Amyotrophic lateral sclerosis (ALS) is a rare degenerative condition of the motor neurons. Over 10% of ALS cases are linked to monogenic mutations, with the remainder thought to be due to other risk factors, including environmental factors, genetic polymorphisms, and possibly gene-environmental interactions. We examined the association between ALS and an intermediate CAG repeat expansion in the ATXN2 gene using a meta-analytic approach. Observational studies were searched with relevant disease and gene terms from MEDLINE, EMBASE, and PsycINFO from January 2010 through to January 2014. All identified articles were screened using disease terms, gene terms, population information, and CAG repeat information according to PRISMA guidelines. The final list of 17 articles was further evaluated based on the study location, time period, and authors to exclude multiple usage of the same study populations: 13 relevant articles were retained for this study. The range 30–33 CAG repeats in the ATXN2 gene was most strongly associated with ALS. The meta-analysis revealed that the presence of an intermediate CAG repeat (30-33) in the ATXN2 gene was associated with an increased risk of ALS [odds ratio (OR) = 4.44, 95%CI: 2.91–6.76)] in Caucasian ALS patients. There was no significant difference in the association of this CAG intermediate repeat expansion in the ATXN2 gene between familial ALS cases (OR = 3.59, 1.58–8.17) and sporadic ALS cases (OR = 3.16, 1.88–5.32). These results indicate that the presence of intermediate CAG repeat expansion in the ATXN2 gene is a specific genetic risk factor for ALS, unlike monogenic mutations with an autosomal dominant transmission mode, which cause a more severe phenotype of ALS, with a higher prevalence in familial ALS.
Citation: Wang M-D, Gomes J, Cashman NR, Little J, Krewski D (2014) Intermediate CAG Repeat Expansion in the ATXN2 Gene Is a Unique Genetic Risk Factor for ALS−A Systematic Review and Meta-Analysis of Observational Studies. PLoS ONE 9(8): e105534. https://doi.org/10.1371/journal.pone.0105534
Editor: Harm H. Kampinga, UMCG, Netherlands
Received: March 18, 2014; Accepted: July 24, 2014; Published: August 22, 2014
Copyright: © 2014 Wang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and its Supporting Information files.
Funding: This work was supported by a contribution agreement with Public Health Agency of Canada to the University of Ottawa under the National Population Health Study of Neurological Conditions in Canada, for which D. Krewski served as Principal Investigator. D. Krewski holds the Natural Sciences and Engineering Research Council of Canada Chair in Risk Science at the University of Ottawa. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Amyotrophic lateral sclerosis (ALS) is a rare neurological degenerative condition of the motor neurons leading to paralysis of skeletal muscles, characterized by rapid irreversible progression in most cases . Although the causes of ALS are largely unknown, it is estimated that genetic and environmental risk factors contribute roughly equally to the development of ALS , . Over the last 20 years, over 10% of all ALS cases [∼68% of familial ALS (fALS), ∼11% of sporadic ALS (sALS)] have been linked to monogenic mutations in one of 30 or so genes –. The mutation frequencies for most known causal genes for ALS are rare among ALS cases, with most accounting for less than 1% of ALS cases. Relatively common mutations occur in the SOD1 (superoxide dismutase 1) gene, which represents the first genetic factor linked to ALS in 1993, and for which there is a valid animal model . Nearly 200 mutations in this gene have been reported in literature , . The mutations in this gene are responsible for about 2–5% of all ALS cases across all races , , SOD1 remained the most common mutated gene related to ALS until the discovery of the GGGGCC hexanucleotide repeat expansion in C9orf72 in ALS patients in 2011 , . We now know that the most common monogenic mutated gene in Caucasians with ALS is the GGGGCC hexanucleotide repeat expansion in the C9orf72 , , which is responsible for over 30% of fALS cases, and about 6.0% of sALS in Caucasian populations , , . This repeat mutation arose in Northern Europe a few thousand years ago , and thus occurs with a much lower frequency in Asian ALS populations –, in particular, the mutation frequency has been shown to be extremely low in Chinese, Indian, Japanese and Korean ALS patients , –. This mutation is also associated with about 25% of familial frontotemporal lobe dementia (FTLD) and about 6% sporadic FTLD , . Among 297 carriers of the hexanucleotide repeat expansion in C9orf72, the diagnoses of ALS, ALS/FTLD and FTLD were 43%, 26% and 31%, respectively . Consequently, the incidence of this mutation in Caucasians with ALS may have been underestimated, since the clinical determination of ALS was mainly based on differential diagnosis. Other relatively common mutations occurred in the TARDBP (5% of fALS, 2% of sALS), and FUS (1–5% of fALS, 1.0% of sALS) genes , . Both gene products belong to the RNA/DNA binding protein family, and are involved in the pathological protein aggregation pathway in ALS patients . The contribution of other types of genetic risk factors (polygenic gene mutations, variants of polymorphic genes) to the burden of ALS has also been investigated in over 130 studies involving about 400 variants in more than 100 genes , , . The abnormal expression of this type of gene variant alone is likely to be insufficient to initiate the onset of ALS; rather, exposure to certain environmental agents is thought to render genetic variant carriers susceptible to ALS . To date, extensive studies have not revealed a definitive, universally accepted environmental risk factor for ALS, although many environmental risk factors, such as exposure to heavy metals, exposure to pesticides, exposure to solvent, intake of biological toxicants, history of head injury, smoking, and military service have been reported to be associated with ALS in some studies –.
Variants of approximately 30 genes have been found to be associated with ALS development in at least one study. These associations have been complied in a website summarizing known genetic risk factors for ALS  and have been documented in many peer-reviewed publications. These associations need to be verified in further epidemiological studies, even for the most credible gene variants for sporadic ALS, such as the Paraoxonase 1 polymorphism , the H63D polymorphism in HFE , , and variants of the gene encoding vascular endothelial growth factor .
