T.C. REPUBLIC OF TURKEY HACETTEPE UNIVERSITY GRADUATE SCHOOL OF HEALTH SCIENCES INVESTIGATION OF THE ROLE OF LAP1B IN TRANSCRIPTIONAL REGULATION OF MUSCLE CELLS Gülsüm KAYMAN KÜREKÇİ, M.Sc. Program of Medical Biology DOCTOR OF PHILOSOPHY THESIS ANKARA 2021 T.C. REPUBLIC OF TURKEY HACETTEPE UNIVERSITY GRADUATE SCHOOL OF HEALTH SCIENCES INVESTIGATION OF THE ROLE OF LAP1B IN TRANSCRIPTIONAL REGULATION OF MUSCLE CELLS Gülsüm KAYMAN KÜREKÇİ, M.Sc. Program of Medical Biology DOCTOR OF PHILOSOPHY THESIS ADVISOR OF THE THESIS Prof. Dr. Pervin R. DİNÇER ANKARA 2021 iii HACETTEPE UNIVERSITY GRADUATE SCHOOL OF HEALTH SCIENCES INVESTIGATION OF THE ROLE OF LAP1B IN TRANSCRIPTIONAL REGULATION OF MUSCLE CELLS Gülsüm Kayman-Kürekçi Supervisor: Prof. Pervin R. Dinçer This thesis study has been approved and accepted as a Ph.D. dissertation in “Medical Biology Program” by the assessment committee, whose members are listed below, on 11.01.2021. Chairman of the Committee : Prof. A. Elif, ERSON BENSAN (Signature) Middle East Technical University Member : Prof. Hayat, ERDEM YURTER (Signature) Hacettepe University Member : Assoc. Prof. Işık, YULUĞ (Signature) Bilkent University Member : Assist. Prof., Aybar C., ACAR (Signature) Middle East Technical University Member : Assist. Prof., Uğur, AKPULAT (Signature) Kastamonu University This dissertation has been approved by the above committee in conformity to the related issues of Hacettepe University Graduate Education and Examination Regulation. Prof. Diclehan ORHAN, M.D., Ph.D. Director iv YAYIMLAMA VE FİKRİ MÜLKİYET HAKLARI BEYANI Enstitü tarafından onaylanan lisansüstü tezimin/raporumun tamamını veya herhangi bir kısmını, basılı (kağıt) ve elektronik formatta arşivleme ve aşağıda verilen koşullarla kullanıma açma iznini Hacettepe Üniversitesine verdiğimi bildiririm. Bu izinle Üniversiteye verilen kullanım hakları dışındaki tüm fikri mülkiyet haklarım bende kalacak, tezimin tamamının ya da bir bölümünün gelecekteki çalışmalarda (makale, kitap, lisans ve patent vb.) kullanım hakları bana ait olacaktır. Tezin kendi orijinal çalışmam olduğunu, başkalarının haklarını ihlal etmediğimi ve tezimin tek yetkili sahibi olduğumu beyan ve taahhüt ederim. Tezimde yer alan telif hakkı bulunan ve sahiplerinden yazılı izin alınarak kullanılması zorunlu metinlerin yazılı izin alınarak kullandığımı ve istenildiğinde suretlerini Üniversiteye teslim etmeyi taahhüt ederim. Yükseköğretim Kurulu tarafından yayınlanan “Lisansüstü Tezlerin Elektronik Ortamda Toplanması, Düzenlenmesi ve Erişime Açılmasına İlişkin Yönerge” kapsamında tezim aşağıda belirtilen koşullar haricince YÖK Ulusal Tez Merkezi / H.Ü. Kütüphaneleri Açık Erişim Sisteminde erişime açılır. o Enstitü / Fakülte yönetim kurulu kararı ile tezimin erişime açılması mezuniyet tarihimden itibaren 2 yıl ertelenmiştir. (1) • Enstitü / Fakülte yönetim kurulunun gerekçeli kararı ile tezimin erişime açılması mezuniyet tarihimden itibaren 6 ay ertelenmiştir. (2) o Tezimle ilgili gizlilik kararı verilmiştir. (3) ...../………/…… Gülsüm Kayman Kürekçi i i“Lisansüstü Tezlerin Elektronik Ortamda Toplanması, Düzenlenmesi ve Erişime Açılmasına İlişkin Yönerge” (1) Madde 6. 1. Lisansüstü tezle ilgili patent başvurusu yapılması veya patent alma sürecinin devam etmesi durumunda, tez danışmanının önerisi ve enstitü anabilim dalının uygun görüşü üzerine enstitü veya fakülte yönetim kurulu iki yıl süre ile tezin erişime açılmasının ertelenmesine karar verebilir. (2) Madde 6. 