Transcriptomics Analyses of ALS Post-mortem Motor Cortex highlight alteration and potential biomarkers in the Neuropeptide Signalling pathway

Renata Kabiljo, Heather Marriott, Guy P Hunt, Abigail L Pfaff, Ahmad Al Khleifat, Brett Adey, Ashley Jones, Claire Troakes, John P Quinn, Richard J B Dobson, Sulev Koks, Ammar Al-Chalabi, Alfredo Iacoangeli
doi: https://doi.org/10.1101/2023.05.05.23289551

Abstract

Background Amyotrophic lateral sclerosis (ALS) is a fatal heterogeneous neurodegenerative disease that typically leads to death from respiratory failure within two to five years. Despite the identification of several genetic risk factors, the biological processes involved in ALS pathogenesis remain poorly understood. The motor cortex is an ideal region to study dysregulated pathological processes in ALS as it is affected from the earliest stages of the disease. In this study, we investigated motor-cortex gene expression of cases and controls to gain new insight into the molecular footprint of ALS.

Methods We performed a large case-control differential expression analysis of two independent post-mortem motor cortex bulk RNA-sequencing (RNAseq) datasets from the King’s College London BrainBank (N = 171) and TargetALS (N = 132). Differentially expressed genes from both datasets were subjected to gene and pathway enrichment analysis. Genes common to both datasets were also reviewed for their involvement with known mechanisms of ALS pathogenesis to identify potential candidate genes. Finally, we performed a correlation analysis of genes implicated in pathways enriched in both datasets with clinical outcomes such as the age of onset and survival.

Results Differential expression analysis identified 2,290 and 402 differentially expressed genes in KCL BrainBank and TargetALS cases, respectively. Enrichment analysis revealed significant synapse-related processes in the KCL BrainBank dataset, while the TargetALS dataset carried an immune system-related signature. There were 44 differentially expressed genes which were common to both datasets, which represented previously recognised mechanisms of ALS pathogenesis, such as lipid metabolism, mitochondrial energy homeostasis and neurovascular unit dysfunction. Differentially expressed genes in both datasets were significantly enriched for the neuropeptide signalling pathway. By looking at the relationship between the expression of neuropeptides and their receptors with clinical measures, we found that in both datasets NPBWR1, TAC3 and SSTR1 correlated with age of onset, and GNRH1, TACR1 with survival. We provide access to gene-level expression results to the broader research community through a publicly available web application (https://alsgeexplorer.er.kcl.ac.uk).

Conclusion This study identified motor-cortex specific pathways altered in ALS patients, potential molecular targets for therapeutic disease intervention and a set of neuropeptides and receptors for investigation as potential biomarkers.

Background

Amyotrophic lateral sclerosis (ALS) is a devastating, heterogeneous neurodegenerative disease that leads to death from neuromuscular respiratory failure within two to five years. Despite the heritability of ALS being largely unknown, the discovery of several genetic risk factors has driven the identification of multiple biological processes involved in the disease aetiology1,2. However, the pathogenic mechanisms of ALS remain poorly understood.

The motor cortex is affected from the earliest stages of the disease and is one of the brain regions majorly affected by ALS pathology. Consequently, its disease expression signatures have been investigated to gain insight into the disease aetiology. Recent studies have reported differential expression of genes related to transcription, neurotransmission, protein synthesis and oxidative stress processes3, with extensive RNA expression alterations of neuroinflammatory and synaptic-related pathways identified at both gene and isoform levels4. Nevertheless, studies of this critical region at the whole-transcriptome level usually involve relatively small sample sizes and do not replicate their results in independent datasets.

To study functional changes occurring in the motor cortex of ALS patients, we have utilised two large independent RNAseq datasets of ALS patients and controls from the King’s College London (KCL) BrainBank (171 subjects) and TargetALS (234 samples from 132 subjects). By analysing the two datasets, our goal was to obtain profiles of altered gene expression that would identify the underlying molecular footprint of ALS and use this information to propose novel hypotheses about disease mechanisms, potential drug targets and biomarkers.

Methods

Datasets

The KCL BrainBank RNAseq dataset5,6 consists of post-mortem RNAseq samples from KCL and the Medical Research Council (MRC) London Neurodegenerative Diseases Brain Bank. Frozen human post-mortem tissue was taken from the primary motor cortex. The RNAseq protocol is outlined in the Supplementary Methods. 112 sporadic ALS cases and 59 controls were included in this analysis.

The TargetALS RNAseq dataset comes from TargetALS Human Postmortem Tissue Core and the New York Genome Centre (NYGC). We obtained the TargetALS dataset from https://www.targetals.org/. A list of the sample IDs of 243 samples (from 132 subjects, 112 ALS cases and 20 controls) used in this study is available in Supplementary Table 1. Frozen human post-mortem tissue was taken from the motor cortex (medial, lateral, and unspecified). Samples from different motor cortex regions for the same subjects were included, where available (see Supplementary Methods: TargetALS Sample Selection for more information on the inclusion of multiple samples per subject). Library preparation and sequencing of RNA are described in detail here https://www.targetals.org/wp-content/uploads/2020/11/README-1.zip.

