Imbalance in Unc80 RNA Editing Disrupts Dynamic Neuronal Activity and Olfactory Perception

Abstract

A-to-I RNA editing, catalyzed by the ADAR protein family, significantly contributes to the diversity and adaptability of mammalian RNA signatures, aligning with developmental and physiological needs. Yet, the functions of many editing sites are still to be defined. The Unc80 gene stands out in this context due to its brain-specific expression and the evolutionary conservation of its codon-altering editing event. The precise biological functions of Unc80 and its editing, however, are still largely undefined. In this study, we first demonstrated that Unc80 editing occurs in an ADAR2-dependent manner and is exclusive to the brain. By employing the CRISPR/Cas9 system to generate Unc80 knock-in mouse models that replicate the natural editing variations, our findings revealed that mice with the "gain-of-editing" variant (Unc80^G/G) exhibit heightened basal neuronal activity in critical olfactory regions, compared to the "loss-of-editing" (Unc80^S/S) counterparts. Moreover, an increase in glutamate levels was observed in the olfactory bulbs of Unc80^G/G mice, indicating altered neurotransmitter dynamics. Behavioral analysis of odor detection revealed distinctive responses to novel odors-both Unc80 deficient (Unc80^+/-) and Unc80^S/S mice demonstrated prolonged exploration times and heightened dishabituation responses. Further elucidating the olfactory connection of Unc80 editing, transcriptomic analysis of the olfactory bulb identified significant alterations in gene expression that corroborate the behavioral and physiological findings. Collectively, our research advances the understanding of Unc80's neurophysiological functions and the impact of its editing on the olfactory sensory system, shedding light on the intricate molecular underpinnings of olfactory perception and neuronal activity.

Keywords: RNA editing; Unc80; neuronal activity; olfactory perception.

Figure 1

Expression patterns of ADAR2-mediated RNA editing of Unc80 in the brain and its structural implications. Total RNA was extracted from various tissues and brain regions from wild-type (WT) and Adar2-knockout (KO) mice. The samples were then analyzed using RT-PCR (A) and Sanger sequencing (B). The sequencing chromatograms highlight the absence of guanine (“G”) signals (indicated by arrows) in the KO samples, demonstrating the reliance of Unc80 editing on ADAR2’s enzymatic function. Differences in RNA editing levels across brain regions were also observed. The percentage represents the editing frequency, calculated by taking the peak area of G peak over the sum of A and G peaks. (C) Prediction of 3-dimensional protein structure models of Unc80^WT and Unc80^S2732G. The magnified view of the region of interest highlights the residue change from Ser to Gly due to editing. (D) Immunofluorescence analysis on olfactory bulb coronal sections, specifically localizing Unc80 (green) and NeuN (red) protein. The a and b correspond to magnified views of the white dashed boxes in left panel. The arrows indicate the positions of overlapping fluorescence (green and red). Scale bars = 200 µm in the left panel and 40 µm in the magnified images. (TA: Tibialis anterior muscle, GA: Gastrocnemius muscle, CT: cortex, CB: cerebellum, HP: hippocampus, OB: olfactory bulb).

Figure 2

Generation of Unc80-deleted and site-specific RNA editing mouse models: (A) Experimental schematic of the CRISPR/Cas9-based genetic engineering to generate deficient and knock-in mouse models. gRNA (nucleotide with orange background) together with Cas9 created indels, thus establishing knockout mice. In parallel, the addition of synthesized homologous DNA templates corresponding to the Unc80 sequence region with a substitution for the “pre-edited” (G-form) allele (gain of editing) or the “unedited” version (loss of editing) resulted in the knock-in models. (B) Breeding schemes of mice with site-specific knock-ins at the Unc80 editing site. (C) Genomic DNA sequencing for genotyping demonstrates the WT sequence (top) and mutations in founder strains (Unc80^+/−). Deletions and point mutations are indicated by a red background and purple boxes, respectively, in heterozygous deletion (Unc80^+/−) or knock-in mice (Unc80^S/S, Unc80^G/G, and Unc80^S/G). The green box denotes the targeted editing site. (D) Gross morphological comparison and body weights of newborn (P0) mice across genotypes. (E) Offspring genotyping from heterozygous Unc80^+/− crosses reveals a lower-than-expected birth rate for homozygous deletion offspring, whereas knock-in alleles followed an approximately Mendelian inheritance pattern. (F,G) Unc80 protein and mRNA levels in brain tissue lysates (cortex and olfactory bulb) from CRISPR-engineered Unc80-deficient mice were assessed via immunoblotting (F) and qRT-PCR (OB, (G)).

