Abstract
Aging is a major risk factor in amyotrophic lateral sclerosis (ALS) and other adult-onset neurodegenerative disorders. Whereas young neurons are capable of buffering disease-causing stresses, mature neurons lose this ability and degenerate over time. We hypothesized that the resilience of young motor neurons could be restored by reexpression of the embryonic motor neuron selector transcription factors ISL1 and LHX3. We found that viral reexpression of ISL1 and LHX3 selectively in postnatal motor neurons reactivates aspects of their youthful gene expression program and alleviates key disease-relevant phenotypes in the SOD1G93A mouse model of ALS. Our results suggest that redeployment of lineage-specific neuronal selector transcription factors can be an effective strategy to attenuate age-dependent phenotypes in neurodegenerative disease.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 print issues and online access
$259.00 per year
only $21.58 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to the full article PDF.
USD 39.95
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
All sequencing data generated from this study have been deposited in the Gene Expression Omnibus and are available at GSE261314. Published sequencing data used in this study are available at GSE198767. The STRING database is accessible at https://string-db.org/.
Code availability
Code used to generate figures from sequencing data is available at https://github.com/TulsiPatelLab/Motor-Neuron-Programming-for-ALS. Code used to quantify CADPS2+ nuclei from spinal cord photomicrographs is provided as a Supplementary Note in the Supplementary Information.
References
Masrori, P. & Damme, P. V. Amyotrophic lateral sclerosis: a clinical review. Eur. J. Neurol. 27, 1918–1929 (2020).
Patel, T. et al. Transcriptional dynamics of murine motor neuron maturation in vivo and in vitro. Nat. Commun. 13, 5427 (2022).
Closser, M. et al. An expansion of the non-coding genome and its regulatory potential underlies vertebrate neuronal diversity. Neuron 110, 70–85 (2022).
Mazzoni, E. O. et al. Synergistic binding of transcription factors to cell-specific enhancers programs motor neuron identity. Nat. Neurosci. 16, 1219–1227 (2013).
Rhee, H. S. et al. Expression of terminal effector genes in mammalian neurons is maintained by a dynamic relay of transient enhancers. Neuron 92, 1252–1265 (2016).
Lee, S. et al. Fusion protein Isl1–Lhx3 specifies motor neuron fate by inducing motor neuron genes and concomitantly suppressing the interneuron programs. Proc. Natl Acad. Sci. USA 109, 3383–3388 (2012).
Liu, M.-L., Zang, T. & Zhang, C.-L. Direct lineage reprogramming reveals disease-specific phenotypes of motor neurons from human ALS patients. Cell Rep. 14, 115–128 (2016).
Alkaslasi, M. R. et al. Single nucleus RNA-sequencing defines unexpected diversity of cholinergic neuron types in the adult mouse spinal cord. Nat. Commun. 12, 2471 (2021).
Blum, J. A. et al. Single-cell transcriptomic analysis of the adult mouse spinal cord reveals molecular diversity of autonomic and skeletal motor neurons. Nat. Neurosci. 24, 572–583 (2021).
Martins, L. F. et al. Motor neurons use push-pull signals to direct vascular remodeling critical for their connectivity. Neuron 110, 4090–4107 (2022).
Marquardt, T. et al. Coexpressed EphA receptors and ephrin-A ligands mediate opposing actions on growth cone navigation from distinct membrane domains. Cell 121, 127–139 (2005).
Gallarda, B. W. et al. Segregation of axial motor and sensory pathways via heterotypic trans-axonal signaling. Science 320, 233–236 (2008).
Jaworski, A., Long, H. & Tessier-Lavigne, M. Collaborative and specialized functions of Robo1 and Robo2 in spinal commissural axon guidance. J. Neurosci. 30, 9445–9453 (2010).
Kumar, A. V., Mills, J. & Lapierre, L. R. Selective autophagy receptor p62/SQSTM1, a pivotal player in stress and aging. Front. Cell Dev. Biol. 10, 793328 (2022).
