Dystonia is a kind of hyperkinetic movement disorder characterized by intermittent or sustained muscle contractions causing involuntary movements and/or abnormal postures in one or more parts of the body (1, 2). At present, the etiology and pathogenesis of dystonia remain largely unclear. The majority view is that dystonia is a neural network disorder, and specific gene variation may be one of the important causes underlying dystonia pathogenesis (3–6). With the rapid development of next-generation sequencing technology, more and more pathogenic genes have been discovered, providing perspectives for our in-depth understanding of dystonia development (7–10).

Recently, the collagen-type VI alpha-3 gene (COL6A3) has been reported to be associated with early-onset isolated dystonia (DYT)-27 (11). COL6A3 encodes the alpha-3 chain of type VI collagen, which is an important component of the extracellular matrix, involving the coordination of synaptogenesis and the stability of the synaptic networks (12). It was reported that compound heterozygous mutations in the COL6A3 gene may be responsible for (DYT)-27 syndrome (11, 13, 14). However, to the best of our knowledge, at present, only seven cases of compound heterozygous COL6A3 in dystonia were reported in Caucasian and Indian patients (11, 13, 14). Whether these compound heterozygous mutations or other variants in COL6A3 were present in Chinese dystonia patients have not been reported so far. Therefore, in this study, we analyzed the mutations of COL6A3 in 45 Chinese patients with isolated cervical dystonia by next-generation sequencing and tried to identify potential causal variants of COL6A3.

Mutations in the COL6A3 gene cause autosomal-recessive early-onset isolated dystonia, namely (DYT)-27, with interindividual heterogeneity of focal, segmental, or generalized distribution in the cranio-cervical region, upper limbs, and trunk (19). Recent evidence showed that the compound heterozygous mutations of COL6A3 may be disease-causing (11, 13, 14). Here, we examined genetic information on the COL6A3 gene in 45 Chinese patients with isolated cervical dystonia and found that 18 patients had seven potential causal variants in the COL6A3 gene. Importantly, among these variants, a novel compound heterozygous mutation of COL6A3 (c.1264G>A and c.8965+9G>A) was identified.

The COL6A3 gene encodes the collagen alpha-3 chain, which is one of the three subunits (COL6α1, COL6α2, and COL6α3) of collagen type VI, a microfibrillar component of the extracellular matrix (20, 21). Collagen VI dysfunctions are known to cause two main types of muscle disorders: Ullrich congenital muscular dystrophy and Bethlem myopathy (22, 23). Zech et al. first reported that COL6A3 mutation was associated with autosomal-recessive (DYT)-27 (11), in which they identified disease-segregating compound heterozygous mutations of COL6A3 in five cases affected by isolated dystonia from three unrelated German families, specifically, two siblings with c.9128G>A (p.Arg3043His) and c.9245C>G (p.Pro3082Arg), two siblings with c.7502G>A (p.Arg2501His) and c.8966-1G>C (p.Val2989_Lys3077delinsGlu), and one patient from the other family with c.7660G>A (p.Ala2554Thr) and c.8966-1G>C (p.Val2989_Lys3077delinsGlu). Intriguingly, they found that all affected individuals had at least one pathogenic allele in exon 41, promoting them to postulate that exon 41 may be a hot spot for mutations causing isolated dystonia (11).

In our study, by analyzing the COL6A3 gene by whole-exome sequencing in 45 Chinese patients with isolated cervical dystonia, we found four compound heterozygous mutations (Table 1) and further identified a novel deleterious mutation in patient S10 after harmful filtration, that is, c.1264G>A (p.Val422Met) in exon 4 and c.8965+9G>A in exon 40. This patient was a 47-year-old woman with a complete presentation of cervical dystonia phenotype, consistent with (DYT)-27 manifestation (19). Since in the family pedigree, the patient's father had passed away and her mother did not have either of the two variants while her daughter only had the mutation of c.1264G>A (Figure 1, Supplemental Figure 1), we supposed that c.1264G>A may be from the father and c.8965+9G>A may be de novo. Admittedly, for lacking the father's genetic information, we still did not rule out the possibility that other factors, such as environmental modifiers (e.g., perinatal adversities, drug abuse, infections, general anesthesia, or physical trauma) (6, 24) and other variants in different genes, could co-segregate with identified compound heterozygous mutations and contribute to dystonia pathogenesis. In addition, it is noted that both variants were not in exon 41 that was inconsistent with Zech et al.'s report. We supposed that different ethnicities between German and Chinese might be the responsible factor for the absence of exon 41 mutation in our case. At this point, Panda et al. also identified a pathogenic compound heterozygous mutation, not in exon 41 but in exon 10 [c.7557C>T (p.Gly1517Ser)] and exon 12 [c.4498G>A (p.Pro1894Leu)] of the COL6A3 gene in an Indian case with early-onset isolated dystonia (14). Moreover, although Lohmann et al. found that one German patient carried a compound heterozygous mutation, one in exon 41 [c.9245C>G (p.Pro3082Arg)] and the other in exon 6 [c.2195C>T (p.Thr732Met)] of the COL6A3 gene when examining 955 patients with isolated or combined dystonia or with another movement disorder with dystonic features, this patient was diagnosed as parkinsonism with dystonic posturing due to homozygous PINK1 mutations (13). In our study, both patients S14 and S21 had a missense VUS variant [c.9148G>A (p.Ala3050Thr)] in exon 41 (Table 2), but this variant was filtered as harmlessness (Table 3). Thus, we assume that mutations in exon 41 may be just one of the causes of isolated dystonia. Moreover, as a fact, only a few cases of compound heterozygous COL6A3 in dystonia were reported to date (11, 13, 14), and more new cases in the future should be reported to demonstrate the relationship between compound heterozygous COL6A3 and the occurrence of dystonia.

