Clin Genet 2014 Printed in Singapore. All rights reserved
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd CLINICAL GENETICS doi: 10.1111/cge.12402
Original Article
Antisense-mediated therapeutic pseudoexon skipping in TMEM165-CDG Yuste-Checa P., Medrano C., Gámez A., Desviat L.R., Matthijs G., Ugarte M., Pérez-Cerdá C., Pérez B. Antisense-mediated therapeutic pseudoexon skipping in TMEM165-CDG. Clin Genet 2014. © John Wiley & Sons A/S. Published by John Wiley & Sons Ltd, 2014 Deficiencies in glycosyltransferases, glycosidases or nucleotide-sugar transporters involved in protein glycosylation lead to congenital disorders of glycosylation (CDG), a group of genetic diseases mostly showing multisystem phenotype. Despite recent advances in the biochemical and molecular knowledge of these diseases, no effective therapy exists for most. Efforts are now being directed toward therapies based on identifying new targets, which would allow to treat specific patients in a personalized way. This work presents proof-of concept for the antisense RNA rescue of the Golgi-resident protein TMEM165, a gene involved in a new type of CDG with a characteristic skeletal phenotype. Using a functional in vitro splicing assay based on minigenes, it was found that the deep intronic change c.792+182G>A is responsible for the insertion of an aberrant exon, corresponding to an intronic sequence. Antisense morpholino oligonucleotide therapy targeted toward TMEM165 mRNA recovered normal protein levels in the Golgi apparatus of patient-derived fibroblasts. This work expands the application of antisense oligonucleotide-mediated pseudoexon skipping to the treatment of a Golgi-resident protein, and opens up a promising treatment option for this specific TMEM165-CDG. Conflict of interest
The authors have declared no conflicting interests.
P. Yuste-Checaa,b,c , C. Medranoa,b,c , A. Gámeza,b,c , L.R. Desviata,b,c , G. Matthijsd , M. Ugartea,b,c , C. Pérez-Cerdáa,b,c and B. Péreza,b,c a Centro de Diagnóstico de Enfermedades Moleculares, Centro de Biología Molecular-SO UAM-CSIC, Campus de Cantoblanco, Universidad Autónoma de Madrid, Madrid,Spain, b Centro de Investigación Biomédica en Red de Enfermedades Raras (CIBERER), IDIPaz, Madrid, Spain, c Instituto de Investigación Biomedica, IDIPaz, Madrid, and d Center for Human Genetics, University of Leuven, Leuven, Belgium
Key words: antisense therapy – congenital disorders of glycosylation – intronic mutation – splicing Corresponding author: B. Pérez, Centro de Diagnóstico de Enfermedades Moleculares, Centro de Biología Molecular UAM-CSIC, Universidad Autónoma Madrid, Madrid, Spain. Tel.: +34 91 497 48 68; fax: 34914974868; e-mail: [email protected] Received 20 February 2014, revised and accepted for publication 8 April 2014
The congenital disorders of glycosylation (CDG) are a group of genetic diseases characterized primarily by the impairment of protein and lipid glycosylation. They are caused by defects in proteins involved in the glycosylation pathway (glycosyltransferases, glycosidases, dolichol-and nucleotide sugar transporters and chaperones), or in proteins involved in Golgi trafficking (1, 2) or Golgi pH homeostasis (3). Proteins N-glycosylation defects comprise CDG-I (defects in the synthesis and transfer of the dolichol-linked oligosaccharide) and CDG-II (defective glycan processing) (4, 5). N-glycosylation is an essential post-translational protein modification. Many membrane proteins, receptors,
enzymes, hormones and transporters are glycoproteins. An impairment in the glycosylation pathways thus mostly results in multisystem failure with a broad range of clinical symptoms (6). Mutations that disrupt cis-acting elements required for pre-mRNA splicing represent about 15% of all known point mutations reported to the Human Gene Mutation Database (https://portal.biobase-international. com/cgi-bin/portal/login.cgi). Most are located within conserved splicing sequences, mostly the consensus 5′ and 3′ donor sites at exon–intron junctions, or in branch point sequences. However, some mutations are located in other intronic or exonic regions, thus generating new 5′
1
Yuste-Checa et al. or 3′ cryptic splicing sites and causing aberrant splicing patterns. Frequently, the deep intronic mutations lead to the activation of so-called cryptic splice sites and to the inclusion of an intronic sequence (pseudoexon) which, under normal conditions, would not be recognized by the splicing machinery (7, 8). Generally, the inserted sequence jeopardizes the open reading frame, and generates a transcript that introduces a pre-mature termination codon. Such transcripts are susceptible to degradation by the non-sense-mediated decay machinery which prevents the generation of truncated proteins in the cell (9). Several examples of such pathogenic mutations have been described in rare diseases such as cystic fibrosis, ataxia telangiectasia, neurofibromatosis type I, organic acidemia, PMM2-CDG and tetrahydrobiopterin deficiencies. A diagnostic approach, that is only focused on the coding, exonic regions and flanking splice sites, does underestimate the frequency of deep intronic sequences, However, there are also many examples of antisense morpholino oligonucleotide (AMO) therapy preventing pseudoexon inclusion and the activation of exonic splice sites, allowing the rescue of the normal transcript and therefore the correct protein (10). Recently, a novel CDG-II was discovered, in a newly identified gene, TMEM165. TMEM165 is a hydrophobic protein of 324 amino acids containing seven predicted transmembrane-spanning domains. The protein is found within the Golgi compartment, plasma membrane and late endosomes/lysosomes, and is involved in calcium and pH homeostasis (11–13). Patients with a disrupted protein suffer an intellectual-disability syndrome and severe dysmorphism and skeletal disease. Three of the five reported patients, belonging to two different families, were found homozygous for the TMEM165 intronic change c.792 + 182G > A, while the other two carry two of the following missense mutations in compound heterozygous form: c.377G > A, c.376C > T or c.910G > A (11, 14). This work describes a functional analysis of the deep intronic change c.792 + 182G > A identified in the above three TMEM165-CDG patients. In addition, the use of antisense therapy for rescuing the Golgi-resident protein involved in this untreatable disorder (11) is shown to be successful. Materials and methods Patient- and control-derived fibroblasts
Fibroblasts were derived from a patient carrying the deep intronic change c.792 + 182G > A, from two patients carrying two of the following missense mutations: c.377G > A, c.376C > T or c.910G > A and from a healthy control. Fibroblast culture and transcriptional profiling
Control- and patient-derived fibroblast cell lines were cultured according to standard procedures in minimal essential medium (MEM) supplemented with 1% glutamine, 10% calf serum and antibiotics in a humidified
2
atmosphere containing 5% CO2 at 37∘ C. These cell lines were used as a source of mRNA, which was isolated using the MagnaPure system according to the manufacturer’s protocol (Roche Applied Sciences, Indianapolis, IN). Reverse transcription-polymerase chain reaction (RT-PCR) was then performed. cDNA was synthesized using the Superscript Vilo Kit (Invitrogen, Carlsbad, CA) following the manufacturer’s instructions, and amplified using the following primers (amplifying the entire cDNA sequence): forward 5′ GCTTGCCTTGGGACTAATGA 3′ and reverse 5′ AAGGGGTCAGTGCTGAAAGA 3′ . The primer located at the exon 4–exon 5 junction, 5′ AGCTAGAGAGGACCCCTATG 3′ , was used for amplifying a wild type patient’s transcript. Amplified products were separated by agarose gel electrophoresis and analyzed by direct sequencing. In some cases, the bands were cut out and purified using the QUIEX®II Gel Extraction Kit (Qiagen, Hilden, Germany) prior to sequencing. Messenger RNA quantification