Rudnick, N. D. et al. Distinct roles for motor neuron autophagy early and late in the SOD1G93A mouse model of ALS. Proc. Natl Acad. Sci. USA 114, E8294–E8303 (2017).
Gal, J., Ström, A.-L., Kilty, R., Zhang, F. & Zhu, H. p62 accumulates and enhances aggregate formation in model systems of familial amyotrophic lateral sclerosis. J. Biol. Chem. 282, 11068–11077 (2007).
Mitsui, S. et al. Systemic overexpression of SQSTM1/p62 accelerates disease onset in a SOD1H46R-expressing ALS mouse model. Mol. Brain 11, 30 (2018).
Trist, B. G. et al. Co-deposition of SOD1, TDP-43 and p62 proteinopathies in ALS: evidence for multifaceted pathways underlying neurodegeneration. Acta Neuropathol. Commun. 10, 122 (2022).
Fecto, F. et al. SQSTM1mutations in familial and sporadic amyotrophic lateral sclerosis. Arch. Neurol. 68, 1440–1446 (2011).
Salvany, S. et al. Accumulation of misfolded SOD1 outlines distinct patterns of motor neuron pathology and death during disease progression in a SOD1G93A mouse model of amyotrophic lateral sclerosis. Brain Pathol. 32, e13078 (2022).
Watanabe, M. et al. Histological evidence of protein aggregation in mutant SOD1 transgenic mice and in amyotrophic lateral sclerosis neural tissues. Neurobiol. Dis. 8, 933–941 (2001).
Alexander, G. M. et al. Effect of transgene copy number on survival in the G93A SOD1 transgenic mouse model of ALS. Mol. Brain Res. 130, 7–15 (2004).
Saxena, S., Cabuy, E. & Caroni, P. A role for motoneuron subtype–selective ER stress in disease manifestations of FALS mice. Nat. Neurosci. 12, 627–636 (2009).
Hatzipetros, T. et al. A quick phenotypic neurological scoring system for evaluating disease progression in the SOD1-G93A mouse model of ALS. J. Vis. Exp. 53257 (2015).
Miana‐Mena, F. J. et al. Optimal methods to characterize the G93A mouse model of ALS. Amyotroph. Lateral Scler. 6, 55–62 (2009).
Scott, S. et al. Design, power, and interpretation of studies in the standard murine model of ALS. Amyotroph. Lateral Scler. 9, 4–15 (2009).
Lalancette-Hebert, M., Sharma, A., Lyashchenko, A. K. & Shneider, N. A. Gamma motor neurons survive and exacerbate alpha motor neuron degeneration in ALS. Proc. Natl Acad. Sci. USA 113, E8316–E8325 (2016).
Poplawski, G. H. D. et al. Injured adult neurons regress to an embryonic transcriptional growth state. Nature 581, 77–82 (2020).
Delestrée, N. et al. Adult spinal motoneurones are not hyperexcitable in a mouse model of inherited amyotrophic lateral sclerosis. J. Physiol. 592, 1687–1703 (2014).
Martinez-Silva, M. et al. Hypoexcitability precedes denervation in the large fast-contracting motor units in two unrelated mouse models of ALS. eLife 7, e30955 (2018).
Bączyk, M. et al. Synaptic restoration by cAMP/PKA drives activity-dependent neuroprotection to motoneurons in ALS. J. Exp. Med. 217, e20191734 (2020).
Liu, J. & Wang, F. Role of neuroinflammation in amyotrophic lateral sclerosis: cellular mechanisms and therapeutic implications. Front. Immunol. 8, 1005 (2017).
Challis, R. C. et al. Systemic AAV vectors for widespread and targeted gene delivery in rodents. Nat. Protoc. 14, 379–414 (2019).
Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet J. 17, 10–12 (2011).
Li, B. & Dewey, C. N. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics 12, 323 (2011).
Robinson, M. D., McCarthy, D. J. & Smyth, G. K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26, 139–140 (2010).