Over the last few years, repeat expansions in two genes in ALS patients have been investigated intensively. One is the GGGGCC hexanucleotide repeat expansion in the upstream of the C9orf72 coding region , , which causes ALS among Caucasians and populations with a Spanish ethnic background , but is rare among ALS cases with other ethnic backgrounds –. The mutation frequencies of this gene among ALS patients with various ethnic backgrounds have been summarized , . The second repeat mutated gene demonstrates a higher occurrence of intermediate CAG (coding for glutamine) repeats (polyglutamine, polyQ) in the 5 prime terminal of the ATXN2 gene in ALS patients . However, there is currently no agreement about which range of the intermediate CAG repeat is associated with ALS, nor its relative prevalence in fALS and sALS cases. To address these issues, we examined the association between this intermediate CAG repeat expansion in the ATXN2 gene and ALS using meta-analysis.
Materials and Methods
This systematic review and meta-analysis was conducted following the PRISMA guideline , major features of which are listed in Supporting Information (Checklist S1). Articles related to the association between ALS (disease terms: ALS, MND, amyotrophic lateral sclerosis, motor neuron disease, motoneurone disease, or Lou Gehrig's disease) and intermediate CAG repeat expansion in the ATXN2 gene (gene terms: ataxin-2, ATXN2, ATX2, SCA2, ASL13, TNRC13, polyglutamine, polyQ, CAG repeat) were searched in MEDLINE, EMBASE, PsycINFO, Pubmed, Hugenet and Google Scholar from January 1, 2010 through to January 20, 2014. The detailed search strategies are described in Supporting Information (Search S1). The retrieved articles were exported to Refworks for screening. First, duplicates were removed by comparing the authors and titles of adjacent records after sorting the articles by first author. Second, irrelevant articles were excluded by reading the titles and abstracts of records against our inclusion/exclusion criteria (at least one disease term and one gene term were required for inclusion). Third, relevant or ambiguous articles were retained for next level screening by examining the full PDF copy of the article. At this level of screening, letters to editors, commentaries, generic reviews, case reports only, conference abstracts, mechanistic and animal studies were excluded by examining full PDF. Articles without one of the essential characteristics of the study population (study location, case recruitment time, and method of recruitment) were also excluded at this screening level. Fourth, the remaining articles were assessed by examining the authors, research groups, country of study, case ascertainment, population recruitment periods, analytic methods, results, and conclusions based on the previous publication . Since ALS is a rare neurological condition, care was taken to avoid inclusion of multiple articles with same study participants. When two or more papers were from the same group of authors, special attention was given to information on the centers involved in, and periods of recruitment of the study participants. Any article with duplicate uses of ALS patient populations or any ambiguous description about the studied population was excluded. Fifth, data from the final list of included articles were extracted (including first author, published year, country, CAG repeat information, case and control information, and main results) (Table S1). The screening and data extraction steps were assessed by second reviewer, based on a random sample of 4 of 33 articles available at these steps. Finally, data were synthesized using meta-analytic approaches based on both random and fixed effects models . The synthetic odds ratio (OR), Tau2, I2 and χ2 statistics were reported, and potential bias due to small study effects was examined using funnel plots. Comparisons among population subgroups defined by geographic region and types of ALS were also made.
1. Studies selected for meta-analysis
A flow diagram summarizing article search, screen, selection, assessment, data extraction and analysis is given in Figure 1. No previous systematic review was identified, except one meta-analysis . Seventeen relevant articles were selected for inclusion in the present study (Table S1). Three studies were excluded because of multiple uses of same populations. One study was designed to test the presence of intermediate CAG in ALS-FTLD and familial ALS only, and was also excluded from the meta-analysis. The data from remaining 13 articles (first author, published year, country, CAG repeat information, case and control information, main results), all of which followed a case-control design, were extracted and synthesized by meta-analysis.
2. Determination of the range of intermediate CAG repeats in the ATXN2 gene that is associated with ALS
Since Elden and colleagues reported that a 27–33 CAG repeat expansion in the ATXN2 gene was associated with sporadic ALS in 2010 , , about two dozen studies have sought to verify this association , –, investigate the mechanisms by which ALS may develop , , or investigate different clinical manifestations in ALS patients carrying CAG repeat expansions in the ATXN2 gene –. The initial study focused on the range 27–33 in the number of CAG repeats, and determined this range as an intermediate CAG (PolyQ) repeat using ROC curves. However, no consensus has been reached regarding which CAG repeat range is associated with ALS , , , .
To address this issue, we compared the CAG repeat range between ALS patients and controls based on the information from the 13 studies selected for inclusion in Figure 2 , –. A high frequency of 27Q repeats was noticed in ALS patients, Parkinson's disease (PD) patients, and controls in Europe and North America (Figure 2a), but not in Chinese ALS patients or healthy controls (Figure 2b), where the peak of 27 CAG repeats in the ATXN2 gene was essentially absent , . These results indicated that the peak of 27 CAG repeats is not an ALS disease-specific CAG repeat, but might be a specific genomic marker of ethnicity . Although outside the scope of the present study, the higher rate of 24 or 25 CAG repeats in ALS patients (Figure 2a, 2b) may be relevant to the transmission of new larger CAG repeats to the next generation. Based on the 9 studies in North America and Europe in which detailed repeat information could be identified (Table S1), the pattern of CAG 30–33 repeats in the ATXN2 gene in ALS patients appeared to be different than in controls (Figure 2a). These data suggest that 30–33 CAG repeats is associated with ALS among Caucasian populations. Ethnic variation appears to exist, with the major CAG repeat range associated with ALS appearing to be 31–36 in Chinese subjects (Figure 2b) , . Meta-analyses were then conducted, with and without these two studies from China, to explore this association in quantitative terms.
