Article Text

Download PDFPDF

Screening for TUBA4A mutations in a large Chinese cohort of patients with ALS: re-evaluating the pathogenesis of TUBA4A in ALS
  1. Jiao Li1,
  2. Ji He1,
  3. Lu Tang1,
  4. Lu Chen1,
  5. Yan Ma1,
  6. Dongsheng Fan1,2
  1. 1 Department of Neurology, Peking University Third Hospital, Beijing, People’s Republic of China
  2. 2 Key Laboratory for Neuroscience, Ministry of Education/National Health and Family Planning Commission, Peking University, Beijing, People’s Republic of China
  1. Correspondence to Professor Dongsheng Fan, Department of Neurology, Peking University Third Hospital, Beijing 100191, People’s Republic of China; dsfan2010{at}

Statistics from


Amyotrophic lateral sclerosis (ALS) is a devastating motor neuron disorder with an unclear aetiology that overlaps with frontotemporal dementia (FTD). ALS is classified as either familial or sporadic, and both of these have genetic components. Knowledge of the genetic architecture of ALS has been facilitated by evolving gene-sequencing technologies. A significant excess burden of rare damaging TUBA4A variants was recently demonstrated in ALS cases via an exome scan in a case−control burden analysis.1 Several follow-up TUBA4A gene screenings were subsequently performed in ALS and FTD cohorts of Italian, Belgian and Spanish origin.2–4 However, the pathogenicity of several TUBA4A non-synonymous mutations is difficult to establish in the absence of cosegregation data, partly because ALS pedigrees are small and difficult to collect due to the late onset nature of the disorder. Here, we performed a thorough genetic analysis to determine the prevalence of TUBA4A variants in a hospital-based ALS cohort of Han Chinese origin.


In all, 113 familial ALS (fALS) index cases, 610 patients with sporadic ALS (sALS) and 1005 ethnicity-matched Chinese controls were subjected to TUBA4A sequencing using a Sanger protocol. All identified variants were further studied according to their frequency in the genome aggregation (GnomAD) database. The prediction of pathogenicity was examined using the SIFT, PolyPhen2, Mutation Tester, AASites and NetGene2 programmes.


We screened the coding region of TUBA4A in 723 Chinese patients with ALS and 1005 controls, leading to the identification of three missense variants (p.I42T, p.I234T and p.A383T), one frameshift variant (p.H393Pfs*19) (online Supplementary table 1) and six synonymous substitutions (online Supplementary table 2). We detected a p.I42T variant (chr2:220116831A>G, c.125T>C, rs201865832), which was also present in three of 1005 Chinese controls, in a patient with sALS. Considering its minor allele frequency (0.0023) in East Asians in GnomAD, p.I42T may be a benign polymorphism. The p.I234T (chr2:220115720A>G, c.701T>C) and p.A383T (chr2:220115274C>T, c.1147G>A, rs368743618) variants were each present once in two Chinese sALS cases but were absent in the local ethnically matched controls. TUBA4A p.I234T is a novel variant, whereas p.A383T was previously found in an Italian patient with fALS and two individuals of African ancestry in GnomAD. Both p.I234T and p.A383T carriers have spinal onset and a classical ALS phenotype with ALS onset ages of 51 and 65 years, respectively. We identified a novel truncating TUBA4A variant, p.H393Pfs*19 (chr2:220115243_220115244insG, c.1177_1178insC), in a familial ALS proband who also harboured the SOD1 p.E133V variant (chr21:33040827A>T, c.401A>T). We performed a segregation analysis in this Chinese family of Han descent (figure 1) and revealed that only SOD1 p.E133V segregates with an ALS phenotype, whereas TUBA4A p.H393Pfs*19 failed to demonstrate cosegregation within affected members. Specifically, we detected SOD1 p.E133V in both III:6 and III:16 (affected), whereas TUBA4A p.H393Pfs*19 was absent in generation III:16. Both III:6 and III:16 developed signs of ALS in their fifth decade of life, which first manifested as weakness in the left hand muscles.

