Mechanisms of Mrna Post-Transcriptional Regulation in Mammalian Oocytes

Mechanisms of mRNA post-transcriptional regulation in mammalian oocytes

Karl-Frédéric VIEUX Department of Biology McGill University Montreal, QC, Canada July 2018

A thesis submitted to McGill University in partial fulfillment of the requirements of the degree of Doctor of Philosophy

Abstract ...... iv

Résumé ...... v

Acknowledgements ...... vi

Publications and author contributions ...... viii

List of figures ...... ix

List of tables ...... xi

List of abbreviations ...... xii

Chapter 1 – INTRODUCTION ...... 1

I – OOGENESIS ...... 2 1) In-utero (pre-natal) stages of oogenesis ...... 2 A. The making of germ cells ...... 3 A.1. Repression of intrinsic somatic signals and activation of pluripotency ...... 3 A.2. Epigenetic reprogramming ...... 4 B. PGC migration, proliferation and gonad colonization ...... 7 B.1. Migration and colonization ...... 7 B.2. Proliferation and survival ...... 8 C. Making oogonial: sex determination and meiosis ...... 11 C.1. Gonad differentiation ...... 11 C.2. Initiation of meiosis ...... 12 2) Post-natal stages of oogenesis ...... 15 A. Oocyte Growth ...... 16 A.1. Primordial follicle recruitment, folliculogenesis and oocyte growth ...... 16 A.2. Granulosa cells support oocyte growth ...... 20 A.3. Meiotic and developmental competence ...... 22 B. Oocyte maturation ...... 26 B.1. Maturation: GVBD and chromosome segregation ...... 26 B.2. Meiotic arrest...... 27 B.3. Molecular triggers of meiosis ...... 28

B.4. CDK1/Cyclin B1 orchestrate maturation ...... 29 C. Postnatal neo-oogenesis……………………….……………………………………….……….... 31 II – THE REGULATION OF GENE EXPRESSION IN THE OOCYTE ...... 34 1) Gene expression ...... 34 A. Transcription ...... 34 B. Translation...... 37 2) Post-transcriptional mechanisms of translational regulation ...... 40 A. The role of the 5'-methyl cap and the poly(A) tail in translational regulation 40 A.1. RNA stability ...... 40 A.2. Translational rate...... 42 A.3. Deadenylation ...... 45 B. Cis-elements and the recruitment of trans-factors...... 46 B.1. The untranslated regions, the open-reading frame ...... 46 B.2. Sequence-mediated regulation ...... 47 C. RNA localization and microenvironment (RNPs) ...... 53 C.1. Characteristics ...... 53 C.2. Formation ...... 54 3) RNA processing and management in the oocyte ...... 57 A. The challenges for gene expression in the oocyte ...... 57 A.1. Transcription during Oogenesis ...... 57 A.2. NSN to SN ...... 57 A.3. Histone modification profile in oocytes ...... 58 B. The need for translational regulation ...... 59 B.1. Meiotic and developmental competence: a complex ordeal ...... 59 B.2. Spatio-temporal regulation ...... 61 C. Mechanisms of RNA processing in mammalian oocytes ...... 62 C.1. Polyadenylation and translational activity ...... 62 C.2. Decapping, RNAi and degradation ...... 63 C.3. Oocyte RNPs: germ granules, P-bodies and stress granules ...... 64 D. Rationale and hypothesis ...... 66 Chapter 2 – MANUSCRIPT I ...... 67

Preface ...... 68

Significance statement...... 69 Abstract ...... 71 Introduction ...... 72 Results ...... 75 Discussion ...... 81 Materials and Methods ...... 84 Acknowledgements ...... 89 Figures ...... 90 Supplemental data ...... 103 Chapter 3 – MANUSCRIPT II ...... 106

Preface ...... 107 Significance statement...... 108 Abstract ...... 110 Introduction ...... 111 Materials and Methods ...... 114 Results ...... 119 Discussion ...... 124 Acknowledgements ...... 129 Figures ...... 130 Supplemental data ...... 139 Chapter 4 – DISCUSSION ...... 143

I – Oocytes are the perfect tool to study RNA management and processing events ...... 144 II – Poly(A)-independent translation and degradation ...... 146 III – RNPs in oocytes, RNPs in disease, RNPs in infertility ...... 148 IV – Conclusion: the oocyte transcriptome is an underappreciated parameter of oocyte quality ...... 152 References ...... 154

iii

Abstract

Precise regulation of gene expression is indispensable for proper cell function. This challenge is particularly complex in oocytes where transcription is active during growth but is undetectable in fully grown oocytes and during meiotic maturation. Transcripts required to drive oocyte maturation and early post-fertilization embryogenesis are therefore synthesized in growing oocytes and kept translationally silent until the appropriate time. RNA processing mechanisms play central roles in regulating translation and are indispensable for timely and successful oogenesis. In Manuscript I, an RNA processing mechanism is described in the oocyte. Maternal transcripts are categorized according to their poly(A) tail length at different stages of oocyte development. Type I and Type II transcripts are deadenylated and polyadenylated, respectively, during maturation. Type III transcripts are polyadenylated in the first half of maturation and deadenylated in the second half of maturation. CNOT6, a key component of the CCR4-NOT deadenylase complex, requires Pumilio and FBF binding elements (PBE) to deadenylate type III transcripts. Surprisingly however, the modulation of the poly(A) tail does not affect RNA translation and turnover in the oocyte. In Manuscript II, a novel type of RNP is described in cold-shocked oocytes. Inducible 18033 granules are identified as a novel type of stress granules whose formation is controlled by oocyte cell cycle regulators. More specifically, CDK1 activity prevents the formation of the granules at the later stages of maturation. A functional reporter gene assay also reveals a protective role for 18033 granules, preventing the non-reversible inhibition of translation in cold-shocked oocytes. The results described in this thesis highlight different facets of RNA metabolism in the oocyte and explore their role during maturation in producing eggs.

Résumé

Une régulation précise de l'expression des gènes est indispensable au bon fonctionnement des cellules. Ce défi est particulièrement complexe dans les ovocytes, où la transcription est active pendant la croissance mais indétectable dans les ovocytes à bout de croissance. Les transcrits nécessaires à la maturation des ovocytes et en début d'embryogenèse post-fécondation sont donc synthétisés dans les ovocytes en croissance et gardés silencieux jusqu'au moment opportun. Les mécanismes de maturation de l'ARN jouent un rôle central dans la régulation de la traduction et sont indispensables pour une ovogenèse opportune et réussie. Dans le Manuscrit I, un mécanisme de maturation de l'ARN chez l’ovocyte est décrit. Les transcrits maternels sont classés en fonction de la longueur de leur queue poly (A) durant les différents stades de développement de l'ovocyte. Les transcrits de type I et de type II sont respectivement déadénylés et polyadénylés pendant la maturation. Les transcrits de type III sont polyadénylés au cours de la première moitié de la maturation et déadénylés pendant la seconde moitié de la maturation. CNOT6, un composant clé du complexe de déadénylase CCR4-NOT, nécessite les éléments de liaison Pumilio et FBF (PBE) pour déadényler les transcrits de type III. Étonnamment, cependant, la modulation de la queue poly (A) ne régule pas la traduction et la dégradation de l'ARN dans l'ovocyte. Dans le manuscrit II, un nouveau type de RNP est décrit dans des ovocytes soumis à un choc thermique à froid. Les granules inductibles d'EDC4 sont identifiés comme étant un nouveau type de granules de stress dont la formation est régulée par le cycle cellulaire de l'ovocyte. Plus précisément, l'activité de CDK1 empêche la formation des granules à la fin de la maturation. Un essai fonctionnel de gènes rapporteurs décrit également un rôle protecteur pour les granules d’EDC4, empêchant la diminution de la traduction dans les ovocytes traités à froid. Les résultats décrits dans cette thèse mettent en évidence différentes facettes du métabolisme de l'ARN et explorent leur rôle au cours de la maturation de l’ovocyte et la production d'œufs compétents et de bonne qualité.

Acknowledgements

They often say, “It takes a village”. As I reflect on the last eleven years of my academic career, I

would be remiss if I did not take the time to acknowledge the people who supported me along the

way.

First and foremost, to my mentor, my supervisor, my friend: Hugh Clarke. Working in your lab has been a privilege. I will forever cherish this experience and will have a hard time matching it

as I move on in my career. Thank you for your patience and your wisdom.

I would also like to extend my gratitude to the other members of my supervisory committee, Dr.

Paul Lasko and Dr. Tamara Western, for their advice and their guidance.

To the countless friends I’ve made along the way, all the members of the Clarke lab, the reproductive biology community at McGill, in Montreal and in Quebec, in particular to Laleh

Abassi, Victoria Chiwara, Qin Yang and my writing partner, Kathrin Spiller, thank you for your feedback and your support.

To Marraine and Parrain, my ‘Taties’, ‘Tontons’, cousins and distant relatives, my big Haitian family, you have always been my loudest cheerleaders. To Maggy, Richard, Gina, Faria, Sabrina,

Sophie and Remi in particular, thank you for your endless support.

To my second family, Stephany El-Hayek, Omar Farah, and Nicole Edwards, my best friends, “As we go on, we remember, all the times we had together. And as our lives change from whatever we will still be friends forever” (Vitamin C).

I have three heroes in my life. My father Patrick is humble, honest and is the best role model a son could ask for. My mother Maryse is strong, unapologetically opiniated, loyal and the best person

vi to have in your corner. My brother Alexandre remains the most idealistic and kind-hearted revolutionary; we often disagree but he always challenges me for the best. They have taught me grace and the value of hard work, encouraged my curiosity and my love of knowledge, supported me my whole life and loved me unconditionally.

To my three heroes, I love you. Thank you.

It has not gone unnoticed to me that throughout my career, I have seen and heard of very few people that look like me in positions of power or influence. It is also evident that this is an endemic problem that spreads across all industries. For far too long ‘minorities’ AND women have had to excel to simply survive in white male dominated spaces. There is hope however. We stand today at the precipice of change. The time for antiquated practices, inaction and complicity is gone. We will no longer content ourselves with just surviving but seek to thrive. We want leadership reflective of the complex and diverse fabric of our societies. We demand a seat at the table and a voice in shaping policies and programs that affect our communities. We are paving the road for future generations to live in a truly egalitarian society, full of hope and opportunity for all.

Let it therefore be known that this thesis was written by a proud black Haitian man. Let young girls and boys of colour share in my success and know that they too, can do it. If I can inspire a single one of them, then I will truly have succeeded.

Lastly, I dedicate this thesis to my godson Zion Nicholas and the next generations from the Vieux and Ferrus clans. Go forth knowing you can become whoever and do whatever you want in life.

Be proud of who you are and where you come from. And finally, never let anyone, or anything convince you that you are capable of anything less than greatness.

vii

Publications and author contributions

H.J. Clarke & K.-F. Vieux (2015). Epigenetic inheritance through the female germline: The known, the unknown, and the possible. Semin Cell Dev Biol. 2015 Jul;43:106-116. doi: 10.1016/ j.semcdb.2015.07.003. Epub 2015 Jul 13. (invited chapter)

K.-F. Vieux & H.J. Clarke (2018) CNOT6 regulates a novel pattern of mRNA deadenylation during oocyte maturation. Scientific Reports, volume 8, Article number: 6812 doi:10.1038/s41598-018-25187-0. (Manuscript I)

K.-F. Vieux & H.J. Clarke (in preparation) Maturing oocytes lose the ability to form 18033 (EDC4) granules and protect translation after cold-shock via a CDK1-dependent process (Manuscript II)

Contribution of authors H.J. Clarke & K.-F. Vieux (2015) Contribution: The candidate, K.-F. Vieux and H.J. Clarke performed the literature research and the primary writing of the review. K.-F. Vieux and H.J. Clarke performed the figure preparation. K.-F. Vieux and H.J. Clarke provided the intellectual contribution.

K.-F. Vieux & H.J. Clarke (2018) Contribution: The candidate, K.-F. Vieux, performed the experiments presented in this manuscript. K.-F. Vieux performed the data analysis for all experiments and the figure preparation. K.-F. Vieux and H.J. Clarke wrote the manuscript. K.-F. Vieux and H.J. Clarke provided the intellectual contribution.

K.-F. Vieux & H.J. Clarke (in preparation) Contribution: The candidate, K.-F. Vieux, performed the experiments presented in this manuscript. K.-F. Vieux performed the data analysis for all experiments and the figure preparation. K.-F. Vieux & H.J. Clarke wrote the manuscript. K.-F. Vieux & H.J. Clarke provided the intellectual contribution.

viii

List of figures

Figure 1 – Primordial germ cell differentiation...... 10

Figure 2 – Follicle formation...... 14

Figure 3 – Folliculogenesis and oocyte growth ...... 19

Figure 4 – Growth, meiotic and developmental competence ...... 24

Figure 5 – Oocyte maturation. Maturation begins with the breakdown of the nuclear envelope and the condensation of the chromosome ...... 33

Figure 6 – Gene expression ...... 35-36

Figure 7 – Ribosome structure ...... 39

Figure 8 – Translation initiation, Elongation, Termination and ribosomal recycling...... 43-44

Figure 9 – Cis-regulatory elements ...... 52

Figure 10 – Liquid-liquid phase separation ...... 56

Figure 11 – Spectrum of induced ribonucleoprotein particles and composition ...... 65

Figure 12 – Patterns of mRNA polyadenylation during oocyte meiotic maturation ...... 90

Figure 13 – Expression of deadenylases in oocytes ...... 91

Figure 14 – Localization of CNOT6 in growing and fully-grown oocytes ...... 92

Figure 15 – Co-localization of CNOT6 foci and cytoplasmic ribonucleoprotein (RNP) markers ...... 93

Figure 16 – Reduction of CNOT6 impairs the deadenylation of Slbp and Orc6 during late maturation ...... 94-95

Figure 17 – Terminal portion of the 3′-UTR of different mRNAs showing location of sequences that potentially regulate polyadenylation ...... 96

Figure 18 – Luciferase-Slbp 3′-UTR constructs and injections ...... 97

Figure 19 – Increasing poly(A) tail-length does not affect mRNA translation or degradation ... 98

Figure 20 – Model for differential poly(A) tail length dynamics during meiotic maturation ..... 99

Figure 21 – Supplemental A-B ...... 103 ix

Figure 22 – Supplemental C-E ...... 104

Figure 23 – Supplemental F-H ...... 105

Figure 24 – Formation of novel 18033 granules is induced in cold-shocked oocytes ...... 130-131

Figure 25 – Formation of novel 18033 granules requires stable microtubules ...... 132

Figure 26 – Ability to form 18033 granules in response to cold-shock is lost during maturation ...... 133-134

Figure 27 – Inhibition of CDK1 preserves the ability to form 18033 granules in MII oocytes 135

Figure 28 – Loss of the ability to make 18033 granules during cold-shock coincides with impaired translation following cold-shock ...... 136-137

Figure 29 – Preventing the loss of 18033 granules forming ability late in maturation rescues translation post-cold-shock ...... 138

Figure 30 – 18033 granules are EDC4 granules ...... 139

Figure 31 – MG132 treatment induces the formation of 18033 granules ...... 140

Figure 32 – Canonical stress granule markers do not localize to 18033 granules ...... 141

Figure 33 – Inhibition of the MAPK activity partially preserves the ability to make 18033 granules in MII oocytes ...... 142

List of tables

Table 1 – PCR primer sequences..………………………………………………………………100 Table 2 – RNA-Linker and RL-PAT primer sequences..……………………….……………....101

Table 3 – qPCR primer sequences….…………………………………………………………..102

List of abbreviations

ABCE1 ATP-binding cassette sub-family E member 1 ADAM a disintegrin and metalloproteinase AID activation-induced cytidine deaminase AKT protein kinase B ALDH1 aldehyde dehydrogenase 1 ALS amyotrophic lateral sclerosis AP2γ activating protein 2γ ATP adenosine triphosphate BAX BCL-2-associated x protein / BCL-2-like protein 4 BAK BCL-2 homologous antagonist/killer BAD BCL-2-associated death promoter BIM BCL-2-like protein 11 BMP bone morphogenetic protein BSA bovine serum albumin CAMP cyclic adenosine monophosphate CCNB1 cyclin B 1 CCR4-NOT carbon catabolite repressor 4 – negative on TATA CDK1 cyclin-dependent kinase 1 CER cereberus cGMP cyclic guanosine monophosphate CNOT CCR4-NOT subunit CNP C-type natriuretic peptide CPE cytoplasmic polyadenylation element CPEB CPE-binding element CPSF cleavage and polyadenylation-specificity factor COC cumulus-oocyte complex CYP26B1 Cytochrome P450 Family 26 Subfamily B Member 1 CXCL chemokine (C-X-C motif) ligand CXCR C-X-C chemokine receptor

xii

DAZL deleted in azoospermia-like DKK1 Dickkopf-related protein 1 dsRNA double-stranded RNA eCG equine chorionic gonadotropin EDC4 enhancer of mRNA decapping protein 4 eEF translation elongation factor eIF translation initiation factor eRF translation termination factor FGF fibroblast growth factor FIGLA Folliculogenesis Specific BHLH Transcription Factor FOXO forkhead box O FOXL2 forkhead box protein L2 FSH follicle-stimulating hormone FSHR FSH receptor FTLD frontotemporal lobar degeneration FUS fused in sarcoma GLD2 germ line development 2 GLI glioma-associated oncogene GOC granulosa-oocyte complex GPCR G-protein coupled receptor grP-bodies germline related P-bodies GVBD germinal vesicle breakdown G3BP ATP-dependent DNA helicase VIII / stress granule assembly factor 1 hCG human chorionic gonadotropin HOX homeobox protein H3K9me(2/3) histone 3 lysine 9 (di-/tri-)methylation IAK-1 Aurora Kinase A ICSI intra-cytoplasmic sperm injection IVF in vitro fertilization KIT mast/stem cell growth factor receptor KL KIT ligand

xiii

KLF2 Krüppel-like Factor 2 LGR5 leucine rich repeat rontaining G protein-coupled receptor 5 LIF leukemia inhibitory factor LHX LIM homeobox gene LH luteinizing hormone LHCGR luteinizing hormone receptor MAPK mitogen-activated protein kinase MEM minimal essential medium miRNA micro-RNA MPF maturation-promoting factor MTOC microtubule-organizing center MTORC mechanistic target of rapamycin complex 1 NPPC natriuretic peptide C NPR2 natriuretic peptide receptor B NP95 nuclear protein 95 NSN non-surrounded nucleolus OBOX oocyte-specific homeobox 1 OCT4 octamer-binding transcription factor 4 PABP poly(A)-binding protein PAN poly(A) nuclease PARN poly(A)-specific ribonuclease PBE PUF-binding element PBS phosphate-buffered saline PBST tween- PBS PCR polymerase chain reaction PD postnatal day PDE phosphodiesterase PGC primordial germ cell PKA protein kinase A PUF pumilio and FBF PUM pumilio

xiv

P-bodies processing bodies RNAi RNA interference RNP ribonucleoprotein complex RAR retinoic acid receptor RXR retinoid X receptor siRNA short interfering RNA SLBP stem-loop binding protein SMAD sma-and mad-related SN surrounded nucleolus SOX9 SRY-Box 9 STRA8 stimulated by retinoic acid 8 SRY sex-determining region Y TCFAP2E transcription factor AP-2ε / activating enhancer binding protein 2ε TDG thymine-DNA glycosylase TET1/2 ten-eleven translocation TIA-1 cytotoxic granule associated RNA binding protein TNRC6A trinucleotide repeat-containing gene 6A protein TZP trans-zonal projection UTR untranslated region WNT wingless-Type MMTV integration site family XRN1 5' to 3' exoribonuclease 1

Chapter 1

[INTRODUCTION]

I – OOGENESIS

Reproduction is essential for the preservation and the evolution of all life. Sexual reproduction requires the successful interaction and fusion of gametes through the process known as fertilization. This process is entirely dependent on the contribution of one gamete by each parent, the fusion of which produces an offspring that has both paternal and maternal genetic input. The paternal gamete is called the spermatozoon and the female gamete the oocyte. Both sperm and oocytes arise from a population of primordial germ cells (PGCs) established early in mouse embryo development at E7.25. Distinct differentiation processes however yield the two sex- specific types of germ cells. In the case of oocytes, fertilization can only take place at the end of the female-specific process referred to as ‘oogenesis’. Furthermore, the initiation of embryogenesis is heavily dependent on a variety of factors accumulated during oogenesis. The orderly and successful execution of each of the differentiation steps that constitute oogenesis and the proper management of the oocyte’s assets are critical to the formation of fertilizable and developmentally competent eggs.

1) In-utero (pre-natal) stages of oogenesis

Oogenesis begins in utero when primordial germ cells (PGCs) are first detectable at E7.25 (Figure

1). The source of PGCs and the molecular pathways responsible for their differentiation vary

between species (Richardson and Lehmann 2010). In mice, BMP signalling induces the

differentiation of about 45 posterior proximal epiblast cells in the extraembryonic ectoderm into

PGCs (Ginsburg, Snow et al. 1990, Chang and Matzuk 2001, Tremblay, Dunn et al. 2001, Arnold,

Maretto et al. 2006). More specifically, BMP8b and BMP4 secreted by the extraembryonic ectoderm, as well as BMP2 from the visceral endoderm act synergistically to promote germ cell fate (Ying, Qi et al. 2001, Ying and Zhao 2001). Other signalling pathways, albeit not as well

described, are also involved. WNT signalling originating from the posterior proximal epiblast and

the posterior visceral endoderm appears to be involved in potentiating the epiblast cells, preparing

them to respond to the BMP signals (Ohinata, Ohta et al. 2009, Kumar and DeFalco 2017). Nodal’s

role in patterning all three germ layers also appears to be important in setting up BMP and WNT

signalling and therefore PGC specification; sustained Nodal signalling however, does not seem to

be required (Brennan, Lu et al. 2001, Kumar and DeFalco 2017). Antagonistic signals from the

anterior visceral endoderm including CER, DKK1 and LEFTY1, counter the effects of BMP4,

WNT and NODAL, respectively, repressing the specification of PGCs and restricting their

induction to the posterior epiblast of the embryo (Ohinata, Ohta et al. 2009, Kumar and DeFalco

2017).

A. The making of germ cells

A.1. Repression of intrinsic somatic signals and activation of pluripotency

The emergence of PGCs can be observed within a few days into embryogenesis, when their

defining alkaline phosphatase activity is detectable (Swartz 1982). The role of these enzymes in

PGC differentiation is not understood, but high alkaline phosphatase activity is routinely

associated with ‘stem cell-ness’ (Stefkova, Prochazkova et al. 2015). In fact, the suppression of a

somatic differentiated state and the promotion of a pluripotent naive state are characteristic of

PGCs. BLIMP1 (a PR domain-containing transcription factor also known as PRDM1) expressed

in precursor cells as early as E6.25, plays a critical role in this process. Blimp1 -/- mice display a

limited number of germ cell precursors, and these are unable to migrate to and colonize the

emerging gonads (Ohinata, Payer et al. 2005). Known to be a potent transcriptional repressor,

BLIMP1 in PGCs represses the expression of homeobox genes Hoxa1 and Hoxb1, involved in

somatic cell differentiation (Ohinata, Payer et al. 2005). Other somatic markers, including Lim1,

Evx1, Fgf8 and Snail, are actively repressed in PGCs (Ancelin, Lange et al. 2006, Hayashi, de

Sousa Lopes et al. 2007, Kurimoto, Yabuta et al. 2008). Conversely, BLIMP1 also promotes

pluripotency by upregulating Sox2 and Nanos (Matsui, Zsebo et al. 1992, Ohinata, Payer et al.

2005, Kurimoto, Yabuta et al. 2008). Other pluripotent genes such as Sox2, Nanog, Oct4 are also upregulated and indispensable in PGC differentiation (Okamura, Tokitake et al. 2008, Ewen and

Koopman 2010). A second member of the PR domain-containing transcription factor regulator

family, PRDM14, has been shown to also be necessary for the upregulation of pluripotency genes,

including Sox2 and Stella, and the induction of the alkaline phosphatase activity (Yamaji, Seki et

al. 2008). BLIMP1 and PRDM14 target similar genes and work together with transcription factor

AP2γ to coordinate reacquisition of pluripotency and PGC induction. While activation of

pluripotency markers Oct4, Nanog and Sox2 is indispensable, the activation of Klf2 and Stella is not necessary for PGC induction (Kehler, Tolkunova et al. 2004, Chambers, Silva et al. 2007,

Yamaguchi, Kurimoto et al. 2009, Campolo, Gori et al. 2013, Kumar and DeFalco 2017).

Additionally, BLIMP1 and PRDM14, which can also target mutually exclusive gene groups and work independently of one another (Yamaji, Seki et al. 2008), are also found downstream of

Nanog, further marking and promoting the commitment to the germ cell fate (Murakami,

Gunesdogan et al. 2016).

A.2. Epigenetic reprogramming

Epigenetics is a relatively new field of study, investigating the molecular processes and mechanisms that regulate gene expression and output without changing the DNA sequence of the gene. Amongst them, DNA methylation and histone modifications are well-described transcriptional regulators with important roles in germ cell development. Parent-of-origin-specific

DNA methylation patterns, known as imprints, are established during sex-specific gametogenesis.

Additional layers of DNA methylation and histone modifications are also added shortly after

conception and with every differentiation step, as somatic cells lose pluripotency potential and

progressively commit to highly specialized cell fates.

When PGCs arise, they must rid themselves of the epigenetic marks of their somatic precursors to

remove the repressive effects on pluripotency gene expression and restore bi-allelic gene expression of imprinted loci. Interestingly, Blimp1, Prdm14 and Tcfap2c have also been implicated both directly and indirectly in the reprogramming of the PGC epigenome (Kurimoto, Yamaji et al.

2008, Magnusdottir, Dietmann et al. 2013) . A global DNA demethylation event occurs in PGCs after their specialization, before they commit to a sex-specific germ cell fate. An active demethylation process utilizing the base-excision repair pathway in coordination with Aid, Tdg or

Tet1/2, may be involved in DNA demethylation in PGCs (Kohli and Zhang 2013, Messerschmidt,

Knowles et al. 2014). This mechanism may be responsible for a late wave of demethylation targeting protected loci including imprinted genes and the promoters of genes involved in meiosis after E11.5 but needs to be further investigated (Messerschmidt, Knowles et al. 2014). The majority of DNA methylation however, is removed through a passive and earlier-acting mechanism. By E9.5, the expression of de novo DNA methyltransferases Dnmt3a/b is repressed in PGCs (Yabuta, Kurimoto et al. 2006, Kurimoto, Yabuta et al. 2008). In addition, Np95 expression is also repressed, preventing the transportation of its co-factor, the maintenance methyltransferase, DNMT1, to replication foci (Sharif, Muto et al. 2007, Kurimoto, Yabuta et al.

2008, Seisenberger, Andrews et al. 2012). Moreover, there is an exponential proliferation of PGCs after E9.5, until E12.5, correlating with the suppression of methylation events in PGCs (Kagiwada,

Kurimoto et al. 2013). In the absence of these enzymes and the lack of new and maintenance DNA

methylation, the epigenetic mark is diluted with every cycle of DNA replication and cell division,

and as a result the DNA of PGCs becomes hypomethylated.

During PGC migration, a reshuffle of histone modifications also occurs. First, there is a

replacement of histone H3 by a H3.3 variant that rids the nucleosomes of Lysine 9 (K9) methylated

(-me) and dimethylated marks (-me2) (Hajkova, Ancelin et al. 2008). Given the tight link between

the methylation of H3K9 and DNA methylation in other species, the loss of H3K9me1/2 could

also participate in the passive loss of DNA methylation observed in PGCs (Tamaru and Selker

2001, Jackson, Lindroth et al. 2002, Tamaru, Zhang et al. 2003). Conversely, H3K27me3 and

H4R3me2, respectively, increase in PGCs after E8.25 because of an abundance of the Polycomb

repressive complex 2 (PRC2) lysine and its methyltransferase activity and the translocation to the

nucleus of a BLIMP1-PRMT5 complex with arginine (R) methyltransferase activity (Ancelin,

Lange et al. 2006, Prokopuk, Stringer et al. 2017).

In female (XX) somatic cells, one of the X chromosomes is silenced randomly by epigenetic

mechanisms. Although this inactivation originally persists upon PGC differentiation, it is

eventually reversed in order to establish the proper dosage of X-linked genes required for proper

female germ cell development (Taketo 2015). This reversal begins around E9.5 with a progressive

50-70% decrease of Xist in PGCs and around their inactivated X chromosomes (de Napoles,

Nesterova et al. 2007, Sugimoto and Abe 2007). As a result, PRC2 recruitment to the inactivated

X chromosomes is lost. The subsequent loss of H3K27me3 is followed by an increase in X-linked

gene expression, despite the high levels of PRC2 and the high H3K27me3 levels in PGCs (de

Napoles, Nesterova et al. 2007, Sugimoto and Abe 2007, Chuva de Sousa Lopes, Hayashi et al.

2008). The process is completed after PGCs have colonized the genital ridge and undetermined

sex-non-specific signals from the somatic environment further stimulate X chromosome

reactivation (Chuva de Sousa Lopes, Hayashi et al. 2008).

B. PGC migration, proliferation and gonad colonization

B.1. Migration and colonization

Once specified, PGCs begin their well-described journey to the undifferentiated gonads (Figure

1). First, by undergoing an epithelial-mesenchymal transition, they leave the epiblast and migrate through the posterior end of the primitive streak and into the soon-to-be hindgut in the endoderm

(Anderson, Copeland et al. 2000, Molyneaux, Stallock et al. 2001). As the hindgut develops and extends anteriorly, the PGCs move anteriorly. By E9.5, PGCs move dorsally and begin leaving the hindgut and migrating to the mesoderm through the dorsal mesentery (Molyneaux, Stallock et al.

2001). Once they have reached the mesoderm, they begin to migrate laterally from the midline and into the nascent genital ridge, which is developing from a thickening of the coelomic epithelium covering the mesonephric tissue.

PGC migration is highly regulated and involves signals responsible both for inducing cell motility and movement directionality. The KL/cKit/AKT pathway is essential to the migration of PGCs

(Runyan, Schaible et al. 2006, Farini, La Sala et al. 2007, Gu, Runyan et al. 2009). Cell motility and directionality can also be controlled independently. Particularly, signals in the form of chemoattractant and extra-cellular matrix (ECM) glycoproteins provide directional cues to migrating PGCs. Defects in the expression of cytokine CXCL12, secreted by the somatic environment, and its receptor CXCR4, on the cell surface of PGCs, have been shown to impair gonad colonization without disrupting PGC migration, attesting to their role in directionality of the PGC movement (Ara, Nakamura et al. 2003). Adhesion to a myriad of ECM glycoproteins found to be expressed along the migration path to the developing gonad, including collagen III, IV

and V, fibronectin, laminin and tenascin-C, also appears to be implicated in dictating the

directionality but not motility of the PGC movements (Ewen and Koopman 2010).

Adhesion proteins have been reported to play a role in the final stages of PGC migration. More

specifically between E10.5 and E11.5, clusters of cells form as PGCs establish physical interaction

with one another through cytoplasmic projections (Gomperts, Garcia-Castro et al. 1994). This has been suggested to play an important role in their migration after leaving the hindgut, the coalescence of PGCs and the colonization of the genital ridges (Gomperts, Garcia-Castro et al.

1994, Bendel-Stenzel, Gomperts et al. 2000, Molyneaux, Stallock et al. 2001, Wakayama, Hamada et al. 2003). To that effect, many adhesion molecules expressed in PGCs, including E-cadherin,

PECAM-1, integrins and members of the ADAM family, are believed to play a role in PGC

migration and homing into the genital ridge (Anderson, Fassler et al. 1999, Bendel-Stenzel,

Gomperts et al. 2000, Rosselot, Kierszenbaum et al. 2003, Wakayama, Hamada et al. 2003)

B.2. Proliferation and survival

By the time PGCs begin their journey to the genital ridge, their numbers have increased from ~45

to ~100. During the ensuing migration, PGCs continue to proliferate, increasing to ~3,000 when

they reach the genital ridge at E11 and ~25,000 in the developing gonads at E13.5, just before they

begin their sex-specific differentiation (Tam and Snow 1981). Interestingly, cell migration,

proliferation and survival appear to be tightly linked, ensuring the colonization by a large number

of germ cell precursors in the appropriate tissue. For instance, survival signals such as FGF2 and

FGF4 originate from the mesoderm along the migratory path of PGCs (Takeuchi, Molyneaux et

al. 2005). In fact, ectopic PGCs that have diverged away from the normal migratory path, or that

have been left behind, enter meiosis and become more susceptible to apoptosis (Molyneaux,

Stallock et al. 2001, Runyan, Schaible et al. 2006). Intrinsically, apoptosis can be induced by

apoptosis pathway components Bax, Bak, Bad, Bim and Caspase3, all of which are expressed in

PGCs (Runyan, Schaible et al. 2006). Bax-mediated apoptosis of PGCs however, is avoided by

activation of the receptor c-KIT by KL from the soma along the migratory path; ectopic PGCs that

are not exposed to KL thereby die off (Buehr, McLaren et al. 1993, Farini, La Sala et al. 2007).

The KL/c-Kit/AKT pathway also plays a part in PGC proliferation once they reach the hindgut. Its

involvement in this process becomes critical, however, after they leave the hindgut and is sustained

until the colonization of the genital ridge (Mahakali Zama, Hudson et al. 2005). Proliferation and

migration are often mutually exclusive, so a fine balance must be established by the migrating

PGCs. While FGF2 appears to regulate motility, FGF7 affects PGC proliferation and cell survival

(Takeuchi, Molyneaux et al. 2005). Again, KL, in its two forms (soluble or membrane-bound), is important for this balance. Soluble KL can both repress and enhance FGF-mediated proliferation,

respectively, depending on the presence or the absence of membrane-bound KL in the environment

(Kawase, Hashimoto et al. 2004). KL-stimulation of PGCs intrinsically upregulates cAMP levels,

in coordination with LIF and forskolin signalling, to promote proliferation (Dolci, Pesce et al.

1993). Interestingly LIF and other cytokines like IL-4 and IL-2 also promote cell survival while enhancing proliferation (Ewen and Koopman 2010).

Upon entry into the gonad, PGCs initially continue to proliferate. They begin to express FGF4 and

FGF8 which act in an autocrine fashion to stimulate proliferation; at this stage mitotic proliferation

is FGF-dependent (Kawase, Hashimoto et al. 2004). Synchronous mitotic cell divisions paired

with incomplete cytokinesis form clusters of PGCs, connected through intercellular bridges termed

ring canals (Pepling, de Cuevas et al. 1999). By E13.5-14.5, the upregulation of membrane-bound

KL in the soma of the differentiating gonads, brings germ cell mitotic proliferation to an abrupt

E10.5

E9.5 E7.5 E8

Figure 1 – Primordial germ cell differentiation. In mice, PGCs arise in the ectoderm at E7.5. They then migrate through posterior end

of the primitive streak to the endoderm. At E8, PGCs begin to migrate anteriorly along the developing hindgut. At E9.5-10.5, PGCs first

migrate to the dorsal wall of the hindgut, then through the dorsal mesentery and finally bilaterally from the midline into the nascent

genital ridge (Adapted from Richardson and Lehman, 2010). 10

end, just as sex determination begins (Kawase, Hashimoto et al. 2004). In parallel, PGCs also

begin to lose pluripotency capabilities upon arrival at the genital (Ewen and Koopman 2010). Yet core pluripotency markers maintain their expression past E10.5 (Scholer, Balling et al. 1989,

Scholer, Dressler et al. 1990, Avilion, Nicolis et al. 2003). The downregulation of Nanog, Oct4,

Sox2 and Stella concludes the resumption of pluripotency loss but only follows the sex

determination and initiation of meiosis described below. In particular, the maintenance of OCT4

expression has been shown to be critical for PGC survival until its physiological downregulation

around E13.5 (Kehler, Tolkunova et al. 2004).

C. Making oogonial: sex determination and meiosis

C.1. Gonad differentiation

The decision to develop the genital ridge into a testis or an ovary is entirely dependent on the

presence or absence of the Y-linked gene SRY in somatic cells. Female determination is initiated

by default between E11.5 and E12.5, by the spontaneous upregulation of Wnt4 and Foxl2, which

activates the ovarian differentiation program of the genital ridge (Kashimada and Koopman 2010).

In male (XY) genital ridges, the transient expression of SRY in the soma around E10-10.5,

upregulates Sox9 which in turn upregulates Fgf9 (Colvin, Green et al. 2001, Kim, Kobayashi et al.

2006, Sekido and Lovell-Badge 2009). FGF9 reciprocally enhances the expression of Sox9 which

is also auto-regulated by SOX9, creating positive feedback loops that amplify the male-

determination signal in the gonadal soma and maintains it after SRY levels drop (Kim, Kobayashi et al. 2006, Jakob and Lovell-Badge 2011). FGF9 also upregulate the expression of genes involved in Sertoli cell differentiation and inhibits female-determination genes Wnt4 (Kim, Kobayashi et al.

2006, Jameson, Lin et al. 2012). This kick-starts the masculinization of the somatic cells, now differentiating into Sertoli cells, while deterring the feminization of the gonad. In the absence of

SRY (e.g. in female XX gonads, in the absence of the Y chromosome or a functional Sry gene)

Foxl2 is upregulated in somatic cells of the gonad, acting in coordination with the WNT signalling

pathway to drive the granulosa cell fate while preventing the masculinization of the gonad

(Ottolenghi, Pelosi et al. 2007). The resulting LGR5 and FOXL2 positive pre-granulosa cells,

derived from mitotically arrested ovarian epithelial cells, join and surround the germ cell cysts in

the first of two waves granulosa cell recruitment (Bristol-Gould, Kreeger et al. 2006, Mork,

Maatouk et al. 2012) (Figure 2).

At the same time ALDH1A2, expressed in the mesonephros, catalyses the production of retinoic

acid (RA) in the mesonephric duct and tubules, spreading to the undifferentiated gonads as early

as E10.5 (Bowles, Knight et al. 2006). Interestingly the RA-degrading enzyme CYP26B1 is originally expressed at basal levels in the gonads of both sexes. It is then significantly upregulated or downregulated in the somatic cells, resulting from the enhancement of its expression by Sox9

in developing testes or its inhibition by Foxl2 in developing ovaries respectively (Bowles, Knight

et al. 2006, Kashimada, Svingen et al. 2011). Accordingly, prior to E12.5, there is little to no RA

build-up in the gonads of both sexes but after E12.5, there is an accumulation of RA in developing

ovaries but not in testes (Bowles, Knight et al. 2006). It is this critical difference in RA levels that

establishes the basis for the divergence between germ cells undertaking spermatogenesis and those

undertaking oogenesis.

C.2. Initiation of meiosis

From their emergence, until a few days after their colonization of the genital ridge, PGCs are

bipotential, retaining the ability to take on either the spermatogonial or the oogonial cell fate. It is

also around E12.5 that the two processes begin to diverge. The use of chimeric mice and sex-

reversal studies have shown that both XX and XY germ cells can initiate male germ cell

differentiation if cultured in a testicular environment, or female germ cell differentiation if cultured

in an ovarian environment (Ford, Evans et al. 1975, Palmer and Burgoyne 1991). These findings

showed that sex determination of germ cells is not cell autonomous and confirmed that they in fact

retain the ability to respond to both testicular and ovarian cues until E13.5. Moreover, the sex of

the somatic environment in the gonad dictates the choice between oogenesis and spermatogenesis;

RA is at the centre of this decision.

Undifferentiated germ cells express the RA receptors RAR and the retinoid receptor RXR (Morita

and Tilly 1999, Bowles, Knight et al. 2006). In the presence of retinoic acid and the activation of

RAR and RXR, germ cells upregulate Stra8 (Stimulated by retinoic acid gene 8) and initiate

meiosis (Menke, Koubova et al. 2003, Anderson, Baltus et al. 2008). STRA8, which has

transcription factor properties, is vital for pre-meiotic DNA replication and meiotic initiation,

though the mechanism by which it carries this out remains undetermined (Baltus, Menke et al.

2006, Feng, Bowles et al. 2014). With the expression of Stra8, the up-regulation of other meiosis

markers Sycp3 and Dmc1 ensues and the female-specific part of oogenesis begins (Menke,

Koubova et al. 2003, Baltus, Menke et al. 2006). The commitment to the female germ line fate is

therefore intimately linked to an immediate initiation of meiosis by the oogonial germ cells in the

developing ovary.

Meiosis is integral to the maintenance of ploidy within a species and across generations during

sexual reproduction. It is an extensively studied process (Burgoyne, Mahadevaiah et al. 2009,

MacLennan, Crichton et al. 2015). During the mitosis-to-meiosis transition, the oogonia undergo a rapid round of pre-meiotic DNA replication (meiotic S-phase) duplicating both parental copies of every chromosome. The resulting sister chromatids are held together by cohesins within homologous chromosomes as meiosis then commences with Prophase I. The first stage of prophase

Figure 2 – Follicle formation. Upon their entry in the gonad around E10.5, PGCs become germ cells (blue cells with yellow nuclei) and undergo numerous rounds of mitosis. The incomplete cytokinesis creates germ cell cysts surrounded by pre-granulosa cells (red cells) recruited from the gonadal soma. At E13.5, PGCs in a developing ovary cease to proliferate and begin the process of meiosis – arresting at the prophase I. After birth, pre-granulosa cells invade the germ cell cysts. Regulated apoptosis of germ cells also contributes to the breakdown of the cysts. This results in the formation of primordial follicles comprised of a single oocyte surrounded by several pre- granulosa cells (Adapted from Bristol-Gould et al. 2006).

is leptotene, where double strand breaks (DSB) in the DNA induce the recruitment of repair

proteins, forming recombination foci as the cell tries to repair the breaks. It is then followed by the

pairing of homologous chromosomes as the regions carrying breaks pair up with complementary

regions from the homologous chromosome to provide a template for the repair machinery. This

step is called the zygotene stage and includes the formation of a synaptonemal complex responsible

for holding the synapsed chromosomes together. During the next step, pachytene, the

recombination foci continue to recruit more factors. At this stage, the interaction between the

homologous chromosomes at the sites of DSBs increases the chances of crossover events, with the

exchange of DNA fragments, as the cell tries to repair the initial DSBs. With the resolution of the

DNA damage and the progression to the diplotene stage, synaptonemal complexes are

disassembled, but homologous chromosomes remain tethered to each other through the sites of

recombination. Diplotene is also the stage at which the oogonial cells arrest their progression

through meiosis for weeks or months.