2. Yeni teknik, materyal ve metotların kullanıldığı, henüz makaleye dönüşmemiş veya patent gibi yöntemlerle korunmamış ve internetten paylaşılması durumunda 3. şahıslara veya kurumlara haksız kazanç imkanı oluşturabilecek bilgi ve bulguları içeren tezler hakkında tez danışmanının önerisi ve enstitü anabilim dalının uygun görüşü üzerine enstitü veya fakülte yönetim kurulunun gerekçeli kararı ile altı ayı aşmamak üzere tezin erişime açılması engellenebilir. (3) Madde 7. 1. Ulusal çıkarları veya güvenliği ilgilendiren, emniyet, istihbarat, savunma ve güvenlik, sağlık vb. konulara ilişkin lisansüstü tezlerle ilgili gizlilik kararı, tezin yapıldığı kurum tarafından verilir *. Kurum ve kuruluşlarla yapılan işbirliği protokolü çerçevesinde hazırlanan lisansüstü tezlere ilişkin gizlilik kararı ise, ilgili kurum ve kuruluşun önerisi ile enstitü veya fakültenin uygun görüşü üzerine üniversite yönetim kurulu tarafından verilir. Gizlilik kararı verilen tezler Yükseköğretim Kuruluna bildirilir. Madde 7.2. Gizlilik kararı verilen tezler gizlilik süresince enstitü veya fakülte tarafından gizlilik kuralları çerçevesinde muhafaza edilir, gizlilik kararının kaldırılması halinde Tez Otomasyon Sistemine yüklenir * Tez danışmanının önerisi ve enstitü anabilim dalının uygun görüşü üzerine enstitü veya fakülte yönetim kurulu tarafından karar verilir. v ETHICAL DECLARATION In this thesis study, I declare that all the information and documents have been obtained in the base of the academic rules and all audio-visual and written information and results have been presented according to the rules of scientific ethics. I did not do any distortion in data sets. In case of using other works, related studies have been fully cited in accordance with the scientific standards. I also declare that my thesis study is original except cited references. It was produced by myself in consultation with the supervisor (Prof. Dr. Pervin R. DİNÇER) and written according to the rules of thesis writing of Hacettepe University Institute of Health Sciences. M.Sc. Gülsüm KAYMAN KÜREKÇİ vi ACKNOWLEDGEMENTS ‘’Thinking must never submit itself, neither to a dogma, nor to a party, nor to a passion, nor to an interest, nor to a preconceived idea, nor to whatever it may be, if not to the facts themselves, because, for it, to submit would be the end of its existence.’’ Henri Poincaré (1909) I wish to thank, first and foremost, my supervisor Prof. Pervin Dinçer for her endless support throughout my Ph.D. life in regard to constantly stimulating my research enthusiasm by opening new doors and decisively pushing me to go through them. Her support has widened my horizon and helped to shape my trajectory. I would like to thank the assessment committee members Prof. A. Elif Erson Bensan, Prof. Hayat Erdem Yurter, Assoc. Prof. Işık Yuluğ, Assist. Prof. Aybar C. Acar and Assist. Prof. Uğur Akpulat for their valuable suggestions and interesting feedback. Also thanks to all members of the Department of Medical Biology for the fun work environment. I have been very fortunate to have wonderful laboratory mates and good friends without whom my long Ph.D. life would be bleak. Special thanks to Ecem Kural Mangıt, Gizem Karagedikli Önal, Evrim Aksu Mengeş, Nazlı Eskici, Edibe Avcı and Şeyda Ünsal for all good talks, laughs and memories. I wish to thank Yusuf M. Kürekçi for his unconditional support and patience, waiting for hours for the never-ending ‘last five minutes’ of my experiments. I dedicate this thesis to my daughter Esin, as her name says, she will be my eternal muse. vii ABSTRACT Kayman Kürekçi, G., Investigation of the Role of LAP1B in Transcriptional Regulation of Muscle Cells. Hacettepe University Graduate School of Health Sciences, Ph.D. Thesis in Medical Biology, Ankara, 2021. The