ORIGINAL RESEARCH published: 14 June 2019 doi: 10.3389/fgene.2019.00576 Identification and Characterization of New RNASEH1 Mutations Associated With PEO Syndrome and Multiple Mitochondrial DNA Deletions Lidia Carreño-Gago 1,2 , Cora Blázquez-Bermejo 1,2 , Jordi Díaz-Manera 2,3 , Yolanda Cámara 1,2 , Eduard Gallardo 2,3 , Ramon Martí 1,2 , Javier Torres-Torronteras 1,2* † and Elena García-Arumí 1,2,4* † 1 Departament de Patologia Mitocondrial i Neuromuscular, Hospital Universitari Vall d’Hebron Institut de Recerca (VHIR), Edited by: Universitat Autònoma de Barcelona, Barcelona, Spain, 2 Centro de Investigación Biomédica en Red de Enfermedades Raras Enrico Baruffini, (CIBERER), Instituto de Salud Carlos III, Barcelona, Spain, 3 Servei de Neurologia, Malalties Neuromusculars, Hospital de la University of Parma, Italy Santa Creu i Sant Pau i Institut de Recerca de HSCSP, Universitat Autònoma de Barcelona, Barcelona, Spain, 4 Àrea de Genètica Clínica i Molecular, Hospital Universitari Vall d’Hebron, Barcelona, Spain Reviewed by: Jan-Willem Taanman, University College London, Mitochondrial DNA (mtDNA) depletion and deletion syndrome encompasses a group of United Kingdom Alejandro Horga, disorders caused by mutations in genes involved in mtDNA replication and maintenance. Servicio de Neurología, Hospital The clinical phenotype ranges from fatal infantile hepatocerebral forms to mild adult Clínico San Carlos, Spain onset progressive external ophthalmoplegia (PEO). We report the case of a patient *Correspondence: with PEO and multiple mtDNA deletions, with two new homozygous mutations in Javier Torres-Torronteras javier.torres@vhir.org RNASEH1. The first mutation (c.487T>C) is located in the same catalytic domain as Elena García-Arumí the four previously reported mutations, and the second (c.258_260del) is located in elena.garcia@vhir.org † These the connection domain, where no mutations have been reported. In silico study of authors share senior authorship the mutations predicted only the first mutation as pathogenic, but functional studies showed that both mutations cause loss of ribonuclease H1 activity. mtDNA replication Specialty section: dysfunction was demonstrated in patient fibroblasts, which were unable to recover This article was submitted to Genetic Disorders, normal mtDNA copy number after ethidium bromide-induced mtDNA depletion. Our a section of the journal results demonstrate the pathogenicity of two new RNASEH1 variants found in a patient Frontiers in Genetics with PEO syndrome, multiple deletions, and mild mitochondrial myopathy. Received: 21 February 2019 Accepted: 31 May 2019 Keywords: mtDNA, mitochondrial disease, PEO, multiple mtDNA deletions, RNASEH1 Published: 14 June 2019 Citation: Carreño-Gago L, INTRODUCTION Blázquez-Bermejo C, Díaz-Manera J, Cámara Y, Gallardo E, Martí R, The mitochondrion is a cellular organelle crucial for cell metabolism that integrates various Torres-Torronteras J and metabolic pathways, including oxidative phosphorylation (OXPHOS), fatty acid oxidation, Krebs García-Arumí E (2019) Identification cycle, urea cycle, gluconeogenesis, and ketogenesis (Gorman et al., 2016). Mitochondrial function and Characterization of New is under the control of two genomes, nuclear DNA (nDNA) and mitochondrial DNA (mtDNA). RNASEH1 Mutations Associated With PEO Syndrome and Multiple mtDNA is a 16.6-kb circular and double-stranded DNA molecule that encodes 2 rRNAs, 22 tRNAs, Mitochondrial DNA Deletions. and 13 peptides from OXPHOS complexes I, III, IV, and V. However, most mitochondrial proteins Front. Genet. 10:576. are encoded by nDNA genes (Anderson et al., 1981). Nuclear-encoded mitochondrial proteins doi: 10.3389/fgene.2019.00576 are synthetized in the cytosol and transported into the mitochondria where they perform their Frontiers in Genetics | www.frontiersin.org 1 June 2019 | Volume 10 | Article 576 Carreño-Gago et al. RNASEH1 Mutations Causing PEO Syndrome functions (Herrmann et al., 2012). A large number of these MATERIALS AND METHODS mitochondrial proteins are OXPHOS complex subunits or complex assembly factors and therefore, they are directly All the experimental protocols were performed with involved in the electron transport chain and energy production appropriate informed consent and approval of the Clinical in the form of ATP. Furthermore, a group of nuclear genes are Research Ethics Committee of the Hospital Universitari Vall involved in several mitochondrial processes, such as mtDNA d’Hebron (PR(IR)66/2016). maintenance, transcription and translation, and control of mitochondrial dynamics, which are also essential for proper Histopathological Studies OXPHOS function. When a malfunction occurs in one of A skeletal muscle