Bioinformatics, data processing and expression analysis

Both datasets were processed using the same pipeline, which is documented and distributed here: https://github.com/rkabiljo/RNASeq_Genes_ERVs. The pipeline involves interleaving the paired reads using reformat.sh from BBtools7 and trimming adapters and low-quality reads using bbduk.sh7, before aligning to the Hg38 reference genome using STAR8.Transcripts are then quantified using HTSeq9. The Supplementary Methods provides more details about each one of these steps and the parameters used. DESeq210 was used for normalisation and differential expression analysis. The scripts are available at https://github.com/rkabiljo/DifferentialExpression_Genes. To test whether the number of differentially expressed genes identified in our analysis was due to chance, we performed 1,000 random permutations of case/control labels and performed differential expression analysis on each permutation. For each permutation, we counted the number of genes with false discovery rate (FDR) adjusted p-value lower than our selected cut-off of 0.05. The random permutation scripts are available here https://github.com/rkabiljo/ShufflingLabelsForDE. Functional enrichment was performed using MetaCore™ (available at https://portal.genego.com) for Gene Ontology (GO) processes, Pathway Maps, Process Networks and Disease by Biomarker enrichments, whilst gProfiler211 was used for Reactome12 enrichment. We used BRETIGEA13 to derive cell proportions for the following cell types; astrocytes, endothelial cells, microglia, neurons, oligodendrocytes, and oligodendrocyte progenitor cells (OPCs).

To share the results with the wider community, we developed a web application using the R “shiny” framework (version 1.7.1) to allow quick visualisation of specific gene expression in ALS cases and controls, available at https://alsgeexplorer.er.kcl.ac.uk.

Results

Sample Characteristics

The summary characteristics of the two datasets are shown in Table 1. The KCL BrainBank dataset had significant differences between cases and controls in terms of age, RNA integrity number (RIN) and post-mortem delay (PMD). The TargetALS dataset showed no significant differences in age, sex, and RIN between cases and controls, but did display a significant difference in PMD (p-value = 0.0086). We have accounted for these differences through our choice of covariates included in the differential expression analyses (see Supplementary Methods). Principal component analysis (PCA) plots of samples based on expression data using the 500 most variable genes are in Supplementary Figures 1A (KCL BrainBank) and 1B (TargetALS).

View this table:
Table 1.

KCL BrainBank and TargetALS sample characteristics. Age: Age at death. PMD: Post-mortem delay. RIN: RNA integrity number. *p-value of Chi-square test for categorical variables (Sex) and T-test for continuous variables (Age, RIN Number, PMD)

Differential expression analysis identifies genes linked to ALS pathogenesis shared in both datasets

KCL BrainBank

A total of 2,290 genes were differentially expressed with an adjusted p-value of <0.05, in the KCL BrainBank dataset. Of these, 1,507 of the genes were upregulated and 783 were downregulated in cases (Figure 1A, 1B). None of the 1000 random permutations produced more differentially expressed genes than the actual differential expression analysis, thus confirming the significance of these results. In the vast majority of the permutations, there were no differentially expressed genes (with adjusted p-value <0.05), with the number of significant genes across permutations averaging 4.5.

Figure 1.
Figure 1.

A) KCL BrainBank and C) TargetALS MA plots representing the distribution of log2 fold change (logFC) across all expression levels; B) KCL BrainBank and D) TargetALS Volcano plots. Plot D has been cropped to exclude one point with extremely high log2 fold change. Genes coloured red pass both the p-value and logFC thresholds. Grey genes are not significant, green genes only pass the logFC threshold, and blue genes only pass the logFC threshold. E) Numbers of differentially expressed genes in the two datasets.

TargetALS

A total of 402 genes were differentially expressed, with an adjusted p-value of <0.05, in the TargetALS dataset. Of these, 250 of these genes were downregulated and 152 were upregulated in cases (Figure 1C, 1D). When performing random permutations, 9 (0.9%) produced as many or more differentially expressed genes as the actual differential expression analysis. In most of the permutations, there were no differentially expressed genes (with adjusted p-value <0.05), with an average of 28 significantly expressed genes across permutations.

Shared Differentially Expressed Genes

We found 44 genes significant in both datasets with an adjusted p-value of <0.05. All of these genes had a concordant direction of regulation in both datasets; 29 genes were downregulated and 15 were upregulated (Figure 1E). To estimate the significance of this intersection between differentially expressed genes in the two datasets, we performed a hypergeometric test, using 41,709 as the number of background genes. The p-value obtained was < 1.2e-05. Genes that were significant in both datasets are listed in Table 2.

View this table:
Table 2.

Genes significantly differentially expressed in both datasets. Red: Upregulated; Blue: Downregulated’ LogFC: Log2 of Fold Change

Enrichment Analyses

To compensate for the different numbers of differentially expressed genes in the two datasets when using 0.05 as the corrected p-value threshold (2,290 genes in KCL BrainBank and 402 in TargetALS), we applied an additional threshold of logFC > 0.2 to the KCL BrainBank dataset. This yielded differential expression for 1,013 genes, which were uploaded to MetaCore for enrichment analysis along with the 402 differentially expressed genes in TargetALS. We performed analyses using MetaCore Pathway Maps, Process Networks, Disease Networks and GO processes. Supplementary Tables 2 and 3 list all enrichments with their p-value, false FDR adjusted p-value, and the genes from each dataset contributing to each enrichment.

Process Networks: neuropeptide signalling is altered in ALS in both datasets

Figures 2A and 2B display the most significant MetaCore Process Networks identified in the two datasets. Among the most significant processes in KCL BrainBank are several neurological-related processes (Transmission of nerve impulse, Neuropeptide signalling pathways and Development Neurogenesis Axonal guidance) and processes related to muscle contraction. The most significant process network in TargetALS is Signal Transduction Neuropeptide signalling pathways, which was the third most significant KCL BrainBank process network. In TargetALS, among the most enriched networks (albeit not significant) are several processes related to inflammation (interferon signalling and NK cell cytotoxicity).

Figure 2.
Figure 2.

