Preservation of miR-9-5p and miR-124-3p in ALS-resistant oculomotor neurons contrasts with their downregulation in vulnerable spinal motor neurons, irrespective of TDP-43 pathology

doi:10.21203/rs.3.rs-7537447/v1

Download PDF

Research Article

Preservation of miR-9-5p and miR-124-3p in ALS-resistant oculomotor neurons contrasts with their downregulation in vulnerable spinal motor neurons, irrespective of TDP-43 pathology

https://doi.org/10.21203/rs.3.rs-7537447/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Selective vulnerability of motor neurons is a defining feature of amyotrophic lateral sclerosis (ALS) and provides a valuable framework for uncovering mechanisms that distinguish resilient from vulnerable neuronal populations. We investigated whether dysregulation of neuroprotective microRNAs (miRNAs), miR-9-5p and miR-124-3p, contributes to the differential susceptibility of motor neuron subtypes. We focused on spinal motor neurons (SMNs), which undergo drastic degeneration in ALS, and oculomotor neurons (OMNs), which remain functionally intact and rarely degenerate, allowing preservation of eye movement in ALS patients. Using a modified multiplexed fluorescent in situ hybridization protocol combined with immunofluorescence, we quantified the expression of miR-9-5p and miR-124-3p in cervical SMNs and OMNs from ALS and control cases. We observed significant downregulation of both miRNAs in ALS SMNs, while their expression was maintained in ALS OMNs. Stratification of ALS SMNs by TDP-43 pathological status revealed similarly reduced miRNA expression in neurons with and without cytoplasmic inclusions, suggesting that miRNA downregulation occurs independently of visible TDP-43 pathology. To explore a potential mechanism, we assessed the localization of the Dicer cofactor TRBP and found that it was sequestered into TDP-43 inclusions in ALS SMNs. However, TRBP remained normally localized in neurons without cytoplasmic inclusions, indicating that sequestration cannot fully account for miRNA reduction across all ALS motor neurons. These findings support a model in which early or subtle disruptions, preceding visible pathology, also contribute to miRNA downregulation in ALS. Our study also indicates cell-type-specific preservation of miRNA networks as a possible contributor to oculomotor neuron resilience.

Amyotrophic lateral sclerosis

Selective motor neuron vulnerability

miR-9-5p

miR-124-3p

TDP-43

Fluorescent in situ hybridization

Amyotrophic lateral sclerosis (ALS) is characterized by progressive loss of motor neurons resulting in paralysis and death, yet some motor neuron subtypes remarkably remain resistant[31]. Spinal motor neurons (SMNs) undergo early and severe degeneration, while other populations, such as oculomotor neurons (OMNs) that control eye movement, are largely spared [17, 23]. Post-mortem analysis shows that ALS OMNs typically retain nuclear expression of transactive response DNA-binding protein 43 kDa (TDP-43), an RNA-binding protein strongly associated with ALS pathology, whereas SMNs commonly exhibit TDP-43 mislocalization to the cytoplasm and subsequent formation of biomolecular condensates[4, 8, 42]. Features such as unique cellular architecture, distinct gene expression, specialized metabolism, and reduced glutamate excitotoxicity are suspected to contribute to the resilience of OMNs[3, 5, 11, 20, 27, 40]. Elucidating these and other novel mechanisms that confer protection could help with the development of targeted neuroprotective approaches in ALS.

Dysregulated RNA metabolism is increasingly recognized as an important pathological feature of ALS, and microRNAs (miRNAs) have emerged as both dysregulated molecules and promising biomarkers for disease progression[9, 10, 14, 15, 21, 36]. miRNAs are short (~22 nt) non-coding RNAs that fine-tune gene expression post-transcriptionally by guiding the RNA-induced silencing complex (RISC) to target mRNAs, leading to translational repression or degradation. miRNA production begins in the nucleus, where Drosha and DGCR8 cleave primary transcripts into precursor miRNAs, which are exported to the cytoplasm and further processed by Dicer, aided by TAR RNA-binding protein (TRBP) and PACT. The guide strand is incorporated into RISC, where Argonaute 2 (AGO2) directs it to target mRNAs for translational repression or degradation. Because of their stability and cell-specificity, miRNAs are useful for understanding ALS pathogenesis and providing candidate biomarkers.

Several studies have reported a global downregulation of miRNAs in ALS spinal cord tissue homogenates and lumbar motor neurons, as demonstrated by in situ hybridization, qPCR, and RNA sequencing[9, 13–15, 37]. One proposed mechanism for this downregulation involves impaired Drosha/DGCR8 and Dicer/TRBP function downstream of ALS-causing mutations and cellular stress. In cell culture stress assays, AGO2 was also shown to interact with the stress granule (SG) protein TIA-1, linking miRNA machinery to SG dynamics[14]. Alternatively, other evidence from cell culture suggests that TDP-43 mislocalization, triggered by cellular stress, can disrupt a negative feedback loop involving specific miRNAs, where loss of nuclear TDP-43 leads to decreased expression of these miRNAs in the cytoplasm[19].

miR-9-5p and miR-124-3p have been reported to exhibit reduced expression in spinal cords of individuals with ALS[9, 12, 14, 15, 18, 46]. Targets of both miRNAs are implicated in neuronal development and function, including neurogenesis, cell fate specification, synaptic remodeling, and structural maturation of axons and dendrites, indicating their importance in neuronal integrity[18, 22, 38, 45]. This functional relevance may explain why SMNs are particularly susceptible to ALS-related pathology. However, miRNA expression profiling in ALS-resistant motor neurons, such as OMNs, is lacking. Since OMNs usually maintain nuclear TDP-43 throughout the disease course, exhibit resistance to degeneration compared to SMNs, and miR-9-5p and miR-124-3p targets are responsible for neuronal maintenance, we hypothesized that OMNs and SMNs from ALS patients would differ in their expression of miR-9-5p and miR-124-3p compared to control cases. We assessed the expression of miR-9-5p and miR-124-3p in postmortem sporadic ALS and control tissue using fluorescent in situ hybridization, targeting both OMNs and SMNs. Immunofluorescence (IF) for TDP-43 was included to evaluate miRNA expression and cellular pathology. This approach enabled direct comparison of miRNA expression in vulnerable and resistant motor neuron populations.