loss-of-function of the inner nuclear membrane protein LAP1B (lamina-associated polypeptide 1, isoform B) causes muscular dystrophy and cardiomyopathy in humans. The function of LAP1B in muscle is still unknown. The goal of this thesis is to contribute to the understanding of the molecular pathogenesis underlying muscular dystrophy caused by LAP1B, by determining transcriptional changes occurring throughout differentiation of muscle cells lacking LAP1B. For this purpose, primary fibroblasts isolated from healthy and one patient affected by LAP1B-related muscular dystrophy were immortalized and myoconverted with inducible MyoD gene transfer. Whereas control cells formed mature myotubes within eight days, mutant cells demonstrated very low fusion potential and failed to fully differentiate. Mutations causing knockdown of LAP1A/B expression were created in C2C12 mouse myoblasts and similar results were observed. By RNA sequencing, genes differentially expressed in control and mutant cells within the transcriptome and enriched pathways were identified. Muscle contraction, cell cycle, mitotic chromatid segregation and extracellular matrix organization were among the most significantly enriched pathways. It was shown by cell cycle assay that despite downregulation of p21 expression, mutant cells withdrew from the cell cycle. Finally, upregulation of p53 expression and increase in the number of micronuclei in mutant cells were related to cellular stress and DNA damage. These findings demonstrated that LAP1B is not involved in cell cycle exit but might suggest a role in DNA damage repair necessary for the induction of myogenin expression. Identification of previously unknown pathways for LAP1 will contribute to the discovery of novel targets for therapy. Keywords: LAP1, myogenic differentiation, transcriptome. This study was funded by TÜBİTAK (Project no. 116S307). viii ÖZET Kayman Kürekçi, G., LAP1B Proteinin Kas Hücrelerinin Transkripsiyon Regülasyonundaki Rolünün Araştırılması. Hacettepe Üniversitesi Sağlık Bilimleri Enstitüsü Tıbbi Biyoloji Programı Doktora Tezi, Ankara, 2021. Çekirdek iç membran proteini LAP1B’nin (lamina-ilişkili polipeptit 1, izoform B) işlev kaybı insanda kas distrofisi ve kardiyomiyopatiye neden olmaktadır. LAP1B’nin kas dokusundaki işlevi henüz bilinmemektedir. Tezin amacı, LAP1B’yi ifade etmeyen kas hücrelerinin farklılaşma sürecinde transkriptom düzeyindeki değişikliklerin belirlenmesi ile, insanda neden olduğu kas distrofisinin moleküler patogenezinin anlaşılmasına katkı sağlamaktır. Bu amaçla in vitro model olarak kullanılmak üzere kontrol ve bir LAP1B-ilişkili kas distrofisi hastasına ait primer fibroblastlar ölümsüzleştirilmiş ve uyarılabilir MyoD gen transferi gerçekleştirilmiştir. Kontrol hücrelerin sekiz günde olgun miyotüpler oluşturduğu gözlenirken, mutant hücrelerin füzyon kapasitelerinin çok düşük olduğu ve farklılaşmayı tamamlayamadıkları bulunmuştur. Ayrıca C2C12 fare miyoblastlarında LAP1A/B ifadesinin baskılanmasına neden olan mutasyonlar oluşturulmuş ve aynı bulgular gözlenmiştir. RNA dizileme yöntemi ile kontrol ve mutant hücrelerde tüm transkriptom içerisinde ifadesi değişen genler ve görev aldıkları yolaklar belirlenmiştir. Bu yolaklar arasından kas kasılması, hücre döngüsü, mitozda kromatit segregasyonu ve hücre dışı matriks organizasyonu öne çıkmıştır. Mutant hücrelerin p21 ifadesinin baskılanmış olmasına rağmen hücre döngüsünden çıktıkları hücre döngüsü analizi ile belirlenmiştir. Son olarak mutant hücrelerde p53 ifadesinde ve mikroçekirdek sayısında gözlenen artışın, hücresel stres ve DNA hasarı ile ilişkili olduğu