Theme 02 - Genetics and Genomics

 

 

GEN-01 A case of an uncommon late onset of ALS associated with an undescribed SOD1 [email protected]

Abstract

Background

Amyotrophic Lateral Sclerosis (ALS) is a fatal neurodegenerative disorder involving upper and lower motor neurons. It is mostly sporadic, tough pathogenic mutations are sometime detectable, mainly in those patients with a positive familiar track record for neurodegenerative disorders. More than 50 genes have been put in relation with ALS so far, among which SOD1, C9Orf72, FUS, ATXN2 and TARDBP are the most frequent. SOD1, encoding for the antioxidant enzyme Cu/ Zn superoxide dismutase and described for the first time in 1993, is the first gene that had been put in relation with ALS. To date, over 180 different SOD1 mutations have been described throughout the five exons of the SOD1 gene. However, the exact pathogenic effect of this gene mutations are still unclear. Recently, a gene therapy targeting SOD1 has been approved for ALS patients carrying any SOD1 mutation. This new therapeutic approach has boosted the number of genetic tests performed, sometimes revealing unexpected positivity and sometimes underlying new genes’ mutations. Here we describe a case of a woman with a new SOD1 mutation, not described in literature before and considered as pathogenetic in VarSome.

Case report

A 81 years old woman came to our attention at Santa Chiara Hospital, Pisa, Italy, for the presence of dysarthria onset 3 years prior. She had no familiar history for neurodegenerative diseases, excepting for gait disorder in one sister. She underwent a complete neurological balance disclosing a severe bulbar involvement with anarthria, tongue hyposthenia and oral apraxia. She presented mild upper limbs hyposthenia, limbs spasticity and lower limbs hyperreflexia as well. An electromyogram showed a diffuse neurogenic pattern with fasciculation potentials on all tested muscles and acute denervation on tongue muscles. A complete paraneoplastic, rheumatological and virological screening resulted negative, as well as anti-nerve and anti-neuron antibodies. CSF findings were unremarkable. A diagnosis of ALS was made, therefore genetic tests for ALS-related genes were performed. The NGS showed a heterozygous mutation in SOD1 (c.292_314del, pVal98LeufsTer17 esone4), not reported in databases but predicted to be pathogenic in VarSome. Afterwards, the patient’s clinical picture progressively worsened and after a few months she died.

Conclusions

The occurrence of gene therapy has increased the amount of genetic tests for ALS-related genes, often revealing unexpected positivity. The analysis of large panels of patients may increase our knowledge of pathogenic mechanisms and contribute in discovering new genes or new mutations involved in disease development.

GEN-02 A novel mutation in ZFYVE26 gene (c.195-1G>A) in a young man affected by spastic [email protected]

Abstract

Hereditary spastic paraplegia (HSP) is a heterogeneous group of genetic neurodegenerative disorders characterized by progressive weakness and spasticity of lower limbs. SPG15 is a rare form of autosomal recessive HSP due to mutations in ZFYVE26 gene, which encodes for spastizin, involved in autophagy and endocytosis (1).

Here we describe the case of a 56 years old man who presented to our clinic in November 2020 referring a progressive motor clumsiness of both legs and walking instability during the last five years. No motor impairment at upper limb, no bulbar nor respiratory symptoms. These symptoms had a progressive evolution. In anamnesis only high blood pressure. No familiar history of neurodegenerative diseases. Neurological examination showed a spastic gait, a light hypotrophy and hyposthenia of the left leg, brisk reflexes at lower limb, no Hoffmann nor Babinski signs, no urinary nor intestinal problems. He underwent: blood examination (resulted normal, no HIV, HTLV infections), electroneurography/electromyography study (signs of chronic neurogenic muscle damage at lower limbs), brain MRI (normal), cervical MRI (normal medulla, small discal hernias without radicular or medulla compression), neuropsychological tests (executive and visuo-spatial impairments). On the basis of these examinations, we suspected a primary form of upper motor neuron disease. Therefore, we prescribed rehabilitation (with partial improvement of gait disturbances) and we executed a genetic examination in the suspect of a rare form of HSP. We observed a homozygous mutation in ZFYVE26 gene, which is the cause of a rare form of HSP (the SPG15). In detail, the mutation we found (c.195-1G>A) is a novel deletion in ZFYVE26 gene, never described in literature. This is a mutation in a splicing site and an in-silico evaluation demonstrated its pathogenic role. The clinical features of our patient are superimposable to other cases of SPG15.

Acknowledgements

We thank the patient involved in the study and his family for the participation and support to MND research.

GEN-03 Genetic testing of people with early-onset MND in the West of [email protected]

Abstract

Background

Motor neuron disease (MND) is heterogeneous in presentation, with a mean age of onset of 60 years (1). Earlier onset is thought to be associated with a genetic cause (1). Consequently, testing is recommended in those <50 years old irrespective of family history (2). Accordingly, a clinic was established in the West of Scotland (WoS) in 2018 to offer genetic testing to people with early-onset apparently sporadic MND. In this quality improvement project we evaluated this service in the context of genetic testing of people with early-onset MND over a 20 year period.

Methods

People with MND <50 years onset in the WoS were identified using the Clinical Audit Research and Evaluation of MND (CARE-MND) register. We conducted a retrospective analysis of clinical and genetic case notes. Details regarding demographics, genetic testing, and test results were collected.

Results

196 of the CARE-MND cohort (n = 2545) had disease onset <50 years (7.7%). 76 resided in the WoS (38.8%). 52 were male (68.4%). Mean age of onset was 42 years. 12 (15.8%) had a family history of MND. Genetic testing was discussed in 52 cases (75%) and 33 individuals (43.4%) underwent genetic testing. A pathological variant was found in 11 (33.3% of tests): six SOD1 variants, one FUS variant and four pathogenic C9orf72 expansions. All 12 patients with a family history of MND had genetic testing and a mutation was identified in 11 (91.6%). 21 people with apparently sporadic disease were tested and no mutations found. Rate of testing was 1.3 tests per year before 2018 and 2.8 tests per year after 2018. Of those tested, only 33.3% underwent multi-gene panel testing; the remainder had focussed gene testing.

Discussion

A recent review estimated that rate of “pick-up” of genetic mutations in people with a family history was ∼70%, whereas in apparently sporadic individuals ∼10% (3). Our data suggest that the early-onset population is enriched for pathogenic mutations but those with apparently sporadic disease have a lower than expected testing yield. Increasing availability of wider genetic panels may confound our results and yield of testing with contemporary approaches may increase. Our data supports testing of individuals with early-onset disease but does not provide strong evidence for testing those without a family history. Interval re-evaluation of this cohort may be of benefit to appreciate the utility of an extended panel (26 genes plus C9orf72 established in 2021). Systematic questioning regarding family history of dementia, other neurological and psychiatric diseases may also help to identify at risk individuals with early-onset disease.

GEN-04 The role of nicotinic receptor genes (CHRN) in amyotrophic lateral [email protected]

Abstract

Background

Neuronal nicotinic acetylcholine receptors (nAChRs) are ligand-gated ion channels encoded by different nicotinic cholinergic receptor (CHRN) genes and expressed in both bulbar and spinal motor neurons. They are involved in neuroprotection and in control of the release of many neurotransmitters, including glutamate. It has been previously postulated that rare variants in the subunits of nAChRs might represent one of several genetic risk factors for Amyotrophic Lateral Sclerosis (ALS) (1).

Objective

The aim of this study was to elucidate the contribution of genetic variation of CHRN genes to disease risk in ALS.