2) Post-natal stages of oogenesis

After their differentiation in-utero, both oocytes and pre-granulosa cells remain mitotically inactive throughout embryonic development, until birth. The second major wave of pre-granulosa

cell recruitment occurs right after birth (Mork, Maatouk et al. 2012). Developmentally

programmed apoptosis of oocytes within the germ cysts is then suggested to induce the breakdown

of clusters. While germ cell apoptosis begins in-utero, in the mouse, it peaks around post-natal day

(PD) 2, following extensive reorganization of mitochondria and endoplasmic reticulum (Pepling

and Spradling 2001, Monniaux, Clement et al. 2014). It is however unknown how the formation

and the breaking down of these cysts contributes exactly to the development of oocytes and the establishment of follicles in mice; some suggesting a role reminiscent of nurse cells in Drosophila,

whereby the cells destined to die transfer cytoplasmic factors, including mitochondria, nutrients and mRNA, to the surviving oocytes (Pepling and Spradling 2001). In the meantime, pre-granulosa

cells begin to intrude in the cyst, isolating and enveloping single oocytes to form primordial

follicles and establish the definitive ovarian reserve (Vainio, Heikkila et al. 1999, Trombly,

Woodruff et al. 2009, Xu and Gridley 2013, Vanorny, Prasasya et al. 2014) (Figure 3).

The subsequent post-natal phases of oogenesis can be further divided into two parts: growth and maturation. Shortly after birth, a subset of the primordial follicles established in utero is recruited and begins to grow. These oocytes, however, do not mature and eventually die off by atresia

(Hulas-Stasiak and Gawron 2011). Consecutive waves of growth then start after the onset of puberty, recruiting pools of follicles until the reserve is depleted, and in primates, menopause is initiated. During growth, oocytes increase in volume by more than 100-fold as they accumulate mRNA, protein and organelles, all the while staying at prophase I (Figure 3). In mice, it lasts approximately 21 days and occurs in parallel to important structural changes to the follicle, physiologically required for the sustenance of growth, maturation and ovulation. After the oocyte has reached its full size, maturation follows. It begins when the block at prophase I is removed, and the oocyte resumes meiosis. During this process, the oocyte discards half of its genetic material as it prepares to be fertilized.

A. Oocyte Growth

A.1. Primordial follicle recruitment, folliculogenesis and oocyte growth

When primordial follicles are recruited into the growth phase, the transition from squamous mitotically-arrested pre-granulosa cells to proliferating cuboidal granulosa cells signals the beginning of folliculogenesis. Their activation is followed by the proliferation of multiple layers of the surrounding somatic cells as the follicles transition from the primary stage with a single

layer of granulosa cells to the secondary stage with two or more layers of granulosa cells. As the

granulosa cells continue to proliferate, the accumulation of fluid within the follicle creates a cavity

called the antrum and the follicle becomes an antral follicle. This also results in the differentiation

of two populations of somatic cells: the mural granulosa cells in the outer layers lining the follicular

wall and the cumulus cells, in the inner layers of the follicle, closer in proximity to the oocyte. The

majority of oocyte growth occurs between the establishment of primary follicles and the

progression to the antral stage. As the two populations of granulosa cells continue to proliferate

and the antrum continues to grow, folliculogenesis concludes with the establishment of a Graafian

follicle with an enclosed oocyte ready to be ovulated (Figure 3).

The mechanisms regulating the transition from primordial to primary follicle are not fully understood. It is believed however, that this process is driven by the somatic cells of the follicle

(Hardy, Mora et al. 2018). MTOR is a conserved kinase known to regulate cell growth and

development by upregulating protein synthesis via transcription and translation, both of which

require a lot of energy. Accordingly, MTOR also stimulates ATP production by stimulating

mitochondrial activity (Albert and Hall 2015). In granulosa cells, mTORC1 is required for pre-

granulosa cell differentiation into granulosa cells, the initiation of oocyte growth and oocyte

survival (Zhang, Risal et al. 2014). However, what triggers the activation of the dormant follicles

remains poorlyunderstood. The correlation between follicle activation and a switch from

SMAD2/3 expression in the pre-granulosa cells to SMAD1/5/8 in the granulosa cells also suggest

a role for the transforming growth factor beta (TGF-β)/SMAD signalling pathway in early follicle

development (Fenwick, Mora et al. 2013; Hardy, Mora et al. 2018). In fact, BMP7 injection into

the ovarian bursa appears to enhance the transition from primordial to primary follicle (Lee, Otsuka

et al. 2001). After the initiation of folliculogenesis, granulosa cell survival and proliferation can

also be stimulated by factors from the extra-follicular somatic environment such as BMP4 (Nilsson and Skinner 2003). Interestingly, at later stages of folliculogenesis, the (TGF-β)/SMAD-dependent suppression of Ddit4l, an mTOR signaling inhibitor, is required for cumulus cell development and survival in post-antral follicles (Guo, Shi et al. 2016).

The mTORC1-dependent expression of KL plays a central role in granulosa cell-mediated activation of oocyte growth (Zhang, Risal et al. 2014). FOXO3A is a transcription factor typically associated with apoptosis and cell cycle arrest. In the primordial follicles, it localizes to the oocyte nucleus where it is involved in the repression of growth (Castrillon, Miao et al. 2003, Liu,

Rajareddy et al. 2007, Pelosi, Omari et al. 2013). Overexpression of FOXO3 promotes the expression of genes associated with non-growing oocytes, including Figla, Lhx8 and Clstn2 while repressing genes like Sohlh1, associated to the transition from primordial to primary follicle. This prevents primordial follicle recruitment and preserves the ovarian reserve (Pelosi, Omari et al.

2013). Conversely, repressing Foxo3 expression in the oocyte or disrupting the expression of its downstream target Lhx8, promotes primordial follicle recruitment and exhausts the ovarian reserve

(Choi, Ballow et al. 2008, Liu, Castrillon et al. 2013). c-KIT activation in oocytes by KL translocates FOXO3A out of the nucleus via an PI3K/AKT-phosphorylation-dependent pathway, effectively neutralizing its repressive activity and triggering oocyte growth (Liu, Rajareddy et al.

2007, Li, Kawamura et al. 2010, Saatcioglu, Cuevas et al. 2016).

Activation of mTOR is found upstream of both translation and transcription (Albert and Hall

2015). Over-activating mTOR or PI3K/AKT signaling, independently, promotes oocyte growth activation but impairing either pathway reveals significant redundancy and compensatory mechanisms, highlighting the importance of transcription and translation in growing oocytes

(Reddy, Liu et al. 2008, Jagarlamudi, Liu et al. 2009, Gorre, Adhikari et al. 2014). In fact, oocyte

Figure 3 – Folliculogenesis and oocyte growth. The oocyte grows within the context of a follicle. The accumulation of proteins, RNA and organelles like mitochondria contributes to a ~100-fold increase in volume. Dormant oocytes lay enclosed in primordial follicles, surrounded by a single layer of squamous granulosa cells and a basement membrane. Oocytes begin to grow with the activation of primordial follicles. First, follicles become cuboidal and proliferate as a single layer of granulosa cells in primary follicles and two or more layers in secondary follicles. Simultaneously, oocytes secrete zona pellucida glycoproteins forming a layer called the zona pellucida between itself and the granulosa cells. Theca cells are also recruited to the follicle on the other side of the basement membrane.

Follicles then develop a fluid-filled cavity known as an antrum as the granulosa cells continue to proliferate. In the resulting antral follicle, two populations of granulosa cells develop: mural cells and cumulus cells. In the Graafian follicle, the oocyte has reached its full size and is ready to be ovulated (Adapted from Clarke, 2018).

growth is accompanied by an overall upregulation of transcription, coinciding with the transition

from primordial to primary follicle (Fair, Hyttel et al. 1996). Furthermore, mRNAs in oocytes have unusually long half-lives and accumulate in growing oocytes (De Leon, Johnson et al. 1983, Ma,

Fukuda et al. 2015). The increase in RNA abundance is also paired with protein synthesis

(Eichenlaub-Ritter and Peschke 2002). The number of polyribosomes in the oocyte increases

during growth, resulting in the translation of part of the generated transcriptome (Bachvarova

1992, Fair, Hulshof et al. 1997). Growth, maturation and early embryogenesis are highly energy-

consuming processes. The accumulation of mitochondria in growing oocytes is critical for

oogenesis and embryogenesis (Wai, Ao et al. 2010, Mahrous, Yang et al. 2012). Fair et al. (1997)

accordingly note the presence of replicating mitochondria as early as the primary oocyte stage.

Therefore, the increase in oocyte size is due to many factors including the build up of transcripts

(~100pg), an increase in protein abundance and the accumulation of mitochondria (Moore and

Lintern-Moore 1974, Moore, Lintern-Moore et al. 1974, Moore and Lintern-Moore 1978, De

Leon, Johnson et al. 1983, Bachvarova, De Leon et al. 1985, Mahrous, Yang et al. 2012). The

expansion and proliferation of Golgi complexes and an increase in cytoplasmic vesicles, cortical

granules and lipid droplets also contribute to the oocyte expanding 100-fold in volume (Fair,

Hulshof et al. 1997) (Figure 3).

A.2. Granulosa cells support oocyte growth

Within the follicle, the oocyte is surrounded by granulosa cells. A basement membrane is also

deposited around the primordial follicle further cutting the germ cells off from the surrounding

environment. Meanwhile, oocytes are unable to meet their own metabolic requirements and those

of the early embryo (Biggers, Whittingham et al. 1967, Buccione, Cecconi et al. 1987). They

require a lot of energy in the form of ATP to sustain growth and set up early embryogenesis.

Accordingly, deficiencies in ATP are associated with impaired fertility and suboptimal

developmental efficiency (Van Blerkom, Davis et al. 1995, Igarashi, Takahashi et al. 2005). Yet,

oocytes are deprived of the glycolytic machinery necessary to process glucose as a source of ATP

(Donahue and Stern 1968, Leese and Barton 1985, Sugiura, Pendola et al. 2005). Fortunately,

granulosa cells compensate for this with the ability to metabolize glucose to generate pyruvate,

capable of diffusing to the oocyte (Donahue and Stern 1968, Sugiura, Pendola et al. 2005, Wang,

Chi et al. 2012). This is also supported by previous work determining that pyruvate is the main

source of energy in the oocyte and the early embryo (Biggers, Whittingham et al. 1967). Oocytes also outsource other metabolic functions to somatic cells, including nutrient and amino acid uptake and cholesterol biosynthesis, importing the final products to meet the molecular requirements for growth and maturation (Colonna, Cecconi et al. 1983, Colonna and Mangia 1983, Buccione,

Cecconi et al. 1987, Haghighat and Van Winkle 1990, Sugiura, Pendola et al. 2005, Su, Sugiura et al. 2008).

After the onset of growth, oocytes begin to secrete three glycoproteins in mice, four in primates

(ZP1, ZP2, ZP3, ZP4), that are then assembled into an extra-cellular matrix around the oocytes.

Known as the zona pellucida, it creates a physical barrier between the oocyte and granulosa cells.

However, in addition to paracrine signalling, physical contact is maintained throughout growth.

This is achieved via cytoplasmic extensions from the granulosa cell to the oocyte, establishing channels of communication between germ cell and soma. Most importantly, gap junctions, assembled where the tip of these transzonal projections (TZPs) meet the oocyte cell membrane, allow the diffusion of small molecules (e.g. pyruvate, amino acids, cGMP, etc.) as part of a process called metabolic coupling (Gilula, Reeves et al. 1972, Anderson and Albertini 1976, El-Hayek and

Clarke 2015).

Bi-directional communication between oocytes and granulosa cells is critical for both oogenesis

and folliculogenesis. In addition to providing the oocyte with the nutrients and the energy to

support its growth, granulosa cells also signal the oocyte to trigger the molecular processes that

regulate it. KITL-KIT signaling downstream of mTORC in granulosa cells is necessary for both

follicle activation and initiation of oocyte growth as described above (Zhang, Risal et al. 2014).

Moreover, granulosa cell activation and function are integral to the survival of the follicle.

Foxl2LacZ mutant pre-granulosa cells fail to differentiate into cuboidal granulosa cells in the

absence of functional FOXL2, and as a result the follicle dies by atresia (Schmidt, Ovitt et al.

2004). Reciprocally, oocytes play an important role in regulating folliculogenesis after its

initiation. Repression or enhancement of oocyte growth, hinders or stimulates follicular

development respectively (Castrillon, Miao et al. 2003, Liu, Rajareddy et al. 2007). More

specifically, expression of oocyte derived factors (ODFs: Gdf9, Bmp15 and Fgf8) work in a

paracrine fashion to stimulate granulosa cell proliferation and follicular development (Emori and

Sugiura 2014, Guo, Shi et al. 2016). Furthermore, TZP formation is significantly impaired with

GDF9 deprivation (El-Hayek, Yang et al. 2018). This emphasizes the need for bidirectional means

of communication between soma and germ cells during healthy folliculogenesis and heavily

implicates the oocyte in the indirect regulation of its own growth.

A.3. Meiotic and developmental competence

Many of the changes incurred during growth support the oocytes needs during this period. For

instances, the oocytes enclosed in primary follicle see an upregulation of factors like Gli3, Mrg1,

Tcfap2e and Obox5, thought to be involved in somatic-germ cell communication, follicle development and transcriptional activation, all of which required for growth to occur (Pan,

O'Brien M et al. 2005). The primary role of the increase in size and the accumulation of factors

22 however, is to prepare the oocyte molecularly and metabolically for maturation, fertilization and the initiation of embryogenesis (Reader, Stanton et al. 2017). In-vitro culture work, isolating oocytes form pre-antral follicles, has shown reduced rates of maturation despite the oocytes having completed 70-80% of growth (Szybek 1972). Without the molecular tools to resume meiosis, the oocyte remains arrested at the diplotene stage of prophase I. This meiotic competence promotes the resumption of meiosis and is acquired only at the early antral stage (Szybek 1972) (Figure 4).

Later, pre-implantation embryogenesis is entirely dependent on the conceptus. But before the zygote takes control of its own development, it relies on factors deposited in the egg during growth

(Dworkin and Dworkin-Rastl 1990). Therefore, the ability to successfully initiate embryogenesis is acquired at the end of growth and is referred to as developmental competence (Figure 4); if the oocyte resumes meiosis before accumulating these factors, early embryogenesis is then impaired

(Szybek 1972, Eppig and Schroeder 1989).

Accordingly, distinct changes in gene expression are associated with meiotic and developmental competence. Moreover, at the end of growth, before the onset of maturation, transcription in the oocyte is significantly downregulated and is undetectable in the fully grown oocyte (Clarke 2012).

Gene expression then, must occur during growth. The accumulation of RNA then becomes critical for the oocyte to overcome the lack of transcription during maturation and early embryogenesis.

In the early antral follicle and during the transition from secondary to early antral follicle, dynamic changes to the transcript profile of oocytes reflect the acquisition of meiotic and developmental competence, respectively (Pan, O'Brien M et al. 2005). For instance, the upregulation of Cdc2 and

Ccnb1, and their upstream regulators Cdc25c and Wee1b, occurs in oocytes of antral follicles

(Mitra and Schultz 1996, Kanatsu-Shinohara, Schultz et al. 2000). Cyclin B1 and CDK1, encoded

Figure 4 – Growth, meiotic and developmental competence. Oocyte growth requires the activation and the development of the follicle.

Oocytes reach 70-80% of their size inside antral follicles. When the follicle becomes a Graafian follicle, oocytes have reached their full- size and cease to grow. Transcription is active in growing oocytes but undetectable at the onset of maturation. It only resumes (in mice) in the late 2-cell embryo. In the oocyte, maturation is the resumption of the first meiotic division and the initiation of the second meiotic division. The ability to initiate and complete maturation is called meiotic competence and is acquired in growing oocytes around the antral follicle stage. In the absence of transcription, the initial steps of embryogenesis rely on molecular factors also acquired during growth. Conferring what is known as developmental competence, these factors are acquired after meiotic competence at the final stages of oocyte growth in the Graafian follicle. Once maturation is completed, the oocyte can be successfully fertilized and initiate

24 embryogenesis. by Ccnb1 and Cdc2 respectively, are cell cycle regulators known as the maturation-promoting factors (MPFs); accumulation of the Cdc2 transcripts during oocyte growth is critical for meiotic competence, and the repression of their expression significantly impairs meiotic resumption

(Adhikari, Zheng et al. 2012). Increases in the transcription of other genes involved in cell cycle, biosynthesis, and macromolecular metabolism are also detected and by the antral follicle stage, the oocyte possesses all the machinery necessary to bypass the arrest at prophase I (Pan, O'Brien M et al. 2005). Changes to the transcript profile are also observed between oocytes in early-antral follicles and oocytes in late antral or Graafian follicles, reflecting the acquisition of developmental competence. Factors involved in the processing of the male genome, removal of maternal RNA and protein, activation of the embryonic genome and early embryonic patterning events comprise developmental competence (O'Shea, Mehta et al. 2012). Moreover, many of the mRNAs accumulated during growth are silenced only to be translated later (Bachvarova 1985). The mechanisms regulating this process will be discussed in a latter next section.

Meanwhile, many other factors are accumulated in the form of proteins and their activity is tightly regulated to coordinate maturation, fertilization and early embryogenesis. Many kinases, phosphatases and other protein-modifying enzymes are therefore expressed in oocytes to regulate the activity of other proteins (Pan, O'Brien M et al. 2005). For instance, CDK1 phosphorylation is a central regulator of meiotic resumption and is modulated by the aforementioned Wee1B and

CDC25B (Gautier, Matsukawa et al. 1989, Adhikari and Liu 2014). Wee1B is a kinase that phosphorylates Tyr15 and inhibits CDK1 to uphold prophase I arrest (Han, Chen et al. 2005).

Meanwhile, CDC25B dephosphorylates Thr14 and Tyr15 and activates CDK1 to resume cell cycle progression (Lincoln, Wickramasinghe et al. 2002, Rudolph 2007). The timing of expression of some factors is also crucial for the adequate progression of oogenesis. Disrupting Wee1b

translation for example, can precociously initiate the resumption of meiosis in oocytes (Han, Chen

et al. 2005). Conversely in a Cdc25b -/- mutant, the injection of a wild type Cdc25b mRNA and

its translation in fully-grown oocytes is enough to initiate the resumption of meiosis (Lincoln,

Wickramasinghe et al. 2002). The translation of the RNA involved in meiotic and developmental

competence is also tightly regulated. Many transcripts are therefore silenced, stored or translated

in a stage-specific manner to coordinate the molecular events of maturation, fertilization and early

embryogenesis.

B. Oocyte maturation

B.1. Maturation: GVBD and chromosome segregation

Meiosis I is a reductional division where homologous chromosomes are segregated. While oocyte

meiosis is initiated in-utero, it only resumes post-natum, after growth completes. It consists of a

round of DNA replication (in-utero before the arrest at prophase I) and two consecutive rounds of

cell divisions where homologous chromosomes and then sister chromatids are segregated,

respectively. Oocyte maturation consist of the first meiotic division after the arrest at prophase I

is overcome. It begins with the breakdown of the nuclear envelope (also referred to as the germinal

vesicle breakdown – GVBD). Microtubule organizing centres (MTOCs), dispersed throughout the

ooplasm, then nucleate tubulin polymerization and microtubule formation. The resulting

microtubules are highly disorganized at first. They then progressively align, forming bipolar

spindles along the chromosome length in a chromosome-dependent manner (but independent of the centromere as in mitosis). Chromosomes are then aligned along the metaphase plate first using a chromatin-associated microtubule motor mechanism (Wignall and Villeneuve 2009, Bennabi,

Queguiner et al. 2018) and then using a kinetochore-dependent mechanism, after kinetochore activation (Brunet, Maria et al. 1999). The spindle is then moved to the periphery through

interactions between chromosomes and actin filaments (Yi, Rubinstein et al. 2013). The ensuing

asymmetric cytokinesis then completes meiosis I in oocytes. As a result, half of the chromosomes

are kept in the oocytes, retaining most of its pre-cleavage volume, and the other half is discarded

in the much smaller polar body. Immediately after the completion of meiosis I, meiosis II is

initiated. Spindle assembly and chromosome alignment along the second metaphase plate take

place in a similar fashion to meiosis I. However, meiosis II arrests at metaphase II. This arrest also

known as cytostasis persists for several hours or days, as the oocytes awaits fertilization to resume

and complete meiosis II. The formation of the first polar body after meiosis I with the alignment

of chromosomes along the second metaphase plate, marks the end of oocyte maturation and of

oogenesis (Figure 5).

B.2. Meiotic arrest

By the end of growth, a very complex gene expression profile and proteome has established the

molecular pathways needed to resume meiosis. Yet in the pre-ovulatory Graafian follicle, the fully- grown oocyte, now equipped with the tools to resume meiosis, must actively prevent it until the time of ovulation. The integrity of the follicle and the maintenance of physical communication between the oocyte and the cumulus cells are critical for the meiotic block at the end-stage of oocyte growth. When removed from the follicle, oocytes can autonomously resume meiosis

(Pincus and Enzmann 1935). Within the follicle, the C-type natriuretic peptide (NPPC) produced by mural granulosa cells binds and activates its receptor, the natriuretic peptide receptor 2 (NPR2), on mural and cumulus cells. NPR2 is a guanylyl cyclase, responsible for the production of cGMP by granulosa cells (Zhang, Su et al. 2010). With high levels accumulating in the somatic cells of the follicle, cGMP diffuses to the oocyte via the gap junctions at the tip of transzonal projections.

Oocytes also indirectly promote the production of cGMP. Secretion of oocyte-derived factors

GDF9, BMP15 and FGF8, act in a paracrine fashion to stimulate Npr2 expression in the follicle

and promote TZP projections towards the oocyte, thereby enhancing its own buildup of cGMP

(Zhang, Su et al. 2010, El-Hayek, Yang et al. 2018).

Ccnb1 and Cdc2, accumulate at the second half of oocyte growth and are the main drivers of oocyte

maturation (Polanski, Ledan et al. 1998, Brunet and Maro 2005, Pan, O'Brien M et al. 2005). Yet

before the onset of maturation, Ccnb1 translation is low, the little amount of Cyclin B1 expressed

is continuously degraded by the anaphase-promoting complex (APC/C-CDH1 encoded by Cdc20 and Fzr1) and CDK1 activity, and nuclear localization is prevented by WEE1B phosphorylation

(Han, Chen et al. 2005, Reis, Chang et al. 2006). Protein kinase A (PKA) activity, itself stimulated by high cAMP levels, in turn enhances WEE1B activity (Han and Conti 2006). The presence of high cGMP concentration in the oocyte inactivates the cAMP-degrading phosphodiesterase

PDE3A and the ensuing accumulation of cAMP keeps PKA active (Richard, Tsafriri et al. 2001,

Vaccari, Horner et al. 2008). Ultimately the activation of PKA inhibits Cyclin B1 and CDK1 activity in oocytes and prevents the resumption of meiosis.

B.3. Molecular triggers of meiosis

Maturation promoting factors (or MPFs) were first described in 1971 in frog eggs where the cytoplasm of oocytes hormonally pushed into maturation (by progesterone treatment) was injected into fully-grown oocytes arrested at prophase I. In doing so, the authors observed the resumption of meiosis in the injected oocytes (Matsui et al. 1971). This elegant set of experiments was one of the first to show that cytoplasmic factors, whose expression was induced by progesterone treatment of the follicle, were responsible for orchestrating both nuclear and cytoplasmic maturation, removing the molecular block at tprophase I and promoting entry into metaphase. It wasn’t until the 1980’s, however, that the link to cyclins and cyclin dependent kinases were made and Cyclin

B1 and CDK1 were identified as the main drivers of maturation (reviewed in Masui, 2001; Lohka

et al., 1988).

Physiologically, maturation is triggered by the luteinizing hormone (LH) surge. When LH binds

and activates its receptor LHR on the mural granulosa cells, it induces the loss of cGMP in the

cumulus cells and the oocyte. This is the result of its inhibitory effect on Nppc expression and the

activation of PDE5, both inhibiting the production and enhancing the degradation of cGMP,

respectively, and the desensitization of NPR2 (Vaccari, Weeks et al. 2009, Robinson, Zhang et al.

2012). The drop of cGMP in the cumulus cells reverses the concentration gradient of cGMP in the

oocyte-cumulus complex and as result, cGMP begins to diffuse out of the oocyte significantly

lowering its concentration in the germ cell. In the absence of cGMP, PDE3A is now free to degrade

cAMP in the oocyte. Alternatively, removing oocytes from post-antral follicles also triggers the

resumption of meiosis (Pincus and Enzmann 1935). Supplementing cumulus-enclosed oocytes

with NPPC however, maintains the meiotic arrest of the oocytes (Franciosi, Coticchio et al. 2014).

This suggests that removing cumulus-oocytes complexes for the follicle cuts them form their source of NPPC, the signal responsible for stimulating cGMP-production. The loss of PKA activity which follows the drop in cAMP, liberates Cyclin B1 and CDK1 from the repressive effects of their inhibitors APC/C-CDH1 and WEE1B, and allows for the stimulation of CDK1 activity by

CDC25B to initiate maturation and resume meiosis (Han and Conti 2006).

B.4. CDK1/Cyclin B1 orchestrate maturation

Upon their activation, CDK1 and Cyclin B1 orchestrate the cellular events of maturation. GVBD requires the phosphorylation of nuclear lamins (Heald and McKeon 1990, Peter, Nakagawa et al.

1990, Ward and Kirschner 1990). Regulators of the cell cycle are known to induce the

phosphorylation of the lamin proteins which in turn induces the depolymerization of lamins and

the disassembly of the nuclear envelope (Ward and Kirschner 1990, Peter, Heitlinger et al. 1991).

In fact, activation of CDK1 has been implicated upstream of lamin protein phosphorylation in various mammalian oocytes (Peter, Nakagawa et al. 1990, Adhikari, Zheng et al. 2012).

Accordingly, disruption of CDK1 activity with the inhibitor roscovitine, for instance, prevents

GVBD and the resumption of meiosis (Zhang, Zhang et al. 2017). In Xenopus oocytes, the

translation of c-mos leads to the initiation of a cascade of MAPK phosphorylations and alone can

trigger GVBD (Gebauer, Xu et al. 1994). In mice however, it is not required for the successful disassembly of the nuclear envelope (Verlhac, Kubiak et al. 1996).

MTOC activity is constrained to the proximity of the condensed chromosomes via a RanGTP

dependent pathway involving TPX2 (Targeting protein for Xenopus kinesin-like protein 2), where

TPX2 supports microtubule assembly and regulates their organization (Wittmann, Wilm et al.

2000, Gruss, Carazo-Salas et al. 2001, Hetzer, Gruss et al. 2002, Caudron, Bunt et al. 2005).

Interestingly, CDK1 has been shown to phosphorylate TPX2 at Thr72; phosphorylation at this site

is required for the TPX2 role in regulating microtubule polarity and organization (Shim, Perez de

Castro et al. 2015). CDK1/Cyclin B1 activity has also been linked to interactions between the spindle and chromosomes and is required for spindle-assembly checkpoint (Davydenko, Schultz

et al. 2013, Rattani, Vinod et al. 2014, Yoshida, Kaido et al. 2015).

Following the alignment of homologous chromosomes along the first metaphase plate,

CDK1/Cyclin B1 activity is transiently downregulated allowing the first meiotic division to

conclude. This is the result of Cyclin B1 ubiquitination and degradation by APC/C and the 26S

proteasome, respectively (Peter, Le Peuch et al. 2002). APC/C is itself activated after the release

from the inhibition by the spindle-assembly checkpoint (Lara-Gonzalez and Taylor 2012, Rattani,

Vinod et al. 2014); without Cyclin B1, CDK1 is ineffective at maintaining metaphase (Pines and

Hunter 1989, Hampl and Eppig 1995, Winston 1997). Once the homologous chromosomes have

been segregated and the polar body is extruded, Ccnb1 expression resumes, APC/C activity

recedes, and CDK1/Cyclin B1 activity drives the initiation of the second meiotic division

(Madgwick and Jones 2007). CDK1/Cyclin B1, again, drives spindle formation, alignment of the

chromosome on the metaphase plate and holds the oocytes at metaphase. However, this time, the

oocyte remains at metaphase II, awaiting fertilization.

C. Postnatal neo-oogenesis: a contentious and ongoing debate

The pool of primordial germ cells from which growing follicles are recruited is limited in most

mammalian species. With the number of total follicles available for folliculogenesis determined

by near the time of birth, the balance between the recruitment of primordial follicles and the atresia

of non-growing primordial follicles is therefore critical in determining the longevity of the

reproductive lifespan (Faddy, Telfer and Gosden, 1987; Monniaux et al, 2014). Some reports over

the last decade and half, however, have hypothesized the existence in mice and humans of a stem

cell population in the ovary capable of generating new germ cells and perhaps prolonging their

reproductive lifespans. The first reports of possible oogonial stem cells dates back to 2004, arguing

that the rates of atresia in the ovary challenge the models that posit a finite number of oocytes at

birth: in mice, between birth and puberty, a third of the oocyte reserve is reported as atretic, and

while the rate of atresia reduces at puberty, it appears to increase again with age, affecting as much

as 16% of oocytes in an adult ovary at any given time, decreasing dramatically the pool of

available oocytes with every cycle of oocyte growth (Johnson et al, 2004). The authors, from the

Tilly group, then proceed to show both the proliferation in the surface epithelium of oocyte-looking

cells, also positive for the germ cell marker mouse Vasa homologue (MVH), and cells entering

meiosis in adult ovaries, using the synaptonemal complex protein 3-SCP3 as a marker. They

conclude that proliferating female germ stem cells exist in the ovary and can enter meiosis postnatally to contribute to a replenishing pool of oocytes (Johnson et al, 2004). In 2012, Tilly and collaborators also reported the isolation of a putative germ stem cells from the adult murine and human ovaries, capable of assuming an oocyte-like fate and be incorporated in developing follicles (White et al, 2012). In the same study, White et al. (2012) also report the use of said oocytes to generate live embryos in mice.

It is important to point out however, that most of the work in support of oogonial stem cells have been carried out by a handful of groups, with little reproducibility outside of these labs. Many have remained skeptical of the original use of DDX4, a mostly cytoplasmic protein, for the isolation of oogonial stem cells through cell sorting assays, despite the more recent discovery of a cell-surface form of DDX4 (Johnson et al, 2004; White et al, 2012). Even in the testes, cell-surface DDX4- immunoreactive cells did not express other germ cell markers, such NANOS2, and DAZL despite expressing early primordial germ cell markers, such as PRDM1, IFITM3, and EPCAM, and could not generate germ cells (Kakiuchi et al, 2014). In direct contradiction to the work of the Tilly group, adult ovaries were found to express little to no early meiotic-specific transcripts in addition to the absence of mitotically active germ cells, making it unlikely for them to have sustained neo- oogenesis post-natally (Liu et al, 2007). Moreover, Lei and Spradling (2013) categorically reject the possibility of a proliferating population of germ stem cells in the adult oovary. In addition to showing the lack of oocyte proliferation, they demonstrate by lineage-labeling of single oocytes, that the population of oocytes established in-utero is enough to support oogenesis throughout the mouse reproductive lifespan, without the need of putative germ stem cells (Lei and Spradling,

2013).

Figure 5 – Oocyte maturation. Maturation begins with the breakdown of the nuclear envelope and the condensation of the chromosome.

The meiotic spindle then forms and holds the chromosomes, aligning them along the metaphase plate. The spindle then moves to the periphery and the first division segregates the homologous chromosomes, while asymmetric cytokinesis of the oocyte generates the small first polar body that contains half of the chromosomes. The chromosomes retained in the oocyte realign on a new metaphase plate as the oocyte awaits to be fertilized (Adapted from Clarke, 2018).

II – THE REGULATION OF GENE EXPRESSION IN THE OOCYTE

1) Gene expression

A. Transcription

Gene expression is at the heart of all cellular mechanisms and is therefore indispensable for cell identity, cell function and communication between cells and with their environment. It begins in the nucleus with the transcription of genes embedded in the DNA sequence. Briefly, RNA polymerase enzymes are recruited to genes where they scan the DNA sequences from the 3'-end to the 5'-end, while synthesizing a complementary single stranded RNA molecule from the 5'-end to the 3'-end. The resulting pre-messenger RNA (pre-mRNA) is rapidly capped by a 7- methylguanosine nucleotide at the nascent 5'-end, while it is being transcribed. At the end of transcription, the pre-mRNA is released via a cleavage event mediated by the cleavage and polyadenylation specific factor (CPSF) binding the hexa-nucleotide polyadenylation signal at the

3'-end of transcripts. In addition to providing the catalytic enzyme responsible for cleaving the

RNA (the endonuclease CPSF-75 subunit), the CPSF also recruits a polynucleotide adenylyltransferase, poly(A) polymerase (Mandel, Kaneko et al. 2006, Schonemann, Kuhn et al.

2014). A chain of adenosine monophosphates, known as a poly(A) tail, is then added at the 3'-end immediately after cleavage. Following this nuclear polyadenylation event, splicing of the pre- mRNA follows. In this process, the pre-mRNA sees the removal of the intronic regions of its sequence. Alternative splicing can also lead to transcriptomic variants with the inclusion of some exons and the exclusion of others. The resulting mRNA is then assisted out of the nucleus to the cytoplasm by a nuclear export machinery where it can be modified, translated, silenced, degraded or any combination of the latter (Figure 6).

Figure 6 Gene expression. (1) It begins in the nucleus with the transcription of genes embedded in the DNA. (2) The nascent messenger RNA is immediately capped at the 5'-end. Upon its release from the transcriptional machinery, it is then (3) polyadenylated and (4) spliced, removing the introns. (5) The mRNA is then exported out of the nucleus and into the cytoplasm where it can be

(6) translated, (7) deadenylated (8) decapped, or (9) degraded. Recent evidence has demonstrated that both deadenylation and decapping can be reversed. As such, certain transcripts can be

*readenylated or others **recapped. The mRNA often binds RNA-binding proteins and other

RNAs nucleating ribonucleoprotein particles that regulate different aspects of RNA metabolism.

B. Translation

Translation is initiated by elements in the 5'-untranslated region (UTR). The scanning model posits that the 40S ribosomal subunit is recruited along with eIF2 (in a 40s:eIF2:Met-tRNAi:GTP

complex) to the 5'UTR through interactions with the translation initiation complex eIF4F,

composed of eIF4E, eIF4A and eIF4G (Kozak 1989). The latter forms on the 5'-methyl cap when

the translation initiation factor eIF4E binds the methyl cap and nucleates the assembly of the

complex. eIF4F is involved in unwinding the RNA and facilitating 40s:eIF2:Met-tRNAi:GTP

binding using (Andreou and Klostermeier 2013). After binding to the mRNA, the 40s:eIF2:Met-

tRNAi:GTP complex moves towards the 3'-end, scanning the RNA sequence and stopping when it detects an AUG start codon (Kozak 1989). Alternatively, in transcripts where the 5'-UTR is inundated with sequence motifs and structures that hinder translation initiation, transcripts can opt for internal initiation cues using internal ribosomal entry sites (IRES). In this scenario, the IRES can also recruit the eIF4F complex, in a eIF4A-ATP-dependent manner, to enable 40s:eIF2:Met- tRNAi:GTP binding and initiate translation (Lopez de Quinto, Lafuente et al. 2001).

Once the 40s:eIF2:Met-tRNAi:GTP complex recognizes the start codon, the complementary base pair interactions between the Met-tRNAi and the start codon sets off a chain of GTP-hydrolysis events that leads to the recruitment of the 60S ribosomal subunit, resulting in a 80S ribosomal complex. The 80s ribosome is a ternary complex comprised of ribosomal RNA (rRNA) and ribosomal proteins that support its interactions with the various factors of the translational machinery throughout initiation, elongation and termination, and carry out its function. This includes the three sequential compartments which house the aminoacyl tRNA-codon interaction

(A, P and E sites) and the RNA-rich peptidyl transferase center (PTC) where the peptide bond between amino acids is catalyzed (Figure 7). Upon translation initiation, the Met-tRNAi is in the

P site with the start codon, leaving the next codon in the A (aminoacyl) site ready to interact with

a cognate aminoacyl tRNA to continue translation.

During translation elongation the ribosome travels along the mRNA (5' to 3'), scanning codon after

codon and pairing them up with the corresponding aminoacyl tRNAs. When aminoacyl tRNAs are brought into the ribosome, at the A site, by the translation elongation factor eEF1A, they must successfully base-pair with the current codon in the A site while the previous aminoacyl tRNA-

codon pairing sits in the P site. eEF1A GTP hydrolysis induces conformational changes, that move

the new aminoacyl tRNA in a hybrid state between the A site and the P site while moving the

previous aminoacyl tRNA between the P site and the E site. At the same time, a peptide bond

between the two amino acids of the two aminoacyl tRNA's is established, elongating the nascent

polypeptide chain. Another elongation factor, eEF2, then completes the GTP-dependent

translocation of the two tRNA's to the E and P sites, respectively. This also slides the ribosome

unto the next codon, which will now occupy the emptied A site. The peptidyl-transferase reaction follows, breaking the bond between the amino acid and the rest of the tRNA at the E site before

separately releasing the tRNA (without an aminoacyl group) and starting the process all over again.

Translation ends when the ribosome encounters a stop codon (UAA, UGA or UAG) and a process

known as translation termination ensues. The latter requires two translation termination factors,

eRF1 and eRF3, which work in a GTP-dependent manner to terminate the elongation of the forming polypeptide. eRF3 is a tRNA-like protein and eRF1 acts like eEF1A in facilitating the

eRF1:eRF3:GTP ternary complex interactions with the ribosome. When eRF1 confirms the

presence of a stop-codon in the A site, the stabilization of the interactions with the ribosome

prompts GTP hydrolysis and eRF3 release ensues. At the same time the eRF1 middle domain shifts

into the PTC of the ribosome and promotes peptide release.

a. a.

a.a.

a.a. a.a.

Figure 7 – Ribosome structure. Ribosomes made of the (small) 40S ribosomal subunit and the

(large) 60S ribosomal subunit: complexes of ribosomal RNA (rRNA) and ribosomal proteins.

When the 40S and 60S come together, changes in their individual composition result in the formation of a single 80S ribosomal complex. The 80s ribosome catalyzes translation. It has three major compartments. Incoming aminoacyl tRNAs are recruited to the ribosome at the A where

(aminoacyl) site. Only a cognate tRNA with an anti-codon that complements the codon at the A- site will enter and stably occupy the A-site. In the P (peptidyl) site, the corresponding cognate tRNA is bound to the downstream codon and its aminoacyl group is linked to the growing polypeptide chain. The peptidyl transferase center (PTC) then creates a peptidyl bond between the elongating polypeptide chain and the amino acid group of the tRNA in the A-site. While the tRNA in the P site is translocated to the E (exit) site, it is deacylated and then released from the ribosome, through the E site (Adapted from Frank, 2003).

2) Post-transcriptional mechanisms of translational regulation

Translation is a very complex and energy-dependent process. Translation initiation, elongation and termination are critical junctures, susceptible to regulation (Figure 8). Naturally, the stability and

the degradation of transcripts will also significantly modulate their translational activity.

Moreover, adding, modifying or removing molecular and structural marks on transcripts also

provide additional layers of translational regulation. The 5'-methyl cap and the poly(A) tail are

particularly susceptible to change. These RNA processing events are often mediated by RNA-

binding proteins and are tightly linked to RNA metabolism. More recent is the evidence of an important role for the subcellular localization of mRNAs, in conjunction with the composition of

these cytoplasmic and nuclear microenvironments, in the management, modification and

translational activity of RNA molecules. Autoregulatory elements, as well as the availability and

the nature of the interacting factors, orchestrate the processes and complex molecular cascades that

manage all aspects of the mRNA metabolism.

A. The role of the 5'-methyl cap and the poly(A) tail in translational regulation

A.1. RNA stability

The rate of translation is first and foremost regulated by transcript levels. For any given gene, the

more mRNA available, the higher the rate of protein expression. Establishing an equilibrium

between the rate of RNA synthesis and the rate of RNA decay can therefore modulate translational

output significantly. Accordingly, RNA degradation has been shown to significantly impair

translational output. Knocking down specific transcripts downregulates the expression of the

corresponding protein (Shyu, Wilkinson et al. 2008). Degradation typically begins with the

deadenylation of the poly (A) tail. This is then believed to lead to the removal of the 5'-methyl cap,

leaving transcripts vulnerable to 5' to 3' exonucleases like XRN1 (Chen and Shyu 2011). 40

The 5'-methyl cap and the poly(A) tail protect the RNA from 5' to 3' and 3' to 5' exonuclease activity, respectively. The cap protects transcripts from XRN1 activity, which is essential for

successful 5' to 3' degradation (Hsu and Stevens 1993). There are two decapping proteins: DCP1

and DCP2 work in concert within a multi-subunit complex. DCP2 holds the enzymatic NUDIX

motif responsible for catalyzing the decapping process (Wang, Jiao et al. 2002). The cap is also

susceptible to modifications that can modulate its function. For instance, the methylation of the

cap is required for DCP1/ DCP2-mediated decapping (LaGrandeur and Parker 1998, Wang, Jiao

et al. 2002). Members of the poly(A) binding protein (PABP) family bind the poly(A) tail and

prevent 3' to 5' exoribonuclease shortening of the poly(A) tail and degradation or decapping and

5' to 3' degradation (Tucker, Staples et al. 2002, Vazquez-Pianzola, Urlaub et al. 2011). Shortening

of the poly(A) tail is therefore important in regulating the stability of transcripts and will be

discussed in a later section.

