Title: Biallelic Mutations in TMEM126B Cause Severe Complex I Deficiency with a Variable Clinical Phenotype
Abstract: Complex I deficiency is the most common biochemical phenotype observed in individuals with mitochondrial disease. With 44 structural subunits and over 10 assembly factors, it is unsurprising that complex I deficiency is associated with clinical and genetic heterogeneity. Massively parallel sequencing (MPS) technologies including custom, targeted gene panels or unbiased whole-exome sequencing (WES) are hugely powerful in identifying the underlying genetic defect in a clinical diagnostic setting, yet many individuals remain without a genetic diagnosis. These individuals might harbor mutations in poorly understood or uncharacterized genes, and their diagnosis relies upon characterization of these orphan genes. Complexome profiling recently identified TMEM126B as a component of the mitochondrial complex I assembly complex alongside proteins ACAD9, ECSIT, NDUFAF1, and TIMMDC1. Here, we describe the clinical, biochemical, and molecular findings in six cases of mitochondrial disease from four unrelated families affected by biallelic (c.635G>T [p.Gly212Val] and/or c.401delA [p.Asn134Ilefs∗2]) TMEM126B variants. We provide functional evidence to support the pathogenicity of these TMEM126B variants, including evidence of founder effects for both variants, and establish defects within this gene as a cause of complex I deficiency in association with either pure myopathy in adulthood or, in one individual, a severe multisystem presentation (chronic renal failure and cardiomyopathy) in infancy. Functional experimentation including viral rescue and complexome profiling of subject cell lines has confirmed TMEM126B as the tenth complex I assembly factor associated with human disease and validates the importance of both genome-wide sequencing and proteomic approaches in characterizing disease-associated genes whose physiological roles have been previously undetermined. Complex I deficiency is the most common biochemical phenotype observed in individuals with mitochondrial disease. With 44 structural subunits and over 10 assembly factors, it is unsurprising that complex I deficiency is associated with clinical and genetic heterogeneity. Massively parallel sequencing (MPS) technologies including custom, targeted gene panels or unbiased whole-exome sequencing (WES) are hugely powerful in identifying the underlying genetic defect in a clinical diagnostic setting, yet many individuals remain without a genetic diagnosis. These individuals might harbor mutations in poorly understood or uncharacterized genes, and their diagnosis relies upon characterization of these orphan genes. Complexome profiling recently identified TMEM126B as a component of the mitochondrial complex I assembly complex alongside proteins ACAD9, ECSIT, NDUFAF1, and TIMMDC1. Here, we describe the clinical, biochemical, and molecular findings in six cases of mitochondrial disease from four unrelated families affected by biallelic (c.635G>T [p.Gly212Val] and/or c.401delA [p.Asn134Ilefs∗2]) TMEM126B variants. We provide functional evidence to support the pathogenicity of these TMEM126B variants, including evidence of founder effects for both variants, and establish defects within this gene as a cause of complex I deficiency in association with either pure myopathy in adulthood or, in one individual, a severe multisystem presentation (chronic renal failure and cardiomyopathy) in infancy. Functional experimentation including viral rescue and complexome profiling of subject cell lines has confirmed TMEM126B as the tenth complex I assembly factor associated with human disease and validates the importance of both genome-wide sequencing