Methods

We analyzed 935 sporadic ALS patients of Italian origin from the Piemonte and Aosta Register for ALS and 775 ethnically and geographically matched controls. We tested for associations of variants in the CHRN genes cluster using linear regression for common variants and burden analysis for rare (minor allele frequency, MAF, <1%) variants.

Results

No significant association between CHRN genes mutations and ALS was found after appropriate correction for multiple testing. We did not observe a significant excess of rare nonsynonymous variants or an aggregate contribution of rare and common coding variants to the risk for ALS. No effect was detected when applying the combined test at the gene cluster level.

Conclusion

Our results indicate that common and rare genetic variations in the CHRN gene cluster do not contribute to ALS pathogenesis. However, further studies are needed to understand the possible effect of genetic variability in nAChRs to motor neuron degeneration.

GEN-05 Investigating the genetic causes of primary lateral [email protected]

Abstract

Introduction

PLS is the rarest form of MND, a condition characterized by progressive muscle weakness leading to an insidious loss of mobility. It predominantly affects the UMN system. The disease progression in PLS occurs at a significantly lower rate in comparison to ALS and typically the life span is not shortened (1). One of the biggest challenges faced by people with PLS is delayed diagnosis and misdiagnosis, since the initial symptoms can be similar to UMN-dominant ALS (2). In addition, due to the rare nature of the disease, very few genetic studies have been performed on the PLS and a handful of genes are associated with PLS. In the absence of a concrete genetic test that differentiates PLS from other MNDs, this delay in diagnosis is inevitable.

Objectives

The aim of our study is to investigate the genetic causes of PLS by generating the first large scale genetic dataset of PLS patients.

Methods

Through the international Project MinE and AnswerALS initiative, we got access to the whole genome sequencing and clinical data of 120 PLS patients, 82 with “definite” PLS diagnosis, 6550 ALS patients and 2444 controls. To our knowledge this is the largest cohort of PLS patient genomic data. The cohort consisted of 68 male and 58 female patients. Due to the rare nature of PLS we employed the candidate gene-based approach Variant calling. Our candidate genes included 154 genes that were previously associated with MND, obtained from AlSoD database. To this we added the genes ERLIN2, PARK2, MMACHC, SYNE2 and TTBK2, which were gathered by an exhaustive literature review of previous genetic studies associated with PLS.

Results

We identified known pathogenic variants in FIG4, FUS, SPG7, SPG11, SQSTM1 genes in five different PLS patients. In addition, we identified predicted pathogenic variants in ERLIN1 and UNC13A genes in two different PLS patients. We also investigated the role of pathogenic expansions in PLS. In our PLS cohort no pathogenic expansions were found in the ATXN2 & C9orf72. Finally, we observed that the mean age of onset was 58.4 years and 60.05 years for our PLS and ALS cohort, respectively. This contrasts with the previously reported mean age of onset of 50 years and 60 years between the PLS and ALS patients, respectively.

Funding

M.K. and A.I. are supported by South London and Maudsley NHS Foundation Trust; MND Scotland; Motor Neurone Disease Association; National Institute for Health Research; Darby Rimmer MND Foundation; Spastic Paraplegia Foundation and Rosetrees Trust.

GEN-06 Genetics in ALS: A change in attitude, no change in [email protected]

Abstract

Introduction

Until the emergence of specific drugs for certain genetic pathologies, the request for a genetic study was basically focused on the possibility of being able to carry out adequate genetic counselling. Now that it seems that the antisense oligonucleotide tofersen will soon be available for forms of ALS due to mutations in the copper-zinc superoxide dismutase (SOD1) gene, many scientific societies recommend carrying out a genetic study, at least of this SOD1 gene, in patients with ALS.

Materials and methods

We collected transversally the results of the genetic studies carried out on our patients, in June 2022, and compared them with the results in 2019.

Results

Of the 36 patients who are being followed up in our ALS unit, only two (5.56%) are positive for some type of genetic mutation (alterations in SOD 1 and in compound heterozygosity for TANK binding kinase (TBK1) and angiogenin (ANG)), after sequencing in 25 (69.44%) of them, a wide ALS panel, as well as copy number determination for the C9orf72 gene. In absolute values, these are the same patients who had genetic alterations in the past, when compared with the results of 2019. In that moment, only those with a family history underwent a genetic study (in this case, two patients had alterations in C9orf72 and another in SOD1). In addition, patients with genetic alterations in our current cohort also have relevant family history.

Conclusions

It seems that despite carrying out more genetic studies, no more patients with ALS due to genetic causes are discovered. However, despite this, it seems reasonable to continue with this behavior, since although the probability is very low (and this must be correctly informed to the patient), the detection of mutations in SOD1, at the present time, can determine a very relevant change in the prognosis of the disease.

GEN-07 A genome-wide association study on the New England and Ohio Amyotrophic Lateral Sclerosis [email protected]

Abstract

Background

Amyotrophic lateral sclerosis (ALS) is a rare and fatal neurodegenerative disease. Patients experience progressive motor neuron degeneration, leading to paralysis and death, often from respiratory failure within 3–5 years of diagnosis (1). While approximately 10% of ALS cases are familial, the genetic architecture of ALS is still largely unknown, particularly in conjunction with local environmental risk factors that may increase susceptibility to disease.

Methods

To address this gap in understanding, we conducted a genome-wide association study to understand the genetic risk factors using a regional cohort of 435 ALS cases and 279 controls primarily based in New England and Ohio. We performed SNP genotyping of 714 subjects using the NeuroChip array, which focuses on curated variants implicated in neurological diseases (Illumina Inc.). To minimize the effects of population stratification, we excluded subjects of non-European descent from the analysis. Standard quality control procedures on the genomic data left 242,090 SNPs included in the analysis. We used covariate adjusted logistic regression to screen all the SNPs’ associations with ALS case-control status. Wald test was used to obtain the p-values followed by Bonferroni correction for multiple comparisons. Covariates included sex, age at symptom onset and the first 10 principal components. The genomic inflation factor was 1.02 after adjusting for covariates.

Results

We found a marginally significant (Bonferroni adjusted p-value <0.1) SNP in the ALS-linked gene, TARDBP: Rs367543041. This SNP was previously reported in association with ALS and was found to share a close affinity with the Sardinian haplotype.

Discussion

Our results support future evaluation of genetic risk factors in regional cohorts to better understand potential genetic contributors to ALS risk as well as possible disease susceptibility in combination with local environmental and lifestyle factors.

Funding

This study was funded by a grant from the Centers for Disease Control and Prevention ATSDR 1R01TS000245 and by the Intramural Research Program of the NIH, National Institute on Aging [Z01-AG000949-02].

GEN-08 Unraveling clinico-pathological heterogeneity in motor neuron diseases using long-read [email protected]

Abstract

Background

While classic ALS is characterized by varying degrees of upper motor neuron (UMN) and/or lower motor neuron (LMN) dysfunction, primary lateral sclerosis (PLS) and progressive muscular atrophy (PMA) represent extreme phenotypes characterized respectively by pure upper motor neuron (UMN) and lower motor neuron (LMN) pathology. Theorizing that splicing defects may contribute to phenotypic heterogeneity in motor neuron diseases, we have used long-read RNA sequencing technologies to accurately capture the full-length transcriptome.

Objectives

To obtain RNA signatures of PLS, classic ALS, and PMA, which may aid in finding drivers of UMN versus LMN pathology.

