Essential tremor (ET) is one of the most common movement disorders, affecting 5% of the general population older than 65 years.1 Sometimes referred to as “familial” tremor, ET has long been appreciated to be a heritable condition. Twin studies demonstrated high heritability with a concordance of 69-93% in monozygotic twins.2 While a large number of relatively rare monogenic movement disorders have been defined with the identification of genetic causes (i.e., pathogenic variants) over the past years, the genetic contributions to common movement disorders such as ET still await elucidation.3 With the hypothesis of “common disease – common variants” in mind, researchers have performed genome-wide association studies (GWAS) to identify genetic risk factors by comparing the frequencies of common variants in thousands of patients to those in thousands of healthy controls. Applying this approach for Parkinson’s disease (PD) has identified ~100 risk variants for PD.4 In contrast, limited insights have been gained into the genetic basis of ET until recently. Only a few risk loci were suggested for ET, but none of these loci showed a genome-wide significant signal, nor could they be convincingly replicated in subsequent studies.5 However, a recent GWAS in ET by Liao and colleagues,6 using a large sample size comprising ~7,000 patients and >450,000 controls, revealed five genome-wide significant risk loci, which lacked clear overlap to previously implicated genes. These common variants were estimated to explain only 18% of the heritability of ET, which is notably less than the estimated heritability based on twin studies. Interestingly, the newly identified genome-wide significant risk variants may be related to axonogenesis and point to a role of the cerebellum in ET. Another interesting finding of this recent GWAS was the potential role of transcriptional changes in BACE2, which encodes a β-secretase homolog capable of cleaving amyloid β precursor protein to result in the formation of amyloid-β protein.
What would you consider the major discoveries in the genetics of ET in the past years?
ET is an inherited condition in 50-70% of the cases, so there is no doubt that genetic factors play a substantial etiological role. However, many families with ET do not follow a Mendelian inheritance pattern, not even when considering incomplete penetrance, suggesting that ET has a multifaceted (genetic) basis. Initial approaches to investigate the genetic causes of ET have been through linkage analyses of families with many members diagnosed with ET. Linkage studies have mapped several susceptibility loci in families of different ethnicity (ETM1, ETM2, and ETM3 loci). However, no clear candidate genes for ET were identified in these regions, and the loci have been found only in single families. The first GWAS in ET identified mutations in LINGO17 and SLC1A28, but replication studies did not confirm their role as risk factors for ET. Additional GWAS have nominated other ET loci, but none of these loci was statistically significant at a genome-wide level, likely owing to the rather small size of the cohorts examined. Further, exome sequencing studies identified potential mutations in several genes, but replication studies failed to identify these genes as responsible for ET in other families.
For me, the major discoveries in the genetics of ET were the results of the GWAS run in Canada involving subjects from 16 centers worldwide, analyzing about half a million subjects, and revealing five independent genome-wide significant loci able to explain about one-fifth of ET heritability.6 Functional analyses found significant enrichment of the GWAS hits in the cerebellar hemisphere, cerebellum, and axonogenesis pathways, offering insights into biological mechanisms involved in ET pathogenesis. All this was very exciting.
How can we make further progress?
Well, genetics in ET is a curious thing. Before Liao and colleagues finished the paper6, I thought we wouldn't get anywhere with our current cohorts. There were just decades of research, and essentially all that came out was that everything is complicated. I think it would be helpful, if there was any solution to create homogeneous cohorts based on clinical data which then have a more common genetic background. I think we need to phenotype much deeper, and then we need to go back with this information to genetics. We still have controversies about clinical features and will need a much more detailed description of the course of the disease. We have cofactors in our current cohorts that mask the hidden, heterogeneous genetic basis. Therefore, in my opinion, more precise, ‘next-generation phenotyping’ and possibly also imaging as well as long-term follow-up examinations will generate more homogeneous cohorts of patients that are also more homogeneous in terms of their genetic basis. For this, we need to find resources that will enable a coordinated global effort. This is what the tremor community is currently working on.
It’s common for geneticists to claim that additional data and studies are needed to verify findings. Liao and colleagues6 did a good step forward into deciphering the genetic basis of ET using several cohorts, including clinical cohorts and population-based cohorts such as UK Biobank and 23andMe. The still-missing heritability (18% resolved by the GWAS findings but ~70% estimated heritability by twin studies) can be explained by the limitations of GWAS and estimation methods. The Linkage Disequilibrium Score regression (LDSC) that is used as a mathematical tool to measure the genetic correlation between two summary statistics is only based on the common variants genotyped and imputed during the GWAS analysis. In contrast, twin studies also reflect other heritable genomic components such as rare variants as well as the shared environment, especially in early life, in their heritability estimates.
An interesting outcome, and something that must be investigated in future studies, is the genetic correlation between included cohorts in a study (i.e., a breakdown of each included cohort by size, age distribution, sex distribution, etc.) reassuring that the finding is robust by demonstrating a high correlation between cohorts. The genetic correlation can be reported using an rg-value ranging from -1 to 1, where -1 is a perfect negative correlation, 0 means no correlation, and 1 perfect positive correlation. In the most recent GWAS,6 the genetic correlation was high between the two clinical cohorts included (rg>0.87). However, there was a lower correlation (rg=0.52) when compared with the 23andMe control cohort, which is a bit concerning. This might be attributed to the fact that 23andMe phenotyping is often based on self-report by the customers. Of note, the genetic correlation was significantly lower than, for example, the genetic correlation between 23andMe’s self-reported PD information when compared to clinically-determined PD (rg=0.85) and also lower compared to UK Biobank proxy cases, defined as a person having a parent with PD (rg=0.84).4 Thus, further studies are warranted in independent, reasonably sized clinical and control cohorts.