biopsy from the left biceps was performed. these proteins, the mitochondrial OXPHOS system is unable Five-micron sections of frozen muscle were double-stained to produce sufficient energy to maintain cell requirements, with cytochrome c oxidase (COX) and succinate dehydrogenase and this deficit leads to a number of mitochondrial diseases (SDH) (Dubowitz and Sewry, 2007). (Wallace, 1999). Therefore, mitochondrial diseases are caused by mutations in mtDNA genes and in a very long list of nuclear mitochondrial-related genes (Cerritelli et al., 2003; Ruhanen Cell Culture Conditions et al., 2011; Uhler and Falkenberg, 2015). Patient fibroblasts and fibroblasts from healthy age-matched Multiple mtDNA genome copies are present in cells to provide donors were obtained by skin biopsy and used in the study. the components for energy production. mtDNA replication is Fibroblasts were grown in high glucose (4.5 g/L) DMEM unsynchronized with the cell cycle as it occurs in dividing (Gibco) supplemented with 10% FBS, 200 mM L-glutamine, and postmitotic cells. Correct mtDNA maintenance requires 100 mM sodium pyruvate 100 U/mL penicillin and 0.1 mg/mL adequate expression of a number of nuclear-encoded proteins streptomycin. In order to force OXPHOS for ATP production, in involved in mtDNA replication, mitochondrial deoxynucleotide some experiments cells were grown in DMEM without glucose pool synthesis, mitochondrial dynamics, and other proteins (Gibco) supplemented with 10% FBS, 1 g/L galactose, 200 mM with uncertain function. Changes in these genes are associated L -glutamine, 100 mM sodium pyruvate, 100 U/mL penicillin and with secondary mtDNA changes, such as mtDNA deletions 0.1 mg/mL streptomycin. or depletion, which cause OXPHOS dysfunction and result in To induce mtDNA depletion in fibroblasts, the same number a group of disorders known as mitochondrial depletion and of cells was seeded and cultured to confluence (day 0). Fibroblasts deletion syndromes (MDDSs). The clinical phenotype associated were then treated with 15 ng/mL ethidium bromide (EtBr) for with MDDSs ranges from fatal infantile hepatocerebral forms 4 days. After EtBr was withdrawn (day 4), cells were kept in to mild adult-onset syndromes, such as progressive external culture media for 10 additional days (day 14). Cells were collected ophthalmoplegia (PEO) (Suomalainen and Kaukonen, 2001; on days 0, 2, 4, 7, 9, and 14, and maintained frozen at −20◦ C until El-Hattab et al., 2017). further DNA extraction. The human RNASEH1 gene encodes an endonuclease involved in DNA replication in both the nucleus and Genetic Studies mitochondria. Ribonuclease H1 (RNase H1) is the only We sequenced the exonic and intron flanking regions of known ribonuclease of this type in the mitochondria 17 nuclear genes (DGUOK, DNA2, FBXL4, MGME1, MFN2, (Al-Behadili et al., 2018; Rusecka et al., 2018) and, in particular, MPV17, OPA1, POLG, POLG2, RNASEH1, RRM2B, SLC25A4, this enzyme is implicated in the RNA degradation of the SUCLA2, SUCLG1, SPG7, TK2, and TWNK) involved in mtDNA double-stranded RNA-DNA hybrid created to prime DNA replication and maintenance by next-generation sequencing, synthesis during mtDNA replication (Cerritelli et al., 2003; using a previously designed panel with GeneRead Custom Panel Ruhanen et al., 2011; Uhler and Falkenberg, 2015; Al-Behadili V.2 (Qiagen) technology. Libraries were prepared with the et al., 2018). RNASEH1 mutations have been described in NEBNext Ultra II DNA Library Prep Kit for Illumina (New patients with clinical MDDS associated with adult-onset PEO England Biolabs), sequenced on the MiSeq platform (Illumina) and multiple mtDNA deletions (Reyes et al., 2015; Bugiardini and analyzed with GeneRead Targeted Enrichment Exon Panel et al., 2017; Sachdev et al., 2018). The predominant clinical Data Analysis (Qiagen) software. The mutations identified were traits of patients with RNASEH1 mutations are PEO, ptosis, verified by conventional Sanger sequencing. dysphagia, facial and/or proximal weakness, ataxia, and respiratory impairment. To date, only four mutations in RNASEH1, all occurring in Quantification of mtDNA Copy Number the catalytic RNase H1 domain, have been identified in a total and Analysis of Multiple Deletions of 14 patients with a mild clinical phenotype (Reyes et al., 2015; Total DNA was isolated from fibroblast pellets or muscle biopsies Bugiardini et al., 2017; Sachdev et al., 2018). Here, we report the with the QIAamp DNA Mini kit (Qiagen). The mtDNA copy case of a patient diagnosed with mild mitochondrial myopathy number was quantified by real-time PCR in the ABI PRISM 7500 characterized by PEO and multiple mtDNA deletions caused by Sequence Detection System (Applied Biosystems). mtDNA