and proteomic approaches in characterizing disease-associated genes whose physiological roles have been previously undetermined. Complex I deficiency is the most common biochemical phenotype observed in subjects with mitochondrial disease.1Janssen R.J. Nijtmans L.G. van den Heuvel L.P. Smeitink J.A. Mitochondrial complex I: structure, function and pathology.J. Inherit. Metab. Dis. 2006; 29: 499-515Crossref PubMed Scopus (222) Google Scholar It can occur as an isolated complex deficiency, where biochemical assessment of enzyme activities of other respiratory-chain components (complexes II, III, and IV) is normal, or as part of a multiple-respiratory-chain-complex deficiency with the involvement of other parts of the oxidative phosphorylation (OXPHOS) system. The latter is suggestive of a global mitochondrial defect involving, for example, mitochondrial maintenance, protein translation, or mitochondrial import. Mitochondrial complex I deficiency is phenotypically diverse, such that clinical presentations range from subacute necrotizing encephalomyelopathy (Leigh syndrome [MIM: 256000]) to pure myopathy and exercise intolerance.1Janssen R.J. Nijtmans L.G. van den Heuvel L.P. Smeitink J.A. Mitochondrial complex I: structure, function and pathology.J. Inherit. Metab. Dis. 2006; 29: 499-515Crossref PubMed Scopus (222) Google Scholar, 2Bugiani M. Invernizzi F. Alberio S. Briem E. Lamantea E. Carrara F. Moroni I. Farina L. Spada M. Donati M.A. et al.Clinical and molecular findings in children with complex I deficiency.Biochim. Biophys. Acta. 2004; 1659: 136-147Crossref PubMed Scopus (222) Google Scholar In cases of isolated complex I deficiency, the genetic basis can be attributed to defects in the mitochondrial DNA (mtDNA) genes encoding seven structural subunits, in the nuclear genes encoding any of 37 other structural subunits, or in the increasing number of ancillary proteins that are responsible for faithful biogenesis and assembly of complex I. Such heterogeneity results in complicated diagnostic pipelines for clinical subjects. Massively parallel sequencing (MPS) strategies, whether in the form of whole-exome sequencing (WES)3Haack T.B. Haberberger B. Frisch E.M. Wieland T. Iuso A. Gorza M. Strecker V. Graf E. Mayr J.A. Herberg U. et al.Molecular diagnosis in mitochondrial complex I deficiency using exome sequencing.J. Med. Genet. 2012; 49: 277-283Crossref PubMed Scopus (140) Google Scholar or targeted capture (e.g., Ampliseq),4Calvo S.E. Compton A.G. Hershman S.G. Lim S.C. Lieber D.S. Tucker E.J. Laskowski A. Garone C. Liu S. Jaffe D.B. et al.Molecular diagnosis of infantile mitochondrial disease with targeted next-generation sequencing.Sci. Transl. Med. 2012; 4: 118ra10Crossref PubMed Scopus (356) Google Scholar are proving extremely effective at establishing genetic diagnoses, particularly when mutations occur within known or candidate disease-associated genes. To date, mutations have been identified in all seven mtDNA-encoded structural subunits of complex I and 20 nuclear-encoded structural genes;5Fassone E. Rahman S. Complex I deficiency: clinical features, biochemistry and molecular genetics.J. Med. Genet. 2012; 49: 578-590Crossref PubMed Scopus (211) Google Scholar, 6Angebault C. Charif M. Guegen N. Piro-Megy C. Mousson de Camaret B. Procaccio V. Guichet P.O. Hebrard M. Manes G. Leboucq N. et al.Mutation in NDUFA13/GRIM19 leads to early onset hypotonia, dyskinesia and sensorial deficiencies, and mitochondrial complex I instability.Hum. Mol. Genet. 2015; 24: 3948-3955Crossref PubMed Scopus (35) Google Scholar, 7van Rahden V.A. Fernandez-Vizarra E. Alawi M. Brand K. Fellmann F. Horn D. Zeviani M. Kutsche K. Mutations in NDUFB11, encoding a complex I component of the mitochondrial respiratory chain, cause microphthalmia with linear skin defects syndrome.Am. J. Hum. Genet. 