An Inversion Affecting the GCH1 Gene as a Novel Finding in Dopa‐Responsive Dystonia

Dopa-responsive dystonia (DRD) is most commonly caused by autosomal dominant GTP cyclohydrolase 1 deficiency,1 also known as Segawa disease (formerly DYT5a). Although more than 100 clinically relevant coding GCH1 variants have been identified, a number of DRD patients lack an identifiable disease-causing variant detectable with conventional gene sequencing.1 Some of these patients are found to have structural variants, such as deletions in the GCH1 gene.2, 3 Here we report the first known case of DRD due to a large inversion disrupting the GCH1 gene. A 57-year-old Han Chinese male developed abnormal head posturing and gait disturbance with diurnal fluctuation following a trivial head injury at age 10 years. He was otherwise well with normal developmental milestones. He underwent neck muscle denervation surgery aged 15 due to deterioration in head movements, which resulted in temporary improvement. Low-dose levodopa/benserazide 200/25 mg twice daily was commenced soon thereafter due to recurrence of involuntary head movements, with marked but incomplete improvement in symptoms. At presentation to our clinic aged 42, he had cervical dystonia characterized by anterocollis, left torticollis, right laterocollis, and side-to-side head tremor. He had associated moderate bilateral bradykinesia, hypophonia, mild slowing of gait, and mildly impaired postural righting reflexes. There was no evidence of limb dystonia, tremor, rigidity, or ataxia (Video 1). He remained responsive to the same low dose of levodopa after several decades with only mild progressive deterioration in gait over 15 years of follow-up. Prior attempts to wean levodopa resulted in reversible deterioration of head movements and gait. Botulinum toxin and trihexyphenidyl were trialed with modest effectiveness but were not tolerated. There were no affected family members, including two daughters aged 11 and 14 (Fig. S1). Magnetic resonance brain imaging was unremarkable. Blood test screening for Wilson's disease was negative. Single gene sequencing of GCH1 and TOR1A did not identify a causative variant. Multiplex ligation-dependent probe amplification (MLPA) was not performed. Further genetic studies were undertaken on a research basis with written informed consent and ethics approval (2019/PID14033). Short-read whole genome sequencing (WGS) was performed as described4 and using the ClinSV tool we detected a possible a ~5 Mb inversion on chromosome 14 disrupting GCH1 (Fig. S2). The inversion is part of a complex structural variant (NC_000014.8:g.[55343254_55346605del;55346606_60822142inv;60822143_60823119del]). No other clinically-relevant variants were detected in the DRD-associated genes GCH1, TH, SPR, or other known dystonia genes (see4). We were unable to confirm the breakpoints using Sanger sequencing, likely because repetitive sequences in the breakpoint region hindered design of specific sequencing primers. However, targeted long-read sequencing performed using Oxford Nanopore's "ReadUntil" function,5 focusing on the candidate inversion region, confirmed the breakpoints at both ends of the putative inversion (Fig. S3). Further confirmation of the inversion was attained through Bionano optical genome mapping (OGM)6 (Fig. 1A). Taken together, these findings confirmed the presence of a chromosomal inversion disrupting GCH1 by displacement of the promoter and first exon (Fig. 1B). The inversion is a component of a complex structural variant belonging to the Paired-Deletion Inversion (delINVdel) subclass.7 There are no repeat sequences implicated in recombination and so the molecular basis for generation of the variant remains unclear. To our knowledge, this is the first report of DRD due to an inversion affecting the GCH1 gene, which is unlikely to have been detected using traditional approaches to genetic testing. Despite the absence of a causative variant on initial standard GCH1 gene sequencing, clues to the diagnosis of DRD included young age at onset, diurnal fluctuations, and sustained response to low-dose levodopa without development of dyskinesia. Given the high degree of clinical suspicion of DRD, we pursued short-read WGS on a research basis, which enabled identification of a complex structural variant, a component of which is an inversion impacting GCH1. This finding was confirmed using two orthogonal approaches, namely long read nanopore sequencing and Bionano OGM. We note studies which have used other methods (eg, fluorescence