The level of TMEM165 wild type transcripts in patient-derived fibroblast before and after AMO treatment was analyzed by quantitative (qRT-PCR) using a LightCycler®480 instrument. For mRNA isolation, cultured patient fibroblasts were harvested and mRNA obtained using TriPure Isolation Reagent (Roche Applied Sciences, Indianapolis, IN) following the manufacturer’s recommendations. All the RNA obtained was subjected to retrotranscription using the High-Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA), and amplified using the LightCycler®480 SYBR Green I Master Kit (Roche Applied Sciences). The forward primer directed to the exon 4–exon 5 junction, and the reverse primer, are the primers used for cDNA amplification. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA expression served as an endogenous expression control. PCR was performed in a total volume of 20 μl, containing 5 ng of cDNA product, 5 μl of LightCycler®480 SYBR Green I Master (Roche Applied Sciences), and PCR primers at a final concentration of 125 nM each. The data were analyzed using lightcycler® software (Roche Applied Sciences), correlating the initial template concentration with the cycle threshold (Ct) to obtain the relative quantity (RQ) of RNA. The RQ is defined as RQ = 2−ΔΔCt , where ΔΔCt is the ΔCt of the patient cell line minus the ΔCt of the control cell line, and ΔCt is the Ct of the target gene (TMEM165) minus the Ct of the housekeeping gene (GAPDH). Minigene construction and in vitro splicing analysis
pSPL3 vector was used to assess in vitro splicing. The gene fragment corresponding to the pseudoexon region, exon 4, and part of the intronic sequence (962 bp) (all in patient fibroblasts) were amplified using the forward primer, 5′ GCAGGAACTGTATTTTATAATTTTACC 3′ and the reverse primer 5′ CTCAACCTCCCAGGCTCA 3′ . This region was cloned into the TOPO vector (Invitrogen) and subsequently excised with Eco RI and cloned
Antisense-mediated therapeutic pseudoexon into the pSPL3 vector. A control construct was generated by PCR mutagenesis using the QuickChange Mutagenesis Kit (Agilent Technologies, Santa Clara, CA). Clones containing the desired normal and mutant inserts in the correct orientation were identified by sequencing. For minigene assays, 1.5 μg of the wild type or mutant pSPL3 construct were transfected into Hep3B human hepatoma cells using the JetPEI reagent (Polyplus Transfection, New York, NY), following the manufacturer’s recommendations. At 24 h post-transfection, cells were harvested and total RNA purified using TriPure Isolation Reagent (Roche Applied Sciences). RT-PCR was performed as described above, using the pSPL3-specific primers SD6 and SA2 (Exon Trapping System, Gibco BRL; Life Technologies, Grand Island, NY). The amplified products were separated by agarose gel electrophoresis and the excised bands further analyzed by direct sequencing after extraction using the QUIEX®II Gel Extraction Kit (Qiagen). Oligonucleotide treatment and analysis
The 25-mer AMO, 5′ GCTTGGTTACAAGAATTTTACCTGC 3′ , directed toward the cryptic 5′ splice site of the pseudoexon activated by the intronic mutation c.792 + 182G > A, was designed, synthesized and purified by Gene Tools (Philomath, OR). Endo-Porter reagent (Gene Tools) was used as the delivery mechanism according to the manufacturer’s instructions. The scrambled AMO sequence used as a negative control was 5′ AAGAGCGAGACTCTGTTTCAAAAAA 3′ . Patient fibroblasts were harvested 24 h after treatment with AMO (20 or 30 μM). RNA extraction and RT-PCR were performed as described above. Immunoblotting
Cells that had undergone or not undergone AMO treatment were harvested with trypsin after 48 h
and resuspended in PBS ×1 plus protease inhibitors (Roche Applied Sciences, Mannheim, Germany). Protein extracts were obtained by freeze–thaw cycles; concentrations were determined by Bio-Rad Protein Assay (Bio-Rad Laboratories, Munich, Germany). Samples were prepared in ×4 NuPage®LDS sample buffer, Dithiothreitol (DTT) and 50 μg of total protein extract, and were separated using 4–12% NuPAGE Novex Bis-Tris mini gels (Invitrogen). They were then transferred to a nitrocellulose membrane. These membranes were blocked 1 h at 4∘ C with phosphate saline buffer (PBS) containing 0.05% Tween-20 and 5% non-fat milk. Blots were incubated with anti-TMEM165 (Sigma-Aldrich, St. Louis, MO) diluted 1:1000 in blocking solution, and with anti-tubulin antibody (Sigma Aldrich) diluted 1:5000 in blocking solution. After incubation with the corresponding secondary antibodies (horseradish peroxide conjugated goat anti-rabbit 1:7500, and goat anti-mouse 1:10,000, respectively), protein bands were detected by ECL Western Blotting System. Protein analysis was performed using a calibrated GS-800 densitometer (Bio-Rad Laboratories). Immunofluorescence microscopy
Cells were grown on glass coverslips with or without AMO for 48 h and fixed with 10% formalin for 20 min at room temperature. The coverslips were rinsed twice with 0.1 M glycine in PBS for 15 min. The cells were then blocked with blocking solution (0.1% triton, 1% BSA, 20% fetal bovine serum) for 30 min. The fixed cells were incubated for 1 h at room temperature with primary antibodies, TMEM165 (Sigma Aldrich) and GM130 (BD Biosciences, San Jose, CA) diluted 1:300 in blocking solution. After washing for three times with PBS, anti-rabbit Alexa 488 and anti-mouse Alexa 594 secondary antibodies (Invitrogen) diluted 1:500 in blocking solution were applied for 1 h at room temperature. The cells were then washed with PBS and incubated with DAPI (Merck, Whitehouse Station, NJ) diluted 1:2500
Fig. 1. Transcriptional profile analysis of patient-derived-fibroblast carrying the c.792+182G>A intronic mutation in the TMEM165 gene, and functional analysis of this splicing mutation using minigenes. (a) Diagram showing the intronic sequence and transcript obtained in patient-derived fibroblasts via reverse transcription-polymerase chain reaction (RT-PCR) analysis. The sequence of the antisense morpholino oligonucleotide (AMO) is included. The intrinsic strength of the TMEM165 splice sites was estimated using the human splicing finder database (BDGP server: http://www.fruitfly.org). Transcriptional profile analysis of healthy control-derived fibroblasts (lane 1) and patient-derived fibroblasts (lane 2). (b) Transcriptional profile returned by ex vivo splicing assays using minigenes for the wild type (c.792+182G) and splicing mutation (c.792+182A). E: empty vector, V: vector exonic sequences.
3
Yuste-Checa et al. in PBS for 5 min. Samples were observed using an Axiovert200 inverted microscope and photographs taken using a C9100-02 digital camera (Hamamatsu, Hamamatsu City, Japan). Shadow and background correction were applied for quantification using fiji software. Statistical analysis
qRT-PCR and immunofluorescence microscopy results were analyzed using one-way anova and Bonferroni’s post hoc test. All statistical analyses were performed using IBM spss Statistics 21 for Windows.
Results
Transcriptional profile analysis of patient-derivedfibroblasts bearing the nucleotide change c.792 + 182G > A in homozygous fashion revealed the presence of two different transcripts. Sequence analysis indicated both to have a 117 bp sequence inserted into intron 4, plus either the deletion of exon 4 or of both exons 4 and 3 (Fig. 1a). The pseudoexon insertion is the result of the activation of a new 5′ cryptic splice site in intron 4. The intrinsic strength of the new 5′ cryptic splicing site increased from 0.5 in the wild type allele to 0.94 in the
Fig. 2. Antisense morpholino treatment in patient-derived fibroblast restores the normal splicing pattern as well as levels of the protein TMEM165. (a) Transcript profile analysis of control (lane 1), untreated patient-derived fibroblasts (lane 2), and patient-derived fibroblasts treated with 20 μM (lane 3) or 30 μM (lane 4) of antisense morpholino oligonucleotide (AMO) or a scrambled morpholino (lane 5). (b) Western blot analysis of control-derived fibroblast, untreated patient-derived fibroblast (−), and patient-derived fibroblasts treated with the AMO (20 μM) or scrambled morpholino (SCR). Not reaching statistically significance. (c) Indirect double immunofluorescence testing of control-derived fibroblasts (c) and fibroblasts from patients P1, P2 and P3 (genotypes indicated in the table in Fig. 2) untreated and treated with AMO (20 μM). Cells were double-labeled with TMEM165 (green fluorescence) and the Golgi marker GM130 (red fluorescence) antibodies. The figure shows the quantification of TMEM165 via the intensity of fluorescence recorded. Data was collected for two independent experiments; at least 60 photos were examined. *** p < 0.001.