Degradation can take place both in the nucleus and in the cytoplasm and is a tightly regulated

process. Intrinsic sequence elements that play an important role in the recruitment of RNA-binding

proteins also play an important role in regulating RNA stability and are discussed at length in a

later section (Day and Tuite 1998). Deletion or insertion of certain elements can significantly

impact the half-life of certain transcripts. The AU- rich element (ARE) for instance, when inserted

at the 3'-end of stable transcript, can significantly destabilize them and impair their expression

(Shaw and Kamen 1986). Conversely, the recruitment of IRE-binding proteins (IREBPs) in low

iron conditions, to iron-response elements (IREs), can protect the transferrin receptor mRNA (TfR)

from endonucleolytic cleavage (Binder et al. 1994) (Figure 9).

A.2. Translational rate

Translation is also regulated during initiation, elongation and termination. The eIF4F complex,

bound to the 5'-methyl cap mediates the initiation of translation. Consequently, many of the

subunits of the complex can regulate translation initiation and efficiency. eIF4E for instance is the

subunit responsible for binding the 5'-methyl cap. Remarkably, the phosphorylation of eIF4E

increases binding of the cap and binding to the rest of the complex 3 to 4-fold and enhances protein

synthesis (Goss, Woodley et al. 1987, Minich, Balasta et al. 1994, Rhoads, Joshi et al. 1994).

Moreover, low eIF4E concentrations can limit the rates of translation (Hiremath, Webb et al. 1985,

Duncan, Milburn et al. 1987). Similarly, eIF2 of the 40s:eIF2:Met-tRNAi:GTP complex is recycled

for translation initiation (Goss, Parkhurst et al. 1984). Phosphorylation of the eIF2α prevents the

exchange of GDP for GTP, impedes eIF2 recycling and therefore impairs translation (Shi, Vattem

et al. 1998, Harding, Zhang et al. 2000).

Furthermore, translation termination is often coupled with initiation, optimizing translational

efficiency (Figure 8). The ATP-binding cassette protein E1 (ABCE1) potentially plays a

significant role in this process. First, it enhances peptidyl tRNA hydrolysis during termination and

participates in the dissociation of the 40S and 60S subunits (Shoemaker and Green 2011). Second,

ABCE1 also interacts with components of the initiation complex machinery therefore linking reinitiation to recycling (Schuller and Green 2017). Circularization of transcripts in a closed-loop

conformation facilitates this process by positioning the ribosomes terminating translation near the cap and the translation initiation complex, to be recruited right away. The closed-loop

conformation results from the interaction between the cap structure and the poly(A) tail and has a

synergistic effect on the rate of translation (Gallie 1991, Wells, Hillner et al. 1998). Long poly(A)

tails can bind the poly(A)-binding protein (PABP) family more easily, often recruiting multiple

Figure 8 – Translation initiation, Elongation, Termination and ribosomal recycling. Translation initiation starts in the 5'-UTR where the eIF4E of the translation initiation complex, bound to the 5'-methyl cap, recruits other translational initiation factors (eIF) and the small 40S ribosomal subunit. The eIF-40S complex then begins to scan the mRNA. When a start codon (green codon) is detected the large 60S subunit joins the

40S subunit and translation begins. The ribosome the translates the open reading frame, decoding codon after codon (pink, blue and purple codons) and recruiting cognate tRNA and the corresponding amino acid resulting in the elongation of the nascent polypeptide. When the ribosome stumbles upon a stop codon (red codon), the ribosome recruits eRF1, catalyzing the hydrolysis of the release of peptidyl chain with the help of ABCE1. ABCE1 also enhance the separation of the 40S and 60S subunits for recycling. eIF4G bind both the PABP, bound to the poly(A) tail, and eIF4E circularizing the mRNA in a closed loop conformation. This conformation allows the recycled subunits to be used right away and re-initiate translation. ABCE-1 is believed to play a role in the re-initiation process (Adapted from Schuller, 2017).

proteins (Kuhn and Pieler 1996, Melo, Dhalia et al. 2003). The yeast PABP family member, Pab1,

interacts with eIF4G of the translation initiation complex, to promote the closed-loop conformation

of transcripts and enhance translation (Beilharz and Preiss 2007, Archer, Shirokikh et al. 2015).

Therefore, long poly(A) tails are associated with active translation and short poly(A) tails are

associated with translational silence or degradation.

A.3. Deadenylation

The shortening of the poly(A) tail is a process known as deadenylation and is associated with

various translational inhibitory mechanisms (Shyu, Belasco et al. 1991, Braun, Huntzinger et al.

2013). Deadenylases are 3' to 5' exoribonucleases with an affinity for poly(A) chains. There are

two main types of deadenylases, which differ in their structure (Yan 2014). Endonuclease-

exonuclease-phosphatase (EEP)-type deadenylases hold in their active site, five conserved motifs

with highly conserved residues: NEDDH (Asn-Glu-Asp-Asp-His). DEDD deadenylases have in

their active site four conserved amino acid residues, DEDD (Asp-Glu-Asp-Asp), after which they are named. The enzymatic activity associated with the CCR4-NOT complex however, is typically the most important deadenylase activity at least in murine cells (Yamashita, Chang et al. 2005).

Moreover, CCR4-NOT deadenylation has been reported in the cytoplasm of cells; where mRNA silencing and storage occurs in oocytes (Yamashita, Chang et al. 2005).

The complex is multifunctional and is comprised of multiple subunits. At any time, the complex binds a DEED deadenylase, either CNOT7 or CNOT8, which in turn recruits an EEP deadenylase,

CNOT6 or CNOT6L (Lau, Kolkman et al. 2009, Doidge, Mittal et al. 2012). Other subunits of the

CCR4-NOT complex play various roles ranging from pre-transcriptional gene expression regulation to post-translational modifications, earning the complex the nickname of ‘control freak’

(Miller and Reese 2012). The CCR4-NOT complex is typically recruited along with another

complex containing deadenylase proteins PAN2 and PAN3 (Yamashita, Chang et al. 2005,

Doidge, Mittal et al. 2012). The role of the latter is still unclear, but it has been proposed that the

PAN2-PAN3 complex is responsible for initiating the deadenylation of mRNAs allowing the

CCR4-NOT complex to then come in and complete the digestion of the tail (Yamashita, Chang et

al. 2005).

B. Cis-elements and the recruitment of trans-factors

B.1. The untranslated regions, the open-reading frame

Sequence elements within the transcript also regulate initiation and elongation and therefore the

rate of translation and will be discussed later. Most transcripts comprise an open reading frame

(ORF) flanked by two untranslated regions (UTRs) (Figure 9). The ORF is delineated by the start

codon at the 3'-end and the stop codon at the 5'-end and corresponds to the part of the transcript that is translated into a protein. The 5'-UTR and 3'-UTR precede and follow the ORF respectively.

While all three elements play a significant role in modulating translational output and RNA metabolism many of the regulatory elements are concentrated in the 5'- and 3'-UTRs.

The 5'-UTR is a major regulator of translation. Arrick et al. (1994) attributed the differences in expression levels of two TGFβ3 mRNA variants to the truncation of the 5'-UTR in one of the isoforms: the shorter transcript, in the absence of unidentified repressive elements, was expressed more efficiently. Conversely, the insertion of synthetic structures like hairpins can modulate the translational activity of transcripts in Xenopus oocytes (Fu, Ye et al. 1991). The 5'-UTR is also the site of the addition of the methyl cap, with which it coordinates the formation of the preinitiation complex, the recruitment of the 40s:eIF2:Met-tRNAi:GTP complex and other translational factors

to promote translation initiation (Furuichi, Morgan et al. 1975, Merrick 2004) .

The regulatory elements of the 3'-UTR, like the 5'-UTR, are embedded within the sequence and

the secondary and tertiary structures of the region. The removal of the 3'-UTR can therefore disturb the regulation of translation and have deleterious effects. This is the case in some paroxysmal nocturnal hemoglobinuria patients where chromosomal translocations within the Hmga2 oncogene remove the 3'-UTR of the corresponding transcript. Because of this truncation, essential regulatory elements imbedded in the 3'-UTR sequence are no longer capable of recruiting and binding repressive elements and HMGA2 is overexpressed and contributes to the development of the disease (Young and Narita 2007, Ikeda, Mason et al. 2011).

B.2. Sequence-mediated regulation

B.2.1. Sequence motifs and 3D structure

Sequence motifs have been shown to regulate different facets of RNA translation. They modulate both the 3-D structure of the transcript and its interactions with other types of RNA and RNA- binding proteins (Figure 9). These interactions are known to regulate translation and many other

RNA processing events. The sequence elements are at the very the basis of these interactions and their specificity. They therefore play an important regulatory role in the metabolism of RNA. Many such elements, dispersed throughout the transcript, regulate translation initiation, elongation, termination, output and RNA processing events. For instance, the cytoplasmic polyadenylation element (CPE - UUUUUAU) and the polyadenylation signal (PAS – canonically AAUAAA) work in tandem to mediate the polyadenylation of transcripts in the cytoplasm. The CPE-binding protein and the CPSF recognize and bind the CPE and the PAS, respectively, and nucleate the formation of a complex that includes the cytoplasmic poly(A) polymerase, Gld2, and the deadenylase PARN in Xenopus oocytes (Kim and Richter 2006). In different conditions, the complex promotes the

47 activity of Gld2 or PARN, while repressing the activity of the other, to elongate or shorten the poly(A) tail, respectively.

During translation, after the start signal is recognized, the mRNA is read by the recognition of sequential nucleotide triplets, referred above to as codons. Sequential combinations of three nucleotide are recognized by complementary aminoacyl tRNAs attached to specific amino acids. tRNAs specifically interact with the corresponding codons and drive the enzymatic activity of the scanning ribosome through the complementary base pairing of their nucleotide triplet in the anti- codon loop with the corresponding codon. It is through this recognition of codons and the complementary interactions with tRNAs that translation leads to the synthesis of sequence specific polypeptide chains (proteins). The start signal consists of an AUG codon, recognized by the Met- tRNAi. Coincidentally, the presence of upstream AUG (uAUG) elements preceding the authentic

AUG start codon in the 5'-UTR can significantly impair translation initiation by either diverting the resources away from the open reading frame of interest or by stalling 80S activity and creating a ribosomal blockade (Morris and Geballe 2000, Meijer and Thomas 2002).

There is a total of 61 possible nucleotide triplets (in addition to the stop codons discussed above) but only 21 amino acids. There is an important amount of redundancy in the pairing of some codons: synonymous codons recruit the same amino acids. Yet, synonymous codons are not created equal despite their redundancy; some nucleotide triplets are more commonly used than others to code for specific amino acids: this is known as codon bias (Grantham, Gautier et al. 1980,

Grantham, Gautier et al. 1980, Klumpp, Dong et al. 2012). In fact, more recent work has shown that different codons can affect the pace of ribosome scanning along the mRNA thereby modulating the rate of elongation and translational output. More specifically, codon bias has been linked to ribosomal load and protein abundance (Klumpp, Dong et al. 2012). Different codons

modulate translation elongation due to the corresponding tRNA availability: common codons have

higher chances of being paired to their complementary tRNAs which are present in high

concentrations. Conversely, the tRNA of rare codons are less common and slow down translation

elongation as the ribosome stalls to pair codons to the proper tRNAs (Pop, Rouskin et al. 2014,

Quax, Claassens et al. 2015). As protein folding occurs concurrently with translation, changing the pace of translation has also been linked to the regulation of protein folding, thereby explaining how codon bias can also regulate translational output in other ways (Zhang, Hubalewska et al.

2009, Hu and Coller 2012, Hu, Wang et al. 2013). In early Zebrafish embryos, non-optimal codon

usage has been shown to promote poly(A) tail shortening and degradation of maternal transcripts

during the maternal to zygote transition (Bazzini, Del Viso et al. 2016).

Despite being single-stranded, RNA molecules can often form complex 3-dimensional structures.

This requires complementary cis sequences where intramolecular base pairings can fold the RNA

strand. These secondary and tertiary structures are also important regulators of translation. They

can play critical roles in promoting or hindering translation initiation (Pelletier and Sonenberg

1988, Fu, Ye et al. 1991). Hairpins and the complex tertiary structures of IRES described above,

for instance, can recruit and use all the subunits of the translation initiation complex without the

cap-binding element eIF4E, to initiate translation (Le and Maizel 1997, Meijer and Thomas 2002).

In the ORF, secondary and tertiary structures can physically hinder ribosomal scanning, slowing

down elongation and therefore impair the rate and efficiency of translation as well as protein folding (Mauger, Siegfried et al. 2013).

Complex interplay between cis-elements also modulate their regulatory role; their effect on translation and their mode of action is contextual. The presence of stem-loop structures near the

5'-7-methylguanosisne cap prevents the recruitment of the preinitiation complex and hinders

translation initiation while similar structures downstream impair translation by obstructing the

ribosome associated helicase activity and RNA unwinding (Cazzola and Skoda 2000, Wang and

Rothnagel 2004). Interestingly, the presence of hairpin structures within 14nt downstream of an

AUG codon slows down ribosome scanning but enhances interaction with the codon and promotes

translation initiation of the respective ORF (Kozak 1990). Otherwise, the helicase activity of the

subunit eIF4A is enhanced by eIF4G within the eIF4F complex to remodel complex RNA structures and change the topography of the transcripts thereby facilitating the ribosomal movement along the transcript and translation initiation (Marintchev, Edmonds et al. 2009, Jin,

Rajabi et al. 2013, Marintchev 2013).

B.2.2. RNAi

Translation is also subject to regulation by small non-coding RNA (sncRNA) through a process known as RNA interference (RNAi). Three main classes are described in animal cells (reviewed in (Carthew and Sontheimer 2009). Piwi-interacting RNA (piRNA) repress the expression of retrotransposons via a PIWI and MILI-dependent mechanism and protect the integrity of the genome (Russell, Stalker et al. 2017). micro-RNA (miRNA) and short interfering RNA (siRNA) are more broadly found in most eukaryotes (Shabalina and Koonin 2008). miRNA biogenesis begins with the transcription of primary miRNA (pri-miRNA) from miRNA genes by RNA polymerase II. The pri-RNA possess complementary sequences that induce the formation of a single or multiple hairpin. The hairpins are then trimmed by a microprocessor complex with the

RNase III type protein, Drosha, resulting in ~70 nucleotides long (precursor) pre-miRNA hairpins

(Slezak-Prochazka, Durmus et al. 2010). siRNA derive from double stranded RNA (dsRNA) that originate from complementary interactions between two RNA strands (endogenous siRNA) or from exogenous dsRNA

(exogenous siRNA) (Provost, Dishart et al. 2002). In the cytoplasm, the double stranded nature of both pre-miRNA hairpins and dsRNA is recognized and cleaved by the DICER complex yielding two small RNA strands (~20 nucleotides in size), one of which or the both of which qualify as miRNAs or siRNA, respectively. Both miRNA and siRNA interact with Argonaute proteins

(AGO1 and AGO2) and nucleate the formation of the RNA-induced silencing complex as they are being processed (Wang, Sheng et al. 2008). Seed sequences imbedded in the mRNA targets are then recognized by complementary base pair interactions with either the miRNA or the siRNA

(Wang, Juranek et al. 2008). This targets the activity of the RISC complex, with the sncRNA and the Argonaute protein, to the target mRNA (Figure 9) (Carthew and Sontheimer 2009). The RISC complex typically mediates the shortening of the poly(A) tail, the repression of translation or the degradation of the transcripts (Mishima, Fukao et al. 2012, Braun, Huntzinger et al. 2013).

Physiologically, RNAi is often involved in developmental processes. Lee et al. identified the first miRNA, lin-4, targeting lin-14 and repressing LIN-14 expression in C. elegans (Lee, Feinbaum et al. 1993). Previous work had shown that the loss of lin-4 expression in lin-4 loss-of-function mutants significantly disrupts late stage cell lineage patterning in the same way ectopic expression of LIN-14 in lin-14 gain-of-function mutants affects C. elegans development (Chalfie, Horvitz et al. 1981, Ambros and Horvitz 1987). The loss of Drosha and DICER or disrupting sncRNA biogenesis is linked to gene expression dysregulation, impaired development, infertility and in the case of total knock-outs, embryonic lethality (Hong, Lee et al. 2000, Bernstein, Kim et al. 2003,

Wu, Song et al. 2012).

Figure 9 – Cis-regulatory elements. Three major regions comprise messenger RNAs. The open reading frame (ORF) is delineated by the start codon at the 5'-end and the stop codon at the 3'-end. It is therefore the region of the transcript which is translated. The ORF is flanked by two untranslated regions (UTRs): the 5'-UTR at the 5'-end and the 3'-UTR at the 3'-end. Both UTRs are inundated with regulatory elements imbedded in the sequence of the transcripts. Certain elements can bind complementary sncRNA and the associated

RISC complex to regulate poly(A) tail length and translational activity. Some motif elements are directly recognized by RNA-binding proteins (RBPs). Other motifs hold complementary sequences that bind each other creating secondary and tertiary structures that can also be recognized by RBPs. RBPs can in turn modulate the different facets of RNA metabolism.

C. RNA localization and microenvironment (RNPs)

C.1. Characteristics

As described above, transcripts interact with other forms of RNA and RNA-binding proteins. Both in the nucleus and in the cytoplasm, these interactions often nucleate more complex structures known as ribonucleoprotein particles (RNPs). This creates membrane-less organelles with specialized micro-environments that regulate RNA metabolism and translation, as well as various

RNA processing, transport and cellular localization mechanisms. In fact, throughout most of their lifespan, mRNAs reside within RNPs. The congregation of transcripts and multiple factors is believed to facilitate RNA-protein interactions and optimize RNA metabolism. Yet, with few functional assays reported, it is still unclear what exactly is the role of these different complexes.

Most studies postulate roles based on their molecular content. For instance, RNPs with protein involved in RNA turnover are believed to play a role in degradation (Sheth and Parker 2003,

Cougot, Babajko et al. 2004); the polyribosome, where multiple ribosomes are bound to a single mRNA molecule, simultaneously scanning and translating the transcript enhances translation (Qin and Fredrick 2013). Both processing bodies (P-bodies) and stress granules result from loss of translation (Kimball, Horetsky et al. 2003, Mazroui, Sukarieh et al. 2006, Eulalio, Behm-Ansmant et al. 2007). P-bodies are generally the site of physiological RNA silencing, active decapping and

RNA degradation (Sheth and Parker 2003, Cougot, Babajko et al. 2004, Eulalio, Behm-Ansmant et al. 2007). Accordingly, proteins involved in deadenylation, decapping and exonuclease activity are all commonly found in P-bodies (Buchan and Parker 2009). Stress granules on the other hand are believed to protect the transcriptome of cells under duress (Noble, Allen et al. 2008, Schisa

2012). Thus, while factors associated to translation regulation are present in stress granules,

deadenylases, decapping enzymes and exonucleases are typically excluded (Buchan and Parker

2009).

C.2. Formation

The agglomeration of associated factors in RNPs relies on the interactions between RNA

molecules and RNA-binding proteins as well as the interaction between proteins. More

specifically, many RNP-associated proteins house low-complexity domains (LCDs) like the

serine-rich domain in the C. elegans MEG1 and MEG3 (Wang, Smith et al. 2014), intrinsically

disordered regions (IDRs) like the arginine/glycine rich N-terminal of the C. elegans LAF-1

(Elbaum-Garfinkle, Kim et al. 2015) and prion-like domains (PLDs) like the glutamine/asparagine

(Q/N)-rich domain in many of the factors found in P-bodies including the yeast Lsm4p, Ccr4p,

Pop2p and Dhh1p (Reijns, Alexander et al. 2008). Such motifs promote protein-protein

interactions and are critical to the formation of RNPs (Reijns, Alexander et al. 2008, Wang, Smith

et al. 2014, Elbaum-Garfinkle, Kim et al. 2015). The agglomeration of RNP associated proteins and RNA use a mechanism by the name of liquid-liquid phase separation (LLPS), to sequester these components out of the cytosol and into the RNP (Brangwynne, Eckmann et al. 2009, Lin,

Protter et al. 2015). LLPS is the localized condensation and dissolution of liquid droplets within an heterogenous liquid, similar to when oil droplets form in a mixture of water and oil; except more dynamic and reversible (Figure 10). The process is regulated by the modulation of protein-

protein interactions through LCDs, IDRs and PLDs, protein abundance and concentration, salt

concentration and temperature in the cell (Hyman, Weber et al. 2014, Lin, Protter et al. 2015,

Protter, Rao et al. 2018).

Owing to the absence of membranes, RNPs are highly dynamic structures with a constant in- and

outflow of proteins and RNA (calculated exchange rates of granule components via FRAP from

multiple studies summarized in Buchan and Parker, 2009). Transcripts are commonly found within

RNPs, suggesting a constant exchange in RNA and proteins with the environment and between different RNPs. In fact, many RNPs form at the expense of other types: for instance, polyribosomes components impair the formation of P-bodies (Weidner, Wang et al. 2014). The link between the various types of RNPs represents more of a spectrum rather than a clearly-defined set of differences: many characteristics and components are shared across the board. In C. elegans, general components of translational inhibition like Xrn1, Ago, TTP, Rap55, BRF1 and Lin28 are detected in both P-bodies and stress granules of oocytes (Buchan and Parker 2009). Moreover, P- body formation often precedes stress granule formation and the two types of RNPs can be seen physically interacting. It has been suggested that P-bodies can morph into stress granule under sustained stress, although evidence also support independent pathways of formation (Kedersha,

Stoecklin et al. 2005, Buchan, Muhlrad et al. 2008, Mollet, Cougot et al. 2008, Ohn, Kedersha et al. 2008).

The dissolution and condensation of RNPs can also be regulated by protein modifications like

phosphorylation, glycosylation, acetylation and ubiquitination (Buchan and Parker 2009). For instance, the kinase MBK-2/DYRK phosphorylation and the PP2A phosphatase dephosphorylation of the MEG1 and MEG3 serine-rich LCDs, appears to impair and enhance protein-protein interactions, respectively, to modulate RNP formation and disassembly (Wang,

Smith et al. 2014). Similarly, HuR, hnRNP C1 and C2 binding of a 5'-UTR-U-rich sequence regulates RNP formation and translation of p27 mRNA in a cell cycle-dependent manner (Millard,

Vidal et al. 2000). Additionally, structures such as microtubules also regulate the formation of certain RNPs. Stress granules for instance appear to require microtubule as well as retrograde and anterograde transport to assemble and disassemble respectively (Loschi, Leishman et al. 2009).

Mixing and Demixing

Figure 10 – Liquid-liquid phase separation. In a heterogenous liquid (left panel), different components interact with one another. If the interactions between the different components are random and equal, then the liquid remains an heterogenous mix. If, however, certain interactions are favored over others, then the liquid demixes, and a liquid droplet forms within the liquid. This results in a rapid concentration of certain components within the newly formed compartment (right panel). This process is highly dynamic and reversible.

Unlike organelles, these foci do not form membranes, allowing the free diffusion of components in and out of the droplet. The properties of each component, their concentration, the temperature of the milieu, the pH, osmolarity and other parameters, that modulate the interactions between the different components of the liquid can modulate the dynamics of both mixing and demixing.

3) RNA processing and management in the oocyte

A. The challenges for gene expression in the oocyte

A.1. Transcription during Oogenesis

In the oocyte, gene expression is very complex (Figure 4). Basal levels of transcription can be detected in oocytes of primordial follicles, before they are recruited for growth. With the activation of growth, transcription is upregulated, and oocytes amass copious amounts of RNA throughout growth (Moore and Lintern-Moore 1974). Before the oocyte completes its growth, however, transcriptional activity is sharply down-regulated and becomes undetectable (Moore, Lintern-

Moore et al. 1974, Edwards, Farookhi et al. 2015). It remains as such in fully-grown GV oocytes, throughout maturation and the initiation of embryogenesis, only to resume later after a few mitotic cell cycles, in the late 2-cell stage in mouse (Edwards, Farookhi et al. 2015). Interestingly, these changes in transcriptional activity accompany landmark changes in the chromatin organization in the GV (discussed below) and precede chromosome condensation, GVBD, and the resumption of meiosis.

A.2. NSN to SN

Extensive remodeling of the heterochromatin and the nucleolar architecture of the GV during growth has been described, as early as 1975 in murine oocytes (Chouinard 1975). Mattson and

Albertini (1990) were the first to categorize into four sequential stages (I, II, III, IV) the evolution of these changes from a homogenous diffuse distribution of chromatin with multiple foci to a more condensed, restricted chromatin. This classification was then simplified by Debey et al. (1993) who characterized two major GV states, coining the terms used today: non-surrounded nucleolus

(NSN) where DNA staining is homogenous throughout the nucleus, with no obvious delineation of substructures (i.e. nucleolus) but the presence of foci of varying sizes; and surrounded nucleolus

(SN) where heterogenous DNA staining shows a condensation of the chromatin which now

encapsulates the nucleolus.

Additional work also showed that these nuclear changes are paired with oocyte and follicular

growth: oocytes of primordial follicles are all at the NSN state; with their recruitment and the

initiation of growth, the GV oocytes remain at NSN; as the oocytes near the end of growth, they

begin to switch to SN and before the initiation of maturation almost all GV oocytes are in SN

(Mattson and Albertini 1990, Wickramasinghe, Ebert et al. 1991, Debey, Szollosi et al. 1993,

Zuccotti, Piccinelli et al. 1995). Correspondingly, meiotic competence was also shown to correlate

with the state of the GV. Oocytes at the SN stage fared better in resuming meiosis than GV oocytes at the NSN stage (Wickramasinghe, Ebert et al. 1991, Debey, Szollosi et al. 1993). Most

importantly, a correlation was originally shown between the NSN or SN states and transcriptional

activity or inactivity, respectively (Bouniol-Baly, Hamraoui et al. 1999).

A.3. Histone modification profile in oocytes

Epigenetic markers including histone modifications are typically associated with changes of the chromatin and are known to have effects on transcription. NPM2 is a nuclear factor involved in protamine disassembly, nucleosome assembly and chromatin decondensation (Burns, Viveiros et al. 2003). In the absence of NPM2, in Npm2 -/- mutants, murine oocytes exhibit abnormal chromatin structures and cannot transition from NSN to SN (De La Fuente, Viveiros et al. 2004).

Similarly, oocytes see the accumulation of a myriad of histone modifications during growth.

H3K9, H3K18, H4K5, H4K12 acetylation and H3K4, H3K9 di- and trimethylation steadily increase in growing oocytes (Clarke and Vieux 2015). Conversely, global loss of these histone modifications is observed with the resumption of meiosis, prior to germinal vesicle break down

(Clarke and Vieux 2015)

H3K4me3 and H3K9me2/me3 are associated with active transcription and heterochromatin,

respectively (Kouzarides 2007, Eissenberg and Reuter 2009, Zhou and Zhou 2011). Yet, although

the knockout of methyltransferase Mll2, significantly prevents the accumulation of H3K4me3

(H3K4me and H3K4me2 were unaffected), this does not disrupt transcription, oocyte growth or

maturation, apart from significant spindle abnormalities (Andreu-Vieyra, Chen et al. 2010).

Moreover, Npm2 -/- oocytes show that the transition from NSN to SN at the end of growth and

can be uncoupled from transcriptional repression: preventing chromatin condensation does not

disrupt loss of transcription (De La Fuente, Viveiros et al. 2004). In contrast, the loss of

acetyltransferase (Hdac1 and Hdac2) expression leads to the disruption of transcription and the

arrest of oocyte growth, suggesting a more critical role for H3K18ac, H4K5a and H4K12ac (Ma,

Pan et al. 2012). This perhaps attests to the critical role of micro changes to the DNA architecture

(i.e. histone lysine acetylation) in regulating transcription while macro changes (i.e. chromatin

density) are dispensable. Nevertheless, the exact mechanism by which transcription is

downregulated in oocytes is not fully understood.

B. The need for translational regulation

B.1. Meiotic and developmental competence: a complex ordeal

Both post-transcriptional regulation of gene expression and post-translational regulation of protein

function are crucial for the normal and timely development of oocytes and the early steps of

embryogenesis. In particular, the inhibition of translation critically impairs oocyte maturation and

early embryogenesis (Fulka, Motlik et al. 1986). De-novo protein synthesis must therefore be sustained for meiotic and developmental competence, acquired during growth, to circumvent the drop in transcriptional activity in fully-grown oocytes. The translation of individual transcripts,

such as c-mos and Ccnb1 for instance, have been shown to be critical for meiotic resumption and

other molecular cascades driving maturation (Yew, Mellini et al. 1992, Gebauer, Xu et al. 1994,

Abrieu, Doree et al. 2001). This means that certain mRNAs transcribed during growth are not

utilized until days later, during maturation or the early stages of embryogenesis. Accordingly,

studies have concluded that, contrary to transcripts in most somatic cells whose half-lives are in

the range of minutes and hours, transcripts can be stored and survive in oocytes for days (Jahn,

Baran et al. 1976, Brower, Gizang et al. 1981, Groisman, Huang et al. 2001).

In vitro cultures yield MII oocytes with reduced developmental competence (Eppig and O'Brien

1996, Balakier, Sojecki et al. 2004). Interestingly, the transcript profile of cultured oocytes differs

only slightly from oocytes matured in vivo. Noticeably, however, this small difference does vary

between culture conditions. The transcriptome of MII oocytes cultured from primordial follicles,

the transcriptome of oocytes from cultured secondary follicles and the transcriptome of oocytes

developed in vivo all differ (Pan, O'Brien M et al. 2005). Whether this is the result of changes in

the dynamics of transcription during growth or direct detrimental effects on the transcriptome is

uncertain. However, poor developmental outcomes after in vitro maturation (Balakier, Sojecki et

al. 2004), occurring in the absence of transcription, suggest that this effect is at least in part

downstream of transcription. Nevertheless, changes to levels of certain transcripts including Ctcf

and Dnmt3a, regulating imprint establishment in the early embryo, shows an effect on

developmental competence and exemplify the high degree of precision required in managing

transcripts (Pan, O'Brien M et al. 2005). Regulating their translation requires various post-

transcriptional mechanisms and is instrumental for successful maturation and initiation of

embryogenesis. In fact, studies from the last four decades, on gene expression and RNA

metabolism in the oocyte have shed a bright light on the intricacies of RNA processing and its role

in RNA activity and stability.

B.2. Spatio-temporal regulation

The spatio-temporal regulation of translation adds an additional layer of complexity to gene expression in the oocyte and is critical for the orderly and successful progression of maturation and early embryogenesis. For instance, the expression of Aurora Kinases and the cytoplasmic polyadenylation-element-binding protein (CPEB) precedes and is required for the translation of

Ccnb1 and then c-mos, both of which are critical for meiotic progression (Sheets, Wu et al. 1995,

Uzbekova, Arlot-Bonnemains et al. 2008). Interestingly, in Xenopus oocytes, the ectopic expression of c-mos alone, is capable of setting of a series of molecular events that can initiate maturation (Yew, Mellini et al. 1992, Barkoff, Ballantyne et al. 1998). Similarly, overexpression of the CDK-activating kinase subunits, Cdk7 or Ccnh, can accelerate CDK1 activity, induce Ccnb1 expression and precociously resume meiosis (Fujii, Nishimura et al. 2011). In fact, constitutive

MPF activity in turn, also upregulates the translation of other factors (Wasserman, Richter et al.

1982). Conversely, the premature degradation of maternal transcripts is also associated with growth and meiotic defects. If the maternal mRNAs are unstable during oocyte growth, growth slows down, spindle formation is impaired and maturation fails (Medvedev, Pan et al. 2011).

Furthermore, the timely progression of events appears to be essential for successful maturation and embryogenesis; extended GV arrest at prophase I, after transcriptional shut down, is linked to lower blastocyst rates (De La Fuente and Eppig 2001). In other species, other mechanisms involved in RNA localization also play a significant role in sequestering specific transcripts to cellular compartment or later cell populations to orchestrate cell differentiation events during early embryogenesis (Riechmann and Ephrussi 2001, Updike and Strome 2010). Many studies therefore have tried to study the multifaceted aspects of gene expression in reproduction, but much of the mechanics that regulate RNA storage and translation in mammalian oocytes remain elusive.

C. Mechanisms of RNA processing in mammalian oocytes

C.1. Polyadenylation and translational activity

By the end of growth and the initiation of maturation, the oocyte shifts from a growth program to a cell cycle and early embryogenesis program as it prepares itself for fertilization. Accordingly,

the polyribosome profile of the oocyte, representative of translationally active transcripts, changes.

Over 7600 transcripts are translated in maturing oocytes including mos and the key meiotic regulator Ccnb1 (Chen, Melton et al. 2011). Mirroring these results, the lengthening of the poly(A) tail of specific maternal transcripts has also been described at the onset of maturation (Gohin,

Fournier et al. 2014, Sousa Martins, Liu et al. 2016). But the different patterns of translation during maturation suggest different regulatory mechanisms. Certain transcripts, for example, require the synergistic effects of two cis elements, the CPE and a DAZL-binding element in the 3'-UTR, to be polyadenylated and successfully translated (Sousa Martins, Liu et al. 2016).

The CPE, as previously mentioned, has been shown to induce both polyadenylation with translation and deadenylation with translational inhibition. In fact, the role of the CPE and it binding partner, CPEB, was first elucidated in Xenopus oocytes. CPEB homologues were identified in various mammals including humans and mice (Gebauer and Richter 1996, Welk,

Charlesworth et al. 2001). In mice, CPEB is critical for both germ cell differentiation and oogenesis

(Tay and Richter 2001, Racki and Richter 2006). More specifically, during growth, the CPE/CPEB interaction supports the polyadenylation of several oocyte mRNAs, including Smad1, Smad5, spindlin, Bub1b, Mos, H1foo, Obox1, Dnmt1o, TiParp, Trim61 and Gdf9 and promotes oocyte development. At the onset of maturation, IAK1 phosphorylation of CPEB drives the translation of

Ccnb1 and promotes meiotic resumption (Hodgman, Tay et al. 2001).

C.2. Decapping, RNAi and degradation

The selective repression of translation and RNA decay is also important in oocytes. Bachvarova et al. (1985) described a large-scale RNA degradation event in maturing oocytes that depleted 20% of the oocyte’s total RNA store. This process is tightly regulated. Despite basal levels in growing oocytes, both DCP1A and DCP2 are mostly accumulated during maturation. Preventing this accumulation or disrupting their role in RNA degradation delays zygotic genome activation in embryos (Ma, Flemr et al. 2013, Ma, Fukuda et al. 2015). More recent work has identified a specific transcript targeted by degradation, whose turnover is critical for early embryogenesis to resume timely and successfully (Treen, Heist et al. 2018). Moreover, the germ-cell specific RNA- binding protein MSY2 confers additional stability to maternal transcripts during growth. When

MSY2 expression is disrupted, poly(A)-containing RNA levels drop by ~25% and oocyte growth is stunted and maturation significantly impaired (Medvedev, Pan et al. 2011). Interestingly, coinciding with the phosphorylation of decapping proteins, the phosphorylation of MSY2 by

CDK1 is important for the maternal transcript degradation (Medvedev, Yang et al. 2008).

Levels of the Argonaute family of proteins are very low in mouse oocytes. Yet, the injection of exogenous AGO2 did not affect the pattern of expression of maternal transcripts (Freimer,

Krishnakumar et al. 2018). This suggest a high degree of redundancy in the mechanisms keeping maternal transcripts stable during oocyte growth. Concurrently, the loss of DICER expression in the oocyte, affecting both miRNA and siRNA production, severely disrupts folliculogenesis in the fetal ovary and oocyte maturation in the adult ovary (Liu, Tang et al. 2010, Yuan, Ortogero et al.

2014). Other studies however, suggest a role only for siRNA in oocytes (Ma, Flemr et al. 2010).

DGCR8 is a Drosha co-factor and another essential component of the nuclear microprocessor complex involved in the generation of miRNA but not siRNA. Surprisingly, the loss of Dgcr8

expression does not impair follicular development or oogenesis (Suh, Baehner et al. 2010).

Accordingly, while both miRNA and siRNA have been widely used for mouse knock down

experiments in molecular biology labs, siRNA tend to be the preferred tool for studies in the mouse

oocyte.

C.3. Oocyte RNPs: germ granules, P-bodies and stress granules

In addition to polyribosomes, commonly found in most cells, two additional types of RNPs have been reported in oocytes to date (Figure 11). First, there are the germ granules that are involved in the storage and the transmission of maternal transcripts from the oocyte to the zygote. In

Drosophila, C. elegans, Zebrafish and Xenopus, they are often involved in the segregation of

transcripts required for early patterning of the embryo and germ cell differentiation (Riechmann

and Ephrussi 2001, Updike and Strome 2010, Schisa 2012). Second, P-bodies-like RNPs have also been reported in mammalian oocytes (Flemr, Ma et al. 2010). These RNPs however, have been suggested to be involved in the storage of transcripts meant to be used during oocyte maturation and early embryogenesis: a role reminiscent of the P-body-like RNPs described in C. elegans oocytes (Boag, Atalay et al. 2008, Schisa 2012). Interestingly, the translational activity of

Ccnb1 has been shown to strictly correlate with the transcripts localization in and out of similar foci (Kotani, Yasuda et al. 2013). But a more functional approach is needed to address a causative effect. In C. elegans, another type of RNP has been described. The stress induced grP-bodies comprise of canonical P-bodies but are reminiscent of stress granules (Jud, Czerwinski et al. 2008,

Noble, Allen et al. 2008). They are also believed to protect the transcriptome of oocytes (Schisa

2012). No such RNPs has been described in mammalian oocytes however.

Figure 11 – Spectrum of induced ribonucleoprotein particles and composition. There are two main categories of inducible RNPs described in eukaryotic cells. P-bodies are the site of RNA decapping and degradation. As such, subunits of the decapping complex, deadenylases and ribonuclease helicases localize to P-bodies. Other factors also associated to translational repression are often found in P-bodies as well. Stress granules composed of factors with high prion-like domain, facilitating their aggregation in adverse conditions. Most factors found in stress granules are RNA- binding proteins. Some helicase, translational inhibitors as well as some of the subunits of the translation initiation complex can also be found in stress granules. Germ granule-related RNPs (or grP-bodies) are a type of granule that form in C. elegans oocytes arrested at ovulation. They can also be induced by various form of stress including heat shock. They share commonalities with both P-bodies and stress granules (Adapted from Buchan and Parker, 2009).

D. Rationale and hypothesis

The transition from oocyte growth to the resumption of meiosis relies on very dynamic molecular

changes including the translation of maternal transcripts. My thesis work focuses on RNA

management during maturation. In Chapter 2, the role of the poly(A) tail and the mechanisms

regulating its length are explored. While the polyadenylation of maternal transcripts has been well

described in mammalian oocytes, the enzymes responsible for deadenylation and their mode of

action remain more elusive. I therefore hypothesized that the CCR4-NOT-associated deadenylase,

CNOT6, is required for maternal transcript deadenaylation. This was tested using an RNAi approach to knock-down CNOT6 in oocytes and determine the impact on poly(A) tail lengths at distinct stages. In doing so, the role of the poly(A) tail on translation and degradation is also assessed. In Chapter 3, a novel type of RNP is identified and characterized in oocytes in duress. I hypothesized that these granules protect the oocyte. Regulatory factors are then identified and targeted (by chemical treatment) to test the functionality of the stress induced granules. The work presented in this thesis shed light on novel mechanisms of mRNA post-transcriptional regulation in mammalian oocytes and provide further evidence for the essential role of maternal transcripts in determining oocyte quality and competence.

Chapter 2

[Manuscript I] CNOT6 regulates a novel pattern of mRNA deadenylation during oocyte meiotic maturation

Preface

RNA processing is at the heart of translational regulation. More specifically, the length of the

poly(A) tail has typically been associated with translational activity (Munroe and Jacobson 1990,

Beilharz and Preiss 2007). In the oocyte, RNA management is critical for successful maturation

and embryogenesis (Kotani, Yasuda et al. 2013, Ma, Flemr et al. 2013, Ma, Fukuda et al. 2015,

Treen, Heist et al. 2018, Treen, Heist et al. 2018). Previous work from others, describe different

patterns of translation and an alternative degradation-independent mechanism of translational

repression (Chen, Melton et al. 2011). Moreover, cis-regulatory elements regulate the pattern of

expression of maternal transcripts. Both the CPE and DAZL-binding element work synergistically

to promote polyadenylation and translation (Chen, Melton et al. 2011, Sousa Martins, Liu et al.

2016). This suggests that in contrast, deadenylation may be involved in the silencing of maternal

transcripts. To test this, the following objectives were designed:

1- Describe the evolution of the poly(A) tail of maternal transcripts during oocyte maturation and

identify candidates that are deadenylated.

2- Identify the deadenylase(s) responsible for the shortening of their poly(A) tail.

3- Identify cis-regulatory factors modulating deadenylate activity.

4- Determine the role of the poly(A) tail in regulating their translation and degradation.

This work is discussed in Chapter 2 (Manuscript I). Chapter 2 is a modified version of the published work.

Significance statement

Proper timing of the translational silencing and the activation of maternal transcripts is critical for successful oogenesis and early embryogenesis. RNA processing events, including the regulation of the poly(A) tail length, are believed to play an essential role in that process but are incompletely understood. Here we describe three adenylation patterns during oocyte maturation. Actb and Ngfr are deadenylated, Ccnb1 is poly adenylated and Orc6 and Slbp are first polyadenylated and then deadenylated, mirroring their expression in maturing oocytes. We then identify CNOT6 as the deadenylase responsible for keeping the poly (A) tails of Actb and Slbp short both at the GV stage and at MII, via a CPEB and PBE-regulated mechanism. Deadenylation is also known to regulated translation and RNA turnover. In the mammalian oocyte, we have shown however, that the poly(A) tail functions in a context-dependent way and can be dissociated from translation and degradation.

More potent regulators of RNA metabolism therefore supersede the changes to the poly(A) tail.

CNOT6 regulates a novel pattern of mRNA deadenylation during oocyte meiotic maturation

Karl-Frédéric Vieux1,3 & Hugh J. Clarke1,2,3

1Department of Biology, McGill University, Montreal, Quebec, Canada.

2Department of Obstetrics and Gynecology, McGill University, Montreal, Quebec, Canada.

3Research Institute of the McGill University Health Centre, Montreal, Quebec, Canada.

Scientific Reports, volume 8, Article number: 6812 (2018) doi:10.1038/s41598-018-25187-0.