Methods

Blood samples from age- and sex-matched individuals with clinical diagnoses of PLS (N∼20), classic ALS (N∼20) or PMA (N∼20) were obtained from the PGB (Phenotype-Genotype-Biomarker, NCT02327845) study of the CReATe (Clinical Research in ALS and Related Disorders for Therapeutic Development) Consortium. RNA was extracted using the PAXgene Blood RNA extraction kit (Qiagen). Only samples with an RNA integrity number (RIN) above 7 were included. Long-read RNA sequencing was performed at Mayo Clinic’s Genome Analysis Core using the Iso-Seq method on the Sequel II platform (Pacific Biosciences). Transcriptomic data was analyzed using a custom workflow that incorporates various tools (Minimap2, SQANTI3, tappAS, Integrative Genomics Viewer, etc.).

Results

In total, we obtained roughly one million reads per individual. The average read length was more than 2 kb, which demonstrates our ability to produce long reads. We detected over 20,000 unique genes. Importantly, multiple transcripts were observed for most genes. Comparisons between groups (i.e. PLS, classic ALS, and PMA) are currently on-going, aiming to identify determinants of UMN and/or LMN dysfunction.

Conclusions

Our long-read sequencing approach in a well-characterized clinical cohort of patients with PLS, classic ALS, and PMA should assist in identifying some of the biological underpinnings of the phenotypic heterogeneity of these diseases, which vary not only in the relative degree of UMN versus LMN pathology, but also with respect to rates of progression and survival. Notably, extensive studies will be performed in clinical and pathological specimens to replicate and validate our preliminary observations from blood.

Funding

CReATe PGB study participants, research staff, and funding for CReATe Consortium and CReATe Biorepository (NIH NCATS/NINDS U54NS092091 and ALS Association 16-TACL-242).

GEN-09 Functional verification and clinical phenotypes of amyotrophic lateral sclerosis with novel variants of VAPB [email protected]

Abstract

Objective

Vesicle-associated membrane protein-Associated Protein B(VAPB) has been identified as a causative gene of Amyotrophic Lateral Sclerosis (ALS). However, VAPB mutations have been rarely reported. Here, mutations in the VAPB gene was sequenced in a cohort of Chinese ALS patients.

Method

We sequenced for VAPB mutations in a cohort of 15 familial ALS (fALS) indexes and 275 sporadic ALS (sALS) patients of Chinese origin by targeted next-generation sequencing. Pathogenicity analysis of the variants was performed by software such as SIFT, Polyphen2 and VarCard. In order to investigate the pathogenic effects of these three mutations, we performed insoluble assays and immunofluorescence experiment to determine protein expression and localization.

Result

Three novel heterozygous missense mutations in the VAPB gene, c.38A>C (p.Gln13Pro), c.511G>A (p.Glu171Lys) and c.659T>G (p.Leu220Arg) were each detected in one sporadic ALS patient, respectively. The three patients carrying these VAPB variants all had an aggressive progression and a short survival. All these three mutations are predicted probably damaging by several prediction softwares. Further functional verification indicated by immunofluorescence that p. Q13P and p.E171K is prone to form small cytoplasmic aggregates and lead to changes in the localization of proteins in HEK293 cells.

Conclusion

The three patients carrying these VAPB variants all had an aggressive progression and a short survival. p.Q13P, p.E171K mutations of the VAPB gene may diminish the activity of VAPB to mediate unfolded protein response (UPR). Thus we speculate that the malfunction of VAPB to mediate UPR, caused by the two new mutations, may contribute to the development of motoneuronal degeneration linked to VAPB/ALS8.

GEN-10 Study into a possible common founder effect for the SOD1 p.V48A mutation amongst Malaysian Chinese and Mainland Chinese ALS [email protected]

Abstract

Background

Over 200 mutations have been identified in SOD1 in ALS populations across the world and recently, our group reported two unrelated Malaysian Chinese patients with a p.V48A mutation. This mutation was previously reported in six patients of Chinese ancestry and we postulated whether there could be a common founder effect in our Malaysian Chinese population with those in mainland China, similar to what has been reported for the p.A4V mutation in a North American ALS cohort. The shared ancestry could have arisen during the period of mass migration of Chinese citizens, particularly from Southern China, to Peninsular Malaysia, attracted by economic incentives offered by the ruling party in the 1800–1900s. Determining the presence of a founder effect would help focus efforts in screening for SOD1 mutations in Malaysian Chinese patients, as well as highlight the importance of this mutation for therapeutic development or recruitment into relevant clinical trials.

Objectives

To determine the possible presence of a common founder effect for the SOD1 p.V48A mutation in Malaysian Chinese patients.

Methods

Thirteen SNPs flanking the SOD1 p.V48A mutation over an 84Mb genomic region, were selected from other SOD1 studies and genotyped by Sanger sequencing in the two patients. These SNPs include rs4817415, rs2070422, rs1008270, rs9974610, rs2173962, rs4816405, rs2070424, rs1041740, rs2833475, rs16988427, rs2833481, rs2070423 and rs2833483. The p.V48A mutation was also screened in 50 neurologically healthy Malaysian Chinese controls.

Results

The p.V48A mutation was not present in any of the 50 controls screened, excluding it as a common polymorphism. In addition, there are currently no reported heterozygote in any genome browser, suggesting that it is a rare allele. The two Malaysian Chinese patients in our study shared a similar haplotype across the SOD1 locus, however as this haplotype pattern was also seen in a subset of controls, it cannot be claimed to be a disease-specific haplotype.

Discussion and conclusion

Our findings indicate that the SOD1 p.V48A mutation is likely to be pathogenic. However, it does not lie on a commonly shared, disease-specific haplotype between the Malaysian Chinese ALS patients and those in Mainland China.

Funding

This study is funded by the Malaysian Ministry of Education Fundamental Research Grant Scheme [FRGS/1/2018/ SKK08/UM/01/1], ALS Association [IF008-2019] and Sydney Southeast Asia Centre [IF012-2019].

Acknowledgments

We would like to extend our deep gratitude to all patients who willingly participated in this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

GEN-11 Comprehensive genetic study of genes that alter NAD + levels in Australian sporadic amyotrophic lateral [email protected]

Abstract

Background: Nicotinamide adenine dinucleotide (NAD+) is a central component for cell metabolism, and a vital factor for function of various groups of enzymes involved in pathways such as DNA repair, inflammation, genomic stability and stress resistance. NAD + is synthesised through three pathways; the Kynurenine, Preiss-Handler and salvage pathways. Conversely, NAD + is consumed to its reduced form nicotinamide adenine dinucleotide (NADH) via poly (ADP-ribose) polymerases (PARPs), sirtuins (SIRTs), ADP-ribosyl cyclases (CD38/CD157), and Sterile alpha and TIR motif containing 1 (SARM1).

Reduced NAD + levels caused by impairment of enzymes involved in NAD + biosynthesis or up-regulation of enzymes that consume NAD + have been associated with aging and several neurodegenerative diseases such as amyotrophic lateral sclerosis (ALS). While gene mutations are known to cause ALS, the genetics of ALS is highly complex, whereby only approximately 60–70% of familial and 10% of sporadic ALS cases have a known genetic cause. In additional to causal gene mutations, multiple gene variants may also act synergistically, with or without environmental influence, to cause or confer risk to the development of ALS.

Objectives: We aimed to determine the genetic contribution of 36 genes involved in NAD + biosynthesis (focusing on Preiss handler and Salvage pathway, Kynurenine pathways), and NAD + consumption (PARPs, SIRTs, CD38/CD157, SARM1) by identification of novel gene variants and gene burden in a large cohort of Australian sporadic ALS cases.

Methods: Whole-genome sequencing data from 609 Australian sporadic ALS cases was screened through the 36 genes. Novel variants absent in public control databases, and rare protein-altering variants (for gene burden analysis) were identified. For burden analysis, variants in genes involved in NAD + synthesis and each consumption pathway were grouped, and compared to two independent control databases.