4
Antisense-mediated therapeutic pseudoexon mutated allele (BDGP server: http://www.fruitfly.org). Because the normal transcript was not detected by RT-PCR in patient-derived fibroblasts, a complementary analysis was performed to check for its presence. An oligonucleotide comprising the sequence between exons 4 and 5 was designed to amplify just the wild type transcript. The result of RT-PCR with the specific primers showed the slight presence of wild type transcripts in patient fibroblasts (data not shown). The effect of the mutation c.792 + 182G > A on the splicing process was examined using an ex vivo splicing assay involving minigenes. The minigene constructs contained the normal and mutant pseudoexon sequences along with an exon 4 construct. Both the wild type and mutant minigenes were used to transfect Hep3B cells. After 24 h, the transcriptional profile was analyzed; it was found to be similar to that recorded for the mRNA of both the control- and patient-derived fibroblasts (Fig. 1b). Antisense therapy for preventing the insertion of the pseudoexon was tested using a specific AMO targeted toward the intronic 5′ cryptic splicing site (Fig. 1b). Patient-derived fibroblasts bearing the splicing mutation c.792 + 182G > A were transfected with two different concentrations of AMO (20 and 30 μM) and also with a scrambled AMO. After 24 h, the mRNA was analyzed by RT-PCR and the fragments obtained were sequenced. The results showed the complete restoration of the transcriptional profile in a dose and sequence-specific manner (Fig. 2a). To quantify the recovery of the wild type transcript in patient-derived fibroblast treated with AMO 20 μM, qRT-PCR was performed with the primers designed to amplify wild type transcripts in patient’s fibroblasts. Wild type transcripts were found to be present in the patient-derived fibroblasts, although only about 2% of that recorded for control-derived fibroblast. Significant (p < 0.001) recovery was achieved, however, using AMO at 20 μM; indeed, at this concentration it increased 10-fold compared to the basal expression and reached 20% of the presence recorded in controls (data not shown). To determine whether the recovery of the transcriptional profile achieved by AMO translated into protein rescue, Western blotting was performed. Patient-derived fibroblasts were treated with AMO (20 μM) and after 48 h protein analysis was performed. Western blotting revealed the presence of immunoreactive protein in untreated cells, and an increase of 3-fold in treated cells over non-treated cells (Fig. 2b). To quantify the protein recovery and determine the localization of the produced protein, an indirect double immunofluorescence assay was performed. Cells were double-labeled with antibodies against TMEM165 and the Golgi marker GM130. The results showed only basal protein expression in the Golgi of patient cells (around 20% compared to control-derived fibroblasts) but a significant increase of threefold (p < 0.001) after AMO treatment (Fig. 2c). No increase in TMEM65 protein was seen after AMO treatment in the patient-derived fibroblasts bearing the missense changes, confirming the protein restoration to be sequence-specific (Fig. 2c).
Discussion
Splicing defects represent the second most important pathogenic causes of genetic disease after missense changes (HGMD, https://portal.biobase-international. com). The most common mutations affecting the splicing process are those involving the 5′ and 3′ splice-site consensus sequences. Nevertheless, mutations outside of these may account for another significant fraction of pathogenic mutations, such as deep intronic mutations, an emerging cause of genetic diseases. Importantly, they provide a promising potential target for personalized treatment. After performing transcriptional profile analysis on patient-derived fibroblasts, functional analysis of nucleotide changes using ex vivo splicing is recommended to confirm the pathogenic effect of genomic changes (15, 16). Functional analysis of the deep intronic variant change c.792 + 182G > A using the minigene method showed the same transcription profile to be present in patient- and control-derived fibroblasts. These results confirm that the deep intronic mutation only activates the insertion of a 117 bp intronic pseudoexon between exons 4 and 5, leading to a putative truncated protein (if it is translated at all) caused by a pre-mature stop codon. Minigene analysis also showed that the insertion of the pseudoexon causes the additional skipping of exon 4 detected in patient-derived fibroblasts. These results are not easy to explain because the skipping of exon 4 cannot rescue the lost frame of the protein. From a therapeutic point of view, pseudoexons represent excellent targets for antisense-mediated ‘exon skipping’ therapy; the natural splice sites of the surrounding exons are intact, and the use of specific small oligonucleotide makes possible the rescue of the aberrant splicing process by forcing pseudoexon skipping. In this study, the aberrant TMEM165 mRNA was targeted using oligonucleotide technology. Wild type transcripts were restored in patient-derived fibroblast, reaching 20% of the level seen in control fibroblasts. These rescued transcripts were efficiently translated and produced 60% of protein seen in the control cells. Further, the protein was located in the Golgi apparatus where it is probably to be active. This therapeutic approach is confirmed to be mutation-specific because that it was unable to increase protein levels in patient-derived fibroblasts bearing missense changes. The treatment of cells with AMO-based therapy that prevents pathogenic splicing defects is a promising treatment for molecular lesions considered ‘undruggable’. To date, deep intronic mutations have been extensively described in different genes expressed in accessible tissues in which mRNA analysis is feasible (17–19). Recent advances in whole genome sequencing have revealed the presence of this type of mutation in many genes (20–22). It is probably that more of these types of pathogenic mutation will be detected, thanks to massively parallel sequencing of entire intronic sequences or RNA Seq, which should allow them to be detected directly in genomic DNA.
5
Yuste-Checa et al. Antisense therapy has been successfully used in cellular models of several genetic disorders, among them inherited metabolic diseases such as phosphomanomutase deficiency (PMM2-CDG), the most common congenital disorder of glycosylation (23). In all cases the splicing correction is dose- and sequence-specific; no obvious cell cytotoxicity has been observed (10). The challenge now is the safe and targeted delivery of the antisense therapy to specific tissues. Successful rescue ameliorating the disease phenotype in mouse models has been reported for Hutchinson–Gilford progeria syndrome, β-thalassemia, myotonic dystrophy and Duchenne muscular dystrophy (24, 25). Further advances in the medicinal chemistry of antisense oligonucleotides are, however, still required for optimizing antisense behavior in damaged tissues and organs. The next generation of chemically modified oligonucleotides is now being developed with the aim of improving their stability, affinity and delivery, and to enhance their pharmaceutical properties. Locked nucleic acids (LNA), peptide nucleic acid (PNA), 2′ -O-methoxyethyl RNA (2′ -MOE) and Vivo-Morpholino are being tested in cellular and animal models (25). In addition, a variety of cell-penetrating peptides directly conjugated to antisense oligonucleotides have been shown to enhance delivery in Duchenne model systems, to improve systemic distribution, and to be more effective than ’naked’ antisense oligonucleotides (26). In summary, progress in nanoparticle development, along with a new generation of disease models, will bring the bench closer to the bedside; this may prove important in disorders with no other available therapy. A clinical trial of a modified phosphorothioate oligonucleotide known as Vitravene (Formivirsen) for the treatment of cytomegalovirus-induced retinitis (27) has recently begun, and an antisense drug for the treatment of familiar hypercholesterolemia has been approved by the Food and Drug Administration (FDA) (28, 29). The therapeutic potential of antisense treatment depends, however, on the tissues and organs involved; each poses different challenges for targeted delivery. Owing to the pre-dominant clinical features of TMEM165-CDG are skeletal abnormalities, (growth retardation not responsive to growth hormone, osteoporosis, and skeletal dysplasia), antisense oligonucleotides should reach the bones. Systemic administration may therefore be sufficient to ameliorate disease outcome. In the absence of TMEM165-CDG animal models with specific splice defects amenable to antisense treatment, testing antisense therapies in induced, pluripotent, human stem cells might provide a model for evaluating antisense oligonucleotide-mediated pseudoexon skipping (30). In conclusion, this paper reports proof-of-concept for treating a Golgi-resident protein defect using mutation-specific antisense therapy. AMO therapy may be an excellent candidate for the treatment of disorders, for which no other treatments option is available.
