The oocyte-to-embryo transition (OET) occurs in the absence of new transcription and relies on post-transcriptional gene regulation, including translational control by mRNA poly(A) tail regulation, where cytoplasmic polyadenylation activates translation and deadenylation leads to translational repression and decay. However, how the transcriptome-wide landscape of mRNA poly(A) tails shapes translation across the OET in mammals remains unknown. Here, we performed long-read RNA sequencing to uncover poly(A) tail lengths and mRNA abundance transcriptome-wide in mice across five stages of development from oocyte to embryo. Integrating these data with recently published ribosome profiling data, we demonstrate that poly(A) tail length is coupled to translational efficiency across the entire OET. We uncover an extended wave of global deadenylation during fertilization, which sets up a switch in translation control between the oocyte and embryo. In the oocyte, short-tailed maternal mRNAs that resist deadenylation in the oocyte are translationally activated, whereas large groups of mRNAs deadenylated without decay in the oocyte are later readenylated to drive translation activation in the early embryo. To investigate the mechanism, we have identified 3’UTR sequence motifs in activated and repressed transcripts and identified stage-specific candidates for RNA binding protein regulators. Our findings provide an important resource and insight into the mechanisms by which cytoplasmic polyadenylation and deadenylation dynamically shape poly(A) tail length in a stage-specific manner to orchestrate development from oocyte to embryo in mammals.

The oocyte-to-embryo transition is arguably one of the most dramatic transitions in biology. Remarkably, this entire transition takes place in the absence of de novo transcription. Transcription by RNA Polymerase II (RNAPII) is globally repressed in the growing oocyte (Global Transcriptional Silencing, GTS) and does not get fully reactivated until the 2-cell embryo stage in mice and 8-cell embryo stage in humans (Embryonic Genome Activation, EGA). Importantly, oocytes that fail to undergo GTS have significantly diminished developmental competence, with reduced rates of maturation, fertilization, and embryo development. Temporally linked to transcriptional silencing, oocytes undergo a DNA configuration change from non-surrounded nucleolus (NSN) to surrounded nucleolus (SN), where the DNA becomes condensed around the nucleolus. Like GTS, oocytes that fail to transition from NSN to SN configuration also have lower rates of fertilization and subsequent embryo development. While GTS and NSN-to-SN transition are critical for the oocyte-to-embryo transition, little is known about the mechanisms that facilitate them.

The goal of my dissertation research is to delineate the mechanisms through which the RNAPII activity is repressed in the growing oocyte as GTS is established and uncover the relationship between transcriptional silencing and global chromatin condensation that leads to the NSN-to-SN transition. In Chapter 1, I introduce and summarize the current state of knowledge on the oocyte-to-embryo transition, GTS, NSN-to-SN transition, regulation of RNAPII during GTS, and oocyte developmental competence. In Chapter 2, I present findings on the regulation of groups of genes critical for RNAPII transcription at the onset of GTS. In Chapter 3, I present evidence that manipulating abundance or activity of these key players in otherwise transcriptionally active oocytes leads to global silencing of transcription and chromatin condensation, thereby facilitating the NSN-to-SN transition, and explore the role of histone modifications in this process. In Chapter 4, I discuss these findings in the context of previous studies and suggest future research directions related to these results.

Mouse embryonic stem cells (mESCs) are the in vitro counterpart of the epiblast, possessing the ability to self-renew and differentiate into all cell types of the organism. mESC pluripotency and loss thereof is controlled by an immense regulatory network of factors, for which both post-transcriptional and epigenetic regulation play important roles. To investigate thepotential role of the cytoplasmic RNA binding protein and decay activator ZFP36L2 in this context, I generated mESCs lacking ZFP36L2 expression using the Cre-loxP system. Together, this is a powerful cell system, which allows Zfp36l2 to be knocked out at different points of cell differentiation in vitro. To facilitate the detection of KO cells, the lines I created also included the mTmG double reporter that can be used to easily monitor and sort for cells after Cre recombination. Using this system in mESCs, I discovered that knockout of ZFP36L2 potentially acts to reinforce pluripotency and 2-cell-like conversion. I also found that ZFP36L2 may also be necessary for normal differentiation, as knockout leads to aberrant expression of mesendoderm markers. Together, these data point to a potential novel role for RNA decay by ZFP36L2 as a repressor of pluripotency and regulator of germ layer specification.