McCarthy, D. J., Chen, Y. & Smyth, G. K. Differential expression analysis of multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids Res. 40, 4288–4297 (2012).
Acknowledgements
The Virology Core at the Zuckerman Institute of Columbia University, led by D. Ng, aided in viral production. Y. Weitzner was instrumental in validating the staining and quantification protocols for SQSTM1 round bodies. This work was supported by funding from Project ALS (H.W.) and by National Institutes of Health grants R01NS116141 (H.W. and T.P.), R01NS109217 (H.W.), R01NS140764 (H.W. and E.R.L.) and K99NS121136 (T.P.). The funders had no role in study design, data collection and analysis, preparation of the paper or decision to publish.
Author information
Authors and Affiliations
Contributions
E.R.L., T.P. and H.W. designed all experiments. E.R.L., T.P. and H.W. wrote the paper and prepared the figures. T.P. designed and cloned the ChatE and AAV–Isl1 and AAV–Lhx3 constructs. G.N., F.L., E.C., S.A. and T.P. prepared the AAVs. E.R.L., T.P., E.C. and G.N. injected pups with AAVs. E.C., S.T. and H.M. prepared tissue for immunohistochemistry. S.T. and I.D. performed immunohistochemistry and confocal imaging. E.C., S.T. and I.D. quantified MNX1, CADPS2, SQSTM1 and SOD1 staining. H.M. assessed tremor onset and lifespan. E.R.L. performed all statistical analyses for histological and behavioral assays. T.P. isolated nuclei for single-nuclei sequencing. T.P., J.A.C. and E.R.L. analyzed single-nuclei sequencing data.
Corresponding authors
Ethics declarations
Competing interests
E.R.L., T.P. and H.W. have filed a US provisional patent application for the use of ISL1/LHX3 in neurodegeneration (no. 63/359,527). The other authors declare no competing interests.
Peer review
Peer review information
Nature Neuroscience thanks Aaron Gitler and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Extended Data Fig. 1 Endogenous ISL1 and LHX3 expression decreases with time.
a, Endogenous Il1 and Lhx3 expression decrease sharply after their peak at E13.5 (RNAseq data from2; n = 2-3 mice per timepoint; expression for each time point normalized to values at E13.5). b, Endogenous ISL1 and LHX3 are undetectable by immunofluorescence in CHAT+ cells in the L4-L5 ventral horn in untreated animals at P45. Images are maximum intensity projections of 15 z sections taken 3 µm apart. Scale bar represents 20 µM. Similar immunostaining experiments were performed in n = 12 mice.
Extended Data Fig. 2 Design and in vivo characterization of the cell type specificity of ChatE.
a, Schematic of ATAC-seq data from2 showing accessibility of ChatE (pink) in spinal motor neurons over time. b, Design of ChatE constructs. c, Schematic of the 1000 bp ChatE demonstrating the presence of transcription factor binding motifs enriched during motor neuron specification or maturation. d,e, Quantification and representative images of mCherry distribution across spinal cord and dorsal root ganglia (DRG) sections from n = 3 mice injected with low titer AAV-mCherry (3.19E + 10 vg/animal) at P1 and analyzed at P45. In D, tissues from the same animal were used to quantify the incidence of mCherry+ cells across the indicated cell types. Circles represent females; triangles represent males. In E, whole cord images were taken at 5x on an epifluorescence microscope (scale bar = 200 µm). Unfilled arrowhead indicates V0c interneurons; dashed lines indicate PGCs represented in inset. PGC inset images were taken at 40x on a confocal microscope are maximum projections of 7 z sections taken 3 µm apart (scale bar = 20 µm). DRG images were taken at 20x on an epifluorescence microscope (scale bar = 200 µm). Filled arrowhead indicates mCherry+ sensory neuron. f, Representative 5x epifluorescence images of whole L4-L5 spinal cord sections from n = 2 animals injected at P1 with low titer AAV-NLS-GFP (2.10E + 10 vg/animal). Dashed yellow line indicates region shown in inset.