Patient tissue

Fixed tissue sections of cervical spinal cord and superior midbrain were used for fluorescent in situ hybridization and immunofluorescence (FISH-IF) experiments. Formalin-fixed, post-mortem tissue control patients or patients diagnosed with sporadic ALS according to El Escorial criteria (World Federation of Neurology) by MJS were processed, embedded in paraffin, sectioned at 7 um thickness and mounted on glass slides[6]. Basic demographic information of the patients is provided in Supplementary file1: Table S1. The cervical spinal cord was selected for analysis because it is significantly affected in ALS while still retaining a substantial population of observable motor neurons. SMNs of the ventral horn were identified in the anterior region of the spinal cord sections by the larger size of the ventral horn compared to the dorsal horn and the presence of prominent motor neuron cell bodies. The oculomotor nucleus was identified anatomically in midbrain sections at the level of the superior colliculus paired posteromedial nuclei ventral to the periaqueductal gray matter, medial to the median longitudinal fasciculus. and from which ventrally exiting fascicles are present. In several cases, the midbrain was initially cut on the mid-sagittal plane and sectioned into halves prior to cutting and mounting for examination.

Fluorescent in situ hybridization and Immunofluorescence

Double fluorescent in situ hybridization and immunofluorescence (dFISH-IF) was used to detect target miRNA miR-9-5p or miR-124-3p with candidate normalizers miR-132-3p or U6 small nuclear RNA (U6 snRNA) and TDP-43 in motor neurons, along with the motor neuron marker NeuroTrace435/455 (Nissl-staining substance; does not stain nucleus). U6 snRNA was used previously for normalization in control and ALS lumbar spinal cord in situ experiments[14]. FISH-IF was performed similarly to a published protocol and is briefly outlined here with modifications[34]. Tissue sections were dewaxed with xylenes and rehydrated with a series of ethanol dilutions. Antigen retrieval was performed in 10 mM sodium citrate, pH 6.0, using a pressure cooker for 15 minutes at high pressure at 100°C. dFISH was performed with inventoried digoxigenin (DIG) or 6-carboxyfluorescein (FAM) labelled mirCURY LNA™ miRNA detection probes (GeneGlobe, Qiagen; for probe details see Supplementary file1: Table S2). Two differently labeled LNA probes (DIG or FAM) with similar hybridization temperatures (within 10°C) were added to tissue sections, and each was detected by sequential HRP antibody-tyramide signal amplification (TSA; Cyanine 3 or 5, Akoya). Triple FISH (tFISH) was performed similarly to the above method, followed by sequential hybridization and detection of the third probe. Immunofluorescence of the same slides was performed using an anti-TDP-43 antibody (ProteinTech) at 1:150 in 1xPBS with 2% BSA (bovine serum albumin) after blocking in 1xPBS with 5% BSA and incubated overnight at 4°C (Supplementary file1: Table S2). An Alexa Fluor 488-conjugated secondary antibody (Invitrogen) at 1:250 in 1xPBS with 2% BSA was used to detect the TDP-43 antibody. NeuroTrace 435/455 (Invitrogen) was used to label neurons as directed by the manufacturer, except permeabilization buffer consisted of 0.3% TritonX-100, and NeuroTrace was diluted 1:200 before addition to sections. Tissue sections were mounted with Dako Fluorescence Mounting Medium (Agilent).

For immunofluorescence without in situ hybridization, we examined 4 control cases and 5 ALS cases. The same procedure described above was performed, except blocking and antibodies (anti-TDP-43 mouse antibody at 1:200 and anti-TRBP rabbit antibody at 1:100) were added directly after antigen retrieval. NeuroTrace 435/455 staining and mounting were then performed as described above. using immunofluorescence with antibodies to TDP-43 and TRBP.

Confocal microscopy

Images of dFISH-IF experiments performed for quantification were captured with Stellaris 5 or SP8 confocal microscopes and Leica Application Suite-X (LAS-X) software (Leica Microsystems). Navigator was used to acquire tiled images of the tissue sections with a 20x objective. Fluorescent dyes detected were NeuroTrace435/455 (Nissl-staining substance), Alexa Fluor 488 (TDP-43), Cyanine 3 (target miRNAs), and Cyanine 5 (reference RNA). A narrow detection window was used to detect each fluorophore, and no crosstalk was observed. For each experiment, confocal settings were optimized to the slide containing the highest motor neuron RNA fluorescent signal, such that no saturation was detected in the image for each fluorophore to be quantified (miRNA and U6 snRNA) and applied equally to all other slides during image acquisition. Representative dFISH-IF images for publication were taken with a Leica SP8 confocal microscope using a 63x objective with 1.4 NA and deconvolved by Lightning software (Leica Microsystems).

Quantification

Tiled images were opened in Fiji-ImageJ and regions of interest were drawn freehand around individual motor neurons[39]. NeuroTrace 435/455 staining, cellular anatomy, and motor neuron size/morphology were used as a combined reference to select motor neurons. The number of cases used per experiment ranged from 3 to 5 for controls and 8-11 for ALS (Supplementary file1: Table S1). The number of quantified motor neurons was typically greater than 50 per case slide. All observable motor neurons that fit the criteria were selected for quantification. Fluorescence representing miRNA expression per cell was measured as integrated fluorescence density. Target miRNA signal was either corrected to background fluorescence or corrected and then normalized to another cellular RNA species for expression[28]. MiR-132-3p in midbrain sections or U6 snRNA in cervical spinal cord sections were evaluated as candidate reference species by comparing corrected expression between control and ALS in two experiments. Cervical SMNs quantified in ALS cases were subdivided into three TDP-43-pathological subgroups based on the TDP-43 immunofluorescence pattern. Cells with an observable nucleus (based on lack of Nissl staining and U6 snRNA staining) were sorted into “ALS_nuclear” (largely nuclear TDP-43), “ALS_cyto no inclusions” (largely cytoplasmic TDP-43 without inclusions), or “ALS_cyto with inclusions” (largely cytoplasmic TDP-43 with inclusions) subtypes. At this magnification (20x objective) fine or subtle inclusions could not be reliably resolved, therefore a conservative approach was adopted whereby only clearly discernible inclusions were recorded in the “with inclusions” subtype and the remaining motor neurons with nuclear loss were included in “no inclusions”.