6
Acknowledgements This work was funded by the Fondo de Investigaciones Sanitarias (PI10/00455 to B. P. and PI12/02078 to C. P. C.). P. Y. was supported by a grant from the Universidad Autónoma de Madrid. An institutional grant from the Fundación Ramón Areces to the Centro de Biología Molecular Severo Ochoa is gratefully acknowledged.
References 1. Wu X, Steet RA, Bohorov O et al. Mutation of the COG complex subunit gene COG7 causes a lethal congenital disorder. Nat Med 2004: 10 (5): 518–523. 2. Reynders E, Foulquier F, Leao Teles E et al. Golgi function and dysfunction in the first COG4-deficient CDG type II patient. Hum Mol Genet 2009: 18 (17): 3244–3256. 3. Kornak U, Reynders E, Dimopoulou A et al. Impaired glycosylation and cutis laxa caused by mutations in the vesicular H + -ATPase subunit ATP6V0A2. Nat Genet 2008: 40 (1): 32–34. 4. Jaeken J, Hennet T, Matthijs G et al. CDG nomenclature: time for a change!. Biochim Biophys Acta 2009: 1792 (9): 825–826. 5. Schachter H, Freeze HH. Glycosylation diseases: quo vadis? Biochim Biophys Acta 2009: 1792 (9): 925–930. 6. Jaeken J. Congenital disorders of glycosylation. Ann N Y Acad Sci 2010: 1214: 190–198. 7. Buratti E, Baralle M, Baralle FE. Defective splicing, disease and therapy: searching for master checkpoints in exon definition. Nucleic Acids Res 2006: 34 (12): 3494–3510. 8. Pagani F, Baralle FE. Genomic variants in exons and introns: identifying the splicing spoilers. Nat Rev Genet 2004: 5 (5): 389–396. 9. Maquat LE. Nonsense-mediated mRNA decay: splicing, translation and mRNP dynamics. Nat Rev Mol Cell Biol 2004: 5 (2): 89–99. 10. Perez B, Rodriguez-Pascau L, Vilageliu L, Grinberg D, Ugarte M, Desviat LR. Present and future of antisense therapy for splicing modulation in inherited metabolic disease. J Inherit Metab Dis 2010: 33 (4): 397–403. 11. Foulquier F, Amyere M, Jaeken J et al. TMEM165 deficiency causes a congenital disorder of glycosylation. Am J Hum Genet 2012: 91 (1): 15–26. 12. Demaegd D, Foulquier F, Colinet AS et al. Newly characterized Golgi-localized family of proteins is involved in calcium and pH homeostasis in yeast and human cells. Proc Natl Acad Sci U S A 2013: 110 (17): 6859–6864. 13. Rosnoblet C, Legrand D, Demaegd D et al. Impact of disease-causing mutations on TMEM165 subcellular localization, a recently identified protein involved in CDG-II. Hum Mol Genet 2013: 22 (14): 2914–2928. 14. Zeevaert R, de Zegher F, Sturiale L et al. Bone dysplasia as a key feature in three patients with a novel congenital disorder of glycosylation (CDG) type II due to a deep intronic splice mutation in TMEM165. JIMD Rep 2013: 8: 145–152. 15. Cooper TA. Use of minigene systems to dissect alternative splicing elements. Methods 2005: 37 (4): 331–340. 16. Desviat LR, Perez B, Ugarte M. Minigenes to confirm exon skipping mutations. Methods Mol Biol 2012: 867: 37–47. 17. Rincon A, Desviat LR, Ugarte M et al. Propionic and methylmalonic acidemia: antisense therapeutics for intronic variations causing aberrantly spliced mRNA. EMBO Conference Series "pre-mRNA processing and disease", Cortina d’Ampezzo, Italy, 2007. 18. Perez B, Rincon A, Jorge-Finnigan A et al. Pseudoexon exclusion by antisense therapy in methylmalonic aciduria (MMAuria). Hum Mutat 2009: 30 (12): 1676–1682. 19. Davis RL, Homer VM, George PM, Brennan SO. A deep intronic mutation in FGB creates a consensus exonic splicing enhancer motif that results in afibrinogenemia caused by aberrant mRNA splicing, which can be corrected in vitro with antisense oligonucleotide treatment. Hum Mutat 2009: 30 (2): 221–227. 20. Cavalieri S, Pozzi E, Gatti RA, Brusco A. Deep-intronic ATM mutation detected by genomic resequencing and corrected in vitro by antisense morpholino oligonucleotide (AMO). Eur J Hum Genet 2013: 21 (7): 774–778.
Antisense-mediated therapeutic pseudoexon 21. Webb TR, Parfitt DA, Gardner JC et al. Deep intronic mutation in OFD1, identified by targeted genomic next-generation sequencing, causes a severe form of X-linked retinitis pigmentosa (RP23). Hum Mol Genet 2012: 21 (16): 3647–3654. 22. Flanagan SE, Xie W, Caswell R et al. Next-generation sequencing reveals deep intronic cryptic ABCC8 and HADH splicing founder mutations causing hyperinsulinism by pseudoexon activation. Am J Hum Genet 2013: 92 (1): 131–136. 23. Vega AI, Perez-Cerda C, Desviat LR, Matthijs G, Ugarte M, Perez B. Functional analysis of three splicing mutations identified in the PMM2 gene: toward a new therapy for congenital disorder of glycosylation type Ia. Hum Mutat 2009: 30 (5): 795–803. 24. Havens MA, Duelli DM, Hastings ML. Targeting RNA splicing for disease therapy. Wiley Interdiscip Rev 2013: 4 (3): 247–266. 25. Osorio FG, Navarro CL, Cadinanos J et al. Splicing-directed therapy in a new mouse model of human accelerated aging. Sci Transl Med 2011: 3 (106): 106ra107.
26. Betts CA, Wood MJ. Cell penetrating peptide delivery of splice directing oligonucleotides as a treatment for Duchenne muscular dystrophy. Curr Pharm Des 2013: 19 (16): 2948–2962. 27. Roehr B. Fomivirsen approved for CMV retinitis. J Int Assoc Physicians AIDS Care 1998: 4 (10): 14–16. 28. Raal FJ, Santos RD, Blom DJ et al. Mipomersen, an apolipoprotein B synthesis inhibitor, for lowering of LDL cholesterol concentrations in patients with homozygous familial hypercholesterolaemia: a randomised, double-blind, placebo-controlled trial. Lancet 2010: 375 (9719): 998–1006. 29. Visser ME, Wagener G, Baker BF et al. Mipomersen, an apolipoprotein B synthesis inhibitor, lowers low-density lipoprotein cholesterol in high-risk statin-intolerant patients: a randomized, double-blind, placebo-controlled trial. Eur Heart J 2012: 33 (9): 1142–1149. 30. Dick E, Kalra S, Anderson D et al. Exon skipping and gene transfer restore dystrophin expression in human induced pluripotent stem cells-cardiomyocytes harboring DMD mutations. Stem Cells Dev 2013.
