Neurogenetics Original research Loss of seryl-­tRNA synthetase (SARS1) causes complex spastic paraplegia and cellular senescence Edgard Verdura ‍ ‍,1,2 Bruno Senger,3 Miquel Raspall-­Chaure,4,5 Agatha Schlüter,1,2 Nathalie Launay,1,2 Montserrat Ruiz,1,2 Carlos Casasnovas,1,2,6 Agustí Rodriguez-­Palmero,1,7 Alfons Macaya,4,5,8 Hubert Dominique Becker,3 Aurora Pujol ‍ ‍1,2,9 ► Additional supplemental ABSTRACT material is published online Background Aminoacyl-­tRNA synthetases (ARS) are WHAT IS ALREADY KNOWN ON THIS TOPIC only. To view, please visit ⇒ Patients affected with microcephaly, intellectual the journal online (http://​dx.​ key enzymes catalysing the first reactions in protein doi.​org/​10.​1136/​jmg-​2022-​ synthesis, with increasingly recognised pleiotropic roles disability and ataxia harbouring biallelic 108529). in tumourgenesis, angiogenesis, immune response and variants in the seryl-­tRNA synthetase 1 (SARS1) lifespan. Germline mutations in several ARS genes have were reported. For numbered affiliations see ⇒ Up to now, recessive variants in aminoacyl-­ end of article. been associated with both recessive and dominant neurological diseases. Recently, patients affected tRNA synthetases (ARS) genes have been with microcephaly, intellectual disability and ataxia predominantly associated with early onset Correspondence to Dr Aurora Pujol, Neurometabolic harbouring biallelic variants in the seryl-­tRNA synthetase diseases impacting the central nervous system, Disease Lab, IDIBELL, Barcelona, encoded by seryl-­tRNA synthetase 1 (SARS1) were compared with dominant variants. Catalonia, Spain; reported. WHAT THIS STUDY ADDS ​apujol@​idibell.​cat Methods We used exome sequencing to identify ⇒ We describe a novel clinical phenotype and EV and BS contributed equally. the causal variant in a patient affected by complex mode of inheritance (de novo dominant) for a spastic paraplegia with ataxia, intellectual disability, Received 1 March 2022 variant in SARS1 gene, and unveil the role of developmental delay and seizures, but without Accepted 25 July 2022 SARS1 as a regulator of cell growth, division Published Online First 30 August microcephaly. Complementation and serylation assays and senescence. 2022 using patient’s fibroblasts and an Saccharomyces cerevisiae model were performed to examine this HOW THIS STUDY MIGHT AFFECT RESEARCH, variant’s pathogenicity. PRACTICE OR POLICY Results A de novo splice site deletion in SARS1 was ⇒ We underscore the need for caution interpreting identified in our patient, resulting in a 5-­amino acid variants in ARS enzyme genes according to in-­frame insertion near its active site. Complementation known inheritance modes, with important assays in S. cerevisiae and serylation assays in both implications for genetic diagnosis and yeast strains and patient fibroblasts proved a loss-­of-­ counselling. function, dominant negative effect. Fibroblasts showed ⇒ Furthermore, we reveal pivotal roles of this an abnormal cell shape, arrested division and increased enzyme in the control of cell growth and beta-­galactosidase staining along with a senescence-­ senescence, underlying the pathogenetic associated secretory phenotype (raised interleukin-­6, mechanisms of these diseases, and open the p21, p16 and p53 levels). way to further research. Conclusion We refine the phenotypic spectrum and modes of inheritance of a newly described, ultrarare neurodevelopmental disorder, while unveiling the role of SARS1 as a regulator of cell growth, division and transferred from AMP to the 3’-end of its cognate senescence. tRNA, producing AA-­tRNA and AMP.2 Mutations in ARS enzymes have been implicated in both recessive and dominant diseases affecting a broad range of tissues.3 While recessive inheritance INTRODUCTION is very frequently associated with severe, early Aminoacyl-­tRNA synthetases (ARS) are ubiq- onset diseases which impact the central nervous uitously expressed and highly evolutionarily system and other organs, dominant mutations cause conserved enzymes responsible for esterification specifically late-­onset peripheral neuropathies in a © Author(s) (or their employer(s)) 2022. Re-­use of amino acids (AA) to their cognate transfer RNA restricted list of cytoplasmic ARS genes.4 It is still permitted under CC BY. (tRNA), an essential reaction in decoding mRNA unclear why mutations in these ubiquitous enzymes Published by BMJ. into protein.1 ARS catalyse a two-­ step reaction cause such a highly specific phenotype. Seryl-­tRNA To cite: Verdura E, Senger B, in which the given amino acid is condensed with synthetase 1 (SARS1) is a cytoplasmic ARS enzyme Raspall-­Chaure M, et al. ATP to form tightly bound aminoacyl adenylate that charges serine onto its cognate tRNASer but J Med Genet (AA-­AMP), with the simultaneous release of pyro- also onto the non-­cognate selenocysteine tRNASec. 2022;59:1227–1233. phosphate PPi. The activated amino acid is then Subsequently, serine-­ bound tRNASec undergoes Verdura E, et al. J Med Genet 2022;59:1227–1233. doi:10.1136/jmedgenet-2022-108529 1227 Neurogenetics tRNA conversion of the bound serine to selenocysteine by a Three-dimensional modelling four-­step reaction.5 The domain structure of the SARS1 gene was retrieved from SARS1 is essential for vascular development as shown in a the UniProt database and NCBI. The three-­dimensional (3D) zebrafish model,6 and also in a rapid network formation assay structure of wild-­type (WT) SARS1 was retrieved from https:// of human umbilical vein endothelial cells.7 This non-­canonical www.rcsb.org/, and a model of the mutated SARS1 was obtained function is linked to the UNE-­S domain, a unique domain which using the SWISS-­ MODEL utility on the ExPASy webpage has only been found in the C-­terminus of vertebrate SARS1. (https://swissmodel. expasy.org). Structures were visualised with UNE-­S harbours a nuclear localisation signal and can divert a PyMOL. fraction of the SARS1 protein into the nucleus to control the expression of vascular endothelial growth factor A.8 Fibroblast cultures Two recent articles reported homozygous loss-­ of-­ function Patient dermal fibroblasts