in situ hybridization, large-insert whole-genome sequencing) to detect complex chromosomal rearrangements affecting GCH1,8 and also the use of long-read sequencing to resolve a complex structural variant in the PRKN gene.9 MLPA has been used to detect copy number variants (eg, duplications, deletions) in GCH1,10 but would not have utility for detecting copy number neutral variants such as the inversion observed in our study. For patients with a phenotype consistent with DRD in whom conventional sequencing is unrevealing, we recommend the use of WGS, nanopore long-read sequencing, and OGM as novel approaches to detect and confirm complex structural events affecting the GCH1 gene. Clinicians should be aware that unsolved cases of DRD may be due to structural variation, such as mega-base scale inversion, affecting GCH1. This report highlights the value of applying emerging genomic technologies to genetically unsolved cases to detect genetic variants that are missed using conventional approaches. This project was funded by the Paul Ainsworth Family Foundation to K.R.K. Optical genome mapping was funded by a Sydney Cancer Partners (Pilot Study Grant to V.M.H. and M.M.H.). K.R.K. receives funding from the Medical Research Future Fund (Grants 2023126, 2023357, and 2024888), unrelated to the current project. Open access publishing facilitated by University of New South Wales, as part of the Wiley - University of New South Wales agreement via the Council of Australian University Librarians. (1) Research project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript: A. Writing of the first draft, B. Review and Critique. S.E.: 1A, 1B, 1C, 3A, 3B A.F.: 1B, 1C, 3B M.H.: 1B, 1C, 3A, 3B J.C.: 1B, 1C, 3B I.D.: 1B, 1C, 3B I.S.: 1C, 3B M.S.: 1C, 3B D.Z.: 1C, 3B R.N.: 1C, 3B B.G.: 1C, 3B L.W.: 1C, 3B A.C.: 1B, 1C, 3B D.Y.: 1C, 3B L.I.R.: 1C, 3B M.M.H.: 1B, 1C, 3B V.M.H.: 1B, 1C, 3B M.K.: 1C, 3B K.R.K.: 1A, 1B, 1C, 3A, 3B M.H.: 1A, 1B, 1C, 3B Ethical compliance statement: Written informed consent was obtained from the patient. Ethics approval was obtained for undertaking genetic analysis on a research basis (2019/PID14033). We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines. Funding Sources and Conflict of Interest: The work presented here was funded by the Paul Ainsworth Family Foundation. The authors declare that there are no conflicts of interest relevant to this work. Financial Disclosures for the Previous 12 Months: Md Mehdi Hasan and Vanessa M. Hayes have a grant from Sydney Cancer Partners unrelated to this research. Kishore R. Kumar has speaking honoraria from Abbvie, Seqirus and Research Review Australia. Kishore R. Kumar has research grants from the Paul Ainsworth Family Foundation and Medical Research Future Fund (National Health and Medical Research Council). Kishore R. Kumar receives royalties from Oxford University Press. All other authors declare that there are no additional disclosures to report. Supplementary Figure S1. Pedigree of the family. Arrow indicates proband who underwent genetic studies, filled symbol indicates affected, squares represent males and circles represent females. d. MVA = died in motor vehicle accident. Additional family members were not available for testing. Supplementary Figure S2. Illumina short read whole genome sequencing data indicating a structural variant (NC_000014.8:g.[55343254_55346605del;55346606_60822142inv;60822143_60823119del]) on chromosome 14 affecting GCH1. Chimaeric ("split") reads identified by the ClinSV tool are visualized in the IGV genome browser. To ease interpretation the alignments of segments of two representative reads are highlighted (A00488:195:HGJN7DSX2:3:2336:15573:12743 in red, and A00488:195:HGJN7DSX2:3:2160:31503:1219 in blue). Other tracks show the deletions which flank the inversion, and the exonic structure of GCH1 transcript NM_000161.3. (A) The left-hand (centromere proximal) breakpoint region, associated with a 3.4 kb deletion. (B) The right-hand (centromere distal) breakpoint region, associated with a smaller (1.0 kb) deletion. Supplementary Figure S3. Oxford Nanopore long read sequencing (LRS) data supporting the proposed structural variant. Chimaeric nanopore sequences are visualized in the IGV genome browser. The sequence alignments confirm the breakpoints indicated by short read sequencing analysis, and extend wide enough for a high level of confidence in read locations. (A, B) Inversion breakpoint regions as in Fig. S2, but in a 50 kb window. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.