2015; 96: 640-650Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar similarly, subjects have been reported with defects in nine assembly factors.5Fassone E. Rahman S. Complex I deficiency: clinical features, biochemistry and molecular genetics.J. Med. Genet. 2012; 49: 578-590Crossref PubMed Scopus (211) Google Scholar However, even after WES analysis, a significant proportion of subjects lack a genetic diagnosis—a common explanation is that their mutations affect an uncharacterized protein.8Calvo S.E. Clauser K.R. Mootha V.K. MitoCarta2.0: an updated inventory of mammalian mitochondrial proteins.Nucleic Acids Res. 2016; 44: D1251-D1257Crossref PubMed Scopus (861) Google Scholar, 9Pagliarini D.J. Calvo S.E. Chang B. Sheth S.A. Vafai S.B. Ong S.E. Walford G.A. Sugiana C. Boneh A. Chen W.K. et al.A mitochondrial protein compendium elucidates complex I disease biology.Cell. 2008; 134: 112-123Abstract Full Text Full Text PDF PubMed Scopus (1501) Google Scholar Here, we describe a cohort of six subjects who all harbor recessive mutations within the gene encoding TMEM126B, a protein recently identified as a complex I assembly factor by a proteomic study of knockdown cell lines.10Heide H. Bleier L. Steger M. Ackermann J. Dröse S. Schwamb B. Zörnig M. Reichert A.S. Koch I. Wittig I. Brandt U. Complexome profiling identifies TMEM126B as a component of the mitochondrial complex I assembly complex.Cell Metab. 2012; 16: 538-549Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar Complexome profiling revealed TMEM126B to be a component of the mitochondrial complex I assembly (MCIA) complex alongside proteins ACAD9, ECSIT, NDUFAF1, and TIMMDC1, thus establishing TMEM126B (MIM: 615533) as a candidate gene for complex I deficiency.10Heide H. Bleier L. Steger M. Ackermann J. Dröse S. Schwamb B. Zörnig M. Reichert A.S. Koch I. Wittig I. Brandt U. Complexome profiling identifies TMEM126B as a component of the mitochondrial complex I assembly complex.Cell Metab. 2012; 16: 538-549Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar, 11Guarani V. Paulo J. Zhai B. Huttlin E.L. Gygi S.P. Harper J.W. TIMMDC1/C3orf1 functions as a membrane-embedded mitochondrial complex I assembly factor through association with the MCIA complex.Mol. Cell. Biol. 2014; 34: 847-861Crossref PubMed Scopus (66) Google Scholar With access to subjects harboring putative TMEM126B defects, we provide functional evidence to support the pathogenicity of these TMEM126B variants, unequivocally establishing this gene as a cause of complex I deficiency in association with either a severe multisystem presentation in infancy or pure myopathy in later child- or adulthood. This report describes the clinical, biochemical, and molecular findings in six cases of TMEM126B-related mitochondrial disease and validates the importance of proteomic approaches in identifying disease-associated genes whose physiological roles have been previously undetermined. Subject 1 (family 1 subject II-1 in Figure 1A) was born to American parents without known consanguinity. He presented in childhood with pure exercise intolerance without muscle weakness. Exercise (running and swimming) caused leg fatigue, shortness of breath, and a rapid heart rate, often provoking vomiting and severe headache. Cardiology review in early adulthood showed normal electrocardiography (ECG) and echocardiography. Treadmill exercise testing caused fatigue after 2 min with a heart rate of 180 and elevated blood lactate (16 mmol/L; normal range < 2.0 mmol/L), characteristic of mitochondrial dysfunction. He had normal creatine kinase (CK) levels, and there was no pigmenturia. Physical