stem-loop containing reporter mRNAs used in the
tethered assay. B. Bar chart of relative luciferase activity when reporters with two MS2-
binding stem-loops (black, immature; white, mature) or 4 MS2-binding stem-loops (dark
grey, immature; light grey, mature) were tethered to MS2-Zarl or MS2-Zar2, showing
that there was no significant increase in the amount of translation repression observed
when extra stem-loops were present, nor was progesterone-dependent change in function
affected by extra stem-loops.
43


To determine whether extra stem-loops affected translation regulation by MS2-
Zar in mature oocytes, oocytes expressing MS2-Zar were injected with 2 stem-loop and 4
stem-loops reporters and half were given progesterone.
In mature oocytes, MS2-Zarl repressed translation of fluc-2 SL-A30 (white bars)
by about 20% but only repressed translation of fluc-4SL-A3o (light grey bars) by about
10%, and MS2-Zar2 repressed translation of fluc-2SL-A30 by about 30% and of fluc-4SL-
A30 by about 20%. These differences were not considered to be significantly different,
therefore the inclusion of extra stem-loops did not affect progesterone-dependent relief of
Zar-mediated repression. Furthermore, no significant differences in translational
regulation activity were observed when lower doses of the MS2-Zar fusion protein were
administered (not shown) or when the extra stem-loops were inserted into other firefly
luciferase reporters (not shown).
Aim 2: Show the mechanism of Zar-mediated translational regulation is consistent
with that established for the TCS
If Zar proteins are functional partners of the TCS, then the mechanism of Zar-
mediated translation regulation and that of TCS-mediated translation regulation should
have certain fundamental similarities. If the TCS sequence imparts some characteristic of
translation regulation or polyadenylation of the mRNA, then the Zar protein should
impart the same characteristic. This is because my hypothesis proposes that Zar proteins
are the trans-factors that bind to the TCS and transmit the translation regulation
information from the c/.s-element to the translation machinery.
Mechanistic features of translation regulation that are commonly investigated are
requirement for a 5-cap and requirement for a 3-poly(A) tail, because they play such
44


prominent roles in most known mechanisms. There is already experimental information
available about the role of the poly(A) tail in TCS-mediated translation regulation, so
determining the role of the poly(A) tail in Zar-mediated regulation provided a way to test
whether the two mechanisms are mutually consistent. The TCS does not require a
poly(A) tail to repress translation, the TCS imparts cytoplasmic polyadenylation and
translational activation in mature oocytes and translational activation is dependent on the
polyadenylation of the mRNA. Both the role of the poly(A) tail in translation repression
and the process of polyadenylation during maturation were evaluated.
m7GTP
firefly
fluc-2SL-A33C17
m7GTP----
fluc-2SL-A0
firefly
A33C
17
Ao
m7GTP-----
fluc-2SL-A30
firefly
m7GTP
firefly
fluc-2SL-A100+
A
30
A10o+
Figure 14. Cartoon of luciferase reporter mRNAs of various poly(A) tail lengths.
Firefly luciferase-coding mRNAs with poly(A) tails of varying lengths were used to
assess the requirement of a poly(A) tail for Zar-mediated translation repression. All
reporters contain an m7GTP cap, a firefly coding region, and 2 MS2-binding stem-loops.
The fluc-2SL-A33Ci7 reporter contains 33 adenylate residues followed by 17 cytidines at
the 3 end. The fluc-2SL-Ao construct does not contain any adenylate residues and the
fluc-2SL-A3o and the fluc-2SL-Ai00+ reporters end in 3 poly(A) tails of 30 and greater
than 100 adenylate residues, respectively. The fluc-2SL-Aioo+ reporter mRNA was made
by adding a poly(A) tail to fluc-2 SL-Ao by using poly(A) polymerase in vitro; so while
the first 3 mRNAs are homogeneous in length, the last tailed reporter is actually a
heterogeneous population of different length reporters.
45


Several new luciferase reporters with various poly(A) tail structures were
constructed to address these issues (fig. 14). The luciferase reporter that has been used to
show repression by Zar proteins in the past (fluc-2SL-A33Ci7) has a poly(A) tail of about
30 adenosines flanked by 17 cytidines at the 3 end, but results could not be compared to
what we know about the TCS so new reporters were made.
The first new reporter (fluc-2SL-Ao) ends with no (A)s after the polyadenylation
hexanucleotide and is most similar to the reporters that were used to evaluate TCS-
mediated translation repression in the past. This was compared to a reporter with 30
terminal (A)s (fluc-2SL-A3o) and to one with greater than 100 (A)s (fluc-2SL-Aioo+).
These lengths were chosen because it is common for mRNAs that are being
translationally repressed to have about 30 (A)s and for mRNAs that are translationally
active to have greater than 100 (A)s. The fluc-2SL-Aioo+ reporter was unique among the
luciferase reporters as it was generated using a poly(A) tailing kit to in vitro polymerize a
poly(A) tail onto fluc-2SL-Ao, resulting in a heterogeneous population of mRNA
reporters of various lengths, with the population centering on a mean length of greater
than 100 adenosines.
Do Zarl and Zar2 require a poly(A) tail to repress translation?
Rationale:
The TCS imparts translational repression in immature Xenopus oocytes when
placed in a (3-globin 3UTR without a poly(A) tail (Wang et al., 2008). If Zar proteins
repress translation in immature oocytes through regulation of the TCS, then they too
should be able to repress translation without a poly(A) tail.
46


Deadenylation of mRNA leads to translational repression; recruitment of
deadenylases is a common mechanism of repression by RNA-binding translation
repressors. While the TCS represses translation of a reporter without a poly(A) tail and
therefore does not require deadenylase activity to repress translation, this does not
preclude the possibility that recruitment of a deadenylase is part of the mechanism. Some
deadenylases have translational repression activity independent of their deadenylase
activity (Cooke et al., 2010) and shortening of tail length may play a role when mRNAs
have a tail, although this has not been tested for the TCS. Therefore, the role of
deadenylation in Zar-mediated repression was examined as well.
Results:
To test whether Zar proteins require a poly(A) tail to repress translation, oocytes
expressing MS2-Zarl or MS2-Zar2 were injected with luciferase reporters with or
without a poly(A) tail, fluc-2SL-Ao or fluc-2SL-A3o respectively. Immature oocytes were
collected and relative luciferase activity compared.
When tethered to fluc-2SL-A3o, MS2-Zarl (MZ1) repressed translation by 40%
and MS2-Zar2 (MZ2) repressed translation by about 50% when 30 ng of mRNA was
injected (fig 15A). When MS2-Zar2 was tethered to fluc-2SL-Ao, about 40% translation
repression was observed, but when MS2-Zarl was tethered there was no repression
observed, even though MS2-Zar fusion proteins were expressed to similar levels (fig
15A, inset). So while Zar2 did not require a poly(A) tail to repress translation,
surprisingly, repression by Zarl did require a poly(A) tail.
47


B
Poly(A)
assay
I P I I
__________________I
I P I I
I P I I
I______________________I

.<> /V
G
vv
*

.jy
G
& &
Figure 15. Zar-mediated repression on reporters varying poly(A) tail length.
Translation repression by Zarl required a poly (A) tail, whereas repression by Zar2 did
not, and an extended poly(A) tail prevented repression by both Zarl and Zar2. A. Bar
chart showing translation repression observed in immature oocytes when MS2-Zarl
(MZ1), MS2-Zar2 (MZ2) or no fusion protein (control) was tethered to firefly luciferase
reporters with no poly(A) tail (fluc-Ao), a tail of 30 adenylate residues (fluc-A3o) or a tail
of more than 100 residues (fluc-Aioo+). Inset, MS2 western shows fusion proteins were
expressed to similar levels. B. Translation repression by MS2-Zar was not accompanied
by deadenylation of the poly(A) tail. Poly(A) assay (RNA-ligation coupled RT-PCR)
showing that reporter mRNA was not shortened when MS2-Zar was tethered in immature
oocytes. Poly(A) tail length was measured in immature (I) and progesterone-matured (P)
control oocytes to demonstrate tail shortening, as flue reporters with poly(A) tails were
deadenylated during maturation: fluc-Ao reporter was not shortened, while both fluc-A3o
and fluc-Aioo+ were.
48


To determine whether an extended poly(A) tail could prevent Zar-mediated
repression, MS2-Zar expressing oocytes were injected with fluc-2SL-Aioo+ and relative
luciferase activity measured. The relative luciferase activity of MS2-Zarl treated oocytes
was equal to that of control oocytes. Although the luciferase activity measured in MS2-
Zar2 treated oocytes appeared to be about 20% greater than that for control oocytes, the
variation was great, almost 30%, and so this was not considered a significant difference.
So when the poly(A) tail was extended to greater than 100 adenylate residues, there was
no translation repression observed for either MS2-Zarl or MS2-Zar2. No destabilization
of reporter mRNAs was observed (not shown).
One possible explanation for the poly(A) dependency was that Zarl recruited
deadenylases that shortened the length of the poly(A) tail, thereby inhibiting full
formation of the translation initiation complex. To determine whether mRNAs
translationally repressed by Zar were deadenylated, total mRNA was extracted after
tethered assay experiments and the lengths of mRNA reporter poly(A) tails were
measured by RNA ligation-coupled RT-PCR (fig 15B). The luciferase reporters, due to
the (3-globin UTR background, underwent default deadenylation in response to
progesterone in untreated oocytes. Thus, comparison of tail length in mature versus
immature control oocytes provided a positive control for deadenylation of the reporter.
The poly(A) tail of fluc-2 SL-Ao was not shortened during meiotic maturation, which was
expected as there was no poly(A) tail to deadenylate, and treatment with MS2-Zarl or
MS2-Zar2 had no effect on tail length. The fluc-2SL-A30 reporter mRNA was
deadenylated during maturation in control oocytes, and so deadenylation of this reporter
was detected with this assay. Treatment with MS2-Zarl or MS2-Zar2 had no effect on
49


poly(A) tail length and the size of the PCR product was no smaller than in control
oocytes. The fluc-2SL-Aioo+ reporter also underwent progesterone-dependent
deadenylation in control oocytes from a larger heterogeneous product (black arrow) to a
more homogeneous product of smaller size. Again, treatment with MS2-Zarl or MS2-
Zar2 did not result in deadenylation of the reporter mRNA in immature oocytes.
Therefore, translation repression by Zar did not result in deadenylation of any of the
tethered reporter mRNAs.
Do Zarl and Zar2 require the process of polyadenylation to activate translation
during maturation?
Rationale:
The TCS sequence has been shown to mediate extension of the poly(A) tail
during meiotic maturation. The current consensus sequence for the TCS is
(A/U)UU(A/G)UCU and when either AUUGUCU or UUUGUCU is inserted into a |3-
globin 3 UTR, the reporter RNA is polyadenylated in response to progesterone. When a
17-nucleotide region of the Weel mRNA containing both TCS sequences was inserted in
to a GST P-globin reporter RNA with a functional polyadenylation hexanucleotide, not
only was GST protein accumulation reduced in immature oocytes, but protein
accumulation was increased about 2-fold in mature oocytes relative to those expressing
GST P-globin reporters lacking a TCS (Wang et al., 2008). This translational activation
was accompanied by polyadenylation of the reporter RNA. Polyadenylation appears to
be required for translational activation as when the polyadenylation hexanucleotide was
disrupted in the weel 3UTR, translational activation was inhibited (Charlesworth et al.,
2000). The experiments demonstrate that the TCS imparts polyadenylation of reporters
50


in mature oocytes and suggest that a poly(A) polymerase is involved the mechanism of
TCS-mediated translation regulation.
Relief of Zar-mediated repression has been observed on the fluc-2SL-A33Ci7
reporter. However, MS2-Zar dependent luciferase accumulation in mature oocytes was
similar to that seen when Zar was not tethered, whereas the TCS showed about a two-fold
increase compared to its appropriate control. It was thought that the mRNA reporter
might not be competent for polyadenylation and this was inhibiting the ability of Zar to
activate translation. Poly(A) tails have been shown to be deadenylated when 10 cytidines
are 3 of the poly(A) tail but not when 30 cytidines are (Vamum and Wormington, 1990).
The 3UTR of the IIUC-2SL-A33C17 reporter mRNA ends in 33 adenylate residues
followed by 17 cytosine residues and a few other non-adenylate residues after that, about
20 nucleotides total and intermediate of the two tested values, so it was not known
whether the reporter could be deadenylated or polyadenylated. If the fluc-2SL-A3o
reporter changed length during maturation while the IIUC-2SL-A33C17 reporter did not,
then comparison of translation of the two reporters in mature oocytes would provide a
way to evaluate the effect of the process of polyadenylation on Zar-mediated translational
activation.
Results:
To test whether the IIUC-2SL-A33C17 and fluc-2SL-A3o reporters changed length
when tethered to MS2-Zar, RNA was extracted from immature and mature oocytes
injected with luciferase reporters and tail length measure by RNA ligation-coupled RT-
PCR (fig 14, lower). The size of the tail of IIUC-2SL-A33C17 was the same in mature
oocytes as it was in immature, whereas the tail of fluc-2 SL-A30 was slightly smaller in
51


mature oocytes due to deadenylation. Therefore, the 3 poly(C) tract of fluc-2SL-A33Ci7
protected the reporter RNA from the default deadenylation observed for reporters with a
terminal poly(A) tail, suggesting that the reporter is not competent for changes in poly(A)
tail length and showing that the fluc-2 SL-A30 can change length.
To determine whether Zar activates translation to a greater extent when the
reporter could be polyadenylated, oocytes expressing MS2-Zarl or MS2-Zar2 were
injected with either fluc-2SL-A33Ci7 or fluc-2SL-A3o reporter mRNA and half were given
progesterone (fig 16, upper). Relative luciferase activity was evaluated for a correlation
between translation activation and polyadenylation. MS2-Zarl and MS2-Zar2 repressed
translation of fluc-2SL-A33Ci7 in immature oocytes by about 30-40% relative to control
oocytes. In comparison, translation of the fluc-2SL-A30 reporter was repressed by about
30-40% by MS2-Zarl and by about 50% by MS2-Zar2. Repression was relieved during
maturation; MS2-Zarl repressed translation by about 20% and MS2-Zar2 repressed by
about 30% in mature oocytes relative to control oocytes. There was not a higher level of
translation in mature oocytes when MS2-Zar fusions were tethered to fluc-2 SL-A30 than
when tethered to IIUC-2SL-A33C17. Therefore, MS2-Zar repressed to similar levels in
immature oocytes and the progesterone-dependent relief of repression during maturation
was similar whether the reporter had a terminal 3poly(A) tail or one flanked with non-
adenylate residues.
MS2-Zarl and MS2-Zar2 fusion proteins were expressed to similar levels and the
fusion proteins were relatively stable throughout maturation, particularly for MS2-Zarl.
Zar 1-mediated translation was relieved during maturation even when the same amount of
52


MS2-Zarl protein was present, suggesting that Zarl protein changes function during
maturation.
poly(A)
assay
J_____P_^ J.____J.________P
control MZ1 MZ2
i 1
I p 1 p 1 p
control MZ1 MZ2
fluc-A33Ci7
fluc-A30
Figure 16. The effect of polyadenylation on Zar-mediated translational regulation.
Upper, bar chart of relative luciferase activity when MS2-Zarl (MZ1) or MS2-Zar2
(MZ2) was tethered to luciferase-coding mRNA reporters with either a terminal poly(A)
tail (fluc-A3o) or a poly(A) tail with a poly(C) region 3 of the poly(A) tail (fluc-A33Ci7).
The A33C17 modification was expected to prevent changes to the length of the poly(A)
tail, while the fluc-A3o reporter was expected to be competent for deadenylation or
polyadenylation. Tethering of MS2-Zar to both reporter mRNAs resulted in similar
translation regulation: translation was repressed to the same extent in immature oocytes
(I, black) and repression was relieved during progesterone-induced maturation (P, white)
with no greater translation activation observed for fluc-A3o compared to IIUC-A33C17.
Middle, MS2 western showing that MS2-Zarl and MZ2-Zar2 were expressed to similar
levels and MS2-Zar fusion protein was stable during maturation. Lower, poly(A) assay
(RNA ligation-coupled RT-PCR) of reporter mRNA showing that the IIUC-A33C17 mRNA
tail length was the same in immature (I) or progesterone-matured (P) oocytes, whereas
fluc-A3o reporters were shorter in matured oocytes compared to reporters in immature
oocytes. Treatment of MS2-Zar did not result in deadenylation of the reporter mRNA in
immature oocytes compared to control, nor did it protect the mRNA from progesterone-
induced deadenylation.
53


To test whether tethering of Zar results in polyadenylation during maturation, the
tail lengths of reporter mRNAs were measured from immature and mature MS2-Zar
expressing oocytes injected with either fluc-2SL-A33Ci7 or fluc-2SL-A3o. The tail length
of the fluc-2SL-A33Ci7 was the same regardless of MS2-Zar or progesterone treatment.
The fluc-2 SL-A30 reporter however, underwent deadenylation in control oocytes in
response to progesterone. The reporter tail length in MS2-Zarl and MS2-Zar2 treated
mature oocytes was the same as in controls. Therefore, tethering of MS2-Zarl or MS2-
Zar2 to the reporter mRNA did not result in polyadenylation of the mRNA during
maturation even when the reporter was competent for changes in tail length, nor did it
protect the mRNA from progesterone-dependent deadenylation.
Aim 3: Show that Zar proteins interact with other translation factors in the oocyte
What proteins do Zarl and Zar2 interact with in Xenopus oocytes?
Rationale:
Translation factors do not generally operate in isolation within the cell. They
interact with other proteins, such as deadenylases and poly(A) polymerases, as previously
mentioned, and with components of the translation initiation complex and ribosome. As
a result, translation factors tend to interact with other translation factors and it would be
predicted that if Zar proteins function as translation factors in the oocyte, then they too
would interact with other translation factors.
Results:
To isolate the proteins that interact with Zar in the Xenopus oocyte, a GST affinity
purification was performed using lysate from immature oocytes injected with GST-Zarl.
54


Immature oocyte lysate was chosen, as it is at this point in development that Zar proteins
repress translation, and the objective was to identify proteins interacting with Zar while it
is functioning. Proteins were purified on glutathione sepharose and extracted with LDS,
then visualized by Coomassie (fig. 17).
kDa
<250
Symplekin, DHX9 Q
eIF3c, Tudor7, ILF3B Q
XPatla Q
<150
<100
< 75
ePABP, Vera
CPEB1, Rap55 C
Xp54, Frgy2, Rap42, T
EF1A3, EF1GB L
GST
RS20, RL36, r
RL22, RL31 L
<3
< 50
< 37
< 25
< 20
< 15
GST-
Zarl
GST GST-
Zarl
Figure 17. GST-Zarl protein interactions in immature oocytes.
Several bands were present only in immature oocytes and many translation factors were
identified in those bands. Lysates from immature oocytes that were injected with 3.7 ng
of GST or 20 ng of GST-Zarl coding mRNA were purified on glutathione sepharose
beads, extracted from the beads with LDS, separated by electrophoresis and visualized
using a Coomassie stain. Molecular marker sizes in kDa are indicated to the right.
Exogenous GST and GST-Zarl are indicated by white arrow and pieces of the gel that
were excised for sequencing are indicated by grey boxes. Proteins that were identified by
mass spectrometry and have previously been implicated in translation regulation are
indicated to the left.
55


Exogenous GST and GST-Zarl proteins were expressed to similar levels (fig. 17,
white arrows). While faint bands were present in the GST alone lane, several darker
bands were detected in the GST-Zarl lane that were not present in the negative control
lanes, GST alone or uninjected (not shown). This suggested that the proteins in these
bands interacted with Zar.
To identify the proteins that interacted with Zar, six of the bands or multiplets that
were unique to the GST-Zarl lane were excised from the gel and delivered to the mass
spectrometry core facility for analysis (fig. 17, grey boxes). A list of protein matches was
generated using peptide sequences retrieved from mass spectrometry analysis that were
BLASTed against a Xenopus protein database.
Matching sequences were scored based on percent coverage of the matching
sequence by the retrieved peptides and the number of retrieved peptides BLASTing to
that sequence. About 45 sequences were identified that met the requirements of the core
facility to be considered present in the sample. Table 1 lists the identified proteins as
they are named in the SWISS-Prot database along with predicted molecular weight,
grouped by the band in which they were retrieved.
Many of the identified proteins are previously characterized and implicated in
translational regulation (fig. 17, brackets), providing evidence that Zar proteins function
as translation factors in Xenopus oocytes. Symplekin, a scaffold protein, and DHX9, an
RNA helicase, play roles in translation initiation and were recovered in the 140 kDa
band. A subunit of the translation initiation factor 3 (eIF3c), a component of cytoplasmic
RNA granules (Tudor7) and a component of the translation inhibitor interleukin
enhancer-binding factor 3 (ILF3b) were found in the 120 kDa band, the translation
56


regulator Xpatla was recovered in the 90 kDa band and in the 70 kDa band, the oocyte
version of poly(A)-binding protein (ePAB) and an mRNA localization factor (Vera) were
identified.
In the 60 kDa band, the well-characterized translation regulator CPEB and
cytoplasmic granule component Rap55. In the 50 kDa band, several translation factors
were identified: the RNA helicase Xp54, Frgy2, Rap 55 and two components of the
translation elongation factor 1 (EF1 A3 and EF1GB). In the 20 kDa band, several
ribosomal proteins were identified: RS20, RL36, RL22 and RL31. These results suggest
that Zar proteins interact with translation factors in the cell.
There were no deadenylases or poly(A) polymerases identified, consistent with
the tethered assay results. There were several other proteins that were not identified as
translation factors. Some have roles in mRNA metabolism not directly pertaining to
translation regulation, such as splicing factor 3B (SF3B) found in the 140 kDa band.
Others have not been directly implicated in mRNA metabolism, such as the glycolytic
enzymes a-enolase and pyruvate kinase. These proteins may represent non-specific
interactions or may actually be involved in Zar function, but pursuing the meaning of
these interactions is not in line with the aims of this study.
The preponderance of proteins with known roles in translation regulation provides
support for the notion that Zar proteins function as translation regulators in the Xenopus
oocyte. Identification of these protein interactions also provided a list of candidates for
proteins that may be involved directly in the mechanism of Zar-mediated translational
regulation.
57


Table 1. Mass spectrometry results from GST-Zarl affinity purification.
Candidate interacting proteins are arranged by size of the band excised from the gel (band
size). Molecular weight refers to the predicted size of the protein based on amino acid
sequence. Protein name refers to the official abbreviated name of the protein in the
SwissProt database. Order of proteins within each band size is based on percent coverage
of the protein sequence and number of peptides matching to that sequence.
Band Size Protein Name Alternative Names Molecular weight (kDa) %_ Coverage tt of Peptide
140 DHX9 DEAH box protein 9 140.4 14 19
Symplekin 134.4 12 15
SF3B Splicing factor 3B 155 kDa subunit 146.1 3 4
XCAP-E Chromosome-associated protein E 136.3 3 3
Polyubiquitin Polyubiquitin (fragment) 18.8 11 2
DNA ligase 1 Polydeoxyribonucleotide synthase [ATP] 1 120.2 2 2
120 DDB1 Damage-specific DNA-binding protein 1 126.8 14 18
ILF3b Interleukin enhancer-binding factor 3-B 98.6 14 13
USPlOa Ubiquitin-specific-processing protease 10-A 87.5 10 7
elF3C Eukaryotic translation initiation factor 3 subunit C (subunit 8) 106.9 6 5
TDRD7 Tudor domain-containing protein 7 120.9 3 3
Polyubiquitin Polyubiquitin (fragment) 18.8 11 2
90 kDa xPatla Protein PAT1 homolog a 83.3 7 5
MVP Major vault protein 95.6 7 5
p97 ATPase Transitional endoplasmic reticulum ATPase 89.2 5 3
70 kDa ePABP-B Embryonic poly(A)-binding protein B 70.5 36 29
ePABP-A Embryonic poly(A)-binding protein A 70.7 35 28
Vera-A Vgl localization element binding and ER association protein 65.6 30 19
BiP Binding immunoglobulin protein 72.6 18 10
60 kDa PKM Pyruvate kinase muscle isozyme 57.5 29 14
RAP55Aa RNA-associated protein 55A-A 51.1 17 7
RAP55Ab RNA-associated protein 55A-B 51.2 20 9
CPEB1-B Cytoplasmic polyadenylation element-binding protein 1-B 62.6 18 9
TCP-l-y T-complex protein 1 subunit y 60.6 13 7
FRGY2a Frog Y-box binding protein 2-A 37.2 13 3
ePABPb Embryonic poly(A)-binding protein B 70.5 4 2
LOH12CR1 Loss of heterozygosity 12 chromosomal region 1 protein 22 5 2
50 kDa Xp54 ATP-dependent RNA helicase pS4 54 38 17
FRGY2a Frog Y-box binding protein 2-A 37.2 37 13
FRGY2b Frog Y-box binding protein 2-B 36.2 40 13
a-enolase 2-phosphoglycerate dehydratase 47.5 30 11
RAP42 RNA-associated protein 42 41.8 35 12
Tubulin p-4 49.8 24 10
EFl-a-Ol Elongation factor 1-a, oocyte form 50.2 16 7
EFl-y-B Elongation factor 1 y-B 50.2 11 5
Tubulin a 49.8 8 4
EP45 Estrogen-regulated protein 45 49.5 5 2
15 kDa XSODB Superoxide dismutase [Cu-Zn] B 15.4 31 4
S22 40S ribosomal protein S20 13.3 29 4
rpl36 60S ribosomal protein L36 12.3 31 4
rpl22 60S ribosomal protein L22 14.8 28 3
rpl31 60S ribosomal protein L31 14.5 15 2
NDK A1 Nucleoside diphosphate kinase Al, NM23 17.5 25 3
H2A1 Histone H2A type 1 14 12 2
58


To determine whether Zar protein interactions change during maturation, proteins
were purified from immature and matured oocytes expressing GST, GST-Zarl or GST-
Zar2 (fig. 18). Purified proteins were eluted from the glutathione sepharose beads with
10 mM reduced glutathione, separated by electrophoresis and visualized using a high
sensitivity Coomassie-based dye.
250 kDa
150 kDa
100 kDa
75 kDa
50 kDa
37 kDa
25 kDa
20 kDa
15 kDa
I P
GST
I P
GST-
Zarl
I P
GST-
Zar2
Figure 18. GST-Zarl and GST-Zar2 affinity purification during maturation.
Lysates from immature (I) or progesterone-matured (P) oocytes were injected with 3.7 ng
of GST, 20 ng of GST-Zarl or 20 ng of GST-Zar2 were purified on glutathione sepharose
beads and then eluted with reduced glutathione. Weight of molecular markers is
indicated at left. Exogenous GST and GST-Zar fusion proteins are indicated by white
arrows. Bands that changed during maturation are indicated by asterisks (*).
59


There were some faint differences between GST-Zarl and GST-Zar2 in the 30-70
kDa range but they did not image well and are not indicated. GST-Zarl and GST-Zar2
showed very similar banding patterns to each other, suggesting that Zarl and Zar2
interacted with many of the same proteins. This was not surprising as they have very
similar translation repression functions and high levels of amino acid sequence
homology, so it seemed likely that they would bind many of the same proteins.
In both the GST-Zarl and GST-Zar2 lanes, bands were present in immature
oocytes around 90-100 kDa (upper asterisk), 60 kDa (middle asterisk) and 50 kDa that
were not present in lysates from mature oocytes. These are all bands that were sequenced
and all three contained translation factors, such as Xpatla, Xp54 and ePAB, suggesting
that Zar interacted with translation factors and these interactions changed during meiotic
maturation, coincident with change in Zar function.
Do Zarl and Zar2 interactions with translation factors change during maturation?
Rationale:
Many translation regulation mechanisms disrupt the assembly of the translation
initiation complex and the circularization of the mRNA. The translation initiation
complex is composed in part of ePAB and the eukaryotic initiation factors, eIF4E-la and
eIF4G (which are hereafter referred to simply as 4E-la and 4G). For example,
deadenylation ultimately acts through inhibiting initiation complex formation as the
shorter poly(A) recruits fewer ePAB molecules, thereby weakening the effect of 4G
stabilization of the 4E-la: 5-cap interaction. As Zar proteins did not appear to recruit
deadenylases to the mRNA, thereby shortening the poly(A), an alternative option was
60


that Zar proteins prevent assembly through disruption of interactions among translation
initiation complex components.
Figure 19. Cartoon of select translation initiation factors.
Translation is activated through formation of the translation initiation complex. Binding
of 4E-la to the 5 cap (m7-GTP) enhances ribosome loading and translation initiation.
The ePAB:4G interaction strengthens between 4E-la and the 5 cap. The 4e-lb and 4E-T
proteins play repressive roles in translation regulation by disrupting binding of 4E-la to
the 5 cap.
Because most cap-dependent mechanisms depend on 4E-la binding to the 5 cap
(m7 GTP) and to 4G, disruption of these interactions results in translational repression.
For example, the translation repressor Maskin binds 4E-la directly and competes with
binding to 4G, thereby sequestering 4E-la from the rest of the translation initiation
complex (fig 3 A). Formation of the complex can also be prevented through disruption of
the cap:4E interaction. The repressor 4E-T recruits 4E-lb to the mRNA rather than 4E-
la, and 4E-lb has lower affinity for the cap, so translation is inhibited (fig. 19).
Disruption of other interactions, such the ePAB:4G interaction, also reduces translation.
The RNA-binding translation repressor Musashi operates through disrupting this
interaction.
61