were generated from a skin biopsy missense mutations in SARS1 in two independent families exhib- using standard cell culture procedures. Fibroblasts from the iting a neurodevelopmental syndrome including microcephaly, patient and age-­matched and sex-­matched controls were grown intellectual disability, seizures and ataxia,9 together with other in DMEM (Dulbecco's Modified Eagle's Medium)(Ref: 31885-­ anomalies such as cardiomyopathy, deafness and decompen- 023) containing 1% penicillin/streptomycin and 10% fetal sation during fever.10 In this work, we identified a de novo bovine serum (FBS). Cultures were maintained at 37°C under inherited novel pathogenic variant of SARS1 causing a complex 5% CO2 and 95% humidity. spastic paraplegia and ataxia phenotype, thus confirming the recent implication of mutations in this gene in neurological and RT-PCR and qRT-PCR movement disorders.9 10 Furthermore, this variant illustrates the RNA was extracted from patient and control fibroblasts using concept that dominant negative mutations in ARS genes can give the RNeasy Mini Kit (QIAGEN). Total RNA quantity and rise to phenotypes similar to those caused by recessive muta- quality were assessed by 0.8% agarose gel electrophoresis and tions, rather than the neuropathies frequently linked to domi- NanoDrop spectrophotometry. cDNA was synthesised using the nantly inherited variants. SuperScript IV kit (Life Technologies) following the manufac- turer’s instructions. cDNA was amplified by PCR using primers MATERIALS AND METHODS designed with Primer3 using the whole coding sequence of Patients and samples SARS1 (NM_001330669) (forward: ​GGAA​CAGG​CTCT​CATC​ The patient was recruited and examined at Vall d’Hebron CAGTA; reverse: A ​ GGAGACCAACTCACGGAAG). Products Hospital, Barcelona. Detailed neurological examination was were separated by agarose gel electrophoresis (3%), purified performed. Blood samples and skin-­derived fibroblast cell lines (QIAquick Gel Extraction Kit, QIAGEN) and sequenced by were obtained using standard methods. DNA was extracted the Sanger method using standard methodology. cDNA PCR using the Gentra Puregene Blood Kit (QIAGEN). products were cloned into pCMV-­Tag2 plasmid using standard methods, and several colonies were extracted for DNA and sequenced to isolate WT and mutated sequences. qRT-­ PCR Whole-exome sequencing analysis was performed in a LightCycler 480 device using SYBR Green Genomic DNA was extracted from peripheral blood using stan- I Master Mix (Roche). All reactions were performed in tripli- dard methods. Whole-­exome sequencing (WES) was performed cate. Each gene mRNA level was normalised to that of H36B4 on patient’s DNA sample using the SureSelect XT Human using the 2–∆∆Ct method. A two-­sided Student’s t-­test was used All Exon V5 50 Mb kit (Agilent) for DNA capture. WES was to compare the expression ratios of patients and controls. Data performed using the HiSeq 2000 Platform (Illumina) at Centre are presented as the mean±SD. All primers are shown in online Nacional d’Anàlisi Genòmica (Barcelona). Variants were filtered supplemental table 1. using an in-­house pipeline based on GATK best practice guide- lines and prioritised based on accurate patient characterisation Immunofluorescence with Human Phenotype Ontology (HPO) terms, interaction Fibroblasts from the index patient and three age-­matched and networks at physical and functional levels and variant intoler- sex-­matched controls were seeded on coverslips and cultured in ance scores generated by the Exome Aggregation Consortium a 6-­well plate (200 000 per well) at 37°C for 24 hours. Cover- (ExAC) and gnomAD consortia. We prioritised non-­synonymous slips were fixed with 4% paraformaldehyde, blocked and incu- coding variants that had a frequency lower than 0.01 in the bated overnight with mouse anti-α-tubulin (dilution 1/10 000, ExAC, 1000 Genomes and gnomAD control databases. Candi- Ref: ab80779, Abcam) and rabbit anti-­SARS1 (dilution 1/10, date variants were validated and tested for cosegregation in all Ref: ab183025, Abcam) primary antibodies. Slides were incu- family members by Sanger sequencing. bated with goat antirabbit IgG Alexa Fluor 488 (dilution 1/1000, Ref: A-­21428, Dako) and antimouse IgG Alexa Fluor 647 (dilu- Evolutionary conservation of SARS1 tion 1/1000, Ref: A-­21236, Dako) secondary antibodies. DAPI SARS1 protein orthologues from multiple species were derived (4′,6-­diamidino-­2-­phenylindole) was used to stain cell nuclei. from the following GenBank accessions: XP_009425317.1 Confocal microscopy images were acquired with a Leica TCS SL (Pan troglodytes), NP_001191908.1 (Mus musculus), laser scanning confocal spectral microscope using a 63× objec- NP_001026563.1 (Gallus gallus), NP_998852.1 (Xenopus trop- tive. Images were treated using FIJI software. icalis), NP_001003882.1 (Danio rerio), NP_501804.1 (Caenor- habditis elegans), NP_309005.1 (Escherichia coli), NP_608743.2 Western blot analysis (Drosophila melanogaster) and NP_010306.1 (Saccharomyces For western blot analyses, fibroblasts were homogenised in RIPA cerevisiae). Calculation of amino acid sequence identity with (Radioimmunoprecipitation assay) buffer (150 mM NaCl, 1% Homo sapiens and alignment were performed using Clustal (w/v) Nonidet P40, 0.5% (v/v) sodium deoxycholate, 0.1% Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/). sodium dodecyl-­ sulfate (SDS) (w/v), 50 mM Tris-­ HCl, pH 1228 Verdura E, et al. J Med Genet 2022;59:1227–1233. doi:10.1136/jmedgenet-2022-108529 Neurogenetics 8.0), sonicated for 2 min at 4°C, centrifuged, mixed with 4× In vitro analysis of SARS1 activity NuPAGE LDS Sample Buffer (Invitrogen) and heated at 70°C To analyse hSARS serylation activity, we prepared S100 extracts for 10 min. Protein samples (25 μg) were subjected to poly- from S. cerevisiae yeast and fibroblasts (see S. cerevisiae yeast acrylamide gel electrophoresis at 120 V in NuPAGE MOPS SDS S100 preparation). These crude enzymatic preparations were Running Buffer (Invitrogen) supplemented with 5 mM sodium then used in tRNA aminoacylation assays with 14C-­radiolabelled bisulfite (Ref: 243973, Sigma-­ Aldrich). Proteins were trans- serine (Serine, L-[3-­ 14C] PerkinElmer Ref NEC827050UC, ferred to nitrocellulose membranes using an iBlot 2 Gel Transfer Lot 2361822 52.4 mCi/mmol ; 50 µCi/mL). All reactions were Device (Invitrogen). After blocking in 5% bovine serum albumin carried out at 30°C for yeast extracts and 37°C for fibroblast (BSA, Sigma-­Aldrich) and 0.05% TBS-­Tween for 1 hour at room extracts under the following conditions: 50 mM HEPES (4-­(2-­h temperature, membranes were incubated with primary antibodies ydroxyethyl)-­1-­piperazineethanesulfonic acid), KOH buffer pH overnight at 4°C. The primary antibodies used included mouse 7.5, 20 mM KCl, 15 mM MgCl2, 4 mM DTT (Dithiothreitol), anti-β-actin (dilution 1/10 000, Ref: A2228, Sigma-­ Aldrich); 5 mM ATP, 0.1 mg/mL BSA, 440 µM total yeast tRNA and 74 rabbit H2AX (Chomatin Immunoprecipitation grade, dilution µM of [14C] serine. tRNA concentration was 22 µM. Data were 1/1000, Ref: ab11175, Abcam); mouse γ-H2AX S139 (dilution: corrected according to the extract’s concentration. Ten micro- 1/1000, Ref: ab22551, Abcam), rabbit anti-­ SARS1 (dilution litre aliquots from a 62 µL reaction mixture were spotted onto 1/1000, Ref: ab183025, Abcam) and γ-tubulin (dilution 1/10 Whatman paper and quenched with 5% trichloroacetic acid 000, Ref: T6557, Sigma). After incubation with secondary anti- (TCA) at various time intervals. After three TCA washes, papers bodies for 1 hour at room temperature, proteins were detected were washed three times with ethanol and dried. Radioactivity with a ChemiDoc Touch Imaging System (Bio-­Rad). was measured in a Beckman LS 6500 scintillation counter with a toluene-­based scintillant. The reaction times were 2, 4, 8, 16, 24, 32 min for yeast extracts and 2, 8, 60, 180, 360 min for β-Galactosidase staining fibroblasts extracts. Experiments were performed five times Fibroblasts from the index patient and two controls were seeded independently on coverslips and cultured in a 6-­well plate (200 000 per well) at 37°C for 24 hours. Staining was performed following the S. cerevisiae yeast S100 preparation manufacturer’s instructions (Senescence Cells Histochemical Cells were grown in 500 mL SDC-­leu-­Ura on a rotary shaker Staining Kit, Sigma). Photos were taken using a Nikon Eclipse at 30°C and harvested when OD600 nm=1 by centrifugation at 80i microscope, 10× objective and NIS-­Elements BR software. 5000× g for 5 min at 4°C. The pellet was resuspended in 5 mL We counted a minimum of 50 cells per individual using the FIJI of lysis buffer containing 50 mM Na-­HEPES pH 7, 30 mM KCl, Cell Counter Tool. Experiments were performed twice. A two-­ 10% glycerol, 0.1 mM EDTA, 5 mM β-mercaptoethanol and sided Student’s t-­test was used to compare the proportion of protease inhibitors tablet (Roche). One volume of glass beads (Ø β-galactosidase (β-gal)-­positive cells (%) between the patients 0.25–0.5 mm, Roth) was added and cell lysis was performed with and controls. Data are presented as the mean±SD. a FastPrep-­24 apparatus (6×1 min at 6.5 m/s, with 1 min on ice between each cycle). Cells debris were removed by centrifugation at 500× g for 10 min at 4°C and the supernatant is centrifuged Construction, culture and analysis of S. cerevisiae yeast at 100 000× g for 1 hour. The resulting soluble fraction (S100) strains is recovered, then dialysed O/N against the storage buffer (same The SES1 open reading frame (ORF) was replaced by either as for fibroblasts extracts). Strains used were hSARS, hSARS+h- human WT SARS1 (plasmid pRS316-­ hSARS) or the mutant SARSmut or WT (see online supplemental table 2). version of hSARS (plasmid pRS316-­hSARSm) using isothermal assembly.11 Each plasmid was transformed into the Y23962 yeast strain (EUROSCARF collection), which is heterozygous diploid Fibroblast S100 preparation for SES1 and has the following genotypes: MATa/MATα; ura3Δ0/ Fibroblasts were grown in DMEM media supplemented with ura3Δ0; leu2Δ0/leu2Δ0; his3Δ1/his3Δ1; met15Δ0/MET15; LYS2/ FBS and antibiotics (streptomycin and penicillin) and harvested lys2Δ0; YDR023w/YDR023w::kanMX4. at 80% confluence and issued from three 75 cm2 flask. They Transformed Y23962 strains were grown on SDC media were lysed in the following buffer : 30 mM Na-­HEPES pH 7.9, lacking uracil (SDC-­Ura) (6.75 g/L yeast nitrogen base without 10 mM KCl, 1.5 mM MgCl2, 5 mM β-mercaptoethanol and amino acids, 20 g/L glucose, 0.55 g complete supplemental protease inhibitors tablet (Roche). Lysis was completed in an media and 20 g/L agar) for 12 hours and transferred to sporula- appropriate douncer with 12 strokes. Cell debris were removed tion media (10 g/L yeast extract, 20 g/L peptone, 10 g/L potas- by centrifugation at 4°C at 8 000× g. The supernatant was then centrifuged at 100 000× g for 1 hour. The extracts wad then sium acetate pH 6.5, adjusted with 6 N HCl and 20 g/L agar) for dialysed against 50 mM Na-­HEPES pH 7.5, 30 mM KCl, 50% 3–5 days. On sporulation, tetrads were dissected on complete glycerol, 0.1 mM EDTA and kept at −20°C. Experiments were YPD(Yeast extract Peptone Dextrose) solid media (10 g/L yeast performed five times independently. extract, 20 g/L peptone, 20 g/L glucose and 20 g/L agar), and viable spores were analysed using a Singer dissection micro- scope on YPD plates. Individual spores were then restreaked Data availability on (i) SDC-­Ura to test for the presence of the pRS316 plasmid The authors confirm that the data supporting the findings of this and (ii) YPD+G418 (500 µg/mL) to test for the presence of the study are available within the article and its online supplemental disrupted genomic copy of SES1. Drop tests were performed by materials. growing yeast strains in SDC-­Ura-­Leu liquid media to an optical density (OD)600 nm of 1. Serial dilutions were performed in water, RESULTS and 6 µL of each dilution was spotted on an SDC-­Ura-­Leu plate Clinical description and genetic findings and incubated at 30°C. Pictures were taken at various times of The patient is an early adolescent Caucasian male with no relevant incubation. family history, born at term after an uneventful pregnancy and Verdura E, et al. J