3. The synthesized odds ratio (OR) estimate from multiple studies based on meta-analysis
The 13 studies included in the present meta-analysis involved a total of 154 positive carriers of 30–33 CAG repeats in the ATXN2 gene among 9,042 ALS cases (including fALS cases with no known monogenic mutations), and 46 positive carriers of the same CAG repeats among 13,116 controls. The crude prevalence rate of the CAG repeats (1.70%) in ALS cases was significantly higher than in controls (0.35%, P<0.0001). Comparing cases to controls, the odds ratio (OR) for ALS in relation to the 30–33 CAG repeat sequence was estimated to be OR = 3.93 (95% CI: 2.49–6.20) using a random effects model (Figure 3), with moderate heterogeneity across included studies as reflected by the I2 statistic [35%, P = 0.10]. Similar results (OR = 3.60, 2.54–5.09) were obtained with a fixed effects model (data not shown).
The data of intermediate CAG 30–33 repeats in ATXN2 were extracted from 13 included studies and the OR of intermediate CAG repeat among ALS and control subjects was synthesized with meta-analysis using random effects model. Similar results were also obtained when use fixed effects model (data not shown).
When the two studies from China were excluded, heterogeneity decreased (I2 = 13%). The remaining 11 articles reported a total of 134 positive carriers of 30–33 CAG repeats among 7,625 ALS cases (including sALS and fALS cases), and 37 positive carriers of the same CAG repeats among 12,555 controls (see Table S1). Excluding these studies, the odds ratio was estimated to be 4.44 (2.91–6.76) using a random effects model (Figure 4). A similar estimate was obtained with a fixed effects model (data not shown). Funnel plots failed to identify evidence of small study effects (data not shown).
4. The prevalence of intermediate CAG repeat in the ATXN2 gene in fALS cases is not different from that in sALS cases
Based on the above analyses, we concluded that the intermediate 30–33 CAG repeat expansion in the ATXN2 gene is a genetic risk factor for ALS. Further analyses were undertaken in an attempt to compare differences in risk between fALS and sALS. High CAG repeat expansions (usually greater than 34) in coding regions of the ATXN2 gene are the cause of spinocerebellar ataxia type 2 (SCA2) , transmitted in an autosomal dominant manner. However, if intermediate CAG 30–33 repeat expansion of ATXN2 also causes ALS in an autosomal dominant mode, then it might be expected that its prevalence in fALS cases would be higher than that in sALS cases (For example, the frequency of SOD1 mutations is higher in fALS than in sALS patients). To address this issue, we identified four included articles that provided relevant genomic information for fALS (defined as two or more ALS cases identified in a family) and sALS cases. Meta-analysis using a random effects model showed that the OR for the presence of the intermediate CAG repeat expansion in fALS cases [OR = 3.59 (1.58–8.14)] was not significantly higher than the OR in sALS cases [OR = 3.01 (1.77–5.11)] (Figure 5). The pooled prevalence of the intermediate CAG repeats was 1.32% and 1.58% among sALS and fALS cases, respectively, which are not significantly different from each other (χ2 = 0.25, p = 0.62). There was no evidence of heterogeneity (I2 = 0%) or publication bias across the four included studies. Similar results were also obtained with a fixed effects model (data not shown).
The relative risks (OR) among fALS cases (top panel) or sALS cases (lower Panel) compared to controls were synthesized with meta-analysis using extracted data from 4 included case-control studies. The results from random effects model were presented. Similar results were also obtained when use fixed effects model (data not shown).
5. The association between CAG repeat length and the age at onset or survival time among ALS patients remains unclear
Out of 13 included articles, 9 articles provided partial numerical data about the age of onset of ALS (116 ALS cases), disease duration (29 cases), and the corresponding CAG repeat numbers in ATXN2 (27,31,35,36,38,40,43–45) (Table S2). Although linear regression analysis showed that the number of CAG repeats in the ATXN2 gene was associated with neither the age at onset (R2 = 0.004) nor the disease duration (R2 = 0.02), these data are insufficient in our opinion to draw firm conclusions about these associations (see Table S2 for further details). No differences in disease progression between ALS patients with and without the intermediate CAG repeat expansion in ATXN2 were noted.
Discussion and Conclusions
In this study, we synthesized data from published articles related to intermediate CAG repeat expansions in the ATXN2 gene in individuals with ALS, and found that an intermediate CAG expansion with a range of 30–33 repeats was associated with an increased risk of ALS. Since significant differences with respect to the prevalence of these expansions among fALS cases compared to sALS cases were not identified, heritability is unlike the dominant autosomal transmission mode of inheritance of SCA2 observed for CAG expansions greater than 34 repeats in the ATXN2 gene. The intermediate CAG repeat may have been inherited differently in relation to ALS (see discussion below). Therefore, the intermediate CAG repeat expansion in the ATXN2 gene may not be a simple modulator of ALS , . We conclude that the intermediate CAG repeat expansion in ATXN2 is a unique genetic risk factor for ALS.