examination remains normal at 23 years of age. Subjects 2 and 3 (family 2 subjects II-1 and II-2, respectively, in Figure 1A) are brothers who were born to non-consanguineous parents in Belgium. They presented in their early teens with exercise-induced dyspnea (subject 2), exercise intolerance (subjects 2 and 3), and post-exertional myalgia (subjects 2 and 3). Exertion was often followed by nausea and vomiting. Now in adulthood, currently aged 40 and 37 years, respectively, subjects 2 and 3 are wheelchair bound and have significantly impaired muscle strength affecting the lower limbs, particularly hip flexion and extension. Strength in the upper limbs is normal. Forced vital capacity, cardiac ultrasound, and cognitive development are normal, and neither subject has epilepsy, neuropathy, diabetes, or hearing impairment. Subject 2 has mild visual impairment (macular and peripheral retinal pigment migration) and had mild left ventricular hypertrophy in his twenties. CK was normal, but blood lactate (2.3–3.0 mmol/L in subject 2 and 3.2–3.8 mmol/L in subject 3) and cerebrospinal fluid lactate (5.8 mmol/L in subject 3) were elevated. Subjects 4 and 5 (family 3 subjects II-1 and II-2, respectively, in Figure 1A) are affected siblings who were born in Belgium to unrelated parents with no other children. Their father died at the age of 47 years and complained of mild exercise intolerance; their mother is alive and complains of fatigue. Subjects 4 and 5 (currently aged 33 and 30 years, respectively) presented in adolescence with fatigue, exercise intolerance, and exercise-induced nausea. No other organs are affected, although subject 5 reports gastrointestinal problems. Cardiac, ophthalmic, and nephrologic examination, intellectual capacity, and CK were normal for both subjects. Cycloergometry (for both siblings) showed very low submaximal and maximal capacity. Both subjects are able to walk but cannot ride a bike or run, and they have reported improvements following co-enzyme Q supplementation (200 mg/d). Subject 6 (family 4 subject II-1 in Figure 1A) is female and the second child of healthy, unrelated parents living in Poland. She was born at 37 weeks of gestation with a weight of 2,150 g (third percentile [−1.88 SD]) and an Apgar score of 10. Patent ductus arteriosus and an atrial septal defect without ventricular hypertrophy were observed, and transient assisted respiration was required in the early neonatal period. At the age of 2 months, she was admitted to the hospital with very poor weight gain and vomiting, and during this period she went into cardiac arrest, attributed to gastroesophageal reflux and protracted renal failure with severe tubular acidosis (pH 7.21 [normal range = 7.35–7.43], 13.5 mmol/L NaHCO3 [normal range = 22.0–26.0 mmol/L], 6.0 mmol/L potassium [normal range = 3.6–5.8 mmol/L], and 124 mmol/L sodium [normal range = 136–145 mmol/L]). Progressive hypertrophic cardiomyopathy, failure to thrive, and elevated blood lactate (8.1 mmol/L) prompted suspicion of mitochondrial disease. Currently aged 6 years, she is in good general condition and has age-appropriate motor and mental development but shows chronic renal failure (stage IV) and a marked growth deficit (−5.1 SD). She requires continuous administration of erythropoietin because of anemia and is supplemented with citrate and sodium because of tubular acidosis. Muscle and/or skin biopsy was performed for each subject, and biochemical, histochemical, and molecular investigations were undertaken (Table 1). Informed consent for diagnostic and research studies was obtained for all subjects in accordance with the Declaration of Helsinki protocols and approved by local institutional review boards.Table 1Biochemical and Clinical Findings in Individuals with TMEM126B VariantsSubject DetailsTMEM126B VariantsOXPHOS Activities in Skeletal MuscleClinical FeaturesIDSexcDNA (GenBank: ), Protein (GenBank: )RCCMean Enzyme ActivityAbsolute ValuesControl Mean (Reference Range)Age at OnsetClinical CourseOther Clinical Features and Relevant Family HistorySubject 1aInvestigated by WES.malec.