was two previously unreported mutations in the RNASEH1 gene, one detected using a custom-designed TaqMan probe and primers for located in the catalytic domain and the other in the RNase H1 the 12S rRNA gene (Andreu et al., 2009), and nuclear DNA using connection domain. the TaqMan RNase P Control Reagent kit (ThermoFisher) in a Frontiers in Genetics | www.frontiersin.org 2 June 2019 | Volume 10 | Article 576 Carreño-Gago et al. RNASEH1 Mutations Causing PEO Syndrome multiplex reaction with the TaqMan Universal PCR Master Mix RNase H1 Activity Assay with UNG (Applied Biosystems). The RNase H1 activity assay is based on the capability of The presence of multiple mtDNA deletions was analyzed by recombinant protein to degrade RNA using a radiolabeled long-range PCR (LPCR) (Nishigaki et al., 2004). RNA-DNA heteroduplex as substrate (Wu et al., 2001; Alla and Nicholson, 2012). For heteroduplex generation, we labeled Western Blot the 50 -end RNA oligo with 32 P using [ϒ −32P ]ATP. The Western blot was used to analyze the protein extracts obtained labeling reaction was performed by incubating 200 pmol RNA from the fibroblast mitochondria-enriched fractions using oligonucleotide (50 -GAAAUACGGUCCGAAACGUUG-30 ) with the Mitochondria Isolation Kit for Cultured Cells (Abcam) 3 pmol [ϒ −32P ]ATP (3000 Ci/mmol), 290 pmol ATP, and following the manufacturer’s instructions. Mitochondrial pellets 10 units of T4 polynucleotide kinase in 25 µl of T4 buffer were resuspended in the resuspension reagent provided by (10 mM MgCl2 , 5 mM DTT and 70 mM Tris–HCl, pH 7) the supplier, supplemented with a protease inhibitor cocktail for 1 h at 37◦ C. The 32 P-labeled RNA oligonucleotide was (cOmplete EDTA-free, Roche). Samples were heat-denatured in then purified with Quick Spin Columns for Radiolabeled RNA loading buffer (60 mM Tris–HCl pH6.8, 20% glycerol, 20% SDS, Purification (Roche). For RNA/DNA heteroduplex annealing, 5% β- mercaptoethanol, 0.05% bromophenol blue) and separated 1 µM 32 P-labeled RNA was incubated with 100 µM DNA by 12% SDS-PAGE. Proteins were transferred to Immun- oligonucleotide (50 -CAACGTTTCGGACCGTATTTC-30 ), 0.5 M Blot PVDF membranes (Bio-Rad) and detected using primary KCl, 250 mM HEPES, and 0.1M EDTA at 90◦ C for 5 min, slowly mouse monoclonal antibodies against human RNase H1 protein cooled-down to 37◦ C and then placed on ice. (ab56560, Abcam) and rabbit polyclonal antibodies against The enzyme activity assay was performed by incubating succinate dehydrogenase (SDHA; NB-22-14256, Neo Biotech). 250 nM 32 P-labeled DNA/RNA hybrid with 40 nM of recombinant RNase H1 protein and reaction buffer (0.5 M KCl, 500 nM DTT, 250 nM HEPES and 500 µM MgCl2 ) Recombinant Human RNase H1 Protein for 1 h at 30◦ C. The reaction was stopped by adding an Production equal volume of stopping solution [95% (v/v) formamide, and We produced 4 different plasmids containing the following 20 mM EDTA]. Samples were heated at 90◦ C for 2 min and RNASEH1 cDNA sequences: single mutant c.487T>C then resolved on 15% polyacrylamide gel with 7 M urea and [p.(Tyr163His)], single mutant c.258_260del [p.(Gln86del)], TBE buffer. The gel was dried and the labeled bands were double mutant c.487T>C/c.258_260del [p.(Tyr163His)/ visualized by autoradiography. p.(Gln86del)] and wild-type (WT). We obtained the various RNase H1 proteins as previously described (Cerritelli and Statistics and Quantification Crouch, 1998). Briefly, total RNA was extracted from patient Data were analyzed using the GraphPad PRISM software and control fibroblasts using the RNeasy Mini kit (Qiagen). (GraphPad software) and are presented as the mean ± SD. The double mutant and WT RNASEH1 cDNAs were generated The protein and radiolabeled RNA bands were quantified by using the High-capacity cDNA Reverse Transcription kit densitometry using ImageJ software (NIH). (Applied Biosystems), amplified by specific PCR using the Expand High Fidelity PCR System (Roche) with a specific pair of primers (forward 50 -ATGTTCTATGCCGTGAGGAG-30 RESULTS and reverse 50 -TCAGTCTTCCGATTGTTTAGC-30 ) and cloned into the pCR 2.1 TOPO TA vector (Invitrogen). Case Report The single mutant cDNAs were obtained by site-directed A 53-year-old patient was consulted for progressive palpebral mutagenesis of the WT cDNA using the Q5 Site-Directed ptosis without diplopia of 2 years duration, associated with Mutagenesis kit (New England Biolabs). Two pairs of a feeling of early fatigue with exercise, mild dysphagia and primers were used to introduce the c.487T>C point mutation dysarthria. Physical examination showed moderate, fatigue- (forward 50 -AATCGGCGTTCACTGGGGGCCA-30 and reverse unrelated palpebral ptosis, associated with limitations in bilateral 50 -CCTGCTCGCGGCCTTCTACGC-30 ) and the c.258_260del ocular movement. No significant muscle weakness was observed, deletion (forward 50 -TGGACAAGAATCGGAGGCGAAA-30 except for a slight lingual weakness. The patient had slightly and reverse 50 -TGATTTTCATGC CCTTCTGAAACTTCC-30 ). elevated serum creatine kinase concentration (<500 U/L). We confirmed the absence of non-specific mutations by direct Examination of biceps muscle biopsy showed some ragged Sanger sequencing. red fibers and abundant COX-negative fibers, consistent with Each RNASEH1 cDNA (WT, double mutant, mutant mitochondrial myopathy (Figure 1A). No fiber size variability or c.487T>C, and mutant c.258_260del) was subcloned into central nuclei were seen. LPCR of DNA extracted from muscle the pET-15b expression vector (Novagen) and transformed biopsy revealed multiple deletions (Figure 1B) suggestive of into Escherichia coli One Shot BL21(DE3) pLysS chemically mtDNA instability syndrome, and real-time PCR showed normal competent cells (Invitrogen). Recombinant protein synthesis was mtDNA copy number (Figure 1C). induced in the bacterial cell culture with 1 mM IPTG, collected Three years after the first visit there was little disease after cell lysis and purified using the Ni-NTA Fast Start Kit progression, with the exception of respiratory muscle weakness (Qiagen) based on histidine tag affinity columns. requiring non-invasive nocturnal ventilation. Frontiers in Genetics | www.frontiersin.org 3 June 2019 | Volume 10 | Article 576 Carreño-Gago et al. RNASEH1 Mutations Causing PEO Syndrome FIGURE 1 | Histological and molecular findings in muscle biopsy specimens. (A) COX/SDH double staining of the patient’s muscle biopsy. The yellow scale bar represents 200 µm. Arrows show COX-negative fibers. (B) LPCR performed in DNA from the patient (P) and control (C) muscle samples. After electrophoresis, mtDNA deletions are seen as multiple bands in the agarose gel. (C) mtDNA copy number measured by qRT-PCR in skeletal muscle samples from patient (P) and controls (C1, C2, C3, and C4). Results are expressed as the ratio between the 12S mtDNA gene and the RNase P nuclear gene. Bars represent the mean (+SD, n = 3). Genetic and Molecular Studies Characterization of the Mutation in Skin DNA sequencing from muscle biopsy specimens with the Fibroblasts custom panel V2 (Qiagen) disclosed 2 potentially pathogenic RNase H1 steady state levels were measured in skin fibroblast variants in homozygosis in the RNASEH1 gene: c.258_260del, mitochondrial protein extracts from the patient and a controls. a three-nucleotide in-frame deletion in exon 3 causing a Western blot analysis showed a marked RNase H1 reduction in glutamine residue deletion [p.(Gln86del)], and c.487T>C, the patient’s sample: residual protein level was only 49% of the a single nucleotide change in exon 4 causing a missense mean levels in control fibroblasts (Figure 3A). mutation [p.(Tyr163His)] (NM_002936.3, NP_002927.2) We compared the capacity for growth of patient fibroblasts (Figures 2A,B). and two control cell lines. Cells were cultured in medium with First, we analyzed the population frequency of the variants, glucose for 12 days with regular medium changes, and cell and found that both were extremely rare; c.258_260del was counts were performed at several time points. No differences in found in only 35 of 276096 alleles (0.013%), none in growth were observed between the patient and control cell lines homozygosis in the gnomAD database, and the c.487T>C (Figure 3B). In order to force cells to preferentially use oxidative variant had not been previously reported in the databases phosphorylation for ATP production, we grew cells in culture consulted. To study the implication of these mutations in medium supplemented with galactose instead of glucose, and the patient’s phenotype, we performed an amino acid residue observed an expected growth rate reduction in all cell lines, with conservation study and in silico prediction of pathogenicity. no differences between control and patient cells (Figure 3B). The conservation studies showed that the glutamine residue Mitochondrial DNA content and integrity was analyzed in at position 86 in RNase H1 protein is only conserved in patient fibroblasts, but neither deletions nor mtDNA depletion primates, whereas the tyrosine residue at position 163 and were detected. As the fibroblasts did not show a spontaneous the surrounding region was conserved in all species analyzed mtDNA phenotype, we induced mtDNA depletion in samples (Figure 2B). The in silico pathogenicity studies performed from the patient and two controls. To this end, we treated with on-line open tools classified the c.487T>C variant as fibroblasts with EtBr and then assessed