Results: Forty-two novel variants identified in 26/36 genes were identified. Burden analysis of rare variants performed on genes in combined Preiss-handler and Salvage pathway (NAD + synthesis), and genes belonging to the SIRT family (NAD + consumption) was statistically significant (p = 0.0014 and p = 0.006 respectively) when compared to Project MinE controls.

Discussion: Some of these variants may represent ALS risk factors that disrupt NAD + biosynthesis or consumption, hence altering NAD + levels in ALS patients. Overall, our genetic findings suggest that rare genetic mutations in genes that alter NAD + levels may be associated with sporadic ALS in Australia. There is potential for therapeutic drugs that boost NAD + levels in ALS patients, and future studies can assess the specific pathogenic role of novel variants identified in patients to improve our understanding underlying the mechanisms of disease.

Funding

This work was funded by the Australian NHMRC and Macquarie University.

Acknowledgements

The authors thank the patients involved in our research, and The Neurodegenerative Disease Biobank and S Furlong, E Cachia, and L Adams for their assistance in sample collection.

GEN-12 Prediction of pathogenicity of ALS-causing gene missense variants using structural information predicted by [email protected]

Abstract

Background

Rare variants in the causative gene of familial amyotrophic lateral sclerosis (FALS) are found in 10–30% of sporadic ALS (SALS). However, the pathological significance of rare variants found only in SALS is often unknown. To predict the pathogenicity of the variants, in silico analysis methods are commonly used. In some ALS causal genes, the mutations are concentrated in specific regions, and the accuracy of pathogenicity prediction can be improved by taking into account the positional information of the variants. However, existing methods have not considered information on the position of variants in the protein’s structure.

Objectives

We developed “a method of missense variant pathogenicity using AlphaFold2” (MOVA), which applies the positional information of variants on the 3D structure predicted by AlphaFold2 as a feature, and machine-learns the pathogenicity of variants for each gene.

Methods

We compared predictive accuracy of rare variants of 6 ALS causative genes (TARDBP, FUS, SETX, OPTN, TBK1, SOD1) for pathogenicity between MOVA and existing methods (PolyPhen-2, EVE, AlphScore, CADD, REVEL) (1–5).

Results

MOVA showed AUC ≥0.70 for 3 (TARDBP, FUS, SOD1) of 6 genes. In PolyPhen-2, only one (TBK1) of six genes showed AUC ≥0.70. In EVE, 2 (OPTN, SOD1) of 4 genes (TARDBP, SETX, OPTN, SOD1) listed in the database (https://evemodel.org/) showed AUC ≥0.70. In AlphScore and CADD, 2 genes (TBK1, SOD1) of 6 genes, and in REVEL, 5 genes (TARDBP, FUS, TBK1, OPTN, SOD1) of 6 genes showed AUC ≥0.70. MOVA was particularly useful for genes where pathogenic gene mutations were concentrated at specific sites (TARDBP, FUS).

Discussion

Of MOVA and five existing methods, REVEL performed best for all genes. MOVA performed almost as well as REVEL, except for TBK1 and OPTN. Guidelines for genetic diagnosis from the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP) recommend that in silico analysis methods are useful only when multiple methods predict that the variant is deleterious (6). The same results from MOVA and existing methods such as REVEL may provide a more accurate assessment of the pathogenicity of ALS-causing gene mutations.

GEN-13 Sex-stratified analysis of ∼133k samples identifies novel associations with amyotrophic lateral [email protected]

Abstract

Background

ALS shows both higher incidence and prevalence in males (1), suggesting sex is an important factor in disease risk. Sex hormones play a role in this, with evidence that male sex hormones increase risk [2) and female sex hormones decrease risk (3). However, recently observed differences in ALS heritability between males and females (4) suggest genetics may also play a significant role in the sex differences observed in ALS.

Objective

Our study aims to identify sex-specific ALS risk variants through sex-stratified analysis of the European samples from the latest ALS GWAS dataset (N∼133k) (5).

Methods

We ran SAIGE on male-only (N = 65,692) and female-only (N = 67,957) strata to generate sex-stratified GWAS summary statistics. To test for global differences in genetic architecture, genetic correlation between sexes was estimated with LD score regression. We performed genome-wide scans for sex-specific variants, defined as variants passing a 1% FDR threshold for association in one sex with no evidence of association in the other (p>0.05). Finally we integrated expression data for 13 brain tissues from GTEx with our sex-specific GWAS statistics in a transcription wide association study (TWAS). This tested for association between gene expression and ALS in males and females.

Results

Despite near perfect genetic correlation between sexes (LD score regression: rg =0.99; se =0.16), sex-specific genome-wide scans identified significantly associated variants in several novel male-specific (CHD2 and RESP18) and female-specific (FGFRL1 and MEF2C) loci. TWAS analysis revealed significant sex-specific associations between expression of two of these genes (MEF2C and RESP18) in brain tissues and ALS.

Discussion

Among our novel sex-specific associations both MEF2C and CHD2 show promising links to ALS in the existing literature. In brief, CHD2 has been shown to interact with a hallmark ALS pathogenic protein TDP-43, leading to reduced expression of stress response genes (6) and the transcription factor MEF2C has been shown to be upregulated in sporadic and SOD1+ ALS patients (7). Thus our analysis reveals several novel sex-specific associations with ALS supported by tissue specific expression and existing literature.

Funding

This work was supported by Motor Neurone Disease Association 879-971.

GEN-14 Whole-genome sequencing reveals that variants in the IL18RAP 3’UTR protect against [email protected]

# These authors have contributed equally to this work.

Abstract

Background

The noncoding genome is substantially larger than the protein-coding genome but has been largely unexplored by genetic association studies.

Methods

Here, we performed region-based rare variant association analysis of >25,000 variants in untranslated regions of 6,139 amyotrophic lateral sclerosis (ALS) whole genomes and the whole genomes of 70,403 non-ALS controls.

Results

We identified interleukin-18 receptor accessory protein (IL18RAP) 3’ untranslated region (3’UTR) variants as significantly enriched in non-ALS genomes and associated with a fivefold reduced risk of developing ALS, and this was replicated in an independent cohort. These variants in the IL18RAP 3’UTR reduce mRNA stability and the binding of double-stranded RNA (dsRNA)-binding proteins. Finally, the variants of the IL18RAP 3’UTR confer a survival advantage for motor neurons because they dampen neurotoxicity of human induced pluripotent stem cell (iPSC)-derived microglia bearing an ALS-associated expansion in C9orf72, and this depends on NF-κB signaling.

Discussion

This study reveals genetic variants that protect against ALS by reducing neuroinflammation and emphasizes the importance of noncoding genetic association studies.

GEN-15 Genetic analysis of ALS in [email protected]

Abstract

Background

Genetic background of ALS varies greatly across populations. Variants in known ALS-associated genes have different frequencies in samples of patients recruited in different parts of the world, and carrier frequencies also vary, sometimes with striking disproportion, e.g. for SOD1, C9orf71. Genetic studies in ALS may benefit patients willing to be enrolled in clinical trials of new therapies, including gene therapies. Moreover, understanding the genetic causes of ALS is crucial for elucidation of molecular pathogenesis of the disease, developing new therapeutic approaches and adjusting healthcare strategies to the needs of patients. Russian populations are underrepresented in public genetic databases and ALS genetics studies and may have features of European and Asian populations as well as unique peculiarities.

Objectives

To explore genetic heterogeneity of ALS on whole genome level in a sample of patients recruited in Russia.

Methods

Whole genome sequencing was performed for 98 ALS patients (61 female and 37 male) and 50 healthy donors on Illumina NovaSeq6000 platform. 92 of 98 patients had the diagnosis of definitive ALS. Bioinformatic pipeline included quality control, small variant calling, copy number variants (CNV) calling and repeat expansion analysis. Clinical interpretation of genetic variants was carried out according to ACMG guidelines (1).