Data preparation and statistical analysis

Separate mixed-effects models were fit to each experiment to compare ALS and control groups for normalized and corrected RNA expression. This analysis compares RNA expression outcomes between ALS and control cases while accounting for patient-level variability, which was apparent from the data and confirmed in the analysis[1, 2]. Pairwise comparisons between ALS motor neuron subtypes were performed using Wald Z-tests on fixed effects from the mixed-effects models.

Outliers were removed prior to analysis using GraphPad Prism’s ROUT method (Q = 1%). Because preliminary modeling suggested non-normality and unequal variance for some datasets, all data were log-transformed to stabilize variance; this also enabled interpretation as a percent change in ALS versus controls. A constant offset (|minimum|+1) was added where values were ≤0. Linear mixed-effects models were fitted with disease status as the fixed effect (control as reference) and a random intercept for each patient to account for repeated measures (non-independent observations). Predicted means were back-transformed to compute percent differences, and 95% confidence intervals were obtained by cluster bootstrapping (1,000 iterations). Significance was set at 0.05, and p-values were corrected within each miRNA-tissue family using the Benjamini-Yekutieli procedure to control false discovery rate under correlation (i.e. corrected miRNA expression comparisons are correlated with normalized miRNA comparisons).

To visualize distributions of miRNA expression in motor neuron subpopulations by TDP-43 pathology, overlaid histograms and kernel density estimates (KDEs) were generated. Histograms were normalized using density scaling to allow direct comparison despite unequal group sizes, and smoothed KDE curves were plotted to show overall distribution shapes.

All data processing and analysis were performed in Python 3.13.3[35]. The pandas package (2.2.3) was used for data manipulation and handling of tabular datasets, while numpy (2.2.4) was employed for numerical operations. Statistical modeling was conducted using the statsmodels package (0.14.4). The scipy library (1.15.2) was used for statistical functions and residual diagnostics. Data visualization was carried out using matplotlib (3.10.1) and seaborn (0.13.2). File path operations were managed using Python's built-in os module.

Multiplex dFISH-IF allows co-detection of miRNAs, reference RNAs, and TDP-43
We developed a multiplexed fluorescent in situ hybridization method for use on human postmortem tissue to detect multiple low-abundance RNA targets and proteins. This approach allowed simultaneous visualization of up to three RNA species in the same tissue section (Supplementary file1: Fig S1). We applied a modified version of this method to profile miR-9-5p or miR-124-3p expression simultaneously with TDP-43 localization. We successfully detected miRNA, reference RNA, and TDP-43 in all cases analyzed (dFISH-IF).

miR-132-3p was selected as a candidate reference RNA for OMNs based on small-scale preliminary testing that showed little difference in its expression between ALS and control cases and because it produced a consistently strong fluorescent signal in this tissue with little background (data not shown). Preliminary testing also indicated that miR-132-3p was reduced in ALS SMNs thus we looked for a substitute reference for spinal cord comparisons (data not shown). Because U6 snRNA has previously been used to normalize these miRNAs in published FISH studies, we adopted this approach for normalization of miRNA expression in spinal cord. Candidate reference RNAs were validated for normalization based on corrected expression showing no significant difference between ALS and control tissues (Supplementary file1: Fig S2).

miR-9-5p and miR-124-3p expression are downregulated in ALS cervical spinal motor neurons

We used multiplexed dFISH-IF in cervical spinal cord tissue to assess whether the previously reported downregulation of miR-9-5p and miR-124-3p in ALS could be detected with this method. Representative confocal images of motor neurons found and analyzed by dFISH-IF in cervical spinal cord control and ALS cases are shown in Fig. 1. We observed that most ALS cases contained SMNs with TDP-43-positive cytoplasmic inclusions (or aggregates), as well as motor neurons with either nuclear TDP-43 localization or nuclear clearance of TDP-43 (Fig. 1). Subsequent quantification of normalized miRNA expression revealed that both miR-9-5p and miR-124-3p were reduced in the ALS group compared to control for SMNs (Fig. 1 and 4a, c, and e). miR-9-5p was reduced in ALS by 52.3% (95% CI: 39.0, 63.3; p = 0.00222), and miR-124-3p was reduced in ALS by 69.7% (95% CI: 47.1, 83.2; p = 0.00647) (Fig. 4a, c and e). Similar results were obtained with quantification data that was not normalized but background-corrected (Supplementary file1: Fig S3). These findings validate our dFISH-IF protocol as a robust method for detecting disease-associated changes in miRNA expression alongside TDP-43 patterns, and they align with previous studies reporting downregulation of these miRNAs in ALS.

miR-9-5p and miR-124-3p expression are preserved in ALS oculomotor neurons

We then extended the analysis to midbrain tissue to assess the levels of miR-9-5p and miR-124-3p in ALS OMNs. To assist with anatomical orientation, low-magnification images of the oculomotor nucleus from control and ALS cases illustrating miR-9-5p, miR-124-3p, and TDP-43 expression are shown in Fig. 2. Representative confocal images of individual OMNs analyzed by dFISH-IF from control and ALS cases at higher magnification are shown in Fig. 3. We did not detect cytoplasmic TDP-43 inclusions or loss of nuclear TDP-43 in OMNs from control or ALS cases (Fig. 3). Notably, neither miRNA showed a statistically significant difference between ALS and control OMNs. Quantification revealed that miR-9-5p expression showed a non-significant 90.5% increase compared to control (p = 0.657), while miR-124-3p showed a non-significant 42.9% decrease (p = 0.144) (Fig. 4b, d, and e). This indicates that both miRNAs show altered regulation in OMNs compared to SMNs. Given that downregulation of the target miRNAs was statistically significant in ALS SMNs, the lack of significant differences in OMNs between ALS and control is consistent with the hypothesis of regional differential regulation.