[635G>T];[635G>T],p.[Gly212Val];[Gly212Val]I36% (↓)bBelow the normal range.1.85.0 ± 0.8 (n = 28)8 yearsalive at 21 yearsexercise intolerance, unable to perform sustained aerobic exercise, normal strength, normal ECG and echocardiography, normal resting lactate, normal CKII210% (↑↑)4.22.0 ± 0.6 (n = 44)III219% (↑↑)23.610.8 ± 2.3 (n = 29)IV218% (↑↑)8.53.9 ± 1.5 (n = 44)CS196% (↑↑)24.112.3 ± 2.7 (n = 44)Subject 2cInvestigated by targeted gene analysis (AmpliSeq capture or carrier testing).malec.[401delA];[635G>T],p.[Asn134Ilefs∗2];[Gly212Val]I48% (↓)bBelow the normal range.1429 ± 13 (n = 30)12 yearsalive at 39 years, wheelchair boundexercise intolerance, muscle weakness in lower limbs and pelvis, normal echocardiography, mild basal increases of lactate, normal CK, normal intelligence, retinitis pigmentosaII138%4734 ± 14 (n = 30)IIINDND96 ± 31 (n = 30)IV82%137167 ± 58 (n = 30)CS237% (↑↑)412174 ± 70 (n = 30)Subject 3cInvestigated by targeted gene analysis (AmpliSeq capture or carrier testing).malec.[401delA];[635G>T],p.[Asn134Ilefs∗2];[Gly212Val]I14% (↓↓)bBelow the normal range.429 ± 13 (n = 30)10 yearsalive at 36 years, wheelchair boundclinically affected sibling of subject 2, exercise intolerance, muscle weakness in lower limbs and pelvis, normal echocardiography, mild basal increases in lactate, normal CK, normal intelligence, no retinitis pigmentosaII179% (↑)6134 ± 14 (n = 30)IIINDND96 ± 31 (n = 30)IV103%172167 ± 58 (n = 30)CS281% (↑↑)489174 ± 70 (n = 30)Subject 4aInvestigated by WES.malec.[401delA];[635G>T],p.[Asn134Ilefs∗2];[Gly212Val]I10% (↓↓)bBelow the normal range.329 ± 13 (n = 30)8 yearsalive at 32 yearsexercise intolerance and fatigueII253% (↑↑)8634 ± 14 (n = 30)III172% (↑)16596 ± 31 (n = 30)IV126%210167 ± 58 (n = 30)CS201% (↑↑)350174 ± 70 (n = 30)Subject 5aInvestigated by WES.femalec.[401delA];[635G>T],p.[Asn134Ilefs∗2];[Gly212Val]I10% (↓↓)bBelow the normal range.329 ± 13 (n = 30)15 yearsalive at 29 yearsclinically affected sibling of subject 4, exercise intolerance and fatigueII288% (↑↑)9834 ± 14 (n = 30)III129%12496 ± 31 (n = 30)IV238% (↑↑)398167 ± 58 (n = 30)CS259% (↑↑)451174 ± 70 (n = 30)Subject 6aInvestigated by WES.femalec.[635G>T];[635G>T],p.[ Gly212Val];[Gly212Val]I17% (↓↓)bBelow the normal range.317 ± 8 (n = 15)2 monthsalive at 5.5 yearsmultiorgan involvement manifesting in infancy (respiratory failure, cardiomyopathy, and renal acidosis), severe growth failure, chronic renal insufficiency, elevated serum lactateII135%1310 ± 3 (n = 15)III64%5890 ± 52 (n = 15)IV82%1012 ± 9 (n = 15)CS228% (↑↑)458200.9 ± 48.5 (n = 15)For subject 1, respiratory-chain enzyme activities are expressed as U/min/g wet weight.12Kirby D.M. Thorburn D.R. Turnbull D.M. Taylor R.W. Biochemical assays of respiratory chain complex activity.Methods Cell Biol. 2007; 80: 93-119Crossref PubMed Scopus (309) Google Scholar For subjects 2–6, enzyme activities are expressed as nanomoles of substrate/min/mg protein.13Vanlander A.V. Menten B. Smet J. De Meirleir L. Sante T. De Paepe B. Seneca S. Pearce S.F. Powell C.A. Vergult S. et al.Two siblings with homozygous pathogenic splice-site variant in mitochondrial asparaginyl-tRNA synthetase (NARS2).Hum. Mutat. 