MicroRNA‐183‐5p is stress‐inducible and protects neurons against cell death in amyotrophic lateral sclerosis

Abstract Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease characterized by the death of motor neurons. A fundamental pathogenesis of ALS is the prolonged cell stress in neurons, which is caused by either accumulation of protein aggregates or reactive oxygen species. However, the mechanistic link between stress sensing and cell death is unsettled. Here, we identify that miR‐183‐5p, a neuron‐enriched miRNA, couples stress sensing and cell death programming in ALS. miR‐183‐5p is immediately induced by hydrogen peroxide, tunicamycin or TNF‐α in neurons. The overexpression of miR‐183‐5p increases neuron survival under stress conditions, whereas its knockdown causes neuron death. miR‐183‐5p coordinates apoptosis and necroptosis pathways by directly targeting PDCD4 and RIPK3, and thus protects neurons against cell death under stress conditions. The consistent reduction of miR‐183‐5p in ALS patients and mouse models enhances the notion that miR‐183‐5p is a central regulator of motor neuron survival under stress conditions. Our study supplements current understanding of the mechanistic link between cell stress and death/survival, and provides novel targets for clinical interventions of ALS.


| INTRODUC TI ON
Amyotrophic lateral sclerosis (ALS) is a fatal motor neuron disease characterized by the progressive degeneration of motor neurons in the brain and spinal cord. 1,2 A pathological hallmark of motor neuron degeneration in ALS is the accumulation of aggregated proteins and concomitant endoplasmic reticulum stress (ER stress) and oxidative stress. 3,4 Assessment of distinct cell stress levels is potential to be biomarkers for the diagnosis of ALS. 5,6 Beyond diagnostic application, mechanistic studies on stress response in ALS also provide therapeutic targets. 3,7,8 Thus, investigations into the diverse mechanisms of neuronal response to cellular stress are beneficial to identify therapeutic targets for ALS.
In the central nervous system, neuron is highly activated and faces many energetic challenges. Therefore, neuron is frequently exposed to stressful environments and particularly susceptible to stress-induced damages. 9,10 Neuronal intrinsic and environment-induced cellular stress causes a broad spectrum of structural and functional changes, such as protein aggregation, mitochondrial dysfunction and programmed cell death, which are shared in neurodegenerative diseases including ALS. 11,12 Recent discoveries shed light on critical functions of stress response in neurodegeneration, with an emerging realization that failure of stress response in neurons influences the initiation and progression of neurodegenerations disorders. 13,14 Neuronal stress response relies on a complicated network of stress response pathways, including autophagy, unfolded protein response and multiple signalling pathways. 14 Emerging evidence establishes that perturbations in non-coding RNAs (ncRNAs) are involved in stress sensing and neuron death/survival. In particular, microR-NAs (miRNAs) that regulate gene expression either by translational inhibition or targeted mRNA cleavage 15,16 are extensively altered in neurodegenerative diseases. 17,18 The expression of miRNA is dynamically regulated, and thus miRNA functions in stress responses. 19 Therefore, miRNA is positioned at the coupling of cell stress sensing and response. miR-183-5p, together with miR-96-5p and miR-182-5p, comprises the miR-183/96/182 cluster, which is expressed in neural cells and involved in long-term memory 20 and neuropathic pain. 21 Here, we report that miR-183-5p is highly enriched in motor neurons and coordinates stress sensing and cell death in ALS.

| Cell culture
Cortical neuron cultures were based on the original methods described by Kaech and Bankerde. 22 In brief, cerebral cortex from mouse embryos of 17.5 days was dissected and digested with Papain (Worthington) and DNAse (Sigma) at 37°C for 15 minutes. The dissociated neurons were plated in 6-well culture plates with a density of 1 × 10 6 cells per well for further experiments. 23 Astroglia were cultured from cerebral cortex from newborn mouse as described. 22 Briefly, cerebral cortex was dissected free of meninges and digested with 0.25% trypsin (Gibco) at 37°C. Then, the tissue pieces were dissociated and planted in 75 cm 2 in MEM (Gibco) supplemented with 10% Horse Serum (Hyclone), 1% Glutamine (Invitrogen) and antibiotics (Gibco

| RNA extraction and RT-qPCR assay
Total RNAs, including miRNA, were extracted and collected using TRIzol. The RNA quality and quantity were measured by NanoDrop 2000c (Thermo). The reverse-transcription PCR and qPCR for miRNA were performed following standard protocols as described previously. 25 The expression of miRNA was normalized to U6, quantified by the 2 −ΔΔCt method.