Enrichment analysis results in the two datasets. The ten most significant enrichments are shown, sorted by adjusted p-value, or fewer if there are less than ten significant after multiple testing correction. A) KCL BrainBank MetaCore Process Networks; B) TargetALS MetaCore Process Networks; C) KCL BrainBank MetaCore Pathway Maps; D) TargetALS MetaCore Pathway Maps; E) KCL BrainBank Reactome; For clarity, in panel E, one of the two redundant enrichments was omitted (Respiratory electron transport, which is more generalised of the one included, Respiratory electron transport, ATP synthesis by chemiosmotic coupling, and heat production by uncoupling proteins). F) TargetALS Reactome. Following pathway abbreviations/names have been introduced: PF: Protein Folding; NP: Neurophysiological process; ST: Signal transduction; CS: Cytoskeleton; MC: Muscle contraction; POMC, alpha-MSH, AGRP, food intake, energy exp in hypothalamus: POMC, alpha-MSH and AGRP in regulation of food intake and energy expenditure in obesity in hypothalamus; Regulation of mp, excitability of cortical pyramidal neurons: Regulation of intrinsic membrane properties and excitability of cortical pyramidal neurons; Respiratory ET, ATP synthesis by CC, heat production: Respiratory electron transport, ATP synthesis by chemiosmotic coupling, and heat production by uncoupling proteins.

Pathway Maps: KCL BrainBank is enriched for neurophysiological processes, oxidative stress and signal transduction, Target ALS is for POMC processing

Figures 2C and 2D present the most significantly enriched pathway maps in the two datasets. Prominent themes in the KCL BrainBank enrichments were neurophysiological processes, oxidative stress and signal transduction. There were three significant TargetALS pathway maps; two related to pro-opiomelanocortin (POMC) processing and one to immune response. Interestingly, the most significant network in TargetALS, POMC processing (Figure 3), includes several agonists of MC4R which is significantly upregulated in both datasets.

Figure 3.
Figure 3.

Protein folding and maturation POMC processing pathway map from MetaCore. Blue barometers represent downregulated genes from the TargetALS dataset. The descriptions of what the other symbols represent are available in Supplementary Figure 1.

KCL BrainBank is enriched for nervous system-specific processes while TargetALS is for immunological processes

Enrichments for MetaCore Diseases by Biomarkers, GO Biological Processes and Reactome pathways are concordant in the two datasets. The KCL dataset showed clear nervous system related enrichments. These included disease enrichments (Heredodegenerative diseases, Huntington’s disease, Central Nervous System diseases among others), GO Processes relevant to nervous system development, and Reactome pathways related to the functioning of the nervous system, such as the transmission of chemical synapses. TargetALS was enriched for immune system related diseases and processes, including diseases connected with response to various viral infections, GO Processes related to immune defence, and Reactome pathways pertaining to interferon signalling (Figures 2E and 2F).

Correlation of neuropeptides and receptors expression with ALS phenotypic attributes

As neuropeptide signalling was the only altered biological network in both datasets, and neuropeptides have been implicated with the ALS clinical phenotype14-16, we decided to test the correlation between the expression of neuropeptides and their receptors with age at onset and disease duration of the ALS patients in both datasets. Several neuropeptides and receptors were found to be significantly correlated with age at onset and disease duration in at least one dataset (Table 3). NPBWR1 was found to be significantly correlated with age at onset in both KCL BrainBank and TargetALS after multiple testing correction, whilst TAC3 was nominally significantly correlated with age at onset in TargetALS and after multiple testing correction in KCL BrainBank. GNRH1 and TACR1 were nominally significantly correlated with survival and SSTR1 with age at onset in both datasets. Interestingly all neuropeptides and receptors that were significantly correlated with the age of onset of ALS in at least one dataset had a negative coefficient. Figure 4A shows NPBWR1 expression against age of onset. Complete results can be found in Supplementary Tables 4, 5, 6 and 7.

View this table:
Table 3.

Results of the correlation analysis between neuropeptides and receptors, and clinical features. Only results that are significant after FDR correction in at least one dataset, or nominally significant in both are displayed. Correlations that are at least nominally significant in both datasets are coloured in green, and those that are also significant after FDR correction in one dataset are also in bold.

Figure 4:
Figure 4:

A) Correlation analysis between NPBWR1 gene expression and age of onset in KCL BrainBank and TargetALS. B) Screenshot from ALS Gene Expression Explorer showing expression of MC4R, which was significantly upregulated in both datasets.

Cell Type Composition

Cell type composition analysis is shown in Figure 5. There was a significantly smaller proportion of endothelial cells in cases compared to controls in both datasets (KCL BrainBank: F = 8.673, p-value = 0.00369; TargetALS: F = 6.423, p-value = 0.0119). In the KCL dataset, cases also displayed higher neuronal cell (F = 25.256, p-value = 1.28 × 10−6) and lower oligodendrocyte levels (F = 13.633, p-value = 0.000301) whereas, in TargetALS, astrocytes were more prevalent amongst cases (F = 5.466, p-value = 0.0203). The full results are available in Supplementary Table 8.

Figure 5.
Figure 5.

Case-control cell composition analysis in KCL BrainBank (left panel) and TargetALS (right panel), using singular decomposition values generated by the BRETIGEA program for six cell types: A) Astrocytes, B) Endothelial Cells, C) Microglia, D) Neurons, E) Oligodendrocytes, F) Oligodendrocyte Progenitor Cells. Significance was assessed using one-way ANOVA. ns = non-significant, ns = non-significant, * < 0.05, ** < 0.01, *** < 0.001. Results were corrected for age and sex. Significance was assessed using one-way ANOVA. Results were corrected for age and sex. Controls are in red and cases in green.

ALS Gene Expression Explorer

The ALS Gene Expression Explorer web application (https://alsgeexplorer.er.kcl.ac.uk) enables users to explore and compare the expression levels for specific genes in ALS cases and controls in the two datasets. Figure 4B displays an example gene (MC4R) which is differentially expressed in both datasets.