görülmektedir. Bu çalışmadan elde edilen bulgular, LAP1B’nin kas farklılaşmasının erken evresinde hücre döngüsü çıkışında rolünün olmadığı ancak miyogenin ifadesinin uyarılması için DNA hasarının tamirinde görevli olabileceğine işaret etmektedir. LAP1’in yeni yolaklardaki olası rolünün tanımlanması, tedavide hedeflenebilir yolakların keşfedilmesine de olanak sağlayacaktır. Anahtar Kelimeler: LAP1, kas farklılaşması, transkriptom. Bu çalışma TÜBİTAK tarafından desteklenmiştir (Proje no. 116S307). ix TABLE OF CONTENTS APPROVAL iii YAYIMLAMA VE FİKRİ MÜLKİYET HAKLARI BEYANI iv ETHICAL DECLARATION v ACKNOWLEDGEMENTS vi ABSTRACT vii ÖZET viii TABLE OF CONTENTS ix ABBREVIATIONS xii FIGURES xiii TABLES xv 1. INTRODUCTION 1 2. BACKGROUND 3 2.1 Lamina-associated Polypeptide 1 (LAP1) 3 2.1.1 General Properties of LAP1 3 2.1.2 Protein Modifications and Interactors of LAP1 4 2.1.3 Human LAP1 Mutations and Related Pathologies 6 2.2 Regulation of In Vitro Myogenic Differentiation 7 2.2.1 Differentiation Commitment and Cell Cycle Withdrawal 9 2.2.2 Differentiation Checkpoint 9 2.3 Role of LAP1 in Muscle 10 2.4 Research Question and Goals of the Thesis 11 3. MATERIALS AND METHODS 12 3.1 Materials 12 3.1.1 Cell Culture 12 3.1.2 Cell Transduction and Transfection 12 3.1.3 CRISPR/Cas9 Genome Editing and Cloning 13 3.1.4 DNA Isolation and DNA Sequencing 13 3.1.5 RNA Isolation and cDNA Synthesis 14 3.1.6 mRNA Sequencing 14 x 3.1.7 Quantitative Real-time PCR 14 3.1.8 Protein isolation and Quantitation 14 3.1.9 Western Blotting 15 3.1.10 Polymerase Chain reaction 15 3.1.11 Agarose Gel Electrophoresis 15 3.1.12 Immunofluorescence Staining 16 3.1.13 Antibodies 16 3.1.14 Cell Cycle Assay 17 3.2 Methods 17 3.2.1 Cell Culture 17 3.2.2 Lentiviral Transduction 17 3.2.3 Induction of Myogenic Differentiation 18 3.2.4 Immunofluorescence Staining 18 3.2.5 Determination of Fusion and Differentiation Indices 19 3.2.6 CRISPR/Cas9-mediated Tor1aip1 Mutations in C2C12 Myoblasts 19 3.2.7 Total mRNA Sequencing 24 3.2.8 cDNA Synthesis and Quantitative Real-time PCR 28 3.2.9 Protein Isolation and Quantitation 29 3.2.10 Western Blotting 30 3.2.11 Cell Cycle Assay 31 3.2.12 Micronuclei Count 32 4. RESULTS 33 4.1 Immortalization and Myoconversion of Primary Fibroblasts 33 4.1.1 Determination of Transduction Efficiency 33 4.1.2 Myoconversion of Fibroblasts 35 4.1.3 Determination of Differentiation and Fusion Indices 37 4.2 CRISPR/Cas9-mediated Tor1aip1 Mutations in C2C12 Myoblasts 38 4.2.1 Generation of CRISPR/Cas9-expressing Vectors 39 4.2.2 Genotyping of Mutant Clones and Determination of LAP1 Expression 39 4.2.3 Myogenic Differentiation Potential of Tor1aip1-mutant C2C12 Myoblasts 42 xi 4.3 Transcriptome Profiling 43 4.3.1 Statistics of Differential Expression Analysis 43 4.3.2 Validation of RNA-seq Data by qPCR 46 4.3.3 Gene Set Enrichment Analysis 47 4.3.4 Analysis of MyoD Target Genes 53 4.4 Investigation of Cell Cycle Exit in Mutant Cells 55 4.4.1 Expression of Proteins Involved in Cell Cycle Exit 55 4.4.2 Cell Cycle Assay 56 4.5 Investigation of DNA Damage in Mutant Cells 57 5. DISCUSSION 60 6. CONCLUSION AND SUGGESTIONS 65 7. REFERENCES 67 8. APPENDICES 73 APPENDIX 1: Ethical Approval APPENDIX 2: DEGs Lists APPENDIX 3: Expression of ERK1/2 in control and LAP1B-mutant cells APPENDIX 4: Thesis Originality Report APPENDIX 5: Digital Receipt 9. CURRICULUM VITAE xii ABBREVIATIONS °C degree Celsius aa amino acid bp base pair cDNA complementary deoxyribonucleic acid DNA deoxyribonucleic acid dNTP deoxyribonucleotide triphosphate DEG differentially expressed gene mRNA messenger ribonucleic acid ml milliliter mM millimolar ng nanogram qPCR quantitative polymerase chain reaction RNA ribonucleic acid RNA-seq RNA sequencing Tm Melting Temperature WT wild type μg microgram μl microliter xiii FIGURES Figure Page 2.1. Structure of human TOR1AIP1 gene and LAP1 protein isoforms. 