TRBP colocalizes with cytoplasmic TDP-43 inclusions in ALS cervical spinal motor neurons

We then asked whether the observed downregulation of mature miR-9-5p and miR-124-3p in ALS might result from sequestration of miRNA biogenesis factors, as proposed by Emde et al.[14] To directly assess disruption of the Dicer complex, we focused on TRBP and evaluated its subcellular localization with TDP-43. In all control cases, TDP-43 was detected in the nucleus, whereas TRBP was cytoplasmic, often spatially associated with the Nissl marker at the rough endoplasmic reticulum (Fig. 5). In ALS SMNs with nuclear TDP-43 and no cytoplasmic inclusions (ALS_nuclear), TRBP was also detected in the cytoplasm, with a similar pattern as control motor neurons (Fig. 5). ALS motor neurons with a complete cytoplasmic shift of TDP-43 from the nucleus to the cytoplasm, along with small or diffuse TDP-43-positive cytoplasmic inclusions (ALS_{cyto, diffuse}), showed TRBP colocalization with these inclusions (white arrows for examples, Fig. 5). The colocalization between TRBP and TDP-43 was most apparent in ALS motor neurons with overt cytoplasmic inclusions, either spheroid or filamentous/skein-like (white arrows for examples in ALS_{cyto, spheroid} or ALS_{cyto, skein}). We also detected this same pattern of colocalization with a second TDP-43 antibody and the same TRBP antibody using an immunofluorescence method to limit background signal and found the same colocalization pattern (Supplementary file1: Fig. S4). TDP-43-TRBP colocalization was also maintained throughout the motor neurons, as shown by a representative maximum projection of a z-stack acquisition from a SMN with a filamentous inclusion (Supplementary file1: Fig. S4b). These findings suggest that TRBP is found within TDP-43-specific cytoplasmic condensates, while its localization remains normal in ALS motor neurons without overt TDP-43 pathology, similar to representative control motor neurons.

miR-9-5p and miR-124-3p downregulation occurs irrespective of overt TDP-43 pathology

TDP-43 pathology in ALS exists along a spectrum, ranging from nuclear localization to complete nuclear clearance accompanied by cytoplasmic diffuse, spheroid, skein-like, or filamentous inclusions (Fig. 1). We observed motor neurons displaying these various subtypes of TDP-43 localization in the ALS cervical spinal cord tissues analyzed, but not within OMNs, where TDP-43 was nuclear. Because miRNA expression was preserved in OMNs without TDP-43 pathology but reduced in cervical spinal cord with TDP-43 pathology, we stratified ALS SMNs into subtypes based on TDP-43 localization (nuclear, cytoplasmic, or cytoplasmic with inclusions) to assess its influence on miRNA expression.

We compared background-corrected miRNA expression in control cases to three subpopulations of ALS motor neurons: those with 1) largely nuclear TDP-43 or 2) largely cytoplasmic TDP-43, which was then further divided into 2a) those without visible TDP-43 inclusions or 2b) those with visible TDP-43 inclusions. Control data and ALS subgroup data were selected from the same raw data used for Fig. 4a and c, except that background-corrected target miRNA expression was used, as initial observations suggested that ALS motor neurons with cytoplasmic inclusions often contained little reference RNA. miR-9-5p and miR-124-3p were downregulated in all three ALS subpopulations compared to control and all comparisons were statistically significant except for the miR-9-5p “ALS_cyto no inclusion” group (uncorrected p = 0.0370, corrected p = 0.188), possibly due to a small sample size (10 cases, 43 motor neurons) (Fig. 6). Downregulation in the total ALS SMN population quantified was 60.4% for miR-9-5p and 69.6% for miR-124-3p. Similarly, miR-9-5p was reduced by 57.7% in nuclear TDP-43, 49.7% in cytoplasmic TDP-43 without inclusions, and 61.6% in cytoplasmic TDP-43 with inclusions, while miR-124-3p was reduced by 69.0% in nuclear TDP-43, 60.7% in cytoplasmic TDP-43 without inclusions, and 67.6% in cytoplasmic TDP-43 with inclusions (Fig. 6a and c). These results indicate that SMNs without TDP-43 pathology show similarly reduced miR-9-5p and miR-124-3p expression as motor neurons with obvious evidence of TDP-43 cytoplasmic inclusions. Additionally, there was a similar distribution of miRNA expression for each pathological subgroup, supporting the statistical non-significance of the results (Fig. 5b and d). Together, these findings suggest that reduced miR-9-5p and miR-124-3p expression in ALS SMNs occurs independently of visible cytoplasmic TDP-43 aggregation.

We also compared miRNA expression changes among the ALS TDP-43 subtypes for each miRNA within the context of this analysis. Most subtype comparisons were not statistically significant for either miRNA, however miR-124-3p expression increased by 26.4% in ALS neurons with cytoplasmic TDP-43 mislocalization (no inclusions) compared to ALS motor neurons with nuclear TDP-43 (p = 0.0222) (Fig. 6c). For miR-9-5p, expression was 18.8% higher in the cytoplasmic group (no inclusions) relative to the nuclear group, however this difference was not statistically significant (p = 0.519) (Fig. 6a).

