RNA epitranscriptomics dysregulation: A major determinant for significantly increased risk of ASD pathogenesis

Abstract

Autism spectrum disorders (ASDs) are perhaps the most severe, intractable and challenging child psychiatric disorders. They are complex, pervasive and highly heterogeneous and depend on multifactorial neurodevelopmental conditions. Although the pathogenesis of autism remains unclear, it revolves around altered neurodevelopmental patterns and their implications for brain function, although these cannot be specifically linked to symptoms. While these affect neuronal migration and connectivity, little is known about the processes that lead to the disruption of specific laminar excitatory and inhibitory cortical circuits, a key feature of ASD. It is evident that ASD has multiple underlying causes and this multigenic condition has been considered to also dependent on epigenetic effects, although the exact nature of the factors that could be involved remains unclear. However, besides the possibility for differential epigenetic markings directly affecting the relative expression levels of individual genes or groups of genes, there are at least three mRNA epitranscriptomic mechanisms, which function cooperatively and could, in association with both genotypes and environmental conditions, alter spatiotemporal proteins expression patterns during brain development, at both quantitative and qualitative levels, in a tissue-specific, and context-dependent manner. As we have already postulated, sudden changes in environmental conditions, such as those conferred by maternal inflammation/immune activation, influence RNA epitranscriptomic mechanisms, with the combination of these processes altering fetal brain development. Herein, we explore the postulate whereby, in ASD pathogenesis, RNA epitranscriptomics might take precedence over epigenetic modifications. RNA epitranscriptomics affects real-time differential expression of receptor and channel proteins isoforms, playing a prominent role in central nervous system (CNS) development and functions, but also RNAi which, in turn, impact the spatiotemporal expression of receptors, channels and regulatory proteins irrespective of isoforms. Slight dysregulations in few early components of brain development, could, depending upon their extent, snowball into a huge variety of pathological cerebral alterations a few years after birth. This may very well explain the enormous genetic, neuropathological and symptomatic heterogeneities that are systematically associated with ASD and psychiatric disorders at large.

Keywords: Autism spectrum disorder (ASD); RNA epitranscriptomics; adenosine-to-inosine RNA editing; mRNA alternative splicing; mRNA poly(A)-tail modulation; maternal inflammation.

FIGURE 1

Role of alternative splicing (AS) in mRNA stability, translational activity and subcellular localization. (Top) Examples of relevant AS topologies; mid: mature mRNA containing 5′ and 3′UTRs flanking the protein-coding ORF; (bottom) downstream regulation outcomes of AS events in the corresponding regions. APA, alternative cleavage and polyadenylation; ARE, AU-rich element; NMD, nonsense-mediated decay; NMTR, nonsense-mediated translational repression; uORF, upstream open reading frame (ORF). Figure obtained from Mockenhaupt and Makeyev (2015) under a creative commons license 3 and 4.

FIGURE 2

Diagram representing some of the splicing mechanisms in the brain. Numerous transcripts important in brain and neurological development (green boxes) are cross-regulated (indicated by arrows) by multiple RNA-binding proteins (RBPs; colored ovals on the left and right). Girk2, inwardly rectifying potassium channel Kir3.2; Gabrg2, GABAA receptor subunit gamma 2; Gabbr2, GABAB receptor 2; MBNL, muscleblind-like; NOVA, neuro-oncological ventral antigen; Psd95, postsynaptic density protein 95; PTBP, polypyrimidine tract binding protein; Rest, repressor element 1-silencing transcription factor; RBFOX, RNA-binding protein fox 1 homolog; Snap25, synaptosome-associated protein 25; SRRM4, serine/arginine repetitive matrix protein 4. Figure from Vuong et al. (2016) obtained by Springer Nature under license No 5430810425823.