Med Genet 2022;59:1227–1233. doi:10.1136/jmedgenet-2022-108529 1229 Neurogenetics with a normal neonatal period. He started having focal seizures as a toddler, frequently precipitated by fever, which lasted up to 10 min and on some occasions were followed by brief postictal paresis. Seizures did not respond to phenobarbitone or valproate monotherapy but were eventually controlled with combination of valproate and lamotrigine at middle childhood. He remained seizure-­free despite AED withdrawal at late childhood. Electro- encephalograms were normal or showed infrequent multifocal spikes. Serial brain MRI at the beginning of adolescence showed subtle and non-­progressive punctiform frontal subcortical hyper- intensities. MRS was normal. Funduscopy and electroretino- gram were normal, but visual evoked potentials showed bilateral increased latencies with normal amplitudes. He had a global developmental delay, with late acquisition of independent walking at early childhood, motor clumsiness and delayed speech, being able to produce only a few bisyllables during early childhood. Early clumsiness evolved to overt signs of spastic paraparesis that worsened slowly during middle child- Figure 1 Seryl-­tRNA synthetase 1 (SARS1) de novo variant features. hood but subsequently remained stable. He underwent surgery (A) Family tree. Square: male, circle: female, black symbol: affected for strabismus just before early adolescence. The most recent individual, white symbols: unaffected individuals, WT: wild-­type allele. exam showed prominent and symmetrical signs of spasticity in (B) Representation of variant c.969_969+2delGGT impacting on SARS1 the lower limbs, with hyper-­reflexia, bilateral ankle clonus and splicing. Disruption of exon 7 donor site leads to the use of an alternative right Babinski sign, and a wide-­base unsteady spastic gait. He AG donor site located 17–18 bases downstream. This results into the showed no dysmorphic traits, and head growth was normal. inclusion of 16 intronic bp into the coding sequence, and consequently Relevant complementary exams included repeatedly normal in the in-­frame insertion of 5 amino acids. (C) Primary and secondary plasma and cerebrospinal fluid amino acid profiles (serine: structure elements of SARS1. Both sequences are shown starting at F(Phe) 125 μM, normal value 89–165; glycine: 236 μM, normal value 316. WT residues: black; inserted residues: blue. Underlined residues 147–299; cysteine: 28 μM, normal value 24–54). correspond to those being part of the conserved motif 2, present in class II WES in our patient, followed by variant prioritisation using an aminoacyl-­tRNA synthetases. The two residues in red correspond to F321 in-­house bioinformatics pipeline that uses interactome networks and E325 that are involved in ATP and serine recognition, respectively. based on the clinical presentation (in standard HPO terms),12 Below each sequence (letters in grey), secondary structure elements are revealed a candidate heterozygous deletion variant in SARS1. shown (b=β strand, h=α-helix and -=loop region). (D) Three-­dimensional The variant (chr1:109778053_109778055delGGT, hg19) causes model (based on 4L87, modelled with PyMol) showing the ATP/AMP-­Ser the ablation of a canonical splice site (c.969_969+2delGGT) in binding region of SARS1. Cyan: WT model; dark blue: motif 2; red: mutated the boundary of exon 7 and intron 7. This variant was prior- region (insertion); pink: AMP-­serine analogue. (E) Western blot analysis itised due to: (i) its absence from gnomAD13 and other data- showing SARS1 and γ-tubulin protein levels in fibroblasts from the patient bases of control individuals (ExAC, 1000 Genomes); (ii) being and age-­matched controls (n=3). γ-Tubulin was used as a loading control. a canonical splice site variant in a predicted loss-­ of-­ function (F) SARS1 serylation assays in control and patient-­derived fibroblast S100 intolerant gene and (iii) being located in a gene (SARS1) previ- extracts. tRNA concentration was 22 µM. Data were corrected according ously associated with similar clinical presentation as our patient to the extract’s concentration. Experiments were performed five times (OMIM #617709). After cosegregation analysis in the parents, independently. this variant was found to be de novo. Patient fibroblast cDNA sequencing showed that this change causes the inclusion of 16 intronic bp into the cDNA, which results in the in-­frame inser- domain as well as structural ordering of insertions I and II. tion of 5 amino acids (p.Lys323_Ile324insSerArgTrpValArg) These conformational changes are necessary to correctly (figure 1A,B, online supplemental figure 1). execute the aminoacylation activity and are likely impaired by this mutation.14–16 We next assessed the effects of the variant on protein Structural and functional characterisation stability by western blot analysis, using a polyclonal antibody The deletion is located in evolutionarily conserved motif 2 against human SARS1 (Ref: ab183025, Abcam). Results indi- (residues 291–329) of the SARS1 aminoacylation domain cate that the insertion did not significantly change SARS1 (online supplemental figure 2). A 3D model analysis showed levels in the patient’s fibroblasts (figure 1E). that the inserted residues perturbed the secondary struc- To assess SARS1 aminoacylation activity, we performed ture of SARS1 by disrupting an important β-strand and serylation enzymatic assays using the patient’s fibroblasts displacing the side chain of Phe321 (figure 1C,D). Residue compared with sex-­ matched and age-­ matched controls Phe321 is predicted to be important for the recognition of (n=3). In brief, we measured radioactivity in fibroblast S100 the adenosine ring (A ring) of the AMP molecule through extracts of the index case and controls at different time points stacking interactions,14 15 suggesting that this variant may during a 6-­hour incubation with 14C-­radiolabelled L-­serine. affect the overall aminoacylation activity of SARS1. Further- Five independent assays performed in duplicate were carried more, L-­serine binding to residue 325 could also be compro- out. Mutant fibroblasts showed reduced (~30%) aminoacy- mised.14 15 lation activity (figure 1F). After ACMG/AMP criteria eval- In higher eukaryotes, binding of the aminoacylation reac- uation,17 and taking into account both de novo segregation tion intermediate analogue (Ser-­SA) in the active site induces (PS2) and enzymatic analysis (PS3), this variant is classified conformational change and tightening of the tRNA binding as pathogenic. 