Discussion

By leveraging motor cortex expression data from two independent large datasets (a total of 224 ALS cases and 79 controls), we reported ALS gene expression profiles that indicate the involvement of various biological processes in the pathology of ALS. We also identified genes in the neuropeptide signalling pathway that showed potential as biomarkers of ALS onset and progression. Differential expression analysis identified 2,290 and 402 differentially expressed genes in KCL BrainBank and TargetALS cases, respectively. Synapse-related processes, such as GABAergic neurotransmission, calcium-mediated signalling and postsynaptic signalling were significantly enriched in KCL BrainBank, whilst TargetALS carried an immune-system related signature, with strong enrichments for interferon signalling and NK cell neurotoxicity. This is most likely a product of the molecular heterogeneity of ALS, as these two profiles reflect previously hypothesised mechanisms of ALS pathogenesis and correspond to the predominant cell types identified in these datasets (neuronal cells in KCL BrainBank and astrocytes in TargetALS). Changes in the excitability patterns of GABAergic interneurons influence alterations in motor cortex circuitry, which has been proposed as a possible trigger for the spread of ALS pathology17,18. Previous work has also implicated NK cells, which secrete interferon-γ and are upregulated in the motor cortex and spinal cord of ALS patients19,20 as being modulators of motor neuron degeneration21, with interferon signalling pathways being upregulated in spinal cord astrocytes in a SOD1 ALS mouse model22.

Despite the diversity in the KCL BrainBank and TargetALS gene expression profiles, the neuropeptide signalling pathway was enriched in both datasets, which suggests that this is a key feature of ALS pathogenesis. The KCL BrainBank and TargetALS datasets exhibited both unique and shared differential expression of several neuropeptide ligands and receptors. One gene, melanocortin receptor 4 (MC4R), was significantly upregulated in both datasets. The role of the melanocortin system in ALS has largely been overlooked, despite MC4R activation inducing hypermetabolism, increased energy expenditure and cachexia; three processes associated with increased lower motor neuron weakness and poor prognosis in ALS patients23-27. Furthermore, MC4R-mediated signalling in the motor cortex also enhances the activity of sympathetic preganglionic neurons in the spinal cord, which modulates skeletal muscle contraction28,29, and rare (MAF <0.01) disruptive and damaging missense variants in MC4R might increase the risk of ALS (Beta = 2.15 SE = 1.52, p-value = 0.037, http://databrowser.projectmine.com). Therefore, we suggest further investigation on MC4R as a potential candidate target gene for future biomarker and screening studies.

Several MC4R agonists are involved in the trans-Golgi network protein folding and maturation pathway, which was the most significantly downregulated pathway in TargetALS. This pathway is controlled by the MC4R ligand pro-opiomelanocortin (POMC), which acts as a precursor peptide for the MC4R agonists30. Hence, aberrant protein degradation linked to both neuropeptide signalling dysregulation and endoplasmic reticulum associated stress could be a triggering event for ALS pathogenesis. We observed reduced expression of POMC in TargetALS (LogFC = -1.12, adjusted p-value = 3.6E-05), which would undoubtedly affect the processing of several peptides reliant on POMC cleavage in this pathway (β-endorphin, ⍰-MSH, β-MSH). Their expression is linked to anti-inflammatory processes in neurodegenerative diseases. For instance, β-endorphin, which normally reduces the production of pro-inflammatory cytokines such as interferon-γ, IL-1β and IL-231 is found at lower levels in the cerebrospinal fluid of Alzheimer’s disease patients32. Similarly, both ⍰-MSH and β-MSH are cytokine antagonists that protect against excitotoxic and apoptotic signalling effects in Parkinson’s and Alzheimer’s disease models33,34; actions that are also perturbed in ALS as evidenced by similarly enriched pathways in this study. Therefore, the effects that MC4R-POMC signalling has on inflammatory processes could also explain the significant enrichment of immune-related processes observed in the TargetALS dataset.

To further investigate how neuropeptide signalling affects clinical phenotype, we performed a correlation analysis of neuropeptides and their receptors (102 neurotransmitters and attributes present in KCL BrainBank and 111 in TargetALS) with clinical attributes. Several neuropeptides and receptors significantly correlated with age of onset and survival in both datasets. NPBWR1 was significantly correlated with the age of onset after multiple testing correction in both datasets (KCL BrainBank r = -0.34, adjusted p-value = 0.019; TargetALS r = -0.25, adjusted p-value = 0.02). Interestingly, NPBWR1 is expressed by GABAergic interneurons and when activated, can dampen their inhibitory effects35. Therefore, increased NPBWR1 expression could have a greater impact on the aforementioned alterations in motor cortex circuitry, which would corroborate with the earlier age of onset in patients with higher NPBWR1 levels. Likewise, expression of the somatostatin receptor SSTR1, which also resides on inhibitory GABAergic interneurons36, was nominally negatively correlated with age of onset in the KCL BrainBank. We also found a nominally significant correlation between TACR1 and GNRH1 levels and survival in both datasets. Both TACR1 and GNRH1 can stimulate the production and release of arachidonic acid37,38, a polyunsaturated fatty acid whose levels can distinguish ALS patients from mimics and controls and positively correlates with disease progression39, with higher levels also found in spinal cord motor neuron cultures40. Therefore, the role of the neuropeptide signalling system on lipid metabolism and cortical signalling in ALS is warranted to assess its full impact on clinical phenotype.

When reviewing the 44 differentially expressed genes which were shared in both datasets, we found that many were involved in integral themes of ALS pathogenesis, such as carbohydrate/lipid metabolism, mitochondrial energy homeostasis (SULT1A1, IP6K3, CPT1A)41-43, and neurovascular unit dysfunction (AQP1, APLNR, CA12, AGPT1, CST3)44-48. Further investigations of these genes are needed to determine if they are the drivers of alterations in these pathways in ALS.