4 2.2. Myogenic differentiation and temporal expression pattern of MRFs. 8 3.1. Structure of mouse LAP1 transcripts. 20 3.2. Locations of gRNAs binding sites on exon 1 of mouse Tor1aip1 gene. 21 3.3. Electrophoresis of selected total RNA extracts used for RNA-seq. 26 3.4. Distance matrices of RNA-seq samples before and after removal of outliers. 27 3.5. Gating strategy used for data analysis in flow cytometry. 32 4.1. Expression of LAP1 isoforms (LAP1B and LAP1C) in control 1 Immortalized (indicated as +hTERT) fibroblasts, LAP1B-mutant Primary and immortalized fibroblasts as shown by western blotting. 33 4.2. MyoD expression in controls and LAP1B-mutant immortalized myoconverted fibroblasts induced with doxycycline (+Dox) for 18 hours as shown by immunofluorescence staining. 34 4.3. (A) Percentages of MyoD-positive nuclei in control and LAP1B-mutant cells after 18 hours of doxycycline induction. (B) MyoD expression in LAP1B-mutant cells incubated in differentiation medium without doxycycline (-Dox) for 18 hours. (C) MyoD mRNA expression levels throughout differentiation as assessed by RNA-seq. (D) MyoD protein levels in control and LAP1B-mutant cells at days 0 and 2 as shown by western blotting. 35 4.4. Monitoring of myogenic differentiation of myoconverted fibroblasts by MyHC expression at day 0 (D0), D2, D5 and D8 of differentiation. 36 4.5. MyHC expression in control multinucleated myotubes at day 8. 36 4.6. Morphology of LAP1B-mutant cells differentiated for 17 days. 37 4.7. (A) Differentiation and (B) fusion indices of control and LAP1B-mutant cells at days 5 and 10 of differentiation. 38 4.8. Cloning of gRNA-coding sequences into CRISPR/Cas9-expressing vectors. 39 4.9. Chromatograms showing mutations in exon 1 of Tor1aip1 gene in mutant C2C12 clones. 40 4.10. LAP1 isoforms expression in Tor1aip1-mutant C2C12 clones at days 0 and 5 of differentiation as assessed by western blotting. 41 4.11. (A) Micrographs of WT and Tor1aip1-mutant C2C12 clones differentiated for 5 days and stained with MF20 antibody for MyHC expression. (B) Differentiation and fusion indices of WT and Tor1aip1-mutant xiv C2C12 myoblasts differentiated for 5 days. 42 4.12. (A) Number of DEGs in LAP1B-mutant cells compared to control at each day of differentiation. (B) Number of DEGs between consecutive days throughout differentiation within control or LAP1B-mutant cells. 44 4.13. Cluster heat map of DEGs in control and LAP1B-mutant cells determined by time course regression analysis. Expression fold changes at each day were compared to day 0 (D0). 45 4.14. Validation of RNA-seq data by qPCR. Log2 values of expression fold changes obtained from RNA-seq and qPCR of selected genes in LAP1B-mutant cells compared to controls were plotted. 46 4.15. Top ten most significantly enriched categories containing DEGs in LAP1B-mutant cells compared to controls at day 0, day 2, day 5 and day 8 of differentiation. 48 4.16. Top ten most significantly enriched categories containing DEGs between consecutive timepoints within control or LAP1B-mutant cells. 49 4.17. Most significantly enriched categories from DEGs in time course regression analysis. (A) Top ten most significantly enriched GO-BP categories and (B) clustergram showing DEGs associated with GO-BP categories. (C) Top ten most significantly enriched KEGG Pathways categories and (D) clustergram showing DEGs associated with KEGG Pathways categories. 