Epigenetic regulation has recently been at the forefront of cancer therapeutic research. Multiple therapies that target histone modification factors such as histone lysine deacetylases and histone lysine methyltransferases have entered the market with considerable success. Histone lysine demethylases is a third class of histone modification factors that have yet to make significant stride in the market, but still show considerable promise for treating various cancer types, including leukemia, colon cancer, and breast cancer. Thus, it is of interest that ZFP36L2, an mRNA decay activator, has previously been shown to regulate multiple histone lysine demethylases, including KDM4B, KDM4C, and KDM5B. In addition, mutations in the ZFP36L2 gene has been associated with many cancer types, including acute myeloid leukemia and colorectal cancer. Expanding upon these findings, my study herein aims to investigate the relationship between ZFP36L2 and histone lysine demethylases in malignant cells through both experimental and computational analysis. Towards these goals, I generated HeLa cells with stably repressed ZFP36L2 expression using the CRISPR-Cas9 technique. These ZFP36L2- knockdown cells were used to assess changes in the expression levels of histone lysine demethylases and changes in a major cancer hallmark - cell proliferation. My experiments suggested that these ZFP36L2-knockdown cells had decreased cell proliferation, implying that ZFP36L2 is a negative regulator of cell proliferation. This does not appear to be due ZFP36L2’s previously reported ability to destabilize ARE-containing histone lysine demethylases mRNAs, as my ZFP36L2-knockdown cells had downregulated (rather than upregulated) expression of histone lysine demethylases mRNAs. To assess the effect of ZFP36L2 loss in a different type of cancer, I analyzed a previously a published RNA-seq dataset of ZFP36L2-knockout acute myeloid leukemia (AML) cells using a published algorithm to infer mRNA stability. Specifically, I asked if the loss of ZFP36L2 results in an increase of stability in the transcripts of histone lysine demethylases, as this would be consistent with a conserved mechanistic relationship between ZFP36L2 and histone lysine demethylases expression. I identified a small subset of mRNAs that are either significantly stabilized or destabilized. However, no KDM mRNAs were identified. Additional studies will be needed to validate my findings and further investigate the relationship between ZFP36L2 and regulation of KDM expression in cancer.

Androgens are a class of steroid hormones produced by men and women. Typically, men have ~10-fold higher levels of serum androgens than women, but elevated androgen production in women has undesirable effects. Hyperandrogenism is present in women with polycystic ovary syndrome (PCOS) and other conditions. Due to the correlation between hyperandrogenism and metabolic dysregulation in females, we used a hyperandrogenism mouse model to investigate when this metabolic dysregulation develops and whether gut microbiota transfer could alleviate this dysfunction. One model of hyperandrogenism in female mice is the 10-mm dihydrotestosterone (DHT) model, which has been reported to develop reproductive and metabolic dysfunction after 3 months of treatment. We were interested in understanding when this metabolic phenotype develops. Similar to a recently published study, we found that DHT treatment was sufficient to induce weight gain and insulin resistance after 8 weeks of treatment. Hyperandrogenism is associated with changes in the gut microbiome in humans and rodent models. In previous studies, modulation of the gut microbiome via fecal microbiome transplantation (FMT) or cohousing improved metabolic dysfunction in models of obesity and PCOS. We investigated if cohousing was effective in ameliorating the metabolic dysfunction induced in the DHT mouse model. We found that over 13 weeks, cohousing had a protective effect on the metabolic dysregulation observed in DHT-treated mice but did not eliminate the phenotype. These studies indicate that manipulation of gut microbiota with pre- or probiotics should be explored further as a treatment for PCOS.

Autism Spectrum Disorder is one of the most genetically heterogenous neurodevelopmental disorders known making it difficult to characterize and understand its genetic basis. Efforts to screen and curate Autism genes have revealed an overrepresentation of Chromatin Regulator genes, suggesting a significant role in the etiology of this condition. This paper seeks to establish a novel means of studying chromatin regulator genes involved in neurodevelopmental disorders by utilizing the dTAG inducible protein depletion system which would allow for specific, rapid, and reversible protein depletion. In this paper we construct the CRISPR plasmid vectors for the tagging of the Chromatin Regulator SETD5 with the dTAG cassette for inducible protein depletion. One Cas9 vector targeting the SETD5 Stop codon and two HDR vectors were produced, each with a different selection marker to allow for ease in producing homozygous clones in the future. Two confirmed heterozygous HEK293T cell lines were confirmed to contain the dTAG cassette through sequencing. Partial degradation of the SETD5 protein upon induction was demonstrated as expected with heterozygotes and blotting for the HA-tag further confirmed this result. Overall these plasmid vector constructs provide a means of SETD5 tagging with an effective inducible protein system for further investigation into the effects of haploinsufficiency of SETD5. This framework is ideal for use in human embryonic stem cells which can recapitulate the neurodevelopmental stages during which haploinsufficiency has an effect. Therefore, it also lays a framework for the investigation of other genes implicated in neurodevelopment disorders.