High CAG repeat expansions (usually greater than 34 CAG repeat) in the ATXN2 gene , transmitted in an autosomal dominant manner, is the cause of spinocerebellar ataxia type 2 (SCA2), a progressive neurodegenerative disease of the cerebellum, brain stem and spinal cord, with a somewhat higher incidence in Cuba than elsewhere , , . The intermediate expansion of CAG repeats in SCA2 cases has also been reported , , and coincidence of ALS and SCA2 patients in the same family has been observed , indicating a certain degree of overlap between SCA2 and ALS with CAG repeat expansion in the ATXN2 gene. The expansion of CAG repeat in the ATXN2 gene leads to the production of elongated polyglutamine (polyQ) in the corresponding protein. The locus of the ATXN2 gene has been mapped to chromosome 12, but the function of the ATXN2 genes product is not known . The CAG repeats in the ATXN2 gene are variable in size. The length of the allele in normal subjects was found to range from 14 to 31 repeats, with over 90% of normal subjects demonstrating alleles with 22 repeats . The SCA2 disease allele usually increases its size when transmitted to successive generations, and the longer expansions associate with earlier onset and more severe SCA2 disease in subsequent generations , different from the ALS cases with intermediate CAG repeat expansion (see Result section 5). Whether or not the intermediate repeat length of CAG (within the normal range, up to 31, and ≥33 for SCA2) was associated with some disease conditions was unknown until Elden and colleagues  first linked the intermediate CAG repeat expansion (27–33) to ALS in 2010. Other follow-up reports generally confirmed this initial finding, but there is no consensus with regards to which repeat length is associated with ALS. One study using ROC (receiver operating characteristic) curves showed that more than 29 CAG repeats in the ATXN2 gene was associated with an increased risk of ALS ; however, other studies have linked ALS with greater than 30 CAG repeats in Italy , 30–35 CAG repeats in Germany , and ≥28 CAG repeats in Italy . Here, we used 30–33 repeats as the cut-off for association with ALS because the difference between ALS cases and corresponding controls was not apparent when the CAG repeat in ATXN2 is less than 29 or greater than 34. Nonetheless, this intermediate CAG range for ALS might vary in different studies due to ethnicity , .
The products of the ATXN2 gene with intermediate CAG repeat expansion may not act like a typical gene modulator . The first study conducted by Elden and colleagues suggested that intermediate CAG repeats were a modulator for ALS . Further experiments showed such repeats might regulate RNA processing  and intracellular vesicular trafficking . Importantly, in vitro experiments showed that the expression products of ATXN2 with intermediate CAG repeats could interact with FUS , a DNA/RNA binding protein, which when mutated could cause fALS as a result of impaired RNA intracellular trafficking , . In addition, TDP-43 cytoplasmic inclusions in motor neurons of ALS patients harboring intermediate PolyQ repeats primarily showed skein-like or filamentous TDP-43 pathology, whereas ALS cases without ataxin-2 polyQ expansions (n = 13) exhibited abundant large round TDP-43 inclusions , , and accumulated activated caspase 3 in motor neurons, an upstream event in the TDP-43 disease pathway. The product of ATXN2 with intermediate CAG repeat expansion could increase the presence of TDP-43 in the cytoplasm in a RNA dependent manner . Moreover, plasma TDP-43 protein was robustly associated with CAG repeat length . These biochemical observations indicated that the expression products of ATXN2 with intermediate CAG expansions interact with two important genes (TDP-43 and FUS) in ALS pathogenesis. However, it seems that ataxin-2 with intermediate polyQ does not function like a typical modulator of other genetic risk factors for ALS (specifically TDP-43 and FUS), because the mutation of a modulator alone should not cause primary outcome of interest (i.e., ALS). Even if the expression products of ATXN2 with intermediate CAG expansion were a modulator of TDP-43 or FUS, it must be a weak or very specific modulator, because differences in clinical features of ALS, such as age at onset or survival time, between the CAG repeat carriers and non-carriers have not been established (see Result section 5). During the course of finalizing the present paper, a new article describing the coexistence of intermediate CAG repeat expansion in the ATXN2 gene and the hexanucleotide repeat expansion in C9orf72 was published online . Compared to controls, the frequency (∼2%) of intermediate CAG repeat expansions in the ATXN2 gene was higher among the carriers of the GGGGCC hexanucleotide repeat expansion in C9orf72 . However, this study did not provide clear evidence that the intermediate CAG repeat expansion in the ATXN2 gene was a modifier to the disease caused by the hexanucleotide repeat expansion in gene C9orf72.
If the ATXN2 gene with intermediate CAG repeat expansion associated with ALS were transmitted in an autosomal dominant manner, as in the case of SOD1 mutations, or as in the case of SCA2 for longer CAG repeat expansions in SCA2, then we would expect that the prevalence of the intermediate CAG repeat expansion in the ATXN2 gene be higher in fALS cases than in sALS cases, as is the case for other genetic risk factors for ALS, such as mutations in SOD1 and C9orf72. Although studies from Cuba and the United States suggested a dominant mode of inheritance , , , our meta-analyses did not show a higher prevalence of the CAG intermediate repeat expansion in fALS cases, as compared to sALS cases. Therefore, the intermediate CAG repeat expansion in the ATXN2 gene may not be a typical independent classic genetic risk factor transmitted in an autosomal dominant manner. Further studies with well diagnosed fALS cases are required to determine the mode of transmission of the intermediate CAG repeats in the ATXN2 gene.
The intermediate repeats in ATXN2 may be transmitted in recessive mode. Van Damme  described a consanguineous married couple, the father suffering from an ALS-like disease with two affected sons [elder brother (79 years old) with CAG repeat alleles 31∶33, onset at 75 years, pedigree (middle son, 77 years old) with CAG repeat alleles 33∶33, onset at 71 years] and a normal carrier younger brother (75 years) with 22∶33 CAG repeat alleles. These observations suggest a recessive transmission mode, unlike a SCA2 family with longer CAG repeats descried in the same article . The transmission of intermediate CAG repeat expansion of ATXN2 gene might behave like the Asp90Ala mutation in SOD1, which was transmitted in a recessive manner in the North Scandinavian population, but in a dominant manner in other regions . The longer CAG repeat expansions for SCA2 are transmitted in a dominant manner, whereas this is not the case for ALS with intermediate CAG repeat in ATXN2. However, we could not rule out the possibility of lower inheritance penetration in a dominant autosomal transmission mode.