Keywords: CNOT6, oocyte, RNA, deadenylation, translation, degradation

Corresponding Author: Hugh Clarke MUHC Research Institute 1001 Decarie Blvd, Block E-M0.2218, Montreal, QC CANADA H4A 3J1 [email protected]

Competing Interests: The authors declare no competing interests.

Abstract

In many cell types, the length of the poly(A) tail of an mRNA is closely linked to its fate – a long

tail is associated with active translation, a short tail with silencing and degradation. During

mammalian oocyte development, two contrasting patterns of polyadenylation have been identified.

Some mRNAs carry a long poly(A) tail during the growth stage and are actively translated, then become deadenylated and down-regulated during the subsequent stage, termed meiotic maturation.

Other mRNAs carry a short tail poly(A) tail and are translationally repressed during growth, and their poly(A) tail lengthens and they become translationally activated during maturation. As well, a program of elimination of this ‘maternal’ mRNA is initiated during oocyte maturation. Here we

describe a third pattern of polyadenylation: mRNAs are deadenylated in growing oocytes, become polyadenylated during early maturation and then deadenylated during late maturation. We show that the deadenylase, CNOT6, is present in cortical foci of oocytes and regulates deadenylation of these mRNAs, and that PUF-binding elements (PBEs) regulate deadenylation in mature oocytes.

Unexpectedly, maintaining a long poly(A) tail neither enhances translation nor inhibits degradation of these mRNAs. Our findings implicate multiple machineries, more complex than previously thought, in regulating mRNA activity in oocytes.

Introduction

A hallmark of germ cell differentiation is the central role played by post-transcriptional mechanisms in regulating gene expression. In female vertebrates, post-natal development of the oocyte comprises two phases – a prolonged growth phase during which the oocyte increases enormously in size, and a much shorter phase termed meiotic maturation during which it completes the first meiotic division (Sánchez and Smitz 2012, El-Hayek and Clarke 2016). During growth, oocytes synthesize large quantities of mRNA that owing to its extraordinary stability – its half-life in mice is estimated to be 3 weeks (Ma, Fukuda et al. 2015) – accumulates to form a large stock of ‘maternal’ mRNA. Some of the accumulated mRNA species are immediately translated to support the growth process, whereas other species are stored in ribonucleoprotein particles in a translationally inactive form. When fully grown oocytes enter meiotic maturation, however, the situation changes dramatically – many previously active mRNAs become translationally silenced, whereas previously silent mRNAs become activated. In addition, through a process that is incompletely understood, the stock of maternal mRNA begins to be degraded. Thus, both the translational activity and the stability of mRNAs are dynamically regulated during oocyte growth and meiotic maturation (Kang and Han 2011, Clarke 2012, Reyes and Ross 2016, Susor, Jansova et al. 2016).

In many cell types, changes in mRNA translation and stability are closely linked to changes in the length of the poly(A) tail, a sequence of adenine residues that is added to the 3′-end of most mRNAs co-transcriptionally in the nucleus (Eckmann, Rammelt et al. 2011, Norbury 2013,

Jalkanen, Coleman et al. 2014, Reyes and Ross 2016). The poly(A) tail provides a docking site for the cytoplasmic poly(A) binding protein (PABPC), which together with the translation initiation

factor eIF4G and the cap-binding protein eIF4E promote translation. A long poly(A) tail is also

associated with mRNA stability; in contrast, a short poly(A) is associated with translational

inactivity and degradation. Following the export of polyadenylated mRNAs to the cytoplasm,

competing activities of poly(A) polymerases and deadenylases respectively lengthen or shorten

the poly(A) tail. Lengthening of the poly(A) tail is promoted by coordinated activity of the

cleavage and polyadenylation specificity factor (CPSF), which binds to the polyadenylation signal

(AAUAAA) located near the 3′-end of the mRNA, and poly(A) polymerases, which catalyse the

polyadenylation reaction. Shortening of the tail is promoted by deadenylases, of which two major

cytoplasmic families have been identified (Doidge, Mittal et al. 2012, Wahle and Winkler 2013,

Shirai, Suzuki et al. 2014). The poly(A)-specific nuclease (PAN), comprising the PAN2 catalytic

subunit in a complex with PAN3 is thought to be responsible at least in some cell types for the

initial deadenylation of an mRNA (Yamashita, Chang et al. 2005, Bartlam and Yamamoto 2010).

The major deadenylase activity in most cell types, however, is conferred by the CCR4-NOT

complex. This multi-protein subunit consists of five or more main subunits, including NOT family

members that provide structural integrity to the complex and two proteins that possess deadenylase

activity – one of either CNOT6 or CNOT6L, members of the EEP (exonuclease-endonuclease- phosphatase) family of deadenylases, and one of either CNOT7 or CNOT8, members of the DEDD family of deadenylases that are named for the conserved Asp and Gln in their catalytic domains.

Although both EEP and DEDD members are catalytically active, they can target different mRNAs

(Aslam, Mittal et al. 2009, Mittal, Aslam et al. 2011, Doidge, Mittal et al. 2012). This provides a

potential mechanism for selective deadenylation of specific mRNA species, although the

mechanism underlying the selectivity remains to be identified. Both CCR4-NOT and PAN2/3 can

be recruited to the poly(A) tail of mRNAs, thus mechanistically coupling deadenylation to prior

polyadenylation.

Two patterns of mRNA polyadenylation dynamics have been identified in oocytes (Kang and Han

2011, Clarke 2012, Reyes and Ross 2016). Many mRNAs carry a long poly(A) tail in growing and

fully grown immature oocytes and are actively translated. During maturation, however, these

mRNAs become deadenylated and their translational activity decreases. In contrast, other mRNAs

carry a short poly(A) tail in growing and fully grown immature oocytes and are only weakly

translationally active; during maturation, the poly(A) tail of these mRNAs lengthens and they

become translationally active. Deadenylation of mRNAs during maturation has also been linked to their degradation. The pattern of polyadenylation of an mRNA is strongly influenced by sequences in its 3ʹ-untranslated region (UTR), including a U-rich sequence termed the cytoplasmic polyadenylation element (CPE), the PUF-binding element (PBE), and the Musashi-binding element (Goldstrohm and Wickens 2008, Charlesworth, Meijer et al. 2013, Cragle and MacNicol

2014, MacNicol, Cragle et al. 2015, Reyes and Ross 2016). In particular, the presence or absence, respectively, of a CPE within about 200 nt of the polyadenylation signal frequently dictates whether an mRNA will become polyadenylated or deadenylated during maturation. These complementary patterns of polyadenylation dynamics suggest a simple dichotomous model in which two classes of oocyte mRNA exist in maturing oocytes – one becoming polyadenylated and translationally active, and the other becoming deadenylated, translationally silent, and degraded.

Emerging evidence now indicates that the translational landscape during meiotic maturation is more complex than originally anticipated, as exemplified by changes in the translational activity of different mRNAs at different stages of maturation (Belloc, Pique et al. 2008, Pique, Lopez et al. 2008, Chen, Melton et al. 2011, MacNicol, Cragle et al. 2015) and the selective degradation of

some mRNAs while others remain stable until after fertilization (Su, Sugiura et al. 2007, Ma,

Fukuda et al. 2015). This raises the possibility that mRNA polyadenylation dynamics and function

during maturation might be more varied than implied by the dichotomous model. Here we identify

oocyte mRNAs which display a novel pattern of adenylation during maturation. These mRNAs

carry a short poly(A) tail in growing oocytes, become polyadenylated during the early portion of

oocyte maturation, and then are deadenylated during late maturation coincident with partial

degradation. The CNOT6 deadenylase selectively regulates the deadenylation of these mRNAs,

and we uncover a role for specific sequences in the 3′-UTR of the mRNAs in mediating

deadenylation. Surprisingly, preventing their deadenylation during late maturation by depletion of

Cnot6 neither increases their translation nor prevents their degradation. These results indicate that

the multiple patterns of mRNA adenylation co-exist in maturing oocytes and in addition suggest

that poly(A) tail length may not play a deterministic role in regulating mRNA translation and

stability in these cells.

Results

Identification of a novel pattern of mRNA adenylation during oocyte maturation

To study the relationship between the poly(A) and the activity of an mRNA, we used a method

termed RL-PAT that we and others have developed to measure poly(A) tail-length (Graindorge,

Thuret et al. 2006, Nakanishi, Kubota et al. 2006, Belloc and Mendez 2008, Prasad, Mahadevan

et al. 2008, Yang, Allard et al. 2010). Briefly, an RNA linker is ligated to the 3′-end of the mRNA, which is then subjected to RT-PCR using gene- and linker-specific primers. This generates PCR products of a single size, in contrast to the multiple lengths produced using methods based on oligo-dT, which may be able to anneal at many sites on the poly(A) tail. We focused on the poly(A)

tail dynamics of three mRNAs known to become translationally activated during meiotic maturation: Ccnb1, which encodes cyclin B1; Orc6, which encodes a DNA replication factor; and

Slbp, which encodes a protein that binds to a stem-loop structure in the 3′-untranslated region

(UTR) of histone mRNAs (Arnold, Francon et al. 2008, Murai, Stein et al. 2010). Maternal cyclin

B1 is required for the completion of meiotic maturation, whereas maternal ORC6 and SLBP are required for embryonic development.

Consistent with previous studies, the poly(A) tail of mRNAs that are translated in immature oocytes shortened during maturation (Figure 12A – Actb, Ngfr), whereas that of Ccnb1 lengthened

(Figure 12). Unexpectedly, however, we observed a third pattern of polyadenylation for Orc6 and

Slbp. Both mRNAs carried a short poly(A) tail in immature oocytes, which lengthened substantially during the early portion of maturation, following germinal vesicle breakdown

(GVBD). These results are consistent with previous data showing a putative CPE in the 3′-UTR of each mRNA and the accumulation of both proteins during maturation (Allard, Champigny et al.

2002, Allard, Yang et al. 2005, Murai, Stein et al. 2010). As maturation progressed to metaphase

II, however, the poly(A) tail of both mRNAs subsequently shortened (Figure 12C, MII group)

(Yang, Allard et al. 2010). Thus, at least three different patterns of mRNA polyadenylation occur during oocyte maturation.

CNOT6 is expressed in immature and maturing oocytes

To gain insight into the mechanism of these diverse patterns of adenylation, we began by identifying deadenylases expressed by oocytes, as previous results indicated that deadenylases can display mRNA-specificity (Ma, Fukuda et al. 2015, Yu, Ji et al. 2016). We found that oocytes express mRNAs encoding a wide range of CNOT family members, including its catalytic subunits

76 encoded by Cnot6, Cnot6l, Cnot7 and Cnot8; they also expressed Pan2 and Pan3, the nuclear deadenylase, Parn, and the related deadenylase, Pnldc (Figure 13A). Using available antibodies, we detailed the expression of the CCR-4NOT associated deadenylases CNOT6 and CNOT7. We confirmed that CNOT6 is present in both immature and maturing oocytes Figure 13B). CNOT7 is also expressed, confirming previous reports (Ma, Fukuda et al. 2015, Yu, Ji et al. 2016). Because it is thought to target different mRNAs than those regulated by CONT7, we focused our studies on CNOT6.

In somatic cells, CNOT6 accumulates in cytoplasmic ribonucleoprotein particles (RNPs) that contain mRNAs and proteins implicated in mRNA processing (Cougot, Babajko et al. 2004,

Andrei, Ingelfinger et al. 2005, Kotani, Yasuda et al. 2013). RNPs sharing a similar molecular composition have been identified in mouse oocytes, where they are localized in the cortical region

(Swetloff, Conne et al. 2009, Flemr, Ma et al. 2010, Flemr and Svoboda 2011), and so we examined the distribution of CNOT6 in oocytes. We collected growing and fully grown immature oocytes and mature oocytes, and examined CNOT6 localization using immunofluorescence. We observed a narrow band of prominent foci of staining located near the periphery of the oocytes at all three stages of development (Figure 14A). A small number of foci were also apparent more centrally within the cytoplasm. To accurately establish the intracellular location of the prominent foci, we used phalloidin to label the actin-rich cortex (Figure 14B). This revealed that the foci lie within the cortical region of the oocyte. To verify this distribution, we also stained histological sections of ovaries which revealed, as observed using the whole-mount specimens, brightly stained CNOT6 foci at the periphery of oocytes (Figure 14C).

We then tested whether the CNOT6 foci might be related to the previously described cortical

RNPs. TNRC6 (GW182) and EDC4, which are recognized by antiserum 18033, and the RNA-

binding protein, LSM14, are components of RNPs known as P-bodies (Eystathioy, Jakymiw et al.

2003, Marnef, Sommerville et al. 2009, Flemr, Ma et al. 2010, Braun, Huntzinger et al. 2013).

Antiserum 18033 and anti-LSM14 antibodies each stained the cortex of oocytes as previously reported (Flemr, Ma et al. 2010, Flemr and Svoboda 2011) and displayed a pattern remarkably similar to that observed for CNOT6 (Figure 15). Because the CNOT6 antibody and serum 18033 are derived from different host species (rabbit and human, respectively), we were further able to test whether these proteins are co-localized. We found that the signals overlapped in many although not all cases (Figure 15B). Two statistics were calculated to better evaluate this overlap:

Mander’s overlap coefficient, which represents the fraction of pixels positive for both signals; and

Pearson’s correlation coefficient, which represents the correlation in the fluctuation of the two signals from pixel to pixel (Dunn, Kamocka et al. 2011). Quantitative analysis of the staining patterns at the cortex of oocytes revealed a Mander’s overlap coefficient of 0.73 ± 0.02 (12 oocytes analysed), indicating a significant colocalization between CNOT6 and 18033 at the cortex, and a

Pearson’s correlation coefficient of 0.32 ± 0.02 (12 oocytes analysed) indicating a positive linearity in the incidence of the two signals. Taken together, these results indicate that the cortical CNOT6 is present in structures that share characteristics with P-bodies.

CNOT6 selectively deadenylates specific mRNAs

Having established that CNOT6 is expressed in oocytes, we then tested whether it regulates mRNA poly(A) tail length. We injected RNAi targeting Cnot6 or a negative control (not expressed in oocytes) into fully grown immature oocytes, incubated them overnight, then either continued to maintain them at the GV stage or allowed them to undergo maturation (Figure 16A). By 24 hr after injection, both Cnot6 mRNA and CNOT6 in cortical foci had been significantly depleted (Figure

16B). Analysis using RL-PAT revealed no effect of CNOT6 depletion on the poly(A) tail of Actb

and Ngfr, which becomes deadenylated during maturation, or of Ccnb1, which becomes stably

polyadenylated during maturation (Figure 16C). In striking contrast, however, CNOT6 depletion

significantly affected the poly(A) tail of Orc6 and Slbp. In both immature and mature oocytes, the length of the poly(A) tail was increased by about 100 nt, as compared to oocytes injected with a control RNAi (Figure 16C). Thus, CNOT6 selectively regulates the poly(A) tail length of specific mRNAs.

PUF-binding elements (PBE) are required for deadenylation during late maturation

To understand the basis for the mRNA species-selectivity of CNOT6, we searched for specific sequence elements in the 3ʹ-UTR of these mRNAs. As expected, given that their poly(A) tail lengthens when maturation begins, Orc6 and Slbp contained a putative CPE located near the polyadenylation signal (Figure 17). U-rich sequences were also present directly upstream or overlapping with the CPEs. Strikingly, both Orc6 and Slbp also contained putative PBE sequences

(UGU(A/U)N(AU/UA) lying between the 3′-most CPE and the polyadenylation signal. In contrast,

Actb, Ngfr and Ccnb1 did not contain PBEs in this position on the 3ʹ-UTR. PBE sequences have previously been associated with deadenylation by the CCR-NOT complex (Goldstrohm and

Wickens 2008, Van Etten, Schagat et al. 2012, Joly, Chartier et al. 2013). To test whether the putative PBEs mediate CNOT6-dependent deadenylation, the 3′-UTR of Slbp mRNA was ligated

to the firefly luciferase coding sequence (Fig. 7A). Two constructs were generated: one carrying

the wild-type sequence of the Slbp 3′-UTR (wt-Slbp-3′UTR), and a second one in which the three

putative PBEs were deleted (ΔPBE-Slbp-3′UTR). cRNAs prepared using these constructs were

individually microinjected into immature oocytes, which were then maintained in the immature

condition or allowed to mature, after which the length of the poly(A) tail was determined using

RL-PAT.

The wt-Slbp-3′UTR construct carried a short poly(A) tail in immature oocytes, which lengthened

during the early phase of maturation and then shortened at the completion of maturation (Figure

18B,C). This pattern mirrors the polyadenylation dynamics observed for endogenous Slbp and thus

demonstrates that this region of the Slbp 3′-UTR is sufficient to dictate the polyadenylation pattern of Slbp during oocyte maturation. Deletion of the putative PBE sequences had no detectable effect on the length of the poly(A) tail in immature oocytes. In contrast, deletion of these sequences led to an increase in the length of the poly(A) tail in mature oocytes. These results suggest that the putative PBE sequences are required for the CNOT6-dependent deadenylation of mRNAs during late maturation.

Lengthening the poly(A) tail is not sufficient to increase mRNA translation or inhibit mRNA

degradation

Because the poly(A) tail has been implicated in the regulation of mRNA translation and

degradation, and CNOT6 regulates the poly(A) tail length of Slbp and Orc6 in both immature and

mature oocytes, we tested the effect of depleting CNOT6 on the translation and degradation of

these mRNAs. Confirming the previous reports (Arnold, Francon et al. 2008, Murai, Stein et al.

2010), we observed that both SLBP and OCT6 increased in abundance during maturation of

unmanipulated oocytes (Figure 19A, left). Depletion of CNOT6 did not, however, increase the

amount of either protein in immature or mature oocytes (Figure 19A, right). Thus, an increase in

the length of the poly(A) tail was not sufficient to increase translation of either mRNA at these

two stages of oocyte development.

Next, we examined mRNA degradation. We found that the quantities of Slbp and Orc6 declined

by about 50% during maturation of unmanipulated oocytes (Figure 19B). Actb and Ngfr, whose

poly(A) tail shortens during maturation, and Ccnb1, whose poly(A) tail lengthens, also declined during this period. These results are consistent with previous reports that oocyte mRNAs begin to be degraded during maturation (Bachvarova, De Leon et al. 1985, Su, Sugiura et al. 2007, Ma,

Fukuda et al. 2015, Yu, Ji et al. 2016, Sha, Dai et al. 2017). When we depleted CNOT6, we observed no change in the amount of Slbp and Orc6 in mature oocytes (Figure 19C). Thus,

experimental lengthening of the poly(A) of Slbp and Orc6 did not protect either mRNA from the partial degradation that occurs during maturation.

Discussion

Previous studies of post-transcriptional regulation of gene expression during oocyte maturation

identified two patterns of polyadenylation. Many mRNAs follow what is considered a default

pathway – they carry a long poly(A) tail in immature oocytes, which shortens during maturation,

when the mRNAs also become translationally suppressed (Fox and Wickens 1990). In contrast, mRNAs that bear one or more CPE in the 3′-UTR within ~200 nt of the polyadenylation signal carry a short poly(A) tail in immature oocytes, which lengthens during maturation when the mRNAs become translationally activated. Here we describe a third pattern of polyadenylation, also recently noted by others (Yu, Ji et al. 2016), in which mRNAs carry a short poly(A) tail in immature oocytes that becomes transiently lengthened during the early portion of maturation as the mRNAs become translationally activated and then becomes shortened during late maturation.

Deadenylation of mRNAs can reportedly follow their translation. Ccnb1 however, is actively translated during maturation, yet does not become deadenylated during late maturation. This suggests that another mechanism is responsible for the late deadenylation, and we provide evidence that putative PBE sequences in the 3ʹ-UTR of Slbp and Orc6 may mediate this process.

Figure 20 presents a mechanism by which late deadenylation might occur, based on evidence that both PUF proteins (Goldstrohm and Wickens 2008, Van Etten, Schagat et al. 2012) and the CPE- binding protein (CPEB) 1 (Ogami, Hosoda et al. 2014, Shirai, Suzuki et al. 2014) can recruit the

CCR4-NOT complex to mRNAs and that CPEB1 becomes phosphorylated during early maturation and is subsequently largely degraded in mature oocytes (Yang, Allard et al. 2010, Chen,

Melton et al. 2011, Sha, Dai et al. 2017). We propose that, in immature oocytes, CPEB recruits

CCR4-NOT to the 3′-end of a target mRNA, promoting its deadenylation. During early maturation,

CPEB becomes phosphorylated, which impairs its ability to recruit CCR4-NOT. Nonetheless, its

sterically hinders binding of PUF proteins to PBE sequences located close to the CPE. In the absence of an associated deadenylase activity, the poly(A) tail of the mRNA lengthens. During late maturation, the degradation of CPEB enables PUF proteins to bind to the PBE present on type

III mRNAs and recruit the CCR4-NOT deadenylase complex.

We found that CNOT6 regulated the deadenylation of this subset of mRNAs, but not of other

mRNAs, in both immature and mature oocytes. Intriguingly, CNOT7 deadenylase regulates the poly(A) tail-length of other mRNAs in the oocyte (Ma, Fukuda et al. 2015). This suggests that different CNOT deadenylases may interact with different mRNAs in the oocyte. In this context, it is worth noting that this type of selectivity has previously been described in somatic cells (Aslam,

Mittal et al. 2009, Mittal, Aslam et al. 2011, Doidge, Mittal et al. 2012). Although the basis for the

selectivity remains to be determined, we detected CNOT6 in cortical foci of the oocyte where it

frequently co-localizes with 18033 antigens. Previous studies have suggested that cortically

associated ribonucleoprotein aggregates in mouse oocytes – particularly those containing 18033

antigens – may be sites of mRNA storage or processing (Swetloff, Conne et al. 2009, Kotani,

Yasuda et al. 2013, Susor, Jansova et al. 2015, Rosario, Filis et al. 2016, Trapphoff, Heiligentag

82 et al. 2016). Although we do not know what proportion of the cellular CNOT6 is present in the foci, it may be speculated that both CNOT6 and its target mRNAs accumulate in cortical foci and that this physical proximity underlies their selective interaction.

Because reducing the quantity of CNOT6 on the oocyte increased the length of the poly(A) tail of

Slbp and Orc6, we were able to study the effect of the longer tail on the activity of these mRNAs.

We found that the longer poly(A) did not increase the steady-state level of either encoded protein in immature or mature oocytes. This suggests that an increase in the length of the poly(A) tail, at least of these mRNAs, does not directly drive increased translation. Prior studies, largely utilizing reporter constructs bearing wild-type or mutant CPEs, have nonetheless consistently identified a close link between lengthening of the mRNA poly(A) tail and translational activation (Ma, Flemr et al. 2013, Ma, Fukuda et al. 2015, Sousa Martins, Liu et al. 2016, Yu, Ji et al. 2016, Yang, Yang et al. 2017). These results may be reconciled by proposing that an increase in poly(A) tail length is a marker of increased translation, but is not an essential part of the mechanism.

We also found that increasing the poly(A) tail length did not prevent the partial degradation of

Slbp or Orc6 during maturation. Similarly, increasing the poly(A) tail length of mRNAs regulated by CNOT7 did not prevent their partial degradation during maturation (Ma, Fukuda et al. 2015).

These results suggest that the mechanism of maternal mRNA degradation during maturation does not depend on deadenylation. The partial degradation of Ccnb1 during maturation even though it becomes polyadenylated at this time, together with prior studies in Xenopus (Audic, Omilli et al.

1997) supports this interpretation. A recent study reported much more extensive mRNA degradation during maturation, and further showed that this degradation was suppressed when

BTG1, which promotes deadenylation, was depleted (Yu, Ji et al. 2016). Although the basis for these differences remains to be elucidated, they may be related to the method used to prepare the

cDNA (discussed in Ma, et al. 2015 and Su, et al. 2007). In any case, by identifying a distinct

pattern of polyadenylation dynamics during maturation and identifying a role for CNOT6 in

regulating this process, our results reveal a mechanism by which the activity of specific mRNAs may be selectively regulated in the oocyte.

Materials and Methods

Mice

All experiments were performed using CD1 mice (Charles River Canada, St.-Constant, QC,

Canada) in compliance with the regulations and policies of the Canadian Council on Animal Care

and were approved by the Animal Care Committee of the Royal Victoria Hospital (protocol 5991).

Oocyte collection

Fully grown oocytes containing a visible intact germinal vesicle (GV) and a diameter greater than

70 μm were collected from post-natal day (PD) 19 females at the GV stage. Ovaries were dissected

into several fragments and incubated at 37 °C in air in minimal essential medium (MEM, Life

Technologies) buffered at pH 7.2 using HEPES (MEM-H), containing sodium pyruvate (0.25 mM,

Sigma), penicillin G (63 mg/L, Sigma), streptomycin (50 mg/L, Sigma), and bovine serum albumin

(BSA, 1 mg/mL, Sigma). Large follicles were punctured using fine needles. The collection media

was supplemented with dibutyryl cyclic AMP (dbcAMP, 0.1 mg/ml, Sigma) to prevent resumption

of meiosis. Oocytes were collected using a mouth- controlled micropipette, transferred to fresh

medium, and the diameter was measured using an ocular micrometer. To obtain oocytes that had

undergone germinal vesicle breakdown (GVBD) and completed maturation to metaphase II,

respectively, GV-stage oocytes were transferred to MEM-NAHCO3 containing pyruvate,

antibiotics and BSA, and incubated at 37 °C in 5% CO2 in air for 5 hr or overnight, respectively, to allow meiotic maturation. GVBD oocytes were identified by the absence of the GV and mature oocytes were by the presence of a polar body.

Reverse transcription-polymerase chain reaction (RT-PCR)

RNA was extracted from oocytes using the ARCTURUS® PicoPure® RNA Isolation Kit

(KIT0204; Life Technologies Applied Biosystems) following the manufacturer’s protocol including the optional DNA removal step. RNA samples were then used for reverse transcription and PCR as described (Allard, Yang et al. 2005). Primers used for PCR are listed in Table 1.

RNA ligation-mediated polyadenylation test (RL-PAT)

RL-PAT was performed as previously described (Yang, Allard et al. 2010). Following RNA extraction, an RNA linker was ligated to the 3′-end of the RNA. Following reverse-transcription, the cDNA was subjected to PCR amplification using nested primers corresponding to the mRNA

of interest at the 5′-end and the ligated RNA at the 3′-end. Sequences for the RNA linker and the

primers used are listed in Table 2. The products were separated using a 15% agarose gel.

Quantitative PCR (qPCR)

Following oocyte collection, 4 pg of rabbit globin mRNA (Hba1) (Sigma; R1253) was added to

each sample tube prior to RNA extraction as an external control. cDNA was then generated, and

qPCR performed using primers listed in Table 3 and EvaGreen-SYBR master mix (Montréal

Biotech Inc.; MBI-E500). cDNA amplification was performed using a Rotor-Gene 6000 (Montréal

Biotech Inc., Montreal, QC) and analysed using software provided by the manufacturer. Fold-

changes in target transcript levels were determined by the 2-ΔΔCT method, using globin as the normalizer.

Immunoblotting

Oocytes were transferred to the base of a plastic microtube containing 10 μl of 2X Laemmli buffer

(Biorad 161-0737) and denatured by heating at 100 °C for 5 min. Samples were then used for

immunoblotting as previously described27. Antibodies used were against CNOT6 (Sigma

SAB2100457, 10 μg/mL), CNOT7 (Sigma WH0029883M1, 5 μg/mL), ORC6 (Cell Signalling

4737 S, 1:1000), SLBP (Abnova H00007884-M01, 200 ng/mL), tubulin (Sigma T8203, 10 μg/mL)

and TACC3 (Upstate Biotechnology Cat#07- 233, 1 ng/μL). Signals were generated using

enhanced chemiluminescence (Thermo Scientific 80196) and detected using film and a

phosphorimager (Amersham Storm). Intensities were quantified using ImageJ (National Institutes

of Health).

Whole-cell immunofluorescence

Oocytes were fixed for 20 min in a solution of 2% para-formaldehyde (Fisher Scientific 04042) in

phosphate buffered saline (PBS) containing 0.1% Triton X-100 (ACROS 9002-93-1) and stored

in a blocking solution (3% BSA + 0.1% Triton X-100 in PBS). Fixed oocytes were incubated

overnight at 4 °C on a shaker in primary antibody diluted appropriately in the blocking solution.

Following two 5-min washes in blocking solution on a shaker at room temperature, oocytes were

then incubated in secondary antibody diluted in the blocking solution for 1 hr at room temperature.

DRAQ5 (5 μM, New England Biolabs 4084S) and Phalloidin-TRITC (5 μg/mL, Sigma P1951)

were added to the secondary antibody solution to label the nucleus and the actin-cytoskeleton,

respectively. Oocytes were then washed twice with blocking solution, and mounted onto slides, in

a PBS drop covered with mineral oil using spacers (GBL654008, Sigma) to avoid crushing the

samples. Primary antibodies were: CNOT6 (1 μg/mL, Sigma SAB2100457), 18033 (1:10000, a

kind gift of Dr. M. Fritzler), and LSM14A (1:500, a kind gift of Dr. J. Lykke-Anderson).

Secondary antibodies were Alexa 488-conjugated goat anti-rabbit antibody (10 μg/mL, Life

Technologies A11008), used to detect CNOT6 and LSM14A, and Dylight 549-conjugated goat anti-human antibody, used to detect GW182. Preparations were mounted on glass slides and imaged using a Zeiss CLSM 510 confocal scanning laser microscope.

Cortical signal intensity of CNOT6 following siRNA injection was quantified using the CLSM

510 software. Two concentric circles were drawn outside and inside the oocyte cortical region, respectively. The areas outside the outer circle and inside the inner circle were masked and the signal within the remaining area (encompassed by the outer but not inner circle) was measured.

The mean signal intensity was then calculated for each experimental group and compared using the t-test.

Co-localization of CNOT6 and GW182 was analyzed using ImageJ. The Pearson correlation coefficient (Rr) and Mander’s overlap coefficient (R) were calculated to determine the correlation of incidence and percentage overlap, respectively, of the two signals.

Immunohistofluorescence

Ovaries were collected from 6-week old mice and fixed overnight in 4% para-formaldehyde at 4

°C. They were then washed three times for 15 min each in PBS and then stored at 4 °C. The specimens were then embedded in paraffin, sectioned, and mounted on glass slides. Antigen retrieval and antibody staining were performed as described (El-Hayek and Clarke 2015).

siRNA microinjection

Fully grown oocytes were transferred in groups of 15 to a pre-warmed 25-μl drop of MEM-HEPES

supplemented with dbcAMP in a 35-mm plastic culture dish and covered with mineral oil. The

dish was then moved to an inverted microscope (Zeiss; Observer.Z1 AX70) equipped with

micromanipulators (Ependorff, Transferman 4r). Using a micro-injector (Medical Systems Corp;

PL-100) as previously described (Yang et al. 2010), 5 μl of an siRNA solution was injected into

the cytoplasm of each oocyte. The following siRNAs were used: Cnot6 (20 μM, Thermo Scientific

1299001), Rspo1 (20 μM, Life Technologies MSS247584). Rspo1 is not detectable in oocytes (S.

El-Hayek, unpublished). Following injection, oocytes were incubated overnight in MEM

supplemented with dbcAMP. Oocytes to be analyzed at the GV stage were then incubated an

additional day in the presence of dbcAMP. Oocytes to be analyzed at metaphase II were transferred

to dbcAMP-free MEM and incubated an additional day. The oocytes were then used either for

whole-cell immunofluorescence or for RNA analysis by quantitative PCR or RL-PAT. cRNA injections

Luciferase-Slbp 3′-UTR constructs were generated using the previously describe methods (Yang et al. 2010). The entire Slbp 3′-UTR (starting from the stop codon to the end of the polyadenylation signal) was amplified using primers incorporating a XhoI site. The amplified product was then inserted into a pCS2+ plasmid (gift from Dr. M. Featherstone), downstream of a Luciferase ORF at the XhoI site. Alternatively, the plasmid was also PCR-amplified using primers that excluded the three putative PUF- binding elements (PBEs) located 92 nt–104 nt upstream of the polyadenylation signal, modelled after previously described methods. This generated the ΔPBE-

Slbp-3′UTR construct that lacked the three putative PBEs in the Slbp 3′-UTR. The constructs were

then amplified within a product that also included the SP6 promoter from the plasmids, using

primers that also incorporated a 30-nt long poly(A) tail at their 3′-end (Table 1). Both constructs were transcribed in-vitro to yield cRNAs with a 30-nt long poly(A) tail using the mMessage mMachine SP6 Transcription kit (Invitrogen; AM1340). cRNAs (100 nM) were then individually injected in GV oocytes.

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Acknowledgements

We thank Dr. M. Fritzler and Dr. J. Lykke-Anderson for generously supplying antibodies.

Supported by the grants from the Natural Sciences and Engineering Research Council of Canada

(RGPIN-402138), and the Research Institute of the McGill University Health Centre to H.J.C.

K.-F.V. was supported by Réseau Québécois en Reproduction CREATE scholarship. We thank

Dr. Qin Yang for her troubleshooting expertise and the other members of the Clarke lab (past and present) for their critical feedback.

Figures

Figure 12 – Patterns of mRNA polyadenylation during oocyte meiotic maturation. Immature

oocytes containing an intact germinal vesicle (GV) were obtained from antral follicles and

collected immediately (GV), or allowed to mature for 6 hr (germinal vesicle breakdown, GVBD)

or 12–15 hr (metaphase II, MII). Poly(A) tail length of the indicated mRNAs was analyzed used

the RL-PAT technique. Red bars indicate the product length that would be generated using a non-

adenylated mRNA. Molecular weight markers (100 bp ladder) are shown in left-most lane. (A)

Actb and Ngfr become progressively deadenylated. (B) Ccnb1 becomes progressively polyadenylated. (C) Slbp and Orc6 become polyadenylated and subsequently deadenylated.

Figure 13 – Expression of deadenylases in oocytes. Oocytes were obtained as in Fig. 1. (A) mRNAs were assayed using the polymerase chain reaction and the primers shown in Table 2. mRNAs encoding all major deadenylases are present in both immature (GV) and mature (MII) oocytes. Actb is used as a loading control. (B) CNOT6 and CNOT7 protein were assayed by immunoblotting using commercially available antibodies. Panels shows representative blots.

Figure 14 – Localization of CNOT6 in growing and fully-grown oocytes. (A) Whole-cell immunofluorescence of growing, fully grown and mature (MII) oocytes. Foci of CNOT6 staining are detectable at the periphery of the oocyte. (B) Whole-cell immunofluorescence of fully grown oocytes co-stained using anti-CNOT6 (green) and phalloidin (red) to reveal the actin-rich cortex.

CNOT6 foci are predominantly located within the cortex of the oocytes. The nucleus is labelled using DRAQ5 (blue). (C) Immunohistochemistry of sections of paraffin embedded ovaries confirms the cortical distribution of CNOT6 foci (arrowheads) in oocytes. Bar = 20 μm.

Figure 15 – Co-localization of CNOT6 foci and cytoplasmic ribonucleoprotein (RNP) markers.

(A) Whole-cell immunofluorescence of fully grown GV-stage oocytes stained for RNP markers

18033 (left panel) or LSM14A (right panel). Both are localized in the cortex. Bar = 20 μm. (B)

Whole-cell immunofluorescence of fully grown GV-stage oocytes co-stained for CNOT6 (green) and 18033 (red). Merged image shows overlapping foci. Bar = 5 μm. Overlap was quantified using

ImageJ (R = 0.73 ± 0.02, Rr = 0.32 ± 0.02; n = 12).

Figure 16 – Reduction of CNOT6 impairs the deadenylation of Slbp and Orc6 during late

maturation. (A) Fully grown immature oocytes were injected with siRNA targeting Cnot6 or a control or were left uninjected. The following day, oocytes were collected and Cnot6 and Actb mRNA were assessed using RT-PCR. (B) Upper: Oocytes were injected as above, incubated for

24 hr, then stained using anti-CNOT6. Lower: Quantification of cortical fluorescence in oocytes injected with a control (39 oocytes) or Cnot6 siRNA (41 oocytes). Oocytes used for quantification were collected in 3 independent experimental replicates. Asterisk indicates p < 0.05. (C) Fully grown immature oocytes were injected with the indicated siRNA and collected for analysis at the

GV stage or after maturation to metaphase II (MII). RL-PAT was used to assess the poly(A) tail length of the indicated mRNAs. Left side shows representative images of gels after electrophoretic separation of PCR products. Right side shows quantification of gel from top (longer poly(A) tail) to bottom. Cnot6-siRNA injected oocytes shown in red, control siRNA-injected oocytes in blue.

The experiment was repeated 3 times.

Figure 17 – Terminal portion of the 3′-UTR of different mRNAs showing location of sequences that potentially regulate polyadenylation. Locations are indicated relative to the polyadenylation signal (PAS, black bar; sequence: AUUAAA). Orc6 and Slbp mRNAs contain one or more potential PBE sequences (blue bars) located between the CPE (red bar) lying closest to the PAS and the PAS itself, whereas Actb, Ngfr and Ccnb1 do not.

Figure 18 – Luciferase-Slbp 3′-UTR constructs and injections. (A) The luciferase ORF was ligated

either to the wild-type Slbp-3′UTR (upper construct) or to a mutant Slbp 3′UTR lacking the three

putative PBE sequences downstream of the URS-CPE (lower construct). (B) mRNA synthesized in vitro was injected into fully grown immature oocytes, which were maintained at the GV-stage

or allowed to mature and collected after GVBD or at MII. The length of the poly(A) tail was

assessed by RL-PAT using primers that do not amplify cDNA derived from the 3′-UTR of

endogenous Slbp mRNA. (C) Line graphs show the length of the cDNA products derived from the

injected mRNAs. wt-Slbp-3′UTR shown in blue, ΔPBE-Slbp-3′UTR in red. The experiment was

performed 2 times.

Figure 19 – Increasing poly(A) tail-length does not affect mRNA translation or degradation. (A)

Left: Immunoblot showing relative quantity of ORC6 and SLBP in immature (GV) and mature

(MII) oocytes. TACC3 is a loading control. Right: Immature oocytes were injected with control

(Ctrl) or Cnot6 siRNA and either maintained in the immature state (GV) or allowed to mature

(MII), then immunoblotted. (B) Relative quantity of the indicated mRNAs at different stages of maturation. Data analysed using ANOVA with Tukey HSD; different letters above bars indicate statistically significant difference (p < 0.05). (C) Relative quantity of the indicated mRNAs in mature oocytes following depletion of CNOT6 as in (A). Data analysed using t-test.

Figure 20 – Model for differential poly(A) tail length dynamics during meiotic maturation. In immature (GV) oocytes, the CPE of Ccnb1, Orc6 and Slbp provides a binding site for CPEB, which directly or indirectly recruits the CCR4-NOT deadenylase complex. After maturing oocytes undergo GVBD, CPEB becomes phosphorylated. This may inhibit its ability to interact with

CCR4-NOT. CPEB nonetheless prevents association of PUM proteins with the nearby PBE sites.

In the absence of an associated CCR4-NOT activity, polyadenylation is favored. Later during maturation, CPEB becomes degraded. This permits PUM proteins to associate with the previously masked PBE sites on Orc6 and Slbp, enabling CCR4-NOT to be recruited to the mRNA. Thus, the mRNA becomes deadenylated. In Ccnb1, by contrast, which lacks a PBE in the appropriate location, CCR4-NOT is not recruited and the mRNA continues to be polyadenylated.

Table 1: PCR primer sequences

Transcript Sequence

F: 5'-GCGACCCGGCAGTTATACGG-3' Cnot6 R: 5'-AAGTGATGTGGGCAGCCGC-3' F: 5'-TGGCGCGCTGAATGCAGACTAA-3' Cnot6l R: 5'-TAGCTGGAAGAGCCGGCCAAGT-3' F: 5'-CCCCGGTGGCTTTCGTCCTC-3' Cnot7 R: 5'-CACACCGGTCCCTACACGCG-3' F: 5'-ACAGGGGTGGCCCAGAAGCA-3' Cnot8 R: 5'-TGGTGGCAGGAGGCAGCAGA-3' F: 5'-ATGGCCCGTGGAGGGCTCAT-3' Pan2 R: 5'-CACAGGTGGTGCTCGCCTGG-3' F: 5'-ACTGAAGGAGCAGGCGCTGC-3' Pan3 R: 5'-TCCCTCCAGAGTCAGGCCGC-3' F: 5'-AGGACGCTGAGTCCCGACCC-3' Parn R: 5'-GCTGCTGCACCGTGAACCCT-3' F: 5'-TTGTCGCGGGTGCGGACTTC-3' Pnldc R: 5'-CCAGCGTTGTGGCTGGCTCA-3' F: 5'-TAGGGACTCAGGCTATCGCA-3' Cnot1 R: 5'-ATGAACTTGGTTCAGCGGAC-3' F: 5'- Cnot2 GGCAACCCAACTCCATTAATAAACC-3' R: 5'-CCCTGGTAATCCATACCCGC-3' F: 5'-CCCTGAATCCCCTACTCGGT-3' Cnot3 R: 5'-TTGTCCAGCATTCGCAGGAT-3' F: 5'-TGGGAGATGAAGGGTGTGGA-3' Orc6 R: 5'-AATGTCCATGTGAGGTGCCC-3' F: 5'-ACCCACCAAAGTCAGACACG-3' Slbp R: 5'-CCAGGCAGAGCCACGTATTT-3' F: 5'-GGCTGTATTCCCCTCCATCG-3' Actb R: 5'-CCAGTTGGTAACAATGCCATGT-3' Luciferase- F: 5'-CCCAAGCTTGATTTAGGTGAC-3' Slbp R: 5'-(T)30CAGTTAAAGGGTCTTTA-3'

100

Table 2. RNA-Linker and RL-PAT primer sequences

Min length Transcript Sequence (bp) RNA 5’-P-CGUAGGCUCAGCUCGGAAUC-ddC-

Linker 3’ Linker primer 5’-GGATTCCGAGCTGAGCCTACG primer outer primer: 5'- GTTGGTTGGAGCAAACATCC-3' Actb 273 inner primer: 5'- GGAAGGTGACAGCATTGCTT-3' outer primer: 5'- TCCTTGGCCTGTTCTGTTTT-3' Ngfr 451 inner primer: 5'- ACAGACTGACTGCCATCCCT-3' outer primer: 5'- AAAAGAATTTGCCCCCAAGT-3' Ccbn1 348 inner primer: 5'- CAGCATTCCTTTCAATGCCT-3' outer primer: 5'- TGCAGTCTGTCTGGGAGATG-3' Orc6 153 inner primer: 5'- TACAGACAGTGAGCGTTCC-3' outer primer: 5'- GCCTTCAGTTGCCACTTTTC-3' Slbp 347 inner primer: 5'- GCTCTGGACAAAGGATGCTAA-3' outer primer: 5'- Luciferase- AGATCGTGGATTACGTCGCC-3' 347 Slbp inner primer: 5'- GCTCTGGACAAAGGATGCTAA-3'

101

Table 3. qPCR primer sequences

Transcript Sequence

F: 5'-GGCTGTATTCCCCTCCATCG-3' Actb R: 5'-CCAGTTGGTAACAATGCCATGT-3' F: 5'-GACTCCTTTACCCACGAGGC-3' Ngfr R: 5'-TGGTCAGAAGCAAAGGTCCC-3' F: 5'-CTGCACTTCCTCCGTAGAGC-3' Ccnb1 R: 5'-TGGTGTCCATTCACCGTTGT-3' F: 5'-TTTCCATCTCACTGCAGGCAT-3' Orc6 R: 5'-AATGGTCCCAATTTCACCCCA-3' F: 5'-ATGGCCTGCAGACCTAGAAG-3' Slbp R: 5'-CTGGCCCAGTCAGAACATCT-3' F: 5'-GCAGCCACGGTGGCGAGTAT-3' Hba1 R: 5'-GTGGGACAGGAGCTTGAAAT-3'

102

Supplemental data

A. Full-length gels corresponding to Figure 11.

Actb Ngfr Ccnb1 Orc6 Slbp GV GVBD MII GV GVBD MII GV GVBD MII GV GVBD MII GV GVBD MII

B. Full-length gels corresponding to Figure 12A. Red line in lane L is 600 nt marker. Lanes correspond to oocytes at early, mid- and fully grown stages, and mature eggs. Lanes in red are shown in the manuscript.