2015; 36: 222-231Crossref PubMed Scopus (41) Google Scholar The following abbreviations are used: ↓, decreased; ↓↓, markedly decreased; ↑, increased; ↑↑, markedly increased; ECG, electrocardiography; and ND, not determined.a Investigated by WES.b Below the normal range.c Investigated by targeted gene analysis (AmpliSeq capture or carrier testing). Open table in a new tab For subject 1, respiratory-chain enzyme activities are expressed as U/min/g wet weight.12Kirby D.M. Thorburn D.R. Turnbull D.M. Taylor R.W. Biochemical assays of respiratory chain complex activity.Methods Cell Biol. 2007; 80: 93-119Crossref PubMed Scopus (309) Google Scholar For subjects 2–6, enzyme activities are expressed as nanomoles of substrate/min/mg protein.13Vanlander A.V. Menten B. Smet J. De Meirleir L. Sante T. De Paepe B. Seneca S. Pearce S.F. Powell C.A. Vergult S. et al.Two siblings with homozygous pathogenic splice-site variant in mitochondrial asparaginyl-tRNA synthetase (NARS2).Hum. Mutat. 2015; 36: 222-231Crossref PubMed Scopus (41) Google Scholar The following abbreviations are used: ↓, decreased; ↓↓, markedly decreased; ↑, increased; ↑↑, markedly increased; ECG, electrocardiography; and ND, not determined. Histochemical analysis of all subjects’ muscle biopsy revealed subsarcolemmal accumulation of mitochondria, suggestive of mitochondrial proliferation and evolving pathology of ragged-red fibers (Figure S1). Biochemical analysis of muscle respiratory-chain activities revealed a marked isolated complex I deficiency in all subjects, suggestive of a defect involving mtDNA or a nuclear-encoded protein implicit in complex I structure or assembly. The genetic basis was identified by previously described MPS strategies involving either a custom, targeted AmpliSeq panel (subjects 2 and 3) or WES (subjects 1 and 4–6) as described elsewhere.14Ploski R. Pollak A. Müller S. Franaszczyk M. Michalak E. Kosinska J. Stawinski P. Spiewak M. Seggewiss H. Bilinska Z.T. Does p.Q247X in TRIM63 cause human hypertrophic cardiomyopathy?.Circ. Res. 2014; 114: e2-e5Crossref PubMed Scopus (82) Google Scholar, 15Haack T.B. Hogarth P. Kruer M.C. Gregory A. Wieland T. Schwarzmayr T. Graf E. Sanford L. Meyer E. Kara E. et al.Exome sequencing reveals de novo WDR45 mutations causing a phenotypically distinct, X-linked dominant form of NBIA.Am. J. Hum. Genet. 2012; 91: 1144-1149Abstract Full Text Full Text PDF PubMed Scopus (260) Google Scholar For all cases, biallelic variants in TMEM126B (GenBank: NM_018480.4 and NP_060950.3) were identified—just two TMEM126B genotypes, either a homozygous c.635G>T (p.Gly212Val) missense variant (subjects 1 and 6) or a compound-heterozygous c.401delA (p.Asn134Ilefs∗2) and c.635G>T (p.Gly212Val) genotype (subjects 2–5), account for the clinical phenotype of each subject in our cohort (Table 1 and Figures 1A and 1B). Where familial samples were available from parents and unaffected siblings, these variants were found to segregate with a clinically affected status. The c.401delA (p.Asn134Ilefs∗2) variant is absent from dbSNP, the National Heart, Lung, and Blood Institute (NHLBI) Exome Sequencing Project Exome Variant Server (ESP6500), and the Exome Aggregation Consortium (ExAC) Browser (as of February 10, 2016). The c.635G>A (p.Gly212Val) variant is referenced in dbSNP (rs141542003) and recorded in ESP6500 (Europeans: 16/8,598 alleles [0.2%]) and the ExAC Browser (Europeans: 146/72,144 alleles [0.2%]; non-Europeans: 10/38,138 [0.02%]). No homozygous cases have been recorded (according to ExAC, ESP6500, and dbSNP data as of February 10, 2016), and subject 1 was the only individual to have rare potentially pathogenic biallelic TMEM126B variants in over 7,500 samples sequenced at the Institute of Human Genetics in Munich (where the c.635G>A variant was present in 15/15,134 alleles [0.1%]). Both TMEM126B variants have been submitted to ClinVar (see Accession Numbers). Because the structure of TMEM126B has not been solved, in silico modeling of TMEM126B tertiary structure was performed with I-TASSER,16Yang J. Yan R. Roy A. Xu D. Poisson J. Zhang Y. The I-TASSER Suite: protein structure and function prediction.Nat. Methods. 