Work Description

03Sep20 Dataset Title: Prenatal androgen exposure alters the developmental trajectory of both GnRH neuron and preoptic area RNA transcripts in female mice. Dataset Creators: Laura L. Burger, Elizabeth R. Wagenmaker, Chayarndorn Phumsatitpong, David P. Olson and Suzanne M. Moenter Dataset Contact: Laura Burger llburger@umich.edu Funding: National Institute of Health/Eunice Kennedy Shriver National Institute of Child Health and Human Development P50 HD28934. Key Points: The goal of this study was to investigate changes in the GnRH neuron translatome (mRNAs in the process of translation) via Translating Ribosome Affinity Purification (TRAP) combined with RNA sequencing (RNAseq; TRAPseq). TRAPseq was performed on PNA and controls before puberty at three weeks and in adulthood (~110d). Adults tissue was collected on diestrus. Preoptic area (POA) punches from 2-5 littermates per treatment were pooled for RNAseq. Each pool of POA punches is considered as a single biological replicate. There were 5-6 biological replicates per treatment. Prominent in GnRH neurons were transcripts associated with protein synthesis and cellular energetics, in particular oxidative phosphorylation. The GnRH neuron transcript profile was affected more by the transition from prepuberty to adulthood than by PNA treatment, however PNA did change the developmental trajectory of GnRH neurons. This included families of transcripts related to both protein synthesis and oxidative phosphorylation, which were more prevalent in adults than in prepubertal mice but were blunted in PNA adults. These findings suggest that prenatal androgen exposure can program alterations in the translatome of GnRH neurons, providing a mechanism independent of changes in the genetic code for altered expression. Research Overview: Polycystic ovary syndrome (PCOS) is the most common form of infertility in women. The causes of PCOS are not yet understood and both genetics and early-life exposure have been considered as candidates. With regard to the latter, circulating androgens are elevated in mid-late gestation in women with PCOS, potentially exposing offspring to elevated androgens in utero; daughters of women with PCOS are at increased risk for developing this disorder. Consistent with these clinical observations, prenatal androgenization (PNA) of several species recapitulates many phenotypes observed in PCOS. There is increasing evidence that symptoms associated with PCOS, including elevated luteinizing hormone (LH) (and presumably gonadotropin-releasing hormone (GnRH)) pulse frequency emerge during the pubertal transition. To examine if PNA alters the translatome of the GnRH neuron we employed Translating Ribosome Affinity Purification (TRAP) combined with RNA sequencing (RNAseq; TRAPseq) both in controls and PNA mice across development from prepuberty to adulthood. Methodology: Using CRE-LoxP technology GnRH neuron ribosomes are labelled with GFP and GnRH neuron enriched RNA is isolated by immunoprecipitation (IP). This results in 2 RNA samples per each biological replicate: one enriched for GnRH neuron transcripts (IP) and the other depleted. cDNA libraries were created and 50 base, paired-end sequencing was done by the University of Michigan DNA Sequencing Core. Analysis was done using Cufflinks/CuffDiff and HTSeq/DESeq2 for expression quantitation and normalization, and for differential expression using UCSC mm10.fa as the reference genome sequence by the University of Michigan Bioinformatics Cores. The threshold for gene differential expression (enriched vs de-enriched in the GnRH neuron) was set at FoldChange �log2|0.585| (linear fold change of 1.5), with FDR adjusted p value (padj) �0.05. In these Deseq2 outputs, enrichment or de-enrichment is indicated by the column identified as 'Call'. YES = meets the criteria for log2FC�|0.585| and padj<0.05, NO = does not meet one or both of those criteria. We added additional quality controls and removed from consideration transcripts that met the criteria for enriched or de-enriched, but had basemean (mean depth normalized CPM across all samples) <50CPM as these transcripts are very low abundance. This additional filter is considered in "Call � 50CPM". Last, we filtered out transcripts that are erroneously enriched. See Burger_TRAPSeq_Data_1 and figure 2 of the accompanying manuscript. These are shown in red font through the data files. The unprocessed data is available at Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/gds) with accession GSE155314. Files contained here: Burger_TRAPSeq_Data_1 Microsoft Excel File of GnRH neuron enriched (and de-enriched) transcripts and those that are erroneously enriched in negative control mice. Burger_TRAPSeq_Data_2 Microsoft Excel File Comparing GnRH neuron enriched (and de-enriched) transcripts in adult diestrus females from a 2018 study (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6287592/) and the current study (2020) to determine technique reproducibility. Burger_TRAPSeq_Data_3 Microsoft Excel File Comparing GnRH neuron enriched (and de-enriched) transcripts in among 3Wk and adult mice prenatally androgenized (or controls). This file contains individual groups (e.g. 3Wk.CON). Burger_TRAPSeq_Data_4 Microsoft Excel File Comparing GnRH neuron enriched (and de-enriched) transcripts in among 3Wk and adult mice or controls and PNA mice. This file contains data where treatments, CON and PNA, are combined irrespective of age and ages, 3Wk and adult, are combined irrespective of treatment. The goal was to increase power and determine if development or prenatal treatment had the larger effect. Burger_TRAPSeq_Data_5 Microsoft Excel File Comparing the depleted RNA fraction (non-GnRH neuron) between individual groups (e.g. 3Wk.CON).The goal was to examine transcript changes the hypothalamic preoptic area, outside of the GnRH neuron, that occur either developmentally or with prenatal androgenization. Burger_TRAPSeq_Data_6 Microsoft Excel File Gene ontology for GnRH neurons. Burger_TRAPSeq_Data_7 Microsoft Excel File Gene ontology for Non-GnRH neurons.