Results:
To determine whether Zar proteins disrupt formation of the translation initiation
complex, Zar2 immunoprecipitation was performed. However, as Zar2 protein levels
decline during maturation, which is a confounding factor, GST-Zar affinity purification
was also performed with lysates from immature and progesterone-matured oocytes
expressing GST, GST-Zarl or GST-Zar2. GST-Zar fusion proteins were stable during
maturation and so this allowed evaluation of protein interactions, independent of
changing Zar levels. Proteins purified in the presence of RNAse, were eluted with
reduced glutathione, run next to uninjected total lysate (1/20 input) and blotted with
antibodies against 4G, 4E-T, 4E, and ePAB (fig. 20A). Zar2 immunoprecipitation was
performed in the presence of RNAse with a-Zar2 antibody or without, and purified
proteins were blotted with the same panel of antibodies used to evaluate the GST affinity
purified proteins (fig. 20B). Zarl was not immunoprecipitated as available Zarl
antibodies have not immunoprecipitated (not shown).
Proteins were blotted with antibodies against GST to verify that proteins were
expressed and to determine whether they were expressed to similar levels. The fusion
proteins were expressed and similar amounts of protein were purified, although there was
slightly more GST-Zar2 than GST-Zarl. The blot also shows that GST-Zarl and GST-
Zar2 were stable during maturation. Membranes were blotted with Zarl and Zar2
antibodies to determine expression of fusion proteins relative to endogenous Zar, and
fusion proteins were overexpressed about 3-5 fold relative to endogenous levels. The
Zar2 from the immunoprecipitation showed that Zar2 was purified and there was less
Zar2 in mature oocytes.
62


A
GST affinity purification
GST
Zarl
Zar2
4G
4E-T
ePAB
4E-lb
B Zar2 immunoprecipitation
Zar2
4G
4E-T
ePAB
o
4E
*L

MU
ip i p i p
____i i___i i__
1/20
input
no ab
a-
Zar2
Figure 20. Zar protein interactions with translation initiation factors.
GST-Zar affinity purifications and Zar2 immunoprecipitation were performed to identify
interactions with translation initiation factors and to determine if these interactions
changed during meiotic maturation. A. Proteins were purified from lysates expressing
GST, GST-Zarl or GST-Zar2 on glutathione sepharose beads and eluted with reduced
glutathione. Purified proteins were blotted with GST, Zarl, Zar2, 4G, 4E-T and 4E
antibodies. Representative data is shown from n = 2. B. Lysates from immature (I) and
mature (P) oocytes were immunoprecipitated with an anti-Zar2 antibody using protein A
dynabeads and LDS extraction. Purified proteins were blotted with Zar2, 4G, 4E-T and
4E antibodies, n = 1. (1/20 input, uninjected cell lysate; no ab, sham purified with no
antibody; white circle, 4E-la protein; black circle, 4E-lb protein.)
Membranes were blotted with antibodies against 4G, which was present in equal
amounts in immature and mature oocyte lysates. Faint bands were present after GST
purification in negative control lanes as well as in GST-Zar lanes. As the signal was very
weak relative to input and appeared in negative lanes, it was concluded that the observed
signal could be attributed to background. It is worth noting that a background 4G binding
63


signal has been observed before when using glutathione sepharose purification (Minshall
2007). When blotted against the Zar2 immunoprecipitated proteins no distinct band was
present, although there was a blurred band of similar size present in both immature and
mature oocytes, similar in shape and intensity to other background bands. This band was
also observed when a non-specific antibody (a-GST IgG) was used to immunoprecipitate
(not shown). Therefore, it was concluded that neither Zarl nor Zar2 interacted with 4G.
Proteins were blotted against 4E-T and the antibody recognized two bands, one at
about 100 kDa and one at about 130 kDa, representing the short and long isoforms of
Xenopus 4E-T respectively. Both isoforms were present in immature and mature oocytes
and both isoforms were shifted upward in mature oocytes, likely due to phosphorylation
of the protein. Both GST-Zarl and GST-Zar2 interacted strongly with both isoforms of
4E-T in immature oocytes and the interaction was lessened in mature oocytes, although
some interaction was still detectable. Interestingly, there was more of the long isoform of
4E-T bound in mature oocytes, relative to the short isoform. This pattern was repeated
when Zar2 immunoprecipitates were blotted with the antibody. Both isoforms were
present in equal amounts in immature oocytes but only the long isoform interaction was
detected in mature oocyte lysates. This suggests Zarl and Zar2 interact strongly with
both isoforms of 4E-T in immature oocytes and the interaction is removed for the short
isoform but only reduced for the long isoform. As only behavior of the short isoform has
been described (Minshall et al., 2007), this finding provides new information about the
behavior of 4E-T in translational regulation complexes.
Antibodies against ePAB showed that ePAB interacted with GST-Zarl and GST-
Zar2, but not with GST alone and the interaction was reduced in mature oocytes,
64


although only slightly for Zarl. More ePAB protein was detected in the GST-Zar2 lane
than in GST-Zarl. When Zar2 immunoprecipitates were blotted with the ePAB antibody
a band was present in immature oocyte lysates that was fainter in lysates from mature
oocytes. This suggests that Zar proteins interacted with ePAB in immature oocytes and
the interaction decreased during maturation.
To evaluate the interactions of Zar with 4E-la and 4E-lb, two antibodies were
used, one that recognizes both proteins and one that only recognizes 4E-lb. When the 4E
antibody (recognizes both 4E-la and 4E-lb) was blotted against GST purified proteins, a
strong band at 25 kDa was recognized in all lanes that made results uninterpretable (not
shown). However, the 4E-lb antibody did not recognize the interfering band as strongly.
4E-lb interacted strongly with GST-Zarl and GST-Zar2 in immature oocytes and the
interaction was reduced during maturation, although some protein was still detectable.
To verify the 4E-lb interaction and to determine if Zar2 interacted with 4E-lb, the Zar2
immunoprecipitates were blotted with the 4E antibody that recognizes both proteins.
Both 4E-la (white circle) and 4E-lb (black circle) were detected in total cell lysate, but
only 4E-lb was detected in the purified immunoprecipitate. These data provide no
indication that Zarl or Zar2 interacted with 4E-la, but suggest that both Zar proteins
interacted with 4E-lb and the interaction was reduced during maturation.
It was surprising that there were not differences in the evaluated protein
interactions of Zarl compared to those of Zar2. To determine whether GST-Zar fusion
proteins were interacting with their endogenous counterparts, the GST affinity purified
proteins were blotted with Zarl and Zar2 antibodies. Endogenous Zar protein levels were
not higher in GST-Zarl or GST-Zar2 lanes compared to GST alone. In contrast, blots
65


with Zar2 antibody showed that there was more endogenous Zar2 protein in both GST-
Zarl and GST-Zar2 lanes than in GST alone. Surprisingly, endogenous Zar2 was co-
purified by both GST-Zarl and GST-Zar2, while endogenous Zarl was not detected at
high levels in proteins co-purified with GST-Zarl or with GST-Zar2. The implications of
this finding have not yet been resolved, but it suggests a difference in how Zarl and Zar2
interact with translational regulation complexes.
66


CHAPTER V
DISCUSSION
The primary purpose of this study was to show that Zar proteins are translation
factors in Xenopus oocytes and that their mechanism of action is consistent with that
established for the TCS. Beyond that, this study sought to further characterize the
mechanism of Zar-mediated translation repression by evaluating the requirement of the
poly(A) tail and identifying protein:protein interactions.
Major findings
New MS2-Zar fusion proteins, redesigned with longer inter-domain linkers,
repressed translation to a greater extent than the Zar-MS2 fusion proteins with minimal
linker regions. New MS2-Zar fusion proteins expressed more stably in immature oocytes
and the new MS2-Zarl fusion proteins were more stable during maturation, behaving
more like endogenous Zarl. The inclusion of extra stem-loops did not affect Zar-
mediated translational regulation.
While Zar2 did not require a poly(A) tail to repress translation, repression by Zarl
did require a poly(A) tail. Neither Zarl nor Zar2 repressed translation on a reporter with
an extended poly(A) tail. Neither Zarl nor Zar2 imparted cytoplasmic polyadenylation
or translational activation of reporter mRNAs during maturation, but relief of repression
was indicative of a change in Zar function.
Zar proteins interacted with other translation factors, including CPEB, Xp54,
Xpatla, 4E-lb, 4E-T and ePAB. Interactions changed during meiotic maturation,
coincident with change in Zar function.
67


Redesign of MS2-Zar fusion proteins improved stability and function
The tethered assay had been used to show translation repression by the N-terminal
region of Zarl and Zar2, but the effect was modest (about 30%) and the full length Zar
fusion proteins showed even less translation repression. This raised concern as to
whether the translational repression effect was great enough to be physiologically
relevant. The new MS2-Zar fusion proteins repressed translation by about 50%. Xp54
(Minshall and Standard, 2004), xPatla (Nakamura et al., 2010) and 4E-T (Minshall et al.,
2007) have all been shown to repress by about 50-60% in Xenopus oocytes. CPEB
represses by about 30-40% in neurons Therefore, Zar-mediated translation repression
was on par with other known translation repressors, strengthening the case that Zar
proteins function as translational regulators.
The more robust fusion proteins were technically advantageous as well, providing
a wide enough range of translation repression that mechanistic studies could be
performed and Zarl and Zar2 could be compared to each other effectively. Stabilization
of MS2-Zarl in response to progesterone, such that it behaved more like endogenous,
allowed for the determination that Zarl protein function changed during maturation
independent of its absolute protein levels, with less repression in mature oocytes despite
the same amount of MS2-Zarl fusion protein.
This is similar to CPEB, which represses translation in immature oocytes and
although protein levels decrease by about 80% during maturation, the remaining portion
changes behavior: the remaining CPEB is translationally activating in mature oocytes
(Thom et al., 2003). Musashi function also changes independent of protein levels.
Musashi represses translation in immature oocytes and while protein levels increase
68


during maturation, Musashi protein interactions change and so translation repression is
relieved during maturation (Kawahara et al., 2008). In contrast, the level of translation
repression by Xpatla is correlated with the amount of Xpatla protein present (Marnef et
al, 2010).
It is possible that the improvement in protein stability and function in the tethered
assay was due to the increased length between the domains and/or to the reversed
orientation of the two domains. It is notable that the GST-Zar fusion proteins were also
redesigned with the same inter-domain linkers and showed considerable improvement in
protein stability (data not shown). The old GST-Zar fusion proteins were already
constructed with the GST tag at the N-terminus of the protein and Zar at the C-terminus
and so the only change to their structure was the inclusion of the linker. This suggests the
increased linker length likely contributed improved stability to both GST-Zar fusion
proteins and to MS2-Zar fusion proteins.
The inclusion of 4 stem-loops rather than 2 in the firefly luciferase reporters did
not significantly affect Zar-mediated translational regulation. Other labs have shown that
3 stem-loops were better than 1 and that 9 were not better than 3 (Collier et al., 2005) and
that 2 stem-loops were better than 1 (Bardwell and Wickens, 1990). We have used 2
stem-loops in the past and this study showed that the inclusion of 4 stem-loops was not
better than 2, suggesting that 2 or 3 stem-loops behave similarly.
Zarl required a poly(A) tail to repress translation; Zar2 did not
While both Zarl and Zar2 repressed translation to similar levels on a reporter with
a poly(A) tail of 30 adenylate residues, only Zar2 was able to repress when the reporter
was lacking a poly(A) tail. This result was surprising and suggests that Zarl and Zar2
69


employ different mechanisms of translational repression. As the TCS does not require a
poly(A) tail to repress translation (Wang et al., 2008), these results suggest that the
mechanism of Zarl-mediated translation repression is not consistent with that established
for the TCS, although that of Zar2 may be. The requirement of a poly(A) tail by Zarl is
similar to that seen for Musashi, which requires a poly(A) tail and disrupts the ePAB:4G
interaction (Kawahara et al., 2008). The lack of a poly(A) tail requirement by Zar2 is
similar to that seen for Xp54 (Minshall et al., 2001), which interacts with 4E (Minshall
and Standard, 2004), and to that seen for Rap55 (Tanaka et al., 2006).
Both Zarl- and Zar2-mediated translation repression was overcome by an
extended poly(A) tail. This is similar to Xp54 whose repression is overcome by a
poly(A) tail (Minshall et al., 2009) and in contrast to Rap55 which represses even in the
presence of a long poly(A) tail.
Translation regulation by Zar did not involve polyadenylation
As poly(A) tail length and level of translation are correlated in Xenopus oocytes,
regulating the length of the tail is a common and effective method of regulating protein
expression. There was no indication that Zar-mediated translational regulation involved
changes to the length of the poly(A) tail, neither deadenylation nor polyadenylation.
The poly(A) tail was not shortened by Zar in tethered assay experiments, no
deadenylases were recovered in GST pull-downs (fig. 17 and table 1) and work from
other members of our lab has shown that the deadenylase PARN did not interact with Zar
(not shown). This suggests that deadenylation of the mRNA was not required for Zar-
mediated repression and that Zar proteins did not recruit or employ deadenylase activity.
The lack of a requirement for deadenylase activity was consistent with observations about
70


the TCS, as the TCS does not require a poly(A) tail to repress translation, implying that
deadenylase activity is not necessary.
Recruitment of deadenylases is part of the mechanisms of CPEB and Pumilio:
CPEB recruits PARN (Kim and Richter, 2006) and Pumilio recruits the CCR4/Not/Pop
deadenylase complex (Chritton and Wickens, 2010). Zar did not appear to employ this
type of mechanism. It should be noted though, that the CCR4/Not/Pop complex is an
example of a deadenylase that can repress translation even in the absence of its
deadenylase activity (Collart and Panasenko, 2012). Therefore, the lack of Zar-mediated
deadenylase activity does not preclude the possibility that Zar proteins recruit
deadenylases to bring about translation repression.
Zar proteins did not impart cytoplasmic polyadenylation to the reporter mRNAs
in the tethered assay. This is in stark contrast to the behavior of the TCS and suggests
that Zar proteins were not responsible for TCS-mediated cytoplasmic polyadenylation
and translational activation in mature oocytes.
Perhaps Zar proteins (particularly Zar2) mediate translation repression by the TCS
in immature oocytes, while some unidentified factor binds the TCS to bring about
polyadenylation and translational activation in mature oocytes. This factor could be
similar to CPEB or Musashi, which recruit the poly(A) polymerase Gld-2 during meiotic
maturation, to promote polyadenylation and translational activation of the mRNA
(Radford et al., 2008).
Zar proteins interacted with components of translational regulation complexes
Zar proteins interacted with Xp54, Rap 55, CPEB, and Xpatla in immature
oocytes, all proteins with identified roles in translational regulation. Zar also interacted
71


with the translation initiation factors 4E-lb, 4E-T and ePAB, and these interactions were
reduced during maturation, consistent with relief of Zar-mediated repression.
Proposed Models of Zarl- and Zar2-mediated translation regulation
Taken together, the tethered assay and affinity purification data suggest that Zar
proteins are translation regulators that interact with the translation initiation complex.
Proposed models of Zar2-mediated translational regulation
Immature
Translational repression
Low translation
Relief of repression
Medium translation
B
Mature
Translational activation
High translation
Figure 21. Proposed models of Zar2-mediated translation regulation.
Tethered assay and protein interaction data were used to create models of proposed Zar2-
mediated translational regulation. A. Zar2 (grey circle) is proposed to repress translation
by recruiting 4E-T and 4E-lb. B. Zar2 may also disrupt ePAB interactions. C. Release
of 4E-T, 4E-lb and ePAB would result in relief of repression and moderate levels of
translation. D. Replacement of some other polymerase activity promoting factor onto
the TCS may account for polyadenylation and higher levels of translational activation.
Zar2-mediated translation repression was poly(A) tail independent and Zar2
protein levels were reduced during maturation, although about 20% remained. Zar2
72


interacted strongly with 4E-T, 4E-lb and ePAB in immature oocytes and these
interactions decreased during maturation.
I propose that Zar2 recruits 4E-T and 4E-lb to TCS-containing mRNAs, thereby
preventing cap:4E-la:4G interactions and repressing translation (fig 21 A). It is possible
that Zar2 also interacts with ePAB to prevent initiation complex formation (fig 21B).
During meiotic maturation, the interactions would be released and Zar2 would no longer
prevent cap:4E-la:4g binding. Repression would be relieved with moderate levels of
translation on mRNAs where Zar2 was still bound (fig 21C). TCS-containing mRNAs
not bound to Zar2 would bind some unidentified polymerase activity promoting factor,
leading to cytoplasmic polyadenylation and higher levels of translation (fig 21D).
Proposed models of Zarl-mediated translational regulation
The Zarl data was inconsistent and so no clear model could be developed.
However, there are other protein interactions that could be involved. For example, the
translation factor Paip (poly(A)-binding protein-interacting protein) binds to and
represses ePAB (Kahvejian et al., 2001). Zarl could recruit Paip to repress translation, in
a poly(A)-dependent manner.
Alternatively, Zarl could recruit Xpatla and Xp54 to recruit translation. As
Xpatla is degraded during maturation (Nakamura et al., 2010), Xp54 would be released,
thus relieving repression. This model would not explain the poly(A) dependence of Zarl-
mediated translation repression, but might address the observed relief of repression
during maturation despite Zarl protein levels remaining constant. These models will be
the subject of future studies.
73


REFERENCES
Aitken, C.E., Lorsch, J.R., 2012. A mechanistic overview of translation initiation in eukaryotes.
Nature structural & molecular biology.
Amiel, A., Leclere, L., Robert, L., Chevalier, S., Houliston, E., 2009. Conserved functions for
Mos in eumetazoan oocyte maturation revealed by studies in a cnidarian. Curr Biol 19,
305-311.
Arthur, P.K., Claussen, M., Koch, S., Tarbashevich, K., Jahn, O., Pieler, T., 2009. Participation
of Xenopus Elr-type proteins in vegetal mRNA localization during oogenesis. The
Journal of biological chemistry 284, 19982-19992.
Bardwell, V.J., Wickens, M., 1990. Purification of RNA and RNA-protein complexes by an R17
coat protein affinity method. Nucleic Acids Res 18, 6587-6594.
Benoit, P., Papin, C., Kwak, J.E., Wickens, M., Simonelig, M., 2008. PAP- and GLD-2-type
poly(A) polymerases are required sequentially in cytoplasmic polyadenylation and
oogenesis in Drosophila. Development (Cambridge, England) 135, 1969-1979.
Bobe, J., Nguyen, T., Mahe, S., Monget, P., 2008. In silico identification and molecular
characterization of genes predominantly expressed in the fish oocyte. BMC Genomics 9,
499.
Bontems, F., Stein, A., Marlow, F., Lyautey, J., Gupta, T., Mullins, M.C., Dosch, R., 2009.
Bucky ball organizes germ plasm assembly in zebrafish. Curr Biol 19, 414-422.
Brevini, T.A., Cillo, F., Colleoni, S., Lazzari, G., Galli, C., Gandolfi, F., 2004. Expression
pattern of the maternal factor zygote arrest 1 (Zarl) in bovine tissues, oocytes, and
embryos. Mol Reprod Dev 69, 375-380.
Brook, M., Smith, J.W., Gray, N.K., 2009. The DAZL and PABP families: RNA-binding
proteins with interrelated roles in translational control in oocytes. Reproduction 137, 595-
617.
Carabatsos, M.J., Elvin, J., Matzuk, M.M., Albertini, D.F., 1998. Characterization of oocyte and
follicle development in growth differentiation factor-9-deficient mice. Dev Biol 204, 373-
384.
Charlesworth, A., Cox, L.L., MacNicol, A.M., 2004. Cytoplasmic polyadenylation element
(CPE)- and CPE-binding protein (CPEB)-independent mechanisms regulate early class
maternal mRNA translational activation in Xenopus oocytes. The Journal of biological
chemistry 279, 17650-17659.
74


Charlesworth, A., Ridge, J.A., King, L.A., MacNicol, M.C., MacNicol, A.M., 2002. A novel
regulatory element determines the timing of Mos mRNA translation during Xenopus
oocyte maturation. The EMBO journal 21, 2798-2806.
Charlesworth, A., Welk, J., MacNicol, A., 2000. The temporal control of Weel mRNA
translation during Xenopus oocyte maturation is regulated by cytoplasmic
polyadenylation elements within the 3'-untranslated region. Developmental biology 227,
706-719.
Charlesworth, A., Wilczynska, A., Thampi, P., Cox, L.L., MacNicol, A.M., 2006. Musashi
regulates the temporal order of mRNA translation during Xenopus oocyte maturation.
The EMBO journal 25, 2792-2801.
Charlesworth, A., Yamamoto, T.M., Cook, J.M., Silva, K.D., Kotter, C.V., Carter, G.S., Holt,
J.W., Lavender, H.F., MacNicol, A.M., Ying Wang, Y., Wilczynska, A., 2012. Xenopus
laevis zygote arrest 2 (zar2) encodes a zinc finger RNA-binding protein that binds to the
translational control sequence in the maternal Weel mRNA and regulates translation.
Dev Biol 369, 177-190.
Chen, J.Y., Li, X.X., Xu, Y.K., Wu, H., Zheng, J.J., Yu, X.L., 2014. Developmental competence
and gene expression of immature oocytes following liquid helium vitrification in bovine.
Cryobiology 69, 428-433.
Chen, X., Zaro, J.L., Shen, W.-C.C., 2013. Fusion protein linkers: property, design and
functionality. Advanced drug delivery reviews 65, 1357-1369.
Choo, S., Heinrich, B., Betley, J.N., Chen, Z., Deshler, J.O., 2005. Evidence for common
machinery utilized by the early and late RNA localization pathways in Xenopus oocytes.
Developmental biology 278, 103-117.
Christensen, A.K., Kahn, L.E., Bourne, C.M., 1986. Circular polysomes predominate on the
rough endoplasmic reticulum of somatotropes and mammotropes in the rat anterior
pituitary. The American journal of anatomy 178, 1-10.
Chritton, J.J., Wickens, M., 2010. Translational repression by PUF proteins in vitro. RNA (New
York, N.Y.) 16, 1217-1225.
Colegrove-Otero, L.J., Minshall, N., Standart, N., 2005. RNA-binding proteins in early
development. Crit Rev Biochem Mol Biol 40, 21-73.
Collart, M.A., Panasenko, O.O., 2012. The Ccr4not complex. Gene 492, 42-53.
75


Coller, J., Wickens, M., 2002. Tethered function assays using 3' untranslated regions. Methods
(San Diego, Calif.) 26, 142-150.
Coller, J., Wickens, M., 2007. Tethered function assays: an adaptable approach to study RNA
regulatory proteins. Methods in enzymology 429, 299-321.
Collier, B., Gorgoni, B., Loveridge, C., Cooke, H.J., Gray, N.K., 2005. The DAZL family
proteins are PABP-binding proteins that regulate translation in germ cells. EMBO J 24,
2656-2666.
Cooke, A., Prigge, A., Wickens, M., 2010. Translational repression by deadenylases. J Biol
Chem 285, 28506-28513.
Copeland, P.R., Wormington, M., 2001. The mechanism and regulation of deadenylation:
identification and characterization of Xenopus PARN. RNA 7, 875-886.
de Moor, C.H., Meijer, H., Lissenden, S., 2005. Mechanisms of translational control by the 3'
UTR in development and differentiation. Seminars in Cell & Developmental Biology 16,
49-58.
Deshler, J.O., Highett, M.I., Schnapp, B.J., 1997. Localization of Xenopus Vgl mRNAby Vera
protein and the endoplasmic reticulum. Science (New York, N. Y.) 276, 1128-1131.
Dumont, J.N., 1972. Oogenesis in Xenopus laevis (Daudin). I. Stages of oocyte development in
laboratory maintained animals. JMorphol 136, 153-179.
Dupre, A., Jessus, C., Ozon, R, Haccard, O., 2002. Mos is not required for the initiation of
meiotic maturation in Xenopus oocytes. EMBO J 21, 4026-4036.
Elis, S., Batellier, F., Couty, I., Balzergue, S., Martin-Magniette, M.L., Monget, P., Blesbois, E.,
Govoroun, M.S., 2008. Search for the genes involved in oocyte maturation and early
embryo development in the hen. BMC Genomics 9, 110.
Farley, B., Ryder, S., 2008. Regulation of maternal mRNAs in early development. Critical
reviews in biochemistry and molecular biology 43, 135-162.
Ferrell, J.E., 1999. Xenopus oocyte maturation: new lessons from a good egg. Bioessays.
Fox, C.A., Sheets, M.D., Wickens, M.P., 1989. Poly(A) addition during maturation of frog
oocytes: distinct nuclear and cytoplasmic activities and regulation by the sequence
UUUUUAU. Genes Dev 3, 2151-2162.
76


Gallie, D.R., 1991. The cap and poly (A) tail function synergistically to regulate mRNA
translational efficiency. Genes & Development.
George, R.A., Heringa, J., 2002. An analysis of protein domain linkers: their classification and
role in protein folding. Protein Eng 15, 871-879.
Gray, N.K., Coller, J.M., Dickson, K.S., Wickens, M., 2000. Multiple portions of poly(A)-
binding protein stimulate translation in vivo. The EMBO journal 19, 4723-4733.
Hake, L.E., Richter, J.D., 1994. CPEB is a specificity factor that mediates cytoplasmic
polyadenylation during Xenopus oocyte maturation. Cell 79, 617-627.
Hanahan, D., lessee, I, Bloom, F.R., 1991. Plasmid transformation of Escherichia coli and other
bacteria. Methods Enzymol 204, 63-113.
Howard, E.L., Charlesworth, A., Welk, I, MacNicol, A.M., 1999. The mitogen-activated protein
kinase signaling pathway stimulates mos mRNA cytoplasmic polyadenylation during
Xenopus oocyte maturation. Molecular and cellular biology 19, 1990-1999.
Hu, I, Wang, F., Zhu, X., Yuan, Y., Ding, M., Gao, S., 2010. Mouse ZARl-like (XM_359149)
colocalizes with mRNA processing components and its dominant-negative mutant caused
two-cell-stage embryonic arrest. Dev Dyn 239, 407-424.
Huarte, J., Stutz, A., O'Connell, M.L., Gubler, P., Belin, D., Darrow, A.L., Strickland, S.,
Vassalli, J.D., 1992. Transient translational silencing by reversible mRNA deadenylation.
Cell 69, 1021-1030.
Kahvejian, A., Roy, G., Sonenberg, N., 2001. The mRNA closed-loop model: the function of
PABP and PABP-interacting proteins in mRNA translation. Cold Spring Harb Symp
Quant Biol 66, 293-300.
Kawahara, H., Imai, T., Imataka, H, Tsujimoto, M., Matsumoto, K., Okano, H., 2008. Neural
RNA-binding protein Musashil inhibits translation initiation by competing with eIF4G
for PABP. J Cell Biol 181, 639-653.
Keryer-Bibens, C., Barreau, C., Osborne, H., 2008. Tethering of proteins to RNAs by
bacteriophage proteins. Biology of the cell / under the auspices of the European Cell
Biology Organization 100, 125-138.
Kim, J.H., Richter, J.D., 2006. Opposing polymerase-deadenylase activities regulate cytoplasmic
polyadenylation. Mol Cell 24, 173-183.
77