| Fluorescence in situ hybridization
Fluorescence in situ hybridization was carried out as previously described. 26 Briefly, the frozen tissue sections were thawed, then digested with pepsin and hybridized with a biotin-labelled probe corresponding to mature miR-183-5p or negative control. Finally, DAPI was used to stain the nuclei. Images were acquired using a confocal laser-scanning microscope (ZEISS 880+ Airyscan).

| Cell death assay
The cell death was examined by flow cytometry using Annexin V-APC/7-AAD kit (4A Biotech) according to the manufacturer's protocol. In brief, NSC-34 cells were transfected with miR-183-5p inhibitors, mimics or scramble sequence (100 nmol/L) for 24 hours and then treated with H 2 O 2 200 μmol/L for 2 hours or tunicamycin 1 μg/ F I G U R E 1 miR-183-5p is enriched in neurons and dynamically altered in progression of ALS mice. A and B, RT-qPCR shows the expression pattern of miR-183-5p in multiple organs (A) and different part of central nervous system (B). Data represent mean ± SEM. *P < .05, **P < .01 and ***P < .001, n = 6. C, RT-qPCR result indicates the expression of miR-183-5p in neurons, astrocytes and oligodendrocytes. Data represent mean ± SEM. *P < .05, **P < .01 and ***P < .001, n = 3. D, RT-qPCR shows the miR-183-5p expression in the primary neurons during the formation of neuronal network. Data represent mean ± SEM. *P < .05, **P < .01 and ***P < .001, n = 3. E, Fluorescent in situ hybridization (FISH) analysis of miR-183-5p expression in the grey matter of the anterior horn of spinal cord. Scale bar 25 μm. F and G, RT-qPCR results shows the dynamic changes of miR-183-5p in the lumbar spinal cord (F) and motor cortex (G). Pre-sym: pre-symptomatic stage, Early-sym: early-symptomatic stage, Late-sym: late-symptomatic stage. Data represent mean ± SEM. *P < .05, **P < .01 and ***P < .001. n = 10 | 8617 LI et aL. mL for 2 hours or TNF-α 100 ng/mL for 24 hours, respectively. The NSC-34 cells were stained for 20 minutes in dark and then analysed by flow cytometry.

| Dual-luciferase reporter assay
To examine the interaction between miR-183-5p and PDCD4 or RIPK3, plasmids of wild type or mutant sequence of 3′UTR of PDCD4 or RIPK3 and miR-183-5p mimics or negative control (miR-NC) were cotransfected to NSC-34 cells by Lipofectamine 2000 (Thermo). 48 hours after transfection, the cells were harvested and the luciferase activities were measured. The relative Renilla luciferase activity was normalized to firefly luciferase activity.

| Protein extraction and western blots
Protein extraction and Western blots were performed as previously reported. 25 In brief, equivalent proteins in each group were isolated using SDS-PAGE and transferred to PVDF membranes (Millipore).
Subsequently, the membranes were blocked with 5% skim milk on a rocker and then incubated with primary antibodies at 4°C overnight, followed by appropriate second antibody (Abcam) for 2 hours.
Finally, the membranes were treated with ECL reagent (Thermo Fisher Scientific) for exposure. The ImageJ software was used for optical den-

| SOD1 G93A transgenic mice
SOD1 G93A transgenic mice were purchased from the Jackson Laboratory. Mice were monitored weekly and staged according to previously report. 31 The pre-symptomatic stage (Pre-sym), earlysymptomatic stage (Early-sym) and late-symptomatic (Late-sym) SOD1 G93A mice were used for experiments and gender-matched littermates were used as controls. Mouse experiments were conducted in compliance with the Laboratory Animal Care Guidelines authorized by Sichuan University.