Our study has some limitations which are primarily linked to the fact that we identified gene expression signatures in post-mortem tissue. Firstly, as the cells we derive expression for have survived the neurodegenerative processes of ALS, we cannot infer if the changes we are observing are causing or are a consequence of the ALS phenotype. We do not have information on the staging of ALS in these patients so cannot attribute the expression profiles to a certain point of ALS progression. Furthermore, there were 5.7 times more differentially expressed genes in KCL BrainBank than in TargetALS. Although this might be linked to the larger number of controls in the KCL datasets (∼3 times more than in TargetALS), there could also be several external factors influencing these datasets, such as library preparation, RNA extraction, sample quality, and population variability in gene expression.

In summary, we identified several key modulatory themes of ALS pathogenesis as well as identified genes in the neuropeptide signalling pathway that showed potential as diagnostic and prognostic biomarkers. There is a great need for accurate biomarkers in ALS, particularly for biomarkers that predict the age of onset. These could be explored to inform the study of the pre-symptomatic stages of ALS and support the development of pre-symptomatic diagnostic approaches49.

Data Availability

KCL BrainBank RNAseq data is available, on reasonable request, from the corresponding author. All raw RNAseq data from TargetALS samples can be requested by emailing cgnd_help@nygenome.org. Results for individual genes can be visually explored at https://alsgeexplorer.er.kcl.ac.uk/. All data analysis scripts used in this study are available at https://github.com/rkabiljo/RNASeq_Genes_ERVs and https://github.com/rkabiljo/DifferentialExpression_Genes.

https://alsgeexplorer.er.kcl.ac.uk/

https://github.com/rkabiljo/RNASeq_Genes_ERVs

https://github.com/rkabiljo/DifferentialExpression_Genes

Ethics declaration

Post-mortem tissue samples from King’s College London were collected under the ethical approval of the MRC London Neurodegenerative Diseases Brain Bank and under the regulations of the Human Tissue Act UK 2014. All post-mortem tissue was donated to the MRC London Neurodegenerative Diseases Brain Bank under standard ethical and Human Tissue Act procedures, with informed consent provided by the next of kin. Data generated from this material were anonymized and analysed on a high-performance computing cloud (https://www.maudsleybrc.nihr.ac.uk/facilities/rosalind/) with data protection protocols in accordance with Department of Health Policy (UK) and the security standards set by the National Data Guardian. Ethical approval to process and analyse post-mortem samples stored at King’s College London was provided by a local ethics committee at the Institute of Psychiatry, Psychology & Neuroscience, King’s College London, and the MRC London Neurodegenerative Diseases Brain Bank.

Data availability

KCL BrainBank RNAseq data is available, on reasonable request, from the corresponding author. All raw RNAseq data from TargetALS samples can be requested by emailing cgnd_help@nygenome.org. Results for individual genes can be visually explored at https://alsgeexplorer.er.kcl.ac.uk/. All data analysis scripts used in this study are available at https://github.com/rkabiljo/RNASeq_Genes_ERVs and https://github.com/rkabiljo/DifferentialExpression_Genes.

Acknowledgements

The TargetALS Human Postmortem Tissue Core, New York Genome Center for Genomics of Neurodegenerative Disease, the Amyotrophic Lateral Sclerosis Association and the Tow Foundation.

Human post-mortem tissue was obtained from the Medical Research Council London Neurodegenerative Diseases Brain Bank (King’s College London). The London Neurodegenerative Diseases Brain Bank at KCL has received funding from the MRC and through the Brains for Dementia Research project (jointly funded by Alzheimer’s Society and Alzheimer’s Research UK)’.

This work was carried out on King’s College HPC Create: King’s College London. (2022). King’s Computational Research, Engineering and Technology Environment (CREATE). Retrieved March 2, 2022, from https://doi.org/10.18742/rnvf-m076, and on the Pawsey Supercomputing Research Centre with funding from the Australian Government and the Government of Western Australia.

RK is funded by MND Scotland, grant number RE14937.

HM is funded by GlaxoSmithKline and the KCL funded centre for Doctoral Training (CDT) in Data-Driven Health.

GH is funded by Perron Institute for Neurological and Translational Science and the KCL funded centre for Doctoral Training (CDT) in Data-Driven Health.RJBD is supported by the following: (1) NIHR Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London, London, UK; (2) Health Data Research UK, which is funded by the UK Medical Research Council, Engineering and Physical Sciences Research Council, Economic and Social Research Council, Department of Health and Social Care (England), Chief Scientist Office of the Scottish Government Health and Social Care Directorates, Health and Social Care Research and Development Division (Welsh Government), Public Health Agency (Northern Ireland), British Heart Foundation and Wellcome Trust; (3) The BigData@Heart Consortium, funded by the Innovative Medicines Initiative-2 Joint Undertaking under grant agreement No. 116074. This Joint Undertaking receives support from the European Union’s Horizon 2020 research and innovation programme and EFPIA; it is chaired by DE Grobbee and SD Anker, partnering with 20 academic and industry partners and ESC; (4) the National Institute for Health Research University College London Hospitals Biomedical Research Centre; (5) the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London; (6) the UK Research and Innovation London Medical Imaging & Artificial Intelligence Centre for Value Based Healthcare; (7) the National Institute for Health Research (NIHR) Applied Research Collaboration South London (NIHR ARC South London) at King’s College Hospital NHS Foundation Trust.