1230 Verdura E, et al. J Med Genet 2022;59:1227–1233. doi:10.1136/jmedgenet-2022-108529 Neurogenetics Figure 2 Functional testing using Saccharomyces cerevisiae complementation/serylation assays confirm seryl-­tRNA synthetase 1 (SARS1) mutation’s deleterious effects. (A) Growth analysis of S. cerevisiae strains by drop test analysis on SDC-­Ura-­Leu media. Yeast strain detailed genotypes are detailed in online supplemental table 2. Doubling time of the various strains was determined in liquid SDC-­Ura-­Leu media in three independent experiments. The results are depicted on the right side. (B) SARS1 serylation assays in S. cerevisiae S100 extracts. tRNA concentration was 22 µM. Data were corrected according to the extract’s concentration. Experiments were performed five times independently. Figure 3 Patient-­derived fibroblasts show senescence-­related features. Analysis of hSARS and its mutant in S. cerevisiae yeast (A) Proliferation measurements. Number of culture days needed for 1 revealed dominant-negative effects million fibroblasts to reach confluency in 150 mm plates. Patient fibroblasts To further study the effect of the mutation, we performed proliferated at a very much slower rate than controls (n=3). Measures complementation studies in the yeast strain Schizosaccharo- were taken thrice. (B) Immunofluorescence images showing seryl-­tRNA myces pombe, which harbours a SARS1 homologue (seryl-­tRNA synthetase 1 (SARS1) localisation (green) in patient (passage 13) and synthetase SES1, YDR023w). This is an essential gene, thus a control fibroblasts (n=4, passages 12–15). Blue: DAPI. Red: gamma-­ plasmid copy of the SES1 must be provided on deletion to sustain tubulin. Patient fibroblasts’ nuclei are structurally affected and micronuclei growth of the haploid strain. We thus constructed a centromeric can be observed (white arrow). Patient fibroblasts also show cell division yeast plasmid (pRS316-­ySARS) carrying the URA3 auxotrophic abnormalities. (C) Western blot analysis showing H2AX and phosphorylated marker and the SES1 gene, including ORF and promoter regions. H2AX (γ-H2AX) protein levels in fibroblasts from the patient (passage Subsequently, the SES1 ORF was replaced by either human WT 13) and age-­matched controls. γ-Tubulin was used as a loading control. SARS1 cDNA (plasmid pRS316-­hSARS) or the mutant version (D) β-Galactosidase (β-gal) staining in patient (passage 13) and control of hSARS (plasmid pRS316-­hSARSm) (figure 2A). fibroblasts (n=4, passages 12–15). Positive β-gal cells (in blue) are Spore analysis showed that substituting SES1 for WT human highly increased in the patient. Left: representative microscopy images. SARS (hSARS) allowed the generation of viable strains, with Right: quantification of β-gal-­positive cells (%). (E) qRT-­PCR results show similar growth capacities figure 2B, indicating that human SARS1 upregulation of senescence-­associated secretory phenotype (SASP) genes can compensate for SES1 absence. In contrast, the mutant plasmid in patient’s fibroblasts compared with controls (n=5). SARS1 and SARS2 carrying the insertion identified in our patient (hSARSm), did genes were also quantified. All experiments were performed three times, not allow the generation of viable haploid strains, meaning that independently. The values are represented as the means+SD, and Student’s mutant hSARS1m is not functional in yeast (online supplemental t-­test was performed (*p<0.05; **p<0.01; ***p<0.001). figure 3). However, when the hSARS strain was cotransformed with a plasmid expressing hSARSm (emulating heterozygosis), an important growth defect was observed, strongly suggesting a revealed increased senescence in patient’s fibroblasts compared dominant negative effect of hSARSm over the WT SARS1 form with controls (figure 3D). We thus set out to investigate the main (figure 2A). molecular effectors of cellular senescence and DNA damage, To assess hSARS aminoacylation activity, we performed such as phosphorylated histone H2AX (γ-H2AX).18 Indeed, similar serylation assays as those performed in fibroblasts, using western blot analysis showed an increase of γ-H2AX levels in yeast extracts transfected with hSARS, or cotransfected with patient’s fibroblasts compared with controls (figure 3C). Finally, hSARS+hSARSm. S100 extracts were incubated with 14C-­radio- the mRNA levels of senescence-­associated secretory phenotype labelled L-­serine for 30 min. Interestingly, the hSARS+hSARSm (SASP) genes such as interleukin-­ 6, p21, p16 and p53 were strain showed a 30% lower activity than hSARS alone, consistent markedly increased in the patient’s fibroblasts, confirming a with previous results in fibroblasts, and again suggesting a domi- senescent phenotype. SARS1 and SARS2 mRNA levels did not nant negative effect of hSARSm over hSARS (figure 2B). vary compared with controls (figure 3E). Loss of SARS1 activity causes cellular senescence DISCUSSION Further functional analysis indicated that the patient’s fibroblasts In this study, WES identified for the first time a de novo variant proliferated very slowly, reaching confluence at least four times in SARS1, challenging the described recessive mode of inheri- slower than controls (figure 3A). Mutant fibroblasts showed tance, in a patient affected with complex spastic paraparesis abnormal shape, being more enlarged and rounded than controls, with ataxia, seizures and intellectual disability. Two recent arti- with nuclear structural alterations such as micronuclei and cles reported homozygous missense mutations in SARS1 also in abnormal mitosis, suggesting an underlying cell division defect close proximity to the active site (c.514G>A, p.Asp172Asn and (figure 3B). The distribution of