Kim, K.H., Lee, K.A., 2014. Maternal effect genes: Findings and effects on mouse embryo
development. Clin Exp Reprod Med 41, 47-61.
Krieg, P.A., Melton, D.A., 1984. Functional messenger RNAs are produced by SP6 in vitro
transcription of cloned cDNAs. Nucleic Acids Res 12, 7057-7070.
Kwak, J.E., Wang, L., Ballantyne, S., Kimble, J., Wickens, M., 2004. Mammalian GLD-2
homologs are poly(A) polymerases. Proceedings of the National Academy of Sciences of
the United States of America 101, 4407-4412.
Kwon, S., Abramson, T., Munro, T.P., John, C.M., Kohrmann, M., Schnapp, B.J., 2002.
UUCAC- and vera-dependent localization of VegT RNA in Xenopus oocytes. Curr Biol
12, 558-564.
Lasko, P., 2011. Posttranscriptional regulation in Drosophila oocytes and early embryos. Wiley
Interdiscip Rev RNA 2, 408-416.
Lee, M., Bang, K., Kwon, H., Cho, S., 2013. Enhanced antibacterial activity of an attacin-
coleoptericin hybrid protein fused with a helical linker. Mol Biol Rep 40, 3953-3960.
Lindeman, R.E., Pelegri, F., 2010. Vertebrate maternal-effect genes: Insights into fertilization,
early cleavage divisions, and germ cell determinant localization from studies in the
zebrafish. Mol Reprod Dev 77, 299-313.
MacNicol, M.C., Pot, D., MacNicol, A.M., 1997. pXen, a utility vector for the expression of
GST-fusion proteins in Xenopus laevis oocytes and embryos. Gene 196, 25-29.
Mailer, J.L., Schwab, M.S., Gross, S.D., Taieb, F.E., Roberts, B.T., Tunquist, B.J., 2002. The
mechanism of CSF arrest in vertebrate oocytes. Mol Cell Endocrinol 187, 173-178.
Marnef, A., Maldonado, M., Bugaut, A., Balasubramanian, S., Kress, M., Weil, D., Standard, N.,
2010. Distinct functions of maternal and somatic Patl protein paralogs. RNA (New York,
N.Y.) 16, 2094-2107.
Marnef, A., Sommerville, J., Ladomery, M., 2009. RAP55: insights into an evolutionarily
conserved protein family. The international journal of....
Masui, Y., 2001. From oocyte maturation to the in vitro cell cycle: the history of discoveries of
Maturation-Promoting Factor (MPF) and Cytostatic Factor (CSF). Differentiation 69, 1-
17.
Masui, Y., Wang, P., 1998. Cell cycle transition in early embryonic development of Xenopus
laevis. Biol Cell 90, 537-548.
78


Matzuk, M.M., 2000. Revelations of ovarian follicle biology from gene knockout mice. Mol Cell
Endocrinol 163, 61-66.
McGrew, L.L., Dworkin-Rastl, E., Dworkin, M.B., Richter, J.D., 1989. Poly(A) elongation
during Xenopus oocyte maturation is required for translational recruitment and is
mediated by a short sequence element. Genes Dev 3, 803-815.
Michailidis, G., Argiriou, A., Avdi, M., 2010. Expression of chicken zygote arrest 1 (Zarl) and
Zarl-like genes during sexual maturation and embryogenesis. Vet Res Commun 34, 173 -
184.
Minami, N., Suzuki, T., Tsukamoto, S., 2007. Zygotic gene activation and maternal factors in
mammals. Journal of Reproduction and ....
Minshall, N., Allison, R., Marnef, A., Wilczynska, A., Standard, N., 2010. Translational control
assessed using the tethered function assay in Xenopus oocytes. Methods 51, 165-169.
Minshall, N., Kress, M., Weil..., D., 2009. Role of p54 RNA helicase activity and its C-terminal
domain in translational repression, P-body localization and assembly. Molecular biology
of the ....
Minshall, N., Reiter, M.H., Weil, D., Standard, N., 2007. CPEB interacts with an ovary-specific
eIF4E and 4E-T in early Xenopus oocytes. The Journal of biological chemistry 282,
37389-37401.
Minshall, N., Standard, N., 2004. The active form of Xp54 RNA helicase in translational
repression is an RNA-mediated oligomer. Nucleic acids research 32, 1325-1334.
Minshall, N., Thom, G., Standard, N., 2001. A conserved role of a DEAD box helicase in mRNA
masking. RNA 7, 1728-1742.
Murakami, M.S., Copeland, T.D., Vande Woude, G.F., 1999. Mos positively regulates Xe-Weel
to lengthen the first mitotic cell cycle of Xenopus. Genes Dev 13, 620-631.
Murakami, M.S., Moody, S.A., Daar, I.O., Morrison, D.K., 2004. Morphogenesis during
Xenopus gastrulation requires Weel-mediated inhibition of cell proliferation.
Development 131, 571-580.
Nakamura, Y., Tanaka, K.J., Miyauchi, M., Huang, L., Tsujimoto, M., Matsumoto, K., 2010.
Translational repression by the oocyte-specific protein PI00 in Xenopus. Developmental
biology 344, 272-283.
79


Nath, A., Sharma, V., Dubey, P.K., Pratheesh, M.D., Gade, N.E., Saikumar, G., Sharma, G.T.,
2013. Impact of gonadotropin supplementation on the expression of germ cell marker
genes (MATER, ZAR1, GDF9, and BMP 15) during in vitro maturation of buffalo
(Bubalus bubalis) oocyte. In Vitro Cell Dev Biol Anim 49, 34-41.
Nishizawa, M., Okazaki, K., Furuno, N., Watanabe, N., Sagata, N., 1992. The 'second-codon
rule' and autophosphorylation govern the stability and activity of Mos during the meiotic
cell cycle in Xenopus oocytes. EMBO J 11, 2433-2446.
Paris, J., Richter, J.D., 1990. Maturation-specific polyadenylation and translational control:
diversity of cytoplasmic polyadenylation elements, influence of poly(A) tail size, and
formation of stable polyadenylation complexes. Mol Cell Biol 10, 5634-5645.
Pennetier, S., Uzbekova, S., Perreau, C., Papillier, P., Mermillod, P., Dalbies-Tran, R., 2004.
Spatio-temporal expression of the germ cell marker genes MATER, ZAR1, GDF9,
BMP15,andVASA in adult bovine tissues, oocytes, and preimplantation embryos. Biol
Reprod 71, 1359-1366.
Pique, M., Lopez, J.M., Foissac, S., Guigo, R., Mendez, R., 2008. A combinatorial code for
CPE-mediated translational control. Cell 132, 434-448.
Prasad, C.K., Mahadevan, M., MacNicol, M.C., MacNicol, A.M., 2008. Mos 3' UTR regulatory
differences underlie species-specific temporal patterns of Mos mRNA cytoplasmic
polyadenylation and translational recruitment during oocyte maturation. Molecular
reproduction and development 75, 1258-1268.
Radford, H.E., Meijer, H.A., de Moor, C.H., 2008. Translational control by cytoplasmic
polyadenylation in Xenopus oocytes. Biochimica et biophysica acta 1779, 217-229.
Rassa, J.C., Wilson, G.M., Brewer, G.A., Parks, G.D., 2000. Spacing constraints on reinitiation
of paramyxovirus transcription: the gene end U tract acts as a spacer to separate gene end
from gene start sites. Virology 274, 438-449.
Richter, J.D., 2007. CPEB: a life in translation. Trends Biochem Sci 32, 279-285.
Sangiorgio, L., Strumbo, B., Brevini, T.A.L., Ronchi..., S., 2008. A putative protein structurally
related to zygote arrest 1 (Zarl), Zarl-like, is encoded by a novel gene conserved in the
vertebrate lineage. ... and Physiology Part B: ....
Smith, R.W., Anderson, R.C., Smith, J.W., Brook, M., Richardson, W.A., Gray, N.K., 2011.
DAZAP1, an RNA-binding protein required for development and spermatogenesis, can
regulate mRNA translation. RNA 17, 1282-1295.
80


Stebbins-Boaz, B., Cao, Q., de Moor, C.H., Mendez, R., Richter, J.D., 1999. Maskin is a CPEB-
associated factor that transiently interacts with elF-4E. Mol Cell 4, 1017-1027.
Stumpff, J., Kellogg, D.R., Krohne, K. A., Su, T.T., 2005. Drosophila Weel interacts with
members of the gammaTURC and is required for proper mitotic-spindle morphogenesis
and positioning. CurrBiol 15, 1525-1534.
Summers, A.F., Pohlmeier, W.E., Sargent, K.M., Cole, B.D., Vinton, R.J., Kurz, S.G., McFee,
R.M., Cushman, R.A., Cupp, A.S., Wood, J.R., 2014. Altered theca and cumulus oocyte
complex gene expression, follicular arrest and reduced fertility in cows with dominant
follicle follicular fluid androgen excess. PLoS One 9, el 10683.
Tadros, W., Lipshitz, FL, 2009. The maternal-to-zygotic transition: a play in two acts.
Development (Cambridge, England) 136, 3033-3042.
Takahashi, K., Kotani, T., Katsu, Y., Yamashita, M., 2014. Possible involvement of insulin-like
growth factor 2 mRNA-binding protein 3 in zebrafish oocyte maturation as a novel cyclin
B1 mRNA-binding protein that represses the translation in immature oocytes. Biochem
Biophys Res Commun 448, 22-27.
Tanaka, K.J., Ogawa, K., Takagi, M., Imamoto, N., Matsumoto, K., Tsujimoto, M., 2006.
RAP55, a Cytoplasmic mRNP Component, Represses Translation in Xenopus Oocytes.
Journal of Biological Chemistry 281.
Thom, G., Minshall, N., Git, A., Argasinska, J., Standard, N., 2003. Role of cdc2 kinase
phosphorylation and conserved N-terminal proteolysis motifs in cytoplasmic
polyadenylation-element-binding protein (CPEB) complex dissociation and degradation.
Biochem J 370, 91-100.
Tominaga, Y., Li, C., Wang, R.H., Deng, C.X., 2006. Murine Weel plays a critical role in cell
cycle regulation and pre-implantation stages of embryonic development. Int J Biol Sci 2,
161-170.
Tunquist, B.J., Mailer, J.L., 2003. Under arrest: cytostatic factor (CSF)-mediated metaphase
arrest in vertebrate eggs. Genes & Development.
Uzbekova, S., Roy-Sabau, M., Dalbies-Tran, R., Perreau, C., Papillier, P., Mompart, F., Thelie,
A., Pennetier, S., Cognie, J., Cadoret, V., Royere, D., Monget, P., Mermillod, P., 2006.
Zygote arrest 1 gene in pig, cattle and human: evidence of different transcript variants in
male and female germ cells. Reprod Biol Endocrinol 4, 12.
81


Varnum, S.M., Wormington, W.M., 1990. Deadenylation of maternal mRNAs during Xenopus
oocyte maturation does not require specific cis-sequences: a default mechanism for
translational control. Genes Dev 4, 2278-2286.
Wang, Y.Y., Charlesworth, A., Byrd, S.M., Gregerson, R., MacNicol, M.C., MacNicol, A.M.,
2008. A novel mRNA 3' untranslated region translational control sequence regulates
Xenopus Weel mRNA translation. Developmental biology 317, 454-466.
Wu, X., Viveiros, M.M., Eppig, J.J., Bai, Y., Fitzpatrick, S.L., Matzuk, M.M., 2003a. Zygote
arrest 1 (Zarl) is a novel maternal-effect gene critical for the oocyte-to-embryo transition.
Nat Genet 33, 187-191.
Wu, X., Wang, P., Brown, C.A., Zilinski, C.A., Matzuk, M.M., 2003b. Zygote arrest 1 (Zarl) is
an evolutionarily conserved gene expressed in vertebrate ovaries. Biol Reprod 69, 861 -
867.
Xu, E.Y., Moore, F.L., Pera, R.A., 2001. A gene family required for human germ cell
development evolved from an ancient meiotic gene conserved in metazoans. Proc Natl
Acad Sci U S A 98, 7414-7419.
Yamamoto, T.M., Cook, J.M., Kotter, C.V., Khat, T., Silva, K.D., Ferreyros, M., Holt, J.W.,
Knight, J.D., Charlesworth, A., 2013. Zarl represses translation in Xenopus oocytes and
binds to the TCS in maternal mRNAs with different characteristics than Zar2. Biochimica
et biophysica acta 1829, 1034-1046.
Yamamoto, T.M., Iwabuchi, M., Ohsumi, K., Kishimoto, T., 2005. APC/C-Cdc20-mediated
degradation of cyclin B participates in CSF arrest in unfertilized Xenopus eggs. Dev Biol
279, 345-355.
Zhao, X., Singh, B., Arlinghaus, R.B., 1991. Inhibition of c-mos protein kinase blocks mouse
zygotes at the pronuclei stage. Oncogene 6, 1423-1426.
82


APPENDIX: SPECIFIC PLASMID CONSTRUCTION
pXenl: This plasmid (MacNicol et al., 1997) (kindly provided by Angus
MacNicol, University of Arkansas for Medical Sciences, AR) provided the backbone for
all plasmids constructed for this project. The plasmid contains an SP6 promoter (Krieg
and Melton, 1984)and a GST coding region in a (3-globin UTR (both 5 and 3 UTR is
present).
pXen N-MS2: PCR primers were designed to amplify the MS2 coding sequence
from pJC5 (Gray et al., 2000) (kindly provided by Jeff Coller, Case Western Reserve
University, OH) with a 5' Ncol site and a 3' Clal site (underlined), forward primer
5'GAT CCCATGGCT TCT AAC TTT ACT CAG TTC and reverse primer 5'-GAT
CAT CGA TGC GTA GAT GCC GGA GTT TGC TGC. Digested PCR product was
ligated into Ncol/Clal digested pXenl, replacing the GST coding sequence.
pXen MS2-Xp54: PCR primers were designed to amplify the Xp54 coding
sequence from Xp54 in the MSP vector (Minshall et al., 2001) (kindly provided by
Nancy Standard, University of Cambridge, UK) with a 5' Kpnl site and a 3' BamHI site
(underlined), forward primer 5'GAT CGG TAC CCA TGA GCA CCG and reverse
primer 5'GAT CGG ATC CTT AAG GTT TGT. Digested PCR product was ligated
into KpnI/BamHI digested pXen N-MS2.
pXen AGST: pXenl (MacNicol et al., 1997) was digested with Ncol and Clal to
remove the GST coding sequence, then treated with Klenow to blunt ends and self-
ligated.
pXen C-MS2: MS2 coding sequence was amplified from pJC5 using primers with
a 5 Xmal site and a 3 Xbal site (underlined), forward primer 5CTA GCC CGG GCT
ATG GCT TCT AAC TTT ACT CAG TTC and reverse primer 5-GAT CTC TAG
AGT TAG TAG ATG CCG GAG TTT GCT G. Digested PCR product was then ligated
into Xmal/Xbal digested pXen AGST.
pXen Zarl-MS2: Full length Zarl was cloned by RT-PCR from total RNA from
immature Xenopus oocytes using primers: Zarl forward 5'ATG GTA CCC TCG AGG
ATG GCT AGC TTC TCA GAG and Zarl reverse 5'-CCT AGC CCG GGC AAT GAT
ATA CTT GAA GCT. PCR products were digested with Xhol and Xmal (underlined)
and ligated into pXen C-MS2 cut with Xhol and Xmal. Full length Zarl was fused 5' of
MS2.
pXen N-Zarl-MS2: 1-159 aa were kept and the C-terminal 160-307 aa were
deleted from pXen Zarl-MS2. Ncol sites (underlined) were introduced by PCR to
remove the C-terminal domain: N-terminal Zarl reverse primer 5'CGA TCC ATG GCT
CAC CCT TCT CTT CCA G and MS2 forward primer 5'-ATG CCC ATG GCC CGG
GAT GGC TTC TAA CTT TAC. The PCR product was cut with Ncol and self-ligated.
83


pXen N-Zar2-MS2: Amino acids 1-158 of zar2b was amplified from pENTR-
zar2b with a 5' Kpnl site and a 3' BamHI site (underlined), forward primer 5'GAT CGG
TAC CAT GGC GGG CTT and reverse primer 5'-GAT CGGATCCGC TCT CTT
CAG, appropriately digested, and ligated into KpnI/BamHI digested pXen C-MS2.
pXen Zar2-MS2: Full length Xenopus laevis Zar2b was amplified from pENTR-
Zar2b with a 5 Kpnl site and a 3 BamHI site (underlined), forward primer 5GAT
CGG TAC CAT GGC GGG CTT TAT GTA TGC GCC G and reverse primer 5-GAT
CGG ATC CGA CGA TGT ACT TGT AGC TGT AAG TG, appropriately digested, and
ligated into KpnI/BamHI digested pXen C-MS2.
pXen MS2-AP: The AP linker, was purchased from Integrated DNA
Technologies (IDT) as a 5 phosphorylated duplex, forward 5phosCGA TCG CCC
CAG CCC CCG CTC CCG CCC CAG CCC CTG CTC CTG CCC CAG CTC CAG
CTC, reverse 5phos-TCG AGA GCT GGA GCT GGG GCA GGA GCA GGG GCT
GGG GCG GGA GCG GGG GCT GGG GC GAT, and was ligated into Clal/Xhol
digested pXen N-MS2.
pXen MS2-EK: The EK linker was purchased from Integrated DNA Technologies
(IDT) as a 5 phosphorylated duplex, forward 5phosCGA TCG CAG AAG CTG CTG
CTA AGG AGG CCG CTG CAA AAG AAG CTG CCG CAA AAG CTC and reverse
5phosTCG AGA GCT TTT GCG GCA GCT TCT TTT GCA GCG GCC TCC TTA
GCA GCA GCT TCT GCG AT, and ligated into Clal/Xhol digested pXen N-MS2.
pXen MS2-GS: The GS linker was purchased from Integrated DNA Technologies
(IDT) as a 5 phosphorylated duplex, forward 5phosCGA TCG CTG GAG GGG GCG
GCA GTG GGG GTG GTG GGA GCG GAG GAG GAG GGA GTG CTC and reverse
5phosTCG AGA GCA CTC CCT CCT CCT CCG CTC CCA CCA CCC CCA CTG
CCG CCC CCT CCA GCG AT, and ligated into Clal/Xhol digested pXen N-MS2.
pXen MS2-AP-Zarl: The full length Zarl sequence was amplified from pXen
Zarl-MS2 with a 5 Xhol restriction site and a 3 Xmal restriction site (underlined),
forward primer 5-GAT CCT CGA GGC CAT GGC TAG CTT CTC AGA G and
reverse primer 5-GAT CCC CGG GTC AAA TGA TAT ACT TGA AGC. Amplified
product and pXen MS2-AP were digested with Xhol and Xmal, then ligated together.
pXen MS2-EK-Zarl: The full length Zarl sequence was amplified from pXen
Zarl-MS2 with a 5 Xhol restriction site and a 3 Xmal restriction site, forward primer
5GAT CCT CGA GGC CAT GGC TAG CTT CTC AGA G and reverse primer
5GAT CCC CGG GTC AAA TGA TAT ACT TGA AGC. The amplified product was
digested with Xhol and Xmal and ligated into Xhol/Xmal digested pXen MS2-EK.
pXen MS2-GS-Zarl: The full length Zarl sequence was amplified from pXen
Zarl-MS2 with a 5 Xhol restriction site and a 3 Xmal restriction site, forward primer
5GAT CCT CGA GGC CAT GGC TAG CTT CTC AGA G and reverse primer
5GAT CCC CGG GTC AAA TGA TAT ACT TGA AGC. The amplified product was
digested with Xhol and Xmal and ligated into Xhol/Xmal digested pXen MS2-GS.
84


pXen MS2-AP-Zar2: The full length Zar2 sequence was amplified from pXen
Zar2-MS2 with a 5 Xhol restriction site and a 3 Xmal restriction site (underlined),
forward primer 5-GAT CCT CGA GGC CAT GGC GGG CTT TAT GTA TGC G and
reverse primer 5-GAT CCC CGG GTC AAT CCG CGA CGA TGT ACT TGT AGC.
The amplified product was digested with Xhol and Xmal and ligated into Xhol/Xmal
digested pXen MS2-AP.
pXen MS2-EK-Zar2: The full length Zar2 sequence was amplified from pXen
Zar2-MS2 with a 5 Xhol restriction site and a 3 Xmal restriction site (underlined),
forward primer 5-GAT CCT CGA GGC CAT GGC GGG CTT TAT GTA TGC G and
reverse primer 5-GAT CCC CGG GTC AAT CCG CGA CGA TGT ACT TGT AGC.
The amplified product was digested with Xhol and Xmal and ligated into Xhol/Xmal
digested pXen MS2-EK.
pXen MS2-GS-Zar2: The full length Zar2 sequence was amplified from pXen
Zar2-MS2 with a 5 Xhol restriction site and a 3 Xmal restriction site (underlined),
forward primer 5-GAT CCT CGA GGC CAT GGC GGG CTT TAT GTA TGC G and
reverse primer 5-GAT CCC CGG GTC AAT CCG CGA CGA TGT ACT TGT AGC.
The amplified product was digested with Xhol and Xmal and ligated into Xhol/Xmal
digested pXen MS2-GS.
pXen rluc: PCR primers were designed to amplify Renilla luciferase (rluc) coding
sequence (plasmid kindly provided by Nancy Standart) with a 5' Ncol site and a 3' Clal
site (underlined), forward primer 5'CAT GCC ATG GCT TCG AAA GTT TAT GAT
CCA and reverse primer 5'-GAT CAT CGA TTT ATT GTT CAT TTT TGA GAA CTC
G. Digested PCR product was then ligated into Ncol/Clal digested pXenl, replacing the
GST coding sequence. This plasmid was linearized with EcoRI prior to in vitro
transcription.
pXen flue: The firefly luciferase (flue) coding sequence was amplified from JC18
(Gray et al., 2000) (kindly provided by Jeff Coller, Case Western Reserve University,
OH) using primers with a 5' Ncol site and a 3' Xhol site (underlined), forward primer
5'GAT CCC ATG GAA GAC GCC AAA AAC AT A AAG and reverse primer
5'GAT CCT CGA GTT AC A ATT TGG ACT TTC CGC C. The PCR product was
inserted into pXenl using Ncol/Xhol, replacing the GST coding sequence.
pXen IIUC-2SL-A33C17: (pXen flue with stem-loops, previously called fluc-2X-
SL): An Nde I site was inserted into the P-globin 3'UTR of pXen flue, 59 nucleotides
upstream of the polyadenylation hexanucleotide, using QuikChange site directed
mutagenesis (Stratagene) to make pXen fluc-Ndel. A DNA duplex (IDT) containing the
sequence of two MS2 stem-loops (2SL) (Bardwell and Wickens, 1990) with 5' and 3'
Ndel sites was digested with Ndel and ligated into Ndel digested pXen fluc-Ndel. This
plasmid was linearized with SacI prior to in vitro transcription.
85


pXen fluc-2SL-Ao: A SacI site (underlined) was QuickChanged (Agilent) into
pXen fluc-2SL- A33C17 just 3 of the polyadenylation hexanucleotide using forward
primer 5-AAG AAA GTT TCT TCA GAG CTC AAA AAA AAA AAA AAA and
reverse primer 5-TTT TTT TTT TTT TTT GAG CTC TGA AGA AAC TTT CTT. The
plasmid was linearized with SacI to remove the poly(A) tail.
pXen fluc-2SL-A3o: QuickChange (Agilent) site directed mutagenesis was used
to remove the A33C17 tract from pXen IIUC-2SL-A33C17 and replace it with an A30 tract
ending in a Bglll site (underlined), forward primer 5AAG TTT CTT CAC ATT CTA
AAA AAA AAA AAA AAA AAA AAA AAA AAA AAG ATC TCT GCA GGT and
reverse primer 5-ACC TGC AGA GAT CTT TTT TTT TTT TTT TTT TTT TTT TTT
TTT TAG AAT GTG AAG AAA CTT. Bglll cuts between the A and the G (underlined)
and leaves a 3 adenosine residue, so when the plasmid is linearized with Bglll, the
poly(A) is terminal and composed of 30 adenosine residues.
pXen IIUC-4SL-A33C17: pXen IIUC-2SL-A33C17 was digested with Ndel, opening
the plasmid and removing the 2SL insert, then ligated with an excess of the two MS2
stem-loop (2SL) DNA duplex from Integrated DNA Technologies (IDT). Colonies were
screened to isolate plasmids containing two inserts (a total of 4 stem-loops) in the correct
orientation.
pXen fluc-4SL-Ao: A SacI site (underlined) was QuickChanged (Agilent) into
pXen fluc-4SL- A33C17 just 3 of the polyadenylation hexanucleotide using forward
primer 5-AAG AAA GTT TCT TCA GAG CTC AAA AAA AAA AAA AAA and
reverse primer 5-TTT TTT TTT TTT TTT GAG CTC TGA AGA AAC TTT CTT. The
plasmid was linearized with SacI to remove the poly(A) tail.
pXen fluc-4SL-A3o: QuickChange (Agilent) site directed mutagenesis was used to
remove the A33C17 tract from pXen fluc-4SL-A33Ci7 and replace it with an A30 tract
ending in a Bglll site (underlined), forward primer 5AAG TTT CTT CAC ATT CTA
AAA AAA AAA AAA AAA AAA AAA AAA AAA AAG ATC TCT GCA GGT and
reverse primer 5-ACC TGC AGA GAT CTT TTT TTT TTT TTT TTT TTT TTT TTT
TTT TAG AAT GTG AAG AAA CTT. Bglll leaves a 3 adenosine residue so when the
plasmid is linearized with Bglll, the poly(A) is terminal and composed of 30 adenylate
residues.
pXenl Zarla: This plasmid codes for GST-Zarl (with minimal linker) and its
construction has been previously described (Yamamoto et al., 2013).
pXen GST -AP: The AP linker was purchased from (IDT) as a 5 phosphorylated
duplex, forward 5phos-CGA TCG CCC CAG CCC CCG CTC CCG CCC CAG CCC
CTG CTC CTG CCC CAG CTC CAG CTC and reverse 5phos-TCG AGA GCT GGA
GCT GGG GCA GGA GCA GGG GCT GGG GCG GGA GCG GGG GCT GGG GC
GAT, and ligated into Clal/Xhol digested pXenl.
86


Full Text

PAGE 1

ZYGOTE ARREST (ZAR) PROTEIN MEDIATED TRANSLATION AL CONTROL IN XENOPUS LAEVIS DEVELOPMENT: REQUIREMENT FOR THE POLY(A) TAIL AND PROTEIN INTERACT IONS by JONATHAN MICHAEL COO K B.S., Metropolitan State University of Denver 2008 B.A., Metropolitan State Uni versity of Denver 2008 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Integrative Biology Program 2015

PAGE 2

201 5 JONATHAN MICHAEL COOK ALL RIGHTS RESERVED

PAGE 3

ii This thesis for the Master of Science degree by Jonathan Michael Cook has been approved for the Integrative Biology Program by Amanda Cha r lesworth Chair Aaron Johnson Bradley Stith April 20, 2015

PAGE 4

iii Cook, Jonathan Michael (M.S., Biology) Zygote arrest (Zar) protein mediated translational control in Xenopus development: a requirement for the poly(A) tail and protein interactions Thesis directed by Assistant Professor Amanda Charlesworth ABSTRACT Zygo te arrest (Zar) proteins, Zar1 and Zar2, are required for successful fertilization and embryogenesis. Synthesis of developmentally important proteins is crucial to embryogenesis and is regulated by translation of mRNA. Zar proteins bind the mRNA of devel opmentally important proteins via a specific sequence found in the RNA called the Translation al Control Sequence (TCS) It is already known that RNAs containing a TCS are translation ally represse d in immature oocytes and activate d in mature oocytes It i s also known that the N termini of both Zar 1 and Zar2 repress translation when tethered to reporter mRNA. The purpose of this study wa s to show Zar proteins are bona fide translation factors and are candidates for mediating translation al regulation by the TCS. In a dual luciferase tethered assay, both Zar proteins repressed translation up to 50% in immature Xenopus oocytes and repression was relieved during oocyte maturation, consistent with translational regulation of developmentally important mRNAs. I nterestingly, Zar1 required the reporter mRNA have a poly ( A ) tail to repress translation w hereas Zar2 did not Furthermore, Zar1 and Zar2 interact ed with overlapping b ut distinct sets of proteins. P roteins recovered from GST affinity purifications inclu de d many known translation factors, such as CPEB, 4E T and embryonic poly(A) binding

PAGE 5

iv protein. In particular, eukaryotic initiation factors were identified such as eIF4E 1b These interactions changed during maturation, coincident with change in Zar funct ion. Together, these data suggest Zar proteins do have roles as translation regulators and may mediate repression by the TCS. They also suggest mechanistic di fferences between Zar1 and Zar2. The form and content of this abstract are approved. I recommend its publication. Approved: Amanda Charlesworth

PAGE 6

v DEDICATION To Jonathan David Lee Henley I love you son. I am always thinking of you, even when we are not together. M y parents I treasure the pieces of me that come from each of you. Kris, Liz, Ka tie, Randi, Steph and Ben W hat an honor to be a member of a tribe of such varied and amazing individuals. Sean Bergman Y ou know more about poly(A) tails than you ever wanted to, and I am better for sharing this part of the journey with you. Thank you to all of my family and friends, whom I love and who love me.

PAGE 7

vi ACKNOWLEDGEMENTS I would first like to acknowledge and thank my advisor and mentor Amanda Charlesworth who has taught me so much about science and what it means to be a scientist. Yo u have chang ed the way I think. I would also like to acknowledge my other committee members Brad Stith and Aaron Johnson as well as other members of the University of Colorado Denver biology and chemistry departments who have contributed to my education and to t he development of this project. T hank you Kenneth Valles, Courtney Warren and Elana Costanza for proof reading and technical support. Services were provided by S trometry C ore Facility and the University of Colorado Den ver DNA S equencing and Analysis Core. Funding for this project has come from the A merican Cancer Society and from the University of Colorado.