51 4.18. Protein expression of myogenin in control and LAP1B-mutant cells throughout differentiation as shown by western blotting. 55 4.19. Expression of proteins involved in cell cycle exit in control and LAP1B- mutant cells throughout differentiation. (A) Expression of p21 was normalized to GAPDH. (B) Expression of total Rb. Left graphic shows total Rb normalized to GAPDH with expression fold changes of all samples relative to control at day 0. Right graphic shows expression fold changes of hyposphophorylated Rb relative to total Rb in control and mutant cells. 56 4.20. Analysis of cell cycle phases in control and LAP1B-mutant cells at days 0, 1 and 3 of differentiation by propidium iodide staining and flow cytometry. 57 4.21. (A) Micronuclei in control and LAP1B-mutant cells at days 0, 5 and 8 of differentiation were monitored by DAPI staining. (B) gH2AX expression co-localized with micronuclei. 58 4.22. Expression of p53 in control and LAP1B-mutant cells throughout differentiation. Expression was normalized to GAPDH, the bar graphic shows expression fold changes of samples relative to control at day 0. 59 xv TABLES Table Page 2.1. Human TOR1AIP1 mutations, affected isoforms and associated phenotypes. 7 3.1. Sequences of oligonucleotides used for gRNA synthesis and genotyping primers. 21 3.2. Concentrations and OD values of RNA samples used in RNA-seq. 25 3.3. Comparison types used for differential expression analyses. 28 3.4. Sequences of primers used in qPCR. 29 3.5. Composition of gels used in SDS-PAGE. 30 4.1. Expression fold changes of differentially expressed MyoD target genes in LAP1B-mutant cells compared to controls at day 2 of differentiation. 54 1 1. INTRODUCTION The nuclear envelope consists of inner and outer nuclear membranes, and nuclear lamina. So far, more than 80 integral inner nuclear membrane proteins have been identified (1). It is known that the nuclear proteome displays tissue specificity; however, inner nuclear membrane proteins are often conserved and ubiquitously expressed in a variety of tissues (2). Mutations in genes encoding proteins of the nuclear lamina and transmembrane proteins of the inner nuclear membrane such as emerin, lamin B receptor, nesprins, MAN1, LAP2 and LAP1 have been associated with muscular dystrophies, cardiomyopathy, lipodystrophies and neuropathies (3). Therefore, a common feature of nuclear envelopathies covering at least 15 different phenotypes is the tissue-specific nature of the pathology. It is suggested that this tissue specificity arises from the tissue-specific interactions of ubiquitously expressed inner nuclear membrane proteins with other nuclear proteins or chromatin (3). Muscular dystrophies are inherited muscle disorders presenting with progressive muscle degeneration and weakness showing clinical and genetic heterogeneity (4). In Turkey, high consanguinity rates reaching 39% in some regions increase the incidence and prevalence of autosomal recessively inherited disorders (5,6). To develop efficient therapeutic strategies, the understanding of the molecular pathogenesis underlying these diseases needs to be improved. Loss of the inner nuclear membrane protein LAP1B (lamina-associated polypeptide 1, isoform B) causes autosomal recessive muscular dystrophy and cardiomyopathy (7–9). In mitotically active cells, LAP1 isoforms have been implicated in the DNA damage response, and in the regulation of nuclear envelope breakdown during mitosis (10–13). LAP1 has been shown to interact with lamin A/C and emerin, which are involved in myogenic differentiation (14,15). However, the role of LAP1 in normal muscle function is still poorly understood. 