A novel Oct4/Pou5f1-like non-coding RNA controls neural maturation and mediates developmental effects of ethanol

Abstract

Prenatal ethanol exposure can result in loss of neural stem cells (NSCs) and decreased brain growth. Here, we assessed whether a noncoding RNA (ncRNA) related to the NSC self-renewal factor Oct4/Pou5f1, and transcribed from a processed pseudogene locus on mouse chromosome 9 (mOct4pg9), contributed to the loss of NSCs due to ethanol. Mouse fetal cortical-derived NSCs, cultured ex vivo to mimic the early neurogenic environment of the fetal telencephalon, expressed mOct4pg9 ncRNA at significantly higher levels than the parent Oct4/Pou5f1 mRNA. Ethanol exposure increased expression of mOct4pg9 ncRNA, but decreased expression of Oct4/Pou5f1. Gain- and loss-of-function analyses indicated that mOct4pg9 overexpression generally mimicked effects of ethanol exposure, resulting in increased proliferation and expression of transcripts associated with neural maturation. Moreover, mOct4pg9 associated with Ago2 and with miRNAs, including the anti-proliferative miR-328-3p, whose levels were reduced following mOct4pg9 overexpression. Finally, mOct4pg9 inhibited Oct4/Pou5f1-3'UTR-dependent protein translation. Consistent with these observations, data from single-cell transcriptome analysis showed that mOct4pg9-expressing progenitors lack Oct4/Pou5f1, but instead overexpress transcripts for increased mitosis, suggesting initiation of transit amplification. Collectively, these data suggest that the inhibitory effects of ethanol on brain development are explained, in part, by a novel ncRNA which promotes loss of NSC identity and maturation.

Keywords: FASD; Fetal neural progenitor cells; Neural development; Oct4/Pou5f1; Pseudogene; miRNAs; ncRNA.

Figure 1:

Identification and characterization of Oct4 pseudogenes:(a) Genomic location and expression status of Oct4 and Oct4 pseudogenes in embryonic stem cells (ESC) and neural stem cells (NSC) in neurosphere culture. (b) RNA transcripts aligning with the genomic location of mOct4pg9, ± 2000 nucleotides, in different brain tissues at various developmental time points. Thickness of the bar indicates expression level (FPKM) of the transcript. (c) Analysis of publicly available single-cell RNAseq data from the Tabula Muris consortium compendium (Schaum et al., 2018) showing RNA transcripts bracketing the region of mOct4pg9 in astrocytes, neurons, oligodendrocytes, and oligodendrocyte progenitor cells (OPC) from adult C57bl/6JN mice. (d) Expression levels of Oct4/Pou5f1 and mOct4pg9 during the time course of differentiation of neurospheres, with data expressed as −ΔΔCT relative to neurosphere culture (NeuroCult), bars represent mean ± SEM, individual points are independent biological replicates, * indicates Fisher LSD post hoc relative to neurosphere culture (NeuroCult) p-value <0.05. (e) Expression of POU5F1, POU5F1B, POUF1P3 and POU5F1P4 through neural differentiation time-course of human embryonic stem cells (hESCs) (Pluripotency (day 0); differentiation initiation (days 2, 4, and 6); neural commitment (days 8–10) and neural progenitor cell proliferation (days 12, and 14, GSE103715, (Li et al., 2017)).

Figure 2:

Characterization of mOct4pg9 regulation of Oct4/Pou5f1 (a) Analysis of western immunoblot assay for OCT4/POU5F1 protein and β-actin proteins in lysates of neurospheres treated with 0, 60, 120 or 320 mg/dl of ethanol. Bar-graph represents mean expression of Oct4/Pou5f1, normalized to β-actin expression, in each group compared to average expression in the controls. (b) mOct4pg9 ncRNA expression, in ethanol-treated neurosphere cultures, expressed as −ΔΔCT, where a difference of 1 ΔΔCT unit represents an ~2-fold change in expression. (c) Venn diagram showing the overlap between the 2000 most abundantly expressed genes in Oct4/Pou5f1^+ve, and mOct4pg9^+ve compared to the most abundant gene transcripts in all cells. (d,e) RT-PCR analysis of Oct4/Pou5f1 mRNA (d) and western immunoblot assays for the expression of OCT4/POU5F1 protein (e) in lysates from mOct4pg9 overexpression (OE) and control neurospheres and mOct4pg9 knockdown (KD) and control neurosphere cultures. Top panel (e) shows the quantification of the immunoblots normalized to loading control protein (tubulin) represented as a ratio to the average intensity in the control. Bars represent mean ± SEM, individual points represent independent biological replicates, * indicates unpaired two-tailed t-test p-value< 0.05 (full uncropped Western immunoblots are presented in supplementary figure 1j).

Figure 3:

mOct4pg9 lncRNA association with miRNAs and Argonaut chaperone proteins: (a) Schematic diagram (ggmsa R package) showing the sequence homology between Oct4/Pou5f1 3’-UTR and mOct4pg9 (Red, Blue, Orange and Green represent A, C, G and T nucleotides respectively). (b) Electrophoretic fractionation of PCR amplicons obtained from reverse-transcribed RNA amplified with primers for mOct4pg9 following immunoprecipitation with anti-pan-Ago or anti-Ago2 antibodies, or control IgG (IG) of subcellular fractionated lysates from neurosphere cultures. (c) Ratio of fluc/Rluc intensity resulting from miRNA Target clone control vector and Oct4/Pou5f1 3’-UTR luciferase reporter vector co-electroporated with either mOct4pg9 overexpression or control overexpression plasmid. * indicates p-value <0.05, Fisher LSD post hoc test. (d, e) Volcano plots showing statistical significance (-log[p-value]) compared to Hedges’s g effect size of the change in miRNA expression following mOct4pg9 knockdown (d) and overexpression (e). Red line marks the threshold of p<0.05 (unpaired two-tailed t-test).

Figure 4:

Effects of mOct4pg9 ncRNA manipulation on neurosphere size and DNA synthesis: (a) Representative images and quantification of sizes of neurospheres after mOct4pg9 siRNA mediated knockdown and overexpression (OE). (b) Rate of DNA synthesis, expressed as mean fluorescence intensity (MFI) of EdU, after either knockdown of Oct4/Pou5f1 and/or mOct4pg9 or following mOct4pg9 overexpression. * indicates Fisher LSD post-hoc (left) or unpaired two tailed t-test (right), p-value<0.05 compared to control. Scale bar, 100 um.

Figure 5:

RNA expression level of key stem cell and differentiation-associated genes in lysates from of neurosphere cultures treated with 0, 60, 120 or 320 mg/dl of ethanol, overexpressing mOct4pg9 (OE), or treated with siRNA to knockdown (KO) or CRISPR sgRNA to knockdown (KO) mOct4pg9. Data is expressed as −ΔΔCT compared to the control of each experiment. Bars represent mean ± SEM, individual points represent independent biological replicates, * indicates p-value <0.05 for either Fisher LSD post hoc (compared to the control) or unpaired two-tailed t-test.