PAGE 8

vii TABLE OF CONTENTS CHAPTER I. INTRODUCTION ................................ ................................ ............................... 1 Zygote arrest (Zar) proteins and maternal control of development .................. 1 Zar proteins are essential for early vertebrate development ....................... 1 Gametogenesis, meiotic maturation and embryogenesis ............................ 2 Maternal control and post transcriptional regulation ................................ 4 Xenopus laevis as a model for translational regulation ................................ .... 6 Maternal control of early development in Xenopus laevis ......................... 6 Cytoplasmic polyadenylation and translational regulation ......................... 9 Mechanisms of mRNA translational regulation ................................ ............ 10 The ................................ ..... 10 The polyadenylation hexanucleotide and the poly(A) tail ........................ 12 Specificity factors : cis elements, trans factors, combinatorial control ..... 14 Zar proteins bind mRNA and regulate mRNA translation ............................. 21 Zar proteins bin d the TCS of the maternal mRNAs mos and wee1 .......... 21 Both Zar and the TCS repress translation in immature Xenopus oocytes 22 II. HYPOTHES IS ................................ ................................ ................................ ... 23 Zar proteins are bona fide translation factors and mediate the translation regulation effects of the Translational Control Sequence (TCS) in Xenopus laevis oocytes ................................ ................................ ............................... 23 III. METHODS ................................ ................................ ................................ ........ 24 Cloning and plasmid preparation ................................ ................................ .. 24

PAGE 9

viii Oocyte isolation, culture and microin jection ................................ ................. 24 MS2 tethered assay ................................ ................................ ...................... 25 mRNA stabilization assays ................................ ................................ ........... 27 mR NA polyadenylation assays ................................ ................................ ..... 28 Western blot ................................ ................................ ................................ 29 GST affinity purification ................................ ................................ .............. 30 Immunoprecipitation ................................ ................................ .................... 31 IV. EXPERIMENTS AND RESULTS ................................ ................................ ..... 32 Aim 1. Show a larger effect for Zar mediated translation repression. ............ 32 Do Zar fusion proteins repress better when stabilized with inter domain linkers? ................................ ................................ ................................ ... 33 Do MS2 Zar fusion proteins repress bette r when tethered to reporter mRNAs with 4 stem loops rather than 2 stem loops? .............................. 41 Aim 2: Show the mechanism of Zar mediated translational regulation is consistent with that established for the TCS ................................ .................. 44 Do Zar1 and Zar2 require a poly(A) tail to repress translation? ............... 46 Do Zar1 and Zar2 require the process of polyadenylatio n to activate translation during maturation? ................................ ................................ 50 Aim 3: Show that Zar proteins interact with other translation factors in the oocyte ................................ ................................ ................................ .......... 54 What proteins do Zar1 and Zar2 interact with in Xenopus oocytes? ......... 54 Do Zar1 and Zar2 interactions with translation factors change during maturation? ................................ ................................ ............................ 60

PAGE 10

ix V. DISCUSSION ................................ ................................ ................................ .... 67 Major findings ................................ ................................ .............................. 67 Redesign of MS2 Zar fusion proteins improved stability and functi on .... 68 Zar1 required a poly(A) tail to repress translation; Zar2 did not .............. 69 Translation regulation by Zar did not involve polyade nylation ................ 70 Zar proteins interacted with components of translational regulation complexes ................................ ................................ .............................. 71 Proposed Models of Zar1 and Zar2 m ediated translation regulation ............ 72 Proposed models of Zar2 mediated translational regulation .................... 72 Proposed models of Zar1 mediated translational regulation .................... 73 REFERENCES ................................ ................................ ................................ .. 74 APPENDIX: Specific Plasmid Construction ................................ ....................... 83

PAGE 11

x LIST OF FIGURES FIGURE 1. Representative cartoon of early Xenopus laevis development. ................................ ...... 7 2. Cartoon of the closed loop model of translation in itiation. ................................ ......... 11 3. Models of CPEB mediated translational regulation. ................................ ................... 17 ................................ ............... 19 5. Representative cartoon of Zar1 and Zar2 protein sequences. ................................ ...... 21 6. Cartoon of MS2 fusion proteins and reporter mRNAs used in tether ed assay. ............ 26 7. Schematic of GST affinity purification assay. ................................ ............................ 30 8. Cartoon of MS2 fusion proteins: old and new constructs. ................................ ........... 34 9. Translation repression by old versus new MS2 Zar fusion proteins. ........................... 37 10. Dose response of translation repression by MS2 Zar. ................................ ............... 38 11. Translation regulation by Zar during maturation. ................................ ..................... 39 12. Amount of MS2 Zar protein remaining after maturation. ................................ ......... 40 13. Effect of extra stem loops on Zar mediated translational regulation. ........................ 43 14. Cartoon of luciferase reporter mRNAs of various poly(A) t ail lengths. ..................... 45 15. Zar mediated repression on reporters varying poly(A) tail length. ............................ 48 16. The effect of polyadenylation on Za r mediated translational regulation. ................... 53 17. GST Zar1 protein interactions in immature oocytes. ................................ ................ 55 18. GST Zar1 and GST Zar2 affi nity purification during maturation. ............................ 59 19. Cartoon of select translation initiation factors. ................................ ......................... 61 20. Zar protein interactions wi th translation initiation factors. ................................ ........ 63 21. Proposed models of Zar2 mediated translation regulation. ................................ ....... 72

PAGE 12

xi LIST OF TABLES TABLE 1. Mass spectrometry results from GST Zar1 affinity purification. ................................ 58

PAGE 13

xii LIST OF ABBREVIATIONS 4E Initiation factor 4E (eIF4E) 4G Initiation factor 4G (eIF4G) 4E BP eIF4E binding protein 4E T eIF4 E transporter protein AMP Adenosine monophosphate AP Alanine/proline rich amino acid linker AV Animal vegetal cAMP Cyclic AMP (adenosine monophosphate) cdc Cell division cycle protein cdk Cyclin dependent kinase Co IP Co immunoprecipitation C PE Cytoplasmic polyadenylation element CPEB Cytoplasmic polyadenylation element binding CPSF Cleavage and polyadenylation specificity factor DAZL Deleted in azoospermia like DNA Deoxy ribonucleic acid eIF Eukaryotic translation initiation fact or EK Glutamate/lysine rich amino acid linker ePAB Embryonic poly(A) binding protein Erk Extracellular signal regulated kinase fluc Firefly luciferase G2 phase 2 n d gap phase (cell cycle) Gdf9 Growth and differentiation factor 9 Gld 2 Germline deficient 2 GS Glycine/serine rich amino acid linker GST Glutathione S transferase GVBD Germinal vesicle breakdown hex Polyadenylation hexanucleotide IgG Immunoglobulin G LDS Lithium dodecyl sulfate m7 GTP MA PK Mitogen activated protein kinase Mater Maternal antigen that embryos require MBE Musashi binding element MBT M id blastula transition MEK MAPK/Erk kinase MPF Maturation promoting factor (M phase promoting factor) M phase Mitotic phase (cell cycle) miRNA M icro RNA (ribonucleic acid) mRNA M essen ger RNA mRNP Message ribonucleoprotein complex MZ1 MS2 Zar1 MZ2 MS2 Zar2 Npm2 Nucleoplasmin protein 2

PAGE 14

xiii PABP Poly(A) binding protein Paip Poly(A) bind protein interacting protein PAP Po ly(A) polymerase PARN Poly(A) specific ribonuclease PAGE Polya crylimide gel electrophoresis Pat1 Topoisomerase II associated protein 1 PBE Pumilio binding element PCR Polymerase chain reaction PKA Protein kinase A rluc Renilla luciferase RNA Ribonucleic acid RT PCR Reverse transcription PCR (polymerase chain reaction) SDS Sodium dodecyl sulfate siRNA S hort interfering RNA SL Stem loops S phase Synthesis phase (cell cycle) TCS Translation al control sequence UTR Untranslated regi on Vera vgLE binding and ER association protein A Xp54 Xenopus protein 54 Xpat1a Xenopus protein Pat1 homolog a Zar Zygote arrest ZGA Zygotic genome activation

PAGE 15

1 CHAPTER I INTRODUCTION Zygote arrest (Zar) proteins a nd maternal control of development Zar proteins are essential for early vertebrate development The recently described Zygote arrest (Zar) family of proteins comprises Zar1 and Zar2 that can be distinguished from each other based on sequence homology and g ene context (Yamamoto et al., 2013) Zar protein sequences have been found only in vertebrate species, and are evolutionaril y conserved in all vertebrate species where sequence data is available (Wu et al., 2003b) Zar1 and Zar2 are germ cell specific: mRNA and protein products are expressed primarily in ovaries, oocytes and eggs in mammals ( (Uzbekov a et al., 2006) ; (Brevini et al., 2004) ), birds ( (Elis et al., 2008) ; (Michailidis et al., 2010) ) and fish (Bobe et al., 2008) Female Zar1 knockout mice are grossly normal, but infertile due to a complete block to embryogenesis (Wu et al., 2003a) No Zar2 knockouts or Zar1/Zar2 double knockouts have been described, but w hen dominant negative strategies have been used to disr upt Zar2 function, embryogenesis is also blocked (Hu et al., 2010) Zar protein sequences are found in all vertebrates examined and share striking features of homology including twelve cysteines whose location and spacing are completely invariant across all identified Zar sequences ( (Sangiorgio et al., 2008) ; (Yamamoto et al., 2013) ) Bec ause Zar proteins share high levels of sequence homology in vertebrate animals it is likely that their function is conserved. While there are fundamental differences between amphibian and mammalian oogenesis, it is tempting to speculate that Zar proteins are essential for early development in vertebrates in general.

PAGE 16

2 Zar proteins have been identified as markers of germ cells ( (Pennetier et al., 2004) ; (Nath et al., 2013) ) a nd of developmental competence ( (Chen et al., 2014) ; (Summers et al., 2014) ) in sever al livestock animals suggesting they play a role in early development in these species as well No Zar protein homologs have been identified in invertebrates to date, including in non vertebrate chordates. Although a couple of sequences have been ident ified in the cephalochordate amphioxus and in the tunicate Ciona that show limited homology to Zar protein sequences (NCBI accession numbers XP_002607077.1 and XP_002130560 .1 ), neither share all of the features characteristic of Zar sequences. Gametogenes is meio tic maturation and embryogenesis G ametogenesis involve s the d ifferentiation and growth of a n immature germ cell followed by 1 2 rounds of meiotic division to yield functional gamete cells Spermatocytes and oocytes have distinct morphologies and p lay very different roles in fertilization and so their mechanisms of gametogenesis are quite different from one another. The primary role of the sperm atocyte is to migrate to the egg and fuse with the plasma membrane, delivering genetic information and triggering fertilization events in the egg. The mature sperm at ocyte needs to be able to produce a large amount of energy to power movement and needs modifications to protect and deliver its genetic cargo, but does not need to maintain metabolism for an ex tended duration of time. The mature spermatocyte contains : a single large, coiled mitochondria to power to the flagella, an acrosome involved in plasma membrane fusion, a nucleus crystallized with protamines

PAGE 17

3 and very little cytoplasm. Meiosis is symmetri c and two rounds of meiotic division pro duce four functional spermatozoa Th e egg (mature oocyte) does not have the same motility requirements as the sperm but must maintain its metabolism beyond fertilization and provide the material and information nece ssary for the early stages of embryogenesis. The oocyte cytoplasm is large and complex compared to that of the sperm, with yolk proteins, a complete set of organelles to maintain the metabolism of the large cell and factors required for the regulation of fertilization and early embryogenesis events The immature oocyte arrests in the early stages of meiosis I until the appropriate cue, and then must reenter the cell cycle to become a fertilizable gamete in a process known as meiotic maturation. Meiosis o ccurs asymmetrically producing one large functional egg cell with cytoplasm and t wo smaller polar bodies without cytoplasm that are discarded. In most vertebrates the mature oocyte (egg) is arrested in meiosis II until fertilization by the sperm, while in invertebrates, sperm entry occurs at other points during meiotic maturation, such as the primary oocyte stage or afte r the comple tion of meiosis II (Tadros and Lipshitz, 2009) Fertilization is the fusion of the sperm and the egg plasma membranes and nuclei, resulting in the combination of the genetic material developmental pro gram. Upon fertilization the egg completes meiosis II and the one cell embryo undergoes the first mitotic division. In the frog, t he one cell embryo, or zygote, enters several rounds of synchronous cleavage (mitotic divisions and cytokin esis) to become an embryo composed of a ball of cells called a blas tula. When the embryo reaches a specific size and cell number (which varies by species) the cell cycles become asynchronous and cell movement called

PAGE 18

4 gastrulation, occurs (Masui and Wang, 1998) This process results in the formation of the gut and tissues within the embryo which then u ndergoes neuru lati on and later stage s of embryogenesis. Maternal control and post transcriptional regulation DNA transcription is largely inactivated during meiosis and does not resume until after the two haploid pronuclei fuse followed by the activation of the n ew diploid genome of the zygote. At this point, transcription increases and genes from both maternal and paternal chromosomes are transcribed in a regulated fashion. Until zygotic genome activation ( ZGA ) regulation of gene express ion is under maternal control and mediated by factors present in the oocyte before in a dormant state until the time t hat their activities are required (Farley and Ryder, 2008) These factors include maternal mRNAs that are transcribed from the genome of the vertebrate oocyte well before fertilization and are post transcriptionally regulated throughout meiosis and fertilization. In some animals, particularly invertebrates, there is significant maternal contribution from neighboring somatic cells. Maternal mRNAs regulate cell cycle and patterning in the absence of transcription, until the zygotic genome is activated. Early embr yogenesis, including early cleavages and the actual activation of the zygotic genome, is under maternal control, as there are primarily maternal transcripts present at this time The timing, rate and localization of transcript expression is regulated by o ther factors in the cell, such as protein translation factors (Colegrove Otero et al., 2005) An example of translational regulation under mater nal control comes from anterior posterio r axis formation in Drosophila embryos, which depends on the initial

PAGE 19

5 polarity of the oocyte that is established well before fertilization. Proper posterior and anterior development require the transcription factors Caudal and H unchback respectively. Whil e the caudal and hunchback mRNAs are generally distributed about the oocyte, their protein products are expressed in gradients across the cell because of localization of the maternal translat ion factors, bicoid and nanos. Establishment of these gradients in the oocyte is essential for proper axis formation in the transcriptionally active embryo (Lasko, 2011) It should be noted that bicoid also fun ctions as a transcription factor in Drosphila embryogenesis. Maternal control of development has been described in vertebrate species as well, such as in zebrafish ( (Bontems et al., 2009) ; (Lindeman and Pelegri, 2010) ) and in mouse ( (Minami et al., 2007) ; (Kim and Lee, 2014) ). The zar1 gene was first identified in a subtractive hybridization screen for maternal effect genes using Gdf9 ( growth and differentiation factor 9) knockout mice (Wu et al., 2003a) F olliculogenesis is disrupted in these mice so that maternal effect gene products accumulate above wild type levels (Carabatsos et al., 1998) The zar1 gene and several other mammalian maternal effect genes showed altered expression in this Gdf9 knockout (Matzuk, 2000) When Zar1 was k nocked out, most of the zygotes arrested before the first cell division in either S or G2 phase and none of the embryos successfully progressed beyond the 2 cell stage (Wu et al., 2003a) In another study, 1 cell mouse embryos were injected with a dominant negative form of Zar2, a similar phenotype was observed, and the cells arrest within one cleavage (Hu et al., 2010) In both of these cases the zy gotic genome wa s not a ctivated. Therefore Zar is regulating maternal effects.

PAGE 20

6 Xenopus laevis as a model for translational regulation Xenopus laevis the African clawed frog, is a common model organism for studying early development. The frogs are available in large numbers a nd are easy to rear in the laboratory and produce large numbers of oocytes and eggs that can be collected year round. The mechanisms that regulate the involved processes are highly aking them easy to manipulate. The oocytes are about a millimeter in diameter and a microliter in volume so RNA and protein can be easily injected into the oocytes and large biochemical samples can be collected. Due to its pseudotetraploid nature, Xenopu s is not a good model system for studying genetics because most genes have a redundant copy that makes knockouts and transgenics difficult. However it is a good model for studying the cell cycle and early vertebrate developmen t from a biochemical perspec tive. Maternal control of early development in Xenopus laevis D uring early oogenesis, maternal mRNAs, and the proteins that regulate the m, are synthesized in the growing oocyte, while transcription is active Xenopus oogenesis is divided into six stages i n the Dumont classification system represented by R oman numerals I VI (Dumont, 1972) At all stages, the immature Xenopus oocyte is arrested in prophase of meiosis I. A progesterone hormone signal induces the immature oocyte to reenter the cell cycle to become a mature oocyte, capable of fertilization ( fig 1 ) Immature Xenopus oocytes are darkly pigmented in the hemisphere of the animal pole and more lightly colored in the vegetal hemisphere. As the oocyte reaches metaphase I, the nucleus, or germinal vesicle, bre aks down (referred to as GVBD) and t t here is a redistribution of c ortical granules, resulting in the formation of a white spot at

PAGE 21

7 the animal pole. This white spot is a convenient marker for meiotic progression and events that occur during meiotic maturation are often grouped as to whether th ey occur before (early) or after (late) GVBD. DNA transcription is largely resume until the zygotic genome is activated in the blastula stage embryo and so Xenopus gene expression is under maternal control du ring this period. Zygotic genome activation Xenopus until after 12 rounds of mitosis, making this an ideal system for studying the role of translational regulation during development. Figure 1 Representative ca rtoon of early Xenopus laevis development. The images show representative anatomy of the oocyte and embryo at different stages of oogenesis, meiosis and embryogenesis. Anatomical stages are labeled below (Roman numerals, Du mont stages of oogenesis; GVBD, germinal vesicle breakdown) and relevant meiotic stages are labeled immediately above images. Major signaling events are marked by red arrows (MBT, mid blastula transition) and dominant form of gene expression is indicate d by upper arrow (light yellow tr anscription; dark green translation). Upon recognition of a progesterone signal, there is a decrease in cyclic adenosine monophosphate (cAMP) levels and protein kinase A (PKA) signaling that ultimately results in the activation of m aturation promoting fa ctor (MPF) (Ferrell, 1999) MPF is compo sed in part of cyclin B and cdk1 (cyclin dependent kinase 1) and is responsible for both the resumption of meiosis I and t he arrest of the cell cycle at metaphase II (Masui,

PAGE 22

8 2001) This activation of MPF is brought about in part through a mitogen activated protein kinase ( MAPK ) cascade (Maller et al., 2002) The mos mRNA is translationally activated so that the Mos protein which is a MAP kinase kinase kinase, is synthesized initiating a phosphorylation cascade that leads to the activation of cyclin B /cdk1 an d promotion of the cell cycle (Tunquist and Maller, 2003) Mos mRNA is translationally activated early during maturation, before GVBD (Charlesworth et al., 2002) Mos activity is further enhanced and stabilized by a MAPK dependent positive feedback loop ( (Howard et al., 1999) ; (Nishizawa et al., 1992) ) Mos activity is also required after GVBD to maintain the metaphase II cell cycle arrest ( (Yamamoto et al., 2005) ; (Dupr et al., 2002) ). Another example of a maternal mRNA that plays an impor tant role in the cell cycle is w ee1. Wee1 protein inhibits cdk1 activity and extends the length of the first mitotic M ph ase to allow for pronuclear fusion and completion of fertilization in the first cycle (Murakami et al., 1999) Wee1 also plays an important role in gastrulation when the cell cycles of th e embryo become asynchronous and mass cell movements begin, extending the cell cycles of some ce lls so they divide slower than their neighbors leading to the necessary contortion of tissues (Murakami et al., 2004) Wee1 is translationally activated late during meiotic maturation, after GVBD, but before fertilization (Charlesworth et al., 2000) The maternal proteins Mos and Wee1 play important roles in cell cycle regulation acros s met azoa in vertebrates such as Xenopus (Murakam i et al., 2004) and mouse ( (Zhao et al., 1991) ; (Tominaga et al., 2006) ), as well as in invertebrates ( (Amiel et al., 2009) ; (Stumpff et al., 2005) ).

PAGE 23

9 Cytoplasmic polyadenylation and translation al regulatio n Because DNA transcription is largely shut down during meiosis a nd early embryogenesis in Xenopus proteins required to regulate developmental events are translated from pre existing maternal mRNAs that have been stored in a dormant state in the egg. It has been noted that, at least in early development, the level of translation activation from an mRNA is positively correlated to the length of the poly(A) tail of the mRNA. RNAs with longer poly(A) tails are translated more robustly than RNAs with shorter poly(A) tails. Cytoplasmic polyadenylation is distinct from th e nuclear polyadenylation that occurs just after transcription. Immature mRNAs are polyadenylated in the nucleus, but after processing and export, maternally regulated mRNAs undergo specific deadenylation so that they have a short poly(A) tail and are tra nslated at a low basal level. In many UTR of the mRNA called the cytoplasmic polyadenylation element (CPE). Extension of the poly(A) tail results in translation above ba sal levels. Cytoplasmic polyadenylation plays a prominent role in translati onal regulation across metazoa and h as been well studied and characterized in Xenopus laevis, as it is a good biochemical model. Both mos and w ee1 mRNAs are polyadenylated coinc ident with their translational activation ( (Prasad et al., 2008) ; (Wang et al., 2008) ) While mos is polyadenylated and activated before GVBD, w ee1 is polyadenylated and acti vated after GVBD. In Xenopus t he protein M os contributes to the signaling that brings about GVB D, while the protein Wee1 would inhibit GVBD if translated prior to the completion of meiosis I

PAGE 24

1 0 Mechanisms of mRNA translation al r egulation T he closed loop model of translation initiation Translation consists of three phases: initiation, elongation an d termination. Any of these phases can be regulated to control the rate of protein production, but most mechanisms of translation al regulation involve the disruption or facilitation of the initiation step. Translation initiation consists of recruitment o f the ribosome to the an initiation codon. Recruitment of the ribosome to the mRNA is a multi step process that involves several proteins that cooperatively interact with t he mRNA and the subunits of the ribosome itself (Aitken and Lorsch, 2012) As a result this process of ribosome recruitment offers many possibilities for translation control. It is thought that ribosome recruitment i s facilitated by a circularization of the mRNA brought about by the interactions of a conserved family of eukaryotic translation initiation factors (eIFs) that interact with the poly(A) methyl ated cap (m 7 GTP) idence for this circularization comes from a poly(A) tail are present compared to when either one alone is present (Gallie, 1991) Further evidence comes from electron micrographs of rough endopl asmic reticulum showing a predominance of circular polysomes (Christensen et al., 1986) The circularization is primarily mediated by poly(A) binding prot ein (PABP) and eIFs In the Xenopus oocyte, the predominate isoform of PABP is embryoni c poly(A) binding protein ( ePAB ) ePAB binds to the polya denosine tracts of mRNA and when bound to the poly(A) tail, it has a strong affinity for eIF4G, a scaffolding p rotein that also

PAGE 25

11 binds to the protein eIF4E (fig. 2B) cap of the mRNA. Figure 2 Cartoon of the closed loop model of translation initiation. The commonly accepted view of translation in itiation involves the binding of several canonical translation fac tors to circularize the mRNA. A. The translationally inactive mRNA has a short poly(A) tail (each A represents approximately 5 adenylate residues) and recruits relatively few e PAB m olecules to the mRNA (4e BP, eIF4E binding protein; 4E, eIF4E ; m 7 GTP, methyl guanosine cap; AUG, translation initiation start codon ; stop, stop codon; hexagon, polyadenylation hexanucleotide; A tract, poly(A) tail; ePAB, embryonic poly(A) binding protein). B. Translationally active mRNAs generally have longer poly(A) tails and bind many ePAB molecules recruit ing other translation fac tors (labeled 3, 4A, 4E and 4G) and creating a protein bridge between the the mRNA, facilitating the binding of the 40S ribosome. The interaction between ePAB and eIF4G strengthens the interactions between eIF4G and eIF4E, as well as the interaction between cap of the mRNA This link of RNA:protein and protein:protein interactions results in th ends of the mRNA being brought together in a closed loop (Kahvejian et a l., 2001) In addition to binding ePAB and the cap bind ing protein eIF4E, eIF4G also binds necessary for 40S ribosome loading and may also assist the ribosomal subun it along the mRNA while it scans for the initiation codon. eIF3 interacts directly with subunits of the 40S ribosome and facilitates ribosomal loading. Once the 40S ribosome recognizes the initiation codon, most of the eIFs are released, the 60S ribosoma l subunit joins and translation begins.

PAGE 26

12 One notable example of modulators of translation initiation are the eIF4E bind ing proteins, or 4E BPs (fig. 2A ) which bind to eIF4E and compete for binding to eIF4G, thereby preventing the formation of the closed lo op translation initiation complex and preventing translation (de Moor et al., 2005) 4E BPs can be phosphorylated, changing their interactions with eIF4E, and coupling translation regulation to signaling pathways that regulate early develo pment. Disruption of any of these eIFs or their protein interactions can prevent translation and most known translation factors modulate these interactions and the formation of a closed loop structure. Translation activators generally promote this close d loop formation and translation repressors generally inhibit its formation. The polyadenylation hexanucleotide and the poly(A) tail translationally silenced have short poly(A) tails an d are not circularized, while mRNAs that are translationally active have longer poly(A) tails and are circularized. This suggests that the length of the poly(A) tail is correlated to the amount of protein produced from the mRNA and that controlling the si ze of the poly(A) tail is a potential mechanism for regulating translation initiation. There is strong evidence for a correlation between poly(A) tail length and amount of translation repression or activation in Xenopus oocytes (Gray et al., 2000) After an immatu re m RNA is transcribed it is capped and po lyadenylated before being exported from the nucleus as a mature mRNA. In the polyadenylation hexanucleotide (hex), with a consensus sequence of AAUAAA. The cleavage and polyadenylation specificity factor (CPSF) complex binds to t he hex

PAGE 27

13 sequence in the mRNA and contains an endonucl the hex site. CPSF also recruits nuclear poly(A) polymerases (PAPs) to the mRNA and the poly(A) tail is greatly extended to about 250 adenosine residues immediately follo wing the polyadenylation hexanucleotide sequence in a process called nuclear polyadenylation (Fox et al., 1989) This is the approximate length of the poly(A) tail for most mature mRNAs exported from the nucleus. Approxima tely one ePAB molecule can bind per 25 adenosine residues. This means that the l onger the poly(A) tail, the more e PAB molecules can bind the mRN A. It is thought that the larger the number of e PAB molecules bound to the mRNA, the greater the enhancement of eIF interactions and the greater the level of translation. The poly(A) tract o f a typical mRNA exported from the nucleus is long enough to bind several ePAB molecules and the mRNA is translationally active. It is thought that the default state for mRNAs exiting the nucleus is to be translated and mRNAs that are not yet to be transl ated are silenced through deadenylation. This allows for decoupling of the timing of transcription and translation and allows for the coordinated regulation of gene expression in an environment where transcription is not active, such as in the maturing oo cyte. Deadenylation of maternally regulated mRNAs and their transient translation silencing after nuclear export has been observed for mouse oocytes (Huarte et al., 1992) and like in Xenopus (McGrew et al., 1989) translational activation is correlated with polyadenylation Deadenylation of mRNA s to be silenced is brought about by deadenylases, such as the poly(A) specific ribonuclease (PARN) (Copeland and Wormington, 2001) When PARN is bound to the mRNA, the poly(A) tail is shortened to about 30 adenosines or less

PAGE 28

14 in many repressed mRNAs. This means that b etween zero and a couple of ePAB molecules can bind to shortened p oly(A) tail, which is not enough to stimulate translation initiation. When the mRNA is to be translationally activated, the poly(A) tail is extended through interactions between components of the same CPSF complex and the AAUAAA hex sequence that were inv olved in nuclear polyadenylation CPSF binds to the hex sequence and recruits the protein G erm line deficient 2 (Gld 2), which is a poly(A) polymerase that is required for cytoplasmic polyadenylation in oocytes. When the polyadenylation hexanucleotide s equence is mutated CPSF does not bind to the mRNA, polyadenylation does not occur and translation of the mRNA is generally reduced. This suggests that regulation of the poly(A) tail length can regulate the level of translation and that the AAUAAA hex seq uence is necessary for regulation of poly(A) tail length ( (Fox et al., 1989) ; (Paris and Richter, 1990) ). There is also an antagonist relationship between poly(A) polymerases such as Gld 2 and deadenylases such as PARN W hen Gld 2 function is disrupted, the poly(A) ta il remains short and the mRNA is poorly translated and when PARN function is disrupted, the poly(A) ta il remains long and the mRNA is translated more robustly. Modulation of the interacti ons between CPSF and Gld 2 and those between CPSF and PARN is an impor tant mechanism for regulating translation initiation. Specificity factors: cis elements trans factors, combinatorial control The polyadenylation hexanucleotide (AAUAAA) that recruits CPSF to the mRNA is present in all mRNAs except for those of histones b ut the mRNAs that are present in the cell have very different poly(A) tail lengths that are specific to mRNA species. This means that regulation of the binding of CPSF to the hex sequence cannot

PAGE 29

15 be the only mechanism of regulating the length of the poly(A ) tail, and that there must be other cytoplasmic polyadenylation specificity factors that modulate the activity of CPSF. Several such factors have been identified that change the activity of CPSF, ePAB and eIFs to regulate the levels of polyadenylation an d translational activation of maternal mRNAs in the developing oocyte. These translation factors will be grouped into two categories for the purpose of this discussion: cis elements and trans factors. Sequences within the mRNA itself, such as the AAUAAA hex sequence that regulate translation are cis elements (as they are in the same molecule that is being regulated) that recruit other molecules that can regulate translation, such as proteins or small RNAs known as trans factors (as they are regulating a different molecule). While micro RNAs (miRNAs) and short interfering RNAs (siRNAs) have been shown to play a role in translation regulation, this discussion will focus exclusively on protein trans factors. The cis element mRNA sequence and its associated trans factor binding protein work together to regulate the translation of the mRNA. Several cis elements have been identified in maternal mRNAs that are 200 nucleotide s of the hex seque nce and therefor e the site of polyadenylation These include: the cytoplasmic polyadenylation element (CPE), th e M usashi binding element (MBE), the Pumilio binding element (PB E) and the translational control sequence (TCS) The CPE recruits CPE bind ing protein (CPEB), the MBE recruits Mus ashi, the PBE recruits Pumilio and the TCS recruits Zar.