The different modes of transmission might be dictated by the different nature of the CAG repeat expansions in ALS and SCA2. Compared to the longer CAG repeats in SCA2 patients, the unique feature in ALS patients with intermediate CAG repeats in the ATXN2 gene is that the intermediate CAG repeats in ALS patients were interrupted by a CAA triplet in almost all ALS cases with this CAG intermediate repeat expansion . This finding has been confirmed by other similar observations . The ATXN2 gene without CAA triplets, which stabilizes the CAG repeats , may produce ataxin-2 protein with longer PolyQ repeats which is more toxic to the host cells, compared to the ataxin-2 protein produced by the gene with the CAA triplet. This type of triplet is less common in longer CAG repeats in the ATXN2 gene in SCA2 patients , , , . The CAA interruption may explain why the intermediate CAG repeat expansion in ATXN2 is associated with ALS, and does not generally produce a mild type of SCA2. Other mechanisms may also influence this comparison. For example, the co-occurrence of fALS with SCA2, or with Parkinson disease, would lead to non-diagnosis of ALS in fALS , , , thus underestimating the occurrence of the intermediate CAG repeats among familial ALS cases.
Collectively, the results of the present study suggest that CAG intermediate (30–33) repeat expansions in the ATXN2 gene is a genetic risk factor for ALS. During the course of this work, a meta-analysis of the association between intermediate CAG repeats and ALS for a family with 3 ALS cases linked to CAG repeat mutations in ATXN2, apparently with an autosomal dominant transmission mode, was reported from Cuba . This article adopted a similar inclusion and exclusion strategy, and identified many of the same studies as were used in the present meta-analysis, with the exception that we also included the two most recent published studies , . However, that study compared the synthetic ORs to determine the cut-off for the intermediate CAG repeat expansion (≥30 CAG repeats) that associates with ALS, compared the range used in the present analysis (30–33 CAG repeats). Because they also included gene carriers with more than 34 CAG repeats (which usually causes SCA2), the synthetic risk was artificially underestimated (OR = 2.16), as compared to our estimate (OR = 4.44, Figure 4). We argue that our strategy of determining the optimal range of the intermediate CAG repeats for ALS is preferable, because it is known that if the number of CAG repeats is greater than 34, then the repeat is more likely to be associated with SCA2, rather than ALS .
Further studies are needed to verify our conclusion that intermediate CAG repeat expansion in the ATXN2 gene is a unique genetic risk factor for ALS. Future research should focus on the mechanisms involved in the etiology of ALS among intermediate CAG repeat carriers, and explore the variation in repeats among ALS patients with different ethnic backgrounds. Since our conclusions were drawn mainly from studies conducted in Europe, North America, and China, further observational studies from other regions would be helpful in establishing the generalizability of our findings.
Summary of studies included in meta-analyses of the ATXN2 gene as a risk factor for ALS.
Age at onset, disease duration and CAG repeat number in ALS patients.
PRISMA checklist and the section location of each item in the manuscript.
The article search process was greatly helped by Lindsey Sikora, research librarian at the University of Ottawa with expertise in systematic review, and Mona Hersi, who conducted a systematic review of factors affecting the onset and progression of Alzheimer's disease at the University of Ottawa as part of her MSc in epidemiology at the University of Ottawa.
Conceived and designed the experiments: MDW JG NRC DK JL. Performed the experiments: MDW JG. Analyzed the data: MDW JG DK. Contributed reagents/materials/analysis tools: MDW JG DK. Contributed to the writing of the manuscript: MDW NRC JL DK.
- 1. Talman P, Forbes A, Mathers S (2009) Clinical phenotypes and natural progression for motor neuron disease: analysis from an Australian database. Amyotroph Lateral Scler 10: 79–84.
- 2. Al-Chalabi A, Fang F, Hanby MF, Leigh PN, Shaw CE, et al. (2010) An estimate of amyotrophic lateral sclerosis heritability using twin data. Journal of neurology, neurosurgery, and psychiatry, December 01 PMC2988617: 1324–1326.
- 3. Wingo TS, Cutler DJ, Yarab N, Kelly CM, Glass JD (2011) The heritability of amyotrophic lateral sclerosis in a clinically ascertained United States research registry. PLoS One 6: e27985.
- 4. Al-Chalabi A, Jones A, Troakes C, King A, Al-Sarraj S, et al. (2012) The genetics and neuropathology of amyotrophic lateral sclerosis. Acta Neuropathol 124: 339–352.
- 5. Al-Chalabi A, Lewis CM (2011) Modelling the effects of penetrance and family size on rates of sporadic and familial disease. Hum Hered 71: 281–288.
- 6. Baumer D, Talbot K, Turner MR (2014) Advances in motor neurone disease. J R Soc Med 107: 14–21.
- 7. Renton AE, Chio A, Traynor BJ (2014) State of play in amyotrophic lateral sclerosis genetics. Nat Neurosci 17: 17–23.
- 8. Ripps ME, Huntley GW, Hof PR, Morrison JH, Gordon JW (1995) Transgenic mice expressing an altered murine superoxide dismutase gene provide an animal model of amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A 92: 689–693.
- 9. ALS Online Genetics Database, World Federation of Neurology and European Network to Cure ALS (Jan. 2014) 2014.
- 10. Mulligan VK, Chakrabartty A (2013) Protein misfolding in the late-onset neurodegenerative diseases: common themes and the unique case of amyotrophic lateral sclerosis. Proteins 81: 1285–1303.
- 11. Jones CT, Brock DJ, Chancellor AM, Warlow CP, Swingler RJ (1993) Cu/Zn superoxide dismutase (SOD1) mutations and sporadic amyotrophic lateral sclerosis. Lancet 342: 1050–1051.
- 12. Ticozzi N, Tiloca C, Morelli C, Colombrita C, Poletti B, et al. (2011) Genetics of familial amyotrophic lateral sclerosis. Arch Ital Biol 149: 65–82.
- 13. Renton AE, Majounie E, Waite A, Simon-Sanchez J, Rollinson S, et al. (2011) A hexanucleotide repeat expansion in C9ORF72 is the cause of chromosome 9p21-linked ALS-FTD. Neuron 72: 257–268.