Figure 21

103

75KDa C. Full-length gels corresponding to Figure 12B. Left: Blot probed for CNOT6 and then for tubulin. The CNOT6 same blot was cut as shown to blot 50KDa for CNOT7. Right: Sample was 50KDa CNOT6 divided into two aliquots that were Tubulin run on separate gels. One was blotted for CNOT6, the other for tubulin. This blot is shown in the manuscript. 50KDa CNOT7 Tubulin

D. Full-length gels corresponding to Figure 15C. Note the images are displayed in reverse black-white in the manuscript. Note lanes are switched in the Orc6-GV panel

Cnot6 Actb

E. Full-length gel corresponding to Figure 15A.

1kb

Figure 22

104

wt-Slbp-3'UTR ΔPBE-Slbp-3'UTR L F. Full-length gels corresponding L GV GVBD MII GV GVBD MII to Figure 17B.

GV MII G. Full-length gels corresponding GV MII Ctrl Cnot6 Ctrl Cnot6 to Figure 18A (ORC6). Dotted line shows where gels were cut before immunoblotting with indicated antibody.

50KDa TACC3 75KDa

50KDa 75KDa

ORC6

GV MII GV MII Ctrl Cnot6 Ctrl Cnot6 H. Full-length gels corresponding

to Figure 18A (SLBP). Dotted line shows where gels were cut before immunoblotting with indicated antibody. 50KDa TACC3 75KDa

50KDa 75KDa

SLBP

Figure 23

105

Chapter 3

[Manuscript II] Maturing oocytes lose the ability to form 18033 granules and protect translation after cold-shock via a CDK1-dependent process

106

Preface

When oocytes are incubated or manipulated in-vitro, the rates of successful maturation,

embryogenesis and live births significantly decrease. This implies that oocytes, under these

conditions, are stressed and compromised. In somatic cells, and in the germ line of other species

such as C. elegans, stress induces the formation of RNPs, called stress granules. In MANUSCRIPT

I, we described the presence of cortical P-body-like foci believed to be the site of mRNA storage in mammalian oocytes. These RNPs are only observed in physiological conditions and therefore differ from canonical stress granules. Stress granule assembly and disassembly, like for most

RNPs, depend on the dynamics of it components. These dynamics can be modulated by

temperature, pH, osmolarity, protein concentration and post-translational modifications (Jud,

Czerwinski et al. 2008, Schisa 2012). Understanding how such granules form in oocytes may help determine how strenuous environments impair oocyte competence and quality. To investigate this, the following objectives were designed:

1- Determine whether stress granules can form in mammalian oocytes.

2- Characterize these stress granules.

3- Determine their role in oocytes.

This work is discussed in Chapter 3 (Manuscript II). Chapter 3 is currently being prepared for the publication.

107

Significance statement

Stress granules play an important role in regulating gene expression in stressed cells. On the one

hand, they repress the translation of non-essential factors and promote the expression of stress gene

transcripts. On the other hand, they are believed to protect the transcriptome. Oocytes that are

manipulated in vitro are subject to variable environmental parameters (e.g. temperature, pH,

glucose availability, etc.). They are therefore subject to stress. Here we describe the formation of

a novel type of stress granule regulated by CDK1 and the cell cycle in mammalian oocytes. We then provide evidence for a role for these 18033 granules in the preservation of translational capability during the recovery period following stress. Oocyte manipulated at stages when they cannot form these stress granules are therefore susceptible to significant and permanent deterioration of their translational ability.

108

Maturing oocytes lose the ability to form 18033 (EDC4) granules and protect translation after cold-shock via a CDK1-dependent process Karl-Frédéric Vieux1,3 & Hugh J. Clarke1,2,3

1Department of Biology, McGill University, Montreal, Quebec, Canada.

2Department of Obstetrics and Gynecology, McGill University, Montreal, Quebec, Canada.

3Research Institute of the McGill University Health Centre, Montreal, Quebec, Canada.

Manuscript in preparation for submission to Human Reproduction.

Keywords: EDC4, RNP, stress granule, translation, oocyte, ART, IVF

Corresponding Author: Hugh Clarke MUHC Research Institute 1001 Decarie Blvd, Block E-M0.2218, Montreal, QC CANADA H4A 3J1 [email protected]

Competing Interests: The authors declare no competing interests.

109

Abstract

In vitro fertilization (IVF) of oocytes is an increasing recourse in couples seeking assistance in their reproductive needs. Despite decades of use and progress, however, the process remains

relatively inefficient. Procedures associated with IVF are potential sources of stress that may

damage the oocyte. Maturing oocytes harbour large quantities of messenger RNAs (mRNAs)

whose activity must be correctly regulated. Strikingly, many cells under stress store mRNAs in

ribonucleoprotein particles (RNPs) called stress granules, for safekeeping. In this study, we cold-

shocked growing and fully grown meiotically immature oocytes at 4oC and describe the formation

of cytoplasmic foci containing the canonical RNP marker EDC4, labelled by the 18033 antiserum.

Like stress granules, these 18033 granules require stable microtubules to assemble and disappear

after the oocytes are returned to 37oC. Cold-shock also induces 18033 granules at the time of germinal vesicle breakdown (GVBD). In contrast, 18033 granules cannot be induced at metaphase

I or metaphase II, even though the EDC4 protein remains detectable by immunoblotting. When oocytes are cultured in the presence of the CDK1 inhibitor roscovitine, the ability to form 18033 granules is retained late in maturation. In addition, while the resumption of translation after cold- shock is impaired in oocytes that cannot form stress granules, it is retained in oocytes where the ability to make 18033 granules is preserved. These results therefore demonstrate a role for these

stress granules in protecting gene expression in stressed oocytes.

110

Introduction

The term assisted reproductive technologies refers to various clinical procedures and methods in

which gametes or embryos are handled in-vitro for the purpose of facilitating fertility. In-vitro fertilization (IVF) is one such technique, where female patients are typically subjected to daily rounds of FSH injection for up to 2 weeks to stimulate antral follicular growth. A trigger injection of hCG (or an hCG analogue) to promote oocyte maturation is then administered and is followed

by egg retrieval using an ultrasound guided needle. Egg collection is strictly coordinated with the

time of the trigger injection to avoid ovulation (Shrestha, La et al. 2015). The male and female

gametes are then selected in incubated together in-vitro, allowing for fertilization to occur.

Alternatively, the sperm head can also be separated from the tail and injected directly into the

oocyte in a procedure known as intra-cytoplasmic sperm injection (ICSI) (Zini 2018). The

resulting embryo is then incubated for several days before it is transferred back into the patient.

The use of IVF and other assisted reproductive technologies is a rapidly growing industry,

expected to reach over $31.4 billion annually by 2023 (Global Market Insight 2016). Despite the

growing interest in such practice, the success rates of IVF remain relatively low. In Canada for

instance, the number of live births per embryo transfer after IVF for women under 35 years of age,

was only 41% in 2016; falling to as low as 6% in women 43 years of age or older (CFAS 2016).

With global rates of infertility increasing (Mascarenhas, Flaxman et al. 2012), and a recent

recourse to social egg freezing made more available (CFAS 2016), more efficient methods are

required to alleviate the socio-economic burden on women seeking treatment.

Since the original use of IVF in the 1970s, many changes have been made to the protocol with

regards to the handling of gametes and the resulting embryo, consistently improving the outcome

(Wade, MacLachlan et al. 2015). The guidelines for oocyte and embryo selection, however, remain

111 superficial and rudimentary (Mikkelsen and Lindenberg 2001). This is in part due to the fact that the evaluation of gamete or embryo quality and competence is limited by the invasiveness of alternative methods required to assess more novel cellular and molecular parameters. In the meantime, understanding how environmental cues regulate the oocytes meiotic and developmental competence is therefore essential in improving culture methods for fertility treatments. Embryo quality and survival are dependent on gamete development (Eppig and Schroeder 1989, Balaban and Urman 2006, Sakkas, Ramalingam et al. 2015). This is particularly true for the first mitotic cell cycles, where early embryogenesis relies solely on maternally deposited factors, acquired during oocyte growth. Transcription becomes undetectable at the onset of oocyte maturation, only to resume in the late 2-cell embryo in mice. The pressure on the oocyte to properly manage its assets is therefore critical for the successful initiation of embryogenesis (Kang and Han 2011,

Clarke 2012, Ma, Flemr et al. 2013, Jansova, Tetkova et al. 2018). These factors are accumulated in the form of protein and RNA whose activity and expression are finely tuned to orchestrate the complex molecular events that comprise maturation, fertilization and early embryogenesis. The expression of transcripts and the activity of proteins during maturation and early embryogenesis are then dependent on transcription during oocyte growth, post-transcriptional regulation of translation and posttranslational modifications during oocyte maturation and early embryogenesis.

Regulatory events in the oocyte can also be modulated by extrinsic factors to coordinate oocyte development with changes in the environment. Under physiological conditions, somatic cells of the follicle, termed the granulosa cells, mediate the arrest of the oocyte at prophase I by producing cGMP (Zhang, Su et al. 2010). Subsequently, cGMP diffuses into the oocyte to prevent the activation of the kinase cascade responsible for the resumption of meiosis, allowing the oocyte to grow before it begins maturing. The presence of granulosa cells at the end stages of growth also

112 support structural chromatin changes in the germinal vesicle conformation from a non-surrounded nucleolus (NSN) to a surrounded nucleolus (SN), accompanying the downregulation of transcription (Sun, Zhu et al. 2016). Additionally, FSH signaling through the granulosa cells, upregulates the translation of transcripts linked to oocyte quality and competence through a transient activation of the phosphatidyl-inositol 3-phosphate/AKT cascade in the oocyte

(Franciosi, Manandhar et al. 2016). In-vitro culture of fully grown oocytes in the absence of granulosa cells, such as seen in in-vitro maturation (IVM) protocols, results in the retention of the

NSN stage. Transcriptional sites remain uniformly distributed in the nucleus and the transcriptional activity is prolonged (De La Fuente and Eppig 2001). In the absence of cGMP coming from the granulosa cells, fully-grown oocytes also resume meiosis (Szybek 1972, Zhang, Su et al. 2010). in-vitro maturation also significantly impairs developmental competence, affecting the oocyte post-transcriptionally, in the absence of transcription (Nogueira, Sadeu et al. 2012). In-vitro manipulations, like the procedures associated with IVF, are therefore potential sources of stress that may compromise the oocyte. Yet the mechanisms by which stress modulates quality and competence of mammalian gametes and embryos remain mostly elusive.

Ribonucleoprotein particles (RNP) are non-membrane-bound cellular organelles where RNA and

RNA-binding proteins aggregate. They are evolutionarily conserved structures found to regulate various aspects of RNA metabolism in most cell types (Warner, Knopf et al. 1963). Diverse types of RNPs are implicated in different cellular functions, including active translation, RNA decay, symmetry breaking and more (Cougot, Babajko et al. 2004, Hasler and Strub 2006, Lee, Occhipinti et al. 2015). Stress granules are an induced type of dynamic RNP that form when translation is stalled at initiation (Kimball, Horetsky et al. 2003), and can be induced by extreme temperatures, arsenic poisoning, glucose deprivation and many other conditions that result in suboptimal cell

113

conditions (reviewed in Buchan and Parker 2009). Stress granule assembly and disassembly are

generally believed to form through liquid-liquid phase separation like most RNPs (Hyman, Weber

et al. 2014). They are also believed to mediate translational control and protect the transcriptome

during stress. It is important to mention that while some RNPs have been observed, it is currently

unknown whether stress granules specifically, can form in mammalian oocytes (Flemr, Ma et al.

2010, Vieux and Clarke 2018).

In this study, we examined whether mammalian oocytes can form stress granules. We described

in cold-shocked oocytes, the induction of novel stress granules, positive for EDC4, using the 18033

antiserum. EDC4 is a common RNP marker and a regulator of DCP2 activity, associated to the

decapping complex (Flemr, Ma et al. 2010, Chang, Bercovich et al. 2014). We also identified

CDK1 as a key regulator of granule formation in response to the cold stress during maturation.

Consequently, growing oocytes form 18033 granules in response to cold-shock but mature oocytes

lose that ability. We then used a reporter gene assay to assess translation in stressful conditions.

During their recovery, if oocytes form 18033 granules during the cold-shock, they resume

translation unabashed. In contrast, if cold-shocked at later stages of maturation, when they cannot

form the foci, translation is permanently impaired. It can however be rescued in oocytes cold- shocked at the later stages of maturation, if they are treated with the CDK1 inhibitor and retain the ability to make 18033 granules. Our results thereby provide the first evidence for the involvement of stress granules in the preservation of translational capacity while the oocyte is under duress.

Materials and Methods

Mice

114

All experiments were performed using CD1 mice (Charles River Canada, St.-Constant, QC,

Canada) in compliance with the regulations and policies of the Canadian Council on Animal Care and were approved by the Animal Care Committee of Research Institute of the McGill University

Health Centre (RI-MUHC) (protocol 5168).

Oocyte collection

The ovaries of female pups were collected and fragmented in MEM (Life Technologies) buffered

at pH 7.2 using HEPES (MEM-H), containing sodium pyruvate (0.25 mM, Sigma), penicillin G

(63 mg/L, Sigma), streptomycin (50 mg/L, Sigma), and bovine serum albumin (BSA, 1mg/mL,

Sigma). Growing follicles were collected from PD12 pups after pipetting ovarian fragments up

and down in media supplemented with collagenase (100 μg/ml, Worthington MAC7120). Fully

grown oocytes were collected from PD18 pups, opting out of the enzymatic digestion of the ovary,

but instead isolating oocytes by bursting follicles with 28G insulin needles and pipetting at regular

intervals. To prevent meiotic maturation, fully grown oocytes were collected in the presence of

dibutyryl cyclic-AMP (dbcAMP, 0.1 mg/ml, Sigma) in the collection or the culture media. For the

collection of GVBD and MI oocytes, fully grown oocytes were transferred to MEM-NAHCO3

containing pyruvate, antibiotics and bovine serum albumin, and without dbcAMP (allowing for

the resumption of meiosis to begin). These oocytes were then incubated at 37ºC in 5% CO2 (in air) for 4 hr or 12-16 hr to get GVBD and MII oocytes respectively.

Fully grown, primed GV oocytes were collected from PD18 pups injected with 7IU of the FSH

analogue, pregnant mare serum gonadotropin (PMSG; Prospec - hor-272-a), 44-48h before

collection. The collection was done in the same way as unprimed fully-grown oocytes as described above. PMSG-injected mice were also injected with 5IU of human chorionic gonadotropin (hCG;

115

Sigma – C1063-1VL) 44-48h after the PMSG injection, to induce oocyte maturation and ovulation.

Expanded cumulus-oocyte complexes (COCs) were collected from the oviduct of these mice, 16hr post hCG injection, in MEM-HEPES. Ovulated in-vivo MII oocytes were then isolated by supplementing the media with 20µL hyaluronidase (Sigma – H4272; 10mg/mL) and pipetting briefly to break down the complexes.

Cold-shock and chemical treatment of oocytes

For cold-shock oocytes were transferred to 1mL of fresh MEM-HEPES in a 35mm dish and incubated at 4ºC for 5 hr or overnight. GV oocytes were cold-shocked in the presence of dbcAMP

(0.1 mg/ml) unless otherwise specified. Control incubations were performed in MEM-NAHCO3

at 37ºC in parallel to cold-shock. Recovery experiments were also carried out in MEM-NAHCO3

at 37ºC overnight after an initial 5-hr cold-shock. For treatment cultures, oocytes were pre- incubated at 37ºC in MEM- NAHCO3 supplemented with nocodazole (5μg/mL; Sigma M1404),

latrunculin A (5 μM; Sigma L5163) or roscovitine (10µM; Sigma – R7772) for 2hr. They were then transferred to MEM-HEPES supplemented with the same drug and incubated at 4ºC for 5 hr.

Whole-cell immunofluorescence

Follicles and oocytes were fixed for 20 min in a solution of 2% para-formaldehyde (Fisher

Scientific 04042) in phosphate buffered saline (PBS) containing 0.1% TritonX-100 (ACROS

9002-93-1) and washed in a blocking solution (3% BSA + 0.1% TritonX-100 in PBS). Samples

were then transferred into a primary antibody solution (3% BSA + 0.1% TritonX-100 in PBS with

antibody diluted accordingly) and incubated overnight at 4ºC on a shaker. The following primary

antibodies were used as indicated: CNOT6 (1 μg/mL, Sigma SAB2100457), EDC4 (1:100; Cell

Signaling Technologies 2548S), TNRC6A (2.5μg -5μg/mL; Sigma SAB2102506), G3BP

116

(2.5μg/mL; Abcam ab56574), TIA1 (5μg/mL; Abcam ab40693) and the 18033 antiserum

(1:10,000, a kind gift of Dr. M. Fritzler). Two 10-min washes in blocking solution followed, on a

shaker at room temperature. Oocytes were again transferred into an antibody solution with the

corresponding secondary antibodies (at appropriate concentrations) and the nuclear marker

DRAQ5 (5μM, New England Biolabs 4084S) for 1hr at room temperature. Secondary antibodies

included Alexa 488-conjugated goat anti-rabbit antibody (10 μg/mL, Life Technologies A11008),

and Alexa594 goat anti-human antibody (Invitrogen – A21216). Finally, oocytes were washed

twice more in blocking solution for 10min on a shaker and mounted onto glass slides to be imaged through a Zeiss CLSM 880 confocal scanning laser microscope as previously described (Vieux and Clarke 2018).

Immunoblotting

Oocyte samples were lysed in 10μL of 2X Laemmli buffer (Biorad 161-0737) and the lysates denatured at 95°C for 5min. Samples were then used for immunoblotting as previously described

(Yang et al. 2010). The 18033 antiserum (1:10,000, a kind gift of Dr. M. Fritzler) and a tubulin antibody (Sigma T8203, 10 μg/mL) were used to blot the membranes. Signals were detected using

Supersignal West Pico PLUS chemiluminescence (Thermo Scientific 34577) and detected using a phosphoimager (Amersham Imager 600). cRNA microinjection

Two constructs were generated and transcribe in-vitro with a ~30nt long poly(A) tail as previously described: first with the Slbp 3′-UTR and second with the Ccnb1 3′-UTR (Vieux and Clarke 2018).

A 30-µL drop of prewarmed MEM-HEPES with 3% polyvinylpyrolidone (supplemented with dbcAMP) and a 2-µL drop of cRNA solution were plated on a 35mm plastic culture dish and

117

covered with mineral oil. Fully-grown GV oocyte were transferred into the pre-warmed drop of

MEM-HEPES and the dish moved to an inverted microscope (Zeiss; Observer.Z1 AX70) equipped

with micromanipulators (Ependorff, Transferman 4r). Using a micro-injector (Medical Systems

Corp; LP-100) approximately 5 µl of cRNA solution was injected into the cytoplasm of roughly

200 oocytes. Oocytes were then transferred, incubated and kept overnight at the GV stage in MEM-

NAHCO3 supplemented with dbcAMP.

Luciferase Assay

Oocytes were transferred into MEM-NAHCO3 in the absence of dbcAMP and allowed to initiate

maturation 16-18hr after cRNA injections. 15 oocytes were collected and lysed in 20 µl of lysis buffer (Promega E4030) at room temperature, before the onset of maturation, 3 hr into maturation at the germinal vesicle breakdown, 8hr into maturation at metaphase I, 13 hr and 19 hr into maturation at metaphase II. 30 additional GVBD and MI oocytes were transferred and cold- shocked in MEM-HEPES. Again 15 oocytes were collected and lysed at room temperature right after (t=0) or 12 hr after both stage-specific cold-shocks (t=12hr). Oocyte lysates were stored at -

20ºC until luciferase assay. 100 µl of LARII (Promega E4030) was added to the wells of a 96-well plate (VWR cat # 624202-969 and #77776-512). 20 µl of oocyte sample lysates were then transferred to the single LARII wells and mixed gently by pipetting. A luminometer (GloMax 96 microplate luminometer – Promega E6501) was then used to measure the luciferase activity of each well. Data was collected and calibrated with the background signal (luciferase activity in a well with 100 µl of LARII and 20 µl of un-injected oocyte cell lysate). The luciferase activity was reported for single experiments. The fold increase was pooled from three experiments and normalized to the average fold increase in luciferase activity of control (non-cold-shocked) untreated oocytes.

118

Results

Inducible 18033 granules form in fully grown GV oocytes after cold-shock

To determine if the formation of ribonucleoprotein complexes (RNPs) can be induced in oocytes in response to a stress stimulus, fully grown oocytes were cold-shocked for 5 hr at 4ºC and stained with 18033 antiserum, a common RNP marker (Flemr, Ma et al. 2010). Cold-shock was used as a model for stress because temperature is a parameter that is easily manipulated and reproduced. We previously described the presence of cytoplasmic foci marked by the 18033 antiserum and the P- body marker CNOT6 that were concentrated in a narrow ring at the cortex of oocytes at 37ºC

(Vieux and Clarke 2018). Here, we began by confirming the cortical pattern of 18033 staining in

fully grown oocytes at 37ºC (Figure 24A, top left panel). Following the 4ºC cold-shock, we

observed a dramatic change in the distribution of the RNP marker. Most of the small foci at the

cortex disappeared and instead, larger dispersed cytoplasmic foci formed (Figure 24A, top right

panel). Because the 18033 antiserum detects both EDC4 and TNRC6A signals (Bloch, Gulick et

al. 2006), we also stained oocytes using specific antibodies. While both signals overlapped with

18033 at the cortex of oocytes at 37ºC, only the EDC4 signal overlapped with the cold-induced

foci detected by 18033 at 4ºC (supplemental; Figure 30). Moreover, the cytoplasmic foci observed

in oocytes at 4ºC did not stain for CNOT6 (Figure 24B). Thus, cold-shock induces the formation of cytoplasmic foci containing EDC4.

Inducible 18033 granules are non-canonical stress granules

The number of cold-inducible 18033 granules varied within a population of oocytes (data not shown). However, the fraction of oocytes within a population that exhibited these foci in response to the cold-shock was consistent, at about 35% (Figure 24C). This frequency, defined as the

119 percentage of oocytes positive for inducible 18033 granules within a population of cold-shocked cells, was therefore chosen as the parameter to study this phenomenon. To determine if the inducible 18033 granules are permanent structures that persist after the cold-shock, oocytes were subjected to a 5-hr cold-shock, then returned to 37ºC or maintained at 4ºC, and incubated overnight. Whereas foci remained visible in oocytes that were maintained at 4ºC overnight (Figure

24A bottom right panel; 24C), they were rarely observed in oocytes that were returned to 37ºC overnight. These instead regained the cortical ring of small 18033 foci characteristic of non- shocked oocytes (Figure 24A, bottom left panel; 24C). Cold-inducible 18033 granules are therefore non-permanent structures that disassemble when oocytes are returned to physiological temperature.

The inducible and reversible nature of the 18033 granules observed at 4ºC are reminiscent of stress granules. Different chemical treatments can induce the formation of stress granules including

MG132 (Aulas, Fay et al. 2017). Accordingly, when treated with MG132 for 5hr, full-grown oocytes also formed the 18033 granules (supplemental; Figure 31). Formation of stress granules typically depends on an intact microtubule network, which likely mediates retrograde transport of structural elements (Loschi, Leishman et al. 2009). Consequently, destabilizing microtubules can significantly impair stress granule formation (Ivanov, Chudinova et al. 2003). We therefore assessed the role of microtubules in the formation of the cytoplasmic 18033 granules using the microtubule-destabilizing agent, nocodazole. Foci were observed in oocytes from all experimental groups after cold-shock (Figure 25A). In oocytes treated with nocodazole, however, inducible

18033 granules formation was significantly hindered, forming only in 16% of cold-shocked oocytes (Figure 25B). In comparison, after treatment with latrunculin, an actin microfilament destabilizing agent, 35% of cold-shocked oocytes formed 18033 granules; comparable to the 34%

120

of cold-shocked untreated oocytes (Figure 24D). The disruption of microtubules specifically,

hinders inducible 18033 granules formation. Microtubules are therefore required for the formation

of the cold-induced 18033 granules, providing further evidence of their similarity to stress

granules.

Oocytes lose the ability to form cold-induced 18033 granules during maturation

To understand the mechanisms regulating this response to the cold-shock, we assessed the formation of the cold-induced 18033 granules at different stages of oogenesis. Foci could be

induced in growing oocytes enclosed within follicles (Figure 26A, B), suggesting that this ability

was a general property of oocytes. To assess the response of oocytes during meiotic maturation,

fully grown oocytes were collected and allowed to initiate maturation in vitro. Samples were

collected at defined time-points and subjected to cold-shock. At the time of GVBD, oocytes retained the ability to form cold-induced foci. In contrast, oocytes at metaphase I or metaphase II did not form EDC4 granules in response to cold-shock (Figure 26A, B). We then tested whether the loss of ability to form foci was a consequence of maturation in vitro. Whereas immature oocytes collected following eCG-priming formed 18033 granules at a high frequency following cold-shock, ovulated oocytes were unable to do so (Figure 26C, D). The loss of the ability to make

18033 granules is therefore characteristic of maturation and is not a byproduct of the in-vitro incubation.

To determine whether the absence of inducible foci was due to a loss of EDC4 protein during maturation, fully grown immature oocytes, GVBD oocytes and mature metaphase II eggs were used for immunoblotting. Using the 18033 antiserum, both EDC4 and TNRC6A were detected at all three stages (Figure 26E). In fact, an additional band appears above the EDC4 signal of oocytes at metaphase II. This is perhaps the result of post-translational modifications to the EDC4 protein.

121

Thus, the loss of 18033 granules at the later stages of maturation is not due to a decrease in the

amount of the protein, but may result from a modification of the protein. Oocytes can therefore

form 18033 granules in response to cold-shock during growth and at the early stages of maturation

but lose this ability during the latter part of maturation.

CDK1 regulates the ability to form 18033 granules

Cyclin-dependent kinase 1 (CDK1) activity drives maturation (Adhikari, Zheng et al. 2012). We

therefore hypothesized that high CDK1 activity may trigger the loss of the ability to form cold-

induced 18033 granules. To test this, oocytes at different stages of maturation were treated with

the CDK1 inhibitor roscovitine for 2 hr, then subjected to cold-shock. As expected, immature

oocytes treated with roscovitine failed to undergo GVBD. Moreover, a high percentage formed

18033 granules following cold-shock (Figure 27A, B – left panels). No foci were observed in

oocytes maintained at 37oC (data not shown). Similar results were observed when oocytes were treated following GVBD (Figure 27A, B – middle panels). Strikingly, when oocytes were treated

near metaphase I, they precociously resumed meiosis and retained their ability to form 18033

granules in response to stress (Figure 27A, B – left panels). These results suggest that the increase

in CDK1 activity that drives maturation also causes oocytes to lose the ability to form 18033

granules in response to cold-shock.

The ability to resume translation after cold-shock requires the ability to form 18033 granules

Stress granules are thought to protect mRNAs so that their translation can resume following relief

from stress. To determine whether the 18033 granules might play a similar role, we examined the translation of a reporter mRNA under different conditions. We generated the reporter by fusing the coding region of the firefly luciferase gene to the 3ʹ-untranslated region of Slbp or Ccnb1

122

(Figure 28A). Both Slbp and Ccnb1 mRNAs are efficiently translated during oocyte maturation, and both gene products are required for normal oocyte development (Abrieu, Doree et al. 2001,

Allard, Champigny et al. 2002, Arnold, Francon et al. 2008, Chao, Chang et al. 2017). We chose

to use a reporter mRNA rather than an endogenous mRNA to avoid the complications of detecting

newly synthesized protein against a background of pre-existing protein and because of the high

sensitivity of the luciferase assay used to detect translation. We injected the reporter mRNA into

immature oocytes, which were then allowed to mature at 37ºC or subjected to cold-shock.

Under control conditions, luciferase activity increased steadily during maturation reflecting the

continuous translation of the reporter mRNA (Figure 28B – grey dotted line). When oocytes were

cold-shocked 4 hr after the initiation of maturation – thus, at a time when they are able to form

18033 granules – luciferase activity continued to increase after recovery from the cold-shock

(Figure 28B, left – red solid line; vertical grey bar shows period of cold-shock). In striking contrast,

when oocytes were cold-shocked 8 hr after the initiation of maturation – at the time that they lose

the ability to form 18033 granules – translation did not resume after the oocytes were returned to

37ºC (Figure 28B, right – black solid line; vertical grey bar shows period of cold-shock).

Quantification of the fold-increase in luciferase activity in oocytes confirmed that the early cold-

shock had no detectable effect whereas late cold-shock significantly reduced subsequent

accumulation of luciferase activity (Figure 28C).

These results demonstrated a correlation between the loss of the ability to form cold-induced 18033

granules and the ability to translate the reporter mRNA following return to 37ºC. To seek evidence

of a functional relationship between these oocyte properties, we treated reporter-injected oocytes

with roscovitine, thereby preserving their ability to form 18033 granules during cold-shock.

Confirming the previous results, luciferase activity increased steadily under control conditions

123

(Figure 29A – solid black line) but did not following recovery from a late cold-shock (Figure 29A

– dotted black line). In contrast, when oocytes were treated with rosocovitne, luciferase activity steadily increased in both oocytes kept at 37oC and in those subjected to the late cold-shock (Figure

29A – solid and dotted red lines, respectively). Quantification of multiple replicates confirmed that oocytes treated with roscovitine maintained the ability to resume mRNA translation following cold-shock (Figure 29B). Thus, conditions that preserve the ability of oocytes to form 18033 granules in response to cold-shock also preserve their ability to translate mRNA following recovery from cold-shock.

Discussion

While stress-induced RNPs have been previously reported in C. elegans and C. remanei oocytes, no such structures have been described in mammalian oocytes. These RNP granules, termed grP- bodies, are P-body-like particles that can form in oocytes in response to various forms of stress, including prolonged cell arrest and heat shock (Jud, Czerwinski et al. 2008, Noble, Allen et al.

2008, Schisa 2012). In mammalian oocytes, P-body-like RNPs associated with translational repression have also been described (Swetloff, Conne et al. 2009, Flemr, Ma et al. 2010, Vieux and Clarke 2018). However, these structures are observed in physiological conditions, and are believed to be the site of RNA storage during normal oogenesis. We describe here novel foci using the 18033 antiserum, commonly used to stain P-bodies. In fact, the foci are positive for EDC4, a common component of P-bodies and a component of the decapping complex, including DCP2, the mammalian orthologue to DCAP-2, found in the stress induced grP-bodies in C. elegans (Noble,

Allen et al. 2008). The 18033 antiserum detects two known components of p-bodies: TNRC6A and EDC4 (Bloch, Gulick et al. 2006). Yet, these 18033 granules differ from the foci found in physiological conditions, in that they are not positive for CNOT6 nor TNRC6A and only form in

124

response to the cold-stress. The induced foci also formed when the oocytes were treated with

MG132 and disappeared when the source of the stress (cold-stress or MG132) was removed, as is

expected of canonical stress granules (Narayanaswamy, Levy et al. 2009, Aulas, Fay et al. 2017).

At the very least then, the 18033 granules are a novel type of stress granule.

G3BP1 is a critical component of canonical stress granules (Aulas, Fay et al. 2017, Markmiller,

Soltanieh et al. 2018). Knocking down its expression significantly impairs stress granule formation

(Ghisolfi, Dutt et al. 2012). Strikingly, however, the 18033 signal did not colocalize with G3BP1

after cold-shock (supplemental; Figure 32A). The emergence of G3BP-negative foci resembling

stress-granules in oocytes may therefore suggests a redundancy in the mechanism regulating their

formation. In fact, two G3PB proteins with an overlap in function have been described - G3BP1 and G3BP2 – although they share only 57% sequence homology. The antibody used in this study was raised against the human G3BP1 protein and can react with its murine orthologue G3BP1 but

not with the paralogue, G3BP2. Individual knock-downs moderately prevent the formation of

stress granules. Only by knocking down both G3BP protein can the formation of stress granules

be almost completely abrogated in somatic cell lines; and still some granules persist (Matsuki,

Takahashi et al. 2013). Alternatively, there is then a diversity in the types of stress granules that can form. Markmiller et al. (2018) identified 53 RNA-binding proteins that localize to stress granules induced by two different types of stress. Interestingly, while the majority of the markers were systematically described in all stress conditions, 23% varied according to cell type and the type of stress. The data is also supported by previous work, describing a context dependent composition of stress granules in somatic cells (Aulas, Fay et al. 2017). The foci that form in oocytes could then comprise a type of stress granule whose formation is G3BP1-independent, perhaps uses G3BP2 to assemble instead or uses neither.

125

The ability to form these foci is detected early in growth and is maintained in fully-grown oocytes

as well as in oocytes at the early stages of maturation. Retrograde and ante-retrograde transport

have both been reported to contribute to the assembly and disassembly of stress granules,

respectively. Accordingly, like stress granules, 18033 granule assembly in GV oocytes relies in

part on microtubules (Ivanov, Chudinova et al. 2003, Loschi, Leishman et al. 2009). Prior to

GVBD, only about one-third of oocytes can form 18033 granules in response to coldshock. The

principles of liquid-liquid phase separation drive the regulation of RNP formation (Hyman, Weber

et al. 2014). EDC4 concentrations is therefore expected to modulate 18033 granule formation. No

significant change in EDC4 expression is detected at GVBD, however (Figure 26E). TIA-1 is

another canonical stress granule marker also found in grP-bodies (Jud, Czerwinski et al. 2008). In the absence of TIA-1, stress granule formation is impaired (Kedersha, Gupta et al. 1999). The small number of foci and the unspecific immunofluorescent signals for TIA-1 (supplemental;

Figure 32B) may also reflect low expression of the protein. Given its key role in canonical stress granule formation, if EDC4 requires interactions with TIA-1 to assemble the foci, then low TIA-1

levels may explain the incomplete penetrance of the phenotype in growing oocytes.

Following GVBD, the incidence of foci decreases dramatically. In fact, no 18033 granules form

in MII oocytes that are cold-shocked – an effect associated with the loss of the ability to make the

foci, not a decrease in the amount of protein. We also identified a role for CDK1 activity in

regulating the formation of 18033 granules, linking the ability to make the foci to meiotic cell

cycle progression. Many of the other subunits of the decapping complex are susceptible to post-

translational modifications, namely phosphorylation by CDK1, during oocyte maturation. Dcp1a

and Dcp2 are maternal effect genes whose translation is significantly upregulated during

maturation (Ma, Flemr et al. 2013). At MII, both DCP1A and DCP2 signals display significant

126

mobility shifts on immunoblots, which can be prevented by the chemical inhibition of either CDK1

or MAPK activity (Ma, Flemr et al. 2013). Preventing the accumulation of CDK1 activity after

the GVBD also preserves the oocyte’s ability to form the foci at the latter stages of oocyte

maturation. Similarly, a timely inhibition of MAPK activity, using the drug U0126, can also

partially preserve the ability to assemble 18033 granules in oocytes cold-shocked late in maturation

(supplemental; Figure 33). Interestingly, during the transition from the GVBD to MII, EDC4 also

displays a second band in immunblots, which might be the result of post-translational protein modification. It can therefore be inferred that CDK1, with some contribution from MAPK, represses the formation of the foci by phosphorylating EDC4. Likewise, in C. elegans, the formation of grP-bodies is inhibited by CEH18 an upstream promoter of meiotic maturation, ovulation and MAPK activity (Jud, Czerwinski et al. 2008).

Using a reporter gene approach, we tried to assess the impact of 18033 granule formation on gene expression. As such we provide indirect evidence for a protective role on the oocyte’s translational activity. After cold-shock, translation of the reporter construct resumes in oocytes that are able to assemble 18033 granules. Conversely, translation is abrogated in oocytes unable to make these foci. This suggests significant damage to the transcriptome or the translational machinery when the oocyte is stressed, and also provides evidence for a mechanism used by the oocyte to protect its transcriptome. The manipulations of oocytes and embryos associated with assisted reproductive technologies induce variable temperatures, pH and other potential sources of stress for the oocytes.

The absence of cold-inducible 18033 granules in mature mouse oocytes may then explain their

vulnerability and the suboptimal yields of quality embryos in IVF. In addition, diminished meiotic

and developmental competence correlates with ≥2-fold decreases in the relative amounts of 145 genes. It is not clear however whether this decrease results from modified transcriptional activity,

127

post-transcriptional regulation or both (Pan, O'Brien M et al. 2005). Interestingly, only 30% of

growing oocyte grown in-vitro, mature successfully; a proportion similar to that of oocytes who

successfully assemble 18033 granules (Pan, O'Brien M et al. 2005). Perhaps then, the formation

of 18033 granules in a subpopulation of oocyte is what protects their developmental competence.

Oocytes maturing in-vitro also display reduced developmental competence (50-60% reach

blastocyst) and is associated with 1528 downregulated transcripts in bovine (Lonergan and Fair

2008, Mamo, Carter et al. 2011).

To improve the outcomes of assisted reproductive technologies, better incubation and manipulation methods need to be developed. Studies of incubators used in ART laboratories, revealed significant temperature variations and therefore discrepancies between expected and real temperatures used (Walker, Butler et al. 2013). Newer methods of incubation therefore require consistency and stability, with minimal perturbations. Automated imaging of embryos in culture while they are kept in the incubator, for example, is a non-invasive method of assessing morphological markers of embryo quality (Wong, Loewke et al. 2010). Ideally, the culture environment would replicate the ovarian and uterine environments perfectly. In the meantime, the use of a broader set of evaluation parameters will give a more complete perspective of quality and competence and will help determine the impact of various culture conditions on the gametes and embryos. Different aspect of oogenesis and embryogenesis have different degrees of tolerance to stress. For instance, cytokinesis and mitosis react differently to 1ºC changes in the temperature: while the nuclear division is unaffected, cytokinesis is impaired, resulting in multinucleated cells and is potentially a problem for embryos at the cleavage stage (Anifandis 2013). Detecting signs of stress like 18033 granule formation, can assess the toxicity of culture environments and help develop better incubation parameters. Reciprocally, understanding how their formation is

128

regulated and how they protect the oocyte, may help develop new culture protocols, inducive to

their formation and the preservation of quality and competence.

Acknowledgements

We thank Dr. M. Fritzler for generously supplying 18033 antiserum. Supported by the grants

from the Natural Sciences and Engineering Research Council of Canada (RGPIN-402138), and the Research Institute of the McGill University Health Centre to H.J.C. K.-F.V. was supported by Réseau Québécois en Reproduction CREATE scholarship. We thank Dr. Qin Yang for her troubleshooting expertise and the other members of the Clarke lab (past and present) for their critical feedback.

129

Figures

A 37°C (5hr) 4°C (5hr) C 18033 DRAQ5

100%

75% b 50% 4°C (O.N.) O.N. Recovery b Frequency 25% with 18033 granules Percentage of oocytes a a 0% 37°C (5hr) 4°C (5hr) 4°C (O.N.) 4°C (5hr) + O.N. recovery

B CNOT6 18033 overlap overlap (3x)

37ºC

(3x)

130 4ºC

Figure 24 – Formation of novel 18033 granules is induced in cold-shocked oocytes. Full-grown oocytes were incubated at 4ºC for 5 hr and stained with various RNP markers. The nuclei were stained with DRAQ5 or DAPI (blue). (A) At 37ºC, 18033 stains small cortical at the cortex of oocytes (top left panel). At the 4ºC larger dispersed cytoplasmic foci are labelled (top right panel). When kept at 4ºC, the induced foci persist (bottom left panel) but if brought back to 37ºC the control phenotype returns (bottom right panel). (B) The foci at 37ºC are also positive for CNOT6 while the foci at 4ºC only stain with 18033. (C) The induced foci formed in 36% of the cold- shocked oocytes compared to the less than 5% of control oocytes. A similar proportion of oocytes cold-shocked overnight displayed the foci. In contrast, only 1% the oocytes that were allowed to recover retained the foci.