PAGE 30

16 CPEB: the first and only specificity factor The CPE and its RNA binding protein CPEB are the most well characterized of the cytoplasmic polyadenylation specific ity factors. For a long time, it was thought that CPEB was the only translation factor that played a role in cytoplasmic polyadenylation in oocytes. While this is no longer considered to be true, CPEB remains the most well understood specificity factor i n oocytes and is involved in many mechanisms of translation regulation. CPEB binds to the CPE sequence (consensus sequence U 4 6 A 1 2 U 1 2 ), which imparts deadenylation and translation repression in immature oocytes and polyadenylation and translation activat ion during meiotic maturation ( (Hake and Richter, 1994) ; (Charlesworth et al., 2004) ) CPEB is involved in the regulation of many mRNAs and has been implicated in several distinct mechanism s of translation repression during early development CPE mediated translation al regulation mechanisms have also been shown to play a role in synaptic plasticity and cellular senescence in somatic cells (Richter, 2007) As a result, there are several proposed models that explain CPEB function, in ter ms of the translation factor ffect on cytoplasmic polyadenylation and disruption of eIF intera ctions. Two of these models include: the Maskin model and the opposing polymerase deadenylase model.

PAGE 31

17 Figure 3 Models of CPEB mediated translational regulatio n. CPEB has been implicated in several mechanisms of translational regulation, including the Maskin model and the opposing deadenylase polymerase model. A. In some translation regulating complexes, CPEB recruits Ma skin, preventing 4E:4G interactions in im mature oocytes. B. In mature oocytes, CPEB is phosphorylated leading to conformational changes in Maskin causing it to release 4E. C. In some complexes CPEB recruits the deadenylase PARN in immature oocytes, preventing elongation of the poly(A) tail and minimizing the number of ePAB molecules recruited to the mRNA. D. In mature oocytes, the Gld 2 that is recruited by CPEB leads to polyadenylation of the mRNA. In the Maskin model (fig. 3A B) CPEB recruits the scaffold protein Symplekin and the eIF4E bind ing protein Maskin to the mRNA. Symplekin acts as a scaffold and allows other translation factors to dock once recruited to CPE containing mRNAs. Maskin acts like a 4E BP and prevents eIF4E from interacting with 4G, thereby preventing translation initiat ion. During meiotic maturation CPEB protein is phosphorylated, which results in a conformational change in Maskin that causes it to release eIF4E and leads to translational ac tivation in the maturing oocyte (Stebbins Boaz et al., 1999)

PAGE 32

18 In the opposing polymerase deadenylase model, both the poly(A) polymerase Gld 2 and the deadenylase PARN are localized to CPE containing mRNAs via CPEB and Symplekin. While Gld 2 functions to extend the poly(A) tail ( (Benoit et al ., 2008) ; (Kwak et al., 2004) ) PARN functions to remove the poly(A) tail (Kim and Richter, 2006) By localizing both enzymes to the same mRNA, they antagonize each other and the poly(A) tail r emains short enough to favor translation repression in immature oocytes. During meiotic maturation though, the phosphorylation and degradation of CPEB result in the ejection of PARN from the complex and Gld 2 is allowed to extend the length of the poly(A) tail and translation is activated (fig. 3 C D) The MBE, PBE and TCS : the introduction of combinatorial control The maternal protein Musashi binds the MBE sequence ((G/A)U 1 3 AGU) and represses translation in immature oocytes. Musashi and the MBE both acti vate cytoplasmic polyadenylation and translation early during meiotic maturation. An important maternal mRNA that is regulated in par t by Musashi and the MBE is mos (Charlesworth et al., 2006) I t should be noted that there appear to be fundamental differences in the way that M os protein expression is regulated among different model organisms, such as bet ween frog and mouse (Prasad et al., 2008) Another maternal trans factor that regulates cytoplasmic polyadenylation and translation is Pumilio protein which binds to the PBE (UGUAU(A/U)UAU) and contribute s to CPE mediated translation repression (Piqu et al., 2008) The Translational Control Sequence (TCS) is a cis of some developmentally important mRNAs. The TCS has been shown to confer repression of mRNA in immature oocytes and ac tivation of translation in mature eggs

PAGE 33

19 (Wang et al., 2008) In isolation, the TCS confers early po lyadenylation to reporter mRNAs and early translational a ctivation that is before GVBD. Figure 4 Cartoo ee1 mRNAs. The diagrams show the location of known cis elements relative to the polyadenyla tion hexanucleotide ( oval, MBE ; circle, CPE ; hexagon, polyadenylation hexanucleotide; square, TCS). C ombination s of these multiple cis elements contribute to the timing and extent of protein translation from these mRNAs. A Th e mos mRNA includes an MBE, a CPE and a TCS. B The w ee1 mRNA includes three CPEs and t wo TCSs. The discovery of multiple trans factors mediating the effects of multiple cis elements in developmentally regulated mRNAs introduced the notion of combinatorial code as an important comp onent of the coordination of timing and extent of mRNA translation (Piqu et al., 2008) In isolation, each cis element has certain properties in terms of the extent to which it represses or activates t ranslation and in terms of when during development its function changes. Most maternal mRNAs have several different cis elements though, with different numbers of each element and different arrangements of elements within the UTRs of the mRNAs (fig. 4). This allows for intricate translation regulation, with built in redundancies and versatility.

PAGE 34

20 Xpat1a, X p54 DAZL and Vera : a growing list of translation regulators In addition to the polyadenylation specificity factors just described, there are many o ther proteins that contribute to the regulation of translation after being recruited to specific mRNAs without changi ng the length of the poly(A) tail The translation regulators Xpat1a and Xp54 work together, along with Rap55, to regulate translation in Xeno pus oocyte s (Mar nef et al., 2009) Xpat1a is a scaffold protein that is recruited to the mRNA by an unknown specificity factor, which binds to and recruits Xp54 to the RNA. Xp54 binds eIF4E in immatur e oocytes (Minshall and Standart, 2004) and it is cap interaction. During meiotic maturation Xpat1a is degraded (Nakamura et al., 2010) which could release Xp54 and allowing cap. The DAZ family of proteins are necessary for germ cell development (Xu et al., 2001) and regulate translation (Smith et al., 2011) DAZL proteins bind mRNA via GUU triplets (Brook et al., 2009) and also bind 4G, strengthening the interaction with 4E and enhancing translation. This has a similar overall effect to that of ePAB. In addition to regulating the extent and timing of mRNA tr anslation, the location of mRNA translation is tightly regulated. Vera is an RNA binding protein that interacts with kinesins and the microtubule network to localize mRNAs to specific locations in the oocyte (Choo et al., 2005) Vera localized mRNAs include vegT (Kwon et al., 2002) and vg1 (Deshler et al., 1997) in Xenopus and cyclinB (Takahashi et al., 2014) in zebrafish. Because Vera localizes mRNAs that it recogniz es, it indirectly localizes all of the proteins and translation machinery (Arthur et al., 2009) This allows the extent of mRNA translation to be reg ulated in both t ime and space.

PAGE 35

21 Zar p roteins bind mRNA and regulate mRNA t ranslation Zar proteins bind the TCS of the maternal mRNAs mos and w ee1 In a yeast 3 mRNA, two pr otein sequences were identified: 1) Musashi, the binding partner of the MB E a cis element found he C terminal of Zar2 (Charlesworth et al., 2006) Zar2 was then shown to immunoprecipitate bot h mos and w ee1 mRNA s in Xenopus oocytes (Charlesworth et al., 2012) Both Zar1 and Zar2 bind the Translation al of the mos and wee1 mRNAs in vitro ( (Charlesworth et al., 2012) ; (Yamamoto et al., 2013) ) Additionally Zar proteins bind to the TCS when inserted in a which does not contain any known tra ns factor binding cis elements Figure 5 Representative cartoon of Zar1 and Zar2 protein sequences. Zar primary protein structure shows areas of high sequence homology (gray sh aded regions) that are proposed to be involved in translation regulation (N terminus) and RNA binding (C terminus) functions and can be generalized to both Zar1 and Zar2 Invariant cysteines thought to be involved in RNA binding a re highlighted in yellow Zar proteins contain two regions that are well conserved across vertebrate species, one region in the N terminal half of the protein and one in the C terminal half of the protein (fig. 5 ) The N terminus of the protein represses translation in the teth ered assay as does full length Zar protein and the C terminus binds mRNA whereas the N terminus does not ( (Charlesworth et al., 2012) ; (Yamamoto et al., 2013) ).

PAGE 36

22 Both Zar and the TCS repress translation in immature Xenopus oocytes Zar proteins have previously been thought to be transcriptional regulators However, Zar proteins bind mRNA, both in vitro and in vivo, and there is evidence that they regulate translation of reporter mRNAs, suggesting they may have a role as translational regulators. When MS2 Zar fusion proteins are tethered to a reporter RNA, N terminals of both Zar1 and Zar2 r educe the amount of protein produced without affecting the stability of the RNA in immature oocytes ( (Charlesworth et al., 2012) ; (Yamamoto et al., 2013) ). They repress translation of reporter mRNAs in immature oocytes and repression is relieved during meiotic maturation. This is consistent with what is known about TCS mediated transla tion regulation, both in terms of repression and a change in function during meiotic maturation. While Zar proteins have been shown to repress translation when tethered to reporter mRNAs in Xenopus oocytes, the effect demonstrated has been small, with Zar proteins repressing translation by only about 30%. While this effect is r eproducibl e and dose responsive, it would provide greater support for Zar being a translation repressor if the effect was more robust, equal in magnitude as that seen for the TCS (ab out 50% repression when compared to a reporter without a TCS) and similar to other translation repressors such as Xp54.

PAGE 37

23 CHAPTER II HYPOTHESI S Zar proteins are bona fide translation factors and mediate the translation regulation effects of the Translation al Control Sequence (TCS) in Xenopus laevis oocytes The following aims were established to test this hypothesis : A im 1. Show a larger effe ct for Zar mediated translation repression. Do Zar fusion proteins repress better when stabilized with inter domain linkers? Do MS2 Zar fusion proteins repress better when tethered to reporter mRNAs with 4 stem loops rather than 2 stem loops? Aim 2: Show the mechanism of Zar mediated translation al regulation is consistent with that established for the TCS Do Zar1 and Zar2 require a poly(A) tail to repress translation? Do Zar1 and Zar2 require the process of polyadenylation to activate translation during maturation? Aim 3: Show that Zar proteins interact with other tr anslation factors in the oocyte. What proteins do Za r1 and Zar2 interact with in Xenopus oocytes? Do Zar1 and Zar2 interactions with translation factors change during maturation?

PAGE 38

24 CHAPTER III METHODS Cloning and plasmid preparation All restriction enzymes, Klenow fragment and Quick Ligase were obtained from New England Biolabs. DNA oligonucleotides were synthesized by Integrated DNA Technologies and Life Technologies Due to the long list of p lasmids constructed for this study, description s of specific pla smids are i n APPENDIX Competent bacteria were prepa red using chemical methods based on (Hanahan et al., 1991) using OneShot Top10 E. coli (Life Technologies) and transformed according to manuf All plasmids we re sequenced to verify integrity using the University of Colorado Denver DNA Sequencing and Analysis Core. For in vitro transcription, all plasmids synthesized in vitro with SP6 mMessage mMachine transcription kit ( Life Technologies ). RNA quality was assessed using gel electrophoresis. Nucleotide and amino acid sequence alignments were performed with MacVector 11.1.2. Oocyte isolation, culture and mic roinjection Adult female Xenopus laevis (Nasco) we re housed and sacri fi ced according to internationally recognized g uidelines and with the approval of the University of Colorado Denver Institutional Animal Care and Use Committee. O ocytes we re isolated and cultured as has been described (Machaca and Haun, 2002). All incubations we re carried out in 0.5X L 15 (MediaTech, Inc.) with penicillin (100 mg/ml) and streptomycin (50 mg/ml). Dumont s tage VI (Dumont, 1972) oocytes we re selected and injected with 23 nl

PAGE 39

25 of the appropriate mRNA using a Drummond NanoInject II microinjector. Oocytes we re incubated to allow for in vivo protein translation. Appropriate samples we re then induced to mature with 2 M progesterone and i mmature samples we re time matched to progester one treated samples (Charlesworth et al., 2012) MS2 tethered assay To test whether Zar proteins regulate translation, a n MS2 tethered assay wa s use d (fig 6 ) The viral protein tag MS2 binds with high affinity to a speci fic primary RNA sequence that forms a secondary stem loop structure (Bardwell and Wickens, 1990) This stem loop s equence was inserted into a reporter mRNA and the MS2 protein was fused to Zar thereby tethering Zar to the reporter mRNA (Coller and Wickens, 2002 ) The reporter mRNA used codes for luciferase, an enzyme that produces light with its substrate directly proportional to the amount of luciferase protein present. The luciferase coding region resides in a globin UTR, which is not known to be transla tionally regulated during development so that any translational regulation effects on the reporter mRNA should be due solely to the tethered MS2 fusion protein. There we r e two luciferase coding mRNAs used in this assay: one codes for firefly luciferase a nd contains the MS2 binding stem loops and the other codes Renilla luciferase and does not contain the stem loops (fig 6B) Because firefly and Renilla luciferases react with disti nct substrates, the mRNAs were co injected into the same oocyte and the amo unts of protein translated from each was measured separately. The stem loop containing, f irefly luciferase coding mRNA was affected by the tethered protein, while the Renilla luciferase coding mRNA, without stem loops, wa s the lo ading control and should n ot have been affected by the fusion protein.

PAGE 40

26 Figure 6 Cartoon of MS2 fusion proteins and reporter mRNAs used in tethered assay. This list of fusion proteins and luciferase coding mRNAs is representative of the constructs used in this study to investigate Zar mediated translation regulation. Modified constructs are described in more detail where appropriate. A MS2 fusion proteins contain an MS2 stem loop binding domain attached to the protein to be tethered t o the reporter mRNA to evaluate translational regulation function (Gray shading, MS2 RNA binding domain; white shading, protein domain to be evaluated for translation regulation activity) MS2 alone is a negative control and MS2 Xp54 is a positive control for translation re pression. B. Luciferase coding mRNAs include either firefly or Renilla luciferase coding region in a globin UTR and either contain MS2 binding stem loops or do not. Firefly luciferase with stem loops tethers MS2 fusion proteins and acts a reporter for translation regulation. Luciferase coding mRNAs without stem loops are used as specificity (firefly luciferase coding) and loading ( Renilla luciferase coding) controls. C. Schematic of the method of the tethered assay. MS2 alone was a negative control fo r translation regulation as it bound to the reporter RNA but should no t have affect ed translation (fig 6A) Xp54, a known translation repressor (Minshall et al., 2009) was a positive control for translation r epression Zar was fused to MS2 to eva luate its effect on translation.

PAGE 41

27 MS2 mRNA was in vitro transcribed from pXen C MS2, MS2 Xp54 was transcribed from pXen MS2 X p54 and Zar MS2 and MS2 Zar mRNAs were transcribed from various pXen Zar MS2 constructs (see APPENDIX), using mMessage mMachine ( Life Technologies ). Oocytes were injected with mRNA for the fusion protein (1 ng 50 ng) and incubated at 18 C for 16 24 hours to allow protein expression (fig 6C) The mRNA coding for the fusion protein was injected rather than the protein itself because mRNA can be rapidly and easily purified and is translated efficiently in Xenopus oocytes. Oocytes were then injected with bot h firefly (100 pg) and Renilla (5 pg) coding mRNA and incubated at 18 24 C to allow for translation of luciferase. In some experiments, half the oocytes were administered progesterone to induce maturation and both immature and mature oocytes were collecte d at the same time, at least 3 hours after progesterone treated oocytes reached GVBD. Each sample consisted of duplicates of 5 oocytes each and were lysed in 250 L passive lysis buffer (Promega). 10 L of cleared lysate was used in the Dual Luciferase R eporter Assay (Promega) according to manufacturer instructions and samples were analyzed using a Synergy HT plate reader (BioT ek). mRNA stabilization assays The relative amount of mRNA was determined by semi quantitative PC R. cDNA was synthesized from 1 5 oocyte equivalent lysate, but from pools of 5 20 oocytes with iScript (Bio Rad) using random hexamer primers and PCR performed with Taq polymerase ( Life Technologies ). The cycle number in which the PCR was in a linear range was determined by running 1 /100 and 1/1000 oocyte equivalent of cDNA from

PAGE 42

28 total RNA and making sure there was a difference in the amount of PCR product between these two amounts of template cDNA. mRNA polyadenylation assays Following the in dicated treatments, total RNA was extracted f rom pools of 5 10 oocytes using TRI reagent ( Life Technologies ) or STAT 60 (TelTest) using manufacturer instructions followed by an additional p henol/chloroform extraction. To obtain high quality RNA, samples were re precipitated with 8 M LiCl (Charlesworth et al., 2000) RNA samples were re suspe nded in water. RNA l iga tion coupled RT PCR is a modified version of the technique described previously ( (Rassa et al., 2000) ; (Cha rlesworth et al., 2004) ). T o assess the length of the poly(A) tail of reporter mRNAs extracted from Xenopus oocytes: 1) a primer (P1) was ligated to the the reverse complement of the ligated primer was used reverse transcribe cDNA and 3) PC R mRNA, including the poly(A) tail. As s horter PCR fragments a re generated from mRNAs with shorter poly(A) tails and longer PCR fragments are generated fr om mRNAs with longer poly(A) tails, changes in the size of the PCR product represented changes in poly(A) tail length. 4 g of total oocyte RNA, from pools of 5 or 6 oocytes, wa s ligated to 0.4 g of P1 anchor primer, in a 10 l reaction using T4 RNA ligas e (New England Biolabs) instructions The whole 10 l RNA ligation reaction wa s used in a 50 l reverse transcription reaction using Superscript III ( Life Technologies ), instructions using 0.4 g l of

PAGE 43

29 this cDNA preparation wa s used in each 50 l PCR using Platinum Taq ( Life Technologies instruction s. PCR wa s performed for 40 cycles, using a 56 C annealing temperature and 1.5 M fi nal concentration of Mg 2+ except where parameters need ed to be changed (Charlesworth et al., 2004) Western blot Pools of 5 10 oocytes we re collected from each sample and lysed in 10 l/oocyte of NP 40 lysis buffer (1% Igepal CA 630, 20 mM Tri s pH 8.0, 137 mM NaCl, 10% glycerol, 2 mM EDTA) supplemented with protease and phosphatase inhibitors (HALT, Pierce). 5 or 10 l of cleared lysate (0.5 or 1 o ocyte equivalent respectively) wa s loade d onto a NuPage Bis Tris polya crylamide gel (Life Technol ogies). Electrophoresis wa s performed using MOPS SDS running buffer ( Life Technologies ), then transferred to 0.45 m Immobilon FL PVDF membrane (Millipore) using an XCell II Blot Module ( Life Technologies ) according to NuPage technic al guide protocol. Membranes we re probed with antibodies against: MS2 (TetraCore), GST (Santa Cruz Biotechnology), Tubulin (Sigma), eIF4G (Cell Signaling Technology ), 4E T (Cell Signaling Technology ), eIF4E (GeneTex ), eIF 4E 1b (Minshall et al., 2007) (kindly provided by Nancy Standart, Univeristy of Cambridge, UK) ePAB (USBiological), Zar 1 (Charlesworth et al., 2012) or Zar2 (Yamamoto et al., 2013) Secondary antibodies were 1:20,000 g oat anti rabbit IR Dye 800 CW and g oat anti mouse I R Dye 680LT (LiCor). M embranes we re imaged on an Odys sey infrared imager and data wa s analyzed usin g Odyssey 2.1 software (LiCor).

PAGE 44

30 GST affinity purification In order to identify interacting proteins, Zar was fused to the tag GST which binds gl utathione sepharose beads, allowing for purification of GST Zar f usion protein along with co purification of proteins that interact with GST Zar from treated oocyte lysates (fig. 7 ) Isolated proteins were separated by electrophoresis and identified by mass spectrometry. Figure 7 Schematic of GST affinity purification assay. Oocytes we re injected with mRNA encoding the GST fusion, incubated to allow for protein expressio n and then lysed. The lysate wa s applied to agarose beads (white circle) covalently attached to glutathione (white hexag ons), which binds GST (grey rectangle). After elution of the beads with reduced glutathione, the GST protein, along with proteins bo und to GST (grey trapezoid), we re isolated from the lysate, then separated by electrophoresi s and visualized by Coomassie o r western blot.

PAGE 45

31 GST alone was used as a negative control and only bands that appeared more strongly in the GST Zar lane compared to GST alone were cut out for identification. GST mRNA was in vitro transcribed using mMessage mMachine ( Life Technologies ) fr om pXen1 (MacNicol et al., 1997) and GST Zar fusion protein coding mRNA was transcribed by pXen GST Zar plasmids described in APPENDIX. Oocytes were injected with 3.7 ng of GST coding mRNA or 20 ng of GST Zar coding mRNA and allowed to incubate at 18 C for 16 24 h ou rs to allow expression of the fusion protein. Oocytes were then collected and lysed in 5 uL per oocyte of either NP 40 lysis buffer or PBS triton and lysates purified on glutathione sepharose beads followed by elution with 10 mM redu ced glutathione. Purified proteins were separated by electrophoresis and visualized by Coomassie or by western blot as appropriate. Immunoprecipitation Pools of 100 oocytes were lysed in 500 l PBS Triton homogenization buffer supplemented with HALT protease and phosphatase inhibitors, RNAse and 5 M DTT. Lysates were clari fi ed by centrifugation 2 x 5 min, 12,000 x g, 4 C. 4 g of Zar2 antibody was added and incubated for 6 h, 4 C. 20 l of protein A DynaBeads ( Life Technologies ) were washed 3 x with PBS triton buffer and added to samples then rotated for 1 h ou r 4 C. Beads were again washed 3 x with PBS Triton buffer and then extracted with 50 uL 1X LDS sample buffer. Purified proteins were separated by electrophoresis and visualized by Coomassie or by western blot as appropriate.

PAGE 46

32 CHAPTER IV EXPERIMENTS AND RESU LTS Aim 1. Show a larger effect for Zar mediated translation repression While both Zar1 and Zar2 have been shown to repress translation of a reporter mRNA using the MS2 tethered assay approach, this repression has not been robust. Repression of up to about 30 40%, relative to when Zar is not tethered to the reporter mRNA (MS2 alone), has been observed for the N terminus of Zar 2 and only about 30% has been observed for the N terminus of Zar1 ( (Charlesworth et al., 2012) ; (Yamamoto et al., 2013) ) This is inconsistent with the level of transla tion repression observed for TCS containing reporter mRNA with 50% repression compared to control in immature Xenopus oocytes (Wang et al., 2008) Zar MS2 fusion proteins have not expressed particularly well and have required 20 100 ng of fusion protein coding mRNA to see significant effects in tethered assay experiments and injection of more than 100 ng was toxic to oocytes (not shown) Poor expression has been particularly p roblematic for full length Zar1 MS2, which has not express ed well in Xenopus oocytes and relat ively littl e fusion protein accumulation was detected at the end of experiments Even whe n the amount of MS2 fusion RNA wa s injected at hi gh levels, very little protein wa s produced (not shown) While the N Zar1 MS2 fusion protein was expressed to levels high enough to show effec t in the tethered assay, there we re similar expression problems with it and much higher levels of RNA had to be injected to get the same level of protein expression and repression effect as that seen for N Zar2 MS2: about 100ng of N Zar1 MS2 wa s required to see the same effect as about 20ng of N Zar2 MS2 (Yamamoto et al., 2013)

PAGE 47

33 If MS2 Zar fusion protein s coul d be made to express better and /or be more stable in oocytes, then perhaps the y could repress translation of reporter mRNAs to a greater extent. Translation repression of about 50% would be considered successful in this aim, as this is the amount of translation repression observed for the TCS and other translation repressors, such as Xp54. T his evidence would provide support for the notion that Zar proteins regulate translation and mediate the effects of the TCS. It would also be technically advan tageous if less MS2 Zar RNA was required to detect the same level of translation repression by minimizing non specific artifacts that may be introduced. In o rder to get better repression changes were made to the design of both the MS2 Zar fusion proteins and the firefly luciferase reporters used in the MS2 tethered ass a y. Do Zar fusion proteins repress better when stabilized with inter domain linkers? Rationale : To improve the function of Zar MS2 fusion proteins in the tethered assay, longer inter domain linkers were inserted between the MS2 and Zar domains with the goal of improving protein stability and preventing unwanted interactions between the domains (Minshall et al., 2010) The old Zar MS2 fusion proteins consist ed of the two domains, a Zar protein or Zar protein truncation and an MS2 tag fused ad jacent to each other with a very short linker region (f ig. 8 A) The Zar domain wa s in the N t erminal position of the fusion and MS2 wa s C terminal. The reasoning behind choosing to not use a linker was to minimize any extra sequence that ma y result in undesired artifact and to take advantage of available restriction site s The Zar domain was fus ed N terminal of MS2 because the N terminal of Zar was shown to be the translational domain (Charlesworth et al., 2012) and this arrangement allows the N terminus of Zar to be unhindered, with hope

PAGE 48

34 of promot ing a mo re natural folding of the domain. However, observations from other labs suggest that at least in some cases the N terminal MS2 tagged fusion protein s were more stable than the C terminal tagged fusion s So it was decided to reverse the orientation of the fusions as another measure to increase protein stability. Figure 8 Cartoon of MS2 fusion proteins: old and new constructs. Longer inter domain linkers were added between MS2 and Zar. A. The MS2 fusion proteins that were used to show moderate repression in previous publications include d very short linker s between MS2 RNA binding domain s and Zar protein domain s Full length Zar and N Zar we re arranged N terminal of MS2 so that the N terminus of Zar wa s free, as would be in the endogen ous protein. B. In the new MS2 fusion protein con structs, a longer linker was inserted and all Zar protein domains, f ull length or truncated, were placed C termina l of MS2. Three linkers were evalua ted, one flexible The amino acid sequenc e of the linker is labeled. Linkers are small amino acid sequences inserted in between domains of a fusion protein to provide physical space between the domains and have been shown to stabi lize and to improve the performance of some fusion proteins (Lee et al., 2013) The use of inter domain linkers to improve fusion protein function comes from observations that many natural multi domain proteins contain linkers of characteristic lengths and chemical

PAGE 49

35 properties that are believed to co ntribute to effective functioning of flanking domains (George and Heringa, 2002) Additionally, the use of synthetic linkers similar to those found in endogenous proteins have been used to improve stability and function of fusion proteins in a variety of experimental systems (Chen et al., 2013) While there is evide nce that particular linker sequences have improved the function of specific fusion proteins, this does not mean that all linkers will improve Zar MS2 fusions and the appropriateness of a linker must ultimately be empirically determined. The new MS2 Zar f usion proteins contained one of three inter domain linkers: a flexible glycine/serine rich linker (GS), a rigid alanine/proline linker (AP) or a rigid gluatamate/ lys ine rich linker (EK) (fig. 8B). Each linker consisted of a repe ating amino acid sequence, 17 amino acids in length, so that each linker fusion wa s the same size as its counterparts. The first linker chosen was A(GGGGS) 3 A and forms a flexible linker with little secondary structure due to the small hydrophilic residues. The second linker was A( EAAAK) 3 A and due to the hydrophil ic interactions between the acidic glutamate residues and the basic lysine residues, forms a rigid alpha helix that provides separation of the domains. The final linker chosen wa s (AP) 8 A, also a rigid linker that provides fixed separation of domains, but forms a rigid structure due to the molecular geometry of the proline residue. The three linkers are her e respectively ( example : MS2 AP Zar1 ) It should be noted that the orientati on of the domains with respect to each other was reversed. In the new constructs the MS2 domain is N terminal of the Zar domain, whereas in the old constructs, the tag was C terminal of Zar. This orientation is similar to other MS2 fusion controls used i n the study, such as MS2 Xp54. Both the inclusion of

PAGE 50

36 linker and inversion of orientation may contribute to a difference in function of the new constructs compared to the old. In addition to the issue of protein stability in immature oocytes, progesterone dependent protein degradation has contributed to difficulties in probing the mechanism of Zar 1 mediated translation regulation during meiotic maturation using the tethered assay. Endogenous Zar1 levels are stable through maturation while Zar2 levels are r educed dramatically in progesterone matured oocytes compared to immature oocytes. Surprisingly, fusion protein levels for both old Zar1 MS2 and old Zar2 MS2 were reduced in response to progesterone though, meaning the relief of repression observed for Zar 1 MS2 from immature to mature oocytes in the tethered assay could be due solely to the degradation of the fusion protein and not represent a change in function of endogenous Zar1 in the mature Xenopus oocyte. In fact, this is why the observation that Zar1 changes function during maturatio n has not yet been published. Results: MS2 tethered assays were performed with the old constructs (fig. 9A)and the new constructs side by side to see if any or all of the new constructs repress ed transl ation to a greater extent (fig 9 B ). MS2 alone did not repress translation of reporter mRNA compared to no fusion protein injected. MS2 Xp54, a positive control for translation repression, repressed translation by 60% compared to no fusion protein or to MS2 alone. Zar1 MS 2 and NZar1 MS2, the old fusion protein constructs, repressed translation by about 20 30% when 50ng of fusion mRNA was injected. The new MS2 Zar1 fusion proteins with longer inter domain linkers repressed translation by about 50% when the same amount of fusion mRNA was injected. The old Zar2 MS2 and N Zar2 MS2 fusion

PAGE 51

37 proteins repressed translation by about 30%, whereas the newly designed MS2 Zar2 constructs repressed translation by about 40 50%, similar to Xp54. Therefore, n ew MS2 Zar fusion proteins ( with longer linkers ) repressed translation to a greater extent than the old Zar MS2 fusion proteins with a minimal linker region and all new constructs repressed to similar levels, regardless of which linker was used. Figure 9 Translation repression by old versus new MS2 Zar fusion proteins. Increased translation repression was observed from MS2 Zar fusion proteins redesigned with longer inter domain linkers. A. Cartoon of MS2 Zar fusion proteins used in tethered assay. The ne wly constructed fusion proteins (lower three) contain longer inter domain linkers and an N terminal orientation with respect to the MS2 tag. B. Bar chart of relative luciferase activity of reporter mRNAs when tethered to MS2 fusion proteins showing that a ll new MS2 Zar fusion proteins ( with longer inter domain linkers ) repressed translation to a greater extent than old fusion proteins, regardless of which linker (AP, EK or GS) was inserted. Results were normalized to MS2 alone ; MS2 Xp54 wa s a positive con trol for translation repression (n = 3 5). Next, a dose response was performed to determine the optimal conditions for the new fusion proteins (fig 10 ). The AP and EK linkers were ch osen for closer analysis as it was thought that the rigid linkers are m ost likely to encourage functional separation of the MS2 and Zar domains. MS2 GS Zar was not used in subsequent experiments (although there was no apparent difference in function among the various linkers).