Results

Variants that were considered causative of monogenic ALS were detected in 6 patients (6%). Among them, repeat expansions in C9orf72 were detected in 2 patients, pathogenic or likely pathogenic variants in SOD1 – in 4 patients (including 2 homozygotes). 59 patients (60%) had variants of unclear relevance to their clinical phenotype. Results of CNV analysis are preliminary and will be reported later.

Discussion

This study is the largest investigation into genetic etiology of ALS using WGS in patients recruited in Russia to date. Nonetheless, the sample size is limited, and further research is needed to establish the genetic spectrum of ALS in this population.

Funding

This study was funded by a grant of the Moscow Government for the realization of scientific and practical projects in medicine № 2412-35 and "Live Now" Charity Foundation.

GEN-16 Genome-wide association study of rare variants in ERV-elements of ALS patients: age at onset of ALS may be regulated by MLT1B [email protected]

Abstract

Background

Transposable elements (TE) comprise about half of the human genome and demonstrated as playing a role in regulation of gene expression by a number of different mechanisms, including as enhancers, promotors, alternative splicing or by modifying chromatin structure. Therefore, TE could modulate genes expression which could correlate with new phenotypes or diseases. The role of human endogenous retroviruses (HERVs) in pathogenesis of neurological and psychiatric diseases has been demonstrated in multiple sclerosis and schizophrenia (1,2). Analysis of muscle biopsies and post-mortem brain tissue from patients with amyotrophic lateral sclerosis (ALS) revealed enhanced level of reverse transcriptase activities (a protein encoded by HERVs) in these tissues, suggesting involvement of HERV elements in pathogenesis (3). HERV family members HERV-K and HERV-W were proposed as one cause of ALS. It was subsequently demonstrated that HML-2 and 3 subfamilies of HERV-K may contribute to development of ALS, by analysis of HERV gene expression post-mortem brains of ALS patients (4,5). To further investigate role of ERVs in development of ALS we conducted genome-wide analysis of rare variants of ERV-elements on large ALS-dataset.

Methods

Multi-marker variable-threshold analysis was performed on 6493 ALS patients from Project Mine dataset. We used age at onset (AAO) as a phenotypic trait in our study to investigate association with rare variants located in ERVs elements. In total 639,487 ERV elements, mapped to hg19 reference genome dataset, were investigated and only rare single nucleotide and small insertions/deletions genetic variants with frequencies less than 0.01 were selected for analysis. As a bioinformatic tool to perform association analysis, we used rvtests program (6) with famAnalytic and famCMC options. Information on sex, population structure and cryptic relatedness was included in the linear mixed model as additional covariates to correct statistics for type I errors.

Results

We found that a MLT1B ERV element, located in the first intron of the UBL3 gene on chromosome 13q12.3, is associated with AAO trait (p-value =3.92E-08). The UBL3 gene, demonstrated high expression in brain tissues and spinal cord. We speculate that MLT1B potentially can affect age at onset of ALS phenotype by altering UBL3 expression which would be predicted to modulate ubiquitination pathways. However, this ERV could also co-ordinately modulate a number of genes via chromatin remodelling or epigenetic modulation.

GEN-17 Fine-tuning a molecular diagnosis algorithm for patients with ALS and [email protected]

Abstract

Background

Amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) are considered to be the extremes of the same neurodegenerative syndrome with three nexuses: neuropathological, genetic and clinical.

Objective

The main aim of this work is the genetic characterisation of the Spaniard population with ALS, ALS-DFT and FTD.

Methods

The cohort used in this study is 1554 index cases. We studied 1198 patients with ALS, 160 with FTD, and, in addition, 196 patients were in the ALS-FTD intermediate. Conventional PCR Triple PCR and panelised whole exome sequencing were used, according to the clinical-familial group to which each of the patients belonged.

Results

Common genes related to ALS and FTD contribute only small percentages in sporadic diseases, emphasising C9orf72 in both sALS and sFTD, SOD1 in sALS and GRN in FTD. These same genes are highly relevant in the familial groups and in the ALS-FTD group, with C9orf72 (19.0%) and SOD1 (20.3%) making a large contribution in fALS, TARDBP in fALS and fFTD and MAPT in fFTD.

Conclusions

Sanger sequencing should start to be considered exceptional in familial cases. The implementation of NGS as a routine method is necessary at least for the study of familial cases. NGS studies are suggested to all familial cases of ALS and FTD and include the comorbidity group of both pathologies (ALS-FTD) regardless of the existence of family history.

Funding

This work was supported by grants PI17/00491 from the Instituto de Salud Carlos III (ISCIII), the support of the Comunidad de Madrid (B2017/BMD-3813 "ELA-Madrid"), the agreement of the CM-Fundación H12O (BOCM 142 - 17/09/2019 pg. 105), and the Plataforma de afectados de ELA.

GEN-18 Exploring the role of HERV-K integrations in amyotrophic lateral [email protected]

Abstract

Background

HERV-K insertions are a class of transposable elements. They originated in the human genome as a result of retroviral infection of a distant ancestor. It is estimated that 8% of the human genome is retroviral in origin (1). Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease, characterised by the death of motor neurons leading to paralysis and death. There is a high degree of missing heritability in ALS and growing evidence that HERV-K insertions may have a role in the disease and explain some heritability.

Objectives

Our first aim is to identify the best bioinformatics tool to detect HERV-K insertions in short read, whole genome sequence (SR-WGS) data. We then aim to use this tool to identify novel HERV-K integrations that are associated with ALS.

Methods

We first benchmarked six bioinformatics tools for the detection of HERV-K integrations in SR-WGS data. This included an in house tool we developed called Retrosnake. We have validated a small number of HERV-K insertions predicted by Retrosnake in participant DNA using a nested PCR protocol. We also applied Retrosnake to a set of 405 genomes from people with ALS and 346 controls. We used correlation, the gradient boosting model and generalised linear models (logistic regression and Cox proportional hazards) to find HERV-Ks associated with ALS incidence or survival duration.

Results

Retrosnake is an accurate, though conservative, detector of HERV-K insertions. Computational and experimental results have confirmed that the HERV-Ks predicted in SR-WGS are indeed present in DNA samples. Logistic regression tests failed to find HERV-Ks associated with ALS incidence, however Cox proportional hazards identified a single HERV-K integration site which significantly predicted survival. We also found that HERV-K burden correlated with survival duration. The gradient boosting model was not able to accurately classify ALS vs control patients using HERV-K genotypes.

Discussion

Retrosnake is a useful tool in detecting HERV-K insertions. The HERV-Ks identified here do not contribute to ALS risk, though a larger sample size may be needed to fully rule them out. HERV-Ks may predict survival duration in ALS.

Acknowledgements

We would like to acknowledge the GSST BRC, Maudsley BRC and NIHR for funding this research. Also, the MNDA, MND Scotland and the ProjectMine consortium. We would finally like to acknowledge all participants who contributed samples and data, making this project possible.

GEN-19 SOD1 D91A variant in the southernmost tip of Europe: a heterozygous ALS patient resident on the island of [email protected]

Abstract

Background

Amyotrophic lateral sclerosis (ALS) is frequently caused by mutations in the SOD1 gene. All SOD1 deleterious variants show autosomal dominant inheritance except the p.Asp91Ala or D91A variant (also known as D90A; dbSNP155 ID rs80265967), which shows recessive inheritance, initially described in ALS cases from Sweden, Norway and Finland. Nonetheless, heterozygous SOD1 D91A ALS patients have now been described in various European populations including those in southern Italy. The archipelago of Malta, a sovereign microstate in the south of Europe, is home to a geographically and culturally isolated population. We have recently showed that ALS patients in Malta do not have deleterious variants in C9orf72, SOD1, TARDBP or FUS genes indicating that the most commonly mutated ALS genes globally do not have a major impact on the Maltese ALS population.