their ability to recover “probably damaging” by Polyphen (score 1), “damaging” by normal mtDNA copy number. mtDNA copy number had SIFT (score 0) and “disease causing” by Mutation Taster. For decreased at 2 and 4 days after EtBr treatment in all fibroblast the c.258_260del, Mutation Taster classified the variant as a lines (Figure 3C). At 5 days after EtBr removal, control fibroblasts polymorphism and the Alamut splicing predictor did not detect had recovered the initial mtDNA copy number, which continued an alternative splicing, although it classified the variant as of to increase up to the end of the experiment (10 days), whereas “unknown pathogenicity.” No results were obtained with the patient fibroblasts were unable to recuperate, showing mtDNA other tools, as these predictors are unable to classify small copy number at only 60% (Figure 3C). deletions or insertions. We measured RNase H1 enzyme activity with an in vitro assay using a DNA/RNA hybrid oligonucleotide as substrate DISCUSSION and the wild-type or mutated recombinant proteins containing each single mutation or a combination of both. Only the wild- Mitochondrial depletion and deletion syndromes are a group of type recombinant protein displayed RNase H1 activity. No mitochondrial disorders caused by mutations in nuclear genes residual activity was detected for any mutated recombinant associated with mtDNA instability. These conditions have a wide protein (Figure 2C). variety of clinical presentations ranging from a spectrum of Frontiers in Genetics | www.frontiersin.org 4 June 2019 | Volume 10 | Article 576 Carreño-Gago et al. RNASEH1 Mutations Causing PEO Syndrome lingual muscle weakness (Bugiardini et al., 2017). Histological and molecular studies in skeletal muscle found ragged red fibers, COX-defective fibers, and multiple mtDNA deletions, which are always present in patients with RNASEH1 mutations. Although the clinical features of our patient matched many of those reported in others with RNASEH1 mutations, the low prevalence of the disease and the large number of genes associated with these symptoms hampered the diagnosis. Our patient showed two mutations in RNASEH1, detected by high-throughput sequencing using a custom panel of genes involved in mtDNA replication and maintenance. High- throughput sequencing techniques have improved the diagnosis of mitochondrial diseases. In fact, all reported RNASEH1 mutations have been detected using target or whole-exome sequencing (Reyes et al., 2015; Bugiardini et al., 2017). The panel used in this study provided a fast and cost-effective way to establish the genetic diagnosis in a patient with a specific MDDS. However, the number of genes known to be involved in these phenotypes is growing, and next-generation sequencing techniques, such as whole exome sequencing, are sometimes needed to reach a definite genetic diagnosis. Maturation of mtDNA during replication involves RNase H1 and at least two additional proteins encoded by the MGME1 and DNA2 genes, which are also implicated in MDDSs with PEO (Kornblum et al., 2013; Ronchi et al., 2013; Viscomi and Zeviani, 2017). RNase H1 protein is essential for terminating replication by removing the replication primer, and defects in its function cause deceleration and stalling of mtDNA replication (Reyes et al., 2015). Here, we investigated the molecular effects of RNASEH1 mutations on the patient’s fibroblasts and found a marked decrease in RNase H1 protein in mitochondria, indicating that one or both variants affected protein stability. However, we did not observe multiple deletions or mtDNA FIGURE 2 | Genetic and enzymatic study. (A) Summary of the known RNASEH1 mutations. Location of the mutations in RNASEH1 and their depletion, which likely explains the normal growing rate seen distribution along the four different RNase H1 protein domains are indicated in the OXPHOS-obligatory culture conditions. This limitation with dashed lines (M, mitochondrial targeting sequence; HBD, hybrid binding of the model, commonly observed in cultured fibroblasts from domain; CD, connection domain; H-Domain, catalytic domain). The two new other MDDSs, is usually bypassed by assessing the ability to variants are boxed within the highlighted squares. (B) Amino acid restore normal mtDNA copy number following induced mtDNA conservation study. CLUSTAL online software amino acid alignment of the RNase H1 regions containing the new variants. The patient’s mutated residues depletion (Stewart et al., 2011; Kornblum et al., 2013; Dalla Rosa are shadowed and the conserved residues are indicated with an asterisk. et al., 2016). In our case, fibroblast mtDNA recovery was clearly (C) Enzyme activity of the wild type (WT), double mutated