PAGE 52

38 Figure 10 Dose res ponse of translation repression by MS2 Zar. Bar chart showing MS2 AP Zar and MS2 EK Zar repressed translation in a dose responsive manner and to similar levels relative to the amount of fusion protein coding mRNA injected (1 ng 50 ng), regardless of whet her the linker was AP (alanine/proline) or EK (glutamate/lysine) or whether the Zar domain was Zar1 or Zar2 (n = 1 5, no error bar indicates n = 1). When 1 3 ng of fusion mRNA was injected, little or no repression was observed for any of the fusion prot eins tested For all four fusion proteins, significant repression was observed when 10 ng, 30 ng or 50 ng was injected. About 20 30% repression was observed when 10 ng was injected, about 30 40% repression was observed when 30 ng was injected and about 4 0 50% repression was observed when 50 ng was injected for all four fusion proteins examined. MS2 AP Zar and MS2 EK Zar repressed translation to similar levels with respect to each other at all doses examined, and so choice of linker did not affect the lev el of repression. Most of the experiments in this study were done with MS2 AP Zar injected at 30 ng of coding mRNA, with 30 ng chosen to get maximal repression while minimizing non specific effects from overexpression.

PAGE 53

39 Figure 11 Translation regulation by Zar during maturation. Bar chart of relative luciferase activity (relarive activity) when mRNA reporters were tethered to MS2 Zar fusion proteins in immature (I, black) and progesterone matured (P, white) oocytes, showing that M S2 AP Zar repressed to a greater extent in immature oocytes than Zar MS2, but progesterone dependent relief of repression was still observed. (n = 1 3, no error bar indicates n = 1). To test whether Zar mediated repression of new constructs changed d uring oocyte maturation, fusion proteins were expressed in oocytes, then reporter s were injected and half the oocytes given progesterone (fig. 11). When Zar1 MS2 was tethered, about 20% translation repression was observed in immature oocytes and about 10% repression observed in mature oocytes. When MS2 Zar2 was tethered, about 30% repression was observed in immature and 10% in mature oocytes. When MS2 AP Zar1 and MS2 AP

PAGE 54

40 Zar2 were tethered, about 40 50% translation repression was observed in immature oocy tes and about 30% repression observed in mature. Although the absolute value of translational activity in mature oocytes was different for Zar2 fusion proteins, the amount of translational activity relative to activity in immature oocytes was similar, ab out a 25 30% increase. These results show that when MS2 AP Zar2 was tethered in progesterone matured oocytes, repression was relieved relative to translation repression in immature oocytes. Figure 12 Amount of MS2 Zar protein remaining after maturation. Lower re presentative MS2 western blot from immature (I) and progesterone matured (P) oocytes, showing new MS2 AP Zar1 fusion protein expressed to higher levels in immature oocytes than old Zar1 MS2 and more fusion protein rema ined after maturation. Upper bar chart showing quantification of the fraction of Zar fusion protein that remained after progesterone treatment compared to immature samples across multiple experiments (n = 3 5). (MS2 AP Zar2 western not representative of all experiments.) To determine why the new proteins were better at repressing translation, the amount of fusion protein expressed was measured (fig. 12, lower). New MS2 AP Zar1

PAGE 55

41 fusion proteins expressed to greater levels than old Zar1 MS2 in immature o ocytes and this effect was observed over several experiments. MS2 AP Zar2 was expressed to only modestly higher levels than Zar2 MS2 in immature oocytes. Both MS2 AP Zar1 and MS2 AP Zar2 fusions produce fewer degradation products than the old Zar2 MS2 c onstructs, suggesting the new fusion proteins are more stable in immature oocytes. To determine whether new MS2 Zar fusion proteins were more stable during maturation, the amount of protein present in mature oocytes was compared to the amount in immature oocytes (fig 12 upper). MS2 AP Zar1 protein levels in mature oocytes were about 90% of the levels observed in immature oocytes over 3 5 experiments, whereas Zar1 MS2 proteins levels in mature oocytes were only about 30% relative to levels in immature oo cytes. Proteins levels for both MS2 AP Zar2 and Zar2 MS2 in mature oocytes were about 60% of levels in immature oocytes. The MS2 AP Zar protein was more stable during maturation than the old Zar1 MS2 and this effect was observed over several expe riments. While a stabilization effect was observed for MS2 AP Zar2 in some experiments, this trend was not consistent across multiple experiments. Do MS2 Zar fusion proteins repress better when tethered to reporter mRNAs with 4 stem loops rather than 2 stem loo ps ? Rationale: The MS2 Zar fusion protein is tethered to the reporter RNA by high affinity interactions between the MS2 domain of the fusion protein and an MS2 binding stem (Coller and Wickens, 2007) There are two tandem copies of the stem loop sequence in the firefly

PAGE 56

42 reporters ( fluc 2SL A 33 C 17 ) that have been used to show repression by N Zar MS2 fusions previously The decision to insert two stem loops rather than one was made because RNA containing two MS2 binding stem loops binds better to MS2 protein coated beads t han RNA conta ining only one stem loop (Bardwell and Wickens, 1990) possibly due to a more stable interaction of the M S2 dimer (Keryer Bibens et al., 2008) The use of multiple MS2 binding stem loops in an MS2 tethered assay has been evaluated by other labs with 3 stem loops inc reasing the effect seen for the MS2 fusion protein evaluated compared to the presence of only one stem loop. The inclusion of 9 stem loops, however, did not further enhance the effect of the fusion proteins used in the assay compared to 3 stem loops (Minshall et al., 2010) As 2 stem loops versus 3 stem loops has not been tested, we were not able to reconcile whether our results were equivalent to their 3 stem loop results. Therefore, firefly luciferase reporters were made with either two stem loops ( fluc 2SL A 30 ) or four stem loops ( fluc 4SL A 30 ) (f ig 13A ). Results: To test whether an increase in the number of MS2 binding stem loops increased M S2 Zar fusion protein mediated translation repression, the levels of translation repression by MS2 Zar1 and MS2 Zar2 on fluc 2SL A 30 and fluc 4SL A 30 were measured and compared to each other (fig. 1 3 B). In immature oocytes, MS2 Zar1 repressed translation of fluc 2SL A 30 (black bars) by about 30 40%, whereas translation of fluc 4SL A 30 (dark grey bars) was repressed by about 20 30%. In immature oocytes MS2 Zar2 repressed translation of fluc 2SL A 30 by about 50% and of fluc 4SL A 30 reporter by about 40%. T herefore, there was no significant increase in the amount of translation repression observed when more stem loops were present.

PAGE 57

43 Figure 13 Effect of extra stem loops on Zar mediated translational regulation A. Cartoon of the 2 s tem loop and 4 stem loop containing reporter mR NAs used in the tethered assay. B. Bar chart of relative luciferase activity when reporters with two MS2 binding stem loops (black, immature; white mature ) or 4 MS2 binding stem loops (dark grey, immature; l ight grey mature ) were tethered to MS2 Zar1 or MS2 Zar2, showing that there was no significant increase in the amount of translation repression observed when extra stem loops were present nor was progesterone dependent change in function affected by extr a stem loops.

PAGE 58

44 To determine whether extra stem loops affected translation regulation by MS2 Zar in mature oocytes, oocytes expressing MS2 Zar were injected with 2 stem loop and 4 stem loops reporters and half were given progesterone. In mature oocytes, M S2 Zar1 repressed translation of fluc 2SL A 30 (white bars) by about 20% but only repressed translation of fluc 4SL A 30 (light grey bars) by about 10% and MS2 Zar2 repressed translation of fluc 2SL A 30 by about 30% and of fluc 4SL A 30 by about 20%. These differences were not considered to be significantly different therefor e the inclusion of extra stem loops did not affect progesterone depend ent relief of Zar mediated repression. Furthermore, n o significant difference s in translational regulation act ivit y were observed when lower doses of the MS2 Zar fusion protein were administered ( not shown) or when the extra stem loops were inserted into other firefly luciferase reporters (not shown). Aim 2: Show the mechanism of Zar mediated translation al regulation is consistent with that established for the TCS If Zar proteins are functional partners of the TCS, then the mechanism of Zar mediated translation re gulation and that of TCS mediated translation regulation should have certain fundamental similarities. If the TCS sequence imparts some characteristic of translation regulation or polyadenylation of the mRNA then the Zar protei n should impart the same characteristic This is because my hypothesis proposes that Zar proteins are the trans factor s that bind t o the TCS and transmit the translation regulation information from the cis element to the translation machinery. Mechanistic features of translation regulation that are commonly investigated are poly(A) tai l, because they play such

PAGE 59

45 prominent roles in most known mechanisms. T here is already experimental information available about the role of the poly(A) tail in TCS mediated translation regulation, so determining the role of the poly(A) tail in Zar mediated regulation provided a way to test whether the two mechanism s are mutually consistent. The TCS does not require a poly(A) tail to repress translation, the TCS imparts cytoplasmic polyadenylation and translational activation in mature oocytes and translatio nal activation is dependent on the polyadenylation of the mRNA. Both the role of the poly(A) tail in translation repression and the process of polyadenylation during maturation were evaluated. Figure 14 Cartoon of luciferase reporter mRNAs of various poly(A) tail lengths. Firefly luciferase coding mRNAs with poly(A) tails of varying lengths were used to assess the requirement of a poly(A) tail for Zar mediated translation repre ssion. All reporters contain an m 7 GTP cap, a fire fly coding region and 2 MS2 binding stem loops. The fluc 2SL A 33 C 17 reporter contains 33 adenylate residues followed by 17 cytidines at The fluc 2SL A 0 construct does not contain any adenylate residues and the fluc 2SL A 3o and the fluc 2SL A 1oo+ than 100 adenylate residues respectively. The fluc 2SL A 100+ reporter mRNA was made by adding a poly(A) tail to fluc 2SL A 0 by using poly(A) polymerase in vitro ; so while the first 3 mRNAs are homog en e ous in length, the last tailed reporter is actually a heterogeneous popu lation of different length repor ters.

PAGE 60

46 Several new luciferase reporters with various poly(A) tail structures were constructed to address these issues (fig. 14). The luciferase repor ter that has been used to show repression by Zar proteins in the past ( fluc 2SL A 33 C 17 ) has a poly(A) tail of about what we know about the TCS so new reporters were m ade. The first new reporter ( fluc 2SL A 0 ) ends with no ( A ) s after the polyadenylation hexanucleotide and is most similar to the reporters that were used to evaluate TCS mediated translation repression in the past. This was compared to a report er with 30 t erminal ( A ) s (fluc 2SL A 30 ) and to one wi th greater than 100 ( A ) s (fluc 2SL A 100+ ). These lengths were chosen because it is common for mRNAs that are being translationally repressed to have about 30 ( A ) s and for mRNAs that are translationally active to ha ve greater than 100 ( A ) s. The fluc 2SL A 100+ reporter wa s unique among the luciferase reporters as it was generated using a poly(A) tailing kit to in vitro polymerize a poly(A) tail onto fluc 2SL A 0 resulting in a heterogen e ous population of mRNA reporte rs of various length s with the population centering on a mean length of greater than 100 adenosines. Do Zar1 and Zar2 require a poly ( A ) tail to repress translation? Rationale: The TCS imparts translational repression in immature Xenopus oocytes when pla ced in a poly(A) tail (Wang et al., 2008) If Zar proteins repress translation in immature oocytes through regulation of the TCS, then they too should be able to repress translation without a poly(A) tail.

PAGE 61

47 Deadenylation of mRNA leads to translational repression; recruitment of deadenylases is a common mechanism of repression by RNA bi nding translation repressors While the TCS represses translation of a reporter without a poly(A) tail and therefor e does not require deadenylase activity to repress translation, this does not preclude the possibility that recruitment of a deadenylase is part of the mechanism. Some deadenylases have translational repression activity independent of their deadenylase activity (Cooke et al., 2010) and shortening of tail length may play a role when mRNAs have a tail although this has not been tested for the TCS. Therefore, the role of deadenylation in Zar mediated repression was examined as well. Results: To test whether Zar proteins require a poly(A) tail to repress translation, oocytes expressing MS2 Zar1 or MS2 Zar2 were injected with luciferase reporters with or without a poly(A) tail, fluc 2SL A 0 or fluc 2SL A 30 respecti vely. Immature oocytes were collected and relative luciferase activity compared. When tethered to fluc 2SL A 30 MS2 Zar1 (MZ1) repressed translation by 40% and MS2 Zar2 (MZ2) repressed translation by about 50% when 30 ng of mRNA was injected (fig 15A). When MS2 Zar2 was tethered to fluc 2SL A 0 about 40% translation repression was observed, but when MS2 Zar1 was tethered there was no repression observed, even though MS2 Zar fusion proteins were expressed to similar levels (fig 15A, inset). So while Zar 2 did not require a poly(A) tail to repress translation, surprisingly, repression by Zar1 did require a poly(A) tail.

PAGE 62

48 Figure 15 Zar mediated repression on reporters varying poly(A) tail length Translation repression by Zar1 req uired a poly ( A ) tail, whereas repression by Zar2 did not, and a n extended poly(A) tail prevented repression by both Zar1 and Zar2 A. Bar chart showing translation repression observed in immature oocytes when MS 2 Zar1 (MZ1), MS2 Zar2 (MZ2) or no fusion pr otein (control) was tethered to firefly luciferase reporters with no poly(A) tail (fluc A 0 ), a tail of 30 adenylate residues (fluc A 30 ) or a tail of more than 100 residues (fluc A 100+ ). Inset MS2 western shows fusion proteins were e xpressed to similar le vels. B Translation repression by MS2 Zar was not accompanied by deadenylation of the poly(A) tail. Poly(A) assay (RNA ligation coupled RT PCR) showing that reporter mRNA was not shortened when MS2 Zar was tethered in immature oocytes. Poly(A) tail len gth was measured in immature (I) and progesterone matured (P) control oocytes to demonstrate tail shortening, as fluc reporters with poly(A) tails were deadenylated during maturation: fluc A 0 reporter was not shortened, while both fluc A 30 and fluc A 100+ w ere

PAGE 63

49 To determine whether an extended poly(A) tail could prevent Zar mediated repression, MS2 Zar expressing oocytes were injected with fluc 2SL A 100+ and relative luciferase activity measured. The relative luciferase activity of MS2 Zar1 treated oocytes was equal to that of control oocytes. Although the luciferase activity measured in MS2 Zar2 treated oo cytes appeared to be about 20% greater than that for control oocytes, the variation was great, almost 30%, and so this was not considered a significant difference. So when the poly(A) tail was extended to greater than 100 adenylate residues, there was no translation repression observed for either MS2 Zar1 or MS2 Zar2. No destabilization of reporter mRNAs was observed (not shown). One possible explanatio n for the poly(A) dependency was that Zar1 recruited deadenylases that shortened the length of the poly(A) tail, thereby inhibiting full formation of the translation initiation complex. To determine whether mRNAs translationally repressed by Zar were dead enylated, total mRNA was extracted after tethered assay experiments and the lengths of mRNA reporter poly(A) tails were measured by RNA ligation coupled RT PCR (fig 15 B ) The luciferase reporters, due to the globin UTR background, underwent default dea denylation in response to progesterone in untreated oocytes Thus, comparison of tail length in mature versus immature control oocytes provided a positive control for deadenylation of the reporter. The poly(A) tail of fluc 2SL A 0 was not shortened during meiotic maturation, which was expected as there was n o poly(A) tail to deadenylate and treatment with MS2 Zar1 or MS2 Zar2 had no effect on tail length The fluc 2SL A 30 reporter mRNA was deadenylated during maturation in control oocytes, and so deadeny lation of this reporter was detected with this assay. Treatment with MS2 Zar1 or MS2 Zar2 had no effect on

PAGE 64

50 poly(A) tail length and the size of the PCR product was no smaller than in control oocytes. The fluc 2SL A 100+ reporter also underwent progesterone dependent deadenylation in control oocytes from a larger heterogen e ous product (black arrow) to a more hom o gen e ous product of smaller size. Again, t reatment with MS2 Zar1 or MS2 Zar2 did not result in deadenylation of the reporter mRNA in immature oocyte s Therefore, translation repression by Zar did no t result in deadenylation of any of the tethered reporter mRNAs. Do Zar1 and Zar2 require the process of polyadenylation to activate translation during maturation ? Rationale: The TCS sequence has been sh own to mediate extension of the poly(A) tail during meiotic maturation. The current consensus sequence for the TCS is (A/U)UU(A/G)UCU and when either AUUGUCU or UUUGUCU is inserted into a 17 nucleotide region of the Wee1 mRNA containing both TCS sequences was inserted in to a GST globin reporter RNA with a functional polyadenylation hexanucleotide not only was GST protein accumulation reduced in immature oocytes, but protein accumulation was increased about 2 fold in mature oocytes relative to those expressing GST globin reporters lacking a TCS (Wang et al., 2008) This translational activation was accompanied by polyadenylation of the reporter RNA. Polyadenylation appears to be required for tra nslational activation as when the polyadenylation hexanucleotide was disrupted translational activation was inhibited (Charlesworth et al., 2000) The experiments demonstrate that the TCS imparts polyadenylation of reporters

PAGE 65

51 in mature oocytes and suggest that a poly(A) pol ymerase is involved the mechanism of TCS mediated translation regulation. Relief of Zar mediated repression has been observed on the fluc 2SL A 33 C 17 reporter. However, MS2 Zar dependent luciferase accum ulation in mature oocytes was similar to that seen wh en Zar was not tethered, whereas the TCS showed about a two fold increase compared to its appropriate control. It was thought that the mRNA reporter might not be competent for polyadenylation and this was inhibiting the ability of Zar to activate translat ion. Poly(A) tails have been shown to be deadenylated when 10 cytidines (Varnum and Wormington, 1990) 2SL A 33 C 17 reporter mRNA ends in 33 adenylate residues followed by 17 cytosine residues and a few other non adenylate residues after that about 20 nucleotides total and intermediate of the two tested values so it was not known whether the reporter could be deadenylated or polyadenylated If the fluc 2SL A 30 reporter changed length during maturation while the fluc 2SL A 33 C 17 reporter did not, then comparison of translation of the two reporters in mature oocytes would provide a way to evaluate the effect of the process of polyadenylation on Zar mediated translational activation. Results: To test whether the fluc 2SL A 33 C 17 and fluc 2SL A 30 reporters changed length when tethered to MS2 Zar, RNA wa s extracted from immature and mature oocytes injected with luciferase reporters and tail length measure by RNA ligation coupled RT PCR (fig 14 lower). The size of the tail of fluc 2SL A 33 C 17 was the same in mature oocytes as it was in immature, whereas t he tail of fluc 2SL A 30 was slightly smaller in

PAGE 66

52 mature oocytes due to deadenylation fluc 2SL A 33 C 17 protected the reporter RNA from the default deadenylation observed for reporters with a terminal poly(A) tail, suggesti ng that the reporter is not competent for changes in poly(A) tail length and showing that the fluc 2SL A 30 can change length. To determine whether Zar activates translation to a grea ter extent when the reporter could be polyadenylated, oocytes expressing M S2 Zar1 or MS2 Zar2 were injected with either fluc 2SL A 33 C 17 or fluc 2SL A 30 reporter mRNA and half were given progesterone (fig 16 upper). Relative luciferase activity was evaluated for a correlation between translation activation and polyadenylation. MS2 Zar1 and MS2 Zar2 repressed translation of fluc 2SL A 33 C 17 in immature oocytes by about 30 40% relative to control oocytes. In comparison, t ranslation of the fluc 2SL A 3o reporter was repressed by about 30 40% by MS2 Zar1 and by about 50% by MS2 Zar2 Repression was relieved during maturation; MS2 Zar1 repressed translation by about 20% and MS2 Zar2 repressed by about 30% in mature oocytes relative to control oocytes. There was not a higher level of translation in mature oocytes when MS2 Zar fusions were tethered to fluc 2SL A 3o than when tethered to fluc 2SL A 33 C 17 Therefore, MS2 Zar repressed to similar levels in immature oocytes and the progesterone dependent relief of repression during maturation was similar whether the reporter had a terminal adenylate residues MS2 Zar1 and MS2 Zar2 fusion proteins were expressed to similar levels and the fusion proteins were relatively stable throughout maturation, particularly for MS2 Zar 1. Zar1 mediated translation w as relieved during maturation even when the same amount of

PAGE 67

53 MS2 Zar1 protein wa s present, suggesting that Zar1 protein chan ges function during maturation. Figure 16 The effect of polyadenylation on Zar mediated translation al regu lation Upper, bar chart of relative luciferase activity when MS2 Zar1 (MZ1) or MS2 Zar2 (MZ2) was tethered to luciferase coding mRNA reporters with either a terminal poly(A) tail (fluc A 30 luc A 33 C 17 ). The A 33 C 17 modification was expected to prevent changes to the length of the poly(A) tail, while the fluc A 30 reporter was expected to be competent for deadenylation or polyadenylation Tethering of MS2 Zar to both reporter mRNAs resulted in similar translation regulation: translation was repressed to the same extent in immature oocytes (I, black) and repression was relieved during progesterone induced maturation (P, white) with no greater translation activation observed for fluc A 30 compared to fluc A 33 C 17 Middle MS2 western showing that MS2 Zar1 and MZ2 Zar2 were expressed to similar levels and MS2 Zar fusion protein was stable during maturation. Lower, poly(A) assay (RNA ligation coupled RT PCR) of reporter mRNA showing that the fluc A 3 3 C 17 mRNA tail length wa s the same in immature (I) or progesterone matured (P) oocytes, whereas fluc A 30 reporters were shorter in matured oocytes compared to reporters in immature oocytes. Treatment of MS2 Zar did not result in deadenylation of the repor ter mRNA in immature oocytes compared to control, nor did it protect the mRNA from progesterone induced deadenylation.

PAGE 68

54 To test whether tethering of Zar results in polyadenylation during maturation, th e tail lengths of reporter mRNAs were measured from im mature and mature MS2 Zar expressing oocytes injected with either fluc 2SL A 33 C 17 or fluc 2SL A 30 The tail length of the fluc 2SL A 33 C 17 was the same regardless of MS2 Zar or progesterone treatment. The fluc 2SL A 30 reporter however, underwent deadenyla tion in control oocytes in response to progesterone. The reporter tail length in MS2 Zar1 and MS2 Zar2 treated mature oocytes was the same as in controls. Therefore, t ethering of MS2 Zar1 or MS2 Zar2 to the reporter mRNA did not result in polyadenylation of the mRNA during maturation even when the reporter was competent for changes in tail length, nor did it protect the mRNA from progesterone dependent deadenylation. Aim 3: Show that Zar proteins interact with other translation factors in the oocyte Wh at proteins do Zar1 and Zar2 interact with in Xenopus oocytes? Rationale: Translation factors do not generally operate in isolation within the cell. They interact with other proteins, such as deadenylases and po l y(A) polymerases as previously mentioned and with components of the translation initiation complex and ribosome. As a result, translation factors tend to interact with other trans lation factors and it would be predicted that if Zar proteins function as translation factors in the oocyte, then th ey too would interact with other translation factors. Results: To isolate the proteins that interact with Zar in the Xenopus oocyte, a GST affinity purification was performed using lysate from immature oocytes injected with GST Zar1

PAGE 69

55 Immature oocyte lysat e was chosen as it is at this point in development that Zar proteins repress translation, and the objective was to identify proteins interacting wit h Zar while it is functioning. Proteins were purified on glutathione sepharose and extracted with LDS, the n visualized by Coomassie (fig. 17 ). Figure 17 GST Zar1 protein interactions in immature oocytes. Several bands were present only in immature oocytes and many translation factors were identified in those bands. Lysates from i mmature oocytes that were injected with 3.7 ng of GST or 20 ng of GST Zar1 coding mRNA were purified on glutathione sepharose bea ds, extracted from the beads with LDS separated by electrophoresis and visualized using a Coomassie stain Molecular marker s izes in kDa are indicated to the right. Exogenous GST and GST Zar1 are indicated by white arrow and pieces of the gel that were excised for sequencing are indicated by grey boxes. Proteins that were identified by mass spectrometry and have previously bee n implicated in translation regulation are indicated to the left.