131

A B Control Nocodazole 18033 DRAQ5

50%

C °

37 a 40% a

30%

b 20% Frequency

with 18033 granules 10% Percentage of oocytes

C ° 4 0% Control Latrunculin Nocodazole

Figure 25 – Formation of novel 18033 granules requires stable microtubules. (A) Full-grown

oocytes were treated with nocodazole to impair tubulin polymerization and prevent microtubule

formation. Oocytes were then stained with 18033 (red) and DRAQ5 (blue). Both control and

treated oocytes displayed inducible 18033 foci after cold-shock. (B) Quantification of the results

show however, that only 15% of oocytes treated with nocodazole form the foci in contrast to the

34% of cold-shocked control oocytes. A third group treated with Latrunculin A, an actin

microfilament destabilizing agent, was used to show the specificity of the microtubule effect.

132

A Growing Full-grown GVBD MI MII

C ° 4

DAPI 18033

B D 100% 100% b

80% 75%

60% 50% a a

40% Frequency Frequency 25% with 18033 granules with 18033 granules 20% Percentage of oocytes Percentage of oocytes C C a a 0% 0% G FG GVBD MI MII Primed GV Primed GV in-vivo mature (37⁰C) (4⁰C) MII (4⁰C)

C E Primed GV in-vivo mature MII

EDC4

18033

4ºC TNRC6A Tubulin

DAPI 18033 133

Figure 26 – Ability to form 18033 granules in response to cold-shock is lost during maturation. (A) Granulosa-growing oocyte complexes were collected, cold-shocked and stained with 18033 (red) and DAPI (blue). Full-grown oocytes were also collected. They were incubated and allowed to initiate maturation. After 4hr, GVBD oocytes were collected. After 8hr, metaphase I (MI) oocytes were collected. After and overnight incubation (16 hr), metaphase II (MII) oocytes were collected. Oocytes from every stage were cold- shocked and stained for whole cell immunofluorescence. All stages (including MI oocytes, albeit at very low frequency) displayed inducible foci after coldshock except MII oocytes (B) The proportion of oocytes at each stage is reported. A similar proportion (33-

36%) of growing and full-grown oocytes form the foci in response to the cold-shock. A significant increase in the number of oocytes that form the foci is observed at the GVBD stage (79%). At MI, the number of oocytes with the 18033 foci dramatically decreases (4%) and at MII the foci are never detected after the cold-shock. (C-D) Primed oocytes were collected from mice that were injected with eCG.

The GV oocytes were collected 44 hr post injections and MII oocytes 12hr after injection of hCG (to induce maturation and ovulation in-vivo). Both groups were cold-shocked and stained. While 75% of primed GV oocytes formed the foci, they were not detected in in- vivo mature MII oocytes. (E) Immunoblot of 3 biological samples, using GV, GVBD and MII oocytes detects the two signals of 18033

EDC4 (top band(s)) and TNRC6A (bottom band); the most representative blot is shown. A second band above the EDC4 signal is also detected in MII oocytes. Tubulin was used as a control. The nature of the EDC4 band was later confirmed with a EDC4-specific antibody

(Signaling Technologies 2548S).

134

A MATURATION

0hr 4hr 8hr

5hr) C ( ° 4

18033 DRAQ5

+ Roscovitine B

***0hr 4hr 8hr 100% 100% 100% ***

75% 75% 75%

50% 50% 50% Frequency 25% 25% 25% Percentage of oocytes 0% with inducible 18033 granules 0% 0% 4°C 4°C 4°C Control Roscovitine Figure 27 – Inhibition of CDK1 preserves the ability to form 18033 granules in MII oocytes. (A)

Full-gown oocytes were treated with roscovitine. These oocytes remained at the GV stage and

were cold-shocked after a 2-hr pre-incubation. Similarly, oocytes were treated with roscovitine 2

hr and 5 hr after the onset of maturation, before and after the completion of the GVBD. After cold-

shock, all three treated groups display 18033 foci, including the MII oocytes. (B) Quantitative

comparison between treated groups and time-matched controls show a significant increase in the

number of oocytes with the foci in treated groups, at the GV and MII stages. Both control and

treated GVBD groups show comparable high numbers of foci-positive oocytes after cold-shock.

135

Slbp 3'-UTR (A) 20

(A)20 Luciferase Ccnb1 ORF 3'-UTR

136

Figure 28 – Loss of the ability to make 18033 granules during cold-shock coincides with impaired translation following cold-shock. (A) Fully grown immature oocytes were injected with reporter a gene construct with either the Slbp 3'-UTR or the Ccnb1 3'-UTR. (B-C) luciferase activity was monitored to assess the translational activity of both constructs. Oocytes were cold-shocked for 5 hr (grey area), 3hr into maturation (early-cold-shock; solid red line) or 8 hr into maturation (late cold-shock; solid black line), and then returned to 37º C. They were then compared to control non- cold-shocked oocytes (grey dotted lines). Data shown represent the results form a single experiment. The luminescence for both constructs increases during maturation in control oocytes.

It also increases after an early cold-shock but remains unchanged after a late cold-shock. (D) The intensity of the signal at the beginning of the recovery (t = 0 hr) was then compared to the signal intensity 5 hr after the end of the cold-shock (t = 5 hr) (arrows). Cumulative data from three sets of experiments shows a higher fold of accumulated Luciferase activity in the 5 hr following an early cold-shocked at the GVBD stage (delineated by red arrows) and significantly less in the 5hr following a late cold-shock at MI (delineated by black arrows).

137

Figure 29 – Preventing the loss of 18033 granules forming ability late in maturation rescues translation post-cold-shock. (A) Fully grown immature oocytes injected with the reporter gene constructs were treated with roscovitine 5 hr into maturation, to inhibit CDK1 activity and maintain foci-making ability (red lines). Both treated and untreated oocytes were then cold-shocked after 8 hr (grey area), returned to 37º C and compared to non-cold-shocked groups. The luminescent signal intensity increases progressively in both treated and untreated control (non-cold-shocked) oocytes.

A similar increase is seen in roscovitine-treated oocytes after cold-shock. Meanwhile the intensity of the signal in untreated oocytes did not change significantly after cold-shock. (B) Again, the intensity of the signal at t = 0 hr and t = 5 hr of recovery (red arrows) was compared. Quantification of three biological replicates confirms that in the 5 hr following a late cold-shock, significantly more luciferase activity is accumulated in treated oocytes in comparison to untreated oocytes.

138

Supplemental data

A TNRC6A 18033 overlap overlap

37ºC

(3x)

4ºC

(3x)

B EDC4 18033 overlap overlap

37ºC

(3x) (3x)

4ºC

(3x) Figure 30 – 18033 granules are EDC4 granules. Full-grown oocytes were cold-shocked and stained with 18033 and either TNRC6A or EDC4. The 18033 antiserum recognizes both TNRC6A and EDC4. (A) At 37ºC, the TNRC6A signal overlaps with the 18033 signal in the cortical foci.

At 4ºC the TNRC6A signal at the cortex disappears as od the cortical foci. Moreover, TNRC6A does not localize to induced 18033 foci after cold-shock. (B) Like TNRC6A, EDC4 overlaps with the 18033 signal at the cortex of oocytes at 37ºC. However, unlike TNRC6A, at 4ºC the EDC4 signal overlaps almost perfectly with the 18033 signal in the induced foci. 139

37ºC 18033DRAQ5

Figure 31 – MG132 treatment induces the formation of 18033 granules. Fully grown oocytes were incubated in MG132 for 5 hr. They were then stained with 18033. Results show the formation of

EDC4 stress granules in treated oocytes at 37C°.

140

A G3BP 18033 overlap overlap

(3x)

37ºC

(3x)

4ºC

B TIA-1 18033 overlap overlap

(3x)

37ºC

(3x)

4ºC

Figure 32 – Canonical stress granule markers do not localize to EDC4 granules. Full-grown oocytes were cold-shocked and stained with 18033 and one of two canonical stress granule markers: G3BP or TIA-1. (A) At 37ºC, the G3BP signal displays a small number of dispersed foci in the cytoplasm. At 4ºC the number of foci increased slightly but their size and distribution remained unchanged. The 18033 signal and the G3BP signal did not overlap in neither of the two conditions. Moreover, TNRC6A does not localize to induced 18033 foci after cold-shock. (B)

Similarly, the TIA-1 signal did not overlap with the 18033 signal in neither condition. In fact, very few TIA-1 foci are observed in oocytes and a very faint background I seen throughout the

141cytoplasm suggesting an unspecific signal.

A B

12hr U0126 pre-treatment 100%

75%

50% C (5hr) C ° 4 * 25% * Percentage of oocytes

with inducible 18033 granules 0% Ctrl + 4⁰C U0126 + 4⁰C U0126 + 37⁰C

Figure 33 – Inhibition of the MAPK activity partially preserves the ability to make EDC4 granules in MII oocytes. Oocytes were treated with U0126, a MAPK inhibitor, 8hr after the onset of maturation. They were then incubated in the presence of U0126 overnight and allowed to complete maturation. (A) The resulting MI and MII oocytes were then cold-shocked for 5hr and stained with the 18033 antiserum. Induced foci were detected in some of the treated oocytes after cold-shock.

(B) More specifically, 25% of oocytes cold-shocked after U0126 treatment maintain the ability to make the foci at MII in comparison to the 6% of control oocytes. It is also important to note that in the control group, the small number of cells with foci after the coldshock, was strictly comprised of MI oocytes; in control conditions, MII oocytes never displayed the foci.

142

Chapter 4

[DISCUSSION]

143

I – Oocytes are the perfect tool to study RNA management and processing events

In somatic cells, different processing events are intimately linked, making it hard to discern their

individual roles. For instance, the length of the poly(A) tail is believed to regulate translational

activity: a long tail enhances translation and a short tail hinders it (Gallie 1991). However, deadenylation commonly precedes decapping and decreases RNA stability and in turn, RNA degradation dampens translation (Couttet, Fromont-Racine et al. 1997). It is therefore difficult to assess the direct impact of the poly(A) tail on translation without the confounding effect of degradation. In contrast, oocytes are the ideal cells to study RNA processing events and their role in regulating RNA metabolism. As discussed in Chapter 1, translational regulation is critical for the orderly and timely progression of maturation, fertilization and the initial stages of embryogenesis. Moreover, a dissociation between the mechanisms of translational silencing and degradation in the oocyte is suggested in Chapter 2 and further discussed in the next section.

Transcripts are therefore subject to extensive processing and robust regulation, underlined by a plethora of regulatory mechanisms to be identified and studied.

Chapter 2 also describes three populations of transcripts according to changes in their poly(A) tail length during maturation. A recent report shows that in oocytes, knocking-out Cnot6l, a paralogue of Cnot6, significantly dysregulated the expression of certain maternal transcript and impaired oocyte developmental competence (Horvat, Jankele et al. 2018). The Cnot6 knock-down described in chapter 2 only had a specific effect on a subset of transcripts and did not disrupt maturation; developmental competence was not assessed. The deadenylation of both Orc6 and Slbp requires the enzymatic activity of CNOT6. Meanwhile, the two other populations of transcripts described were unaffected by the knockdown of CNOT6. While 80% of the Cnot6 transcripts is depleted,

the levels of CNOT6 protein only drop by about 25%. The poly(A) tail length of Orc6 and Slbp

144

were significantly impacted by this loss, revealing a high sensitivity to the expression levels of

CNOT6. The lack of effect on the other maternal transcripts assessed does not however rule out a

role for CNOT6; these transcripts may require a complete knock-out before an effect on their

poly(A) tail is detected.

Molecular and proteomic changes in the oocyte also provide additional insight on the regulators

of RNA metabolism and allow to assess the context-dependent role of certain processing events.

For instance, the CPE and its recruitment of the CPEB has been shown to be bivalent in Xenopus.

In growing oocytes, it mediates deadenylation and the silencing of transcripts (Hodgman, Tay et

al. 2001, Racki and Richter 2006). In maturing oocytes however, it is reportedly subject to phosphorylation by Eg2 and is required for the polyadenylation of maternal transcripts (Hodgman,

Tay et al. 2001). In mice, the interplay between CPEB and other cis elements regulate the dynamics of translational activity as is illustrated in our model for the maturation-induced polyadenylation of Ccnb1, Orc6 and Slbp and work done by others (Sousa Martins, Liu et al. 2016). CPEB activity and stability can be regulated by IAK-1 phosphorylation, linking it to oocyte meiotic progression

(Hodgman, Tay et al. 2001). The role of CPEB in regulating poly(A) tail length and translation is

therefore context-dependent: the availability of co-factors and the presence or absence of

modulators dictate its function.

The results in both Chapter 2 and Chapter 3 also imply the use of different mechanisms to regulate

the processing of different transcripts and RNA management mechanisms regulated by the oocyte

cell cycle. Different elements in the 3'-UTR result in different adenylation fates. The presence of

a PBE near the polyadenylation signal (PAS) in the 3'UTR is critical for the CNOT6-dependent

deadenylation of Slbp. Interestingly removing the PBE replicates the pattern of element observed

in the Ccnb1 3'-UTR and results in a polyadenylation of the transcript similar to that of Ccnb1.

145

Moreover, Chapter 2 proposes a model where the interplay between the different elements in the

3'-UTR regulates the changes to the Ccnb1, Orc6 and Slbp poly(A) tails, in a CPEB and stage- dependent manner. More work is required however to validate this model. For example, the

binding partner of the PBE was not identified. Pumilio2 binding in Xenopus oocytes has been

shown to regulate the translation of the RINGO/Spy maternal transcript (Padmanabhan and Richter

2006). Both PUM1 and PUM2, the murine orthologues, are expressed in ovaries and PUM1

expression is important for primordial oogenesis (Mak, Fang et al. 2016). A role has yet to be

reported for either protein in oocyte maturation. It would also be interesting to assess whether the

removal of the PBE also enhances the translation of the construct in addition to preventing its

deadenylation. This would determine if the PBE and its binding partner confer translational silence

by deadenylating the transcript or if deadenylation and translational inhibition are separately

induced.

II – Poly(A)-independent translation and degradation

A well-established correlation between poly(A) tail length and translational activity exist since the

1970s: a long tail is believed to enhance translation while a short tail is associated to translational

silence (Slater, Slater et al. 1972, Munroe and Jacobson 1990, Beilharz and Preiss 2007).

Correspondingly, in Xenopus oocytes, the precocious recruitment of PAP is enough to catalyze the

elongation of the poly(A) tail and stimulate translation (Dickson, Thompson et al. 2001). Likewise,

the increased luciferase activity of the Slbp- and Ccnb1-3'UTR reporter constructs in Chapter 3,

mirrors the polyadenylation of the endogenous transcripts at the onset of maturation in Chapter 2.

However, the elongation of the Slbp and Orc6 poly(A) tails at the GV stage does not trigger a

precocious expression of either transcript. The poly(A) tail is therefore not sufficient to initiate

translation in oocytes. Conversely, the expression of ORC6 and SLBP did not persist in oocytes

146

when Orc6 and Slbp poly(A) tails were ectopically elongated at MII (Figure 19A). Other groups

even suggest that poly (A) tail length is a byproduct of translational activity. Djuranovic et al.,

(2012) show that miRNA induced deadenylation and decay occurs after translation has been silenced. Deadenylation in this case is then believed to consolidate the effects translational repression, perhaps acting as a failsafe mechanism. The authors do concede however that this order

of events may be cell-specific or transcript-specific (Djuranovic, Nahvi et al. 2012). Similarly,

TNRC6A was shown to have the ability to repress translation independently of poly(A) tail and

5'-methyl cap (Mishima, Fukao et al. 2012). Interestingly, the same study attributed deadenylation-

independent silencing and deadenylation activities of TNRC6A to separate domains: PAM2 and

P-GL motifs respectively.

Essential subunits of the decapping complex, DCP1A and DCP2, are absent in the growing oocyte

and only accumulate during maturation (Ma, Flemr et al. 2013). The massive degradation event at

the latter end of maturation therefore occurs hours or days after the downregulation of transcription

in mice (Figure 4). Meanwhile, the average half-life of transcripts in oocytes correspondingly

spans hours and days (Jahn, Baran et al. 1976, Brower, Gizang et al. 1981). Strikingly, severely

deadenylated but stable transcripts (poly(A) tail length ~0nt) were described and studied in full-

grown oocytes in Chapter 2. In contrast, the induced lengthening of the poly(A) tail could not

prevent the degradation of Orc6 and Slbp in mature oocytes (Figure 19C). This may also explain

the stagnant levels of ORC6 and SLBP at MII, even when their respective transcripts have been

ectopically polyadenylated.

Typically, deadenylation precedes decapping, which in turn is a pre-requisite for 5' to 3'

ribonuclease mediated degradation (Couttet, Fromont-Racine et al. 1997). In fact, disrupting the

expression of decapping enzymes, DCP1A and DCP2, significantly hinders the degradation of

147

maternal transcripts in maturing oocytes and early embryos (Ma, Flemr et al. 2013). RNA stability

in oocytes is therefore stage and context-dependent. Therefore, it can be stipulated that deadenylation only contributes (but is not required) to the silencing of maternal transcripts in growing oocytes, in the absence of decapping enzymes. With maturation and the expression of

DCP1A and DCP2, deadenylaytion then facilitates degradation of maternal transcripts. Other factors however, regulate both processes and supersede the regulatory function of the poly(A) tail.

For instance, recent work, studying a cycle of capping and decapping termed cap homeostasis, has described uncapped mRNAs with long poly(A) tails (Kiss, Oman et al. 2016). These results

therefore suggest the existence of a decapping mechanism independent of the poly(A) tail length.

conversely, deadenylation has been shown to be dissociated from the cap-status of transcripts in certain context (Mishima, Fukao et al. 2012).

III – RNPs in oocytes, RNPs in disease, RNPs in infertility

In Chapter 3, stress-induced RNPs are described in mammalian oocytes. These particles behave like stress granules but have a composition reminiscent of P-bodies. grP-bodies are similar RNPs

described in C. elegans, that also share similar characteristics with both stress granules and P-

bodies (Noble, Allen et al. 2008). The penetrance of the phenotype after coldstress varied

significantly: the number of foci and their size changed within experiments and across biological

replicates (data not shown). Interestingly, however, the formation of the EDC4 stress granules is

observed consistently in a subset of growing oocytes and was therefore chosen to describe the

observation. Nevertheless, the use of a more automated and computerized approach, collecting

data on the number, size, location, intensity, interactions of foci and other components of the

cytoplasm, would provide a more quantitative analysis of 18033 granule formation. For instance,

an inverse correlation between the number of foci and the size of foci or their distribution in the

148

cytoplasm may provide insight into the nature of this variability and may unveil more intricate

dynamics in the formation and behaviours of the 18033 granules. Nevertheless, why the ability to

respond to the cold-stress in such a way is limited to a subset of growing oocytes remains unclear.

Chapter 3 speculates on the role played by the molecular environment, determined by the

expression of known partners of EDC4 (e.g. DCP1A and DCP2) and suggest a link between the

percentage of oocyte that can form granules to the success rate of in-vitro growth. These

hypotheses remain highly speculative and must be further tested.

As described in Chapter 1, the range of RNP types describes a spectrum rather than clear cut

categories. The nomenclature is further muddled by the use of different names in different models

for similar RNPs or the same names for RNPs with similar compositions but different behaviours.

Stress granules for instance, describe all RNPs that form in response to a wide range of stress

stimuli. Reportedly however, these foci vary significantly, depending on the type of cell and the

type of stress (Aulas, Fay et al. 2017, Markmiller, Soltanieh et al. 2018). Furthermore, function is

often stipulated based solely on composition. Most RNP studies lack a functional approach; both

CGH-1+/PATR-1- and CGH-1+/PATR-1+ RNPs in C. elegans are involved in the protection and

the degradation of maternal transcripts respectively, despite similar compositions, and are both

called p-bodies (Boag, Atalay et al. 2008). A functional assay determined that the formation of the

described 18033 granules, protects the translational activity of the oocytes under duress. It is not

clear however whether it is the transcripts or the translational machinery that is protected.

Nevertheless, transcripts, translation initiation factors and ribosomal subunits all localize to stress

granules in somatic cells (Kedersha, Gupta et al. 1999, Kimball, Horetsky et al. 2003). In contrast,

the decapping complex-associated nature and the P-body characteristics of these RNPs suggest a

role in RNA decay instead. Cougot et al. (2004), targeted and knocked-down the 5' to 3'

149

exoribonuclease XRN1 to identify the role of P-bodies. In doing so, they were able to detect the

accumulation of RNA in P-bodies and therefore concluded that they were the site of RNA decay.

Similarly, the accumulation of RNA in P-bodies when DCP1A expression is lost suggest that P-

bodies are also the site of decapping (Sheth and Parker 2003). The formation of grP-bodies in

stressed C. elegans oocytes however, does not impair meiotic and developmental competence and

therefore are unlikely representative of maternal transcripts degradation (Jud, Czerwinski et al.

2008). In fact, the P-body marker CGH-1, in the absence of PATR-1 can confer protective properties to the p-bodies of oocytes in C. elegans (Boag, Atalay et al. 2008). Interestingly, 18033

granules mostly form before the phosphorylation of DCP1A and DCP2 and the widespread degradation of maternal transcript. Alternatively, the formation of the foci may be symptomatic of another process but may not be the structures storing and protecting the RNA or the translational factors. To that effect, it has not been determined whether the ‘protected transcripts’ localize to the

18033 granules during the cold-shock. The foci may then serve as a site where factors that could be detrimental to the transcripts are sequestered. Moreover, ectopic reporter gene constructs were used to assess the impact of foci formation on translation and may not reflect the real biology of transcripts in stressed oocytes. A hallmark of maturation is the initiation of degradation of mRNAs, a process that is completed during the early cleavage stages of embryogenesis. Although it is not known whether injected mRNAs are subject to the same regulation, we considered that roscovitine might prevent this degradation. Defects in RNA-binding proteins involved in RNP formation are often associated with significant pathologies. Whether the latter result from the failure to establish dysregulated interactions between the RNA and the proteins or from the changed dynamics of RNP formation has yet to be determined. The misfolding of RNA-binding proteins can increase their aggregation in stress granules and reduces the dynamic liquid nature of the RNPs into more

150

permanent solid foci (Alberti, Mateju et al. 2017). The RNA-binding protein FUS for instance, is

associated with Amyotrophic lateral sclerosis (ALS) and frontotemporal lobar degeneration

(FTLD). Overexpression of FUS or mutations in the glycine-rich region and the extreme C-

terminal part of the protein lead to toxic aggregations of the protein in the cytoplasm of motor

neurons and the development of the neuro-degenerative diseases (Lagier-Tourenne, Polymenidou

et al. 2010, Sama, Fallini et al. 2017). In fact, numerous mutations affecting RNA-binding proteins

involved RNA processing have been implicated in neurodegenerative diseases (Taylor, Brown et

al. 2016). Accordingly, the study of RNP formation and function has become a very trendy topic

in fields like neurobiology.

Components of the protein quality control including the ubiquitin-binding protein VCP/p97, also

linked to autophagy, have been shown to localize to stress granules (Buchan, Kolaitis et al. 2013).

In fact, Buchan et al. (2013) describe an alternative clearing of stress granules involving VCP-

dependent autophagy. This process has been suggested to rescue cells from aberrant stress granule

formation (Alberti, Mateju et al. 2017), perhaps in conditions of prolonged stress, changed RNP

dynamics (e.g. FUS mutation) or in cells with faulty stress disassembly capabilities. In aging

systems, the properties of key stress granule regulators are changed, reducing the RNP dynamics

therefore increasing the likelihood of neurodegenerative diseases. For instance, in C. elegans, ~700 proteins, many of which are linked to neurodegenerative diseases, decrease in solubility with age

(David, Ollikainen et al. 2010). It would therefore be interesting to know 1) whether in aged oocytes, stress granules form idiopathically, 2) whether they can form in response to stress and 3) whether they disassemble normally when the oocytes are rescued from the source of stress or whether they persist and become toxic, impairing oocyte function, quality and competence.

151

IV – Conclusion: the oocyte transcriptome is an underappreciated parameter of oocyte

quality

Nuclear transfer experiments dating back to the 1950, were the first to show the importance of

cytoplasmic factors in early embryogenesis. Robert Briggs and Thomas King were the first to

generate animal clones using Xenopus eggs (Briggs and King 1952). It was later shown that the

use of cytoplasm from undifferentiated oocytes conferred their status to the injected nuclei from

differentiated somatic cells (Gurdon, Elsdale et al. 1958, Gurdon 1962). The transcriptome and the

proteome are therefore as important as the changes to the nucleus, in establishing developmentally

competent zygotes. They establish the molecular program at the basis of meiotic resumption, the

first mitotic cell cycles of early embryogenesis, genome reprogramming and activation. Factors

critical for nuclear reprogramming, establishment of totipotency, and successful initiation of

embryogenesis are accumulated during oogenesis (Pan, O'Brien M et al. 2005). Arguably, the small amount of material in single cells makes it difficult to gather enough insight on these mechanisms using proteomic tools in the oocyte. The transcriptome, with the high copy number and the adapted tools available today provide a more sensitive approach (Chen, Melton et al. 2011).

The transcriptome is also sensitive to the oocyte’s environment. FSH-FSHR activation in granulosa cells, modulates poly(A) tail length of maternal transcripts by upregulating polyadenylation through an AKT-dependent pathway in the oocyte, thereby increasing translation.

Similarly, high levels of IGF-1 can prevent the global decrease in protein solubility and prevent aberrant stress granule formation associated to aging (David, Ollikainen et al. 2010, Lechler,

Crawford et al. 2017). In the absence of transcription, modulation of the transcriptome can have both rapid and long-lasting effects and can induce irreversible change in the case of damage done to the transcriptome. Furthermore, the transcriptome provides a broad perspective on the molecular

152 state of oocytes and embryos. For instance, in C. Intestinalis, Ccnb3 degradation regulates the timing of mitosis and zygotic genome activation in the early embryo (Treen, Heist et al. 2018,

Treen, Heist et al. 2018). Subsequently, the timing and the duration of the first three rounds of mitosis in the early cleavage stage is indicative of embryo potential. Reciprocally, higher potentials are also associated with a specific transcript profiles (Wong, Loewke et al. 2010). Transcriptomic tools should therefore be used more commonly to study and assess oocyte quality and competence in depth.

153

References

Abrieu, A., M. Doree and D. Fisher (2001). "The interplay between cyclin-B-Cdc2 kinase (MPF) and MAP kinase during maturation of oocytes." J Cell Sci 114(Pt 2): 257-267. Adhikari, D. and K. Liu (2014). "The regulation of maturation promoting factor during prophase I arrest and meiotic entry in mammalian oocytes." Mol Cell Endocrinol 382(1): 480-487. Adhikari, D., W. Zheng, Y. Shen, N. Gorre, Y. Ning, G. Halet, P. Kaldis and K. Liu (2012). "Cdk1, but not Cdk2, is the sole Cdk that is essential and sufficient to drive resumption of meiosis in mouse oocytes." Hum Mol Genet 21(11): 2476-2484. Albert, V. and M. N. Hall (2015). "mTOR signaling in cellular and organismal energetics." Curr Opin Cell Biol 33: 55-66. Alberti, S., D. Mateju, L. Mediani and S. Carra (2017). "Granulostasis: Protein Quality Control of RNP Granules." Front Mol Neurosci 10: 84. Allard, P., M. J. Champigny, S. Skoggard, J. A. Erkmann, M. L. Whitfield, W. F. Marzluff and H. J. Clarke (2002). "Stem-loop binding protein accumulates during oocyte maturation and is not cell- cycle-regulated in the early mouse embryo." J Cell Sci 115(23): 4577-4586. Allard, P., Q. Yang, W. F. Marzluff and H. J. Clarke (2005). "The stem-loop binding protein regulates translation of histone mRNA during mammalian oogenesis." Dev Biol 286(1): 195-206. Ambros, V. and H. R. Horvitz (1987). "The lin-14 locus of Caenorhabditis elegans controls the time of expression of specific postembryonic developmental events." Genes Dev 1(4): 398-414. Ancelin, K., U. C. Lange, P. Hajkova, R. Schneider, A. J. Bannister, T. Kouzarides and M. A. Surani (2006). "Blimp1 associates with Prmt5 and directs histone arginine methylation in mouse germ cells." Nat Cell Biol 8(6): 623-630. Anderson, E. and D. F. Albertini (1976). "Gap junctions between the oocyte and companion follicle cells in the mammalian ovary." J Cell Biol 71(2): 680-686. Anderson, E. L., A. E. Baltus, H. L. Roepers-Gajadien, T. J. Hassold, D. G. de Rooij, A. M. van Pelt and D. C. Page (2008). "Stra8 and its inducer, retinoic acid, regulate meiotic initiation in both spermatogenesis and oogenesis in mice." Proc Natl Acad Sci U S A 105(39): 14976-14980. Anderson, R., T. K. Copeland, H. Scholer, J. Heasman and C. Wylie (2000). "The onset of germ cell migration in the mouse embryo." Mech Dev 91(1-2): 61-68. Anderson, R., R. Fassler, E. Georges-Labouesse, R. O. Hynes, B. L. Bader, J. A. Kreidberg, K. Schaible, J. Heasman and C. Wylie (1999). "Mouse primordial germ cells lacking beta1 integrins enter the germline but fail to migrate normally to the gonads." Development 126(8): 1655-1664. Andrei, M. A., D. Ingelfinger, R. Heintzmann, T. Achsel, R. Rivera-Pomar and R. Luhrmann (2005). "A role for eIF4E and eIF4E-transporter in targeting mRNPs to mammalian processing bodies." RNA 11(5): 717-727. Andreou, A. Z. and D. Klostermeier (2013). "The DEAD-box helicase eIF4A: paradigm or the odd one out?" RNA Biol 10(1): 19-32.

154

Andreu-Vieyra, C. V., R. Chen, J. E. Agno, S. Glaser, K. Anastassiadis, A. F. Stewart and M. M. Matzuk (2010). "MLL2 is required in oocytes for bulk histone 3 lysine 4 trimethylation and transcriptional silencing." PLoS Biol 8(8). Anifandis, G. (2013). "Temperature variations inside commercial IVF incubators." J Assist Reprod Genet 30(12): 1587-1588. Ara, T., Y. Nakamura, T. Egawa, T. Sugiyama, K. Abe, T. Kishimoto, Y. Matsui and T. Nagasawa (2003). "Impaired colonization of the gonads by primordial germ cells in mice lacking a chemokine, stromal cell-derived factor-1 (SDF-1)." Proc Natl Acad Sci U S A 100(9): 5319-5323. Archer, S. K., N. E. Shirokikh, C. V. Hallwirth, T. H. Beilharz and T. Preiss (2015). "Probing the closed-loop model of mRNA translation in living cells." RNA Biol 12(3): 248-254. Arnold, D. R., P. Francon, J. Zhang, K. Martin and H. J. Clarke (2008). "Stem-loop binding protein expressed in growing oocytes is required for accumulation of mRNAs encoding histones H3 and H4 and for early embryonic development in the mouse." Dev Biol 313(1): 347-358. Arnold, S. J., S. Maretto, A. Islam, E. K. Bikoff and E. J. Robertson (2006). "Dose-dependent Smad1, Smad5 and Smad8 signaling in the early mouse embryo." Dev Biol 296(1): 104-118. Aslam, A., S. Mittal, F. Koch, J. C. Andrau and G. S. Winkler (2009). "The Ccr4-NOT deadenylase subunits CNOT7 and CNOT8 have overlapping roles and modulate cell proliferation." Mol Biol Cell 20(17): 3840-3850. Audic, Y., F. Omilli and H. B. Osborne (1997). "Postfertilization deadenylation of mRNAs in Xenopus laevis embryos is sufficient to cause their degradation at the blastula stage." Mol Cell Biol 17(1): 209-218. Aulas, A., M. M. Fay, S. M. Lyons, C. A. Achorn, N. Kedersha, P. Anderson and P. Ivanov (2017). "Stress-specific differences in assembly and composition of stress granules and related foci." J Cell Sci 130(5): 927-937. Avilion, A. A., S. K. Nicolis, L. H. Pevny, L. Perez, N. Vivian and R. Lovell-Badge (2003). "Multipotent cell lineages in early mouse development depend on SOX2 function." Genes Dev 17(1): 126-140. Bachvarova, R. (1985). "Gene expression during oogenesis and oocyte development in mammals." Dev Biol (N Y 1985) 1: 453-524. Bachvarova, R., V. De Leon, A. Johnson, G. Kaplan and B. V. Paynton (1985). "Changes in total RNA, polyadenylated RNA, and actin mRNA during meiotic maturation of mouse oocytes." Dev Biol 108(2): 325-331. Bachvarova, R. F. (1992). "A maternal tail of poly(A): the long and the short of it." Cell 69(6): 895-897. Balaban, B. and B. Urman (2006). "Effect of oocyte morphology on embryo development and implantation." Reprod Biomed Online 12(5): 608-615.

155

Balakier, H., A. Sojecki, G. Motamedi and C. Librach (2004). "Time-dependent capability of human oocytes for activation and pronuclear formation during metaphase II arrest." Hum Reprod 19(4): 982-987. Baltus, A. E., D. B. Menke, Y. C. Hu, M. L. Goodheart, A. E. Carpenter, D. G. de Rooij and D. C. Page (2006). "In germ cells of mouse embryonic ovaries, the decision to enter meiosis precedes premeiotic DNA replication." Nat Genet 38(12): 1430-1434. Barkoff, A., S. Ballantyne and M. Wickens (1998). "Meiotic maturation in Xenopus requires polyadenylation of multiple mRNAs." EMBO J 17(11): 3168-3175. Bartlam, M. and T. Yamamoto (2010). "The structural basis for deadenylation by the CCR4-NOT complex." Protein & Cell 1(5): 443-452. Bazzini, A. A., F. Del Viso, M. A. Moreno-Mateos, T. G. Johnstone, C. E. Vejnar, Y. Qin, J. Yao, M. K. Khokha and A. J. Giraldez (2016). "Codon identity regulates mRNA stability and translation efficiency during the maternal-to-zygotic transition." EMBO J 35(19): 2087-2103. Beilharz, T. H. and T. Preiss (2007). "Widespread use of poly(A) tail length control to accentuate expression of the yeast transcriptome." RNA 13(7): 982-997. Belloc, E. and R. Mendez (2008). "A deadenylation negative feedback mechanism governs meiotic metaphase arrest." Nature 452(7190): 1017-1021. Belloc, E., M. Pique and R. Mendez (2008). "Sequential waves of polyadenylation and deadenylation define a translation circuit that drives meiotic progression." Biochem Soc Trans 36(Pt 4): 665-670. Bendel-Stenzel, M. R., M. Gomperts, R. Anderson, J. Heasman and C. Wylie (2000). "The role of cadherins during primordial germ cell migration and early gonad formation in the mouse." Mech Dev 91(1-2): 143-152. Bennabi, I., I. Queguiner, A. Kolano, T. Boudier, P. Mailly, M. H. Verlhac and M. E. Terret (2018). "Shifting meiotic to mitotic spindle assembly in oocytes disrupts chromosome alignment." EMBO Rep 19(2): 368-381. Bernstein, E., S. Y. Kim, M. A. Carmell, E. P. Murchison, H. Alcorn, M. Z. Li, A. A. Mills, S. J. Elledge, K. V. Anderson and G. J. Hannon (2003). "Dicer is essential for mouse development." Nat Genet 35(3): 215-217. Biggers, J. D., D. G. Whittingham and R. P. Donahue (1967). "The pattern of energy metabolism in the mouse oocyte and zygote." Proc Natl Acad Sci U S A 58(2): 560-567. Bloch, D. B., T. Gulick, K. D. Bloch and W. H. Yang (2006). "Processing body autoantibodies reconsidered." RNA 12(5): 707-709. Boag, P. R., A. Atalay, S. Robida, V. Reinke and T. K. Blackwell (2008). "Protection of specific maternal messenger RNAs by the P body protein CGH-1 (Dhh1/RCK) during Caenorhabditis elegans oogenesis." J Cell Biol 182(3): 543-557.

156

Bouniol-Baly, C., L. Hamraoui, J. Guibert, N. Beaujean, M. S. Szollosi and P. Debey (1999). "Differential transcriptional activity associated with chromatin configuration in fully grown mouse germinal vesicle oocytes." Biol Reprod 60(3): 580-587. Bowles, J., D. Knight, C. Smith, D. Wilhelm, J. Richman, S. Mamiya, K. Yashiro, K. Chawengsaksophak, M. J. Wilson, J. Rossant, H. Hamada and P. Koopman (2006). "Retinoid signaling determines germ cell fate in mice." Science 312(5773): 596-600. Brangwynne, C. P., C. R. Eckmann, D. S. Courson, A. Rybarska, C. Hoege, J. Gharakhani, F. Julicher and A. A. Hyman (2009). "Germline P granules are liquid droplets that localize by controlled dissolution/condensation." Science 324(5935): 1729-1732. Braun, J. E., E. Huntzinger and E. Izaurralde (2013). "The role of GW182 proteins in miRNA- mediated gene silencing." Adv Exp Med Biol 768: 147-163. Brennan, J., C. C. Lu, D. P. Norris, T. A. Rodriguez, R. S. Beddington and E. J. Robertson (2001). "Nodal signalling in the epiblast patterns the early mouse embryo." Nature 411(6840): 965-969. Briggs, R. and T. J. King (1952). "Transplantation of Living Nuclei From Blastula Cells into Enucleated Frogs' Eggs." Proc Natl Acad Sci U S A 38(5): 455-463. Bristol-Gould, S. K., P. K. Kreeger, C. G. Selkirk, S. M. Kilen, R. W. Cook, J. L. Kipp, L. D. Shea, K. E. Mayo and T. K. Woodruff (2006). "Postnatal regulation of germ cells by activin: the establishment of the initial follicle pool." Dev Biol 298(1): 132-148. Brower, P. T., E. Gizang, S. M. Boreen and R. M. Schultz (1981). "Biochemical studies of mammalian oogenesis: synthesis and stability of various classes of RNA during growth of the mouse oocyte in vitro." Dev Biol 86(2): 373-383. Brunet, S., A. S. Maria, P. Guillaud, D. Dujardin, J. Z. Kubiak and B. Maro (1999). "Kinetochore fibers are not involved in the formation of the first meiotic spindle in mouse oocytes, but control the exit from the first meiotic M phase." J Cell Biol 146(1): 1-12. Brunet, S. and B. Maro (2005). "Cytoskeleton and cell cycle control during meiotic maturation of the mouse oocyte: integrating time and space." Reproduction 130(6): 801-811. Buccione, R., S. Cecconi, C. Tatone, F. Mangia and R. Colonna (1987). "Follicle cell regulation of mammalian oocyte growth." J Exp Zool 242(3): 351-354. Buchan, J. R., R. M. Kolaitis, J. P. Taylor and R. Parker (2013). "Eukaryotic stress granules are cleared by autophagy and Cdc48/VCP function." Cell 153(7): 1461-1474. Buchan, J. R., D. Muhlrad and R. Parker (2008). "P bodies promote stress granule assembly in Saccharomyces cerevisiae." J Cell Biol 183(3): 441-455. Buchan, J. R. and R. Parker (2009). "Eukaryotic stress granules: the ins and outs of translation." Mol Cell 36(6): 932-941. Buehr, M., A. McLaren, A. Bartley and S. Darling (1993). "Proliferation and migration of primordial germ cells in We/We mouse embryos." Dev Dyn 198(3): 182-189. Burgoyne, P. S., S. K. Mahadevaiah and J. M. Turner (2009). "The consequences of asynapsis for mammalian meiosis." Nat Rev Genet 10(3): 207-216.

157

Burns, K. H., M. M. Viveiros, Y. Ren, P. Wang, F. J. DeMayo, D. E. Frail, J. J. Eppig and M. M. Matzuk (2003). "Roles of NPM2 in chromatin and nucleolar organization in oocytes and embryos." Science 300(5619): 633-636. Campolo, F., M. Gori, R. Favaro, S. Nicolis, M. Pellegrini, F. Botti, P. Rossi, E. A. Jannini and S. Dolci (2013). "Essential role of Sox2 for the establishment and maintenance of the germ cell line." Stem Cells 31(7): 1408-1421. Carthew, R. W. and E. J. Sontheimer (2009). "Origins and Mechanisms of miRNAs and siRNAs." Cell 136(4): 642-655. Castrillon, D. H., L. Miao, R. Kollipara, J. W. Horner and R. A. DePinho (2003). "Suppression of ovarian follicle activation in mice by the transcription factor Foxo3a." Science 301(5630): 215- 218. Caudron, M., G. Bunt, P. Bastiaens and E. Karsenti (2005). "Spatial coordination of spindle assembly by chromosome-mediated signaling gradients." Science 309(5739): 1373-1376. Cazzola, M. and R. C. Skoda (2000). "Translational pathophysiology: a novel molecular mechanism of human disease." Blood 95(11): 3280-3288. CFAS (2016). "Assisted Reproductive Technologies (ART) in Canada: 2015 Results from the Canadian ART Register (CARTR)". Chalfie, M., H. R. Horvitz and J. E. Sulston (1981). "Mutations that lead to reiterations in the cell lineages of C. elegans." Cell 24(1): 59-69. Chambers, I., J. Silva, D. Colby, J. Nichols, B. Nijmeijer, M. Robertson, J. Vrana, K. Jones, L. Grotewold and A. Smith (2007). "Nanog safeguards pluripotency and mediates germline development." Nature 450(7173): 1230-1234. Chang, C. T., N. Bercovich, B. Loh, S. Jonas and E. Izaurralde (2014). "The activation of the decapping enzyme DCP2 by DCP1 occurs on the EDC4 scaffold and involves a conserved loop in DCP1." Nucleic Acids Res 42(8): 5217-5233. Chang, H. and M. M. Matzuk (2001). "Smad5 is required for mouse primordial germ cell development." Mech Dev 104(1-2): 61-67. Chao, A. S., Y. L. Chang, A. Chao, T. S. Wu, L. Y. Yang, R. Lian and Y. C. Huang (2017). "Seropositivity of influenza A H1NI in mothers and infants following maternal vaccination with trivalent seasonal influenza vaccine after the 2009 pandemic." Taiwan J Obstet Gynecol 56(1): 37- 40. Charlesworth, A., H. A. Meijer and C. H. de Moor (2013). "Specificity factors in cytoplasmic polyadenylation." Wiley Interdiscip Rev RNA 4(4): 437-461. Chen, C. Y. and A. B. Shyu (2011). "Mechanisms of deadenylation-dependent decay." Wiley Interdiscip Rev RNA 2(2): 167-183. Chen, J., C. Melton, N. Suh, J. S. Oh, K. Horner, F. Xie, C. Sette, R. Blelloch and M. Conti (2011). "Genome-wide analysis of translation reveals a critical role for deleted in azoospermia-like (Dazl) at the oocyte-to-zygote transition." Genes Dev 25(7): 755-766.

