Single-cell transcriptomic analysis reveals diversity within mammalian spinal motor neurons.

Spinal motor neurons (MNs) integrate sensory stimuli and brain commands to generate movements. In vertebrates, the molecular identities of the cardinal MN types such as those innervating limb versus trunk muscles are well elucidated. Yet the identities of finer subtypes within these cell populations that innervate individual muscle groups remain enigmatic. Here we investigate heterogeneity in mouse MNs using single-cell transcriptomics. Among limb-innervating MNs, we reveal a diverse neuropeptide code for delineating putative motor pool identities. Additionally, we uncover that axial MNs are subdivided into three molecularly distinct subtypes, defined by mediolaterally-biased Satb2, Nr2f2 or Bcl11b expression patterns with different axon guidance signatures. These three subtypes are present in chicken and human embryos, suggesting a conserved axial MN expression pattern across higher vertebrates. Overall, our study provides a molecular resource of spinal MN types and paves the way towards deciphering how neuronal subtypes evolved to accommodate vertebrate motor behaviors.

Inducible lncRNA transgenic mice reveal continual role of HOTAIR in promoting breast cancer metastasis.

HOTAIR is a 2.2-kb long noncoding RNA (lncRNA) whose dysregulation has been linked to oncogenesis, defects in pattern formation during early development, and irregularities during the process of epithelial-to-mesenchymal transition (EMT). However, the oncogenic transformation determined by HOTAIR in vivo and its impact on chromatin dynamics are incompletely understood. Here, we generate a transgenic mouse model with doxycycline-inducible expression of human HOTAIR in the context of the MMTV-PyMT breast cancer-prone background to systematically interrogate the cellular mechanisms by which human HOTAIR lncRNA acts to promote breast cancer progression. We show that sustained high levels of HOTAIR over time increased breast metastatic capacity and invasiveness in breast cancer cells, promoting migration and subsequent metastasis to the lung. Subsequent withdrawal of HOTAIR overexpression reverted the metastatic phenotype, indicating oncogenic lncRNA addiction. Furthermore, HOTAIR overexpression altered both the cellular transcriptome and chromatin accessibility landscape of multiple metastasis-associated genes and promoted EMT. These alterations are abrogated within several cell cycles after HOTAIR expression is reverted to basal levels, indicating an erasable lncRNA-associated epigenetic memory. These results suggest that a continual role for HOTAIR in programming a metastatic gene regulatory program. Targeting HOTAIR lncRNA may potentially serve as a therapeutic strategy to ameliorate breast cancer progression.

The m6A epitranscriptome on neural development and degeneration.

N6-methyladenosine (m6A) is the most prevalent, conserved, and abundant RNA modification of the mRNAs of most eukaryotes, including mammals. Similar to epigenetic DNA modifications, m6A has been proposed to function as a critical regulator for gene expression. This modification is installed by m6A methylation "writers" (Mettl3/Mettl14 methyltransferase complex), and it can be reversed by demethylase "erasers" (Fto and Alkbh5). Furthermore, m6A can be recognized by "readers" (Ythdf and Ythdc families), which may be interpreted to affect mRNA splicing, stability, translation or localization. Levels of m6A methylation appear to be highest in the brain, where it plays important functions during embryonic stem cell differentiation, brain development, and neurodevelopmental disorders. Depletion of the m6A methylation writer Mettl14 from mouse embryonic nervous systems prolongs cell cycle progression of radial glia and extends cortical neurogenesis into postnatal stages. Recent studies further imply that dysregulated m6A methylation may be significantly correlated with neurodegenerative diseases. In this review, we give an overview of m6A modifications during neural development and associated disorders, and provide perspectives for studying m6A methylation.

Mir-17∼92 Confers Motor Neuron Subtype Differential Resistance to ALS-Associated Degeneration.

Progressive degeneration of motor neurons (MNs) is the hallmark of amyotrophic lateral sclerosis (ALS). Limb-innervating lateral motor column MNs (LMC-MNs) seem to be particularly vulnerable and are among the first MNs affected in ALS. Here, we report association of this differential susceptibility with reduced expression of the mir-17∼92 cluster in LMC-MNs prior to disease onset. Reduced mir-17∼92 is accompanied by elevated nuclear PTEN in spinal MNs of presymptomatic SOD1G93A mice. Selective dysregulation of the mir-17∼92/nuclear PTEN axis in degenerating SOD1G93A LMC-MNs was confirmed in a double-transgenic embryonic stem cell system and recapitulated in human SOD1+/L144F-induced pluripotent stem cell (iPSC)-derived MNs. We further show that overexpression of mir-17∼92 significantly rescues human SOD1+/L144F MNs, and intrathecal delivery of adeno-associated virus (AAV)9-mir-17∼92 improves motor deficits and survival in SOD1G93A mice. Thus, mir-17∼92 may have value as a prognostic marker of MN degeneration and is a candidate therapeutic target in SOD1-linked ALS. VIDEO ABSTRACT.

Dlk1-Dio3 locus-derived lncRNAs perpetuate postmitotic motor neuron cell fate and subtype identity.