AAC is an NIHR Senior Investigator (NIHR202421). This research is part an EU Joint Programme - Neurodegenerative Disease Research (JPND) project. The project is supported through the following funding organisations under the aegis of JPND: Medical Research Council; Economic and Social Research Council and the Motor Neurone Disease Association. Funding for open access charge: UKRI.

AI is funded by the Motor Neurone Disease Association. This study represents independent research partly funded by the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London. BA acknowledges funding through a Pre-doctoral Fellowship from the NIHR (NIHR301067). We acknowledge use of the research computing facility at King’s College London, Rosalind (https://rosalind.kcl.ac.uk), which is delivered in partnership with the National Institute for Health Research (NIHR) Biomedical Research Centres at South London & Maudsley and Guy’s & St. Thomas’ NHS Foundation Trusts and part-funded by capital equipment grants from the Maudsley Charity (award 980) and Guy’s and St Thomas’ Charity (TR130505). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, King’s College London, or the Department of Health and Social Care.

Footnotes

  • Just corrected a typo in the author list

References

  1. 1.↵
    Julien JP. Amyotrophic lateral sclerosis. unfolding the toxicity of the misfolded. Cell. Feb 23 2001;104(5):581–91. doi:10.1016/s0092-8674(01)00244-6
    OpenUrlCrossRefPubMedWeb of Science
  2. 2.↵
    Rowland LP, Shneider NA. Amyotrophic lateral sclerosis. N Engl J Med. May 31 2001;344(5):1688–700. doi:10.1056/NEJM200105313442207
    OpenUrlCrossRefPubMedWeb of Science
  3. 3.↵
    Wang XS, Simmons Z, Liu W, Boyer PJ, Connor JR. Differential expression of genes in amyotrophic lateral sclerosis revealed by profiling the post mortem cortex. Amyotroph Lateral Scler. Dec 2006;7(4):201–10. doi:10.1080/17482960600947689
    OpenUrlCrossRefPubMedWeb of Science
  4. 4.↵
    Dols-Icardo O, Montal V, Sirisi S, et al. Motor cortex transcriptome reveals microglial key events in amyotrophic lateral sclerosis. Neurol Neuroimmunol Neuroinflamm. Sep 2020;7(5) doi:10.1212/NXI.0000000000000829
    OpenUrlAbstract/FREE Full Text
  5. 5.↵
    Iacoangeli A, Fogh I, Selvackadunco S, et al. SCFD1 expression quantitative trait loci in amyotrophic lateral sclerosis are differentially expressed. Brain Commun. 2021;3(4):fcab236. doi:10.1093/braincomms/fcab236
    OpenUrlCrossRef
  6. 6.↵
    Jones AR, Iacoangeli A, Adey BN, et al. A HML6 endogenous retrovirus on chromosome 3 is upregulated in amyotrophic lateral sclerosis motor cortex. Sci Rep. Jul 12 2021;11(5):14283. doi:10.1038/s41598-021-93742-3
    OpenUrlCrossRef
  7. 7.↵
    Brian B. BBMap: A Fast, Accurate, Splice-Aware Aligner. Lawrence Berkeley National Lab (LBNL), Berkeley, CA (United States). 2014;
  8. 8.↵
    Dobin A, Davis CA, Schlesinger F, et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. Jan 1 2013;29(5):15–21. doi:10.1093/bioinformatics/bts635
    OpenUrlCrossRefPubMedWeb of Science
  9. 9.↵
    Anders S, Pyl PT, Huber W. HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics. Jan 15 2015;31(5):166–9. doi:10.1093/bioinformatics/btu638
    OpenUrlCrossRefPubMedWeb of Science
  10. 10.↵
    Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. doi:10.1186/s13059-014-0550-8
    OpenUrlCrossRefPubMed
  11. 11.↵
    Raudvere U, Kolberg L, Kuzmin I, et al. g:Profiler: a web server for functional enrichment analysis and conversions of gene lists (2019 update). Nucleic Acids Res. Jul 2 2019;47(W1):W191–W198. doi:10.1093/nar/gkz369
    OpenUrlCrossRefPubMed
  12. 12.↵
    Gillespie M, Jassal B, Stephan R, et al. The reactome pathway knowledgebase 2022. Nucleic Acids Res. Jan 7 2022;50(D1):D687–D692. doi:10.1093/nar/gkab1028
    OpenUrlCrossRefPubMed
  13. 13.↵
    McKenzie AT, Wang M, Hauberg ME, et al. Brain Cell Type Specific Gene Expression and Co-expression Network Architectures. Sci Rep. Jun 11 2018;8(5):8868. doi:10.1038/s41598-018-27293-5
    OpenUrlCrossRefPubMed
  14. 14.↵
    Ahmed RM, Phan K, Highton-Williamson E, et al. Eating peptides: biomarkers of neurodegeneration in amyotrophic lateral sclerosis and frontotemporal dementia. Ann Clin Transl Neurol. Mar 2019;6(3):486–495. doi:10.1002/acn3.721
    OpenUrlCrossRef
  15. 15.
    Clark CM, Clark RM, Hoyle JA, Dickson TC. Pathogenic or protective? Neuropeptide Y in amyotrophic lateral sclerosis. J Neurochem. Feb 2021;156(3):273–289. doi:10.1111/jnc.15125
    OpenUrlCrossRef
  16. 16.↵