Figure 3

MEMRI-based neuronal activity assessment in the olfactory systems of Unc80^S/S and Unc80^G/G mice: (A) Anatomical MRI images and corresponding color mapping generated during odor stimulation. (B) Variation in MEMRI signal intensity in response to odor stimulation, with error bars representing the SD. (C) In vivo glutamate- and dopamine-sensitive CEST-MRI images showing anatomical and color mapping in various brain nuclei. (D) Quantitative changes in CEST-MRI signals across different brain nuclei for Unc80^S/S and Unc80^G/G mice, depicting both glutamate (left) and dopamine (right) contrasts. Brain regions assessed include the olfactory bulb (OB), anterior olfactory nucleus (AON), and anterior piriform cortex (APC). Error bars indicate SD. Statistical significance is denoted as follows: ns (not significant); p > 0.05; * p < 0.05; ** p < 0.01.

Figure 4

Neurophysiological connection of Unc80 editing event to olfactory sensing and motor control: Habituation and dishabituation behaviors, in response to odors, were analyzed for Unc80 knockout and site-specific editing variants (A,B), along with odor-induced neuronal activity assays (C,D). Mice with different genotypes (wild-type vs. Unc80^+/− in (A), Unc80^S/S vs. Unc80^G/G in (B)) were exposed to odors, and their explorative times near the odor source were recorded and presented as mean ± SD. The cohorts consisted of: WT (n = 13), Unc80^+/− (n = 5), Unc80^S/S (n = 9), Unc80^G/G (n = 8). For the neuronal activation assay, knock-in mice with site-specific edits were either not exposed or exposed to banana oil for up to 60 min and subsequently sacrificed for olfactory bulb isolation, from which total RNA and proteins were prepared for qRT-PCR (C) and Western blot (D) analyses, respectively. Changes in the expression of c-Fos were monitored as a readout for neuronal activation. The bar graph represents the relative mRNA expression levels of Fos. Statistical significance is indicated as follows: ns (not significant); p > 0.05; * p < 0.05; ** p < 0.01; *** p < 0.001.

Figure 5

Transcriptomic analysis of the olfactory bulb in Unc80 editing variant mice. Transcriptome-wide RNA-seq was conducted to identify changes in the olfactory bulb of Unc80^S/S and Unc80^G/G mice. (A–C) The overall transcriptome distribution is represented in a principal component analysis (PCA) plot (A) and a volcano plot (B), delineating genotype-specific gene expression profiles. A heatmap (C) displays genes with significant differential expression (|fold-change| > 1.5, p < 0.05; n = 6 per genotype, with samples pooled from three mice each) between the two strains. (D) Bubble chart of the enriched canonical pathway from IPA analysis. The x-axis represents the significance (p-value), and the y-axis shows the top 17 significantly enriched pathways based on differentially expressed genes (DEGs). The size of the circles corresponds to the number of genes associated with each pathway. The colors of the bubbles denote the Z-score, indicating whether the pathway is activated (orange, positive Z-score) or inhibited (blue, negative Z-score). (E) Normalized read count plots highlight expression variations in gene sets linked to neuronal signaling between Unc80^S/S and Unc80^G/G mice. Statistical significance in this figure is indicated as follows: * p < 0.05; ** p < 0.01; *** p < 0.001.