Objective

Our study aimed at investigating the genetic profile of Maltese patients with ALS, identified throughout a five-year window (2017–2022) to confirm whether Maltese ALS patients are absent for deleterious variants in the four major ALS genes.

Methods

Whole-genome sequencing was performed on patients recruited through Malta’s National ALS Registry and Biobank to determine rare DNA variants (European minor allele frequency ≤0.01) that change the protein-coding sequence of ALS-associated genes.

Results

We identify a patient with a sporadic form of ALS living on the island of Gozo in which the heterozygous SOD1 c.272A>C; p.(Asp91Ala) variant was detected. Gozo is the second-largest island of the Maltese archipelago (total area 67 km²) and has a population that presently numbers around 34,430, which is 6.7% of the total population of Malta (514,564). The patient had a late onset (79 years), sensory impairments and rapid disease progression culminating in respiratory failure. ALS has not yet developed in any of the three additional family members in which the D91A variant was identified. None of the healthy controls from the Maltese population were found to carry this variant.

Discussion and conclusions

Our findings underscore the high prevalence of the D91A variant in Europe, despite the presence of a North-South gradient in its frequency, and confirms that this variant can be associated with dominant cases in Mediterranean countries.

GEN-20 Allele genotype of a SINE-VNTR-Alu (SVA) retrotransposon correlates with differential gene expression at the MAPT [email protected]

Abstract

SINE-VNTR-Alu (SVA) retrotransposons are transposable elements which are capable of propagating throughout the genome and hence represent a source of genetic variation. We previously demonstrated that presence or absence of a human-specific SVA, termed SVA_67, correlated with both isoform- and tissue-specific expression of multiple genes at the MAPT locus, a locus associated with several neurodegenerative diseases including frontotemporal dementia and other tauopathies. A problem in defining regulatory associations is that this region is contained within a large genomic inversion termed H1/H2. We have identified that SVA _67 tags the H1 inversion and we are extending our analysis to distinguish the SVA specific regulation from that associated with the inversion. In the current study we show that SVA_67 is not only a retrotransposon insertion polymorphism (RIP), but also variable in the size of its variable number tandem repeat (VNTR) domain. Four different alleles of the SVA_67 VNTR were identified which demonstrated differential properties of a regulatory domain in a reporter gene assay. We expanded these findings by correlating SVA_67 allele genotypes with expression of genes located around this SVA in the MAPT locus using transcriptomic datasets from the Parkinson’s Progression Markers Initiative (PPMI) cohort. Variation of the SVA_67 VNTR domain was associated with differential expression of multiple genes including KANSL1. As part of the nonspecific lethal complex, KANSL1 was previously identified as an essential gene for autophagy, a key machinery to maintain proteostasis, and dysregulation has been considered to play major roles in neurodegenerative diseases including amyotrophic lateral sclerosis and Parkinson’s disease. This study highlights the potential of SVA_67 to act in concert with other regulatory domains to modulate gene expression leading to phenotypic variation in a population and an additional type of variation to be taken into account when investigating the missing heritability of neurodegenerative diseases.

GEN-21 Utilising genetic data and quality of life metrics within the TONiC initiative to better predict patient outcome, progression and severity in [email protected]

Abstract

The Trajectories of Outcome in Neurological Conditions (TONiC) study is a UK initiative to better understand the external factors that influence the quality of life of people suffering with neurological conditions including Amyotrophic Lateral Sclerosis (ALS). The primary goal of this initiative was to explore the longitudinal associations between onset type, physical and psychological factors, perceived health and quality of life in a large cohort of people with ALS. This provides a well characterised database of patient disease progression which may be utilised for both clinical uses, to help better treat patients on an individual level, but also for genetic research to generate potential biomarkers to better predict patient outcome and severity. We are undertaking a small pilot study of 26 patient derived blood samples from which DNA and RNA have been isolated and these will be analysed using Genome-Wide Association Study (GWAS), Epigenome-Wide Association Study (EWAS) and Oxford Nanopore RNA sequencing. These combined technologies will allow us to identify the genetic differences between patients, which when correlated with data collected in TONiC, will potentially allow us to stratify severity of progression (fast vs slow progressors) as well as changes in the trajectories of outcomes and quality of life. The aim of these proposed analyses is to establish pipelines that will be utilised for the future analysis of the extended TONiC project which aims to collect in excess of 500 patient samples. These studies will provide the foundation for more extensive analysis with the proposal to expand into the national TONiC initiative in the future to include greater sample sizes.

Acknowledgements

We would like to thank all the participants in the TONiC study and the staff who make this work possible for their continued support for this project.

GEN-22 Targeted long-read sequencing of C9orf72 in blood of patients with [email protected]

Abstract

Background

The most prominent genetic cause of ALS is a non-coding hexanucleotide repeat expansion in the gene C9orf72. Little is known about how variation in the C9orf72 expansion contributes to clinico-pathological heterogeneity. Long-read sequencing allows us to study the repeat expansion at a greater resolution than traditional methods, like Southern blotting.

Objective

To uncover variation in the length, sequence content, and methylation levels of the C9orf72 repeat expansion in order to determine whether they may serve as disease modifiers.

Methods

A targeted long-read DNA sequencing technique, called No-Amp sequencing, was used to examine the expanded C9orf72 repeat. High-quality genomic DNA was extracted from blood (Circulomics) of approximately 20 C9orf72 expansion carriers. At Mayo Clinic’s Genome Analysis Core, No-Amp sequencing was performed on the Sequel II platform (Pacific Biosciences), using CRISPR-Cas9 to capture our region of interest. Data was analyzed using a customized workflow that generates circular consensus reads (pbccs) and aligns reads to the genome (pbmm2).

Results

We completed No-Amp sequencing on a well-characterized cohort of ALS patients. We prioritized subjects with detailed clinical and pathological information available. This includes age at onset, age at death (if available), C9orf72 repeat length (based on Southern blotting), C9orf72 promoter methylation, and C9orf72 transcript levels. Of note, we included longitudinal specimens to determine how the expansion changes over time and familial cases to see how the expansion changes in successive generations. Analyses are currently being performed and have already demonstrated that we can successfully sequence the entire repeat expansion in single, long reads. Additionally, we are able to span a wide range of sizes, from several hundreds to thousands of repeats (e.g. over 20 kb). Moreover, we established that the expanded repeat contains sequencing impurities and is methylated.

Discussion

Overall, we have shown that No-Amp sequencing can be used to capture features of the C9orf72 repeat expansion in blood. Ongoing efforts are focused on association analyses and obtaining No-Amp sequencing data in additional regions, such as the frontal and motor cortices.

GEN-23 Genetic and clinical characteristics of ALS patients with NEK1 gene [email protected]

Abstract

Background

NIMA-related kinase 1(NEK1) gene was found to be related to amyotrophic lateral sclerosis (ALS). However, genetic spectrum and clinical characteristics of ALS patients with NEK1 variants was largely unknown.

Methods

A total of 1587 Chinese patients with ALS was performed whole-exome sequencing (WES) and repeat primed-polymerase chain reaction (RP-PCR). We analyzed NEK1 gene variants and used 5 software to predict the protein function of missense variant in the NEK1. Additionally, we comprehensively searched the literatures including PubMed, Embase, and Web of Science.