7
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd CLINICAL GENETICS doi: 10.1111/cge.12402
Original Article
Antisense-mediated therapeutic pseudoexon skipping in TMEM165-CDG Yuste-Checa P., Medrano C., Gámez A., Desviat L.R., Matthijs G., Ugarte M., Pérez-Cerdá C., Pérez B. Antisense-mediated therapeutic pseudoexon skipping in TMEM165-CDG. Clin Genet 2014. © John Wiley & Sons A/S. Published by John Wiley & Sons Ltd, 2014 Deficiencies in glycosyltransferases, glycosidases or nucleotide-sugar transporters involved in protein glycosylation lead to congenital disorders of glycosylation (CDG), a group of genetic diseases mostly showing multisystem phenotype. Despite recent advances in the biochemical and molecular knowledge of these diseases, no effective therapy exists for most. Efforts are now being directed toward therapies based on identifying new targets, which would allow to treat specific patients in a personalized way. This work presents proof-of concept for the antisense RNA rescue of the Golgi-resident protein TMEM165, a gene involved in a new type of CDG with a characteristic skeletal phenotype. Using a functional in vitro splicing assay based on minigenes, it was found that the deep intronic change c.792+182G>A is responsible for the insertion of an aberrant exon, corresponding to an intronic sequence. Antisense morpholino oligonucleotide therapy targeted toward TMEM165 mRNA recovered normal protein levels in the Golgi apparatus of patient-derived fibroblasts. This work expands the application of antisense oligonucleotide-mediated pseudoexon skipping to the treatment of a Golgi-resident protein, and opens up a promising treatment option for this specific TMEM165-CDG. Conflict of interest
The authors have declared no conflicting interests.
P. Yuste-Checaa,b,c , C. Medranoa,b,c , A. Gámeza,b,c , L.R. Desviata,b,c , G. Matthijsd , M. Ugartea,b,c , C. Pérez-Cerdáa,b,c and B. Péreza,b,c a Centro de Diagnóstico de Enfermedades Moleculares, Centro de Biología Molecular-SO UAM-CSIC, Campus de Cantoblanco, Universidad Autónoma de Madrid, Madrid,Spain, b Centro de Investigación Biomédica en Red de Enfermedades Raras (CIBERER), IDIPaz, Madrid, Spain, c Instituto de Investigación Biomedica, IDIPaz, Madrid, and d Center for Human Genetics, University of Leuven, Leuven, Belgium
Key words: antisense therapy – congenital disorders of glycosylation – intronic mutation – splicing Corresponding author: B. Pérez, Centro de Diagnóstico de Enfermedades Moleculares, Centro de Biología Molecular UAM-CSIC, Universidad Autónoma Madrid, Madrid, Spain. Tel.: +34 91 497 48 68; fax: 34914974868; e-mail: [email protected] Received 20 February 2014, revised and accepted for publication 8 April 2014
The congenital disorders of glycosylation (CDG) are a group of genetic diseases characterized primarily by the impairment of protein and lipid glycosylation. They are caused by defects in proteins involved in the glycosylation pathway (glycosyltransferases, glycosidases, dolichol-and nucleotide sugar transporters and chaperones), or in proteins involved in Golgi trafficking (1, 2) or Golgi pH homeostasis (3). Proteins N-glycosylation defects comprise CDG-I (defects in the synthesis and transfer of the dolichol-linked oligosaccharide) and CDG-II (defective glycan processing) (4, 5). N-glycosylation is an essential post-translational protein modification. Many membrane proteins, receptors,
enzymes, hormones and transporters are glycoproteins. An impairment in the glycosylation pathways thus mostly results in multisystem failure with a broad range of clinical symptoms (6). Mutations that disrupt cis-acting elements required for pre-mRNA splicing represent about 15% of all known point mutations reported to the Human Gene Mutation Database (https://portal.biobase-international. com/cgi-bin/portal/login.cgi). Most are located within conserved splicing sequences, mostly the consensus 5′ and 3′ donor sites at exon–intron junctions, or in branch point sequences. However, some mutations are located in other intronic or exonic regions, thus generating new 5′
1
Yuste-Checa et al. or 3′ cryptic splicing sites and causing aberrant splicing patterns. Frequently, the deep intronic mutations lead to the activation of so-called cryptic splice sites and to the inclusion of an intronic sequence (pseudoexon) which, under normal conditions, would not be recognized by the splicing machinery (7, 8). Generally, the inserted sequence jeopardizes the open reading frame, and generates a transcript that introduces a pre-mature termination codon. Such transcripts are susceptible to degradation by the non-sense-mediated decay machinery which prevents the generation of truncated proteins in the cell (9). Several examples of such pathogenic mutations have been described in rare diseases such as cystic fibrosis, ataxia telangiectasia, neurofibromatosis type I, organic acidemia, PMM2-CDG and tetrahydrobiopterin deficiencies. A diagnostic approach, that is only focused on the coding, exonic regions and flanking splice sites, does underestimate the frequency of deep intronic sequences, However, there are also many examples of antisense morpholino oligonucleotide (AMO) therapy preventing pseudoexon inclusion and the activation of exonic splice sites, allowing the rescue of the normal transcript and therefore the correct protein (10). Recently, a novel CDG-II was discovered, in a newly identified gene, TMEM165. TMEM165 is a hydrophobic protein of 324 amino acids containing seven predicted transmembrane-spanning domains. The protein is found within the Golgi compartment, plasma membrane and late endosomes/lysosomes, and is involved in calcium and pH homeostasis (11–13). Patients with a disrupted protein suffer an intellectual-disability syndrome and severe dysmorphism and skeletal disease. Three of the five reported patients, belonging to two different families, were found homozygous for the TMEM165 intronic change c.792 + 182G > A, while the other two carry two of the following missense mutations in compound heterozygous form: c.377G > A, c.376C > T or c.910G > A (11, 14). This work describes a functional analysis of the deep intronic change c.792 + 182G > A identified in the above three TMEM165-CDG patients. In addition, the use of antisense therapy for rescuing the Golgi-resident protein involved in this untreatable disorder (11) is shown to be successful. Materials and methods Patient- and control-derived fibroblasts
Fibroblasts were derived from a patient carrying the deep intronic change c.792 + 182G > A, from two patients carrying two of the following missense mutations: c.377G > A, c.376C > T or c.910G > A and from a healthy control. Fibroblast culture and transcriptional profiling
Control- and patient-derived fibroblast cell lines were cultured according to standard procedures in minimal essential medium (MEM) supplemented with 1% glutamine, 10% calf serum and antibiotics in a humidified
2
atmosphere containing 5% CO2 at 37∘ C. These cell lines were used as a source of mRNA, which was isolated using the MagnaPure system according to the manufacturer’s protocol (Roche Applied Sciences, Indianapolis, IN). Reverse transcription-polymerase chain reaction (RT-PCR) was then performed. cDNA was synthesized using the Superscript Vilo Kit (Invitrogen, Carlsbad, CA) following the manufacturer’s instructions, and amplified using the following primers (amplifying the entire cDNA sequence): forward 5′ GCTTGCCTTGGGACTAATGA 3′ and reverse 5′ AAGGGGTCAGTGCTGAAAGA 3′ . The primer located at the exon 4–exon 5 junction, 5′ AGCTAGAGAGGACCCCTATG 3′ , was used for amplifying a wild type patient’s transcript. Amplified products were separated by agarose gel electrophoresis and analyzed by direct sequencing. In some cases, the bands were cut out and purified using the QUIEX®II Gel Extraction Kit (Qiagen, Hilden, Germany) prior to sequencing. Messenger RNA quantification