FIGURE 3

(A) Alternative splicing regulation of synaptogenesis and synaptic function. At the presynaptic terminal, alternative splicing of synaptosomal-associated protein 25 (Snap25) by RNA-binding protein fox 1 homolog 1 (RBFOX1/A2BP1) and of the calcium-activated potassium channel subunit alpha 1 (Kcnma1) by muscleblind-like 2 (MBNL2) are important to control neurotransmitter release. Differential splicing of the presynaptic neurexins (Nrxns) at AS4 by KHDRBS proteins (SAM68, SLM1, and SLM2) controls targeting to postsynaptic partners. At excitatory synapses, alternative splicing of the transcript encoding the NMDA receptor subunit GluN1, Grin1, is regulated by RBFOX1 (A2BP1) and MBNL2, whereas the polypyrimidine tract binding proteins (PTBPs) control productive splicing of the scaffold protein, postsynaptic density protein 95 (Psd95). Splicing of the transcripts encoding L-type voltage-gated calcium channels, such as the pore-forming subunit Cav1.3 (encoded by Cacna1d), by MBNL2 may allow the voltage sensitivity, conductance, or other properties to be tuned as synapses differentiate. At inhibitory synapses, neuro-oncological ventral antigen 2 (NOVA2) mediates alternative splicing of the transcripts encoding many postsynaptic components such as the metabotropic GABAB receptor (Gabbr2), the inwardly rectifying potassium channel Kir3.2 (Girk2) and the glycine receptor alpha 2 (Glra2). Splicing of the GABAA receptor subunit transcript (Gabrg2) is controlled by multiple splicing regulators including NOVA2, RBFOX1 (A2BP1) and PTBP2. (B) Alternative splicing regulation of synaptogenesis and synaptic function. Alternative splicing controls the expression and function of many synaptic components. Expression of PSD95 is repressed by PTBP-controlled exclusion of exon 18 until late in neuronal maturation when it is required for synaptogenesis. The gene encoding the voltage-gated sodium channel Nav1.6, Scn8a, has multiple alternative exons (such as 5N, 5A, 18N, and 18A as shown in the figure) that can change its gating properties, determine its localization or alter its overall function. GABAAR, GABAA receptor; GABABR, GABAB receptor; GlyR, glycine receptor; NMD, nonsense-mediated decay; NMDAR, NMDA receptor; SAM68, SRC-associated in mitosis 68 kDa protein. Figure from Vuong et al. (2016) obtained by Springer Nature under license No 5430810425823.

FIGURE 4

Diagram depicting the number of genes showing significant expression differences between frontal and temporal cortex in control samples (top) and autism samples (bottom). (A) Top 20 genes differentially expressed between frontal and temporal cortex in control samples. All of the genes shown are also differentially expressed between frontal and temporal cortex in fetal mid-gestation brain, but show no significant expression differences between frontal and temporal cortex in autism. The horizontal bars depict p values for differential expression between frontal and temporal cortex in the autism and control groups. (B) Top A2BP1 (RBFOX1) -specific differential splicing (DS) events. DS events showing the most significant differences in alternative splicing between low-A2BP1 autism cases and controls as well as DS differences consistent with the A2BP1 binding site position. The horizontal axis depicts the percentage of transcripts including the alternative exon. Red-autism samples, black-control samples Figure modified from Voineagu et al. (2011) obtained by Springer Nature under license No 5430811097703 and 5431300363740.

FIGURE 5

CPEB activity in inducing cytoplasmic shortening or elongation of poly(A)-tails. Some mRNAs contain a cytoplasmic polyadenylation element (CPE), which is bound by CPEB. (A) CPEB also interacts with Maskin (or Neuroguidin, a functionally related protein), which in turn interacts with the cap (m7G) binding protein eIF4E. Such mRNAs are translationally inactive because Maskin inhibits the association of eIF4E and eIF4G, another initiation factor that helps recruit the 40S ribosomal subunit to the 5’ end of the mRNA. (B) Cues such as NMDA receptor activation stimulate the kinase Aurora A, which phosphorylates CPEB, an event that causes poly(A) tail elongation. poly(A) binding protein (PABP) binds the newly elongated poly(A) tail and recruits eIF4G. The PABP-eIF4G dimer helps to displace Maskin from eIF4E, allowing eIF4G to bind eIF4E and initiate translation [figure modified from Zukin et al. (2009), obtained under a creative commons license 3 and 4].

FIGURE 6

Recycling of αCaMKII mRNA in neurons. Alternative polyadenylation (APA) of αCaMKII mRNA generates two transcripts with 3′ UTRs of different lengths (1). Both transcripts are polyadenylated in the nucleus, but the composition of the cis-elements in their 3′ UTR modifies their cytoplasmic fates. The short isoform is translated mainly in the soma (2), whereas miRNA binding sites, AU-rich elements (AREs) and cytoplasmic polyadenylation elements (CPEs) mediate deadenylation, repression (3) and transport (4) of the long transcript to dendrites. Upon neuronal stimulation in the post-synaptic densities (PSD), CPEB1 promotes αCaMKII mRNA polyadenylation (5), and αCAMKII local translation (6). Figure from Weill et al. (2012), obtained by Springer Nature under license No 5430811375131.