158

Choi, Y., D. J. Ballow, Y. Xin and A. Rajkovic (2008). "Lim homeobox gene, lhx8, is essential for mouse oocyte differentiation and survival." Biol Reprod 79(3): 442-449. Chouinard, L. A. (1975). "A light- and electron-microscope study of the oocyte nucleus during development of the antral follicle in the prepubertal mouse." J Cell Sci 17(3): 589-615. Chuva de Sousa Lopes, S. M., K. Hayashi, T. C. Shovlin, W. Mifsud, M. A. Surani and A. McLaren (2008). "X chromosome activity in mouse XX primordial germ cells." PLoS Genet 4(2): e30. Clarke, H. J. (2012). "Post-transcriptional control of gene expression during mouse oogenesis." Results Probl Cell Differ 55: 1-21. Clarke, H. J. and K. F. Vieux (2015). "Epigenetic inheritance through the female germ-line: The known, the unknown, and the possible." Semin Cell Dev Biol 43: 106-116. Colonna, R., S. Cecconi, R. Buccione and F. Mangia (1983). "Amino acid transport systems in growing mouse oocytes." Cell Biol Int Rep 7(12): 1007-1015. Colonna, R. and F. Mangia (1983). "Mechanisms of amino acid uptake in cumulus-enclosed mouse oocytes." Biol Reprod 28(4): 797-803. Colvin, J. S., R. P. Green, J. Schmahl, B. Capel and D. M. Ornitz (2001). "Male-to-female sex reversal in mice lacking fibroblast growth factor 9." Cell 104(6): 875-889. Cougot, N., S. Babajko and B. Seraphin (2004). "Cytoplasmic foci are sites of mRNA decay in human cells." J Cell Biol 165(1): 31-40. Couttet, P., M. Fromont-Racine, D. Steel, R. Pictet and T. Grange (1997). "Messenger RNA deadenylylation precedes decapping in mammalian cells." Proc Natl Acad Sci U S A 94(11): 5628- 5633. Cragle, C. and A. M. MacNicol (2014). "Musashi protein-directed translational activation of target mRNAs is mediated by the poly(A) polymerase, germ line development defective-2." J Biol Chem 289(20): 14239-14251. David, D. C., N. Ollikainen, J. C. Trinidad, M. P. Cary, A. L. Burlingame and C. Kenyon (2010). "Widespread protein aggregation as an inherent part of aging in C. elegans." PLoS Biol 8(8): e1000450. Davydenko, O., R. M. Schultz and M. A. Lampson (2013). "Increased CDK1 activity determines the timing of kinetochore-microtubule attachments in meiosis I." J Cell Biol 202(2): 221-229. Day, D. A. and M. F. Tuite (1998). "Post-transcriptional gene regulatory mechanisms in eukaryotes: an overview." J Endocrinol 157(3): 361-371. De La Fuente, R. and J. J. Eppig (2001). "Transcriptional activity of the mouse oocyte genome: companion granulosa cells modulate transcription and chromatin remodeling." Dev Biol 229(1): 224-236. De La Fuente, R., M. M. Viveiros, K. H. Burns, E. Y. Adashi, M. M. Matzuk and J. J. Eppig (2004). "Major chromatin remodeling in the germinal vesicle (GV) of mammalian oocytes is dispensable for global transcriptional silencing but required for centromeric heterochromatin function." Dev Biol 275(2): 447-458.

159

De Leon, V., A. Johnson and R. Bachvarova (1983). "Half-lives and relative amounts of stored and polysomal ribosomes and poly(A) + RNA in mouse oocytes." Dev Biol 98(2): 400-408. de Napoles, M., T. Nesterova and N. Brockdorff (2007). "Early loss of Xist RNA expression and inactive X chromosome associated chromatin modification in developing primordial germ cells." PLoS One 2(9): e860. Debey, P., M. S. Szollosi, D. Szollosi, D. Vautier, A. Girousse and D. Besombes (1993). "Competent mouse oocytes isolated from antral follicles exhibit different chromatin organization and follow different maturation dynamics." Mol Reprod Dev 36(1): 59-74. Dickson, K. S., S. R. Thompson, N. K. Gray and M. Wickens (2001). "Poly(A) polymerase and the regulation of cytoplasmic polyadenylation." J Biol Chem 276(45): 41810-41816. Djuranovic, S., A. Nahvi and R. Green (2012). "miRNA-mediated gene silencing by translational repression followed by mRNA deadenylation and decay." Science 336(6078): 237-240. Doidge, R., S. Mittal, A. Aslam and G. S. Winkler (2012). "Deadenylation of cytoplasmic mRNA by the mammalian Ccr4-Not complex." Biochem Soc Trans 40(4): 896-901. Dolci, S., M. Pesce and M. De Felici (1993). "Combined action of stem cell factor, leukemia inhibitory factor, and cAMP on in vitro proliferation of mouse primordial germ cells." Mol Reprod Dev 35(2): 134-139. Donahue, R. P. and S. Stern (1968). "Follicular cell support of oocyte maturation: production of pyruvate in vitro." J Reprod Fertil 17(2): 395-398. Duncan, R., S. C. Milburn and J. W. Hershey (1987). "Regulated phosphorylation and low abundance of HeLa cell initiation factor eIF-4F suggest a role in translational control. Heat shock effects on eIF-4F." J Biol Chem 262(1): 380-388. Dunn, K. W., M. M. Kamocka and J. H. McDonald (2011). "A practical guide to evaluating colocalization in biological microscopy." Am J Physiol Cell Physiol 300(4): C723-742. Dworkin, M. B. and E. Dworkin-Rastl (1990). "Functions of maternal mRNA in early development." Mol Reprod Dev 26(3): 261-297. Eckmann, C. R., C. Rammelt and E. Wahle (2011). "Control of poly(A) tail length." Wiley Interdiscip Rev RNA 2(3): 348-361. Edwards, N., R. Farookhi and H. J. Clarke (2015). "Identification of a beta-galactosidase transgene that provides a live-cell marker of transcriptional activity in growing oocytes and embryos." Mol Hum Reprod 21(7): 583-593. Eichenlaub-Ritter, U. and M. Peschke (2002). "Expression in in-vivo and in-vitro growing and maturing oocytes: focus on regulation of expression at the translational level." Hum Reprod Update 8(1): 21-41. Eissenberg, J. C. and G. Reuter (2009). "Cellular mechanism for targeting heterochromatin formation in Drosophila." Int Rev Cell Mol Biol 273: 1-47.

160

El-Hayek, S. and H. J. Clarke (2015). "Follicle-Stimulating Hormone Increases Gap Junctional Communication Between Somatic and Germ-Line Follicular Compartments During Murine Oogenesis." Biol Reprod 93(2): 47. El-Hayek, S. and H. J. Clarke (2016). "Control of oocyte growth and development by intercellular communication within the follicular niche." Res Probl Cell Differ 58: 191-224. El-Hayek, S., Q. Yang, L. Abbassi, G. FitzHarris and H. J. Clarke (2018). "Mammalian Oocytes Locally Remodel Follicular Architecture to Provide the Foundation for Germline-Soma Communication." Curr Biol 28(7): 1124-1131 e1123. Elbaum-Garfinkle, S., Y. Kim, K. Szczepaniak, C. C. Chen, C. R. Eckmann, S. Myong and C. P. Brangwynne (2015). "The disordered P granule protein LAF-1 drives phase separation into droplets with tunable viscosity and dynamics." Proc Natl Acad Sci U S A 112(23): 7189-7194. Emori, C. and K. Sugiura (2014). "Role of oocyte-derived paracrine factors in follicular development." Anim Sci J 85(6): 627-633. Eppig, J. J. and M. J. O'Brien (1996). "Development in vitro of mouse oocytes from primordial follicles." Biol Reprod 54(1): 197-207. Eppig, J. J. and A. C. Schroeder (1989). "Capacity of mouse oocytes from preantral follicles to undergo embryogenesis and development to live young after growth, maturation, and fertilization in vitro." Biol Reprod 41(2): 268-276. Eulalio, A., I. Behm-Ansmant, D. Schweizer and E. Izaurralde (2007). "P-body formation is a consequence, not the cause, of RNA-mediated gene silencing." Mol Cell Biol 27(11): 3970-3981. Ewen, K. A. and P. Koopman (2010). "Mouse germ cell development: from specification to sex determination." Mol Cell Endocrinol 323(1): 76-93. Eystathioy, T., A. Jakymiw, E. K. Chan, B. Seraphin, N. Cougot and M. J. Fritzler (2003). "The GW182 protein colocalizes with mRNA degradation associated proteins hDcp1 and hLSm4 in cytoplasmic GW bodies." RNA 9(10): 1171-1173. Faddy MJ, Telfer E, Gosden RG. "The kinetics of pre-antral follicle development in ovaries of CBA/Ca mice during the first 14 weeks of life." Cell Tissue Kinet. (1987) 20(6):551-60 Fair, T., S. C. Hulshof, P. Hyttel, T. Greve and M. Boland (1997). "Nucleus ultrastructure and transcriptional activity of bovine oocytes in preantral and early antral follicles." Mol Reprod Dev 46(2): 208-215. Fair, T., P. Hyttel, T. Greve and M. Boland (1996). "Nucleus structure and transcriptional activity in relation to oocyte diameter in cattle." Mol Reprod Dev 43(4): 503-512. Farini, D., G. La Sala, M. Tedesco and M. De Felici (2007). "Chemoattractant action and molecular signaling pathways of Kit ligand on mouse primordial germ cells." Dev Biol 306(2): 572-583. Feng, C. W., J. Bowles and P. Koopman (2014). "Control of mammalian germ cell entry into meiosis." Mol Cell Endocrinol 382(1): 488-497.

161

Fenwick, M. A., J. M. Mora, Y. T. Mansour, C. Baithun, S. Franks and K. Hardy (2013). "Investigations of TGF-beta signaling in preantral follicles of female mice reveal differential roles for bone morphogenetic protein 15." Endocrinology 154(9): 3423-3436. Flemr, M., J. Ma, R. M. Schultz and P. Svoboda (2010). "P-body loss is concomitant with formation of a messenger RNA storage domain in mouse oocytes." Biology of Reproduction 82(5): 1008-1017. Flemr, M. and P. Svoboda (2011). "Ribonucleoprotein localization in mouse oocytes." Methods 53(2): 136-141. Ford, C. E., E. P. Evans, M. D. Burtenshaw, H. M. Clegg, M. Tuffrey and R. D. Barnes (1975). "A functional 'sex-reversed' oocyte in the mouse." Proc R Soc Lond B Biol Sci 190(1099): 187- 197. Fox, C. A. and M. Wickens (1990). "Poly(A) removal during oocyte maturation: a default reaction selectively prevented by specific sequences in the 3' UTR of certain maternal mRNAs." Genes Dev 4(12B): 2287-2298. Franciosi, F., G. Coticchio, V. Lodde, I. Tessaro, S. C. Modina, R. Fadini, M. Dal Canto, M. M. Renzini, D. F. Albertini and A. M. Luciano (2014). "Natriuretic peptide precursor C delays meiotic resumption and sustains gap junction-mediated communication in bovine cumulus-enclosed oocytes." Biol Reprod 91(3): 61. Franciosi, F., S. Manandhar and M. Conti (2016). "FSH Regulates mRNA Translation in Mouse Oocytes and Promotes Developmental Competence." Endocrinology 157(2): 872-882. Freimer, J. W., R. Krishnakumar, M. S. Cook and R. Blelloch (2018). "Expression of Alternative Ago2 Isoform Associated with Loss of microRNA-Driven Translational Repression in Mouse Oocytes." Curr Biol 28(2): 296-302 e293. Fu, L. N., R. Q. Ye, L. W. Browder and R. N. Johnston (1991). "Translational potentiation of messenger RNA with secondary structure in Xenopus." Science 251(4995): 807-810. Fujii, W., T. Nishimura, K. Kano, K. Sugiura and K. Naito (2011). "CDK7 and CCNH are components of CDK-activating kinase and are required for meiotic progression of pig oocytes." Biol Reprod 85(6): 1124-1132. Fulka, J., Jr., J. Motlik, J. Fulka and F. Jilek (1986). "Effect of cycloheximide on nuclear maturation of pig and mouse oocytes." J Reprod Fertil 77(1): 281-285. Furuichi, Y., M. Morgan, A. J. Shatkin, W. Jelinek, M. Salditt-Georgieff and J. E. Darnell (1975). "Methylated, blocked 5 termini in HeLa cell mRNA." Proc Natl Acad Sci U S A 72(5): 1904-1908. Gallie, D. R. (1991). "The cap and poly(A) tail function synergistically to regulate mRNA translational efficiency." Genes Dev 5(11): 2108-2116. Gautier, J., T. Matsukawa, P. Nurse and J. Maller (1989). "Dephosphorylation and activation of Xenopus p34cdc2 protein kinase during the cell cycle." Nature 339(6226): 626-629.

162

Gebauer, F. and J. D. Richter (1996). "Mouse cytoplasmic polyadenylylation element binding protein: an evolutionarily conserved protein that interacts with the cytoplasmic polyadenylylation elements of c-mos mRNA." Proc Natl Acad Sci U S A 93(25): 14602-14607. Gebauer, F., W. Xu, G. M. Cooper and J. D. Richter (1994). "Translational control by cytoplasmic polyadenylation of c-mos mRNA is necessary for oocyte maturation in the mouse." EMBO J 13(23): 5712-5720. Ghisolfi, L., S. Dutt, M. E. McConkey, B. L. Ebert and P. Anderson (2012). "Stress granules contribute to alpha-globin homeostasis in differentiating erythroid cells." Biochem Biophys Res Commun 420(4): 768-774. Gilula, N. B., O. R. Reeves and A. Steinbach (1972). "Metabolic coupling, ionic coupling and cell contacts." Nature 235(5336): 262-265. Ginsburg, M., M. H. Snow and A. McLaren (1990). "Primordial germ cells in the mouse embryo during gastrulation." Development 110(2): 521-528. Global Market Insight (2016). "Assisted Reproductive Technology (ART) Market Size by Procedure (Fresh Donor, Fresh Nondonor, Frozen Donor, Frozen Nondonor, Egg/Embryo bank), Industry Analysis Report, Regional Outlook (U.S, Canada, UK, France, Germany, Italy, Spain, China, India, Japan, Brazil, Mexico, South Africa), Application Potential, Price Trends, Competitive Market Share & Forecast, 2016 – 202." Gohin, M., E. Fournier, I. Dufort and M. A. Sirard (2014). "Discovery, identification and sequence analysis of RNAs selected for very short or long poly A tail in immature bovine oocytes." Mol Hum Reprod 20(2): 127-138. Goldstrohm, A. C. and M. Wickens (2008). "Multifunctional deadenylase complexes diversify mRNA control." Nat Rev Mol Cell Biol 9(4): 337-344. Gomperts, M., M. Garcia-Castro, C. Wylie and J. Heasman (1994). "Interactions between primordial germ cells play a role in their migration in mouse embryos." Development 120(1): 135- 141. Gorre, N., D. Adhikari, R. Lindkvist, M. Brannstrom, K. Liu and Y. Shen (2014). "mTORC1 Signaling in oocytes is dispensable for the survival of primordial follicles and for female fertility." PLoS One 9(10): e110491. Goss, D. J., L. J. Parkhurst, H. B. Mehta, C. L. Woodley and A. J. Wahba (1984). "Studies on the role of eukaryotic nucleotide exchange factor in polypeptide chain initiation." J Biol Chem 259(12): 7374-7377. Goss, D. J., C. L. Woodley and A. J. Wahba (1987). "A fluorescence study of the binding of eucaryotic initiation factors to messenger RNA and messenger RNA analogues." Biochemistry 26(6): 1551-1556. Graindorge, A., R. Thuret, N. Pollet, H. B. Osborne and Y. Audic (2006). "Identification of post- transcriptionally regulated Xenopus tropicalis maternal mRNAs by microarray." Nucleic Acids Res 34(3): 986-995.

163

Grantham, R., C. Gautier and M. Gouy (1980). "Codon frequencies in 119 individual genes confirm consistent choices of degenerate bases according to genome type." Nucleic Acids Res 8(9): 1893-1912. Grantham, R., C. Gautier, M. Gouy, R. Mercier and A. Pave (1980). "Codon catalog usage and the genome hypothesis." Nucleic Acids Res 8(1): r49-r62. Groisman, I., Y. S. Huang, R. Mendez, Q. Cao and J. D. Richter (2001). "Translational control of embryonic cell division by CPEB and maskin." Cold Spring Harb Symp Quant Biol 66: 345-351. Gruss, O. J., R. E. Carazo-Salas, C. A. Schatz, G. Guarguaglini, J. Kast, M. Wilm, N. Le Bot, I. Vernos, E. Karsenti and I. W. Mattaj (2001). "Ran induces spindle assembly by reversing the inhibitory effect of importin alpha on TPX2 activity." Cell 104(1): 83-93. Gu, Y., C. Runyan, A. Shoemaker, A. Surani and C. Wylie (2009). "Steel factor controls primordial germ cell survival and motility from the time of their specification in the allantois, and provides a continuous niche throughout their migration." Development 136(8): 1295-1303. Guo, J., L. Shi, X. Gong, M. Jiang, Y. Yin, X. Zhang, H. Yin, H. Li, C. Emori, K. Sugiura, J. J. Eppig and Y. Q. Su (2016). "Oocyte-dependent activation of MTOR in cumulus cells controls the development and survival of cumulus-oocyte complexes." J Cell Sci 129(16): 3091-3103. Gurdon, J. B. (1962). "The developmental capacity of nuclei taken from intestinal epithelium cells of feeding tadpoles." J Embryol Exp Morphol 10: 622-640. Gurdon, J. B., T. R. Elsdale and M. Fischberg (1958). "Sexually mature individuals of Xenopus laevis from the transplantation of single somatic nuclei." Nature 182(4627): 64-65. Haghighat, N. and L. J. Van Winkle (1990). "Developmental change in follicular cell-enhanced amino acid uptake into mouse oocytes that depends on intact gap junctions and transport system Gly." J Exp Zool 253(1): 71-82. Hajkova, P., K. Ancelin, T. Waldmann, N. Lacoste, U. C. Lange, F. Cesari, C. Lee, G. Almouzni, R. Schneider and M. A. Surani (2008). "Chromatin dynamics during epigenetic reprogramming in the mouse germ line." Nature 452(7189): 877-881. Hampl, A. and J. J. Eppig (1995). "Analysis of the mechanism(s) of metaphase I arrest in maturing mouse oocytes." Development 121(4): 925-933. Han, S. J., R. Chen, M. P. Paronetto and M. Conti (2005). "Wee1B is an oocyte-specific kinase involved in the control of meiotic arrest in the mouse." Curr Biol 15(18): 1670-1676. Han, S. J. and M. Conti (2006). "New pathways from PKA to the Cdc2/cyclin B complex in oocytes: Wee1B as a potential PKA substrate." Cell Cycle 5(3): 227-231. Harding, H. P., Y. Zhang, A. Bertolotti, H. Zeng and D. Ron (2000). "Perk is essential for translational regulation and cell survival during the unfolded protein response." Mol Cell 5(5): 897-904. Hardy K, Mora JM, Dunlop C, Carzaniga R, Franks S, Fenwick MA (2018). "Nuclear exclusion of SMAD2/3 in granulosa cells is associated with primordial follicle activation in the mouse ovary. " J Cell Sci 131(17)

164

Hasler, J. and K. Strub (2006). "Alu RNP and Alu RNA regulate translation initiation in vitro." Nucleic Acids Res 34(8): 2374-2385. Hayashi, K., S. M. de Sousa Lopes and M. A. Surani (2007). "Germ cell specification in mice." Science 316(5823): 394-396. Heald, R. and F. McKeon (1990). "Mutations of phosphorylation sites in lamin A that prevent nuclear lamina disassembly in mitosis." Cell 61(4): 579-589. Hetzer, M., O. J. Gruss and I. W. Mattaj (2002). "The Ran GTPase as a marker of chromosome position in spindle formation and nuclear envelope assembly." Nat Cell Biol 4(7): E177-184. Hiremath, L. S., N. R. Webb and R. E. Rhoads (1985). "Immunological detection of the messenger RNA cap-binding protein." J Biol Chem 260(13): 7843-7849. Hodgman, R., J. Tay, R. Mendez and J. D. Richter (2001). "CPEB phosphorylation and cytoplasmic polyadenylation are catalyzed by the kinase IAK1/Eg2 in maturing mouse oocytes." Development 128(14): 2815-2822. Hong, Y., R. C. Lee and V. Ambros (2000). "Structure and function analysis of LIN-14, a temporal regulator of postembryonic developmental events in Caenorhabditis elegans." Mol Cell Biol 20(6): 2285-2295. Horvat, F. F., H.; Jankele, R.; Malik, R.; Jun, M.; Solcova, K.; Sedlacek, R.; Vlahovicek, K.; Schultz, R.M.; Svoboda P. (2018). "Role of Cnot6l in maternal mRNA turnover." BioRxiv. Hsu, C. L. and A. Stevens (1993). "Yeast cells lacking 5'-->3' exoribonuclease 1 contain mRNA species that are poly(A) deficient and partially lack the 5' cap structure." Mol Cell Biol 13(8): 4826-4835. Hu, S., M. Wang, G. Cai and M. He (2013). "Genetic code-guided protein synthesis and folding in Escherichia coli." J Biol Chem 288(43): 30855-30861. Hu, W. and J. Coller (2012). "What comes first: translational repression or mRNA degradation? The deepening mystery of microRNA function." Cell Res 22(9): 1322-1324. Hulas-Stasiak, M. and A. Gawron (2011). "Follicular atresia in the prepubertal spiny mouse (Acomys cahirinus) ovary." Apoptosis 16(10): 967-975. Hyman, A. A., C. A. Weber and F. Julicher (2014). "Liquid-liquid phase separation in biology." Annu Rev Cell Dev Biol 30: 39-58. Igarashi, H., T. Takahashi, E. Takahashi, N. Tezuka, K. Nakahara, K. Takahashi and H. Kurachi (2005). "Aged mouse oocytes fail to readjust intracellular adenosine triphosphates at fertilization." Biol Reprod 72(5): 1256-1261. Ikeda, K., P. J. Mason and M. Bessler (2011). "3'UTR-truncated Hmga2 cDNA causes MPN-like hematopoiesis by conferring a clonal growth advantage at the level of HSC in mice." Blood 117(22): 5860-5869. Ivanov, P. A., E. M. Chudinova and E. S. Nadezhdina (2003). "Disruption of microtubules inhibits cytoplasmic ribonucleoprotein stress granule formation." Exp Cell Res 290(2): 227-233.

165

Jackson, J. P., A. M. Lindroth, X. Cao and S. E. Jacobsen (2002). "Control of CpNpG DNA methylation by the KRYPTONITE histone H3 methyltransferase." Nature 416(6880): 556-560. Jagarlamudi, K., L. Liu, D. Adhikari, P. Reddy, A. Idahl, U. Ottander, E. Lundin and K. Liu (2009). "Oocyte-specific deletion of Pten in mice reveals a stage-specific function of PTEN/PI3K signaling in oocytes in controlling follicular activation." PLoS One 4(7): e6186. Jahn, C. L., M. M. Baran and R. Bachvarova (1976). "Stability of RNA synthesized by the mouse oocyte during its major growth phase." J Exp Zool 197(2): 161-171. Jakob, S. and R. Lovell-Badge (2011). "Sex determination and the control of Sox9 expression in mammals." FEBS J 278(7): 1002-1009. Jalkanen, A. L., S. J. Coleman and J. Wilusz (2014). "Determinants and implications of mRNA poly(A) tail size--does this protein make my tail look big?" Semin Cell Dev Biol 34: 24-32. Jameson, S. A., Y. T. Lin and B. Capel (2012). "Testis development requires the repression of Wnt4 by Fgf signaling." Dev Biol 370(1): 24-32. Jansova, D., A. Tetkova, M. Koncicka, M. Kubelka and A. Susor (2018). "Localization of RNA and translation in the mammalian oocyte and embryo." PLoS One 13(3): e0192544. Jin, C., H. Rajabi, C. M. Rodrigo, J. A. Porco, Jr. and D. Kufe (2013). "Targeting the eIF4A RNA helicase blocks translation of the MUC1-C oncoprotein." Oncogene 32(17): 2179-2188. Johnson J, Canning J, Kaneko T, Pru JK, Tilly JL. (2004) "Germline stem cells and follicular renewal in the postnatal mammalian ovary." Nature 428(6979):145-50. Joly, W., A. Chartier, P. Rojas-Rios, I. Busseau and M. Simonelig (2013). "The CCR4 deadenylase acts with Nanos and Pumilio in the fine-tuning of Mei-P26 expression to promote germline stem cell self-renewal." Stem Cell Reports 1(5): 411-424. Jud, M. C., M. J. Czerwinski, M. P. Wood, R. A. Young, C. M. Gallo, J. S. Bickel, E. L. Petty, J. M. Mason, B. A. Little, P. A. Padilla and J. A. Schisa (2008). "Large P body-like RNPs form in C. elegans oocytes in response to arrested ovulation, heat shock, osmotic stress, and anoxia and are regulated by the major sperm protein pathway." Dev Biol 318(1): 38-51. Kagiwada, S., K. Kurimoto, T. Hirota, M. Yamaji and M. Saitou (2013). "Replication-coupled passive DNA demethylation for the erasure of genome imprints in mice." EMBO J 32(3): 340- 353. Kakiuchi K, Tsuda A, Goto Y, Shimada T, Taniguchi K, Takagishi K, Kubota H. (2014) "Cell- surface DEAD-box polypeptide 4-immunoreactive cells and gonocytes are two distinct populations in postnatal porcine testes." Biol Reprod. 90(4):82 Kanatsu-Shinohara, M., R. M. Schultz and G. S. Kopf (2000). "Acquisition of meiotic competence in mouse oocytes: absolute amounts of p34(cdc2), cyclin B1, cdc25C, and wee1 in meiotically incompetent and competent oocytes." Biol Reprod 63(6): 1610-1616. Kang, M. K. and S. J. Han (2011). "Post-transcriptional and post-translational regulation during mouse oocyte maturation." BMB Rep 44(3): 147-157.

166

Kashimada, K. and P. Koopman (2010). "Sry: the master switch in mammalian sex determination." Development 137(23): 3921-3930. Kashimada, K., T. Svingen, C. W. Feng, E. Pelosi, S. Bagheri-Fam, V. R. Harley, D. Schlessinger, J. Bowles and P. Koopman (2011). "Antagonistic regulation of Cyp26b1 by transcription factors SOX9/SF1 and FOXL2 during gonadal development in mice." FASEB J 25(10): 3561-3569. Kawase, E., K. Hashimoto and R. A. Pedersen (2004). "Autocrine and paracrine mechanisms regulating primordial germ cell proliferation." Mol Reprod Dev 68(1): 5-16. Kedersha, N., G. Stoecklin, M. Ayodele, P. Yacono, J. Lykke-Andersen, M. J. Fritzler, D. Scheuner, R. J. Kaufman, D. E. Golan and P. Anderson (2005). "Stress granules and processing bodies are dynamically linked sites of mRNP remodeling." J Cell Biol 169(6): 871-884. Kedersha, N. L., M. Gupta, W. Li, I. Miller and P. Anderson (1999). "RNA-binding proteins TIA- 1 and TIAR link the phosphorylation of eIF-2 alpha to the assembly of mammalian stress granules." J Cell Biol 147(7): 1431-1442. Kehler, J., E. Tolkunova, B. Koschorz, M. Pesce, L. Gentile, M. Boiani, H. Lomeli, A. Nagy, K. J. McLaughlin, H. R. Scholer and A. Tomilin (2004). "Oct4 is required for primordial germ cell survival." EMBO Rep 5(11): 1078-1083. Kim, J. H. and J. D. Richter (2006). "Opposing polymerase-deadenylase activities regulate cytoplasmic polyadenylation." Mol Cell 24(2): 173-183. Kim, Y., A. Kobayashi, R. Sekido, L. DiNapoli, J. Brennan, M. C. Chaboissier, F. Poulat, R. R. Behringer, R. Lovell-Badge and B. Capel (2006). "Fgf9 and Wnt4 act as antagonistic signals to regulate mammalian sex determination." PLoS Biol 4(6): e187. Kimball, S. R., R. L. Horetsky, D. Ron, L. S. Jefferson and H. P. Harding (2003). "Mammalian stress granules represent sites of accumulation of stalled translation initiation complexes." Am J Physiol Cell Physiol 284(2): C273-284. Kiss, D. L., K. M. Oman, J. A. Dougherty, C. Mukherjee, R. Bundschuh and D. R. Schoenberg (2016). "Cap homeostasis is independent of poly(A) tail length." Nucleic Acids Res 44(1): 304- 314. Klumpp, S., J. Dong and T. Hwa (2012). "On ribosome load, codon bias and protein abundance." PLoS One 7(11): e48542. Kohli, R. M. and Y. Zhang (2013). "TET enzymes, TDG and the dynamics of DNA demethylation." Nature 502(7472): 472-479. Kotani, T., K. Yasuda, R. Ota and M. Yamashita (2013). "Cyclin B1 mRNA translation is temporally controlled through formation and disassembly of RNA granules." J Cell Biol 202(7): 1041-1055. Kouzarides, T. (2007). "Chromatin modifications and their function." Cell 128(4): 693-705. Kozak, M. (1989). "The scanning model for translation: an update." J Cell Biol 108(2): 229-241. Kozak, M. (1990). "Downstream secondary structure facilitates recognition of initiator codons by eukaryotic ribosomes." Proc Natl Acad Sci U S A 87(21): 8301-8305.

167

Kuhn, U. and T. Pieler (1996). "Xenopus poly(A) binding protein: functional domains in RNA binding and protein-protein interaction." J Mol Biol 256(1): 20-30. Kumar, D. L. and T. DeFalco (2017). "Of Mice and Men: In Vivo and In Vitro Studies of Primordial Germ Cell Specification." Semin Reprod Med 35(2): 139-146. Kurimoto, K., Y. Yabuta, Y. Ohinata, M. Shigeta, K. Yamanaka and M. Saitou (2008). "Complex genome-wide transcription dynamics orchestrated by Blimp1 for the specification of the germ cell lineage in mice." Genes Dev 22(12): 1617-1635. Kurimoto, K., M. Yamaji, Y. Seki and M. Saitou (2008). "Specification of the germ cell lineage in mice: a process orchestrated by the PR-domain proteins, Blimp1 and Prdm14." Cell Cycle 7(22): 3514-3518. Lagier-Tourenne, C., M. Polymenidou and D. W. Cleveland (2010). "TDP-43 and FUS/TLS: emerging roles in RNA processing and neurodegeneration." Hum Mol Genet 19(R1): R46-64. LaGrandeur, T. E. and R. Parker (1998). "Isolation and characterization of Dcp1p, the yeast mRNA decapping enzyme." EMBO J 17(5): 1487-1496. Lara-Gonzalez, P. and S. S. Taylor (2012). "Cohesion fatigue explains why pharmacological inhibition of the APC/C induces a spindle checkpoint-dependent mitotic arrest." PLoS One 7(11): e49041. Lau, N. C., A. Kolkman, F. M. van Schaik, K. W. Mulder, W. W. Pijnappel, A. J. Heck and H. T. Timmers (2009). "Human Ccr4-Not complexes contain variable deadenylase subunits." Biochem J 422(3): 443-453. Le, S. Y. and J. V. Maizel, Jr. (1997). "A common RNA structural motif involved in the internal initiation of translation of cellular mRNAs." Nucleic Acids Res 25(2): 362-369. Lechler, M. C., E. D. Crawford, N. Groh, K. Widmaier, R. Jung, J. Kirstein, J. C. Trinidad, A. L. Burlingame and D. C. David (2017). "Reduced Insulin/IGF-1 Signaling Restores the Dynamic Properties of Key Stress Granule Proteins during Aging." Cell Rep 18(2): 454-467. Lee, C., P. Occhipinti and A. S. Gladfelter (2015). "PolyQ-dependent RNA-protein assemblies control symmetry breaking." J Cell Biol 208(5): 533-544. Lee, R. C., R. L. Feinbaum and V. Ambros (1993). "The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14." Cell 75(5): 843-854. Lee, W. S., F. Otsuka, R. K. Moore and S. Shimasaki (2001). "Effect of bone morphogenetic protein-7 on folliculogenesis and ovulation in the rat." Biol Reprod 65(4): 994-999. Leese, H. J. and A. M. Barton (1985). "Production of pyruvate by isolated mouse cumulus cells." J Exp Zool 234(2): 231-236. Lei L, Spradling AC. (2013) "Female mice lack adult germ-line stem cells but sustain oogenesis using stable primordial follicles." Proc Natl Acad Sci U S A 110(21):8585-90 Li, J., K. Kawamura, Y. Cheng, S. Liu, C. Klein, S. Liu, E. K. Duan and A. J. Hsueh (2010). "Activation of dormant ovarian follicles to generate mature eggs." Proc Natl Acad Sci U S A 107(22): 10280-10284.

168

Lin, Y., D. S. Protter, M. K. Rosen and R. Parker (2015). "Formation and Maturation of Phase- Separated Liquid Droplets by RNA-Binding Proteins." Mol Cell 60(2): 208-219. Lincoln, A. J., D. Wickramasinghe, P. Stein, R. M. Schultz, M. E. Palko, M. P. De Miguel, L. Tessarollo and P. J. Donovan (2002). "Cdc25b phosphatase is required for resumption of meiosis during oocyte maturation." Nat Genet 30(4): 446-449. Liu, H. C., Y. Tang, Z. He and Z. Rosenwaks (2010). "Dicer is a key player in oocyte maturation." J Assist Reprod Genet 27(9-10): 571-580. Liu, L., S. Rajareddy, P. Reddy, K. Jagarlamudi, C. Du, Y. Shen, Y. Guo, K. Boman, E. Lundin, U. Ottander, G. Selstam and K. Liu (2007). "Phosphorylation and inactivation of glycogen synthase kinase-3 by soluble kit ligand in mouse oocytes during early follicular development." J Mol Endocrinol 38(1-2): 137-146. Liu Y, Wu C, Lyu Q, Yang D, Albertini DF, Keefe DL, Liu L. (2007) "Germline stem cells and neo-oogenesis in the adult human ovary." Dev Biol. 306(1):112-20 Liu, Z., D. H. Castrillon, W. Zhou and J. S. Richards (2013). "FOXO1/3 depletion in granulosa cells alters follicle growth, death and regulation of pituitary FSH." Mol Endocrinol 27(2): 238- 252. Lohka, M.J., Hayes, M.K. and Maller, J.M. (1988) "Purification of maturation-promoting factor, an intracellular regulator of early mitotic events." Proc Natl Acad Sci USA 85:3009–3013. Lonergan, P. and T. Fair (2008). "In vitro-produced bovine embryos: dealing with the warts." Theriogenology 69(1): 17-22. Lopez de Quinto, S., E. Lafuente and E. Martinez-Salas (2001). "IRES interaction with translation initiation factors: functional characterization of novel RNA contacts with eIF3, eIF4B, and eIF4GII." RNA 7(9): 1213-1226. Loschi, M., C. C. Leishman, N. Berardone and G. L. Boccaccio (2009). "Dynein and kinesin regulate stress-granule and P-body dynamics." J Cell Sci 122(Pt 21): 3973-3982. Ma, J., M. Flemr, P. Stein, P. Berninger, R. Malik, M. Zavolan, P. Svoboda and R. M. Schultz (2010). "MicroRNA activity is suppressed in mouse oocytes." Curr Biol 20(3): 265-270. Ma, J., M. Flemr, H. Strnad, P. Svoboda and R. M. Schultz (2013). "Maternally recruited DCP1A and DCP2 contribute to messenger RNA degradation during oocyte maturation and genome activation in mouse." Biol Reprod 88(1): 11. Ma, J., Y. Fukuda and R. M. Schultz (2015). "Mobilization of Dormant Cnot7 mRNA Promotes Deadenylation of Maternal Transcripts During Mouse Oocyte Maturation." Biol Reprod 93(2): 48. Ma, P., H. Pan, R. L. Montgomery, E. N. Olson and R. M. Schultz (2012). "Compensatory functions of histone deacetylase 1 (HDAC1) and HDAC2 regulate transcription and apoptosis during mouse oocyte development." Proc Natl Acad Sci U S A 109(8): E481-489. MacLennan, M., J. H. Crichton, C. J. Playfoot and I. R. Adams (2015). "Oocyte development, meiosis and aneuploidy." Semin Cell Dev Biol 45: 68-76.

169

MacNicol, M. C., C. E. Cragle, K. Arumugam, B. Fosso, G. Pesole and A. M. MacNicol (2015). "Functional Integration of mRNA Translational Control Programs." Biomolecules 5(3): 1580- 1599. Madgwick, S. and K. T. Jones (2007). "How eggs arrest at metaphase II: MPF stabilisation plus APC/C inhibition equals Cytostatic Factor." Cell Div 2: 4. Magnusdottir, E., S. Dietmann, K. Murakami, U. Gunesdogan, F. Tang, S. Bao, E. Diamanti, K. Lao, B. Gottgens and M. Azim Surani (2013). "A tripartite transcription factor network regulates primordial germ cell specification in mice." Nat Cell Biol 15(8): 905-915. Mahakali Zama, A., F. P. Hudson, 3rd and M. A. Bedell (2005). "Analysis of hypomorphic KitlSl mutants suggests different requirements for KITL in proliferation and migration of mouse primordial germ cells." Biol Reprod 73(4): 639-647. Mahrous, E., Q. Yang and H. J. Clarke (2012). "Regulation of mitochondrial DNA accumulation during oocyte growth and meiotic maturation in the mouse." Reproduction 144(2): 177-185. Mak, W., C. Fang, T. Holden, M. B. Dratver and H. Lin (2016). "An Important Role of Pumilio 1 in Regulating the Development of the Mammalian Female Germline." Biol Reprod 94(6): 134. Mamo, S., F. Carter, P. Lonergan, C. L. Leal, A. Al Naib, P. McGettigan, J. P. Mehta, A. C. Evans and T. Fair (2011). "Sequential analysis of global gene expression profiles in immature and in vitro matured bovine oocytes: potential molecular markers of oocyte maturation." BMC Genomics 12: 151. Mandel, C. R., S. Kaneko, H. Zhang, D. Gebauer, V. Vethantham, J. L. Manley and L. Tong (2006). "Polyadenylation factor CPSF-73 is the pre-mRNA 3'-end-processing endonuclease." Nature 444(7121): 953-956. Marintchev, A. (2013). "Roles of helicases in translation initiation: a mechanistic view." Biochim Biophys Acta 1829(8): 799-809. Marintchev, A., K. A. Edmonds, B. Marintcheva, E. Hendrickson, M. Oberer, C. Suzuki, B. Herdy, N. Sonenberg and G. Wagner (2009). "Topology and regulation of the human eIF4A/4G/4H helicase complex in translation initiation." Cell 136(3): 447-460. Markmiller, S., S. Soltanieh, K. L. Server, R. Mak, W. Jin, M. Y. Fang, E. C. Luo, F. Krach, D. Yang, A. Sen, A. Fulzele, J. M. Wozniak, D. J. Gonzalez, M. W. Kankel, F. B. Gao, E. J. Bennett, E. Lecuyer and G. W. Yeo (2018). "Context-Dependent and Disease-Specific Diversity in Protein Interactions within Stress Granules." Cell 172(3): 590-604 e513. Marnef, A., J. Sommerville and M. R. Ladomery (2009). "RAP55: insights into an evolutionarily conserved protein family." Int J Biochem Cell Biol 41(5): 977-981. Mascarenhas, M. N., S. R. Flaxman, T. Boerma, S. Vanderpoel and G. A. Stevens (2012). "National, regional, and global trends in infertility prevalence since 1990: a systematic analysis of 277 health surveys." PLoS Med 9(12): e1001356. Masui, Y. and Markert, C.L. (1971) "Cytoplasmic control of nuclear behavior during meiotic maturation of frog oocytes." J Exp Zool 177(2):129–145

170

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):1–17 Matsui, Y., K. Zsebo and B. L. Hogan (1992). "Derivation of pluripotential embryonic stem cells from murine primordial germ cells in culture." Cell 70(5): 841-847. Matsuki, H., M. Takahashi, M. Higuchi, G. N. Makokha, M. Oie and M. Fujii (2013). "Both G3BP1 and G3BP2 contribute to stress granule formation." Genes Cells 18(2): 135-146. Mattson, B. A. and D. F. Albertini (1990). "Oogenesis: chromatin and microtubule dynamics during meiotic prophase." Mol Reprod Dev 25(4): 374-383. Mauger, D. M., N. A. Siegfried and K. M. Weeks (2013). "The genetic code as expressed through relationships between mRNA structure and protein function." FEBS Lett 587(8): 1180-1188. Mazroui, R., R. Sukarieh, M. E. Bordeleau, R. J. Kaufman, P. Northcote, J. Tanaka, I. Gallouzi and J. Pelletier (2006). "Inhibition of ribosome recruitment induces stress granule formation independently of eukaryotic initiation factor 2alpha phosphorylation." Mol Biol Cell 17(10): 4212- 4219. Medvedev, S., H. Pan and R. M. Schultz (2011). "Absence of MSY2 in mouse oocytes perturbs oocyte growth and maturation, RNA stability, and the transcriptome." Biol Reprod 85(3): 575- 583. Medvedev, S., J. Yang, N. B. Hecht and R. M. Schultz (2008). "CDC2A (CDK1)-mediated phosphorylation of MSY2 triggers maternal mRNA degradation during mouse oocyte maturation." Dev Biol 321(1): 205-215. Meijer, H. A. and A. A. Thomas (2002). "Control of eukaryotic protein synthesis by upstream open reading frames in the 5'-untranslated region of an mRNA." Biochem J 367(Pt 1): 1-11. Melo, E. O., R. Dhalia, C. Martins de Sa, N. Standart and O. P. de Melo Neto (2003). "Identification of a C-terminal poly(A)-binding protein (PABP)-PABP interaction domain: role in cooperative binding to poly (A) and efficient cap distal translational repression." J Biol Chem 278(47): 46357-46368. Menke, D. B., J. Koubova and D. C. Page (2003). "Sexual differentiation of germ cells in XX mouse gonads occurs in an anterior-to-posterior wave." Dev Biol 262(2): 303-312. Merrick, W. C. (2004). "Cap-dependent and cap-independent translation in eukaryotic systems." Gene 332: 1-11. Messerschmidt, D. M., B. B. Knowles and D. Solter (2014). "DNA methylation dynamics during epigenetic reprogramming in the germline and preimplantation embryos." Genes Dev 28(8): 812- 828. Mikkelsen, A. L. and S. Lindenberg (2001). "Morphology of in-vitro matured oocytes: impact on fertility potential and embryo quality." Hum Reprod 16(8): 1714-1718. Millard, S. S., A. Vidal, M. Markus and A. Koff (2000). "A U-rich element in the 5' untranslated region is necessary for the translation of p27 mRNA." Mol Cell Biol 20(16): 5947-5959.