WT and mutant SARS1 was c.638G>T, p.Arg213Leu), in two families affected by a neurode- consistent with a previously reported, mainly cytosolic pattern velopmental syndrome including microcephaly, ataxia, seizures, with some nuclear localisation.8 8 Moreover, β-gal staining moderate intellectual disability and other anomalies such as Verdura E, et al. J Med Genet 2022;59:1227–1233. doi:10.1136/jmedgenet-2022-108529 1231 Neurogenetics cardiomyopathy, deafness and decompensation during fever.9 10 Author affiliations 1 Alterations in the central nervous system, such as MRI abnor- Neurometabolic Diseases Laboratory, Bellvitge Biomedical Research Institute (IDIBELL), L’Hospitalet de Llobregat, 08908, Barcelona, Catalonia, Spain malities including leucoencephalopathy, ataxia and seizures, 2 Centre for Biomedical Research on Rare Diseases (CIBERER), Instituto de Salud have been reported in both cytoplasmic and mitochondrial ARS-­ Carlos III, Madrid, Spain related diseases.3 4 19 20 3 Université de Strasbourg 1, Strasbourg, France 4 In this report, we showed that (i) SARS1 activity was signifi- Pediatric Neurology Research Group, Vall d’Hebron University Hospital, Universitat cantly decreased both in patient’s fibroblasts and in the ‘hetero- Autònoma de Barcelona, 08035, Barcelona, Catalonia, Spain 5 zygous’ yeast strain; (ii) mutated SARS1 protein did not substitute Department of Paediatric Neurology, Vall d’Hebron University Hospital, 08035, Barcelona, Catalonia, Spain SES1 activity in yeast and (iii) cell growth was impaired both in 6 Hospital Universitari de Bellvitge, Barcelona, Spain patient fibroblasts, and in the yeast strain expressing simultane- 7 Pediatrics, Hospital Germans Trias i Pujol, Barcelona, Spain ously WT and mutated SARS1 forms, emulating our heterozy- 8 Institut de Neurociències, Universitat Autònoma de Barcelona, 08193, Barcelona, gous patient. These results, together with the presence of five Catalonia, Spain 9 healthy SARS1 loss-­of-­function variant carriers in the gnomAD Catalan Institution of Research and Advanced Studies (ICREA), 08010, Barcelona, Catalonia, Spain database, suggest that this mutation operates through a domi- nant negative effect, arguing against a haploinsufficiency mech- Acknowledgements We thank Cristina Guilera and Juanjo Martínez for excellent anism. Interestingly, a very recent work described a recurrent de technical support, Noelia Carrión and Laura Argelich for their assistance with novo pathogenic variant in NARS1, which produced a protein fibroblasts experiments and Dr Stéphane Fourcade for support and discussion. lacking the ATP-­binding domain.21 This observation, in addi- Contributors EV, BS and AP designed and conceptualised the study. EV, BS, MR-­C, tion to our work, strongly suggests that mutations impacting AS, NL, MR, CC, AR-­P, AM, HB performed analysis and interpreted the data. EV, BS, the ATP-­binding domain function can yield dominant-­negative MR-­C and AP drafted the manuscript. AP and CC acquired the funding. AP acts as mutation effects. guarantor and accepts full responsibility for the work and/or the conduct of the In addition, functional characterisation of the patient’s study, had access to the data, and controlled the decision to publish. All authors critically revised the manuscript, contributed significantly to this work and declared primary fibroblasts indicated severe growth defects and cellular to meet the ICMJE criteria for authorship. senescence, in combination with a SASP signature. It is intriguing Funding This study was supported by grants from the Undiagnosed Rare that the opposite mechanism, an overexpression of SARS1 in a Disease programme URD-­Cat (SLT002/16/00174) from the Generalitat de transformed BG/HeLa cell line results in increased levels of p21, Catalunya, the Center for Biomedical Research on Rare Diseases (CIBERER) p16 and β-gal and less telomerase recruitment to telomeres, also (ACCI19-­759), the ASL-­HSP France and the Hesperia Foundation to AP. This leading to cellular senescence,22 indicating that a tight regula- study was also funded by the Instituto de Salud Carlos III and ’Fondo Europeo tion of SARS1 levels is pivotal for homeostatic control of cell de Desarrollo Regional (FEDER), Unión Europea, una manera de hacer Europa’ (FIS PI20/00758) to CC and ’La Marató de TV3’ Foundation (202006-­30) to CC growth and cell cycle. Thus, SARS1 emerges as the only known and AP. This study was also funded by Instituto de Salud Carlos III (Sara Borrell tRNA synthetase with such a non-­canonical, specific function programme, CD19/00221) to EV, the Spanish Ministerio de Economia, Industria in cellular senescence. Our results further strengthen the notion y Competividad (Juan de la Cierva programme FJCI-­2016-­28811) to EV, and the that SARS1 dysfunction results into increased senescence levels, Center for Biomedical Research on Rare Diseases (CIBERER) to MR and NL. This and show for the first time how it can be induced by impairing study was supported by the French National Programme Investissement d’Avenir serylation activity, instead of modifying SARS1 expression. administered by the ’Agence National de la Recherche’ (ANR), ’MitoCross’ Laboratory of Excellence (Labex), funded as ANR-­10-­IDEX-­0002-­02, the University In a more global perspective, tRNA metabolism is closely asso- of Strasbourg and CNRS (to HB, BS). The CERCA Programme/Generalitat de ciated with lifespan and ageing through regulation of processes Catalunya provided institutional support. AP is member of the Undiagnosed such as tRNA transcription, modification/transformation of Disease Network International (UDNI). tRNAs, aminoacylation or the effect of the previous on global Competing interests None declared. transcription. As illustrative examples, C. elegans models in Patient consent for publication Consent obtained directly from patient(s) which two tRNA synthetase genes were downregulated resulted Ethics approval This study was approved by Clinical Research Ethics Committee of into increased lifespan (LARS2,23 LARS124), whereas decreased the Bellvitge University Hospital (PR076/14) and VHIR Institute (PR(AG) 