2015; 12: 7-8Crossref PubMed Scopus (3764) Google Scholar Phyre2,17Kelley L.A. Mezulis S. Yates C.M. Wass M.N. Sternberg M.J. The Phyre2 web portal for protein modeling, prediction and analysis.Nat. Protoc. 2015; 10: 845-858Crossref PubMed Scopus (6155) Google Scholar and RaptorX,18Källberg M. Margaryan G. Wang S. Ma J. Xu J. RaptorX server: a resource for template-based protein structure modeling.Methods Mol. Biol. 2014; 1137: 17-27Crossref PubMed Scopus (190) Google Scholar and although confidence was low for overall structure predictions, each tool predicted the Gly212 residue to be located within a helical domain. Glycine, the smallest amino acid and the only one without a carbon-containing side chain, is often critical within helices because it is permissive in structure and allows the helix to twist. Its substitution for a branched-chain amino acid, such as valine, is likely to affect the tertiary structure and thus compromise protein function.19Richards A.J. Lloyd J.C. Ward P.N. De Paepe A. Narcisi P. Pope F.M. Characterisation of a glycine to valine substitution at amino acid position 910 of the triple helical region of type III collagen in a patient with Ehlers-Danlos syndrome type IV.J. Med. Genet. 1991; 28: 458-463Crossref PubMed Google Scholar, 20Javadpour M.M. Eilers M. Groesbeek M. Smith S.O. Helix packing in polytopic membrane proteins: role of glycine in transmembrane helix association.Biophys. J. 1999; 77: 1609-1618Abstract Full Text Full Text PDF PubMed Scopus (232) Google Scholar This is corroborated by in silico prediction tools including SIFT,21Ng P.C. Henikoff S. SIFT: Predicting amino acid changes that affect protein function.Nucleic Acids Res. 2003; 31: 3812-3814Crossref PubMed Scopus (4109) Google Scholar MutationTaster,22Schwarz J.M. Cooper D.N. Schuelke M. Seelow D. MutationTaster2: mutation prediction for the deep-sequencing age.Nat. Methods. 2014; 11: 361-362Crossref PubMed Scopus (2441) Google Scholar and PolyPhen-2,23Adzhubei I.A. Schmidt S. Peshkin L. Ramensky V.E. Gerasimova A. Bork P. Kondrashov A.S. Sunyaev S.R. A method and server for predicting damaging missense mutations.Nat. Methods. 2010; 7: 248-249Crossref PubMed Scopus (9354) Google Scholar which support a detrimental effect due to the p.Gly212Val substitution. Moderate evolutionary conservation of the Gly212 TMEM126B residue was suggested by Clustal Omega alignment of Ensembl-derived orthologs (Figure 1C). Given that just two TMEM126B variants were identified in our ethnically diverse cohort of subjects (from Belgium, the United States, and Poland), we performed SNP genotyping to investigate a possible founder effect (Figure 1D and Tables S1 and S2). The most likely haplotype structure for the subjects was inferred with the SHAPEIT2 algorithm.24O’Connell J. Gurdasani D. Delaneau O. Pirastu N. Ulivi S. Cocca M. Traglia M. Huang J. Huffman J.E. Rudan I. et al.A general approach for haplotype phasing across the full spectrum of relatedness.PLoS Genet. 