FIGURE 7

Adenosine deamination and the ADAR enzyme family. (A) ADAR enzymes catalyze the A-to-I hydrolytic deamination reaction, by which an adenosine loses an amine group and is converted to inosine. (B) There are four main proteins of the ADAR enzyme family: two isoforms of ADAR1 (p110 and p150), ADAR2 and ADAR3. All of these enzymes contain a conserved deaminase domain, shown in blue. The double-stranded (ds) RNA-binding domains, shown in purple, determine substrate specificity. The two ADAR1 isoforms differ in their Z-DNA-binding domains, shown in green. ADAR3 contains an arginine-rich domain, shown in pink, which binds single-stranded RNA. Figure from Slotkin and Nishikura (2013), obtained by Springer Nature under license No 5431360984026.

FIGURE 8

Impact of editing on selected neuronal receptors and proteins. Shown are several receptors and channels in their unedited (labeled in black) and edited (E, labeled in pink) versions. Editing at the Q/R site of ionotropic glutamate receptor GRIA2 (GluA2) subunit decreases Ca2 + permeability and endoplasmic reticulum exit efficiency. Membrane trafficking of the GABA_A receptor is reduced by editing of I to M in the alpha3 subunit. Editing of the serotonin 5-HT2c receptor converts the amino acids I-N-I to V-S-V, V-G-V, or V-N-V. This reduces G-protein coupling in the receptor. The editing-induced I to V exchange in the potassium voltage-gated channel KCNA1 (Kv1.1) alters the interaction with KCNB1. Editing of the IQ motif in voltage-gated calcium channel CACNA1D (Cav1.3) to MR abolishes calmodulin binding. Filamin-α (FLNA) is edited in a region that is known to interact with metabotropic glutamate receptor GRM7 (mGlu7) and some of its relatives. Figure from Tariq and Jantsch (2012), obtained under a creative commons license 3 and 4.

FIGURE 9

5-HT2C receptor signaling pathways. The 5-HT2CR is coupled to PLC in neurons and choroid plexus, and its activation leads to an accumulation of IP3. It inhibits K + conductance in both neurons and choroid plexus and stimulates an apical Cl- conductance in choroid plexus. It may also be coupled to PLD in transfected cells and choroid plexus. Figure from Masson et al. (2012), obtained under a creative commons license 3 and 4.

FIGURE 10

Overview of RNA editing effects on CaV1.3 Ca2 + -dependent inactivation (CDI) and calcium load in neurons. Channel schematic as defined in Figure above. Increased RNA editing (right scenario) favors channels with MQ and other edited versions of the IQ domain, decreasing overall CDI and presumably increasing cellular Ca2 + load (intense yellow-green coloration). Decreased editing (left scenario) favors default channels with IQ version of IQ domain, increasing overall CDI and potentially decreasing cellular Ca2 + load (weak yellow-green coloration). Actual neurons reside on a continuum between these two extremes, as represented by ramp schematics at bottom. Figure from Huang et al. (2012), obtained by Elsevier under license No 5430820583951.

FIGURE 11

Regulation of the Kv1.1 (KCNA1) channel by Kvb1.1 (KCNE5). Kvb1.1 has an inactivation gate which interacts with the unedited (I) form of the Kv1.1 receptor. Editing of Kv1.1 changes the isoleucine to valine (V). This reduces the affinity for Kvb1.1 and enhances recovery from inactivation. While all Kv channels (voltage-gated) function along the same principles (Grizel et al., 2014), different Kv channel subtypes show very different responses to short stimulation (Cameron et al., 2017). Figure from Tariq and Jantsch (2012), obtained under a creative commons license 3 and 4.