The level of TMEM165 wild type transcripts in patient-derived fibroblast before and after AMO treatment was analyzed by quantitative (qRT-PCR) using a LightCycler®480 instrument. For mRNA isolation, cultured patient fibroblasts were harvested and mRNA obtained using TriPure Isolation Reagent (Roche Applied Sciences, Indianapolis, IN) following the manufacturer’s recommendations. All the RNA obtained was subjected to retrotranscription using the High-Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA), and amplified using the LightCycler®480 SYBR Green I Master Kit (Roche Applied Sciences). The forward primer directed to the exon 4–exon 5 junction, and the reverse primer, are the primers used for cDNA amplification. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA expression served as an endogenous expression control. PCR was performed in a total volume of 20 μl, containing 5 ng of cDNA product, 5 μl of LightCycler®480 SYBR Green I Master (Roche Applied Sciences), and PCR primers at a final concentration of 125 nM each. The data were analyzed using lightcycler® software (Roche Applied Sciences), correlating the initial template concentration with the cycle threshold (Ct) to obtain the relative quantity (RQ) of RNA. The RQ is defined as RQ = 2−ΔΔCt , where ΔΔCt is the ΔCt of the patient cell line minus the ΔCt of the control cell line, and ΔCt is the Ct of the target gene (TMEM165) minus the Ct of the housekeeping gene (GAPDH). Minigene construction and in vitro splicing analysis
pSPL3 vector was used to assess in vitro splicing. The gene fragment corresponding to the pseudoexon region, exon 4, and part of the intronic sequence (962 bp) (all in patient fibroblasts) were amplified using the forward primer, 5′ GCAGGAACTGTATTTTATAATTTTACC 3′ and the reverse primer 5′ CTCAACCTCCCAGGCTCA 3′ . This region was cloned into the TOPO vector (Invitrogen) and subsequently excised with Eco RI and cloned
Antisense-mediated therapeutic pseudoexon into the pSPL3 vector. A control construct was generated by PCR mutagenesis using the QuickChange Mutagenesis Kit (Agilent Technologies, Santa Clara, CA). Clones containing the desired normal and mutant inserts in the correct orientation were identified by sequencing. For minigene assays, 1.5 μg of the wild type or mutant pSPL3 construct were transfected into Hep3B human hepatoma cells using the JetPEI reagent (Polyplus Transfection, New York, NY), following the manufacturer’s recommendations. At 24 h post-transfection, cells were harvested and total RNA purified using TriPure Isolation Reagent (Roche Applied Sciences). RT-PCR was performed as described above, using the pSPL3-specific primers SD6 and SA2 (Exon Trapping System, Gibco BRL; Life Technologies, Grand Island, NY). The amplified products were separated by agarose gel electrophoresis and the excised bands further analyzed by direct sequencing after extraction using the QUIEX®II Gel Extraction Kit (Qiagen). Oligonucleotide treatment and analysis
The 25-mer AMO, 5′ GCTTGGTTACAAGAATTTTACCTGC 3′ , directed toward the cryptic 5′ splice site of the pseudoexon activated by the intronic mutation c.792 + 182G > A, was designed, synthesized and purified by Gene Tools (Philomath, OR). Endo-Porter reagent (Gene Tools) was used as the delivery mechanism according to the manufacturer’s instructions. The scrambled AMO sequence used as a negative control was 5′ AAGAGCGAGACTCTGTTTCAAAAAA 3′ . Patient fibroblasts were harvested 24 h after treatment with AMO (20 or 30 μM). RNA extraction and RT-PCR were performed as described above. Immunoblotting
Cells that had undergone or not undergone AMO treatment were harvested with trypsin after 48 h
and resuspended in PBS ×1 plus protease inhibitors (Roche Applied Sciences, Mannheim, Germany). Protein extracts were obtained by freeze–thaw cycles; concentrations were determined by Bio-Rad Protein Assay (Bio-Rad Laboratories, Munich, Germany). Samples were prepared in ×4 NuPage®LDS sample buffer, Dithiothreitol (DTT) and 50 μg of total protein extract, and were separated using 4–12% NuPAGE Novex Bis-Tris mini gels (Invitrogen). They were then transferred to a nitrocellulose membrane. These membranes were blocked 1 h at 4∘ C with phosphate saline buffer (PBS) containing 0.05% Tween-20 and 5% non-fat milk. Blots were incubated with anti-TMEM165 (Sigma-Aldrich, St. Louis, MO) diluted 1:1000 in blocking solution, and with anti-tubulin antibody (Sigma Aldrich) diluted 1:5000 in blocking solution. After incubation with the corresponding secondary antibodies (horseradish peroxide conjugated goat anti-rabbit 1:7500, and goat anti-mouse 1:10,000, respectively), protein bands were detected by ECL Western Blotting System. Protein analysis was performed using a calibrated GS-800 densitometer (Bio-Rad Laboratories). Immunofluorescence microscopy
Cells were grown on glass coverslips with or without AMO for 48 h and fixed with 10% formalin for 20 min at room temperature. The coverslips were rinsed twice with 0.1 M glycine in PBS for 15 min. The cells were then blocked with blocking solution (0.1% triton, 1% BSA, 20% fetal bovine serum) for 30 min. The fixed cells were incubated for 1 h at room temperature with primary antibodies, TMEM165 (Sigma Aldrich) and GM130 (BD Biosciences, San Jose, CA) diluted 1:300 in blocking solution. After washing for three times with PBS, anti-rabbit Alexa 488 and anti-mouse Alexa 594 secondary antibodies (Invitrogen) diluted 1:500 in blocking solution were applied for 1 h at room temperature. The cells were then washed with PBS and incubated with DAPI (Merck, Whitehouse Station, NJ) diluted 1:2500
Fig. 1. Transcriptional profile analysis of patient-derived-fibroblast carrying the c.792+182G>A intronic mutation in the TMEM165 gene, and functional analysis of this splicing mutation using minigenes. (a) Diagram showing the intronic sequence and transcript obtained in patient-derived fibroblasts via reverse transcription-polymerase chain reaction (RT-PCR) analysis. The sequence of the antisense morpholino oligonucleotide (AMO) is included. The intrinsic strength of the TMEM165 splice sites was estimated using the human splicing finder database (BDGP server: http://www.fruitfly.org). Transcriptional profile analysis of healthy control-derived fibroblasts (lane 1) and patient-derived fibroblasts (lane 2). (b) Transcriptional profile returned by ex vivo splicing assays using minigenes for the wild type (c.792+182G) and splicing mutation (c.792+182A). E: empty vector, V: vector exonic sequences.
3
Yuste-Checa et al. in PBS for 5 min. Samples were observed using an Axiovert200 inverted microscope and photographs taken using a C9100-02 digital camera (Hamamatsu, Hamamatsu City, Japan). Shadow and background correction were applied for quantification using fiji software. Statistical analysis
qRT-PCR and immunofluorescence microscopy results were analyzed using one-way anova and Bonferroni’s post hoc test. All statistical analyses were performed using IBM spss Statistics 21 for Windows.
Results
Transcriptional profile analysis of patient-derivedfibroblasts bearing the nucleotide change c.792 + 182G > A in homozygous fashion revealed the presence of two different transcripts. Sequence analysis indicated both to have a 117 bp sequence inserted into intron 4, plus either the deletion of exon 4 or of both exons 4 and 3 (Fig. 1a). The pseudoexon insertion is the result of the activation of a new 5′ cryptic splice site in intron 4. The intrinsic strength of the new 5′ cryptic splicing site increased from 0.5 in the wild type allele to 0.94 in the
Fig. 2. Antisense morpholino treatment in patient-derived fibroblast restores the normal splicing pattern as well as levels of the protein TMEM165. (a) Transcript profile analysis of control (lane 1), untreated patient-derived fibroblasts (lane 2), and patient-derived fibroblasts treated with 20 μM (lane 3) or 30 μM (lane 4) of antisense morpholino oligonucleotide (AMO) or a scrambled morpholino (lane 5). (b) Western blot analysis of control-derived fibroblast, untreated patient-derived fibroblast (−), and patient-derived fibroblasts treated with the AMO (20 μM) or scrambled morpholino (SCR). Not reaching statistically significance. (c) Indirect double immunofluorescence testing of control-derived fibroblasts (c) and fibroblasts from patients P1, P2 and P3 (genotypes indicated in the table in Fig. 2) untreated and treated with AMO (20 μM). Cells were double-labeled with TMEM165 (green fluorescence) and the Golgi marker GM130 (red fluorescence) antibodies. The figure shows the quantification of TMEM165 via the intensity of fluorescence recorded. Data was collected for two independent experiments; at least 60 photos were examined. *** p < 0.001.
