Single-nucleus RNA-sequencing reveals novel potential mechanisms of ovarian insufficiency in 45,X Turner Syndrome

Study questionCan single-nuclei and bulk RNA sequencing technologies be used to elucidate novel mechanisms of ovarian insufficiency in Turner Syndrome (TS)? Summary answerUsing single-nucleus and bulk RNA sequencing approaches, we identified novel potential pathogenic mechanisms underlying ovarian insufficiency in TS including and beyond X chromosome haploinsufficiency. What is known alreadyTurner syndrome (TS) is the most common genetic cause of Primary Ovarian Insufficiency (POI) in humans. Morphological analyses of human fetal 45,X ovaries have demonstrated fewer germ cells and marked apoptosis established by 15-20 weeks post conception (wpc); however, we do not understand why POI develops mechanistically in the first instance. Study design, size, durationSingle-nucleus RNA sequencing (snRNA-seq): two 46,XX and two 45,X (TS) human fetal ovaries at 12-13 wpc. Bulk RNA sequencing: 19 human fetal ovary, 20 fetal testis, and 8 fetal control tissue (n=47 total samples; Carnegie Stage 22-16wpc). Participants/materials, setting, methodsTo identify novel potential mechanisms of ovarian insufficiency in TS and to characterise X chromosome gene expression in the 45,X ovary, we performed snRNA-seq of peri-meiotic 46,XX (n=2) and 45,X (n=2) fetal ovaries at 12-13 weeks post conception (wpc); and 2) a bulk RNA sequencing time-series analysis of fetal ovary, testis, and control samples across four developmental timepoints. Main results and the role of chanceGerm and somatic cell subpopulations were mostly shared across 46,XX and 45,X ovaries, aside from a 46XX-specific/45,X-depleted cluster of oogonia ("synaptic oogonia") containing genes with functions relating to sex chromosome synapsis; histone modification; intracellular protein regulation and chaperone systems. snRNA-seq enabled accurate cell counting localised to individual cell clusters; the 45,X ovary has fewer germ cells than the 46,XX ovary in every germ cell subpopulation, confirmed by histopathological analysis. The normal sequence of X-chromosome inactivation and reactivation is disrupted in 45,X ovaries; XIST was not expressed in 45,X somatic cells but was present in germ cell clusters, albeit with lower expression than in corresponding 46,XX clusters. The 45,X ovary has a globally abnormal transcriptome, with low expression of genes with proteostasis functions (RSP4X); cell cycle progression (BUB1B); and OXPHOS mitochondrial energy production (COX6C, ATP11C). Genes with higher expression in 45,X cell populations were enriched for apoptotic functions (e.g., NR4A1). Limitations, reasons for cautionLimitations include the relatively small sample size of the snRNA-seq analysis and the focus on a fixed meiotic timepoint which may overlook a dynamic process over time. Wider implications of the findingsWe characterise the human fetal peri-meiotic 45,X ovary at single-cell resolution and offer insights into novel pathogenic mechanisms underlying ovarian insufficiency in TS. Although asynapsis due to X chromosome haploinsufficiency likely plays a significant role, these data suggest meiotic failure and ovarian insufficiency may be a combinatorial process characterised by periods of vulnerability throughout early 45,X germ cell development Study funding/competing interest(s)This research was funded in whole, or in part, by the Wellcome Trust Grants 216362/Z/19/Z to SMcG-B and 209328/Z/17/Z to JCA. Human fetal material was provided by the Joint MRC/Wellcome Trust (Grant MR/R006237/1) Human Developmental Biology Resource (http://www.hdbr.org). Research at UCL Great Ormond Street Institute of Child Health is supported by the National Institute for Health Research, Great Ormond Street Hospital Biomedical Research Centre (grant IS-BRC-1215-20012).