The mammalian imprinted Dlk1-Dio3 locus produces multiple long non-coding RNAs (lncRNAs) from the maternally inherited allele, including Meg3 (i.e., Gtl2) in the mammalian genome. Although this locus has well-characterized functions in stem cell and tumor contexts, its role during neural development is unknown. By profiling cell types at each stage of embryonic stem cell-derived motor neurons (ESC~MNs) that recapitulate spinal cord development, we uncovered that lncRNAs expressed from the Dlk1-Dio3 locus are predominantly and gradually enriched in rostral motor neurons (MNs). Mechanistically, Meg3 and other Dlk1-Dio3 locus-derived lncRNAs facilitate Ezh2/Jarid2 interactions. Loss of these lncRNAs compromises the H3K27me3 landscape, leading to aberrant expression of progenitor and caudal Hox genes in postmitotic MNs. Our data thus illustrate that these lncRNAs in the Dlk1-Dio3 locus, particularly Meg3, play a critical role in maintaining postmitotic MN cell fate by repressing progenitor genes and they shape MN subtype identity by regulating Hox genes.

When a gene is active, its DNA sequence is 'transcribed' to form a molecule of RNA. Many of these RNAs act as templates for making proteins. But for some genes, the protein molecules are not their final destinations. Their RNA molecules instead help to control gene activity, which can alter the behaviour or the identity of a cell. For example, experiments performed in individual cells suggest that so-called long non-coding RNAs (or lncRNAs for short) guide how stem cells develop into different types of mature cells. However, it is not clear whether lncRNAs play the same critical role in embryos.Yen et al. used embryonic stem cells to model how motor neurons develop in the spinal cord of mouse embryos. This revealed that motor neurons produce large amounts of a specific group of lncRNAs, particularly one called Meg3. Further experiments showed that motor neurons in mouse embryos that lack Meg3 do not correctly silence a set of genes called the Hox genes, which are crucial for laying out the body plans of many different animal embryos. These neurons also incorrectly continue to express genes that are normally active in an early phase of the stem-like cells that make motor neurons.There is wide interest in how lncRNAs help to regulate embryonic development. With this new knowledge of how Meg3 regulates the activity of Hox genes in motor neurons, research could now be directed toward investigating whether lncRNAs help other tissues to develop in a similar way.

Mir-17∼92 Governs Motor Neuron Subtype Survival by Mediating Nuclear PTEN.

Motor neurons (MNs) are unique because they project their axons outside of the CNS to innervate the peripheral muscles. Limb-innervating lateral motor column MNs (LMC-MNs) travel substantially to innervate distal limb mesenchyme. How LMC-MNs fine-tune the balance between survival and apoptosis while wiring the sensorimotor circuit en route remains unclear. Here, we show that the mir-17∼92 cluster is enriched in embryonic stem cell (ESC)-derived LMC-MNs and that conditional mir-17∼92 deletion in MNs results in the death of LMC-MNs in vitro and in vivo. mir-17∼92 overexpression rescues MNs from apoptosis, which occurs spontaneously during embryonic development. PTEN is a primary target of mir-17∼92 responsible for LMC-MN degeneration. Additionally, mir-17∼92 directly targets components of E3 ubiquitin ligases, affecting PTEN subcellular localization through monoubiquitination. This miRNA-mediated regulation modulates both target expression and target subcellular localization, providing LMC-MNs with an intricate defensive mechanism that controls their survival.

An Lmx1b-miR135a2 regulatory circuit modulates Wnt1/Wnt signaling and determines the size of the midbrain dopaminergic progenitor pool.

MicroRNAs regulate gene expression in diverse physiological scenarios. Their role in the control of morphogen related signaling pathways has been less studied, particularly in the context of embryonic Central Nervous System (CNS) development. Here, we uncover a role for microRNAs in limiting the spatiotemporal range of morphogen expression and function. Wnt1 is a key morphogen in the embryonic midbrain, and directs proliferation, survival, patterning and neurogenesis. We reveal an autoregulatory negative feedback loop between the transcription factor Lmx1b and a newly characterized microRNA, miR135a2, which modulates the extent of Wnt1/Wnt signaling and the size of the dopamine progenitor domain. Conditional gain of function studies reveal that Lmx1b promotes Wnt1/Wnt signaling, and thereby increases midbrain size and dopamine progenitor allocation. Conditional removal of Lmx1b has the opposite effect, in that expansion of the dopamine progenitor domain is severely compromised. Next, we provide evidence that microRNAs are involved in restricting dopamine progenitor allocation. Conditional loss of Dicer1 in embryonic stem cells (ESCs) results in expanded Lmx1a/b+ progenitors. In contrast, forced elevation of miR135a2 during an early window in vivo phenocopies the Lmx1b conditional knockout. When En1::Cre, but not Shh::Cre or Nes::Cre, is used for recombination, the expansion of Lmx1a/b+ progenitors is selectively reduced. Bioinformatics and luciferase assay data suggests that miR135a2 targets Lmx1b and many genes in the Wnt signaling pathway, including Ccnd1, Gsk3b, and Tcf7l2. Consistent with this, we demonstrate that this mutant displays reductions in the size of the Lmx1b/Wnt1 domain and range of canonical Wnt signaling. We posit that microRNA modulation of the Lmx1b/Wnt axis in the early midbrain/isthmus could determine midbrain size and allocation of dopamine progenitors. Since canonical Wnt activity has recently been recognized as a key ingredient for programming ESCs towards a dopaminergic fate in vitro, these studies could impact the rational design of such protocols.