PAGE 70

56 E xogenous GST and GST Zar1 protein s were expressed to similar levels (fig. 17, white arrows). While faint bands were present in the GST alone lane, several darker bands were detected in the GST Zar1 lane that were not present in the negative control lanes, GST alone or uninjected (not shown). This suggested that the proteins in these bands interacted with Zar. To identify the proteins that interacted with Zar, six of the bands or multiplets that were unique to the GST Zar1 lane were excised from the gel and delivered to the mass spectrometry core facility for analysis (fig. 17, grey boxes). A list of protein matches was generated using peptide sequences retrieved from mass spectrometry anal ysis that were BLASTed against a Xenopus protein database. Matching sequences were scored based on percent coverage of the matching sequence by the retrieved peptides and the number of retrieved peptides BLASTing to that sequence. About 45 sequences wer e identified that met the requirements of the core facility to be considered present in the sample. Table 1 lists the identified proteins as they are named in the SWISS Prot database along with predicted molecular weight, grouped by the band in which they were retrieved. Many of the identified proteins are previously characterized and implicated in translational regulation (fig. 17, brackets), providing evidence that Zar proteins function as translation factors in Xenopus oocytes. Symplekin, a scaffold protein, and DHX9, an RNA helicase, play roles in translation initiation and were recovered in the 140 kDa band. A subunit of the translation initiation factor 3 (eIF3c), a component of cytoplasmic RNA granules (Tudor7) and a component of the translation inhibitor interleukin enhancer binding factor 3 (ILF3b) were found in the 120 kDa band, the translation

PAGE 71

57 regulator Xpat1a was recovered in the 90 kDa band and in the 70 kDa band, the oocyte version of poly(A) binding protein (ePAB) and an mRNA localization factor (Vera) were identified. In the 60 kDa band, the well characterized translation regulator CPEB and cytoplasmic granule component Rap55. In the 50 kDa band several tran slation factors were identified: the RNA helicase Xp54, Frgy2, Rap 55 and two com ponents of the translation elongation factor 1 (EF1A3 and EF1GB). In the 20 kDa band several rib osomal proteins were identified: RS20, RL36, RL22 and RL31. These results suggest that Zar proteins interact with translation factors in the cell. There were no deadenylases or poly(A) polymerases identified, consistent with the tethered assay results. There were several other proteins that were not identified as translation factors. Some have roles in mRNA metabolism not directly pertaining to translation r egulation, such as splicing factor 3B (SF3B) found in the 140 kDa band. Others have not been directly implicated in mRNA metabolism, such as the glycolytic enzymes enolase and pyruvate kinase. These proteins may represent non specific interactions or may actually be involved in Zar function, but pursuing the meaning of these interactions is not in line with the aims of this study. The preponderance of proteins wit h known roles in translation regulation provides support for the notion that Zar proteins function as translation regulators in the Xenopus oocyte. Identification of these protein interactions also provided a list of candidates for proteins that may be in volved directly in the mechanism of Zar mediated translational regulation.

PAGE 72

58 Table 1 Mass spectrometry results from GST Zar1 affinity purification. Candidate interacting proteins are arranged by size of the band excised from the ge l (band size). M olecular weight refers to the predicted size of the protein based on amino acid sequence. Protein name refers to the official abbreviated name of the protein in the SwissProt database. Order of proteins within each band size is based on percent coverage of the protein sequence and number of peptides matching to that sequence.

PAGE 73

59 To determine whether Zar protein interactions change during maturation, proteins were purified from immature and matured oocytes expressing GST, GST Zar1 or GST Za r2 (fig. 18). Purified proteins were eluted from the glutathione sepharose beads with 10 mM reduced glutathione, separated by electrophoresis and visualized using a high sensitivity Coomassie based dye. Figure 18 GST Zar1 and GST Zar2 affinity purification during maturation. Lysates from immature (I) or progeste ro ne matured ( P) oocytes were injected with 3.7 ng of GST, 20 ng of GST Zar1 or 20 ng of GST Zar2 were purified on glutathione sepharose beads and then eluted with redu ced glutathione. Weight of molecular markers is indicated at left Exogenous GST and GST Zar fusion proteins are indicated by white arrow s Bands that change d during maturation are indicated by asterisk s (*)

PAGE 74

60 There were some faint differences between G ST Zar1 and GST Zar2 in the 30 70 kDa range but they did not image well and are not indicated. GST Zar1 and GST Zar2 showed very similar banding patterns to each other, suggesting that Zar1 and Zar2 interacted with many of the same proteins. This was not surprising as they have very similar translation repression functions and high levels of amino acid sequence homology, so it seemed likely that they would bind many of the same proteins. In both the GST Zar1 and GST Zar2 lanes, b ands were present in imm ature oocytes around 90 100 kDa (upper asterisk) 60 kDa (middle asterisk) and 50 kDa that were not present in lysates from mature oocytes. These are all bands that were sequenced and all three contain ed translation factors, such as Xpat1a, Xp54 and ePAB suggesting that Zar interacted with translation factor s and these interactions changed during meiotic maturation, coincident with change in Zar function. Do Zar1 and Zar2 interactions with translation factors change during maturation? Rationale: Many tran slation regulation mechanisms disrupt the assembly of the translation initiation complex and th e circularization of the mRNA. The translation initiation complex is composed in part of ePAB and the eukaryotic initiation factors, eIF4E 1a and eIF4G (which a re hereafter referred to simply as 4E 1a and 4G). For example, deadenylation ultimately acts through inhibiting initiation complex formation as the shorter poly(A) recruits fewer ePAB molecules, thereby weakening the effect of 4G stabilization of the 4E 1 cap interaction As Zar proteins did not appear to recruit deadenylases to the mRNA, thereby shortening the poly(A), an alternative option wa s

PAGE 75

61 that Zar proteins prevent assembly through disruption of interactions among translation initiation complex components. Figure 19 Cartoon of select translation initiation factors. Translation is activated through formation of the translation initiation complex. Binding of 4E 1a to the cap (m 7 GTP) enhances ribosome loading and translation initiation. The ePAB:4G interaction strengthen s between 4E The 4e 1b and 4E T proteins play repressive roles in translation regulation by disrupting binding of 4E 1a to Because most cap dependent mechanisms depe nd on 4E 1a binding to the (m 7 GTP ) and to 4G, disruption of these interaction s results in translation al repression. For example, t he translation repressor Maskin binds 4E 1a directly and competes with binding to 4G, thereby sequestering 4E 1a from the rest of the translation initiation complex (fig 3A) Formation of the complex can also be prevente d through disruption of the cap: 4E interaction. The repressor 4E T recruits 4E 1b to the mRNA rather than 4E 1a, and 4E 1b has lower affinity for the c ap, so translation is inhibited (fig. 19 ) D isruption of other interactions, such the ePAB :4G interaction also reduce s translation. T he RNA binding translation repressor Musa shi operates through disrupting this interaction.

PAGE 76

62 Results: To determine whether Zar proteins disrupt formation of the translation initiation complex, Zar2 immunoprecipitation was performed. However, as Zar2 protein levels decline during maturation, which is a confounding factor, GST Zar affinity purification was also performed with lysates from immature a nd progesterone matured oocytes expressing GST, GST Zar1 or GST Zar2. GST Zar fusion proteins were stable during maturation and so this allowed evaluation of protein interactions, independent of changing Zar levels. Proteins purif ied in the presence of RNAse, were eluted with reduced glutathione run next to uninjected total lysate (1/20 input) and blotted with ant ibodies against 4G, 4E T, 4E, and ePAB (fig. 20A ) Za r2 immunoprecipitation was performed in the presence of RNAse wit h Zar2 antibody or without and purified protein s were b lotted with the same panel of antibodies used to evaluate the GST affinity purified proteins (fig. 20B ) Zar1 was not immunoprecipitated as available Zar1 antibodies have not immunoprecipitated (no t shown). Proteins were blotted with antibodies against GST to verify that proteins were expressed and to determine whether they were expressed to similar levels. The fusion proteins were expressed and similar amounts of protein were purified although the re was slightly more GST Zar2 than GST Zar1. The blot also shows that GST Zar1 and GST Zar2 were sta ble during maturation. Membranes were blotted with Zar1 and Zar2 antibodies to determine expression of fusion proteins rel ative to endogenous Zar, and fus ion proteins were overexpressed about 3 5 fold relative to endogenous levels. The Zar2 from the immunoprecipitation showed that Zar2 was purified and there was less Zar2 in mature oocytes.

PAGE 77

63 Figure 20 Zar protein interactions wit h translation initiation factors. GST Zar affinity p urification s and Zar2 immunoprecipitation were performed to identify interactions with tra nslation initiation factors and to determine if these interactions changed during meiotic maturation A. Proteins were purified from lysates expressing GST, GST Zar1 or GST Zar2 on glutathione sepharose beads and eluted with reduced glutathione. Purified proteins were blotted with GST, Zar1, Z ar2, 4G, 4E T and 4E antibodies. Representative data is shown from n = 2. B. Lysates from immature (I) and mature (P) oocytes were immunoprecipitated with an anti Zar2 antibody using protein A dynabeads and LDS extraction. Purified proteins were blotted with Z ar2, 4G, 4E T and 4E antibodies, n = 1. (1/20 input, uninjected ce ll lysate; no ab, sham purified with no antibody; white circle, 4E 1a protein; black circle, 4E 1b protein.) Membranes were blotted with antibodies against 4G, which wa s present in equal amounts in immature and mature oocyte lysates. F aint bands were pres ent after GST purification in negative control lanes as well as in GST Zar lanes. As the signal wa s very weak relative to input and appear ed in negative lanes, it was concluded that the observed signal cou ld be attributed to background. It is worth notin g that a background 4G binding

PAGE 78

64 signal has been observed before when using glutathione sepharose purification (Minshall 2007). When blotted against the Zar2 immunoprecipitated proteins no distinct ba nd was present, although there wa s a blurred band of simi lar size present in b oth immature and mature oocytes, similar in shape and intensity to other background bands. This band was also observed when a non specific antibody ( GST IgG) was used to immunoprecipitate (not shown). Therefore, it was concluded that neither Zar1 nor Zar2 interacted with 4G. Proteins were blotted against 4E T and the antibody recognized two bands, one at about 100 kDa and one at about 130 kDa, repre senting the short and long isoforms of Xenopus 4E T respectively. Both isoforms we re present in immature and mature oocytes and both isoforms we re shifted upward in mature oocytes, likely due to phosphorylation of the protein. Both GST Zar1 and GST Zar2 interacted strongly with both isoforms of 4E T in immature oocytes and the interaction was lessened in mature oocytes, although some interaction was still detectable. Interestingly, there was more of the long isoform of 4E T bound in mature oocytes, relat ive to the short isoform. This pattern was repeated when Zar2 immunoprecipitates were blotted with the antibody. Both isoforms were present in equal amounts in immature oocytes but only the long isoform interaction was detected in mature oocyte lysates. This suggests Zar1 and Zar2 interact strongly with both isoforms of 4E T in immature oocytes and the interaction is removed for the short isoform but only reduced for the long isoform. As only behavior of the short isoform has been described (Minshall et al., 2007) this finding provides new information about the behavior of 4E T in translation al regulation complexes. Antibodies against ePAB showed that ePAB interacted with GST Zar1 and GST Za r2 but not with GST alone and the interaction was reduced in mature oocytes,

PAGE 79

65 although only slightly for Zar1. More ePAB protein w a s detected in the GST Zar2 lane than in GST Zar1. When Zar2 immunoprecipitates were blotted with the ePAB antibody a band w as present in immature oocyte lysates that was fainter in lysates from mature oocytes. This suggests that Zar proteins interact ed with ePAB in immature oocytes and the interaction decreased during maturation. To evaluate the interactions of Zar with 4E 1a and 4E 1b, two antibodies were used, one that recognizes both proteins and one that only recognizes 4E 1b. When the 4E antibody (recognizes both 4E 1a and 4E 1b) was blotted against GST purified proteins, a strong band at 25 kDa was recognized in all l anes that made results uninterpretable (not shown). However, the 4E 1b antibody did not recognize the interfering band as strongly. 4E 1b interacted strongly with GST Zar1 and GST Zar2 in immature oocytes and the interaction was reduced during maturation although some protein was still detectable. To verify the 4E 1b interaction and to determine if Zar2 interacted with 4E 1b, the Zar2 immunoprecipitates were blotted with the 4E antibody that recognizes both proteins. Both 4E 1a (white circle) and 4E 1b (black circle) were detected in total cell lysate, but only 4E 1b was detected in the p urified immunoprecipitate. These data provide no indication that Zar1 or Zar2 interacted with 4E 1a, but suggest that both Zar proteins interacted with 4E 1b and the i nteraction was reduced during maturation. It was surprising that there were not differences in the evaluated protein interactions of Zar1 compared to those of Zar2. To determine whether GST Zar fusion proteins were interacting with their endogenous counte rparts the GST affinity purified proteins were blotted with Zar1 and Zar2 antibodies. Endogenous Zar protein levels were not higher in GST Zar1 or GST Zar2 lanes compared to GST alone. In contrast blots

PAGE 80

66 with Zar2 antibody showed that there was more end ogenous Zar2 protein in both GST Zar1 and GST Zar2 lanes than in GST alone. Surprisingly, endogenous Zar2 was co p urified by both GST Zar1 and GST Zar2, while endogenous Zar1 was not detected at high levels in proteins co purified with GST Zar1 or with GS T Zar2. The implications of this finding have not yet been resolved, but it suggests a difference in how Zar1 and Zar2 interact with translational regulation complexes.

PAGE 81

67 CHAPTER V DISCUSSION The primary purpose of this study was to show that Zar protei ns are translation factors in Xenopus oocytes and that their mechanism of action is consistent with that established for the TCS. Beyond that, this study sought to further characterize the mechanism of Zar mediated translation repression by evaluating the requirement of the poly(A) tail and identifying protein:protein interactions. Major findings New MS2 Zar fusion proteins, redesigned with longer inter domain linkers, repressed translation to a greater extent than the Zar MS2 fusion proteins with minimal linker region s New MS2 Zar fusion proteins expressed more stably in immature oocytes and the new MS2 Zar1 fusion protein s were more stable during maturation, behaving more like endogenous Zar1. The inclusion of extra stem loops did not affect Zar mediat ed translation al regulation. While Zar2 did not require a poly(A) tail to repress translation, repression by Zar1 did require a poly(A) tail. Neither Zar1 nor Zar2 repressed translation on a reporter with an extended poly(A) tail. Neither Zar1 nor Zar2 i mparted cytoplasmic polyadenylation or translational activation of reporter mRNAs during maturation, but relief of repression was indicative of a change in Zar function. Zar proteins interacted with other translation factors including CPEB, Xp54, Xp at1 a, 4E 1b, 4E T and ePAB. I nteractions change d during meiotic maturation coincident with change in Zar function.

PAGE 82

68 Redesign of MS2 Zar fusion proteins improved stability and function The tethered assay ha d been used to show translation repression by the N term inal region of Zar1 and Zar2, but the effect wa s modest (about 30%) and the full length Zar fusion proteins showed even less translation repression. This raised concern as to whether the translation al repression effect was great enough to be physiological ly relevant. The new MS2 Zar fusion proteins repressed translation by about 50%. Xp54 (Minshall and Standart, 2004) xPat1a (Nakamura et al., 2010) and 4E T (Minshall et al., 2007) have all been shown to repress by about 50 60% in Xenopus oocytes CPEB represses by about 30 40% in neurons Therefore, Zar mediated translation repression was on par with other known translation repressors, strengthening the case that Zar proteins functio n as translation al regulators. The more robust fusion proteins were technically advantageous as well, providing a wide enough range of translation repression that mechanistic studies could be performed and Zar1 and Zar2 could be compared to each other ef fectively. Stabilization of MS2 Zar1 in response to progesterone, such that it behaved more like endogenous, allowed for the determination th at Zar1 protein function changed during maturation independent of its absolute protein levels, with less repressio n in mature oocytes despite the same amount of MS2 Zar1 fusion protein. This is similar to CPEB which represses translation in immature oocytes and although protein levels decrease by about 80% during maturation, the remaini ng portion changes behavior: the remaining CPEB is translationally activating in mature oocytes (Thom et al., 2003) Musashi function also changes independent of protein levels. Musashi represses translation in immature oocytes and while protein levels increase

PAGE 83

69 during maturation Musashi protein interactions change and so tr anslation repression is relieved during maturation (Kawahara et al., 2008) In contrast, the level of translation repression by Xpat1a is correlated with the a mount of Xpat1a protein present (Marnef et al., 2010) It is possible t hat the improvement in protein stability and function in the tethered assay was due to the increased length between the domains and/or to the reversed orientation of the two domains. It is notable that the GST Zar fusion proteins were also redesigned with the same inter domain linkers and showed considerable improvement in protein stability (data not shown). The old GST Zar fusion proteins were already constructed with the GST tag at the N terminus of the protein and Zar at the C terminus and so the only change to their structure was the inclusion of the linker. This suggests the increased linker length likely contributed improved stability to both GST Zar fusion proteins and to MS2 Zar fusion proteins. The inclusion of 4 stem loops rather than 2 in the f irefly luciferase reporters did not significantly affect Zar mediated translational regulation. Other labs have shown that 3 stem loops were better than 1 and that 9 were not better than 3 (Collier et al., 2005) and that 2 stem loops were better than 1 (Bardwell and Wickens, 1990) We have used 2 stem loops in the past and this study showed that the inclusion of 4 stem loops was not better than 2, suggesting that 2 or 3 st em loops behave similarly. Zar1 required a poly(A) tail to r epress translation; Zar2 did not While both Zar1 and Zar2 repress ed translation to similar levels on a reporter with a poly(A) tail of 30 adenylate residues only Zar2 was able to repress when the reporter was lacking a poly(A) tail. This result was surpr ising and suggests that Zar1 and Zar2

PAGE 84

70 employ different mechanisms of translational repression. As the TCS does not require a poly(A) tail to repress translation (Wang et al., 2008) these results su ggest t hat t he mechanism of Zar1 mediated translation repression is not consistent with that established for the TCS although that of Zar2 may be The req uirement of a poly(A) tail by Zar1 is similar to that seen for Musashi, which require s a poly(A) tail and disrupt s the ePAB:4G interaction (Kawahara et al., 2008) The lack of a poly(A) tail requirement by Zar2 is similar to that seen for Xp54 (Minshall et al., 2001) which interacts with 4E (Minshall and Standart, 2004) and to that seen for Rap55 (Tanaka et al., 2006) Both Zar1 and Zar2 mediated translation repression was overcome by an extended poly(A) tail. This is similar to Xp54 whose rep ression is overcome by a poly(A) tail (Minshall et al., 2009) and in contrast to Rap55 which represses even in the prese nce of a long poly(A) tail Translation regulation by Zar did not involve polyadenylation As poly(A) tail length and level of translation are correlated in Xenopus oocytes, regulating the length of the tail is a common and effective m ethod of regulating pr otein expression. There was no indication that Zar mediated translation al regulation involved changes to the length of the poly(A) tail, neither deadenylation nor polyadenylation. The poly(A) tail was not shortened by Zar in tethered assay experiments no deadenylases were recovered in GST pull down s (fig. 1 7 and table 1) and work from other members of our lab has shown that the deadenylase PARN did not interact with Zar (not shown). This suggests that deadenylation of the mRNA wa s not required for Zar mediated rep ression and that Zar proteins did not recruit or employ deadenylase activity. The lack of a requirement for deadenylase activity wa s consistent with observations about

PAGE 85

71 the TCS, as the TCS does not require a poly(A) tail to repress translation implying that deadenylase activity is not necessary. Recruitment of deadenylases is part of the mechanisms of CPEB and Pumilio: CPEB recruits PARN (Kim and Richter 2006) and Pumilio recruits the CCR4/Not/Pop deadenylase complex (Chritton and Wickens, 2010) Zar did not appear to employ this type of mechanism. It should be noted t hough, that t he CCR4/Not/Pop complex is an example of a deadenylase that can repress translation even in the absence of its de adenylase activity (Collart and Panasenko, 2012) Therefore, the lack of Zar mediated deadenylase activity does not preclude the possibility that Zar proteins recruit deadenylases to bring about translation repression. Zar proteins did not impart cytoplasmic polyadenylation to the reporter mRNAs in the tethered ass ay. This is in stark contrast to the behavior of the TCS and suggests that Zar proteins were not responsible for TCS mediated cytoplasmic polyadenylation and translational activation in mature oocytes. Perhaps Zar proteins (particularly Zar2) mediate tr anslation repression by the TCS in immature oocytes, while some unidentified factor binds the TCS t o bring about polyadenylati on and translational activation in mature oocytes. This factor could be similar to CPEB or Musashi, which recruit the poly(A) pol ymerase Gld 2 during meiotic maturation, to promote polyadenylation and translational activation of the mRNA (Radford et al., 2008) Zar proteins interacted with components of translation al regulation complexes Zar proteins interacted w ith Xp54, Rap 55, CPEB, and Xpat1a in immature oocytes, all proteins with identified roles in translational regulation. Zar also interacted

PAGE 86

72 with the translation initiation factors 4E 1b, 4E T and ePAB, and these interactions were reduced during maturatio n consistent with relief of Zar mediated repression. Proposed Models of Zar1 and Zar2 mediated translation regulation Taken together, the tethered assay and affinity purification data suggest that Zar proteins are translation regulators that interact wit h the translation initiation complex. Proposed model s of Zar2 mediated translational regulation Figure 21 Proposed models of Zar2 mediated translation regulation. Tethered assay and protein interaction data were used to create m odels of proposed Zar2 mediated translational regulation. A. Zar2 (grey circle) is proposed to repress translation by recruiting 4E T and 4E 1b. B. Zar2 may also disrupt ePAB interactions. C. Release of 4E T, 4E 1b and ePAB would result in relief of re pression and moderate levels of translation. D. Replacement of some other polymerase activity promoting factor onto the TCS may account for polyadenylation a nd higher levels of translational activation. Zar2 mediated translation repression was poly(A) ta il independent and Zar2 protein levels were reduced during m aturation, although about 20% remained. Zar2

PAGE 87

73 interacted strongly with 4E T, 4E 1b and ePAB in immature oocytes and these interactions decreased during maturation. I propose that Zar2 recruits 4 E T and 4E 1b to TCS containing mRNAs, thereby preventing cap:4E 1a:4G interactions and repressing translation (fig 21 A). It is possible that Zar2 also interacts with ePAB to prevent initi ation complex formation (fig 21 B). During meiotic maturation, the interactions would be released and Zar2 would no longer pr event cap:4E 1a:4g binding. R epression would be relieved with moderate levels of translation on mRNAs whe re Zar2 was still bound (fig 21 C). TCS containing mRNAs not bound to Zar 2 would bind some un identified polymerase activity promoting factor leading to cytoplasmic polyadenylation and highe r levels of translation (fig 21 D). Proposed models of Zar1 mediated translational regulation The Zar1 data was inconsistent and so no clear model could be deve loped. However, there are other protein interactions that could be involved. For example, the translation factor Paip (poly(A) binding protein interacting protein) binds to and represses ePAB (Kahvejian et al., 2001) Zar1 could recruit Paip to repress translation, in a poly(A) dependent manner. Alternatively, Zar1 could recruit Xpat1a and Xp54 to recruit translation. As Xpat1a is degraded during maturation (Nakamura et al., 2010) Xp54 would be released, thus relieving repression. This m odel would not explain the poly(A) dependence of Zar1 mediated translation repression, but might address the observed relief of repression during maturation despite Zar1 pr otein levels re maining constant. These models will be the subject of future studies.

PAGE 88

74 REFERENCES Aitken, C.E., Lorsch, J.R., 2012. A mechanistic overview of translation initiation in eukaryotes. Nature structural & molecular biology. Amiel, A., Leclre, L., Robert, L., Chevalier, S., Houliston, E., 2009. Conserved func tions for Mos in eumetazoan oocyte maturation revealed by studies in a cnidarian. Curr Biol 19, 305 311. Arthur, P.K., Claussen, M., Koch, S., Tarbashevich, K., Jahn, O., Pieler, T., 2009. Participation of Xenopus Elr type proteins in vegetal mRNA localiza tion during oogenesis. The Journal of biological chemistry 284, 19982 19992. Bardwell, V.J., Wickens, M., 1990. Purification of RNA and RNA protein complexes by an R17 coat protein affinity method. Nucleic Acids Res 18, 6587 6594. Benoit, P., Papin, C., Kw ak, J.E., Wickens, M., Simonelig, M., 2008. PAP and GLD 2 type poly(A) polymerases are required sequentially in cytoplasmic polyadenylation and oogenesis in Drosophila. Development (Cambridge, England) 135, 1969 1979. Bobe, J., Nguyen, T., Mah, S., Monge t, P., 2008. In silico identification and molecular characterization of genes predominantly expressed in the fish oocyte. BMC Genomics 9, 499. Bontems, F., Stein, A., Marlow, F., Lyautey, J., Gupta, T., Mullins, M.C., Dosch, R., 2009. Bucky ball organizes germ plasm assembly in zebrafish. Curr Biol 19, 414 422. Brevini, T.A., Cillo, F., Colleoni, S., Lazzari, G., Galli, C., Gandolfi, F., 2004. Expression pattern of the maternal factor zygote arrest 1 (Zar1) in bovine tissues, oocytes, and embryos. Mol Repro d Dev 69, 375 380. Brook, M., Smith, J.W., Gray, N.K., 2009. The DAZL and PABP families: RNA binding proteins with interrelated roles in translational control in oocytes. Reproduction 137, 595 617. Carabatsos, M.J., Elvin, J., Matzuk, M.M., Albertini, D.F. 1998. Characterization of oocyte and follicle development in growth differentiation factor 9 deficient mice. Dev Biol 204, 373 384. Charlesworth, A., Cox, L.L., MacNicol, A.M., 2004. Cytoplasmic polyadenylation element (CPE) and CPE binding protein (CPE B) independent mechanisms regulate early class maternal mRNA translational activation in Xenopus oocytes. The Journal of biological chemistry 279, 17650 17659.

PAGE 89

75 Charlesworth, A., Ridge, J.A., King, L.A., MacNicol, M.C., MacNicol, A.M., 2002. A novel regulat ory element determines the timing of Mos mRNA translation during Xenopus oocyte maturation. The EMBO journal 21, 2798 2806. Charlesworth, A., Welk, J., MacNicol, A., 2000. The temporal control of Wee1 mRNA translation during Xenopus oocyte maturation is re gulated by cytoplasmic polyadenylation elements within the 3' untranslated region. Developmental biology 227, 706 719. Charlesworth, A., Wilczynska, A., Thampi, P., Cox, L.L., MacNicol, A.M., 2006. Musashi regulates the temporal order of mRNA translation d uring Xenopus oocyte maturation. The EMBO journal 25, 2792 2801. Charlesworth, A., Yamamoto, T.M., Cook, J.M., Silva, K.D., Kotter, C.V., Carter, G.S., Holt, J.W., Lavender, H.F., MacNicol, A.M., Ying Wang, Y., Wilczynska, A., 2012. Xenopus laevis zygote a rrest 2 (zar2) encodes a zinc finger RNA binding protein that binds to the translational control sequence in the maternal Wee1 mRNA and regulates translation. Dev Biol 369, 177 190. Chen, J.Y., Li, X.X., Xu, Y.K., Wu, H., Zheng, J.J., Yu, X.L., 2014. Devel opmental competence and gene expression of immature oocytes following liquid helium vitrification in bovine. Cryobiology 69, 428 433. Chen, X., Zaro, J.L., Shen, W. C.C., 2013. Fusion protein linkers: property, design and functionality. Advanced drug deliv ery reviews 65, 1357 1369. Choo, S., Heinrich, B., Betley, J.N., Chen, Z., Deshler, J.O., 2005. Evidence for common machinery utilized by the early and late RNA localization pathways in Xenopus oocytes. Developmental biology 278, 103 117. Christensen, A.K. Kahn, L.E., Bourne, C.M., 1986. Circular polysomes predominate on the rough endoplasmic reticulum of somatotropes and mammotropes in the rat anterior pituitary. The American journal of anatomy 178, 1 10. Chritton, J.J., Wickens, M., 2010. Translational r epression by PUF proteins in vitro. RNA (New York, N.Y.) 16, 1217 1225. Colegrove Otero, L.J., Minshall, N., Standart, N., 2005. RNA binding proteins in early development. Crit Rev Biochem Mol Biol 40, 21 73. Collart, M.A., Panasenko, O.O., 2012. The Ccr4 not complex. Gene 492, 42 53.

PAGE 90

76 Coller, J., Wickens, M., 2002. Tethered function assays using 3' untranslated regions. Methods (San Diego, Calif.) 26, 142 150. Coller, J., Wickens, M., 2007. Tethered function assays: an adaptable approach to study RNA regul atory proteins. Methods in enzymology 429, 299 321. Collier, B., Gorgoni, B., Loveridge, C., Cooke, H.J., Gray, N.K., 2005. The DAZL family proteins are PABP binding proteins that regulate translation in germ cells. EMBO J 24, 2656 2666. Cooke, A., Prigge, A., Wickens, M., 2010. Translational repression by deadenylases. J Biol Chem 285, 28506 28513. Copeland, P.R., Wormington, M., 2001. The mechanism and regulation of deadenylation: identification and characterization of Xenopus PARN. RNA 7, 875 886. de Moo r, C.H., Meijer, H., Lissenden, S., 2005. Mechanisms of translational control by the 3' UTR in development and differentiation. Seminars in Cell & Developmental Biology 16, 49 58. Deshler, J.O., Highett, M.I., Schnapp, B.J., 1997. Localization of Xenopus V g1 mRNA by Vera protein and the endoplasmic reticulum. Science (New York, N.Y.) 276, 1128 1131. Dumont, J.N., 1972. Oogenesis in Xenopus laevis (Daudin). I. Stages of oocyte development in laboratory maintained animals. J Morphol 136, 153 179. Dupr, A., J essus, C., Ozon, R., Haccard, O., 2002. Mos is not required for the initiation of meiotic maturation in Xenopus oocytes. EMBO J 21, 4026 4036. Elis, S., Batellier, F., Couty, I., Balzergue, S., Martin Magniette, M.L., Monget, P., Blesbois, E., Govoroun, M. S., 2008. Search for the genes involved in oocyte maturation and early embryo development in the hen. BMC Genomics 9, 110. Farley, B., Ryder, S., 2008. Regulation of maternal mRNAs in early development. Critical reviews in biochemistry and molecular biolog y 43, 135 162. Ferrell, J.E., 1999. Xenopus oocyte maturation: new lessons from a good egg. Bioessays. Fox, C.A., Sheets, M.D., Wickens, M.P., 1989. Poly(A) addition during maturation of frog oocytes: distinct nuclear and cytoplasmic activities and regulat ion by the sequence UUUUUAU. Genes Dev 3, 2151 2162.