Men and women require androgens for proper fertility. However, elevated androgen levels suppress the reproductive axis in women with polycystic ovarian syndrome (PCOS) and transgender men undergoing androgen hormone therapy. The mechanisms behind this reproductive deficit are not fully understood. We therefore investigated how elevated androgens affect a model of the preovulatory LH surge and the critically associated anteroventral periventricular nucleus (AVPV) kisspeptin neurons in the brain. C57BL/6 female mice were implanted with 3 different doses of the androgen dihydrotestosterone (DHT; implant length: 4 mm, 10 mm or 16 mm) which acts exclusively through the androgen receptor (AR). A week after implantation, mice were ovariectomized and implanted with an estradiol (E2) pellet to promote an exogenous LH surge two days later. Brain and blood samples were collected 2 days later either in the morning (AM; non-surge time) or evening (PM; surge time). We measured serum LH levels by radioimmunoassay. Kiss1 expression in the AVPV was measured by in situ hybridization, while AVPV kisspeptin neuron activation was assessed using Kiss1/cfos double-label RNAscope. We found that DHT potently suppressed the LH surge, even at the lowest dose tested. Additionally, AVPV Kiss1 expression was lowered in DHT-treated animals compared to control females. Lastly, AVPV kisspeptin neuron activation during the LH surge was strongly suppressed by DHT. These three experiments demonstrate that the female reproductive axis is sensitive to androgen signaling and is inhibited, in part, through the suppression of AVPV kisspeptin neurons.

The earliest stages of life, including the transition from the fully differentiated oocyte to the totipotent zygote, the first days of embryo cleavage and cell differentiation to form a blastocyst, and implantation of that blastocyst in the wall of the uterus, are somehow beautifully simple and strikingly complex at the same time. These stages, which represent the beginning of life for us and other vertebrates, are difficult to study, owing to a number of experimental and ethical considerations. However, advances in our understanding of nuclear reprogramming, transcriptional quiescence and activation, the maintenance of pluripotency, and what it takes for an embryo to initiate contact with the endometrium to generate a successful pregnancy, will without question have far-reaching influence for many scientific and medical disciplines. Here, I attempt to unravel some of these concepts, combining molecular and developmental biology with genome-wide analysis only recently made possible through advances in techniques with low input. These approaches, in combination, allow us to ask questions about early developmental systems that would not have been possible only years prior.

In the oocyte, global transcriptional silencing is a highly conserved mechanism that is essential for the transition from the differentiated oocyte to the totipotent zygote. Here, I report that global transcriptional silencing in the mouse oocyte depends on an mRNA decay activator. By downregulating master regulators of transcription during oocyte growth, particularly a group of mRNAs encoding demethylases for H3K4 and H3K9, ZFP36L2 enables increased histone methylation that is associated with transcriptional silencing. These results uncover a mRNA decay mechanism that acts a developmental switch during growth of the mammalian oocyte, resulting in wide-spread shifts in chromatin modification, and mediating silencing of transcription in the oocyte.

The pluripotent population of cells in the blastocyst, the inner cell mass, is established in the mouse embryo approximately three days after fertilization. These cells will undergo gastrulation to form the entire organism and can be maintained in culture as embryonic stem cells. I report here that UPF2, a mRNA decay activator, is needed specifically within the pluripotent inner cell mass of the mouse embryo for maintenance of pluripotency in the embryo in vivo, as well as for embryonic stem cells in vitro. That mRNA decay may underly the establishment or maintenance of this intriguing and complex population of cells, is an exciting possibility.

Reproductive success depends on embryo implantation in the uterus, and in fertile and infertile couples alike, failure of the embryo to implant in the wall of the uterus accounts for up to 75% of all lost pregnancies. Here, I provide a qualitative assessment of gene expression and cellular communication networks within the major compartments of the human blastocyst that are most closely associated with successful implantation and ongoing pregnancy. Most strikingly, establishment and/or maintenance of the extraembryonic primitive endoderm lineage—following the second major embryonic differentiation event—most strongly differentiates embryos of high and low implantation potential. Unbiased machine learning identified genes within each embryo compartment most closely associated with implantation. Taken together, this data supports a model in which successful implantation and ongoing pregnancy predominantly depends upon the inner cell mass and highlights a potentially novel role for the extraembryonic primitive endoderm in early pregnancy success.