171

Miller, J. E. and J. C. Reese (2012). "Ccr4-Not complex: the control freak of eukaryotic cells." Crit Rev Biochem Mol Biol 47(4): 315-333. Minich, W. B., M. L. Balasta, D. J. Goss and R. E. Rhoads (1994). "Chromatographic resolution of in vivo phosphorylated and nonphosphorylated eukaryotic translation initiation factor eIF-4E: increased cap affinity of the phosphorylated form." Proc Natl Acad Sci U S A 91(16): 7668-7672. Mishima, Y., A. Fukao, T. Kishimoto, H. Sakamoto, T. Fujiwara and K. Inoue (2012). "Translational inhibition by deadenylation-independent mechanisms is central to microRNA- mediated silencing in zebrafish." Proc Natl Acad Sci U S A 109(4): 1104-1109. Mitra, J. and R. M. Schultz (1996). "Regulation of the acquisition of meiotic competence in the mouse: changes in the subcellular localization of cdc2, cyclin B1, cdc25C and wee1, and in the concentration of these proteins and their transcripts." J Cell Sci 109 ( Pt 9): 2407-2415. Mittal, S., A. Aslam, R. Doidge, R. Medica and G. S. Winkler (2011). "The Ccr4a (CNOT6) and Ccr4b (CNOT6L) deadenylase subunits of the human Ccr4-Not complex contribute to the prevention of cell death and senescence." Mol Biol Cell 22(6): 748-758. Mollet, S., N. Cougot, A. Wilczynska, F. Dautry, M. Kress, E. Bertrand and D. Weil (2008). "Translationally repressed mRNA transiently cycles through stress granules during stress." Mol Biol Cell 19(10): 4469-4479. Molyneaux, K. A., J. Stallock, K. Schaible and C. Wylie (2001). "Time-lapse analysis of living mouse germ cell migration." Dev Biol 240(2): 488-498. Monniaux, D., F. Clement, R. Dalbies-Tran, A. Estienne, S. Fabre, C. Mansanet and P. Monget (2014). "The ovarian reserve of primordial follicles and the dynamic reserve of antral growing follicles: what is the link?" Biol Reprod 90(4): 85. Moore, G. P. and S. Lintern-Moore (1974). "A correlation between growth and RNA synthesis in the mouse oocyte." J Reprod Fertil 39(1): 163-166. Moore, G. P. and S. Lintern-Moore (1978). "Transcription of the mouse oocyte genome." Biol Reprod 18(5): 865-870. Moore, G. P., S. Lintern-Moore, H. Peters and M. Faber (1974). "RNA synthesis in the mouse oocyte." J Cell Biol 60(2): 416-422. Morita, Y. and J. L. Tilly (1999). "Segregation of retinoic acid effects on fetal ovarian germ cell mitosis versus apoptosis by requirement for new macromolecular synthesis." Endocrinology 140(6): 2696-2703. Mork, L., D. M. Maatouk, J. A. McMahon, J. J. Guo, P. Zhang, A. P. McMahon and B. Capel (2012). "Temporal differences in granulosa cell specification in the ovary reflect distinct follicle fates in mice." Biol Reprod 86(2): 37. Morris, D. R. and A. P. Geballe (2000). "Upstream open reading frames as regulators of mRNA translation." Mol Cell Biol 20(23): 8635-8642. Munroe, D. and A. Jacobson (1990). "mRNA poly(A) tail, a 3' enhancer of translational initiation." Mol Cell Biol 10(7): 3441-3455.

172

Murai, S., P. Stein, M. G. Buffone, S. Yamashita and R. M. Schultz (2010). "Recruitment of Orc6l, a dormant maternal mRNA in mouse oocytes, is essential for DNA replication in 1-cell embryos." Dev Biol 341(1): 205-212. Murakami, K., U. Gunesdogan, J. J. Zylicz, W. W. C. Tang, R. Sengupta, T. Kobayashi, S. Kim, R. Butler, S. Dietmann and M. A. Surani (2016). "NANOG alone induces germ cells in primed epiblast in vitro by activation of enhancers." Nature 529(7586): 403-407. Nakanishi, T., H. Kubota, N. Ishibashi, S. Kumagai, H. Watanabe, M. Yamashita, S.-i. Kashiwabara, K. Miyado and T. Baba (2006). "Possible role of mouse poly(A) polymerase mGLD- 2 during oocyte maturation." Developmental Biology 289(1): 115-126. Narayanaswamy, R., M. Levy, M. Tsechansky, G. M. Stovall, J. D. O'Connell, J. Mirrielees, A. D. Ellington and E. M. Marcotte (2009). "Widespread reorganization of metabolic enzymes into reversible assemblies upon nutrient starvation." Proc Natl Acad Sci U S A 106(25): 10147-10152. Nilsson, E. E. and M. K. Skinner (2003). "Bone morphogenetic protein-4 acts as an ovarian follicle survival factor and promotes primordial follicle development." Biol Reprod 69(4): 1265-1272. Noble, S. L., B. L. Allen, L. K. Goh, K. Nordick and T. C. Evans (2008). "Maternal mRNAs are regulated by diverse P body-related mRNP granules during early Caenorhabditis elegans development." J Cell Biol 182(3): 559-572. Nogueira, D., J. C. Sadeu and J. Montagut (2012). "In vitro oocyte maturation: current status." Semin Reprod Med 30(3): 199-213. Norbury, C. J. (2013). "Cytoplasmic RNA: a case of the tail wagging the dog." Nat Rev Mol Cell Biol 14(10): 643-653. O'Shea, L. C., J. Mehta, P. Lonergan, C. Hensey and T. Fair (2012). "Developmental competence in oocytes and cumulus cells: candidate genes and networks." Syst Biol Reprod Med 58(2): 88- 101. Ogami, K., N. Hosoda, Y. Funakoshi and S. Hoshino (2014). "Antiproliferative protein Tob directly regulates c-myc proto-oncogene expression through cytoplasmic polyadenylation element-binding protein CPEB." Oncogene 33(1): 55-64. Ohinata, Y., H. Ohta, M. Shigeta, K. Yamanaka, T. Wakayama and M. Saitou (2009). "A signaling principle for the specification of the germ cell lineage in mice." Cell 137(3): 571-584. Ohinata, Y., B. Payer, D. O'Carroll, K. Ancelin, Y. Ono, M. Sano, S. C. Barton, T. Obukhanych, M. Nussenzweig, A. Tarakhovsky, M. Saitou and M. A. Surani (2005). "Blimp1 is a critical determinant of the germ cell lineage in mice." Nature 436(7048): 207-213. Ohn, T., N. Kedersha, T. Hickman, S. Tisdale and P. Anderson (2008). "A functional RNAi screen links O-GlcNAc modification of ribosomal proteins to stress granule and processing body assembly." Nat Cell Biol 10(10): 1224-1231. Okamura, D., Y. Tokitake, H. Niwa and Y. Matsui (2008). "Requirement of Oct3/4 function for germ cell specification." Dev Biol 317(2): 576-584.

173

Ottolenghi, C., E. Pelosi, J. Tran, M. Colombino, E. Douglass, T. Nedorezov, A. Cao, A. Forabosco and D. Schlessinger (2007). "Loss of Wnt4 and Foxl2 leads to female-to-male sex reversal extending to germ cells." Hum Mol Genet 16(23): 2795-2804. Padmanabhan, K. and J. D. Richter (2006). "Regulated Pumilio-2 binding controls RINGO/Spy mRNA translation and CPEB activation." Genes Dev 20(2): 199-209. Palmer, S. J. and P. S. Burgoyne (1991). "In situ analysis of fetal, prepuberal and adult XX----XY chimaeric mouse testes: Sertoli cells are predominantly, but not exclusively, XY." Development 112(1): 265-268. Pan, H., J. O'Brien M, K. Wigglesworth, J. J. Eppig and R. M. Schultz (2005). "Transcript profiling during mouse oocyte development and the effect of gonadotropin priming and development in vitro." Dev Biol 286(2): 493-506. Pelletier, J. and N. Sonenberg (1988). "Internal initiation of translation of eukaryotic mRNA directed by a sequence derived from poliovirus RNA." Nature 334(6180): 320-325. Pelosi, E., S. Omari, M. Michel, J. Ding, T. Amano, A. Forabosco, D. Schlessinger and C. Ottolenghi (2013). "Constitutively active Foxo3 in oocytes preserves ovarian reserve in mice." Nat Commun 4: 1843. Pepling, M. E., M. de Cuevas and A. C. Spradling (1999). "Germline cysts: a conserved phase of germ cell development?" Trends Cell Biol 9(7): 257-262. Pepling, M. E. and A. C. Spradling (2001). "Mouse ovarian germ cell cysts undergo programmed breakdown to form primordial follicles." Dev Biol 234(2): 339-351. Peter, M., E. Heitlinger, M. Haner, U. Aebi and E. A. Nigg (1991). "Disassembly of in vitro formed lamin head-to-tail polymers by CDC2 kinase." EMBO J 10(6): 1535-1544. Peter, M., C. Le Peuch, J. C. Labbe, A. N. Meyer, D. J. Donoghue and M. Doree (2002). "Initial activation of cyclin-B1-cdc2 kinase requires phosphorylation of cyclin B1." EMBO Rep 3(6): 551- 556. Peter, M., J. Nakagawa, M. Doree, J. C. Labbe and E. A. Nigg (1990). "In vitro disassembly of the nuclear lamina and M phase-specific phosphorylation of lamins by cdc2 kinase." Cell 61(4): 591- 602. Pincus, G. and E. V. Enzmann (1935). "The Comparative Behavior of Mammalian Eggs in Vivo and in Vitro : I. The Activation of Ovarian Eggs." J Exp Med 62(5): 665-675. Pines, J. and T. Hunter (1989). "Isolation of a human cyclin cDNA: evidence for cyclin mRNA and protein regulation in the cell cycle and for interaction with p34cdc2." Cell 58(5): 833-846. Pique, M., J. M. Lopez, S. Foissac, R. Guigo and R. Mendez (2008). "A combinatorial code for CPE-mediated translational control." Cell 132(3): 434-448. Polanski, Z., E. Ledan, S. Brunet, S. Louvet, M. H. Verlhac, J. Z. Kubiak and B. Maro (1998). "Cyclin synthesis controls the progression of meiotic maturation in mouse oocytes." Development 125(24): 4989-4997.

174

Pop, C., S. Rouskin, N. T. Ingolia, L. Han, E. M. Phizicky, J. S. Weissman and D. Koller (2014). "Causal signals between codon bias, mRNA structure, and the efficiency of translation and elongation." Mol Syst Biol 10: 770. Prasad, C. K., M. Mahadevan, M. C. MacNicol and A. M. MacNicol (2008). "Mos 3prime UTR regulatory differences underlie species-specific temporal patterns of Mos mRNA cytoplasmic polyadenylation and translational recruitment during oocyte maturation." Molecular Reproduction and Development 75(8): 1258-1268. Prokopuk, L., J. M. Stringer, K. Hogg, K. D. Elgass and P. S. Western (2017). "PRC2 is required for extensive reorganization of H3K27me3 during epigenetic reprogramming in mouse fetal germ cells." Epigenetics Chromatin 10: 7. Protter, D. S. W., B. S. Rao, B. Van Treeck, Y. Lin, L. Mizoue, M. K. Rosen and R. Parker (2018). "Intrinsically Disordered Regions Can Contribute Promiscuous Interactions to RNP Granule Assembly." Cell Rep 22(6): 1401-1412. Provost, P., D. Dishart, J. Doucet, D. Frendewey, B. Samuelsson and O. Radmark (2002). "Ribonuclease activity and RNA binding of recombinant human Dicer." EMBO J 21(21): 5864- 5874. Qin, D. and K. Fredrick (2013). "Analysis of polysomes from bacteria." Methods Enzymol 530: 159-172. Quax, T. E., N. J. Claassens, D. Soll and J. van der Oost (2015). "Codon Bias as a Means to Fine- Tune Gene Expression." Mol Cell 59(2): 149-161. Racki, W. J. and J. D. Richter (2006). "CPEB controls oocyte growth and follicle development in the mouse." Development 133(22): 4527-4537. Rattani, A., P. K. Vinod, J. Godwin, K. Tachibana-Konwalski, M. Wolna, M. Malumbres, B. Novak and K. Nasmyth (2014). "Dependency of the spindle assembly checkpoint on Cdk1 renders the anaphase transition irreversible." Curr Biol 24(6): 630-637. Reader, K. L., J. L. Stanton and J. L. Juengel (2017). "The Role of Oocyte Organelles in Determining Developmental Competence." Biology (Basel) 6(3). Reddy, P., L. Liu, D. Adhikari, K. Jagarlamudi, S. Rajareddy, Y. Shen, C. Du, W. Tang, T. Hamalainen, S. L. Peng, Z. J. Lan, A. J. Cooney, I. Huhtaniemi and K. Liu (2008). "Oocyte- specific deletion of Pten causes premature activation of the primordial follicle pool." Science 319(5863): 611-613. Reijns, M. A., R. D. Alexander, M. P. Spiller and J. D. Beggs (2008). "A role for Q/N-rich aggregation-prone regions in P-body localization." J Cell Sci 121(Pt 15): 2463-2472. Reis, A., H. Y. Chang, M. Levasseur and K. T. Jones (2006). "APCcdh1 activity in mouse oocytes prevents entry into the first meiotic division." Nat Cell Biol 8(5): 539-540. Reyes, J. M. and P. J. Ross (2016). "Cytoplasmic polyadenylation in mammalian oocyte maturation." Wiley Interdiscip Rev RNA 7(1): 71-89.

175

Rhoads, R. E., B. Joshi and W. B. Minich (1994). "Participation of initiation factors in the recruitment of mRNA to ribosomes." Biochimie 76(9): 831-838. Richard, F. J., A. Tsafriri and M. Conti (2001). "Role of phosphodiesterase type 3A in rat oocyte maturation." Biol Reprod 65(5): 1444-1451. Richardson, B. E. and R. Lehmann (2010). "Mechanisms guiding primordial germ cell migration: strategies from different organisms." Nat Rev Mol Cell Biol 11(1): 37-49. Riechmann, V. and A. Ephrussi (2001). "Axis formation during Drosophila oogenesis." Curr Opin Genet Dev 11(4): 374-383. Robinson, J. W., M. Zhang, L. C. Shuhaibar, R. P. Norris, A. Geerts, F. Wunder, J. J. Eppig, L. R. Potter and L. A. Jaffe (2012). "Luteinizing hormone reduces the activity of the NPR2 guanylyl cyclase in mouse ovarian follicles, contributing to the cyclic GMP decrease that promotes resumption of meiosis in oocytes." Dev Biol 366(2): 308-316. Rosario, R., P. Filis, V. Tessyman, H. Kinnell, A. J. Childs, N. K. Gray and R. A. Anderson (2016). "FMRP Associates with Cytoplasmic Granules at the Onset of Meiosis in the Human Oocyte." PLoS One 11(10): e0163987. Rosselot, C., A. L. Kierszenbaum, E. Rivkin and L. L. Tres (2003). "Chronological gene expression of ADAMs during testicular development: prespermatogonia (gonocytes) express fertilin beta (ADAM2)." Dev Dyn 227(3): 458-467. Rudolph, J. (2007). "Cdc25 phosphatases: structure, specificity, and mechanism." Biochemistry 46(12): 3595-3604. Runyan, C., K. Schaible, K. Molyneaux, Z. Wang, L. Levin and C. Wylie (2006). "Steel factor controls midline cell death of primordial germ cells and is essential for their normal proliferation and migration." Development 133(24): 4861-4869. Russell, S. J., L. Stalker and J. LaMarre (2017). "PIWIs, piRNAs and Retrotransposons: Complex battles during reprogramming in gametes and early embryos." Reprod Domest Anim 52 Suppl 4: 28-38. Saatcioglu, H. D., I. Cuevas and D. H. Castrillon (2016). "Control of Oocyte Reawakening by Kit." PLoS Genet 12(8): e1006215. Sakkas, D., M. Ramalingam, N. Garrido and C. L. Barratt (2015). "Sperm selection in natural conception: what can we learn from Mother Nature to improve assisted reproduction outcomes?" Hum Reprod Update 21(6): 711-726. Sama, R. R., C. Fallini, R. Gatto, J. E. McKeon, Y. Song, M. S. Rotunno, S. Penaranda, I. Abdurakhmanov, J. E. Landers, G. Morfini, S. T. Brady and D. A. Bosco (2017). "ALS-linked FUS exerts a gain of toxic function involving aberrant p38 MAPK activation." Sci Rep 7(1): 115. Sánchez, F. and J. Smitz (2012). "Molecular control of oogenesis." Biochim Biophys Acta 1822(12): 1896-1912. Schisa, J. A. (2012). "New insights into the regulation of RNP granule assembly in oocytes." Int Rev Cell Mol Biol 295: 233-289.

176

Schmidt, D., C. E. Ovitt, K. Anlag, S. Fehsenfeld, L. Gredsted, A. C. Treier and M. Treier (2004). "The murine winged-helix transcription factor Foxl2 is required for granulosa cell differentiation and ovary maintenance." Development 131(4): 933-942. Scholer, H. R., R. Balling, A. K. Hatzopoulos, N. Suzuki and P. Gruss (1989). "Octamer binding proteins confer transcriptional activity in early mouse embryogenesis." EMBO J 8(9): 2551-2557. Scholer, H. R., G. R. Dressler, R. Balling, H. Rohdewohld and P. Gruss (1990). "Oct-4: a germline- specific transcription factor mapping to the mouse t-complex." EMBO J 9(7): 2185-2195. Schonemann, L., U. Kuhn, G. Martin, P. Schafer, A. R. Gruber, W. Keller, M. Zavolan and E. Wahle (2014). "Reconstitution of CPSF active in polyadenylation: recognition of the polyadenylation signal by WDR33." Genes Dev 28(21): 2381-2393. Schuller, A. P. and R. Green (2017). "The ABC(E1)s of Ribosome Recycling and Reinitiation." Mol Cell 66(5): 578-580. Seisenberger, S., S. Andrews, F. Krueger, J. Arand, J. Walter, F. Santos, C. Popp, B. Thienpont, W. Dean and W. Reik (2012). "The dynamics of genome-wide DNA methylation reprogramming in mouse primordial germ cells." Mol Cell 48(6): 849-862. Sekido, R. and R. Lovell-Badge (2009). "Sex determination and SRY: down to a wink and a nudge?" Trends Genet 25(1): 19-29. Sha, Q. Q., X. X. Dai, Y. Dang, F. Tang, J. Liu, Y. L. Zhang and H. Y. Fan (2017). "A MAPK cascade couples maternal mRNA translation and degradation to meiotic cell cycle progression in mouse oocytes." Development 144(3): 452-463. Shabalina, S. A. and E. V. Koonin (2008). "Origins and evolution of eukaryotic RNA interference." Trends Ecol Evol 23(10): 578-587. Sharif, J., M. Muto, S. Takebayashi, I. Suetake, A. Iwamatsu, T. A. Endo, J. Shinga, Y. Mizutani- Koseki, T. Toyoda, K. Okamura, S. Tajima, K. Mitsuya, M. Okano and H. Koseki (2007). "The SRA protein Np95 mediates epigenetic inheritance by recruiting Dnmt1 to methylated DNA." Nature 450(7171): 908-912. Shaw, G. and R. Kamen (1986). "A conserved AU sequence from the 3' untranslated region of GM-CSF mRNA mediates selective mRNA degradation." Cell 46(5): 659-667. Sheets, M. D., M. Wu and M. Wickens (1995). "Polyadenylation of c-mos mRNA as a control point in Xenopus meiotic maturation." Nature 374(6522): 511-516. Sheth, U. and R. Parker (2003). "Decapping and decay of messenger RNA occur in cytoplasmic processing bodies." Science 300(5620): 805-808. Shi, Y., K. M. Vattem, R. Sood, J. An, J. Liang, L. Stramm and R. C. Wek (1998). "Identification and characterization of pancreatic eukaryotic initiation factor 2 alpha-subunit kinase, PEK, involved in translational control." Mol Cell Biol 18(12): 7499-7509. Shim, S. Y., I. Perez de Castro, G. Neumayer, J. Wang, S. K. Park, K. Sanada and M. D. Nguyen (2015). "Phosphorylation of targeting protein for Xenopus kinesin-like protein 2 (TPX2) at threonine 72 in spindle assembly." J Biol Chem 290(14): 9122-9134.

177

Shirai, Y. T., T. Suzuki, M. Morita, A. Takahashi and T. Yamamoto (2014). "Multifunctional roles of the mammalian CCR4-NOT complex in physiological phenomena." Front Genet 5: 286. Shoemaker, C. J. and R. Green (2011). "Kinetic analysis reveals the ordered coupling of translation termination and ribosome recycling in yeast." Proc Natl Acad Sci U S A 108(51): E1392-1398. Shrestha, D., X. La and H. L. Feng (2015). "Comparison of different stimulation protocols used in in vitro fertilization: a review." Ann Transl Med 3(10): 137. Shyu, A. B., J. G. Belasco and M. E. Greenberg (1991). "Two distinct destabilizing elements in the c-fos message trigger deadenylation as a first step in rapid mRNA decay." Genes Dev 5(2): 221-231. Shyu, A. B., M. F. Wilkinson and A. van Hoof (2008). "Messenger RNA regulation: to translate or to degrade." EMBO J 27(3): 471-481. Slater, D. W., I. Slater and D. Gillespie (1972). "Post-fertilization synthesis of polyadenylic acid in sea urchin embryos." Nature 240(5380): 333-337. Slezak-Prochazka, I., S. Durmus, B. J. Kroesen and A. van den Berg (2010). "MicroRNAs, macrocontrol: regulation of miRNA processing." RNA 16(6): 1087-1095. Sousa Martins, J. P., X. Liu, A. Oke, R. Arora, F. Franciosi, S. Viville, D. J. Laird, J. C. Fung and M. Conti (2016). "DAZL and CPEB1 regulate mRNA translation synergistically during oocyte maturation." J Cell Sci 129(6): 1271-1282. Stefkova, K., J. Prochazkova and J. Pachernik (2015). "Alkaline phosphatase in stem cells." Stem Cells Int 2015: 628368. Su, Y. Q., K. Sugiura, K. Wigglesworth, M. J. O'Brien, J. P. Affourtit, S. A. Pangas, M. M. Matzuk and J. J. Eppig (2008). "Oocyte regulation of metabolic cooperativity between mouse cumulus cells and oocytes: BMP15 and GDF9 control cholesterol biosynthesis in cumulus cells." Development 135(1): 111-121. Su, Y. Q., K. Sugiura, Y. Woo, K. Wigglesworth, S. Kamdar, J. Affourtit and J. J. Eppig (2007). "Selective degradation of transcripts during meiotic maturation of mouse oocytes." Dev Biol 302(1): 104-117. Sugimoto, M. and K. Abe (2007). "X chromosome reactivation initiates in nascent primordial germ cells in mice." PLoS Genet 3(7): e116. Sugiura, K., F. L. Pendola and J. J. Eppig (2005). "Oocyte control of metabolic cooperativity between oocytes and companion granulosa cells: energy metabolism." Dev Biol 279(1): 20-30. Suh, N., L. Baehner, F. Moltzahn, C. Melton, A. Shenoy, J. Chen and R. Blelloch (2010). "MicroRNA function is globally suppressed in mouse oocytes and early embryos." Curr Biol 20(3): 271-277. Sun, M. J., S. Zhu, Y. W. Li, J. Lin, S. Gong, G. Z. Jiao, F. Chen and J. H. Tan (2016). "An essential role for the intra-oocyte MAPK activity in the NSN-to-SN transition of germinal vesicle chromatin configuration in porcine oocytes." Sci Rep 6: 23555.

178

Susor, A., D. Jansova, M. Anger and M. Kubelka (2016). "Translation in the mammalian oocyte in space and time." Cell Tissue Res 363(1): 69-84. Susor, A., D. Jansova, R. Cerna, A. Danylevska, M. Anger, T. Toralova, R. Malik, J. Supolikova, M. S. Cook, J. S. Oh and M. Kubelka (2015). "Temporal and spatial regulation of translation in the mammalian oocyte via the mTOR-eIF4F pathway." Nat Comm 6: 6078. Swartz, W. J. (1982). "Acid and alkaline phosphatase activity in migrating primordial germ cells of the early chick embryo." Anat Rec 202(3): 379-385. Swetloff, A., B. Conne, J. Huarte, J. L. Pitetti, S. Nef and J. D. Vassalli (2009). "Dcp1-bodies in mouse oocytes." Mol Biol Cell 20(23): 4951-4961. Szybek, K. (1972). "In-vitro maturation of oocytes from sexually immature mice." J Endocrinol 54(3): 527-528. Taketo, T. (2015). "The role of sex chromosomes in mammalian germ cell differentiation: can the germ cells carrying X and Y chromosomes differentiate into fertile oocytes?" Asian J Androl 17(3): 360-366. Takeuchi, Y., K. Molyneaux, C. Runyan, K. Schaible and C. Wylie (2005). "The roles of FGF signaling in germ cell migration in the mouse." Development 132(24): 5399-5409. Tam, P. P. and M. H. Snow (1981). "Proliferation and migration of primordial germ cells during compensatory growth in mouse embryos." J Embryol Exp Morphol 64: 133-147. Tamaru, H. and E. U. Selker (2001). "A histone H3 methyltransferase controls DNA methylation in Neurospora crassa." Nature 414(6861): 277-283. Tamaru, H., X. Zhang, D. McMillen, P. B. Singh, J. Nakayama, S. I. Grewal, C. D. Allis, X. Cheng and E. U. Selker (2003). "Trimethylated lysine 9 of histone H3 is a mark for DNA methylation in Neurospora crassa." Nat Genet 34(1): 75-79. Tay, J. and J. D. Richter (2001). "Germ cell differentiation and synaptonemal complex formation are disrupted in CPEB knockout mice." Dev Cell 1(2): 201-213. Taylor, J. P., R. H. Brown, Jr. and D. W. Cleveland (2016). "Decoding ALS: from genes to mechanism." Nature 539(7628): 197-206. Trapphoff, T., M. Heiligentag, D. Dankert, H. Demond, D. Deutsch, T. Frohlich, G. J. Arnold, R. Grummer, B. Horsthemke and U. Eichenlaub-Ritter (2016). "Postovulatory aging affects dynamics of mRNA, expression and localization of maternal effect proteins, spindle integrity and pericentromeric proteins in mouse oocytes." Hum Reprod 31(1): 133-149. Treen, N., T. Heist, W. Wang and M. Levine (2018). "Depletion of Maternal Cyclin B3 Contributes to Zygotic Genome Activation in the Ciona Embryo." Curr Biol 28(7): 1150-1156 e1154. Treen, N., T. Heist, W. Wang and M. Levine (2018). "Depletion of Maternal Cyclin B3 Contributes to Zygotic Genome Activation in the Ciona Embryo." Curr Biol 28(8): 1330-1331. Tremblay, K. D., N. R. Dunn and E. J. Robertson (2001). "Mouse embryos lacking Smad1 signals display defects in extra-embryonic tissues and germ cell formation." Development 128(18): 3609- 3621.

179

Trombly, D. J., T. K. Woodruff and K. E. Mayo (2009). "Suppression of Notch signaling in the neonatal mouse ovary decreases primordial follicle formation." Endocrinology 150(2): 1014-1024. Tucker, M., R. R. Staples, M. A. Valencia-Sanchez, D. Muhlrad and R. Parker (2002). "Ccr4p is the catalytic subunit of a Ccr4p/Pop2p/Notp mRNA deadenylase complex in Saccharomyces cerevisiae." EMBO J 21(6): 1427-1436. Updike, D. and S. Strome (2010). "P granule assembly and function in Caenorhabditis elegans germ cells." J Androl 31(1): 53-60. Uzbekova, S., Y. Arlot-Bonnemains, J. Dupont, R. Dalbies-Tran, P. Papillier, S. Pennetier, A. Thelie, C. Perreau, P. Mermillod, C. Prigent and R. Uzbekov (2008). "Spatio-temporal expression patterns of aurora kinases a, B, and C and cytoplasmic polyadenylation-element-binding protein in bovine oocytes during meiotic maturation." Biol Reprod 78(2): 218-233. Vaccari, S., K. Horner, L. M. Mehlmann and M. Conti (2008). "Generation of mouse oocytes defective in cAMP synthesis and degradation: endogenous cyclic AMP is essential for meiotic arrest." Dev Biol 316(1): 124-134. Vaccari, S., J. L. Weeks, 2nd, M. Hsieh, F. S. Menniti and M. Conti (2009). "Cyclic GMP signaling is involved in the luteinizing hormone-dependent meiotic maturation of mouse oocytes." Biol Reprod 81(3): 595-604. Vainio, S., M. Heikkila, A. Kispert, N. Chin and A. P. McMahon (1999). "Female development in mammals is regulated by Wnt-4 signalling." Nature 397(6718): 405-409. Van Blerkom, J., P. W. Davis and J. Lee (1995). "ATP content of human oocytes and developmental potential and outcome after in-vitro fertilization and embryo transfer." Hum Reprod 10(2): 415-424. Van Etten, J., T. L. Schagat, J. Hrit, C. A. Weidmann, J. Brumbaugh, J. J. Coon and A. C. Goldstrohm (2012). "Human Pumilio proteins recruit multiple deadenylases to efficiently repress messenger RNAs." J Biol Chem 287(43): 36370-36383. Vanorny, D. A., R. D. Prasasya, A. J. Chalpe, S. M. Kilen and K. E. Mayo (2014). "Notch signaling regulates ovarian follicle formation and coordinates follicular growth." Mol Endocrinol 28(4): 499-511. Vazquez-Pianzola, P., H. Urlaub and B. Suter (2011). "Pabp binds to the osk 3'UTR and specifically contributes to osk mRNA stability and oocyte accumulation." Dev Biol 357(2): 404- 418. Verlhac, M. H., J. Z. Kubiak, M. Weber, G. Geraud, W. H. Colledge, M. J. Evans and B. Maro (1996). "Mos is required for MAP kinase activation and is involved in microtubule organization during meiotic maturation in the mouse." Development 122(3): 815-822. Vieux, K. F. and H. J. Clarke (2018). "CNOT6 regulates a novel pattern of mRNA deadenylation during oocyte meiotic maturation." Sci Rep 8(1): 6812. Wade, J. J., V. MacLachlan and G. Kovacs (2015). "The success rate of IVF has significantly improved over the last decade." Aust N Z J Obstet Gynaecol 55(5): 473-476.

180

Wahle, E. and G. S. Winkler (2013). "RNA decay machines: deadenylation by the Ccr4-not and Pan2-Pan3 complexes." Biochim Biophys Acta 1829(6-7): 561-570. Wai, T., A. Ao, X. Zhang, D. Cyr, D. Dufort and E. A. Shoubridge (2010). "The role of mitochondrial DNA copy number in mammalian fertility." Biol Reprod 83(1): 52-62. Wakayama, T., K. Hamada, M. Yamamoto, T. Suda and S. Iseki (2003). "The expression of platelet endothelial cell adhesion molecule-1 in mouse primordial germ cells during their migration and early gonadal formation." Histochem Cell Biol 119(5): 355-362. Walker, M. W., J. M. Butler, H. L. Higdon, 3rd and W. R. Boone (2013). "Temperature variations within and between incubators-a prospective, observational study." J Assist Reprod Genet 30(12): 1583-1585. Wang, J. T., J. Smith, B. C. Chen, H. Schmidt, D. Rasoloson, A. Paix, B. G. Lambrus, D. Calidas, E. Betzig and G. Seydoux (2014). "Regulation of RNA granule dynamics by phosphorylation of serine-rich, intrinsically disordered proteins in C. elegans." Elife 3: e04591. Wang, Q., M. M. Chi, T. Schedl and K. H. Moley (2012). "An intercellular pathway for glucose transport into mouse oocytes." Am J Physiol Endocrinol Metab 302(12): E1511-1518. Wang, X. Q. and J. A. Rothnagel (2004). "5'-untranslated regions with multiple upstream AUG codons can support low-level translation via leaky scanning and reinitiation." Nucleic Acids Res 32(4): 1382-1391. Wang, Y., S. Juranek, H. Li, G. Sheng, T. Tuschl and D. J. Patel (2008). "Structure of an argonaute silencing complex with a seed-containing guide DNA and target RNA duplex." Nature 456(7224): 921-926. Wang, Y., G. Sheng, S. Juranek, T. Tuschl and D. J. Patel (2008). "Structure of the guide-strand- containing argonaute silencing complex." Nature 456(7219): 209-213. Wang, Z., X. Jiao, A. Carr-Schmid and M. Kiledjian (2002). "The hDcp2 protein is a mammalian mRNA decapping enzyme." Proc Natl Acad Sci U S A 99(20): 12663-12668. Ward, G. E. and M. W. Kirschner (1990). "Identification of cell cycle-regulated phosphorylation sites on nuclear lamin C." Cell 61(4): 561-577. Warner, J. R., P. M. Knopf and A. Rich (1963). "A multiple ribosomal structure in protein synthesis." Proc Natl Acad Sci U S A 49: 122-129. Wasserman, W. J., J. D. Richter and L. D. Smith (1982). "Protein synthesis during maturation promoting factor- and progesterone-induced maturation in Xenopus oocytes." Dev Biol 89(1): 152-158. Weidner, J., C. Wang, C. Prescianotto-Baschong, A. F. Estrada and A. Spang (2014). "The polysome-associated proteins Scp160 and Bfr1 prevent P body formation under normal growth conditions." J Cell Sci 127(Pt 9): 1992-2004. Welk, J. F., A. Charlesworth, G. D. Smith and A. M. MacNicol (2001). "Identification and characterization of the gene encoding human cytoplasmic polyadenylation element binding protein." Gene 263(1-2): 113-120.

181

Wells, S. E., P. E. Hillner, R. D. Vale and A. B. Sachs (1998). "Circularization of mRNA by eukaryotic translation initiation factors." Mol Cell 2(1): 135-140. White YA, Woods DC, Takai Y, Ishihara O, Seki H, Tilly JL. (2012) "Oocyte formation by mitotically active germ cells purified from ovaries of reproductive-age women." Nat Med. 18(3):413-21. Wickramasinghe, D., K. M. Ebert and D. F. Albertini (1991). "Meiotic competence acquisition is associated with the appearance of M-phase characteristics in growing mouse oocytes." Dev Biol 143(1): 162-172. Wignall, S. M. and A. M. Villeneuve (2009). "Lateral microtubule bundles promote chromosome alignment during acentrosomal oocyte meiosis." Nat Cell Biol 11(7): 839-844. Winston, N. J. (1997). "Stability of cyclin B protein during meiotic maturation and the first mitotic cell division in mouse oocytes." Biol Cell 89(3): 211-219. Wittmann, T., M. Wilm, E. Karsenti and I. Vernos (2000). "TPX2, A novel xenopus MAP involved in spindle pole organization." J Cell Biol 149(7): 1405-1418. Wong, C. C., K. E. Loewke, N. L. Bossert, B. Behr, C. J. De Jonge, T. M. Baer and R. A. Reijo Pera (2010). "Non-invasive imaging of human embryos before embryonic genome activation predicts development to the blastocyst stage." Nat Biotechnol 28(10): 1115-1121. Wu, Q., R. Song, N. Ortogero, H. Zheng, R. Evanoff, C. L. Small, M. D. Griswold, S. H. Namekawa, H. Royo, J. M. Turner and W. Yan (2012). "The RNase III enzyme DROSHA is essential for microRNA production and spermatogenesis." J Biol Chem 287(30): 25173-25190. Xu, J. and T. Gridley (2013). "Notch2 is required in somatic cells for breakdown of ovarian germ- cell nests and formation of primordial follicles." BMC Biol 11: 13. Yabuta, Y., K. Kurimoto, Y. Ohinata, Y. Seki and M. Saitou (2006). "Gene expression dynamics during germline specification in mice identified by quantitative single-cell gene expression profiling." Biol Reprod 75(5): 705-716. Yamaguchi, S., K. Kurimoto, Y. Yabuta, H. Sasaki, N. Nakatsuji, M. Saitou and T. Tada (2009). "Conditional knockdown of Nanog induces apoptotic cell death in mouse migrating primordial germ cells." Development 136(23): 4011-4020. Yamaji, M., Y. Seki, K. Kurimoto, Y. Yabuta, M. Yuasa, M. Shigeta, K. Yamanaka, Y. Ohinata and M. Saitou (2008). "Critical function of Prdm14 for the establishment of the germ cell lineage in mice." Nat Genet 40(8): 1016-1022. Yamashita, A., T. C. Chang, Y. Yamashita, W. Zhu, Z. Zhong, C. Y. Chen and A. B. Shyu (2005). "Concerted action of poly(A) nucleases and decapping enzyme in mammalian mRNA turnover." Nat Struct Mol Biol 12(12): 1054-1063. Yan, Y. B. (2014). "Deadenylation: enzymes, regulation, and functional implications." Wiley Interdiscip Rev RNA 5(3): 421-443.

182

Yang, Q., P. Allard, M. Huang, W. Zhang and H. J. Clarke (2010). "Proteasomal activity is required to initiate and to sustain translational activation of messenger RNA encoding the stem- loop-binding protein during meiotic maturation in mice." Biol Reprod 82(1): 123-131. Yang, Y., C. R. Yang, S. J. Han, E. M. Daldello, A. Cho, J. P. S. Martins, G. Xia and M. Conti (2017). "Maternal mRNAs with distinct 3' UTRs define the temporal pattern of Ccnb1 synthesis during mouse oocyte meiotic maturation." Genes Dev 31(13): 1302-1307. Yew, N., M. L. Mellini and G. F. Vande Woude (1992). "Meiotic initiation by the mos protein in Xenopus." Nature 355(6361): 649-652. Yi, K., B. Rubinstein and R. Li (2013). "Symmetry breaking and polarity establishment during mouse oocyte maturation." Philos Trans R Soc Lond B Biol Sci 368(1629): 20130002. Ying, Y., X. Qi and G. Q. Zhao (2001). "Induction of primordial germ cells from murine epiblasts by synergistic action of BMP4 and BMP8B signaling pathways." Proc Natl Acad Sci U S A 98(14): 7858-7862. Ying, Y. and G. Q. Zhao (2001). "Cooperation of endoderm-derived BMP2 and extraembryonic ectoderm-derived BMP4 in primordial germ cell generation in the mouse." Dev Biol 232(2): 484- 492. Yoshida, S., M. Kaido and T. S. Kitajima (2015). "Inherent Instability of Correct Kinetochore- Microtubule Attachments during Meiosis I in Oocytes." Dev Cell 33(5): 589-602. Young, A. R. and M. Narita (2007). "Oncogenic HMGA2: short or small?" Genes Dev 21(9): 1005-1009. Yu, C., S. Y. Ji, Q. Q. Sha, Y. Dang, J. J. Zhou, Y. L. Zhang, Y. Liu, Z. W. Wang, B. Hu, Q. Y. Sun, S. C. Sun, F. Tang and H. Y. Fan (2016). "BTG4 is a meiotic cell cycle-coupled maternal- zygotic-transition licensing factor in oocytes." Nat Struct Mol Biol 23(5): 387-394. Yuan, S., N. Ortogero, Q. Wu, H. Zheng and W. Yan (2014). "Murine follicular development requires oocyte DICER, but not DROSHA." Biol Reprod 91(2): 39. Zhang, G., M. Hubalewska and Z. Ignatova (2009). "Transient ribosomal attenuation coordinates protein synthesis and co-translational folding." Nat Struct Mol Biol 16(3): 274-280. Zhang, H., S. Risal, N. Gorre, K. Busayavalasa, X. Li, Y. Shen, B. Bosbach, M. Brannstrom and K. Liu (2014). "Somatic cells initiate primordial follicle activation and govern the development of dormant oocytes in mice." Curr Biol 24(21): 2501-2508. Zhang, M., Y. Q. Su, K. Sugiura, G. Xia and J. J. Eppig (2010). "Granulosa cell ligand NPPC and its receptor NPR2 maintain meiotic arrest in mouse oocytes." Science 330(6002): 366-369. Zhang, M., C. X. Zhang, L. Z. Pan, S. Gong, W. Cui, H. J. Yuan, W. L. Zhang and J. H. Tan (2017). "Meiotic arrest with roscovitine and follicular fluid improves cytoplasmic maturation of porcine oocytes by promoting chromatin de-condensation and gene transcription." Sci Rep 7(1): 11574. Zhou, B. O. and J. Q. Zhou (2011). "Recent transcription-induced histone H3 lysine 4 (H3K4) methylation inhibits gene reactivation." J Biol Chem 286(40): 34770-34776.

183

Zini, A. (2018). "ICSI with testicular sperm for couples with sperm DNA damage." Int Braz J Urol 44(4): 664-666. Zuccotti, M., A. Piccinelli, P. Giorgi Rossi, S. Garagna and C. A. Redi (1995). "Chromatin organization during mouse oocyte growth." Mol Reprod Dev 41(4): 479-485.

184