223/2017). lifespan was observed in several Drosophila models with inacti- Participants gave informed consent to participate in the study before taking part. vation or decreased expression of GARS1, MARS2 or MARS1.25 Provenance and peer review Not commissioned; externally peer reviewed. Along these lines, there is previous evidence connecting preco- cious senescence and neurodevelopmental phenotypes in the Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information. literature, including a developmental delay disorder caused by mutations in ALKBH8, a methyltransferase that acts on the Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not wobble uridine base of several tRNA.26 Alkbh8-­deficient mouse have been peer-­reviewed. Any opinions or recommendations discussed are embryonic fibroblasts showed increased β-gal levels, heteroch- solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all romatin foci and senescence-­ associated secretory phenotype liability and responsibility arising from any reliance placed on the content. markers.21 Another tRNA methyltransferase gene, NSUN2, Where the content includes any translated material, BMJ does not warrant the whose recessive loss-­ of-­ function mutations cause intellectual accuracy and reliability of the translations (including but not limited to local disability (OMIM: 611091) is also associated with senescence: regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and NSUN2-­ knockdown human fibroblasts showed repressed cell adaptation or otherwise. proliferation and accelerated replicative senescence.27 28 Open access This is an open access article distributed in accordance with the In summary, we report a novel dominant-­negative mutation Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits in SARS1 and a novel mode of inheritance, underscoring the others to copy, redistribute, remix, transform and build upon this work for any need for caution in interpreting variants in ARS enzyme genes purpose, provided the original work is properly cited, a link to the licence is given, according to inheritance modes, with important implications and indication of whether changes were made. See: https://creativecommons.org/​ for genetic diagnosis and counselling. Furthermore, we unveil licenses/by/4.0/. pivotal roles of this enzyme in the control of cell growth and ORCID iDs senescence, underlying the neurodevelopmental phenotypes of Edgard Verdura http://orcid.org/0000-0003-3856-2060 affected patients. Aurora Pujol http://orcid.org/0000-0002-9606-0600 1232 Verdura E, et al. J Med Genet 2022;59:1227–1233. doi:10.1136/jmedgenet-2022-108529 Neurogenetics REFERENCES Sokol H, Spector T, Sullivan PF, Suvisaari J, Tai ES, Teo YY, Tiinamaija T, Tsuang M, 1 Antonellis A, Green ED. The role of aminoacyl-­tRNA synthetases in genetic diseases. Turner D, Tusie-­Luna T, Vartiainen E, Watkins H, Weersma RK, Wessman M, Wilson Annu Rev Genomics Hum Genet 2008;9:87–107. JG, Xavier RJ, Neale BM, Daly MJ, Genome Aggregation Database Consortium. The 2 Schimmel P. Development of tRNA synthetases and connection to genetic code and mutational constraint spectrum quantified from variation in 141,456 humans. Nature disease. Protein Sci 2008;17:1643–52. 2020;581:434–43. 3 Meyer-­Schuman R, Antonellis A. Emerging mechanisms of aminoacyl-­tRNA 14 Xu X, Shi Y, Yang X-­L. Crystal structure of human seryl-­tRNA synthetase and synthetase mutations in recessive and dominant human disease. Hum Mol Genet Ser-­SA complex reveals a molecular lever specific to higher eukaryotes. Structure 2017;26:R114–27. 2013;21:2078–86. 4 Oprescu SN, Griffin LB, Beg AA, Antonellis A. Predicting the pathogenicity of 15 Wang C, Guo Y, Tian Q, Jia Q, Gao Y, Zhang Q, Zhou C, Xie W. SerRS-­tRNASec complex aminoacyl-­tRNA synthetase mutations. Methods 2017;113:139–51. structures reveal mechanism of the first step in selenocysteine biosynthesis. Nucleic 5 Gonzalez-­Flores JN, Shetty SP, Dubey A, Copeland PR. The molecular biology of Acids Res 2015;43:10534. selenocysteine. Biomol Concepts 2013;4:349–65. 16 Cusack S, Yaremchuk A, Tukalo M. The crystal structure of the ternary complex of 6 Fukui H, Hanaoka R, Kawahara A. Noncanonical activity of seryl-­tRNA synthetase is T.thermophilus seryl-­tRNA synthetase with tRNA(Ser) and a seryl-­adenylate analogue involved in vascular development. Circ Res 2009;104:1253–9. reveals a conformational switch in the active site. Embo J 1996;15:2834–42. 7 Herzog W, Müller K, Huisken J, Stainier DYR. Genetic evidence for a noncanonical 17 Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-­Foster J, Grody WW, Hegde M, function of seryl-­tRNA synthetase in vascular development. Circ Res Lyon E, Spector E, Voelkerding K, Rehm HL, ACMG Laboratory Quality Assurance 2009;104:1260–6. Committee. Standards and guidelines for the interpretation of sequence variants: a 8 Xu X, Shi Y, Zhang H-­M, Swindell EC, Marshall AG, Guo M, Kishi S, Yang X-­L. joint consensus recommendation of the American College of medical genetics and Unique domain appended to vertebrate tRNA synthetase is essential for vascular genomics and the association for molecular pathology. Genet Med 2015;17:405–24. development. Nat Commun 2012;3:681–9. 18 Mah L-­J, El-­Osta A, Karagiannis TC. gammaH2AX: a sensitive molecular marker of 9 Musante L, Püttmann L, Kahrizi K, Garshasbi M, Hu H, Stehr H, Lipkowitz B, Otto DNA damage and repair. Leukemia 2010;24:679–86. S, Jensen LR, Tzschach A, Jamali P, Wienker T, Najmabadi H, Ropers HH, Kuss AW. 19 Fine AS, Nemeth CL, Kaufman ML, Fatemi A. Mitochondrial aminoacyl-­tRNA Mutations of the aminoacyl-­tRNA-­synthetases SARS and WARS2 are implicated in the synthetase disorders: an emerging group of developmental disorders of myelination. J etiology of autosomal recessive intellectual disability. Hum Mutat 2017;38:621–36. Neurodev Disord 2019;11:29. 