PAGE 91

77 Gallie, D.R., 1991. The cap and poly (A) tail function synergistically to regulate mRNA translational efficiency. Genes & Development. George, R.A., Heringa, J., 2002. An analysis of protein domain linke rs: their classification and role in protein folding. Protein Eng 15, 871 879. Gray, N.K., Coller, J.M., Dickson, K.S., Wickens, M., 2000. Multiple portions of poly(A) binding protein stimulate translation in vivo. The EMBO journal 19, 4723 4733. Hake, L.E ., Richter, J.D., 1994. CPEB is a specificity factor that mediates cytoplasmic polyadenylation during Xenopus oocyte maturation. Cell 79, 617 627. Hanahan, D., Jessee, J., Bloom, F.R., 1991. Plasmid transformation of Escherichia coli and other bacteria. Me thods Enzymol 204, 63 113. Howard, E.L., Charlesworth, A., Welk, J., MacNicol, A.M., 1999. The mitogen activated protein kinase signaling pathway stimulates mos mRNA cytoplasmic polyadenylation during Xenopus oocyte maturation. Molecular and cellular biolo gy 19, 1990 1999. Hu, J., Wang, F., Zhu, X., Yuan, Y., Ding, M., Gao, S., 2010. Mouse ZAR1 like (XM_359149) colocalizes with mRNA processing components and its dominant negative mutant caused two cell stage embryoni c arrest. Dev Dyn 239, 407 424. Huarte, J., Stutz, A., O'Connell, M.L., Gubler, P., Belin, D., Darrow, A.L., Strickland, S., Vassalli, J.D., 1992. Transient translational silencing by reversible mRNA deadenylation. Cell 69, 1021 1030. Kahvejian, A., Roy, G., Sonenberg, N., 2001. The mRNA closed loop model: the function of PABP and PABP interacting proteins in mRNA translation. Cold Spring Harb Symp Quant Biol 66, 293 300. Kawahara, H., Imai, T., Imataka, H., Tsujimoto, M., Matsumoto, K., Okano, H., 2008. Neural RNA binding protein Musashi1 inhib its translation initiation by competing with eIF4G for PABP. J Cell Biol 181, 639 653. Keryer Bibens, C., Barreau, C., Osborne, H., 2008. Tethering of proteins to RNAs by bacteriophage proteins. Biology of the cell / under the auspices of the European Cell Biology Organization 100, 125 138. Kim, J.H., Richter, J.D., 2006. Opposing polymerase deadenylase activities regulate cytoplasmic polyadenylation. Mol Cell 24, 173 183.

PAGE 92

78 Kim, K.H., Lee, K.A., 2014. Maternal effect genes: Findings and effects on mouse embr yo development. Clin Exp Reprod Med 41, 47 61. Krieg, P.A., Melton, D.A., 1984. Functional messenger RNAs are produced by SP6 in vitro transcription of cloned cDNAs. Nucleic Acids Res 12, 7057 7070. Kwak, J.E., Wang, L., Ballantyne, S., Kimble, J., Wickens M., 2004. Mammalian GLD 2 homologs are poly(A) polymerases. Proceedings of the National Academy of Sciences of the United States of America 101, 4407 4412. Kwon, S., Abramson, T., Munro, T.P., John, C.M., Khrmann, M., Schnapp, B.J., 2002. UUCAC and ver a dependent localization of VegT RNA in Xenopus oocytes. Curr Biol 12, 558 564. Lasko, P., 2011. Posttranscriptional regulation in Drosophila oocytes and early embryos. Wiley Interdiscip Rev RNA 2, 408 416. Lee, M., Bang, K., Kwon, H., Cho, S., 2013. Enhan ced antibacterial activity of an attacin coleoptericin hybrid protein fused with a helical linker. Mol Biol Rep 40, 3953 3960. Lindeman, R.E., Pelegri, F., 2010. Vertebrate maternal effect genes: Insights into fertilization, early cleavage divisions, and g erm cell determinant localization from studies in the zebrafish. Mol Reprod Dev 77, 299 313. MacNicol, M.C., Pot, D., MacNicol, A.M., 1997. pXen, a utility vector for the expression of GST fusion proteins in Xenopus laevis oocytes and embryos. Gene 196, 25 29. Maller, J.L., Schwab, M.S., Gross, S.D., Taieb, F.E., Roberts, B.T., Tunquist, B.J., 2002. The mechanism of CSF arrest in vertebrate oocytes. Mol Cell Endocrinol 187, 173 178. Marnef, A., Maldonado, M., Bugaut, A., Balasubramanian, S., Kress, M., Weil D., Standart, N., 2010. Distinct functions of maternal and somatic Pat1 protein paralogs. RNA (New York, N.Y.) 16, 2094 2107. Marnef, A., Sommerville, J., Ladomery, M., 2009. RAP55: insights into an evolutionarily conserved protein family. The internatio Masui, Y., 2001. From oocyte maturation to the in vitro cell cycle: the history of discoveries of Maturation Promoting Factor (MPF) and Cytostatic Factor (CSF). Differentiation 69, 1 17. Masui, Y., Wang, P., 1998. Cell cycle transition in early embryonic development of Xenopus laevis. Biol Cell 90, 537 548.

PAGE 93

79 Matzuk, M.M., 2000. Revelations of ovarian follicle biology from gene knockout mice. Mol Cell Endocrinol 163, 61 66. McGrew, L.L., Dworkin Rastl, E., Dworkin, M.B., Richter, J.D., 1989. Poly(A) elongation during Xenopus oocyte maturation is required for translational recruitment and is mediated by a short sequence element. Genes Dev 3, 803 815. Michailidis, G., Argiriou, A., Avdi, M., 2010. Expression of chicken zygote arrest 1 (Zar1) an d Zar1 like genes during sexual maturation and embryogenesis. Vet Res Commun 34, 173 184. Minami, N., Suzuki, T., Tsukamoto, S., 2007. Zygotic gene activation and maternal factors in Minshall, N., Allison, R., Marnef A., Wilczynska, A., Standart, N., 2010. Translational control assessed using the tethered function assay in Xenopus oocytes. Methods 51, 165 169. terminal domain in tra nslational repression, P body localization and assembly. Molecular biology Minshall, N., Reiter, M.H., Weil, D., Standart, N., 2007. CPEB interacts with an ovary specific eIF4E and 4E T in early Xenopus oocytes. The Journal of biological chemistr y 282, 37389 37401. Minshall, N., Standart, N., 2004. The active form of Xp54 RNA helicase in translational repression is an RNA mediated oligomer. Nucleic acids research 32, 1325 1334. Minshall, N., Thom, G., Standart, N., 2001. A conserved role of a DEAD box helicase in mRNA masking. RNA 7, 1728 1742. Murakami, M.S., Copeland, T.D., Vande Woude, G.F., 1999. Mos positively regulates Xe Wee1 to lengthen the first mitotic cell cycle of Xenopus. Genes Dev 13, 620 631. Murakami, M.S., Moody, S.A., Daar, I.O., Morrison, D.K., 2004. Morphogenesis during Xenopus gastrulation requires Wee1 mediated inhibition of cell proliferation. Development 131, 571 580. Nakamura, Y., Tanaka, K.J., Miyauchi, M., Huang, L., Tsujimoto, M., Matsumoto, K., 2010. Translational repres sion by the oocyte specific protein P100 in Xenopus. Developmental biology 344, 272 283.

PAGE 94

80 Nath, A., Sharma, V., Dubey, P.K., Pratheesh, M.D., Gade, N.E., Saikumar, G., Sharma, G.T., 2013. Impact of gonadotropin supplementation on the expression of germ cell marker genes (MATER, ZAR1, GDF9, and BMP15) during in vitro maturation of buffalo (Bubalus bubalis) oocyte. In Vitro Cell Dev Biol Anim 49, 34 41. Nishizawa, M., Okazaki, K., Furuno, N., Watanabe, N., Sagata, N., 1992. The 'second codon rule' and autophos phorylation govern the stability and activity of Mos during the meiotic cell cycle in Xenopus oocytes. EMBO J 11, 2433 2446. Paris, J., Richter, J.D., 1990. Maturation specific polyadenylation and translational control: diversity of cytoplasmic polyadenyla tion elements, influence of poly(A) tail size, and formation of stable polyadenylation complexes. Mol Cell Biol 10, 5634 5645. Pennetier, S., Uzbekova, S., Perreau, C., Papillier, P., Mermillod, P., Dalbis Tran, R., 2004. Spatio temporal expression of the germ cell marker genes MATER, ZAR1, GDF9, BMP15,andVASA in adult bovine tissues, oocytes, and preimplantation embryos. Biol Reprod 71, 1359 1366. Piqu, M., Lpez, J.M., Foissac, S., Guig, R., Mndez, R., 2008. A combinatorial code for CPE mediated trans lational control. Cell 132, 434 448. Prasad, C.K., Mahadevan, M., MacNicol, M.C., MacNicol, A.M., 2008. Mos 3' UTR regulatory differences underlie species specific temporal patterns of Mos mRNA cytoplasmic polyadenylation and translational recruitment duri ng oocyte maturation. Molecular reproduction and development 75, 1258 1268. Radford, H.E., Meijer, H.A., de Moor, C.H., 2008. Translational control by cytoplasmic polyadenylation in Xenopus oocytes. Biochimica et biophysica acta 1779, 217 229. Rassa, J.C., Wilson, G.M., Brewer, G.A., Parks, G.D., 2000. Spacing constraints on reinitiation of paramyxovirus transcription: the gene end U tract acts as a spacer to separate gene end from gene start sites. Virology 274, 438 449. Richter, J.D., 2007. CPEB: a life i n translation. Trends Biochem Sci 32, 279 285. related to zygote arrest 1 (Zar1), Zar1 like, is encoded by a novel gene conserved in the Smith, R.W., Anderson, R.C., Smith, J.W., Brook, M., Richardson, W.A., Gray, N.K., 2011. DAZAP1, an RNA binding protein required for development and spermatogenesis, can regulate mRNA translation. RNA 17, 1282 1295.

PAGE 95

81 Stebbins Boaz, B., Cao, Q., de Moor, C.H., Mendez, R., Richter, J.D., 1999. Maskin is a CPEB associated factor that transiently interacts with elF 4E. Mol Cell 4, 1017 1027. Stumpff, J., Kellogg, D.R., Krohne, K.A., Su, T.T., 2005. Drosophila Wee1 interacts with members of the gammaTURC and is required for proper mitotic spindle morphogenesis and positioning. Curr Biol 15, 1525 1534. Summers, A.F., Pohlmeier, W.E., Sargent, K.M., Cole, B.D., Vinton, R.J., Kurz, S.G., McFee, R.M., Cushman, R.A., Cupp, A.S., Wood, J.R., 20 14. Altered theca and cumulus oocyte complex gene expression, follicular arrest and reduced fertility in cows with dominant follicle follicular fluid androgen excess. PLoS One 9, e110683. Tadros, W., Lipshitz, H., 2009. The maternal to zygotic transition: a play in two acts. Development (Cambridge, England) 136, 3033 3042. Takahashi, K., Kotani, T., Katsu, Y., Yamashita, M., 2014. Possible involvement of insulin like growth factor 2 mRNA binding protein 3 in zebrafish oocyte maturation as a novel cyclin B1 mRNA binding protein that represses the translation in immature oocytes. Biochem Biophys Res Commun 448, 22 27. Tanaka, K.J., Ogawa, K., Takagi, M., Imamoto, N., Matsumoto, K., Tsujimoto, M., 2006. RAP55, a Cytoplasmic mRNP Component, Represses Translation in Xenopus Oocytes. Journal of Biological Chemistry 281. Thom, G., Minshall, N., Git, A., Argasinska, J., Standart, N., 2003. Role of cdc2 kinase phosphorylation and conserved N terminal proteolysis motifs in cytoplasmic polyadenylation element binding pr otein (CPEB) complex dissociation and degradation. Biochem J 370, 91 100. Tominaga, Y., Li, C., Wang, R.H., Deng, C.X., 2006. Murine Wee1 plays a critical role in cell cycle regulation and pre implantation stages of embryonic development. Int J Biol Sci 2, 161 170. Tunquist, B.J., Maller, J.L., 2003. Under arrest: cytostatic factor (CSF) mediated metaphase arrest in vertebrate eggs. Genes & Development. Uzbekova, S., Roy Sabau, M., Dalbis Tran, R., Perreau, C., Papillier, P., Mompart, F., Thelie, A., Penne tier, S., Cognie, J., Cadoret, V., Royere, D., Monget, P., Mermillod, P., 2006. Zygote arrest 1 gene in pig, cattle and human: evidence of different transcript variants in male and female germ cells. Reprod Biol Endocrinol 4, 12.

PAGE 96

82 Varnum, S.M., Wormington, W.M., 1990. Deadenylation of maternal mRNAs during Xenopus oocyte maturation does not require specific cis sequences: a default mechanism for translational control. Genes Dev 4, 2278 2286. Wang, Y.Y., Charlesworth, A., Byrd, S.M., Gregerson, R., MacNicol, M.C., MacNicol, A.M., 2008. A novel mRNA 3' untranslated region translational control sequence regulates Xenopus Wee1 mRNA translation. Developmental biology 317, 454 466. Wu, X., Viveiros, M.M., Eppig, J.J., Bai, Y., Fitzpatrick, S.L., Matzuk, M.M., 2003a Zygote arrest 1 (Zar1) is a novel maternal effect gene critical for the oocyte to embryo transition. Nat Genet 33, 187 191. Wu, X., Wang, P., Brown, C.A., Zilinski, C.A., Matzuk, M.M., 2003b. Zygote arrest 1 (Zar1) is an evolutionarily conserved gene exp ressed in vertebrate ovaries. Biol Reprod 69, 861 867. Xu, E.Y., Moore, F.L., Pera, R.A., 2001. A gene family required for human germ cell development evolved from an ancient meiotic gene conserved in metazoans. Proc Natl Acad Sci U S A 98, 7414 7419. Yama moto, T.M., Cook, J.M., Kotter, C.V., Khat, T., Silva, K.D., Ferreyros, M., Holt, J.W., Knight, J.D., Charlesworth, A., 2013. Zar1 represses translation in Xenopus oocytes and binds to the TCS in maternal mRNAs with different characteristics than Zar2. Bio chimica et biophysica acta 1829, 1034 1046. Yamamoto, T.M., Iwabuchi, M., Ohsumi, K., Kishimoto, T., 2005. APC/C Cdc20 mediated degradation of cyclin B participates in CSF arrest in unfertilized Xenopus eggs. Dev Biol 279, 345 355. Zhao, X., Singh, B., Arl inghaus, R.B., 1991. Inhibition of c mos protein kinase blocks mouse zygotes at the pronuclei stage. Oncogene 6, 1423 1426.

PAGE 97

83 APPENDIX: SPECIFIC P LASMID CONSTRUCTION pXen1 : This plasmid (MacNicol et al., 1997) (kindly provided by Angus MacNic ol, University of Arkansas for Medical Sciences, AR) provided the backbone for all plasmids constructed for this project The plasmid contains an SP6 promoter (Krieg and Melton, 1984) and a GST coding region in a globin UTR is present). pXen N MS2: PCR primers were designed to amplify the MS2 coding sequence from pJC5 (Gray et al., 2000) (kindly provided by Jeff Coller, Case Western Reserve University, OH) e (underlined) forward primer GAT C CC ATG G CT TCT AAC TTT AC GAT C AT CGA T GC GTA GAT GCC GGA GTT TGC TGC. Digested PCR product was ligated into NcoI/ClaI digested pXen1, replacing the GST coding sequence. pXen MS2 Xp54: P CR primers were designed to amplify the Xp54 coding sequence from Xp54 in the MSP vector (Minshall et al., 2001) (kindly provided by (underlined) GAT C G G TAC C CA TG A GCA CCG and reverse GAT C GG ATC C TT AAG GTT TGT. Digested PCR product was ligated into KpnI/BamHI digested pXen N MS2. pXen GST: pXen1 (MacNicol et al., 1997) was digested with NcoI and ClaI to remove the GST coding seq uence, then treated with Klenow to blunt ends and self ligated pXen C MS2: MS2 coding sequence was amplified from pJC5 using primers with a 5 XmaI site and a 3 XbaI site (underlined) forward primer 5 CTA G CC CGG G CT ATG GCT TCT AAC TTT ACT CAG TTC and reverse primer 5 GAT C TC TAG A GT TAG TAG ATG CCG GAG TTT GCT G. Digested PCR product was then ligated into XmaI/XbaI digested pXen GST. pXen Zar1 MS2: Full length Zar1 was cloned by RT PCR from total R NA from immature Xenopus oocytes ATG GTA CC C TCG AG G ATG GCT AGC CCT AG C CCG GG C AAT GAT ATA CTT GAA GCT. PCR products were digested with XhoI and XmaI (underline d) and ligated into pXen C MS2 MS2. pXen N Zar1 MS2: 1 159 aa were kept and the C terminal 160 307 aa were deleted from pXen Zar1 MS2. NcoI sites (underlined) were introduced by PCR to remove the C terminal domain: N terminal Zar 1 CGA T CC ATG G CT CAC CCT TCT CTT ATG C CC ATG G CC CGG GAT GGC TTC TAA CTT TAC. The PCR product was cut with NcoI and self ligated.

PAGE 98

84 pXen N Zar2 MS2: Amino acids 1 158 of zar2b was amplified from pENTR zar2b (underlined) GAT C GG TAC C AT GG GAT C GG ATC C GC TCT CTT CAG, appropriately digested, and ligated into KpnI/BamHI digested pXen C MS2. pXen Zar2 MS2: Full length Xenop us laevis Zar2b was amplified from pENTR (underlined) GAT C GG TAC C AT GGC GGG CTT TAT GTA GAT C GG ATC C GA CGA TGT ACT TGT AGC TGT AAG TG, appropriately digeste d, and ligated into KpnI/BamHI digested pXen C MS2 pXen M S2 AP: The AP linker, was purchased from Integrated DNA CGA TCG CCC CAG CCC CCG CTC CCG CCC CAG CCC CTG CTC CTG CCC CAG CTC CAG CTC, TCG AGA GCT GGA GCT GGG GCA GGA GCA GGG GCT GGG GCG GGA GCG GGG GCT GGG GC GAT and was ligated into ClaI/XhoI digested pXen N MS2. pXen MS2 EK: The EK linker was purchased from Integrated DNA Technologies forward CGA TCG CAG AAG CTG CTG CTA AGG AGG CCG CTG CAA AAG AAG CTG CCG CAA AAG CTC and reverse phos TCG AGA GCT TTT GCG GCA GCT TCT TTT GCA GCG GCC TCC TTA GCA GCA GCT TCT GCG AT and ligated into ClaI/XhoI digested pXen N MS2. pXen MS2 GS: The GS linker was purchased from Integrated DNA Technologies CGA TCG CTG GAG GGG GCG GCA GTG GGG GTG GTG GGA GCG GAG GAG GAG GGA GTG CTC and reverse TCG AGA GCA CTC CCT CCT CCT CCG CTC CCA CCA CCC CCA CTG CCG CCC CCT CCA GCG AT, and ligated into ClaI/XhoI digested pXen N MS2. pXen MS2 AP Zar1: The full length Zar1 sequence was amplified from pXen Zar1 riction site (underlined) GAT C CT CGA G GC CAT GGC TAG CTT CTC AGA G an d GAT C CC CGG G TC AAA TGA TAT ACT TGA AGC. A mplified product and pXen MS2 AP were digested with XhoI and XmaI, then ligated together. pXen MS2 EK Zar1: The full length Zar1 sequence was amplified from pXen Zar1 riction site, forward primer GAT C CT CGA G GC CAT GGC TAG CTT CTC AGA G and reverse primer GAT C CC CGG G TC AAA TGA TAT ACT TGA AGC. The amplified product was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen MS2 EK. pXen MS2 GS Zar1: The full length Zar1 sequence was amplified from pXen Zar1 riction site, forward primer GAT C CT CGA G GC CAT GGC TAG CTT CTC AGA G and reverse primer GAT C CC CGG G TC AAA TGA TAT ACT TGA AGC. The amplified product was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen MS2 GS.

PAGE 99

85 pXen MS2 AP Zar2: The full length Zar2 sequence was amplified from pXen Zar 2 MS2 with (underlined) forward primer GAT C CT CGA G GC CAT GGC GGG CTT TAT GTA TGC G and GAT C CC CGG G TC AAT CCG CGA CGA TGT ACT TGT AGC. The amplified product was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen MS2 AP. pXen MS2 EK Zar2: The full length Zar2 sequence was amplified from pXen Zar2 riction site (underlined) GAT C CT CGA G GC CAT GGC GG G CTT TAT GTA TGC G and GAT C CC CGG G TC AAT CCG CGA CGA TGT ACT TGT AGC. The amplified product was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen MS2 EK. pXen MS2 GS Zar2: The full length Zar2 sequence was amplified from pXen Zar2 riction site (underlined) GAT C CT CGA G GC CAT GGC GGG CTT TAT GTA TGC G and GAT C CC CGG G TC AAT CCG CGA CGA TGT ACT TGT AGC. The amplified product was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen MS2 GS. pXen rluc: PCR primers were designed to amplify Renilla luciferase (rluc) coding site (unde rlined) CAT G CC ATG G CT TCG AAA GTT TA T GAT GAT C AT CGA T TT ATT GTT CAT TTT TGA GAA CTC G. Digested PCR product was then ligated into NcoI/ClaI digested pXen1, replacing the GST coding sequence. This plasmid was linearized with EcoRI prior to in vitro transcription. pXen fluc: The firefly luciferase (fluc) coding sequence was amplified from JC18 (Gray et al., 2000) (kindly provided by Jeff Coller Case Western Reserve University, OH ) (underlined) forward primer GAT C CC ATG G AA GAC GCC AAA AA C ATA AAG and reverse primer GAT C CT CGA G TT ACA ATT TGG ACT TTC CGC C. The PCR product was inserted into pXen1 using NcoI/XhoI, replacing the GST coding sequence. pXe n fluc 2SL A 33 C 17 : (pXen fluc with stem loops, previously called fluc 2X upstream of the polyadenylation hexanucleotide, using QuikChange site directed mutagenesis (Strata gene) to make pXen fluc NdeI. A DNA duplex (IDT) containing the sequence of two MS2 stem NdeI sites was digested with NdeI and ligated into NdeI digested pXen fluc NdeI. This plasmid was linearized wi th SacI prior to in vitro transcription.

PAGE 100

86 pXen fluc 2SL A 0 : A SacI site (underlined) was QuickChanged (Agilent) into pXen fluc 2SL A 33 C 17 g forward AAG AAA GTT TCT TCA GAG CTC AAA AAA AAA AAA AA A and TTT TTT TTT TTT TTT GAG CTC TGA AGA AAC TTT CTT. The plasmid was linearized with SacI to remove the poly(A) tail. pXen fluc 2SL A 30 : QuickChange (Agilent) site directed mutagenesis was used to remove the A 33 C 17 tract from pXen flu c 2 SL A 33 C 17 and replace it with an A 30 tract ending in a BglII site (underlined) AAG TTT CTT CAC ATT CTA AAA AAA AAA AAA AAA AAA AAA AAA AAA A AG ATC T C T GCA GGT and ACC TGC AG A GAT CT T TTT TTT TTT TTT TTT TTT TTT TTT T TT TAG AAT GTG AAG AAA CTT. BglII cuts between the A and the G (underlined) and so when the plasmid is linearized with BglII, the poly(A) is termi nal and composed of 30 adenosine residues. pXen fluc 4SL A 33 C 17 : pXen fluc 2SL A 33 C 17 was digested with NdeI opening the plasmid and removing the 2SL insert, then ligated with an excess of the two MS2 stem loop (2SL) DNA duplex from Integrated DNA Technologies ( IDT ) Colonies were screened to isolate plasmids containing two insert s (a total of 4 stem loops) in the correct orientation. pXen fluc 4SL A 0 : A SacI site (underlined) was QuickChanged (Agilent) into pXen fluc 4SL A 33 C 17 leotide using forward AAG AAA GTT TCT TCA GAG CTC AAA A AA AAA AAA AAA and TTT TTT TTT TTT TTT GAG CTC TGA AGA AAC TTT CTT. The plasmid was linearized with SacI to remove the poly(A) tail. pXen fluc 4SL A 30 : QuickChange (Agilent) site directed mutagenesis was used to remove the A 33 C 17 tract f rom pXen fluc 4SL A 33 C 17 and replace it with an A 30 tract ending in a BglII site (underlined) AAG TTT CTT CAC ATT CTA AAA AAA AAA AAA AAA AAA AAA AAA AAA A AG ATC T C T GCA GGT and ACC TGC AG A GAT CT T TTT TTT TTT TTT TTT T TT TTT TTT plasmid is linearized with BglII, the poly(A) is terminal and composed of 30 adenylate residues. pXen1 Zar1a: This plasmid codes for GST Zar1 (with minimal linker) and its construction has been previously described (Yamamoto et al., 2013) pXe n GST AP: CGA TCG CCC CAG CCC CCG CTC CCG CCC CAG CCC CTG CTC CTG CCC CAG TCG AGA GCT GGA GCT GGG GCA GGA GCA GGG GCT GGG GCG GGA GCG GGG G CT GGG GC GAT, and ligated into ClaI/XhoI digested pXen1.

PAGE 101

87 pXen GST EK: The EK linker was purchased from Integrated DNA Technologies CGA TCG CAG AAG CTG CTG CTA AGG AGG CCG CTG CAA AAG AAG CTG CCG CAA AAG CTC and reverse TCG AGA GCT TTT GCG GCA GCT TCT TTT GCA GCG GCC TCC TTA GCA GCA GCT TCT GCG AT, and ligated into ClaI/XhoI digested pXen1. pXen GST GS: The GS linker was purchased from Integrated DNA Technologies orylated duplex CGA TCG CTG GAG GGG GCG GCA GTG GGG GTG GTG GGA GCG GAG GAG GAG GGA GTG CTC and reverse TCG AGA GCA CTC CCT CCT CCT CCG CTC CCA CCA CCC CCA CTG CCG CCC CCT CCA GCG AT, and ligated into ClaI/XhoI digested pXen1. pXen GST AP Zar1: F ul l length Zar1 sequence was amplified from pXen Zar1 riction site (underlined) forward GAT C CT CGA G GC CAT GGC TAG CTT CTC AGA G and reverse primer GAT C CC CGG G TC AAA TGA TAT ACT TGA AGC. The amplified product was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen GST AP. pXen GST EK Zar1: F ull length Zar1 sequence was amplified from pXen Zar1 riction site (underlined) forward GAT C CT CGA G GC CAT GGC TAG CTT CTC AGA G and reverse primer GAT C CC CGG G TC AAA TGA TAT ACT TGA AGC. The amplified product was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen GST EK. pXen GST GS Zar1: F ull leng th Zar1 sequence was amplified from pXen Zar1 riction site (underlined) forward GAT C CT CGA G GC CAT GGC TAG CTT CTC AGA G and reverse primer GAT C CC CGG G TC AAA TGA TAT ACT TGA AGC. The a mplified product was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen GST GS. pXen GST AP Zar2: F ull length Zar2 sequence was amplified from pXen Zar2 riction site (underlined) forw ard GAT C CT CGA G GC CAT GGC GGG CTT TAT GTA TGC G and reverse GAT C CC CGG G TC AAT CCG CGA CGA TGT ACT TGT AGC. P roduct was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen GST AP. pXen GST EK Zar2: F ull length Zar2 s equence was amplified from pXen Zar2 riction site (underlined) forward GAT C CT CGA G GC CAT GGC GGG CTT TAT GTA TGC G, reverse GAT C CC CGG G TC AAT CCG CGA CGA TGT ACT TGT AGC. P rodu ct was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen GST EK. pXen GST GS Zar2: F ull length Zar2 sequence was amplified from pXen Zar2 riction site (underlined) forward GAT C CT CGA G GC CAT GGC GGG CTT TAT GTA TGC G, reverse GAT C CC CGG G TC AAT CCG CGA CGA TGT ACT TGT AGC. P roduct was digested with XhoI and XmaI and ligated into XhoI/XmaI digested pXen GST GS.