- 14. DeJesus-Hernandez M, Mackenzie IR, Boeve BF, Boxer AL, Baker M, et al. (2011) Expanded GGGGCC hexanucleotide repeat in noncoding region of C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron 72: 245–256.
- 15. Rademakers R, van Blitterswijk M (2013) Motor neuron disease in 2012: Novel causal genes and disease modifiers. Nat Rev Neurol 9: 63–64.
- 16. Cruts M, Gijselinck I, Van Langenhove T, van der Zee J, Van Broeckhoven C (2013) Current insights into the C9orf72 repeat expansion diseases of the FTLD/ALS spectrum. Trends Neurosci 36: 450–459.
- 17. Smith BN, Newhouse S, Shatunov A, Vance C, Topp S, et al. (2013) The C9ORF72 expansion mutation is a common cause of ALS+/−FTD in Europe and has a single founder. Eur J Hum Genet 21: 102–108.
- 18. Tsai CP, Soong BW, Tu PH, Lin KP, Fuh JL, et al. (2012) A hexanucleotide repeat expansion in C9ORF72 causes familial and sporadic ALS in Taiwan. Neurobiol Aging 33: 2232.e11–2232.e18.
- 19. Konno T, Shiga A, Tsujino A, Sugai A, Kato T, et al. (2013) Japanese amyotrophic lateral sclerosis patients with GGGGCC hexanucleotide repeat expansion in C9ORF72. J Neurol Neurosurg Psychiatry 84: 398–401.
- 20. Alavi A, Nafissi S, Rohani M, Shahidi G, Zamani B, et al. (2014) Repeat expansion in C9ORF72 is not a major cause of amyotrophic lateral sclerosis among Iranian patients. Neurobiol Aging 35: 267.e1–267.e7.
- 21. Zou ZY, Li XG, Liu MS, Cui LY (2013) Screening for C9orf72 repeat expansions in Chinese amyotrophic lateral sclerosis patients. Neurobiol Aging 34: 1710.e5–1710.e6.
- 22. Liu Y, Yu JT, Zong Y, Zhou J, Tan L (2014) C9ORF72 mutations in neurodegenerative diseases. Mol Neurobiol 49: 386–398.
- 23. Jiao B, Tang B, Liu X, Yan X, Zhou L, et al. (2014) Identification of C9orf72 repeat expansions in patients with amyotrophic lateral sclerosis and frontotemporal dementia in mainland China. Neurobiol Aging 35: 936.e19–936.e22.
- 24. Majounie E, Renton AE, Mok K, Dopper EG, Waite A, et al. (2012) Frequency of the C9orf72 hexanucleotide repeat expansion in patients with amyotrophic lateral sclerosis and frontotemporal dementia: a cross-sectional study. Lancet Neurol 11: 323–330.
- 25. van Blitterswijk M, Mullen B, Heckman MG, Baker MC, DeJesus-Hernandez M, et al.. (2014) Ataxin-2 as potential disease modifier in C9ORF72 expansion carriers. Neurobiol Aging.
- 26. Polymenidou M, Lagier-Tourenne C, Hutt KR, Bennett CF, Cleveland DW, et al. (2012) Misregulated RNA processing in amyotrophic lateral sclerosis. Brain Res 1462: 3–15.
- 27. Lill CM, Abel O, Bertram L, Al-Chalabi A (2011) Keeping up with genetic discoveries in amyotrophic lateral sclerosis: The ALSoD and ALSGene databases. Amyotroph Lateral Scler 12: 238–249.
- 28. PRIZE4LIFE (nd) ALSgene. Available at: http://www.alsgene.org/. Accessed 2014 Jan 25.
- 29. Al-Chalabi A, Hardiman O (2013) The epidemiology of ALS: a conspiracy of genes, environment and time. Nat Rev Neurol 9: 617–628.
- 30. Meireles A, Al-Chalabi A (2009) Genetic studies of amyotrophic lateral sclerosis: Controversies and perspectives. Amyotrophic Lat Scler 10: 1–14.
- 31. Johnson F, Atchison W (2009) The role of environmental mercury, lead and pesticide exposure in development of amyotrophic lateral sclerosis. Neurotoxicology 30: 761–765.
- 32. Sutedja NA, Veldink JH, Fischer K, Kromhout H, Heederik D, et al. (2009) Exposure to chemicals and metals and risk of amyotrophic lateral sclerosis: a systematic review. Amyotroph Lateral Scler 10: 302–309.
- 33. Sutedja NA, Fischer K, Veldink JH, van der Heijden GJMG, Kromhout H, et al. (2009) What we truly know about occupation as a risk factor for ALS: a critical and systematic review. Amyotrophic lateral sclerosis: official publication of the World Federation of Neurology Research Group on Motor Neuron Diseases 10: 295–301.
- 34. Ahmed A, Wicklund MP (2011) Amyotrophic lateral sclerosis: what role does environment play? Neurol Clin 29: 689–711.
- 35. Migliore L, Coppede F (2009) Genetics, environmental factors and the emerging role of epigenetics in neurodegenerative diseases. Mutat Res 667: 82–97.
- 36. Abel O, Powell JF, Andersen PM, Al-Chalabi A (2012) ALSoD: A user-friendly online bioinformatics tool for amyotrophic lateral sclerosis genetics. Hum Mutat 33: 1345–1351.
- 37. Wills AM, Cronin S, Slowik A, Kasperaviciute D, Van Es MA, et al. (2009) A large-scale international meta-analysis of paraoxonase gene polymorphisms in sporadic ALS. Neurology 73: 16–24.
- 38. van Rheenen W, Diekstra FP, van Doormaal PT, Seelen M, Kenna K, et al. (2013) H63D polymorphism in HFE is not associated with amyotrophic lateral sclerosis. Neurobiol Aging 34: 1517.e5–1517.e7.