Worth mentioning, we have seen with many other GWAS (including PD and Alzheimer's disease) that the number of associated GWAS regions increases relatively linearly, based on the number of included cases. So, in short, to make further progress, we indeed need larger and, importantly, more ancestrally diverse studies.
ET is a highly heritable condition with an additive common variability of ~75%. Despite the overlapping symptomatology of ET and PD with tremor, this high heritability does not seem to be due to the misdiagnosis with PD. Further, ET subjects do not seem to share many genetic etiologic factors with PD subjects. These results suggest that ET is a genetic condition caused, in most cases, by the sum of effects of many loci. Therefore, yet unidentified genetic mutations are likely responsible for ET, but the inheritance of these variants is probably very complex, and other epigenetic and environmental factors can modify the expressivity of ET, obfuscating the genetic basis. The contribution of environmental factors to disease etiology has been examined in numerous epidemiological studies of PD, Alzheimer’s disease, and amyotrophic lateral sclerosis, but surprisingly few studies have been focused on studying the role of environmental factors in ET. Notably, the available studies in ET addressing lifestyle, metals, and dietary factors have shown controversial results.9 Among the environmental factors most studied is harmane, a dietary tremor-producing toxin, whose blood and brain levels have been found to be increased in ET in geographically diverse studies, including New York (US) and Madrid (Spain) as well as the Faroe Islands.10 Thus, larger epidemiological studies are needed to elucidate the environmental factors increasing ET risk, and such risk factors need to be taken into account (as a co-variate) when looking for genetic risk factors.
Do you see any translational value of the identification of genome-wide significant risk factors in ET, including the potential link to amyloid-β protein aggregation?
To be honest, I lack the imagination to draw translational consequences from the current genetic findings. We do not even agree whether there is a pathomorphological correlate of ET. The whole neurodegeneration hypothesis stands on feet of clay. Now there is finally hope that excellent pathologists will solve the problem. The likelihood that ET is just a network disease without any relevant (neuro)degeneration is high. The heterogeneity of the condition with opposite effects regarding such important outcomes like earlier or delayed mortality compared to controls may obscure our view on this.11
The five reported GWAS hits are interesting hits to follow up on, but it is too early to start translating. Some fine-mapping was performed, and the nomination of BACE2 as an ET risk gene is, of course, very interesting with what is already known about BACE2 and Alzheimer's disease. Further fine-mapping and functional validation are needed to confirm these genes as ET risk genes. Additionally, it would be very interesting to see if any of these five reported GWAS hits correlate with any particular ET phenotype (e.g., ET age at onset), which could provide valuable clues about the potential underlying biology of these GWAS hits.
The major recent breakthrough in the genetics of ET is the identification of genome-wide significant risk factors which required reaching the critical sample size for a GWAS through inclusion of about half a million of individuals. As pointed out by the interviewed experts, some caution is still needed when interpreting the results, replication and fine-mapping are required, and the results need to be robustly validated using other clinical and control cohorts. Further steps to improve our understanding of the genetic basis of ET include 1) deep phenotyping to identify subtypes and to generate more homogeneous patient groups, and 2) elucidating the non-genetic, environmental risk factors by epidemiological studies again to define more homogeneous groups in terms of risk factors. For sure, the genetic basis of ET is highly complex, and we are still far away from understanding the disease mechanisms to be able to translate the molecular findings to the development of therapies. An exciting future is awaiting us as the thousands of puzzle pieces are starting to be put together.
1. Louis ED, Ferreira JJ. How common is the most common adult movement disorder? Update on the worldwide prevalence of essential tremor. Mov Disord 2010;25:534-541.
2. Lorenz D, Frederiksen H, Moises H, Kopper F, Deuschl G, Christensen K. High concordance for essential tremor in monozygotic twins of old age. Neurology 2004;62:208-211.
3. Jimenez-Jimenez FJ, Alonso-Navarro H, Garcia-Martin E, Alvarez I, Pastor P, Agundez JAG. Genomic Markers for Essential Tremor. Pharmaceuticals (Basel) 2021;14.
4. Nalls MA, Blauwendraat C, Vallerga CL, et al. Identification of novel risk loci, causal insights, and heritable risk for Parkinson's disease: a meta-analysis of genome-wide association studies. Lancet Neurol 2019;18:1091-1102.
5. Siokas V, Aloizou AM, Tsouris Z, et al. Genetic Risk Factors for Essential Tremor: A Review. Tremor Other Hyperkinet Mov (N Y) 2020;10:4.
6. Liao C, Castonguay CE, Heilbron K, et al. Association of Essential Tremor With Novel Risk Loci: A Genome-Wide Association Study and Meta-analysis. JAMA Neurol 2022.
7. Stefansson H, Steinberg S, Petursson H, et al. Variant in the sequence of the LINGO1 gene confers risk of essential tremor. Nat Genet 2009;41:277-279.
8. Thier S, Lorenz D, Nothnagel M, et al. Polymorphisms in the glial glutamate transporter SLC1A2 are associated with essential tremor. Neurology 2012;79:243-248.
9. Ong YL, Deng X, Tan EK. Etiologic links between environmental and lifestyle factors and Essential tremor. Ann Clin Transl Neurol 2019;6:979-989.
10. Louis ED, Eliasen EH, Ferrer M, et al. Blood Harmane (1-Methyl-9H-Pyrido[3,4-b]indole) and Mercury in Essential Tremor: A Population-Based, Environmental Epidemiology Study in the Faroe Islands. Neuroepidemiology 2020;54:272-280.
11. Deuschl G, Petersen I, Lorenz D, Christensen K. Tremor in the elderly: Essential and aging-related tremor. Mov Disord 2015;30:1327-1334.