    Zuccaro E, Piol D, Basso M, Pennuto M. Motor Neuron Diseases and Neuroprotective Peptides: A Closer Look to Neurons. Front Aging Neurosci. 2021;13:723871. doi:10.3389/fnagi.2021.723871
    OpenUrlCrossRef
  17. 17.↵
    Eisen A, Braak H, Del Tredici K, Lemon R, Ludolph AC, Kiernan MC. Cortical influences drive amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry. Nov 2017;88(11):917–924. doi:10.1136/jnnp-2017-315573
    OpenUrlAbstract/FREE Full Text
  18. 18.↵
    Scekic-Zahirovic J, Fischer M, Stuart-Lopez G, et al. Evidence that corticofugal propagation of ALS pathology is not mediated by prion-like mechanism. Prog Neurobiol. May 2021;200:101972. doi:10.1016/j.pneurobio.2020.101972
    OpenUrlCrossRef
  19. 19.↵
    Gustafson MP, Staff NP, Bornschlegl S, et al. Comprehensive immune profiling reveals substantial immune system alterations in a subset of patients with amyotrophic lateral sclerosis. PLoS One. 2017;12(7):e0182002. doi:10.1371/journal.pone.0182002
    OpenUrlCrossRef
  20. 20.↵
    Murdock BJ, Zhou T, Kashlan SR, Little RJ, Goutman SA, Feldman EL. Correlation of Peripheral Immunity With Rapid Amyotrophic Lateral Sclerosis Progression. JAMA Neurol. Dec 1 2017;74(5):1446–1454. doi:10.1001/jamaneurol.2017.2255
    OpenUrlCrossRefPubMed
  21. 21.↵
    Garofalo S, Cocozza G, Porzia A, et al. Natural killer cells modulate motor neuron-immune cell cross talk in models of Amyotrophic Lateral Sclerosis. Nat Commun. Apr 14 2020;11(5):1773. doi:10.1038/s41467-020-15644-8
    OpenUrlCrossRef
  22. 22.↵
    Wang R, Yang B, Zhang D. Activation of interferon signaling pathways in spinal cord astrocytes from an ALS mouse model. Glia. Jun 2011;59(6):946–58. doi:10.1002/glia.21167
    OpenUrlCrossRefPubMedWeb of Science
  23. 23.↵
    Holm T, Maier A, Wicks P, et al. Severe loss of appetite in amyotrophic lateral sclerosis patients: online self-assessment study. Interact J Med Res. Apr 17 2013;2(5):e8. doi:10.2196/ijmr.2463
    OpenUrlCrossRef
  24. 24.
    Jesus P, Fayemendy P, Nicol M, et al. Hypermetabolism is a deleterious prognostic factor in patients with amyotrophic lateral sclerosis. Eur J Neurol. Jan 2018;25(1):97–104. doi:10.1111/ene.13468
    OpenUrlCrossRefPubMed
  25. 25.
    Kasarskis EJ, Mendiondo MS, Matthews DE, et al. Estimating daily energy expenditure in individuals with amyotrophic lateral sclerosis. Am J Clin Nutr. Apr 2014;99(4):792–803. doi:10.3945/ajcn.113.069997
    OpenUrlAbstract/FREE Full Text
  26. 26.
    Kurihara M, Bamba S, Yasuhara S, et al. Factors Affecting Energy Metabolism and Prognosis in Patients with Amyotrophic Lateral Sclerosis. Ann Nutr Metab. 2021;77(4):236–243. doi:10.1159/000518908
    OpenUrlCrossRef
  27. 27.↵
    Steyn FJ, Ioannides ZA, van Eijk RPA, et al. Hypermetabolism in ALS is associated with greater functional decline and shorter survival. J Neurol Neurosurg Psychiatry. Oct 2018;89(10):1016–1023. doi:10.1136/jnnp-2017-317887
    OpenUrlAbstract/FREE Full Text
  28. 28.↵
    Sohn JW, Harris LE, Berglund ED, et al. Melanocortin 4 receptors reciprocally regulate sympathetic and parasympathetic preganglionic neurons. Cell. Jan 31 2013;152(5):612–9. doi:10.1016/j.cell.2012.12.022
    OpenUrlCrossRefPubMedWeb of Science
  29. 29.↵
    Ye DW, Liu C, Liu TT, Tian XB, Xiang HB. Motor cortex-periaqueductal gray-spinal cord neuronal circuitry may involve in modulation of nociception: a virally mediated transsynaptic tracing study in spinally transected transgenic mouse model. PLoS One. 2014;9(2):e89486. doi:10.1371/journal.pone.0089486
    OpenUrlCrossRef
  30. 30.↵
    Harno E, Gali Ramamoorthy T, Coll AP, White A. POMC: The Physiological Power of Hormone Processing. Physiol Rev. Oct 1 2018;98(5):2381–2430. doi:10.1152/physrev.00024.2017
    OpenUrlCrossRefPubMed
  31. 31.↵
    Pilozzi A, Carro C, Huang X. Roles of beta-Endorphin in Stress, Behavior, Neuroinflammation, and Brain Energy Metabolism. Int J Mol Sci. Dec 30 2020;22(1) doi:10.3390/ijms22010338
    OpenUrlCrossRef
  32. 32.↵
    Jolkkonen JT, Soininen HS, Riekkinen PJ. beta-Endorphin-like immunoreactivity in cerebrospinal fluid of patients with Alzheimer’s disease and Parkinson’s disease. J Neurol Sci. Feb 1987;77(2-3):153–9. doi:10.1016/0022-510x(87)90118-3
    OpenUrlCrossRefPubMed
  33. 33.↵
    Giuliani D, Neri L, Canalini F, et al. NDP-alpha-MSH induces intense neurogenesis and cognitive recovery in Alzheimer transgenic mice through activation of melanocortin MC4 receptors. Mol Cell Neurosci. Jul 2015;67:13–21. doi:10.1016/j.mcn.2015.05.004
    OpenUrlCrossRef
  34. 34.↵