Extended Data Fig. 3 In vivo characterization of AAV-Isl1 and AAV-Lhx3 expression in spinal motor neurons.
a, Distribution of ISL+, LHX3+, ISL + LHX3+, and ISL1-LHX3- cells among CHAT+ cells in the L4-L5 ventral horn at P45 from animals treated with low titer (6-9E+10 vg/animal) or high titer AAVs (3-4+11 vg/animal). Mean percentage of motor neurons in each condition was quantified from 6 hemisections per animal and 15 confocal images per hemisection from 2 (low titer) or 6 (high titer) animals. Error bars represent SEM. b, Quantification of transgene-positive motor neurons in L4-L5 ventral horn at P45 in animals treated with AAV-Isl1 or AAV-Lhx3 under high titer conditions (3.6E+11 - 3.8E+11 vg/animal) at P1. n = 1–4 animals per treatment per time point. Points represent mean transgene-expressing motor neurons per hemisection per animal, quantified across 6 hemisections per animal from 15 confocal images per hemisection. Error bars represent SEM. c, Distribution of AAV+ motor neurons among the high titer animals plotted in c, separated by sex (n = 3 animals per sex). No significant effect of sex was observed by one-way ANOVA (p = 0.3739). d, Quantification of total CHAT+ motor neurons per hemisection in the L4-L5 region of the spinal cord in untreated animals or in AAV-mCherry-treated or AAV-Isl1+AAV-Lhx3-treated injected at P1 and analyzed at P45 in 15 confocal sections from 6 70 µm hemisections per animal. Each point represents the mean motor neuron count per hemisection from one animal; triangles represent males, and circles represent females. Horizontal bars represent animal medians per treatment condition. No significant effect of treatment on motor neuron survival was observed by one-way ANOVA (p = 0.1600).
Extended Data Fig. 4 Characterization of CCA-generated clusters from Single Nuc Seq.
a, RNAseq-derived expression levels of Mnx1 in mouse motor neurons over time (from2). Numbers on the y-axis represent expression values normalized across all genes in all samples. b, Percent control and treatment cells in all clusters. While there is some variation between clusters, all clusters are comprised of both control and treatment cells. c, Violin plot showing expression of the neuronal gene Snap25, the cholinergic gene Chat, the excitatory gene Slc17a6, and the inhibitory gene Gad1 in clusters from Fig. 2c. d, Dot plot showing expression levels and percent cells expressing known alpha, gamma, and type 3 markers in cholinergic clusters. e, Violin plot showing distribution of the four AAV elements across CCA clusters. f, Differential gene expression between AAV transcript+ treatment cells vs. control cells in motor neurons and M5 clusters. Gene expression changes are largely restricted to alpha and type 3 motor neurons.
Extended Data Fig. 5 AAV-driven ISL1 and LHX3 expression are observed in gamma motor neurons.
Representative immunostaining in the L4-L5 region of the ventral horn from animals injected with AAV-Isl1+AAV-Lhx3 (6.38E+10 total vg/animal) at P1 and analyzed at P14 (a) or P21 (b). Images are maximum intensity projections of 8 z sections taken 3 µm apart. Filled arrowheads indicate ectopic ISL1 and/or LHX3-expressing gamma motor neurons, identified as small-diameter CHAT+ cells that lack NEUN expression. Outlined arrowheads indicate gamma motor neurons that do not currently express ISL1 or LHX3. Scale bars represent 20 µM. Similar immunostainings were performed in n = 4 mice at P14 and n = 2 mice at P21.