Results

In our ALS cohort, 42 sporadic ALS patients (2.6%) were detected to carry NEK1 variant. Among the NEK1 variant carriers, 10 patients carried novel heterozygous loss-of-function (LoF) variants, 32 patients carried rare missense variants with 19 predicted pathogenic missense variants. 90% of the 10 patients carrying NEK1 LoF variants had upper limbs onset, while only 63.2% of the pathogenic missense variants carriers had upper limbs onset. The median survival time of LoF variant carriers tend to be shorter than that of pathogenic missense variant carriers (23.80 months vs 42.77 months). Sixteen related case reports and studies on NEK1 gene in ALS were published. There were 167 different NEK1 variants in 237 reported ALS patients, including 62 LoF variants and 105 missense variants, of which about 24% carried additional gene variants. At the same time, it was found that the survival time of LoF variant carriers was significantly shorter than that of missense variant carriers.

Conclusion

Our study further expanded the genotype and phenotype spectrum of ALS patients with NEK1 variants. Our ALS cohort along with previously published cases showed that the majority of the patients who carried NEK1 LoF variants have upper limb onset and shorter survival time. More large studies are needed to further deepen the understanding of the pathogenesis of NEK1 gene.

GEN-24 Reduced-penetrance Huntington’s disease-causing alleles with 39 CAG trinucleotide repeats could be a genetic factor of amyotrophic lateral [email protected]

Abstract

Background

Expanded HTT alleles with 40 or more CAG repeats were recently found to be a rare cause of frontotemporal dementia (FTD) and amyotrophic lateral sclerosis (ALS) spectrum diseases. The aim of this study was to investigate the role of HTT repeat expansions in a Taiwanese cohort with ALS.

Methods

We analyzed the numbers of the CAG repeats in exon 1 of HTT in a cohort of 410 Taiwanese patients with ALS and 1514 control individuals by utilizing polymerase chain reaction and amplicon fragment length analysis.

Results

Only one of the 410 ALS patients carried a reduced-penetrance HD-causing allele with 39 CAG repeats and none had an expanded HTT CAG repeats ≥40. The patient presented with a rapidly progressive bulbar onset ALS with disease onset at age 64 years. He had neither chorea nor cognitive impairment. He had a family history of chorea but no other family member manifested with ALS. None of the 1514 control individuals carried a HTT expanded allele with CAG repeats larger than 37 repeats.

Conclusion

The HTT allele with 39 CAG repeats could be a genetic factor linked to ALS susceptibility.

GEN-25 Discovering novel MND causal mutations through comprehensive assessment of complex genomic variant [email protected]

Abstract

Aim: Gene mutations remain the only known cause of MND, and although heritability studies suggest 40–60% of MND risk is heritable almost 90% of patients remain with an unidentified cause. These unexplained cases include one third of familial patients who carry inherited genetic factors conferring a sizable predisposition to MND.

The majority of MND genetics research has been limited to studying small genetic changes affecting 1–50 nucleotides. As such, most known MND mutations are small protein-coding changes, however, the most common cause of MND, the pathogenic C9orf72 expansion, is more complex spanning thousands of nucleotides and residing in a non-protein coding intronic region. This suggests that the unsolved cases of MND may also be explained by more complex genomic variant types displaying similarities to the C9orf72 expansion, namely large structural variants altering large stretches of DNA, and/or non-coding variants falling outside of protein coding genes. Historically, SVs and non-coding variants have been technically challenging to study, however recent DNA sequencing technology and bioinformatics advances now make this prospect feasible.

We have identified a small family consisting of two distantly related individuals (third cousins) both affected by MND, where small protein-coding genetic variants have been excluded as a cause of MND.

Objectives: We aim to leverage the genetic power of this small family to investigate structural and non-coding variants as the cause of MND in this family, and thereby implicate a novel form of complex genomic variation in the aetiology of MND.

Methods: Whole-genome sequencing data from both individuals was processed to identify (1) small nucleotide level variation and (2) large structural variation using a combination of three bioinformatics tools (Lumpy, Manta and MetaSV). This data was then filtered to variants shared by both affected family members within chromosomal regions inherited from a common ancestor, and absent from control databases (e.g. the genome aggregation database). Remaining candidates were assessed for potential MND pathogenicity by in silico characterisation of features including functional consequence, sequence context and genic association, using tools including ANNOVAR, AnnotSV, CADD-genome, CADD-SV, GeneHancer, RegulomeDB and DeepSEA. Variant validation of top candidates was performed using PCR and Sanger sequencing.

Results: From an initial >7,000 structural and >5,000,000 nucleotide variants, just 5 and 153 candidate MND mutations remain. Of these, two structural, and six non-coding variants have evidence supporting a potential role in MND.

Discussion: After excluding known MND mutations and small protein-coding variation, we have narrowed the search for the causal mutation in this family to <0.00000053% of the genome and complex genomic variant types, tructural and non-coding variants. The implication of such complex variants in MND presents exciting therapeutic targets, especially given their known effects on gene expression, coupled with the recent promise shown by gene silencing therapies in MND.

GEN-26 Association of variants in the KIF1A gene with amyotrophic lateral [email protected]

Abstract

Background

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease. The genetic risk factors and pathogenesis of ALS involve dysregulation of the cytoskeleton and neuronal transport. Notably, sensory and motor neuron diseases such as hereditary sensory and autonomic neuropathy type 2 (HSAN2) and spastic paraplegia 30 (SPG30) share several causative genes with ALS, as well as common clinical phenotypes. KIF1A encodes a kinesin 3 motor that transports presynaptic vesicle precursors (SVPs) and dense core vesicles and has been reported as a causative gene for HSAN2 and SPG30.

Methods

Here, we analyzed whole-exome sequencing data from 941 patients with ALS to investigate the genetic association of KIF1A with ALS.

Results

We identified rare damage variants (RDVs) in the KIF1A gene associated with ALS and delineated the clinical characteristics of ALS patients with KIF1A RDVs. Clinically, these patients tended to exhibit sensory disturbance. Interestingly, the majority of these variants are located at the C-terminal cargo-binding region of the KIF1A protein. Functional examination revealed that ALS-associated KIF1A variants located in the C-terminal region preferentially enhanced the binding of SVPs containing RAB3A, VAMP2, and synaptophysin. The expression of several disease-related KIF1A mutants in cultured mouse cortical neurons led to enhanced colocalization of RAB3A or VAMP2 with the KIF1A motor.

Conclusions

Together, our findings identified KIF1A as a novel causative gene for ALS and characterized the functional roles of the identified KIF1A variants in interactions with cargos, highlighting the importance of KIF1A motor-mediated transport in the pathogenesis of ALS.

GEN-27 Does familiarity for neurodegenerative diseases influence amyotrophic lateral sclerosis presentation? An exploratory clinical [email protected]

Abstract

Objectives

It has been previously suggested that familial and sporadic amyotrophic lateral sclerosis (ALS) entities are clinically indistinguishable. However, there are currently different definitions of familial ALS (fALS), and the identification of other neurodegenerative disorders, in particular frontotemporal dementia (FTD), in patients’ pedigree has been lately proposed as an additional criterion for the classification of this entity. Despite these considerations, studies in fALS patients as recently defined are still rare, and little is known about the influence of familiarity for neurodegenerative conditions on the clinical presentation of ALS. The aim of our study was therefore to investigate whether ALS patients with a positive family history for either ALS or other neurodegenerative diseases may exhibit distinct clinical features compared to sporadic cases.

Materials

98 patients with a clinical diagnosis of ALS, thorough clinical examinations and complete genetic screening were included in the study.