| Statistical analysis
GraphPad Prism 7.0 software was used for statistical analysis. All quantitative data were shown as the mean ± SEM unless otherwise stated. Significant statistical differences between the two groups were determined by the two-tailed unpaired Student's t test and for three or more groups, one-way ANOVA (followed by a Dunnett post hoc test) was used for statistical analysis. The P-value < .05 was assumed as statistically significant.

| miR-183-5p is enriched in neurons of the spinal cord and dynamically regulated in ALS mice
As the first step towards understanding the role of miR-183-5p in neural systems, we performed reverse transcription quantitative PCR (RT-qPCR) to investigate miR-183-5p expression pattern in different organs of mouse. Results showed that miR-183-5p was generally expressed in multiple organs/tissues, particularly in neural tissues, such as spinal cord and brain ( Figure 1A). It's noted that miR-183-5p was extremely enriched in the spinal cord. Thus, we studied the neural expression of miR-183-5p and found that miR-183-5p was highly expressed in the lumbar, thoracic and cervical spinal cord, but lower in the hippocampus and cerebellum ( Figure 1B). By primary cultures of neurons, astrocytes and oligodendrocytes, we found that neuronal expression of miR-183-5p was much higher than its glial counterparts ( Figure 1C). The expression of miR-183-5p was increased during the formation of neuronal network ( Figure 1D Figure 1F). However, the expression of miR-183-5p was not dramatically altered in the mouse cortex ( Figure 1G). All data reveal the neuronal enrichment of miR-183-5p in the spinal cord, implying its potential role in ALS.

| miR-183-5p expression is dynamically regulated by cell stress
To better understand the function of miR-183-5p in neurons, we performed bioinformatic prediction to screen the downstream genes targeted by miR-183-5p. By screening databases of micro-RNA.org, 27 Targetscan, 28 Diana Tools 29 and miRDB, 30 we found a total of 399 potential genes targeted by miR-183-5p in at least two databases, which were concentrated in the pathways of cell stress responses and cell death by GO (Gene Ontology) analysis ( Figure 2A-C). Thus, we next examined the expression of miR-183-5p in response to cellular stress. Neuronal miR-183-5p was induced by H 2 O 2 that evokes oxidative stress, and this induction was recapitulated in NSC-34 and Neuro2a cells ( Figure 2D,E). It's also noticed that the inductive miR-183-5p by oxidative stress was quickly increased within 30 minutes after H 2 O 2 treatment.
To further investigate the dynamic regulation of miR-183-5p, we next examined its response to ER stress. Results showed that neuronal miR-183-5p was increased in close temporal association with ER stress occurring by tunicamycin (TM) ( Figure 2F). Similarly, the inductive miR-183-5p to ER stress in neurons was reproduced in NSC-34 and Neuro2a cells ( Figure 2G). Neural inflammation activates cell stress and causes neuronal death and degeneration.
We wondered whether miR-183-5p was dynamically controlled by external inflammatory cytokines. By TNF-α treatment, we found that miR-183-5p was increased in neuron cultures as well as NSC-34 and Neuro2a cells ( Figure 2H,I). Taken together, our data indicate the dynamic regulation of miR-183-5p by cell stress in neurons.

| miR-183-5p protects against cell death under stress conditions
Considering the neuronal enrichment and stress-responsive regulation of miR-183-5p, we hypothesized that miR-183-5p might play a role in protecting neurons against cell stress. We knocked down or