4
Antisense-mediated therapeutic pseudoexon mutated allele (BDGP server: http://www.fruitfly.org). Because the normal transcript was not detected by RT-PCR in patient-derived fibroblasts, a complementary analysis was performed to check for its presence. An oligonucleotide comprising the sequence between exons 4 and 5 was designed to amplify just the wild type transcript. The result of RT-PCR with the specific primers showed the slight presence of wild type transcripts in patient fibroblasts (data not shown). The effect of the mutation c.792 + 182G > A on the splicing process was examined using an ex vivo splicing assay involving minigenes. The minigene constructs contained the normal and mutant pseudoexon sequences along with an exon 4 construct. Both the wild type and mutant minigenes were used to transfect Hep3B cells. After 24 h, the transcriptional profile was analyzed; it was found to be similar to that recorded for the mRNA of both the control- and patient-derived fibroblasts (Fig. 1b). Antisense therapy for preventing the insertion of the pseudoexon was tested using a specific AMO targeted toward the intronic 5′ cryptic splicing site (Fig. 1b). Patient-derived fibroblasts bearing the splicing mutation c.792 + 182G > A were transfected with two different concentrations of AMO (20 and 30 μM) and also with a scrambled AMO. After 24 h, the mRNA was analyzed by RT-PCR and the fragments obtained were sequenced. The results showed the complete restoration of the transcriptional profile in a dose and sequence-specific manner (Fig. 2a). To quantify the recovery of the wild type transcript in patient-derived fibroblast treated with AMO 20 μM, qRT-PCR was performed with the primers designed to amplify wild type transcripts in patient’s fibroblasts. Wild type transcripts were found to be present in the patient-derived fibroblasts, although only about 2% of that recorded for control-derived fibroblast. Significant (p < 0.001) recovery was achieved, however, using AMO at 20 μM; indeed, at this concentration it increased 10-fold compared to the basal expression and reached 20% of the presence recorded in controls (data not shown). To determine whether the recovery of the transcriptional profile achieved by AMO translated into protein rescue, Western blotting was performed. Patient-derived fibroblasts were treated with AMO (20 μM) and after 48 h protein analysis was performed. Western blotting revealed the presence of immunoreactive protein in untreated cells, and an increase of 3-fold in treated cells over non-treated cells (Fig. 2b). To quantify the protein recovery and determine the localization of the produced protein, an indirect double immunofluorescence assay was performed. Cells were double-labeled with antibodies against TMEM165 and the Golgi marker GM130. The results showed only basal protein expression in the Golgi of patient cells (around 20% compared to control-derived fibroblasts) but a significant increase of threefold (p < 0.001) after AMO treatment (Fig. 2c). No increase in TMEM65 protein was seen after AMO treatment in the patient-derived fibroblasts bearing the missense changes, confirming the protein restoration to be sequence-specific (Fig. 2c).
Discussion
Splicing defects represent the second most important pathogenic causes of genetic disease after missense changes (HGMD, https://portal.biobase-international. com). The most common mutations affecting the splicing process are those involving the 5′ and 3′ splice-site consensus sequences. Nevertheless, mutations outside of these may account for another significant fraction of pathogenic mutations, such as deep intronic mutations, an emerging cause of genetic diseases. Importantly, they provide a promising potential target for personalized treatment. After performing transcriptional profile analysis on patient-derived fibroblasts, functional analysis of nucleotide changes using ex vivo splicing is recommended to confirm the pathogenic effect of genomic changes (15, 16). Functional analysis of the deep intronic variant change c.792 + 182G > A using the minigene method showed the same transcription profile to be present in patient- and control-derived fibroblasts. These results confirm that the deep intronic mutation only activates the insertion of a 117 bp intronic pseudoexon between exons 4 and 5, leading to a putative truncated protein (if it is translated at all) caused by a pre-mature stop codon. Minigene analysis also showed that the insertion of the pseudoexon causes the additional skipping of exon 4 detected in patient-derived fibroblasts. These results are not easy to explain because the skipping of exon 4 cannot rescue the lost frame of the protein. From a therapeutic point of view, pseudoexons represent excellent targets for antisense-mediated ‘exon skipping’ therapy; the natural splice sites of the surrounding exons are intact, and the use of specific small oligonucleotide makes possible the rescue of the aberrant splicing process by forcing pseudoexon skipping. In this study, the aberrant TMEM165 mRNA was targeted using oligonucleotide technology. Wild type transcripts were restored in patient-derived fibroblast, reaching 20% of the level seen in control fibroblasts. These rescued transcripts were efficiently translated and produced 60% of protein seen in the control cells. Further, the protein was located in the Golgi apparatus where it is probably to be active. This therapeutic approach is confirmed to be mutation-specific because that it was unable to increase protein levels in patient-derived fibroblasts bearing missense changes. The treatment of cells with AMO-based therapy that prevents pathogenic splicing defects is a promising treatment for molecular lesions considered ‘undruggable’. To date, deep intronic mutations have been extensively described in different genes expressed in accessible tissues in which mRNA analysis is feasible (17–19). Recent advances in whole genome sequencing have revealed the presence of this type of mutation in many genes (20–22). It is probably that more of these types of pathogenic mutation will be detected, thanks to massively parallel sequencing of entire intronic sequences or RNA Seq, which should allow them to be detected directly in genomic DNA.
5
Yuste-Checa et al. Antisense therapy has been successfully used in cellular models of several genetic disorders, among them inherited metabolic diseases such as phosphomanomutase deficiency (PMM2-CDG), the most common congenital disorder of glycosylation (23). In all cases the splicing correction is dose- and sequence-specific; no obvious cell cytotoxicity has been observed (10). The challenge now is the safe and targeted delivery of the antisense therapy to specific tissues. Successful rescue ameliorating the disease phenotype in mouse models has been reported for Hutchinson–Gilford progeria syndrome, β-thalassemia, myotonic dystrophy and Duchenne muscular dystrophy (24, 25). Further advances in the medicinal chemistry of antisense oligonucleotides are, however, still required for optimizing antisense behavior in damaged tissues and organs. The next generation of chemically modified oligonucleotides is now being developed with the aim of improving their stability, affinity and delivery, and to enhance their pharmaceutical properties. Locked nucleic acids (LNA), peptide nucleic acid (PNA), 2′ -O-methoxyethyl RNA (2′ -MOE) and Vivo-Morpholino are being tested in cellular and animal models (25). In addition, a variety of cell-penetrating peptides directly conjugated to antisense oligonucleotides have been shown to enhance delivery in Duchenne model systems, to improve systemic distribution, and to be more effective than ’naked’ antisense oligonucleotides (26). In summary, progress in nanoparticle development, along with a new generation of disease models, will bring the bench closer to the bedside; this may prove important in disorders with no other available therapy. A clinical trial of a modified phosphorothioate oligonucleotide known as Vitravene (Formivirsen) for the treatment of cytomegalovirus-induced retinitis (27) has recently begun, and an antisense drug for the treatment of familiar hypercholesterolemia has been approved by the Food and Drug Administration (FDA) (28, 29). The therapeutic potential of antisense treatment depends, however, on the tissues and organs involved; each poses different challenges for targeted delivery. Owing to the pre-dominant clinical features of TMEM165-CDG are skeletal abnormalities, (growth retardation not responsive to growth hormone, osteoporosis, and skeletal dysplasia), antisense oligonucleotides should reach the bones. Systemic administration may therefore be sufficient to ameliorate disease outcome. In the absence of TMEM165-CDG animal models with specific splice defects amenable to antisense treatment, testing antisense therapies in induced, pluripotent, human stem cells might provide a model for evaluating antisense oligonucleotide-mediated pseudoexon skipping (30). In conclusion, this paper reports proof-of-concept for treating a Golgi-resident protein defect using mutation-specific antisense therapy. AMO therapy may be an excellent candidate for the treatment of disorders, for which no other treatments option is available.