Methods

By evaluating family history, patients were categorized in ALS with positive family history for either ALS, other neurodegenerative disorders, or both (fALS/ND) and sporadic ALS (sALS). Demographic and clinical features were compared between the two groups using chi-squared and Mann–Whitney U tests, as appropriate.

Results

17 patients (17.34%) were classified as fALS/ND, while the remaining 81 cases (82.66%) as sALS. No significant differences were observed between the two groups in demographic features and onset clinical presentations. Conversely, relative to sALS cases, fALS/ND exhibited significant greater neck flexor weakness (p = 0.05), higher ALSFRS-r progression rate (p = 0.03), more severe impairment in fluency measures (p = 0.03) and, as expected, greater frequency of genetic mutations (p = 0.01).

Discussion

Despite common onset presentations, fALS/ND cases exhibited more severe clinical impairment across multiple domains compared to sALS patients. Notably, all the clinical features found to differentiate fALS/ND from sALS cases (neck flexor weakness, high ALSFRS-r progression rate and fluency impairment) are well-known markers of more severe clinical decline, suggesting a greater vulnerability to more aggressive disease course in familial disease forms.

Conclusions

While further studies are warranted to confirm our preliminary observations, our data suggest that familiarity for neurodegenerative diseases may play a negative prognostic role in ALS.

GEN-28 Neuropeptide signalling processes perturbed in postmortem motor cortex of ALS [email protected]

Abstract

Background

Amyotrophic lateral sclerosis (ALS) is a devastating, heterogeneous neurodegenerative neuromuscular disease that leads to a fatal outcome within two to five years. Despite identification of a number of variants linked to the disease, the diversity of processes involved and the ambiguity of their relative importance in ALS pathogenesis is still not clear.

Methods

Using two separate RNAseq datasets from post mortem motor cortex samples of ALS patients and controls, BrainBank from King’s College London (171 subjects) and TargetALS motor cortex samples (234 subjects), we have performed a large differential expression study of a key brain region implicated in ALS pathology. We have used a wide range of biological databases to perform gene set enrichment analyses with the differentially expressed genes in the two datasets and compared the results. This analysis highlights key biological processes and pathways affected in the ALS postmortem brains.

Results

In both BrainBank and TargetALS datasets, neuropeptide signalling and neurogenesis processes are top perturbed processes in ALS patients versus controls. GO process enrichment reveals significant synapse related processes in BrainBank dataset, while TargetALS dataset carries an immune system related signature. We also reviewed the role of the most significant differentially expressed genes in both datasets in the context of the pathways and processes highlighted by the enrichment analyses and produced a set of candidate target genes. Our findings support already established cortical involvement in ALS and identify new potential drug targets.

GEN-29 Facilitating template structure selection for in silico anaylses of ALS proteins – a comprehensive dataset of all available experimentally solved wildtype and ALS mutant protein [email protected]

Abstract

Background

Our understanding of ALS has been greatly advanced by the continuous identification of disease-associated genes, yet, the downstream molecular mechanisms linking any genetic variant to ALS aetiology remain unknown (for all variants in all genes). A growing number of experimentally solved protein structures harbouring well-known disease-associated mutations are deposited in the Protein Data Bank (PDB). (1) The application of bioinformatics tools to systematically study differences in protein structure and dynamics in ALS mutant compared to wildtype proteins constitutes a promising approach to potentially elucidate the molecular causes linking ALS genetics and pathology. However, it remains challenging to identify suitable template structures for all mutant proteins associated with ALS.

Methods

We have developed MutaPipe, a python-based bioinformatics pipeline to screen the PDB for genes of interest and identify the highest quality protein structures for all available mutant and wildtype proteins associated with the input genes. Variants are annotated using information on variant pathogenicity from ClinVar (2) whenever available. Using MutaPipe, we screened the PDB for data on all genes listed in the ALS online database (ALSoD (3)) and additional genes identified through literature review (n = 167 ALS-associated genes). Based on the results of the run, we curated a comprehensive dataset of the highest quality protein structures for all ALS mutant and wildtype proteins.

Results

The final dataset is publicly available and comprises a wide range of information on the best protein structures for all ALS-associated (mutant and wildtype) proteins. PDB structures were available for 97 out of 167 genes, high-quality wildtype structures have been identified for 23 genes. A total of 214 unique protein mutants with a single amino acid change are available for 49 genes. Out of these, only 25 variants for 5 genes (ANG, PFN1, SOD1, VAPB, VCP) have available information on variant pathogenicity from ClinVar.

Conclusion

The ALS protein structure dataset provides a useful starting point for the study of changes in protein structure and dynamics which are caused by ALS-associated genetic variants. Moreover, MutaPipe is freely available and takes any gene of interest as input, allowing researchers to efficiently screen the PDB for the most suitable template structures for further in silico analyses (e.g. homology modelling, mutagenesis experiments, molecular dynamics simulations).

Funding

This research is part-funded by the Motor Neuron Disease Association and the NIHR Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London.

GEN-30 MIROCALS: Understanding low-dose IL-2 treatment in [email protected]

Abstract

Introduction

Neuroinflammation is a significant contributor to Amyotrophic lateral sclerosis (ALS), characterised by activation of glia, increased pro-inflammatory cytokines and T cell infiltration. Evidence has shown that regulatory T cells (Tregs), modulators of the immune system are negatively correlated with the ALSFRS-R and AALS scores, with a lower Treg count associated with a more aggressive disease course and poorer outcome. The Immune Modulation in ALS study (IMODALS) (https://clinicaltrials.gov/ct2/show/NCT02059759) showed how IL-2 mediates Treg expansion, with low dose (ld) IL-2 treatment improving Treg expansion albeit with a variable response across patients. In the larger longitudinal Modifying Immune Response and OutComes in ALS study (MIROCALS) (https://clinicaltrials.gov/ct2/show/NCT03039673) the main objective was to assess patient survival following ld IL-2 treatment. Firstly, we aimed to determine the transcriptomic signatures associated with ld IL-2 and riluzole treatment. Secondly, establish distinct expression profiles from responders and non-responders to ld IL-2. Thirdly, determine the transcriptomic signatures associated with specific biomarkers of ALS at diagnosis, such as high and low levels of neurofilament light (NFL).

Methods

A cohort of 211 Patients were recruited onto MIROCALS at diagnosis and randomised to either the treatment (2MIU IL-2) or the placebo arm of the trial. Blood was collected at diagnosis (at day -85 (D-85)) and patients received riluzole for 3 months. Blood was collected prior to the first ld IL-2 injection of cycle 1 (D1), 3 days after the 5th ld IL-2 injection of cycle 1 (D8), before the start of the final 5th cycle (D113) and 3 days following the last day of treatment (D120). Blood was sampled and white blood cell collected on LeukoLock filters stored in RNAlater, RNA was extracted and Affymetrix Clariom D human microarrays were used to assess the gene expression profile across the samples.

Results

Quality control of samples across all timepoints (1055 samples) is currently underway using Expression Console. Upon becoming unblinded to each sample demographics we aim to determine the response to ld-IL-2 in the patients. Additionally, we aim to validate the two-gene regression model from the IMODALS study that predicted a patient’s response to ld IL-2 at D120 using expression levels of CD27 and TLR9 at the time of randomisation (D1). We will also establish whether this model can be applied at (D-85). Other genes including FOXP3 and IL2RA which drive Treg expansion will be validated in IL-2 responders vs non-responders alongside other candidate genes that are identified in the study.

Discussion

A very limited number of studies have investigated the longitudinal effect of both IL-2 and riluzole from point of diagnosis. By correlating the data generated with Treg levels and ALS biomarkers, such as NFL levels will allow us to begin to examine the underlying basis for the response to ld IL-2.