- 39. Sutedja NA, Sinke RJ, Van Vught PW, Van der Linden MW, Wokke JH, et al. (2007) The association between H63D mutations in HFE and amyotrophic lateral sclerosis in a Dutch population. Arch Neurol 64: 63–67.
- 40. Lambrechts D, Poesen K, Fernandez-Santiago R, Al-Chalabi A, Del Bo R, et al. (2009) Meta-analysis of vascular endothelial growth factor variations in amyotrophic lateral sclerosis: increased susceptibility in male carriers of the-2578AA genotype. J Med Genet 46: 840–846.
- 41. Dejesus-Hernandez M, Rayaprolu S, Soto-Ortolaza AI, Rutherford NJ, Heckman MG, et al. (2013) Analysis of the C9orf72 repeat in Parkinson's disease, essential tremor and restless legs syndrome. Parkinsonism Relat Disord 19: 198–201.
- 42. Garcia-Redondo A, Dols-Icardo O, Rojas-Garcia R, Esteban-Perez J, Cordero-Vazquez P, et al. (2013) Analysis of the C9orf72 gene in patients with amyotrophic lateral sclerosis in Spain and different populations worldwide. Hum Mutat 34: 79–82.
- 43. Elden AC, Kim H, Hart MP, Chen-Plotkin AS, Johnson BS, et al. (2010) Ataxin-2 intermediate-length polyglutamine expansions are associated with increased risk for ALS. Nature 466: 1069–1075.
- 44. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, et al. (2009) The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol 62: e1–34.
- 45. Downs SH, Black N (1998) The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health 52: 377–384.
- 46. Laffita-Mesa JM, Rodriguez Pupo JM, Moreno Sera R, Vazquez Mojena Y, Kouri V, et al. (2013) De Novo Mutations in Ataxin-2 Gene and ALS Risk. PLoS One 8: e70560.
- 47. Corrado L, Mazzini L, Oggioni GD, Luciano B, Godi M, et al. (2011) ATXN-2 CAG repeat expansions are interrupted in ALS patients. Hum Genet 130: 575–580.
- 48. Conforti FL, Spataro R, Sproviero W, Mazzei R, Cavalcanti F, et al. (Dec 2012) Ataxin-1 and ataxin-2 intermediate-length PolyQ expansions in amyotrophic lateral sclerosis. Neurology 79: 2315–2320.
- 49. Bonini NM, Gitler AD (2011) Model organisms reveal insight into human neurodegenerative disease: ataxin-2 intermediate-length polyglutamine expansions are a risk factor for ALS. J Mol Neurosci 45: 676–683.
- 50. Ross OA, Rutherford NJ, Baker M, Soto-Ortolaza AI, Carrasquillo MM, et al. (2011) Ataxin-2 repeat-length variation and neurodegeneration. Hum Mol Genet 20: 3207–3212.
- 51. Van Langenhove T, van der Zee J, Engelborghs S, Vandenberghe R, Santens P, et al. (May 2012) Ataxin-2 polyQ expansions in FTLD-ALS spectrum disorders in Flanders-Belgian cohorts. Neurobiol Aging 33: e17–e20.
- 52. Gellera C, Ticozzi N, Pensato V, Nanetti L, Castucci A, et al. (Aug 2012) ATAXIN2 CAG-repeat length in Italian patients with amyotrophic lateral sclerosis: Risk factor or variant phenotype? Implication for genetic testing and counseling. Neurobiol Aging 33: e15–e21.
- 53. Lee T, Li YR, Chesi A, Hart MP, Ramos D, et al. (Jun 2011) Evaluating the prevalence of polyglutamine repeat expansions in amyotrophic lateral sclerosis. Neurology 76: 2062–2065.
- 54. Daoud H, Belzil V, Martins S, Sabbagh M, Provencher P, et al. (Jun 2011) Association of long ATXN2 CAG repeat sizes with increased risk of amyotrophic lateral sclerosis. Arch Neurol 68: 739–742.
- 55. Van Damme P, Veldink JH, van Blitterswijk M, Corveleyn A, van Vught PW, et al. (2011) Expanded ATXN2 CAG repeat size in ALS identifies genetic overlap between ALS and SCA2. Neurology 76: 2066–2072.
- 56. Lee T, Li YR, Ingre C, Weber M, Grehl T, et al. (2011) Ataxin-2 intermediate-length polyglutamine expansions in European ALS patients. Hum Mol Genet 20: 1697–1700.
- 57. Gispert S, Kurz A, Waibel S, Bauer P, Liepelt I, et al. (2012) The modulation of Amyotrophic Lateral Sclerosis risk by Ataxin-2 intermediate polyglutamine expansions is a specific effect. Neurobiol Dis 45: 356–361.
- 58. Liu X, Lu M, Tang L, Zhang N, Chui D, et al. (Sep 2013) ATXN2 CAG repeat expansions increase the risk for Chinese patients with amyotrophic lateral sclerosis. Neurobiol Aging 34: e5–e8.
- 59. Chen Y, Huang R, Yang Y, Chen K, Song W, et al. (Oct 2011) Ataxin-2 intermediate-length polyglutamine: A possible risk factor for Chinese patients with amyotrophic lateral sclerosis. Neurobiol Aging 32: e1–e5.
- 60. Laffita J, Bauer PO, Kouri V, Pena Serrano L, Roskams J, et al. (2012) Epigenetic DNA-methylation in the core ataxin-2 gene promoter: Novel physiological and pathological implications. Parkinsonism and Related Disorders 18: S187.
- 61. Lahut S, Omur O, Uyan O, Agim ZS, Ozoguz A, et al. (2012) ATXN2 and its neighbouring gene SH2B3 are associated with increased ALS risk in the Turkish population. PLoS One 7: e42956.