This work identifies two examples of miRNA dysregulation as cell-type-specific events in ALS disease progression, revealing a potential new mechanism of selective neuronal vulnerability. In agreement with previous studies, our profiling method revealed that miR-9-5p and miR-124-3p, two highly enriched neuronal miRNAs, were significantly downregulated in ALS cervical SMNs. Reduced miR-9-5p and miR-124-3p are predicted to derepress targets such as REST, MAP1B, PTBP1, and ROCK1, leading to impaired neuronal transcription, axonal instability, altered splicing, and disrupted mitochondrial function[33, 43, 44]. These pathways are validated in motor neurons and align with ALS phenotypes, including axonal transport defects, mitochondrial impairment, and abnormal neuron-glia interactions. In contrast, both miRNAs were unchanged in ALS OMNs, consistent with their resistance to degeneration and lack of TDP-43 pathology. The preservation of these miRNAs in OMNs likely maintains or enhances regulatory control over these disrupted pathways, conferring resilience against ALS-related stress. Although it is possible that maintained miRNA levels in OMNs reflect a downstream consequence of reduced pathological burden, the functional relevance of these specific miRNAs supports an active protective role. Further experimental validation is needed to confirm this and determine the specific pathways responsible for miRNA-mediated protection in OMNs.

miR-132-3p expression was stable between ALS and control OMNs. Although used here as a reference, this brain-enriched miRNA regulates cytoskeletal plasticity, mitochondrial dynamics, and pro-survival signaling, and its maturation is TDP-43-dependent[24]. Because it is predicted to be downregulated in ALS spinal cord, its preserved expression in OMNs may represent a third example of miRNA preservation in this resilient population[9, 16]. Future work will determine whether this preservation is specific to a subset of neuroprotective miRNAs or reflects broader maintenance of miRNA regulatory networks.

To explore a mechanism for miRNA downregulation in ALS, we examined TRBP localization with TDP-43. We hypothesized that impaired miRNA biogenesis upstream of RISC contributes to pathology, given prior evidence of AGO2 in SGs in cell models[14]. Because TDP-43 is an SG-associated RNA-binding protein in ALS, it provides a disease-relevant scaffold for assessing TRBP interactions in postmortem neurons. TRBP exhibited cytoplasmic localization in control SMNs and ALS SMNs where TDP-43 was nuclear, however, in ALS SMNs with cytoplasmic TDP-43 inclusions, TRBP colocalized with the inclusions. TRBP is a cytoplasmic double-stranded RNA-binding protein that modulates PKR-dependent SG formation, providing a mechanistic basis for its presence in protein-RNA assemblies that overlap with TDP-43 complexes.[32]. Although evidence from cell culture shows that TDP-43 and TRBP can co-exist in Dicer-containing complexes under normal conditions, TRBP is not recovered when immunoprecipitating TDP-43, indicating no stable association[24]. Thus, the co-localization we observed in ALS SMNs more likely reflects a pathological sequestration into condensates or SG-like assemblies. Together, these results suggest that chronic stress and aggregation may aberrantly incorporate miRNA biogenesis factors into persistent SG-like structures in ALS patients, leading to decreased miRNA expression.

The colocalization results implied that dysregulated miRNA biogenesis by sequestration would only apply to neurons with visible TDP-43 inclusions. Furthermore, loss of nuclear TDP-43 is known to impair processing of specific miRNAs and disrupt feedback loops that normally regulate TDP-43 via specific miRNAs, consequently decreasing miRNA expression[19, 24, 30, 46]. These findings provided a rationale for quantifying miR-9-5p and miR-124-3p expression in subgroups of SMNs stratified by TDP-43 pathology. We asked whether nuclear TDP-43-positive motor neurons retain higher miRNA expression compared with neurons showing cytoplasmic or aggregated TDP-43. Contrary to our expectation, we found equivalent downregulation across all TDP-43 pathology subtypes compared to control for both miRNAs, indicating that nuclear TDP-43 alone in ALS is insufficient to maintain normal levels of these miRNAs. This may be caused by early and possibly TDP-43-independent mechanisms affecting RNA regulatory pathways. It also indicates that cytoplasmic TDP-43 inclusions are not necessary for miRNA dysregulation, and that motor neurons without visible aggregates may still be functionally impaired.

These findings support a model in which miRNA downregulation occurs early in the cellular disease process, before the formation of visible TDP-43 inclusions. Even in the absence of visible TDP-43/TRBP inclusions, ALS motor neurons may experience early or subtle molecular disruptions that could impair miRNA biogenesis. Soluble mislocalized TDP-43, low-level cellular stress, or transient stress responses, such as early-stage granule formation, could be sufficient to disrupt Dicer complex function or engage alternative pathways that reduce mature miRNA levels, even in the absence of histologically detectable pathology at our current resolution.

Retention of nuclear TDP-43 does not preclude SMNs from being actively engaged in the disease process. Identifying the molecular changes that precede or contribute to TDP-43 mislocalization is a critical area of ongoing research. Studies have demonstrated that nuclear TDP-43 can exhibit early conformational pathology coinciding with STMN2 (Stathmin-2) cryptic splicing, preceding clinical onset[41]. Similar loss-of-function effects occur at the UNC13A (Unc-13 Homolog A) locus, where cryptic exon inclusion leads to nonsense-mediated decay in ALS/FTD (frontotemporal dementia) patient tissue[7, 25, 29]. Beyond splicing defects, researchers have reported that nonfibrillar TDP-43 accumulates on the rough endoplasmic reticulum early in disease, potentially disrupting protein synthesis prior to visible aggregation[26]. Together, these observations and our findings support the interpretation that nuclear TDP-43-positive neurons in ALS can already be engaged in pathological processes that impair RNA regulation and cellular homeostasis. They also provide a rationale for further investigation into miRNA targets that are disrupted early in ALS cellular progression and developing therapeutic strategies that preserve RNA regulatory pathways to protect vulnerable neurons.