    Peng T, Wang J, Lu J, Lu H, Teng J, Jia Y. Neuroprotective effects of alpha-melanocyte-stimulating hormone against the neurotoxicity of 1-methyl-4-phenylpyridinium. IUBMB Life. May 2017;69(5):315–320. doi:10.1002/iub.1385
    OpenUrlCrossRef
  35. 35.↵
    Chottova Dvorakova M. Distribution and Function of Neuropeptides W/B Signaling System. Front Physiol. 2018;9:981. doi:10.3389/fphys.2018.00981
    OpenUrlCrossRef
  36. 36.↵
    Lukomska A, Dobrzanski G, Liguz-Lecznar M, Kossut M. Somatostatin receptors (SSTR1-5) on inhibitory interneurons in the barrel cortex. Brain Struct Funct. Jan 2020;225(1):387–401. doi:10.1007/s00429-019-02011-7
    OpenUrlCrossRefPubMed
  37. 37.↵
    Naor Z, Catt KJ. Mechanism of action of gonadotropin-releasing hormone. Involvement of phospholipid turnover in luteinizing hormone release. J Biol Chem. Mar 10 1981;256(5):2226–9.
    OpenUrlAbstract/FREE Full Text
  38. 38.↵
    Steinhoff MS, von Mentzer B, Geppetti P, Pothoulakis C, Bunnett NW. Tachykinins and their receptors: contributions to physiological control and the mechanisms of disease. Physiol Rev. Jan 2014;94(1):265–301. doi:10.1152/physrev.00031.2013
    OpenUrlCrossRefPubMed
  39. 39.↵
    Lawton KA, Brown MV, Alexander D, et al. Plasma metabolomic biomarker panel to distinguish patients with amyotrophic lateral sclerosis from disease mimics. Amyotroph Lateral Scler Frontotemporal Degener. Sep 2014;15(5-6):362–70. doi:10.3109/21678421.2014.908311
    OpenUrlCrossRefPubMed
  40. 40.↵
    Lee H, Lee JJ, Park NY, et al. Multi-omic analysis of selectively vulnerable motor neuron subtypes implicates altered lipid metabolism in ALS. Nat Neurosci. Dec 2021;24(12):1673–1685. doi:10.1038/s41593-021-00944-z
    OpenUrlCrossRef
  41. 41.↵
    Sidharthan NP, Minchin RF, Butcher NJ. Cytosolic sulfotransferase 1A3 is induced by dopamine and protects neuronal cells from dopamine toxicity: role of D1 receptor-N-methyl-D-aspartate receptor coupling. J Biol Chem. Nov 29 2013;288(5):34364–74. doi:10.1074/jbc.M113.493239
    OpenUrlAbstract/FREE Full Text
  42. 42.
    Dato S, Crocco P, De Rango F, et al. IP6K3 and IPMK variations in LOAD and longevity: Evidence for a multifaceted signaling network at the crossroad between neurodegeneration and survival. Mech Ageing Dev. Apr 2021;195:111439. doi:10.1016/j.mad.2021.111439
    OpenUrlCrossRef
  43. 43.↵
    Trabjerg MS, Andersen DC, Huntjens P, et al. Downregulating carnitine palmitoyl transferase 1 affects disease progression in the SOD1 G93A mouse model of ALS. Commun Biol. Apr 30 2021;4(5):509. doi:10.1038/s42003-021-02034-z
    OpenUrlCrossRef
  44. 44.↵
    Bonomini F, Rezzani R. Aquaporin and blood brain barrier. Curr Neuropharmacol. Jun 2010;8(2):92–6. doi:10.2174/157015910791233132
    OpenUrlCrossRefPubMed
  45. 45.
    Siddiqui MR, Mayanil CS, Kim KS, Tomita T. Angiopoietin-1 Regulates Brain Endothelial Permeability through PTPN-2 Mediated Tyrosine Dephosphorylation of Occludin. PLoS One. 2015;10(6):e0130857. doi:10.1371/journal.pone.0130857
    OpenUrlCrossRef
  46. 46.
    Benndorf RA. Renal Biomarker and Angiostatic Mediator? Cystatin C as a Negative Regulator of Vascular Endothelial Cell Homeostasis and Angiogenesis. J Am Heart Assoc. Nov 6 2018;7(5):e010997. doi:10.1161/JAHA.118.010997
    OpenUrlCrossRef
  47. 47.
    Hu G, Wang Z, Zhang R, Sun W, Chen X. The Role of Apelin/Apelin Receptor in Energy Metabolism and Water Homeostasis: A Comprehensive Narrative Review. Front Physiol. 2021;12:632886. doi:10.3389/fphys.2021.632886
    OpenUrlCrossRef
  48. 48.↵
    Hosford PS, Wells JA, Nizari S, et al. CO(2) signaling mediates neurovascular coupling in the cerebral cortex. Nat Commun. Apr 19 2022;13(5):2125. doi:10.1038/s41467-022-29622-9
    OpenUrlCrossRef
  49. 49.↵
    Benatar M, Wuu J, McHutchison C, et al. Preventing amyotrophic lateral sclerosis: insights from pre-symptomatic neurodegenerative diseases. Brain. Mar 29 2022;145(5):27–44. doi:10.1093/brain/awab404
    OpenUrlCrossRefPubMed
Back to top
Posted May 08, 2023.
Download PDF

Supplementary Material

Data/Code
Email
Transcriptomics Analyses of ALS Post-mortem Motor Cortex highlight alteration and potential biomarkers in the Neuropeptide Signalling pathway
Renata Kabiljo, Heather Marriott, Guy P Hunt, Abigail L Pfaff, Ahmad Al Khleifat, Brett Adey, Ashley Jones, Claire Troakes, John P Quinn, Richard J B Dobson, Sulev Koks, Ammar Al-Chalabi, Alfredo Iacoangeli
medRxiv 2023.05.05.23289551; doi: https://doi.org/10.1101/2023.05.05.23289551
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area