- 62. Yu Z, Zhu Y, Chen-Plotkin AS, Clay-Falcone D, McCluskey L, et al. (2011) PolyQ repeat expansions in ATXN2 associated with ALS are CAA interrupted repeats. PLoS One 6: e17951.
- 63. Hart MP, Gitler AD (Jul 2012) ALS-associated ataxin 2 polyQ expansions enhance stress-induced caspase 3 activation and increase TDP-43 pathological modifications. The Journal of Neuroscience 32: 9133–9142.
- 64. Drost J, Nonis D, Eich F, Leske O, Damrath E, et al. (2013) Ataxin-2 modulates the levels of Grb2 and SRC but not ras signaling. J Mol Neurosci 51: 68–81.
- 65. Gispert S, Kurz A, Waibel S, Bauer P, Liepelt I, et al. (2012) The modulation of Amyotrophic Lateral Sclerosis risk by ataxin-2 intermediate polyglutamine expansions is a specific effect. Neurobiol Dis 45: 356–361.
- 66. Chen X, Sun H, Zhang C, Zhang Y, Lin K, et al. (2013) Positive selection of CAG repeats of the ATXN2 gene in Chinese ethnic groups. Journal of Genetics and Genomics 40: 543–548.
- 67. Geschwind DH, Perlman S, Figueroa CP, Treiman LJ, Pulst SM (1997) The prevalence and wide clinical spectrum of the spinocerebellar ataxia type 2 trinucleotide repeat in patients with autosomal dominant cerebellar ataxia. Am J Hum Genet 60: 842–850.
- 68. Soraru G, Clementi M, Forzan M, Orsetti V, D'Ascenzo C, et al. (Jun 2011) ALS risk but not phenotype is affected by ataxin-2 intermediate length polyglutamine expansion. Neurology 76: 2030–2031.
- 69. Laffita-Mesa JM, Velazquez-Perez LC, Santos Falcon N, Cruz-Marino T, Gonzalez Zaldivar Y, et al. (2012) Unexpanded and intermediate CAG polymorphisms at the SCA2 locus (ATXN2) in the Cuban population: evidence about the origin of expanded SCA2 alleles. Eur J Hum Genet 20: 41–49.
- 70. Costanzi-Porrini S, Tessarolo D, Abbruzzese C, Liguori M, Ashizawa T, et al. (2000) An interrupted 34-CAG repeat SCA-2 allele in patients with sporadic spinocerebellar ataxia. Neurology 54: 491–493.
- 71. Fernandez M, McClain ME, Martinez RA, Snow K, Lipe H, et al. (2000) Late-onset SCA2: 33 CAG repeats are sufficient to cause disease. Neurology 55: 569–572.
- 72. Tazen S, Figueroa K, Kwan JY, Goldman J, Hunt A, et al. (2013) Amyotrophic lateral sclerosis and spinocerebellar ataxia type 2 in a family with full CAG repeat expansions of ATXN2. Movement Disorders 28: S241–S242.
- 73. Gispert S, Twells R, Orozco G, Brice A, Weber J, et al. (1993) Chromosomal assignment of the second locus for autosomal dominant cerebellar ataxia (SCA2) to chromosome 12q23-24.1. Nat Genet 4: 295–299.
- 74. Magana JJ, Velazquez-Perez L, Cisneros B (2013) Spinocerebellar ataxia type 2: clinical presentation, molecular mechanisms, and therapeutic perspectives. Mol Neurobiol 47: 90–104.
- 75. Farg MA, Soo KY, Warraich ST, Sundaramoorthy V, Blair IP, et al. (2013) Ataxin-2 interacts with FUS and intermediate-length polyglutamine expansions enhance FUS-related pathology in amyotrophic lateral sclerosis. Hum Mol Genet 22: 717–728.
- 76. Lattante S, Rouleau GA, Kabashi E (2013) TARDBP and FUS mutations associated with amyotrophic lateral sclerosis: summary and update. Hum Mutat 34: 812–826.
- 77. Hart MP, Brettschneider J, Lee VMY, Trojanowski JQ, Gitler AD (Aug 2012) Distinct TDP-43 pathology in ALS patients with ataxin 2 intermediate-length polyQ expansions. Acta Neuropathol 124: 221–230.
- 78. Andersen PM, Forsgren L, Binzer M, Nilsson P, Ala-Hurula V, et al. (1996) Autosomal recessive adult-onset amyotrophic lateral sclerosis associated with homozygosity for Asp90Ala CuZn-superoxide dismutase mutation. A clinical and genealogical study of 36 patients. Brain 119 (Pt 4): 1153–1172.
- 79. Choudhry S, Mukerji M, Srivastava AK, Jain S, Brahmachari SK (2001) CAG repeat instability at SCA2 locus: anchoring CAA interruptions and linked single nucleotide polymorphisms. Hum Mol Genet 10: 2437–2446.
- 80. Ramos EM, Keagle P, Gillis T, Lowe P, Mysore JS, et al. (2012) Prevalence of Huntington's disease gene CAG repeat alleles in sporadic amyotrophic lateral sclerosis patients. Amyotrophic Lat Scler 13: 265–269.
- 81. Ramos EM, Martins S, Alonso I, Emmel VE, Saraiva-Pereira ML, et al. (2010) Common origin of pure and interrupted repeat expansions in spinocerebellar ataxia type 2 (SCA2). Am J Med Genet B Neuropsychiatr Genet 153B: 524–531.
- 82. BragaNeto P, Pedroso JL, Felicio AC, Abrahao A, Dutra LA, et al. (Apr 2011) SCA2 presenting as an ataxia-Parkinsonism-motor neuron disease syndrome. Arq Neuropsiquiatr 69(: 405–406.
- 83. Qureshi AI, Wilmot G, Dihenia B, Schneider JA, Krendel DA (1996) Motor neuron disease with parkinsonism. Arch Neurol 53: 987–991.