2014; 10: e1004234Crossref PubMed Scopus (330) Google Scholar As anticipated, there was evidence of two alleles shared by state (0.81 cM region from 91.31 to 92.12 cM) for the Belgian sibling pairs from two apparently unrelated families (subjects 2 and 3 and subjects 4 and 5). Similarly, there was a shared haplotype (1.15 cM region from 91.31 to 92.46 cM) between subjects 2 (Belgian) and 6 (Polish), and this was echoed by a 1.37 cM shared haplotype from 90.75 to 92.12 cM in an analysis involving subjects 4 (Belgian) and 6 (Polish). Together, these data support common ancestors and the c.401delA (p.Asn134Ilefs∗2) and c.635G>T (p.Gly212Val) variants as founder mutations. Subject 1, of European-American ancestry, was found to have a very small homozygosity-by-state (HBS) tract (0.07 cM, ∼500 kb genomic distance), but on a background suggestive of first-cousin parentage. The homozygous c.635G>T (p.Gly212Val) variant occurs within the HBS tract but is intriguingly outside the large identity-by-descent tracts shared as a result of consanguinity. This suggests that a much more distant inbreeding loop leads to this HBS tract and that the first-cousin inbreeding loop is coincidental. The sharing of haplotypes in the cohort of subjects, and that some individuals share several megabases, suggests founder events for both haplotypes; with evidence of shorter shared haplotypes, HBS, and a slightly higher frequency than that of the p.Gly212Val variant, p.Asn134Ilefs∗2 is likely to be the older founder event. Extensive functional characterization of the identified TMEM126B variants was undertaken in muscle and fibroblast cell lines obtained from subjects 1–3. Blue native PAGE (BN-PAGE) analysis of fibroblasts from affected subjects revealed a marked reduction of fully assembled complex I in supercomplex form (Figure 2A) or holoenzyme form (Figure 2B) in subjects 2 and 3, who harbored a truncating mutation in trans with a p.Gly212Val missense variant. Conversely, complex I assembly was normal in fibroblasts from subject 1, suggesting an ability to function despite the biallelic p.Gly212Val variants (Figure S1). The accumulation of subcomplexes containing NDUFS3 in subjects 2 and 3 indicates that the matrix module containing NDUFS3 is made but is unable to be added to the membrane arm. SDS-PAGE and immunoblot analysis of select complex I subunits revealed strongly reduced levels in fibroblasts from subject 2 and 3, but not subject 1 (Figure 2C). Subsequent BN-PAGE analysis of muscle from subject 1 revealed severely diminished levels of fully assembled complex I (Figure 2D).26Oláhová M. Hardy S.A. Hall J. Yarham J.W. Haack T.B. Wilson W.C. Alston C.L. He L. Aznauryan E. Brown R.M. et al.LRPPRC mutations cause early-onset multisystem mitochondrial disease outside of the French-Canadian population.Brain. 2015; 138: 3503-3519Crossref PubMed Scopus (60) Google Scholar These results support a deleterious effect and recapitulate the biochemical enzyme assays in which markedly decreased complex I levels were observed in fibroblasts from compound-heterozygous subjects, whereas the fibroblasts from subject 1 retained complex I activities within the normal range (Figure 2E). Functional analysis of fibroblasts and muscle biopsy from additional individuals, notably subjects 4–6, revealed similar patterns of pathology (Supplemental Data). Two-dimensional BN-PAGE of mitochondria-enriched pellets from muscle biopsy of subjects 4 and 5 revealed a marked reduction of complex I subunits, whereas other complexes remained intact (Figure S2). Double immunofluorescence staining of fibroblasts from subjects 4 and 5 (Figure S3) or subject 6 (Figure S4) revealed decreased signal of TMEM126B directly (subjects 4 and 5) or clear evidence of reduced signal of complex I subunits in the case of subject 6 (NDUFS4 was used as a surrogate marker of complex I signal) in comparison to age-matched control subjects. Most noteworthy is the observation of a complex I biochemical defect in the cells from subject 6, who like subject 1, was homozygous for the p.Gly212Val variant yet presented much earlier in life with a more severe clinical phenotype (Table 1). To provide further evidence that TMEM126B mutations are causative, we performed cellular rescue with