Acknowledgements

The authors acknowledge funding from the Canadian Institutes of Health Research (CIHR) and preliminary work important for this project, but not shown, conducted by Alex Martin. The authors acknowledge the use of ChatGPT (OpenAI, San Francisco, CA, USA) to assist with Python coding.

Author contributions

Conceptualization: RH, MJS, DCM. Methodology and Design: CM, DCM, RH, MJS.

Experimentation/Data Acquisition: CM. Investigation: CM, MJS, DCM. Formal Analysis: CM. Supervision: RH, MJS, DCM. Manuscript (writing): CM. Manuscript (editing): CM, MJS, DCM, RH. All authors read, reviewed, and approved the final manuscript.

Funding

This work was supported by a Canadian Institutes of Health Research (CIHR) grant 201806SOP-411481 (MJS).

Data availability

All data generated or analyzed in this study are included in this published article, found in the supplementary files, or can be made available on reasonable request to the corresponding author.

Conflict of interest

The authors declare that there are no competing interests.

Ethics approval and consent to participate

Ethics approval for the use of human postmortem tissue was obtained from the University of Western Ontario, Health Sciences Research Ethics Board (HSREB) (protocols #103735 and #124855).

Informed consent

Informed consent for autopsy and retention of tissues for research purposes was obtained in written format from all patients antemortem or, if not possible, from their spouse postmortem. No minors participated in this study. To ensure anonymization, at the time of utilization for this research project, all cases were assigned a unique identifier with the master list correlating the unique identifier maintained in a locked cabinet in the office of the principal investigator (MJS) as per HSREB requirements.

Author details

¹Molecular Medicine Group, Robarts Research Institute, Schulich School of Medicine and Dentistry, Western University, London, ON, Canada. ²Department of Pathology, Schulich School of Medicine and Dentistry, Western University, London, ON, Canada. ³Department of Clinical Neurological Sciences, Schulich School of Medicine and Dentistry, Western University, London, ON, Canada.