10 Ravel J-­M, Dreumont N, Mosca P, Smith DEC, Mendes MI, Wiedemann A, Coelho D, 20 Fuchs SA, Schene IF, Kok G, Jansen JM, Nikkels PGJ, van Gassen KLI, Terheggen-­Lagro Schmitt E, Rivière J-­B, Tran Mau-­Them F, Thevenon J, Kuentz P, Polivka M, Fuchs SA, SWJ, van der Crabben SN, Hoeks SE, Niers LEM, Wolf NI, de Vries MC, Koolen DA, Kok G, Thauvin-­Robinet C, Guéant J-­L, Salomons GS, Faivre L, Feillet F. A bi-­allelic Houwen RHJ, Mulder MF, van Hasselt PM. Aminoacyl-­tRNA synthetase deficiencies in loss-­of-­function SARS1 variant in children with neurodevelopmental delay, deafness, search of common themes. Genet Med 2019;21:319–30. cardiomyopathy, and decompensation during fever. Hum Mutat 2021;42:1576–83. 21 Manole A, Efthymiou S, O’Connor E, Mendes MI, Jennings M, Maroofian R, 11 Gibson DG, Young L, Chuang R-­Y, Venter JC, Hutchison CA, Smith HO. Enzymatic Davagnanam I, Mankad K, Lopez MR, Salpietro V, Harripaul R, Badalato L, Walia J, assembly of DNA molecules up to several hundred kilobases. Nat Methods Francklyn CS, Athanasiou-­Fragkouli A, Sullivan R, Desai S, Baranano K, Zafar F, Rana N, 2009;6:343–5. Ilyas M, Horga A, Kara M, Mattioli F, Goldenberg A, Griffin H, Piton A, Henderson LB, 12 Schlüter A, Rodríguez-­Palmero A, Verdura E, Vélez-­Santamaría V, Ruiz M, Fourcade Kara B, Aslanger AD, Raaphorst J, Pfundt R, Portier R, Shinawi M, Kirby A, Christensen S, Planas-­Serra L, Martínez JJ, Guilera C, Girós M, Artuch R, Yoldi ME, O’Callaghan KM, Wang L, Rosti RO, Paracha SA, Sarwar MT, Jenkins D, Ahmed J, Santoni FA, Ranza M, García-­Cazorla A, Armstrong J, Marti I, Mondragón Rezola E, Redin C, Mandel JL, E, Iwaszkiewicz J, Cytrynbaum C, Weksberg R, Wentzensen IM, Guillen Sacoto MJ, Si Y, Conejo D, Sierra-­Córcoles C, Beltrán S, Gut M, Vázquez E, Del Toro M, Troncoso M, Telegrafi A, Andrews MV, Baldridge D, Gabriel H, Mohr J, Oehl-­Jaschkowitz B, Debard Pérez-­Jurado LA, Gutiérrez-­Solana LG, López de Munain A, Casasnovas C, Aguilera-­ S, Senger B, Fischer F, van Ravenwaaij C, Fock AJM, Stevens SJC, Bähler J, Nasar A, Albesa S, Macaya A, Pujol A, Rezola EM, de Munain AL, Arroyo HA, Barrios A, Campo Mantovani JF, Manzur A, Sarkozy A, Smith DEC, Salomons GS, Ahmed ZM, Riazuddin A, Castillo T, Cazorla R, García MA, García A, Hedrera A, Hernández J, Launay N, S, Riazuddin S, Usmani MA, Seibt A, Ansar M, Antonarakis SE, Vincent JB, Ayub M, Lorenzo M, Miranda C, Moreno F, Muñoz A, Narbona J, Pérez MS, Ramos MA, Chaure Grimmel M, Jelsig AM, Hjortshøj TD, Karstensen HG, Hummel M, Haack TB, Jamshidi MR, Roig-­Quilis M, Urtasun MA, Vázquez ME, Vázquez JF, GWMD working group. Y, Distelmaier F, Horvath R, Gleeson JG, Becker H, Mandel J-­L, Koolen DA, Houlden Diagnosis of genetic white matter disorders by singleton whole-­exome and genome H, v AM, Riazuddin SS, De Novo HH, SYNAPS Study Group. De novo and bi-­allelic sequencing using Interactome-­Driven prioritization. Neurology 2022;98:e912–23. pathogenic variants in NARS1 cause neurodevelopmental delay due to toxic gain-­of-­ 13 Karczewski KJ, Francioli LC, Tiao G, Cummings BB, Alföldi J, Wang Q, Collins RL, function and partial loss-­of-­function effects. Am J Hum Genet 2020;107:311–24. Laricchia KM, Ganna A, Birnbaum DP, Gauthier LD, Brand H, Solomonson M, 22 Li Y, Li X, Cao M, Jiang Y, Yan J, Liu Z, Yang R, Chen X, Sun P, Xiang R, Wang L, Shi Y. Watts NA, Rhodes D, Singer-­Berk M, England EM, Seaby EG, Kosmicki JA, Walters Seryl tRNA synthetase cooperates with POT1 to regulate telomere length and cellular RK, Tashman K, Farjoun Y, Banks E, Poterba T, Wang A, Seed C, Whiffin N, Chong senescence. Signal Transduction and Targeted Therapy 2019;4:1–11. JX, Samocha KE, Pierce-­Hoffman E, Zappala Z, O’Donnell-­Luria AH, Minikel EV, 23 Lee SS, Lee RYN, Fraser AG, Kamath RS, Ahringer J, Ruvkun G. A systematic RNAi Weisburd B, Lek M, Ware JS, Vittal C, Armean IM, Bergelson L, Cibulskis K, Connolly screen identifies a critical role for mitochondria in C. elegans longevity. Nat Genet KM, Covarrubias M, Donnelly S, Ferriera S, Gabriel S, Gentry J, Gupta N, Jeandet T, 2003;33:40–8. Kaplan D, Llanwarne C, Munshi R, Novod S, Petrillo N, Roazen D, Ruano-­Rubio V, 24 Nakamura S, Karalay Özlem, Jäger PS, Horikawa M, Klein C, Nakamura K, Latza C, Saltzman A, Schleicher M, Soto J, Tibbetts K, Tolonen C, Wade G, Talkowski ME, Neale Templer SE, Dieterich C, Antebi A. Mondo complexes regulate TFEB via TOR inhibition BM, Daly MJ, MacArthur DG, Aguilar Salinas CA, Ahmad T, Albert CM, Ardissino to promote longevity in response to gonadal signals. Nat Commun 2016;7:1–15. D, Atzmon G, Barnard J, Beaugerie L, Benjamin EJ, Boehnke M, Bonnycastle LL, 25 Zhou Z, Sun B, Yu D, Bian M. Roles of tRNA metabolism in aging and lifespan. Cell Bottinger EP, Bowden DW, Bown MJ, Chambers JC, Chan JC, Chasman D, Cho J, Death Dis 2021;12:1–11. Chung MK, Cohen B, Correa A, Dabelea D, Darbar D, Duggirala R, Dupuis J, Ellinor 26 Monies D, Vågbø CB, Al-­Owain M, Alhomaidi S, Alkuraya FS. Recessive truncating PT, Elosua R, Erdmann J, Esko T, Färkkilä M, Florez J, Franke A, Getz G, Glaser B, mutations in ALKBH8 cause intellectual disability and severe impairment of wobble Glatt SJ, Goldstein D, Gonzalez C, Groop L, Haiman C, Hanis C, Harms M, Hiltunen uridine modification. Am J Hum Genet 2019;104:1202–9. M, Holi MM, Hultman CM, Kallela M, Kaprio J, Kathiresan S, Kim BJ, Kim YJ, Kirov G, 27 Tang H, Fan X, Xing J, Liu Z, Jiang B, Dou Y, Gorospe M, Wang W. Nsun2 delays Kooner J, Koskinen S, Krumholz HM, Kugathasan S, Kwak SH, Laakso M, Lehtimäki T, replicative senescence by repressing p27 (Kip1) translation and elevating Cdk1 Loos RJF, Lubitz SA, Ma RCW MDG, Marrugat J, Mattila KM, McCarroll S, McCarthy translation. Aging 2015;7:1143–55. MI, McGovern D, McPherson R, Meigs JB, Melander O, Metspalu A, Nilsson PM, 28 Abbasi-­Moheb L, Mertel S, Gonsior M, Nouri-­Vahid L, Kahrizi K, Cirak S, Wieczorek D, O’Donovan MC, Ongur D, Orozco L, Owen MJ, Palmer CNA, Palotie A, Park KS, Pato Motazacker MM, Esmaeeli-­Nieh S, Cremer K, Weißmann R, Tzschach A, Garshasbi M, C, Pulver AE, Rahman N, Remes AM, Rioux JD, Ripatti S, Roden DM, Saleheen D, Abedini SS, Najmabadi H, Ropers HH, Sigrist SJ, Kuss AW. Mutations in NSun2 cause Salomaa V, Samani NJ, Scharf J, Schunkert H, Shoemaker MB, Sklar P, Soininen H, autosomal-­recessive intellectual disability. Am J Hum Genet 2012;90:847–55. Verdura E, et al. J Med Genet 2022;59:1227–1233. doi:10.1136/jmedgenet-2022-108529 1233