reduced, thus demonstrating that the pathogenesis of the disease (p.Tyr163His/p.Gln86del), p.Tyr163His, and p.Gln86del recombinant RNase was related to an mtDNA replication defect, in agreement with H1 proteins. The ssOligo corresponds to the uncut RNA chain of the DNA:RNA heteroduplex, whereas the ss processed Oligo is the digested RNA other cases involving RNASEH1 mutations (Holmes et al., 2015; oligomer. Reyes et al., 2015). The RNASEH1 mutations found were homozygous and neither had been previously described as disease-related. Genetic study of the mutations in the patient’s parents was severe early onset neurologic or other organ-specific syndromes not possible, but we know that they were consanguineous, to adult-onset milder encephalomyopathies characterized by which reduces the possibility of apparent homozygosity due PEO (Viscomi and Zeviani, 2017). One possible genetic cause to compound heterozygosity of the two mutations with an of the milder clinical form is the presence of mutations in overlapping long-range gene deletion (Landsverk et al., 2012; RNASEH1, an uncommon occurrence with only 14 patients Martins Rda et al., 2014; Shibata et al., 2015). We also knew that reported to date, all of them showing PEO as a clinical trait neither of the patient’s parents were affected, which concurs with (Reyes et al., 2015; Bugiardini et al., 2017). Here, we report the previously reported autosomal recessive inheritance of the the case of a patient with PEO and additional clinical features disease (Reyes et al., 2015; Bugiardini et al., 2017). commonly reported in other patients with RNASEH1 mutations, In vitro study of recombinant protein function carrying such as dysphagia, dysarthria, respiratory impairment, and slight the two mutations demonstrated a complete loss of activity, Frontiers in Genetics | www.frontiersin.org 5 June 2019 | Volume 10 | Article 576 Carreño-Gago et al. RNASEH1 Mutations Causing PEO Syndrome FIGURE 3 | Functional studies in patient fibroblasts. (A) Western blot analysis of RNase H1 in mitochondria-enriched protein extracts from patient (P) and control (C1, C2, and C3) fibroblasts. The bar chart represents relative quantification of RNase H1 protein by densitometry. Results are expressed as the percentage of the mean (+SD, n = 3) RNase H1 protein density value obtained with control fibroblasts corrected with the SDHA protein load control. (B) Cell growth curves of fibroblasts incubated at 37◦ C in culture media supplemented with glucose (4.5 g/L) or galactose (1 g/L) during 12 days. The black lines correspond to control fibroblasts and the gray line to patient fibroblasts. Results are expressed as the mean fold increase with respect to the total cell count at time 0 (±SD, n = 3). (C) Recovery from mtDNA depletion in fibroblasts. Control (black lines) and patient (gray line) fibroblasts were incubated at 37◦ C in complete culture media supplemented with EtBr (15 ng/mL) during 4 days. EtBr was removed from culture media (dashed vertical line) and cells were maintained in culture during 10 additional days. DNA was extracted on days 0, 2, 4, 7, 9, and 14, and mtDNA copy number was quantified by qRT-PCR as the ratio between the 12S and RNase P gene copy numbers. Results are expressed as the mean mtDNA copy number referred to day 0, for each cell line (±SD, n = 2). which points to them as the cause of the disease. Our that the missense mutation located in the catalytic domain was functional studies also showed that each mutation separately pathogenic. However, pathogenicity prediction of the deletion resulted in complete loss of RNase H1 activity, suggesting located in the connection domain did not concur with the results that either mutation alone can cause the disease. These results obtained from the functional study. One potential explanation were partially consistent with our conservation study and our is that the deletion was predicted to be a polymorphism or of in silico prediction of pathogenesis, since all results indicated unknown pathogenicity because it was found in the population Frontiers in Genetics | www.frontiersin.org 6 June 2019 | Volume 10 | Article 576 Carreño-Gago et al. RNASEH1 Mutations Causing PEO Syndrome in heterozygosis at a frequency of 0.01% (gnomAD database). ETHICS STATEMENT Moreover, only two of the four predictors used were suitable for assessing small deletions and their predictions are based, All the experimental protocols were performed with appropriate among other factors, on the mutation frequency in population informed consent and approval of the Clinical Research databases, evolutionary conservation, splice-site changes, and Ethics Committee of the Hospital Universitari Vall d’Hebron loss of protein