Aarts E, Dolan CV, Verhage M, van der Sluis S (2015) Multilevel analysis quantifies variation in the experimental effect while optimizing power and preventing false positives. BMC Neurosci 16:94. https://doi.org/10.1186/s12868-015-0228-5
Aarts E, Verhage M, Veenvliet JV, Dolan CV, van der Sluis S (2014) A solution to dependency: using multilevel analysis to accommodate nested data. Nat Neurosci 17:491–496. doi: 10.1038/nn.3648
Allodi I, Nijssen J, Benitez JA, Schweingruber C, Fuchs A, Bonvicini G, Cao M, Kiehn O, Hedlund E (2019) Modeling motor neuron resilience in ALS using stem cells. Stem Cell Reports 12:1329–1341. https://doi.org/10.1016/j.stemcr.2019.04.009
Brettschneider J, Arai K, Del Tredici K, Toledo JB, Robinson JL, Lee EB, Kuwabara S, Shibuya K, Irwin DJ, Fang L, Van Deerlin VM, Elman L, McCluskey L, Ludolph AC, Lee VM-Y, Braak H, Trojanowski JQ (2014) TDP-43 pathology and neuronal loss in amyotrophic lateral sclerosis spinal cord. Acta Neuropathol 128:423–437. https://doi.org/10.1007/s00401-014-1299-6
Brockington A, Ning K, Heath PR, Wood E, Kirby J, Fusi N, Lawrence N, Wharton SB, Ince PG, Shaw PJ (2013) Unravelling the enigma of selective vulnerability in neurodegeneration: motor neurons resistant to degeneration in ALS show distinct gene expression characteristics and decreased susceptibility to excitotoxicity. Acta Neuropathol 125:95–109. https://doi.org/10.1007/s00401-012-1058-5
Brooks BR, Miller RG, Swash M, Munsat TL (2000) El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord 1:293–299. https://doi.org/10.1080/146608200300079536
Brown A-L, Wilkins OG, Keuss MJ, Kargbo-Hill SE, Zanovello M, Lee WC et al (2022) TDP-43 loss and ALS-risk SNPs drive mis-splicing and depletion of UNC13A. Nature 603:131–137. https://doi.org/10.1038/s41586-022-04436-3
Brown RH, Al-Chalabi A (2017) Amyotrophic lateral sclerosis. N Engl J Med 377:162–172. https://doi.org/10.1056/NEJMra1603471
Campos-Melo D, Droppelmann CA, He Z, Volkening K, Strong MJ (2013) Altered microRNA expression profile in amyotrophic lateral sclerosis: a role in the regulation of NFL mRNA levels. Mol Brain 6:26. https://doi.org/10.1186/1756-6606-6-26
Cestra G, Rossi S, Di Salvio M, Cozzolino M (2017) Control of mRNA translation in ALS proteinopathy. Front Mol Neurosci 10:85. https://doi.org/10.3389/fnmol.2017.00085
Comley L, Allodi I, Nichterwitz S, Nizzardo M, Simone C, Corti S, Hedlund E (2015) Motor neurons with differential vulnerability to degeneration show distinct protein signatures in health and ALS. Neuroscience 291:216–229. https://doi.org/10.1016/j.neuroscience.2015.02.013
Dash BP, Freischmidt A, Weishaupt JH, Hermann A (2024) An integrative miRNA-mRNA expression analysis identifies miRNA signatures associated with SOD1 and TARDBP patient-derived motor neurons. Hum Mol Genet 33:1300–1314. https://doi.org/10.1093/hmg/ddae072
D’Erchia AM, Gallo A, Manzari C, Raho S, Horner DS, Chiara M, Valletti A, Aiello I, Mastropasqua F, Ciaccia L, Locatelli F, Pisani F, Nicchia GP, Svelto M, Pesole G, Picardi E (2017) Massive transcriptome sequencing of human spinal cord tissues provides new insights into motor neuron degeneration in ALS. Sci Rep 7:10046. https://doi.org/10.1038/s41598-017-10488-7
Emde A, Eitan C, Liou L, Libby RT, Rivkin N, Magen I, Reichenstein I, Oppenheim H, Eilam R, Silvestroni A, Alajajian B, Ben‐Dov IZ, Aebischer J, Savidor A, Levin Y, Sons R, Hammond SM, Ravits JM, Möller T, Hornstein E (2015) Dysregulated miRNA biogenesis downstream of cellular stress and ALS‐causing mutations: a new mechanism for ALS. EMBO J 34:2633–2651. https://doi.org/10.15252/embj.201490493
Figueroa-Romero C, Hur J, Lunn JS, Paez-Colasante X, Bender DE, Yung R, Sakowski SA, Feldman EL (2016) Expression of microRNAs in human post-mortem amyotrophic lateral sclerosis spinal cords provides insight into disease mechanisms. Mol Cell Neurosci 71:34–45. https://doi.org/10.1016/j.mcn.2015.12.008
Freischmidt A, Müller K, Ludolph AC, Weishaupt JH (2013) Systemic dysregulation of TDP-43 binding microRNAs in amyotrophic lateral sclerosis. Acta Neuropathol Commun 1:42. https://doi.org/10.1186/2051-5960-1-42
Gizzi M, DiRocco A, Sivak M, Cohen B (1992) Ocular motor function in motor neuron disease. Neurology 42:1037–1037. https://doi.org/10.1212/WNL.42.5.1037
Hawley ZCE, Campos-Melo D, Strong MJ (2019) MiR-105 and miR-9 regulate the mRNA stability of neuronal intermediate filaments. Implications for the pathogenesis of amyotrophic lateral sclerosis (ALS). Brain Res 1706:93–100. https://doi.org/10.1016/j.brainres.2018.10.032
Hawley ZCE, Campos-Melo D, Strong MJ (2020) Evidence of a negative feedback network between TDP-43 and miRNAs dependent on TDP-43 nuclear localization. J Mol Biol 432:166695. https://doi.org/10.1016/j.jmb.2020.10.029
Hedlund E, Karlsson M, Osborn T, Ludwig W, Isacson O (2010) Global gene expression profiling of somatic motor neuron populations with different vulnerability identify molecules and pathways of degeneration and protection. Brain 133:2313–2330. https://doi.org/10.1093/brain/awq167
Hideyama T, Yamashita T, Aizawa H, Tsuji S, Kakita A, Takahashi H, Kwak S (2012) Profound downregulation of the RNA editing enzyme ADAR2 in ALS spinal motor neurons. Neurobiol Dis 45:1121–1128. https://doi.org/10.1016/j.nbd.2011.12.033
Juźwik CA, Drake S, Zhang Y, Paradis-Isler N, Sylvester A, Amar-Zifkin A, Douglas C, Morquette B, Moore CS, Fournier AE (2019) microRNA dysregulation in neurodegenerative diseases: A systematic review. Prog Neurobiol 182:101664. https://doi.org/10.1016/j.pneurobio.2019.101664
Kanning KC, Kaplan A, Henderson CE (2010) Motor neuron diversity in development and disease. Annu Rev Neurosci 33:409–440. https://doi.org/10.1146/annurev.neuro.051508.135722
Kawahara Y, Mieda-Sato A (2012) TDP-43 promotes microRNA biogenesis as a component of the Drosha and Dicer complexes. Proc Natl Acad Sci U S A 109:3347–3352. https://doi.org/10.1073/pnas.1112427109
Koike Y, Pickles S, Estades Ayuso V, Jansen-West K, Qi YA, Li Z, Daughrity LM, Yue M, Zhang Y-J, Cook CN, Dickson DW, Ward M, Petrucelli L, Prudencio M (2023) TDP-43 and other hnRNPs regulate cryptic exon inclusion of a key ALS/FTD risk gene, UNC13A. PLoS Biol 21:e3002028. https://doi.org/10.1371/journal.pbio.3002028