In addition to compound heterozygous mutations, we also identified other five missense variants in the COL6A3 gene, including c.958G>A (p.Ala320Thr), c.1478T>C (p.Val493Ala), c.1597C>T (p.Arg533Cys), c.1762G>A (p.Asp588Asn), and c.4912G>A (p.Ala1638Thr). It is especially noteworthy that patients S8, S23, and S36 carried c.4912G>A (p.Ala1638Thr) simultaneously (Table 3). In addition, several filtered VUS variants should deserve our attention as well (Table 2), though some of them were synonymous mutations, which, however, are reported as harmful as the non-synonymous mutations that alter proteins (25). For instance, c.4900+9C>T was simultaneously present in patients S33, S39, and S40; c.1065C>T (p.Ala355=) was present in patients S4 and S30; and c.237T>C (p.Ala79=) was present in patients S32 and S33. Therefore, these variants might also have the possibility of contributing to dystonia pathogenesis. Admittedly, whether these variants can really cause isolated dystonia needs further studies to be clarified.

Collagen VI represents a remarkable extracellular matrix molecule known for its roles in muscle and connective tissue. In addition to these, it also functions in the coordination of synaptogenesis and the stability of the synaptic networks (12). Zech M et al. revealed that COL6A3 was expressed in neurons, and the suppression of the exon 41 ortholog caused deficits in an axonal outgrowth without overt collagen defects probably because the exon 41 encodes part of the collagen VI α3 C4 domain (FN-III motif), which might be involved in the organization of structural plasticity (11). Apart from the C4 domain, the von Willebrand factor type-A (vWFA) domains have been shown to bind extracellular matrix proteins, cell-to-cell interaction, and other signaling pathways (14, 22). In Panda et al.'s study, both variants (Gly1517Ser and Pro1894Leu) were found to locate in the vWFA domain of the COL6α3 protein, which was supposed to be involved in other functions in the brain extracellular matrix such as neuronal organization, plasticity, and neuronal circuit formation (14). In our study, for the novel deleterious compound heterozygous mutation, the missense variant c.1264G>A (p.Val422Met) lies in exon 4, located in the vWFA domain of the COL6α3 protein, and the splicing variant c.8965+9G>A located in the downstream of exon 40, probably affecting the C4 domain just as variant c.8966-1G>C located in the upstream of exon 41 reported previously (11).

In conclusion, we identified a novel deleterious compound heterozygous mutation in the COL6A3 gene in Chinese patients with cervical dystonia. To the best of our knowledge, this study is the fourth report [Zech et al. (11); Lohmann et al. (13); Panda et al. (14)] on the compound heterozygous mutation of COL6A3 for the dystonia onset. Therefore, our findings may expand the spectrum of the COL6A3 genotype in the development of isolated dystonia.

The datasets presented in this study can be found in online repositories. The name of the repository and accession number can be found below: National Center for Biotechnology Information (NCBI) ClinVar, https://www.ncbi.nlm.nih.gov/clinvar, SCV002553216, SCV002553244 - SCV002553248, SCV002558734, SCV002558735, and SCV002586278 - SCV002586283.

The study was approved by the Medical Ethics Committee of the Xijing Hospital. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

MS and RW conceived the idea and drafted the manuscript. MS, RW, WD, and HZ contributed to the collection and interpretation of the data. All authors participated in the revision of the manuscript and figure and read and approved the final manuscript.