FIGURE 12

Regulation of RNA interference (RNAi) by adenosine deaminases acting on RNA (ADARs). Two different types of interaction between RNA-editing and RNAi pathways are known, one antagonistic and the other stimulative. (A) In antagonistic interactions, ADAR–ADAR homodimers edit long double-stranded RNA (dsRNA) and certain microRNA (miRNA) precursors. Editing changes the dsRNA structure and makes it less accessible to Drosha and/or Dicer, consequently decreasing the efficacy of RNAi by reducing the production of short interfering RNAs (siRNAs) and miRNAs. (B) In the case of stimulative interactions, ADAR1, as part of a Dicer–ADAR1 heterodimer, promotes RNAi by increasing the Dicer cleavage reaction rate, thereby generating more siRNAs and miRNAs and enhancing RISC (RNA-induced silencing complex) loading and target mRNA silencing. AGO2, Argonaute 2. Figure from Nishikura (2016), obtained by Springer Nature under license No 5430820144326.

FIGURE 13

Overview of neuronal signaling events that influence miRNA biogenesis, activity, and degradation. Each step within the miRNA biogenesis pathway may be stimulated (green arrow) or inhibited (red bar) by intra- and extracellular signaling events. Pri-miRNA levels increase when BDNF or other signals activate transcription factors that stimulate the transcription of miRNA-encoding genes. Pri-miRNA cleavage by the Microprocessor is influenced by the activity of proteins, such as PP1 which inhibits Microprocessor activity. Pre-miRNA cleavage by Dicer is increased in response to glutamate, BDNF signaling, or NMDA receptor activation. BDNF can also inhibit Dicer’s ability to cleave some pre-miRNAs by inducing Lin28 binding to the pre-miRNA terminal loop or by promoting TRBP redistribution and dissociation from Dicer. Neuronal activity and NMDA receptor activation inhibit RISC activity by promoting the degradation of the RISC component Mov10. miRNA interactions with target mRNAs are also influenced by RNA binding proteins such as FMRP or HuD neuron-specific HuD and HuB are subject to adenosine-to-inosine RNA editing at five editing sites each (Enstero et al., 2010), which are regulated by mGluRs (GRMs) and mTORC1 signaling, respectively. P38-induced phosphorylation of Ago2 stimulates mRNA translation by causing the RISC to release its bound miRNA. miRNAs may also be destabilized by increases in neuronal activity, by glutamatergic signaling, or by pilocarpine-induced inhibition of the miRNA stabilizing protein Gld2. Pilocarpine may also stimulate miRNA degradation by increasing expression of the exonuclease Xrn2. Figure from Thomas et al. (2018), obtained under a creative commons license 3 and 4.

FIGURE 14

Biogenesis and regulation pathways of dendritic miRNAs. The miRNA gene is transcribed into a primary miRNA transcript (pri-mRNA) which is cleaved by Drosha to generate a hairpin miRNA precursor (pre-miRNA). After nuclear export, the pre-miRNA is cleaved by Dicer to form the double-stranded miRNA duplex. One strand of this duplex, the mature miRNA, is then incorporated into the miRNA-induced silencing complex (miRISC). The miRISC complex can bind to complementary target mRNAs, thereby repressing their translation. The figure depicts those targets that are in turn regulating miRNA expression in neurons. Figure from Bicker et al. (2014), obtained by Springer Nature under license No 5430820320321.

FIGURE 15

Roles for miR-137 at the glutamatergic synapse. This figure summarizes the findings of Kwon et al. (2013), Zhao et al. (2013), Olde Loohuis et al. (2015), Siegert et al. (2015), and Thomas et al. (2017). miR-137 regulates presynaptic signaling by regulating vesicle trafficking in the axon terminal. miR-137 also targets the mRNA that encodes the L-type calcium channel subunit Cav1.2 (CACNA1C), which has also been linked to schizophrenia with genome-wide significance. Postsynaptically, miR-137 regulates the levels of glutamatergic receptor subunits GluA1 (GRIA1) and GluN2A (GRIN2A), and bioinformatic predictions suggest that miR-137 may target the metabotropic glutamate receptor mGluR5 (GRM5) as well as ErbB4, which regulates the strength of glutamatergic synapses. mGluR5 signaling, in turn, increases miR-137 levels. miR-137 also regulates proteins within the PI3K-Akt-mTOR pathway, e.g., p55g, to regulate neuronal responses to BDNF and Nrg1 signaling. miR-137 may regulate PI3K-Akt-mTOR signaling downstream of mGluR receptors as well. Figure from Thomas et al. (2018), obtained under a creative commons license 3 and 4.