6
Acknowledgements This work was funded by the Fondo de Investigaciones Sanitarias (PI10/00455 to B. P. and PI12/02078 to C. P. C.). P. Y. was supported by a grant from the Universidad Autónoma de Madrid. An institutional grant from the Fundación Ramón Areces to the Centro de Biología Molecular Severo Ochoa is gratefully acknowledged.
References 1. Wu X, Steet RA, Bohorov O et al. Mutation of the COG complex subunit gene COG7 causes a lethal congenital disorder. Nat Med 2004: 10 (5): 518–523. 2. Reynders E, Foulquier F, Leao Teles E et al. Golgi function and dysfunction in the first COG4-deficient CDG type II patient. Hum Mol Genet 2009: 18 (17): 3244–3256. 3. Kornak U, Reynders E, Dimopoulou A et al. Impaired glycosylation and cutis laxa caused by mutations in the vesicular H + -ATPase subunit ATP6V0A2. Nat Genet 2008: 40 (1): 32–34. 4. Jaeken J, Hennet T, Matthijs G et al. CDG nomenclature: time for a change!. Biochim Biophys Acta 2009: 1792 (9): 825–826. 5. Schachter H, Freeze HH. Glycosylation diseases: quo vadis? Biochim Biophys Acta 2009: 1792 (9): 925–930. 6. Jaeken J. Congenital disorders of glycosylation. Ann N Y Acad Sci 2010: 1214: 190–198. 7. Buratti E, Baralle M, Baralle FE. Defective splicing, disease and therapy: searching for master checkpoints in exon definition. Nucleic Acids Res 2006: 34 (12): 3494–3510. 8. Pagani F, Baralle FE. Genomic variants in exons and introns: identifying the splicing spoilers. Nat Rev Genet 2004: 5 (5): 389–396. 9. Maquat LE. Nonsense-mediated mRNA decay: splicing, translation and mRNP dynamics. Nat Rev Mol Cell Biol 2004: 5 (2): 89–99. 10. Perez B, Rodriguez-Pascau L, Vilageliu L, Grinberg D, Ugarte M, Desviat LR. Present and future of antisense therapy for splicing modulation in inherited metabolic disease. J Inherit Metab Dis 2010: 33 (4): 397–403. 11. Foulquier F, Amyere M, Jaeken J et al. TMEM165 deficiency causes a congenital disorder of glycosylation. Am J Hum Genet 2012: 91 (1): 15–26. 12. Demaegd D, Foulquier F, Colinet AS et al. Newly characterized Golgi-localized family of proteins is involved in calcium and pH homeostasis in yeast and human cells. Proc Natl Acad Sci U S A 2013: 110 (17): 6859–6864. 13. Rosnoblet C, Legrand D, Demaegd D et al. Impact of disease-causing mutations on TMEM165 subcellular localization, a recently identified protein involved in CDG-II. Hum Mol Genet 2013: 22 (14): 2914–2928. 14. Zeevaert R, de Zegher F, Sturiale L et al. Bone dysplasia as a key feature in three patients with a novel congenital disorder of glycosylation (CDG) type II due to a deep intronic splice mutation in TMEM165. JIMD Rep 2013: 8: 145–152. 15. Cooper TA. Use of minigene systems to dissect alternative splicing elements. Methods 2005: 37 (4): 331–340. 16. Desviat LR, Perez B, Ugarte M. Minigenes to confirm exon skipping mutations. Methods Mol Biol 2012: 867: 37–47. 17. Rincon A, Desviat LR, Ugarte M et al. Propionic and methylmalonic acidemia: antisense therapeutics for intronic variations causing aberrantly spliced mRNA. EMBO Conference Series "pre-mRNA processing and disease", Cortina d’Ampezzo, Italy, 2007. 18. Perez B, Rincon A, Jorge-Finnigan A et al. Pseudoexon exclusion by antisense therapy in methylmalonic aciduria (MMAuria). Hum Mutat 2009: 30 (12): 1676–1682. 19. Davis RL, Homer VM, George PM, Brennan SO. A deep intronic mutation in FGB creates a consensus exonic splicing enhancer motif that results in afibrinogenemia caused by aberrant mRNA splicing, which can be corrected in vitro with antisense oligonucleotide treatment. Hum Mutat 2009: 30 (2): 221–227. 20. Cavalieri S, Pozzi E, Gatti RA, Brusco A. Deep-intronic ATM mutation detected by genomic resequencing and corrected in vitro by antisense morpholino oligonucleotide (AMO). Eur J Hum Genet 2013: 21 (7): 774–778.
Antisense-mediated therapeutic pseudoexon 21. Webb TR, Parfitt DA, Gardner JC et al. Deep intronic mutation in OFD1, identified by targeted genomic next-generation sequencing, causes a severe form of X-linked retinitis pigmentosa (RP23). Hum Mol Genet 2012: 21 (16): 3647–3654. 22. Flanagan SE, Xie W, Caswell R et al. Next-generation sequencing reveals deep intronic cryptic ABCC8 and HADH splicing founder mutations causing hyperinsulinism by pseudoexon activation. Am J Hum Genet 2013: 92 (1): 131–136. 23. Vega AI, Perez-Cerda C, Desviat LR, Matthijs G, Ugarte M, Perez B. Functional analysis of three splicing mutations identified in the PMM2 gene: toward a new therapy for congenital disorder of glycosylation type Ia. Hum Mutat 2009: 30 (5): 795–803. 24. Havens MA, Duelli DM, Hastings ML. Targeting RNA splicing for disease therapy. Wiley Interdiscip Rev 2013: 4 (3): 247–266. 25. Osorio FG, Navarro CL, Cadinanos J et al. Splicing-directed therapy in a new mouse model of human accelerated aging. Sci Transl Med 2011: 3 (106): 106ra107.
26. Betts CA, Wood MJ. Cell penetrating peptide delivery of splice directing oligonucleotides as a treatment for Duchenne muscular dystrophy. Curr Pharm Des 2013: 19 (16): 2948–2962. 27. Roehr B. Fomivirsen approved for CMV retinitis. J Int Assoc Physicians AIDS Care 1998: 4 (10): 14–16. 28. Raal FJ, Santos RD, Blom DJ et al. Mipomersen, an apolipoprotein B synthesis inhibitor, for lowering of LDL cholesterol concentrations in patients with homozygous familial hypercholesterolaemia: a randomised, double-blind, placebo-controlled trial. Lancet 2010: 375 (9719): 998–1006. 29. Visser ME, Wagener G, Baker BF et al. Mipomersen, an apolipoprotein B synthesis inhibitor, lowers low-density lipoprotein cholesterol in high-risk statin-intolerant patients: a randomized, double-blind, placebo-controlled trial. Eur Heart J 2012: 33 (9): 1142–1149. 30. Dick E, Kalra S, Anderson D et al. Exon skipping and gene transfer restore dystrophin expression in human induced pluripotent stem cells-cardiomyocytes harboring DMD mutations. Stem Cells Dev 2013.
7