Extended Data Fig. 6 Characterization of motor neuron-specific clusters from Single Nuc Seq.
a, Expression of alpha, gamma, and type 3 markers in motor neuron-specific clusters. b, Percent control and treatment cells in motor neuron-specific clusters. Two of the clusters, 6 and 8 (termed type 3 prime and alpha prime, respectively), are composed solely of treatment cells. These clusters represent type 3 (6) and alpha (8) motor neurons from the treatment group. c, Cluster 4 is similar to one found in published single nuclei RNAseq of adult motor neurons (publicly accessible at spinalcoratlas.org) that is found near alpha cells but lacks key marker that label most alpha cells (Stk32a, Sv2b). The 7 markers shown here to be expressed in cluster 4 are also expressed in the corresponding cluster in the published data. d, Percent control and treatment cells in non-motor neuron cholinergic clusters. While there is some variation between clusters, all clusters have both control and treatment cells. e,f, Clustering of non-motor neuron cholinergic cells, and composition of control and treatment nuclei in each cluster.
Extended Data Fig. 7 Additional validation of Single Nuc Seq DEGs and motifs.
a,b, Dotplots showing log2 fold expression change in alpha prime vs. alpha clusters (alpha DEGs) or type 3 prime vs. type 3 clusters on x-axis and log2 fold expression change in the same genes in treated bulk RNAseq data vs. control bulk RNAseq data. There is an overall positive correlation in these gene expression changes despite differences in the methodology and sensitivity of snRNAseq vs. bulk RNAseq. The positive correlation with bulk RNAseq is stronger for alpha (R2 = 0.40, p < 0.0001) than type 3 (R2 = 0.21, p < 0.0001) DEGs. c, Representative immunostaining for RFP in the AAV-mCherry-treated animal included in Fig. 2b showing extensive transgene expression in CHAT+ motor neurons. Scale bar represents 20 µm. d, Enlarged regions of images from Fig. 2b showing MNX1 expression in a subset of ISL1+ cells that show strong nuclear CADPS2 staining (arrowheads). No MNX1 signal was detected in animals injected with control mCherry virus. Scale bar represents 20 µm. e, UMAP from Fig. 1h showing distribution and prevalence of Lhx3 motifs among differentially accessible peaks in each cluster.
Extended Data Fig. 8 Single Nuc seq DEGs indicate that transgene-expressing motor neurons at P21 resemble perinatal motor neurons.
a, Heatmap of R2 values for correlations between AAV-Isl1-Lhx3-mediated changes in gene expression vs. maturation-mediated changes in expression between P21 and the indicated time points for alpha and type 3 DEGs. Values for P4 are reported in Fig. 2g. p values for alpha DEGs are as follows: E13.5, p < 0.0001; P4, p < 0.0001; P13, p < 0.0001. p values for type 3 DEGs are as follows: E13.5, p < 0.0001; P4, p < 0.0001; P13, p = 0.0002. Minimum R2 value = 0.03823; maximum R2 value = 0.4063. b, Bubble plot of FDR values for shared, significantly enriched GO terms between up-regulated alpha and type 3 DEGs. Maximum FDR value = 0.0381; minimum FDR value = 8.09E-12. Bubble size represents the percentage of shared DEGs populating each term.
Extended Data Fig. 9 The effects of AAV-Isl1+AAV-Lhx3 on ALS pathologies are dose-dependent and do not result from interference with endogenous Sqstm1 or Sod1 expression, or with Sod1G93A transgene expression.