Supplementary file 1

Figure 1

Pedigree of the patient with familial amyotrophic lateral sclerosis with the SOD1 p.E133V mutation and the TUBA4A p.H393Pfs*19 variant. Black symbols represent patients affected with ALS; white symbols represent unaffected individuals. The arrow denotes the proband of the family. An asterisk at the top right represents family members whose DNA was available for SOD1 and TUBA4A gene sequencing. SOD1() represents individuals who are not SOD1 mutation carriers; SOD1(+) represents SOD1 p.E133V mutation carriers without the TUBA4A p.H393Pfs*19 variant and SOD1/TUBA4A (+) represents individuals with both the SOD1 p.E133V mutation and the TUBA4A p.H393Pfs*19 variant.


Our study presents the screening results for TUBA4A sequence variations in a large cohort of Chinese patients with ALS. We identified two missense mutations (p.I234T and p. A383T) in two of 610 patients with sALS, indicating that the frequency of TUBA4A mutations in Chinese patients with sporadic ALS is low (0.30%). Additionally, a truncating TUBA4A p.H393Pfs*19 variant was found in a familial ALS individual from a Chinese family, in which the TUBA4A variant failed to demonstrate cosegregation and SOD1 p.E133V segregated with the disease phenotype within this pedigree.

The Chinese p.A383T mutation carriers exhibited a predominantly classical limb-onset ALS phenotype, similar to that of the Italian p.A383T mutant fALS patient,1 and this was predicted to be deleterious by several predication programmes. We identified a novel TUBA4A p.I234T variant, which was predicted to be ‘deleterious’ by SIFT and ‘benign’ by PolyPhen-2. Although no segregation analysis was possible in sALS cases, because p.I234T and p.A383T are rare and located within a highly evolutionarily conserved region of TUBA4A encoded by exon 4, these two missense variants are likely ALS disease causing. The TUBA4A p.H393Pfs*19 frameshift variant coexisted with SOD1 p.E133V in the proband of an ALS family. However, a segregation analysis within the Chinese Han pedigree showed that SOD1 p.E133V, but not TUBA4A p.H393Pfs*19, segregates with the ALS phenotype. SOD1 p.E133V was previously reported in a Chinese fALS pedigree5 and is absent from several public databases. Therefore, the evidence presented here suggests that the SOD1 p.E133V mutation is the primary cause of disease in this family and that the TUBA4A variant may be a neutral polymorphism. Nevertheless, we cannot exclude the possibility that concomitant ALS-related genes within the same individual may influence the ALS phenotype. TUBA4A p.H393Pfs*19 and SOD1 p.E133V were also detected in 39-year-old (IV:1) and 36-year-old (IV:2) sons. Thus, long-term follow-up is required.

Combined with our results, 12 missense, one nonsense, and two frameshift mutations and one splice donor site variant have been identified in TUBA4A in both ALS and FTD cases across different populations.1 2 Among them, TUBA4A p.T145P and p.T381M demonstrated cosegregation in two first-degree relatives with ALS. However, p.H393Pfs*19 in a Chinese patient with fALS and p.K430N in a patient with fALS from the Netherlands did not segregate with disease,1 and judgement of TUBA4A p.T381M in the pathogenesis of ALS is also complicated due to its co-occurrence with C9ORF72 repeat expansion and the NEK1 p.Ser1036* variant.3 Even though individual TUBA4A variants have been demonstrated to have a disruptive effect on cytoskeletal integrity in vitro,1 no single TUBA4A variant is considered pathogenic due to the absence of functional data from primary tissue of TUBA4A mutation carriers. Therefore, assessing the effect of TUBA4A variants on ALS or FTD disease risk remains challenging.

The frequency of TUBA4A mutations suggests that TUBA4A is still a rare cause of ALS in Chinese patients. As genetic screening is more accessible in ALS clinical practice, the interpretation of the effects of TUBA4A variants on ALS risk should be interpreted with caution.


We thank all patients with ALS, family members and controls for their participation.



  • Contributors DF conceived this study and provided financial support. DF and JL designed the study, prepared and revised the manuscript; also had key roles in the study. JH, LT, LC and YM took part in the design of the study and in sample collection. JL, LT, LC and DF conducted data management. JL conducted data follow-up. LC and LT undertook data checking. JH, JL and DF undertook statistical analysis. DF was responsible for project management.

  • Funding This work was supported by the National Natural Science Foundation of China (81030019).

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval Ethics Committee at Peking University Third Hospital.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.