features (Schwarz et al., 2014), thus highlighting (PR(IR)66/2016). the importance of using various pathogenic predictors and functional validation to study new variants. Finally, both variants can be classified as “likely pathogenic” following the American AUTHOR CONTRIBUTIONS College of Medical Genetics and Genomics (ACMG) guidelines for the interpretation of sequence variants because c.478T>C LC-G, CB-B, JD-M, EG, and JT-T acquired the data. LC-G, meets the pathogenicity criteria PS3, PM1, PM2, and PP3 and CB-B, JD-M, EG, YC, RM, JT-T, and EG-A analyzed and c.258_260del meets the pathogenicity criteria PS3 and PM4 interpreted the data. LC-G, JD-M, EG, JT-T, and EG-A drafted (Richards et al., 2015). the manuscript. LC-G, YC, RM, JT-T, and EG-A critically revised the manuscript for important intellectual content. LC-G, JT-T, and EG-A approved the final version of the manuscript CONCLUSION to be published. In conclusion, our data support the pathogenicity of the two RNASEH1 variants found in a patient with PEO FUNDING syndrome, multiple deletions, and mild mitochondrial myopathy, phenotypic traits that are characteristic of other patients with This work was supported by the Spanish Instituto de Salud mutations in the same gene. Carlos III, Fondo de Investigaciones Sanitarias and cofunded with ERDF funds (Grant No. FIS PI15/01428 to EG-A and FIS PI18/01574 to RM), the Spanish Ministry of Industry, DATA AVAILABILITY Economy and Competitiveness (Grant No. SAF2017-87506-R to YC), and the Generalitat de Catalunya (a grant from the The raw data supporting the conclusions of this manuscript will URDCat project PERIS to EG-A and RM). JT-T was funded be made available by the authors, without undue reservation, to by a fellowship granted by the Generalitat de Catalunya (PERIS any qualified researcher. program, SLT002/16/00370). REFERENCES Dubowitz, C., and Sewry, C. A. (2007). Muscle Biopsy a Practical Approach. Philadelphia, PA: Elsevier. Al-Behadili, A., Uhler, J. P., Berglund, A. K., Peter, B., Doimo, M., Reyes, A., et al. El-Hattab, A. W., Craigen, W. J., and Scaglia, F. (2017). Mitochondrial DNA (2018). A two-nuclease pathway involving RNase H1 is required for primer maintenance defects. Biochim. Biophys. Acta 1863, 1539–1555. doi: 10.1016/ removal at human mitochondrial OriL. Nucleic Acids Res. 46, 9471–9483. doi: j.bbadis.2017.02.017 10.1093/nar/gky708 Gorman, G. S., Chinnery, P. F., Dimauro, S., Hirano, M., Koga, Y., Mcfarland, Alla, N. R., and Nicholson, A. W. (2012). Evidence for a dual functional role of a R., et al. (2016). Mitochondrial diseases. Nat. Rev. Dis. Primers 2:16080. conserved histidine in RNA.DNA heteroduplex cleavage by human RNase H1. doi: 10.1038/nrdp.2016.80 FEBS J. 279, 4492–4500. doi: 10.1111/febs.12035 Herrmann, J. M., Longen, S., Weckbecker, D., and Depuydt, M. (2012). Biogenesis Anderson, S., Bankier, A. T., Barrell, B. G., De Bruijn, M. H., Coulson, A. R., of mitochondrial proteins. Adv. Exp. Med. Biol. 748, 41–64. doi: 10.1007/978- Drouin, J., et al. (1981). Sequence and organization of the human mitochondrial 1-4614-3573-0_3 genome. Nature 290, 457–465. Holmes, J. B., Akman, G., Wood, S. R., Sakhuja, K., Cerritelli, S. M., Moss, C., et al. Andreu, A. L., Martinez, R., Marti, R., and Garcia-Arumi, E. (2009). Quantification (2015). Primer retention owing to the absence of RNase H1 is catastrophic for of mitochondrial DNA copy number: pre-analytical factors. Mitochondrion 9, mitochondrial DNA replication. Proc. Natl. Acad. Sci. U.S.A 112, 9334–9339. 242–246. doi: 10.1016/j.mito.2009.02.006 doi: 10.1073/pnas.1503653112 Bugiardini, E., Poole, O. V., Manole, A., Pittman, A. M., Horga, A., Hargreaves, Kornblum, C., Nicholls, T. J., Haack, T. B., Scholer, S., Peeva, V., Danhauser, K., I., et al. (2017). Clinicopathologic and molecular spectrum of RNASEH1- et al. (2013). Loss-of-function mutations in MGME1 impair mtDNA replication related mitochondrial disease. Neurol. Genet. 3:e149. doi: 10.1212/NXG. and cause multisystemic mitochondrial disease. Nat. Genet. 45, 214–219. doi: 0000000000000149 10.1038/ng.2501 Cerritelli, S. M., and Crouch, R. J. (1998). Cloning, expression, and mapping of Landsverk, M. L., Douglas, G. V., Tang, S., Zhang, V. W., Wang, G. L., Wang, J., ribonucleases H of human and mouse related to bacterial RNase HI. Genomics et al. (2012). Diagnostic approaches to apparent homozygosity. Genet. Med. 14, 53, 300–307. doi: 10.1006/geno.1998.5497 877–882. doi: 10.1038/gim.2012.58 Cerritelli, S. M., Frolova, E. G., Feng, C., Grinberg, A., Love, P. E., and Crouch, R. J. Martins Rda, S., Fonseca, A. C., Acosta, F. E., Folescu, T. W., Higa, L. Y., Sad, (2003). Failure to produce mitochondrial DNA results in embryonic lethality in I. R., et al. (2014). Severe phenotype in an apparent homozygosity caused Rnaseh1 null mice. Mol. Cell 11, 807–815. doi: 10.1016/s1097-2765(03)00088-1 by a large deletion in the CFTR gene: a case report. BMC Res. Notes 7:583. Dalla Rosa, I., Camara, Y., Durigon, R., Moss, C. F., Vidoni, S., Akman, G., doi: 10.1186/1756-0500-7-583 et al. (2016). MPV17 loss causes deoxynucleotide insufficiency and slow DNA Nishigaki, Y., Marti, R., and Hirano, M. (2004). ND5 is a hot-spot for multiple replication in mitochondria. PLoS Genet. 