Kon T, Mori F, Tanji K, Miki Y, Nishijima H, Nakamura T, Kinoshita I, Suzuki C, Kurotaki H, Tomiyama M, Wakabayashi K (2022) Accumulation of nonfibrillar TDP-43 in the rough endoplasmic reticulum is the early-stage pathology in amyotrophic lateral sclerosis. J Neuropathol Exp Neurol 81:271–281. https://doi.org/10.1093/jnen/nlac015
Lee H, Lee JJ, Park NY, Dubey SK, Kim T, Ruan K, Lim S Bin, Park S-H, Ha S, Kovlyagina I, Kim K, Kim S, Oh Y, Kim H, Kang S-U, Song M-R, Lloyd TE, Maragakis NJ, Hong Y Bin, Eoh H, Lee G (2021) Multi-omic analysis of selectively vulnerable motor neuron subtypes implicates altered lipid metabolism in ALS. Nat Neurosci 24:1673–1685. https://doi.org/10.1038/s41593-021-00944-z
Lu J, Tsourkas A (2011) Quantification of miRNA abundance in single cells using locked nucleic acid-FISH and enzyme-labeled fluorescence. Methods Mol Biol 680:77–88. https://doi.org/10.1007/978-1-60761-901-7_5
Ma XR, Prudencio M, Koike Y, Vatsavayai SC, Kim G, Harbinski F, Briner A, Rodriguez CM, Guo C, Akiyama T, Schmidt HB, Cummings BB, Wyatt DW, Kurylo K, Miller G, Mekhoubad S, Sallee N, Mekonnen G, Ganser L, Rubien JD, Jansen-West K, Cook CN, Pickles S, Oskarsson B, Graff-Radford NR, Boeve BF, Knopman DS, Petersen RC, Dickson DW, Shorter J, Myong S, Green EM, Seeley WW, Petrucelli L, Gitler AD (2022) TDP-43 represses cryptic exon inclusion in the FTD–ALS gene UNC13A. Nature 603:124–130. https://doi.org/10.1038/s41586-022-04424-7
Magarotto M, Gawne RT, Vilkaite G, Beltrami M, Mason AS, Chen H-J (2025) Familial ALS/FTD-associated RNA-binding deficient TDP-43 mutants cause neuronal and synaptic transcript dysregulation in vitro. Hum Mol Genet 34:1480–1494. https://doi.org/10.1093/hmg/ddaf111
Nijssen J, Comley LH, Hedlund E (2017) Motor neuron vulnerability and resistance in amyotrophic lateral sclerosis. Acta Neuropathol 3:863–885. https://doi.org/10.1007/s00401-017-1708-8
Onomoto K, Sakai M, Watanabe M, Fukao A, Sakamura Y, Miyao M, Tomohiro T, Yamashita A, Fujiwara T, Takahashi T, Ui-Tei K, Yoneyama M (2025) TRBP modulates RLR signaling by inhibiting PKR-mediated antiviral stress granule formation. Sci Rep 15:20678. https://doi.org/10.1038/s41598-025-07121-3
Packer AN, Xing Y, Harper SQ, Jones L, Davidson BL (2008) The bifunctional microRNA miR-9/miR-9* regulates REST and CoREST and is downregulated in Huntington’s disease. J Neurosci 28:14341–14346. https://doi.org/10.1523/JNEUROSCI.2390-08.2008
Planell-Saguer M de, Rodicio MC, Mourelatos Z (2010) Rapid in situ codetection of noncoding RNAs and proteins in cells and formalin-fixed paraffin-embedded tissue sections without protease treatment. Nat Protoc 5:1061–1073. https://doi.org/10.1038/NPROT.2010.62
Python Software Foundation (2023) Python Language Reference
Rinchetti P, Rizzuti M, Faravelli I, Corti S (2018) MicroRNA metabolism and dysregulation in amyotrophic lateral sclerosis. Mol Neurobiol 55:2617–2630. https://doi.org/10.1007/s12035-017-0537-z
Rizzuti M, Filosa G, Melzi V, Calandriello L, Dioni L, Bollati V, Bresolin N, Comi G Pietro, Barabino S, Nizzardo M, Corti S (2018) MicroRNA expression analysis identifies a subset of downregulated miRNAs in ALS motor neuron progenitors. Sci Rep 8:10105. https://doi.org/10.1038/s41598-018-28366-1
Saraiva C, Esteves M, Bernardino L (2017) MicroRNA: Basic concepts and implications for regeneration and repair of neurodegenerative diseases. Biochem Pharmacol 141:118–131. https://doi.org/10.1016/j.bcp.2017.07.008
Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A (2012) Fiji: an open-source platform for biological-image analysis. Nat Methods 9:676–682. https://doi.org/10.1038/nmeth.2019
Shaw P, Eggett CJ (2000) Molecular factors underlying selective vulnerability of motor neurons to neurodegeneration in amyotrophic lateral sclerosis. J Neurol 247:I17–I27. https://doi.org/10.1007/BF03161151
Spence H, Waldron FM, Saleeb RS, Brown A-L, Rifai OM, Gilodi M, Read F, Roberts K, Milne G, Wilkinson D, O’Shaughnessy J, Pastore A, Fratta P, Shneider N, Tartaglia GG, Zacco E, Horrocks MH, Gregory JM (2024) RNA aptamer reveals nuclear TDP-43 pathology is an early aggregation event that coincides with STMN-2 cryptic splicing and precedes clinical manifestation in ALS. Acta Neuropathol 147:50. https://doi.org/10.1007/s00401-024-02705-1
Suk TR, Rousseaux MWC (2020) The role of TDP-43 mislocalization in amyotrophic lateral sclerosis. Mol Neurodegener 15:45. https://doi.org/10.1186/s13024-020-00397-1
Xue Q, Yu C, Wang Y, Liu L, Zhang K, Fang C, Liu F, Bian G, Song B, Yang A, Wang J (2016) miR-9 and miR-124 synergistically affect regulation of dendritic branching via the AKT/GSK3β pathway by targeting Rap2a. Sci Rep 6:26781. https://doi.org/10.1038/srep26781
Yu J-Y, Chung K-H, Deo M, Thompson RC, Turner DL (2008) MicroRNA miR-124 regulates neurite outgrowth during neuronal differentiation. Exp Cell Res 314:2618–2633. https://doi.org/10.1016/j.yexcr.2008.06.002
Zhang W-H, Jiang L, Li M, Liu J (2023) MicroRNA‑124: an emerging therapeutic target in central nervous system disorders. Exp Brain Res 241:1215–1226. https://doi.org/10.1007/s00221-022-06524-2
Zhang Z, Almeida S, Lu Y, Nishimura AL, Peng L, Sun D, Wu B, Karydas AM, Tartaglia MC, Fong JC, Miller BL, Farese R V., Moore MJ, Shaw CE, Gao F-B (2013) Downregulation of microRNA-9 in iPSC-derived neurons of FTD/ALS patients with TDP-43 mutations. PLoS One 8:e76055. https://doi.org/10.1371/journal.pone.0076055

No competing interests reported.

Supplementaryfile1.docx

Download PDF

Editorial decision: Revision requested
19 Oct, 2025
Reviews received at journal
29 Sep, 2025
Reviews received at journal
26 Sep, 2025
Reviewers agreed at journal
11 Sep, 2025
Reviewers agreed at journal
10 Sep, 2025
Reviewers agreed at journal
08 Sep, 2025
Reviewers invited by journal
06 Sep, 2025
Editor assigned by journal
06 Sep, 2025
Submission checks completed at journal
06 Sep, 2025
First submitted to journal
04 Sep, 2025

You are reading this latest preprint version

Preservation of miR-9-5p and miR-124-3p in ALS-resistant oculomotor neurons contrasts with their downregulation in vulnerable spinal motor neurons, irrespective of TDP-43 pathology

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Results

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1