a, Correlation between motor neuron transduction efficiency and incidence of SQSTM1 aggregates in mice from Fig. 3a, b. Animals were treated with a range of AAV-Isl1+AAV-Lhx3 doses at P1 (6.37E+10 – 3.69E+11 vg/animal; n = 17 mice, 6 treated with high titer and 11 treated with low titer). Quantification of SQSTM1 round bodies and ISL1 and LHX3 expression in CHAT+ motor neurons at P45 was performed in 6 70 µm L4-L5 hemisections per animal from 15 confocal images per hemisection. Each point represents one animal. The percentage of all CHAT+ motor neurons per hemisection (whether positive for ISL1 or LHX3 or not) that exhibited SQSTM1 round bodies was correlated with the percentage of CHAT+ motor neurons in the same hemisections that were positive for ectopic ISL1 and/or LHX3 expression by simple linear regression, segregated by sex (Males: R squared=0.5745; p = 0.0180; Females: R squared=0.7023, p = 0.0094). b, Distribution of transgene-expressing CHAT+ motor neurons among cells positive or negative for SQSTM1 round bodies from the high titer AAV-Isl1+AAV-Lhx3-treated animals included in Fig. 3b (n = 6). c, ATAC-seq reads at the genomic locus of Sqstm1 on the left, and violin plots showing gene expression on the right. Expression Sqstm1 is not significantly different between treatment-specific clusters and control clusters. d, Representative immunostaining of L4-L5 ventral horn motor neurons from animals treated with high titer AAV-Isl1-Lhx3-fusion (1.80E+11 vg/animal) at P1 and analyzed at P14. CHAT+ motor neurons show strong and specific expression of both transgenes at this time point. Scale bar represents 20 µm. Similar immunostaining experiments were performed in n = 2 mice. e, qPCR quantification of Isl1-Lhx3-fusion, mCherry, and human SOD1 transgene expression in whole spinal cord lysates from SOD1G93A animals injected at P1 with AAV-mCherry, AAV-Isl1-Lhx3-fusion, or left untreated, and analyzed at P14. Isl1-Lhx3 and mCherry show strong and specific expression in AAV-Isl1-Lhx3-fusion or AAV-mCherry-treated animals, respectively. Expression of the human SOD1 transgene is unchanged under all treatment conditions. Significance was determined two-way ANOVA for the effects of AAV treatment and transgene expression with Dunnett’s multiple comparisons test. ****p < 0.0001. Error bars represent SEM. f, ATAC-seq reads at the genomic locus of endogenous Sod1 on the left, and violin plots showing gene expression on the right. Expression of endogenous Sod1 is not significantly different between treatment-specific clusters and control clusters.
Extended Data Fig. 10 Effects of low-titer AAV-Isl1+AAV-Lhx3 treatment in SOD1G93A mice.
a, The probability of remaining asymptomatic as a measure of tremor onset in AAV-mCherry vs AAV-Isl1+AAV-Lhx3-treated SOD1G93A mice was compared by Log-rank (Mantel-Cox) tests (n = 10–13 animals per treatment per sex). Separate analyses were performed for male vs. female animals. For females, AAV-Isl1+AAV-Lhx3 treatment delayed tremor onset by 15 days (p = 0.0036). For males, tremor onset was delayed by 5 days but was not significant (p = 0.1138). b, The probability of survival in the same animals as (a) compared by Log-rank (Mantel-Cox) test. Mean survival was as follows: AAV-mCherry males = 161 days; AAV-Isl1+AAV-Lhx3 males = 160.5 days; AAV-mCherry females = 166 days; AAV-Isl1+AAV-Lhx3 females = 169 days. No significant effects of treatment on survival were observed for females (p = 0.8807) or males (p = 0.2687). c, Representative immunostaining for CHAT+ motor neuron survival and transgene expression in a subset of animals from a,b collected at endstage (n = 6 mice). mCherry expression generally remains strong, while ISL1 and LHX3 expression are sparse or absent. d, In the subset of AAV-Isl1+AAV-Lhx3-treated animals analyzed at endstage there was a trend toward a significant correlation between the percentage of CHAT+ motor neurons per hemisection expressing ISL1 or LHX3 and the number of surviving motor neurons per hemisection in the L4-L5 region of the spinal cord (n = 6 mice; R2 = 0.6200, p = 0.0630). 6 70 µm hemisections were quantified per animal. Circles represent females; triangles represent males.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Lowry, E.R., Patel, T., Costa, J.A. et al. Embryonic motor neuron programming factors reactivate immature gene expression and suppress ALS pathologies in postnatal motor neurons. Nat Neurosci 28, 2044–2053 (2025). https://doi.org/10.1038/s41593-025-02033-x
Received:
Accepted:
Published:
Version of record:
Issue date:
DOI: https://doi.org/10.1038/s41593-025-02033-x