| miR-183-5p targets RIPK3 and PDCD4 to regulate cell death
To clarify how miR-183-5p participates in regulation of neuronal survival/death, we next examined the potential targets of miR-183-5p. By sequence analysis, we found that RIPK3, a well-known regulator of necroptosis, was a potential candidate of miR-183-5p in NSC-34 cells ( Figure 4A). Results of luciferase reporter assay confirmed the binding of miR-183-5p with RIPK3 ( Figure 4B). To further validate the regulatory effect of miR-183-5p on RIPK3, we investigated RIPK3 protein level in NSC-34 cells. Western blots showed that the protein level of RIPK3 was decreased by miR-183-5p overexpression and increased by its knockdown (Figure 4C,D). The consistent alternations of phospho-MLKL confirmed the effect of miR-183-5p on RIPK3.
Moreover, we found that miR-183-5p also binds to PDCD4, a critical protein in cell apoptosis ( Figure 4E). Similarly, luciferase reporter assay ( Figure 4F) and Western blots ( Figure 4G,H) demonstrated that miR-183-5p directly bound to PDCD4 and negatively regulated the expression of PDCD4. Furthermore, the similar change patterns of cleaved caspase-3 confirmed the effect of miR-183-5p on PDCD4. Therefore, we propose that miR-183-5p targets both RIPK3 and PDCD4 and coordinates necroptosis and apoptosis pathways to control neuronal survival/death.

| D ISCUSS I ON
Emerging evidence establishes that neuronal stress responses are essential in neurodegenerative disease. 14 The complicated relationship among stress sensing and responses, adaptive outcomes and F I G U R E 3 miR-183-5p desensitizes cell to death under stress conditions. A-D, Flow cytometry assay shows that miR-183-5p mimics protect cells from cell death while inhibitors induce cell death at basal (A) and TNF-α (B), H 2 O 2 (C) and tunicamycin (TM) (D) treatment conditions. E and F, The total cell death is statistically analysed in inhibitors (E) and mimics (F) respectively. Data represent mean ± SEM. *P < .05, **P < .01 and ***P < .001, n = 3 neurodegeneration creates challenges for understanding the mechanistic interplay. In this study, we demonstrate that miR-183-5p, a neuron-enriched miRNA, plays a vital role in the coupling of stress sensing and scavenging. miR-183-5p is highly expressed in neurons of the central nervous system and immediately induced in response to oxidative stress and ER stress. Elevation of miR-183-5p increases  [32][33][34] Moreover, miRNAs are involved in pathways that regulate redox biology. 35,36 In this study, we propose that miR-183-5p is a protective factor in neuron survival. The fact that miR-183-5p is stress-inducible indicates miR-183-5p is an immediately response factor to neuronal stress. Moreover, miR-183-5p suppresses neuronal death by controlling two distinct cell death pathways. Therefore, miR-183-5p coordinates stress sensing and responding in neurons and is critical for motor neuron survival under stress conditions. So far, multiple gene mutations of ALS, such as SOD1, TARDBP, FUS and TBK1, have been identified. 37 Regardless of how the primary gene defects are categorized, most of them are associated with common downstream pathologic processes, such as oxidative stress, ER stress and inflammatory stress, which finally cause neuronal death through apoptosis and necroptosis. 38 However, the mechanistic link between the neuronal stress and death is not fully understood. Our results reveal that miR-183-5p is not only a stress sensor in motor neurons, but also an executive factor in neuron death programming. miR-183-5p modulates cell apoptosis by targeting PDCD4 and necroptosis by RIPK3. In ALS, necroptosis is appreciated as a major pathway for motor neuron death. 39 Therefore, miR-183-5p is critical for motor neuron survival and increased expression of miR-183-5p may be beneficial for motor neuron survival in ALS.
Our previous study investigated the miRNA expression profiles of Chinese ALS patients to explore novel biomarkers for ALS diagnosis. We found that miR-183-5p was down-regulated in ALS patients and provided high diagnostic accuracy for ALS. 40 In this study, we examined miR-183-5p expression pattern in the progression of ALS mouse. Intriguingly, we found that the increase of miR-183-5p is correlated with cell stress in motor neurons of ALS in pre-symptomatic and early-symptomatic stages, while in the late-symptomatic stage miR-183-5p was decreased, possibly due to the progressive motor neuron degeneration/death. 41 This finding is similar to our previous work, demonstrating the dynamic changes of a neuronal protective gene LanCL1 in the progression of ALS. 23 Therefore, the dynamic expression of miR-183-5p affords a novel diagnostic and therapeutic target for ALS.
In conclusion, we report that miR-183-5p, a neuronal enriched miRNA, is stress-inducible and protects neurons against cell death.
Our study supplements current understandings of the mechanistic link between cell stress and cell death/survival and provides novel targets for ALS interventions.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.