12:e1005779. doi: 10.1371/journal. atypical mitochondrial DNA deletions in mitochondrial neurogastrointestinal pgen.1005779 encephalomyopathy. Hum. Mol. Genet. 13, 91–101. doi: 10.1093/hmg/ddh010 Frontiers in Genetics | www.frontiersin.org 7 June 2019 | Volume 10 | Article 576 Carreño-Gago et al. RNASEH1 Mutations Causing PEO Syndrome Reyes, A., Melchionda, L., Nasca, A., Carrara, F., Lamantea, E., Zanolini, A., Stewart, J. D., Schoeler, S., Sitarz, K. S., Horvath, R., Hallmann, K., Pyle, A., et al. et al. (2015). RNASEH1 mutations impair mtDNA replication and cause adult- (2011). POLG mutations cause decreased mitochondrial DNA repopulation onset mitochondrial encephalomyopathy. Am. J. Hum. Genet. 97, 186–193. rates following induced depletion in human fibroblasts. Biochim. Biophys. Acta doi: 10.1016/j.ajhg.2015.05.013 1812, 321–325. doi: 10.1016/j.bbadis.2010.11.012 Richards, S., Aziz, N., Bale, S., Bick, D., Das, S., Gastier-Foster, J., et al. (2015). Suomalainen, A., and Kaukonen, J. (2001). Diseases caused by nuclear genes Standards and guidelines for the interpretation of sequence variants: a joint affecting mtDNA stability. Am. J. Med. Genet. 106, 53–61. doi: 10.1002/ajmg. consensus recommendation of the american college of medical genetics and 1379 genomics and the association for molecular pathology. Genet. Med. 17, Uhler, J. P., and Falkenberg, M. (2015). Primer removal during mammalian 405–424. mitochondrial DNA replication. DNA Repair. 34, 28–38. doi: 10.1016/j.dnarep. Ronchi, D., Di Fonzo, A., Lin, W., Bordoni, A., Liu, C., Fassone, E., et al. 2015.07.003 (2013). Mutations in DNA2 link progressive myopathy to mitochondrial DNA Viscomi, C., and Zeviani, M. (2017). MtDNA-maintenance defects: syndromes and instability. Am. J. Hum. Genet. 92, 293–300. doi: 10.1016/j.ajhg.2012.12.014 genes. J. Inherit. Metab. Dis. 40, 587–599. doi: 10.1007/s10545-017-0027-5 Ruhanen, H., Ushakov, K., and Yasukawa, T. (2011). Involvement of DNA ligase Wallace, D. C. (1999). Mitochondrial diseases in man and mouse. Science 283, III and ribonuclease H1 in mitochondrial DNA replication in cultured human 1482–1488. doi: 10.1126/science.283.5407.1482 cells. Biochim. Biophys. Acta 1813, 2000–2007. doi: 10.1016/j.bbamcr.2011. Wu, H., Lima, W. F., and Crooke, S. T. (2001). Investigating the structure of human 08.008 RNase H1 by site-directed mutagenesis. J. Biol. Chem. 276, 23547–23553. doi: Rusecka, J., Kaliszewska, M., Bartnik, E., and Tonska, K. (2018). Nuclear genes 10.1074/jbc.m009676200 involved in mitochondrial diseases caused by instability of mitochondrial DNA. J. Appl. Genet. 59, 43–57. doi: 10.1007/s13353-017-0424-3 Conflict of Interest Statement: The authors declare that the research was Sachdev, A., Fratter, C., and Mcmullan, T. F. W. (2018). Novel mutation in the conducted in the absence of any commercial or financial relationships that could RNASEH1 gene in a chronic progressive external ophthalmoplegia patient. Can. be construed as a potential conflict of interest. J. Ophthalmol. 53, e203–e205. doi: 10.1016/j.jcjo.2018.01.005 Schwarz, J. M., Cooper, D. N., Schuelke, M., and Seelow, D. (2014). Copyright © 2019 Carreño-Gago, Blázquez-Bermejo, Díaz-Manera, Cámara, MutationTaster2: mutation prediction for the deep-sequencing age. Nat. Gallardo, Martí, Torres-Torronteras and García-Arumí. This is an open-access Methods 11, 361–362. doi: 10.1038/nmeth.2890 article distributed under the terms of the Creative Commons Attribution License Shibata, A., Sugiura, K., Suzuki, A., Ichiki, T., and Akiyama, M. (2015). Apparent (CC BY). The use, distribution or reproduction in other forums is permitted, provided homozygosity due to compound heterozygosity of one point mutation and an the original author(s) and the copyright owner(s) are credited and that the original overlapping exon deletion mutation in ABCA12: a genetic diagnostic pitfall. publication in this journal is cited, in accordance with accepted academic practice. No J. Dermatol. Sci. 80, 196–202. doi: 10.1016/j.jdermsci.2015.10.003 use, distribution or reproduction is permitted which does not comply with these terms. Frontiers in Genetics | www.frontiersin.org 8 June 2019 | Volume 10 | Article 576