2 The aim of this thesis was to investigate the role of LAP1B in transcriptional regulation of muscle cells. For this purpose, an in vitro model of human LAP1B- related muscular dystrophy was established by using primary fibroblasts isolated from a patient, fibroblasts were myoconverted and their transcriptomic reprogramming throughout myogenic differentiation was profiled by RNA sequencing. Pathways underlying the impaired myogenic differentiation of LAP1B-mutant cells were identified. 3 2. BACKGROUND 2.1 Lamina-associated Polypeptide 1 (LAP1) 2.1.1 General Properties of LAP1 Lamina-associated polypeptide 1s (LAP1s) are type II transmembrane proteins localized in the inner nuclear membrane and identified for the first time in rat liver as three isoforms (LAP1A [75 kDa], LAP1B [68 kDa] and LAP1C [55 kDa]) generated by alternative splicing and interacting with nuclear lamina (15). In human, after identifying a sequence showing 73.6% homology to rat LAP1B, a sequence coding for human LAP1B was first cloned with its three distinct domains: a central transmembrane domain flanked by an amino-terminal nucleoplasmic domain and a carboxyl terminal lumenal domain (16). Human LAP1B has a length of 584 amino acid (aa), with a predicted molecular weight of 66.3 kDa. More recently, human LAP1C cDNA resulting from an alternative downstream transcriptional start site was also cloned (17). Consequently, human LAP1C is 462 aa long and has a predicted molecular weight of 52.5 kDa with a truncated N-terminal region compared to LAP1B (Figure 2.1.). Both transcripts are encoded by the human TOR1AIP1 gene located on chromosome 1 and composed of ten exons. So far, no human LAP1A isoform has been detected. Despite being ubiquitously expressed in a multitude of tissues such as liver, lung, kidney, spleen, brain and muscles, LAP1 isoforms display differential expression according to the differentiation level of tissues or cells examined. While LAP1C expression was found to be higher compared to LAP1B in undifferentiated cancer cell lines such as SH-SY5Y, HeLa and SKMEL-28; LAP1B is the predominant isoform in heart and brain, and its expression increased during murine neuronal and skeletal muscle differentiation (17,18). Strikingly, only LAP1B was detected in myonuclei in human skeletal muscle biopsies suggesting a critical role in these cells (7). 4 Figure 2.1. Structure of human TOR1AIP1 gene and LAP1 protein isoforms. On TOR1AIP1 gene, arrows indicate alternative transcription start sites, turquoise boxes show exons, light blue boxes show introns. UTR, untranslated region; TM, transmembrane domain. (Created on BioRender). 2.1.2 Protein Modifications and Interactors of LAP1 Components of the nuclear lamina are the first LAP1 protein partners to have been identified (15,19). In vitro, only LAP1A and LAP1B isoforms bind to A-type and B-type lamins (19). LAP1C interaction with nuclear lamina has not been observed in vitro; however, LAP1C was shown to co-precipitate with B-lamins under native conditions (20). In mitotic cells, LAP1s are phosphorylated and localized in the mitotic spindle and centrosomes (11). In a phenotypic screening of cell cycle-related genes in HeLa cells, knockdown of LAP1s have been associated with early mitotic delay caused by impaired spindle formation resulting in polylobed nuclei and apoptosis (12). Similarly, knockdown of LAP1 in SH-SY5Y cells (neuroblastoma cell line) resulted in a decrease in mitosis and altered centrosome positioning (11). Very recently, LAP1 has been shown to attach mitotic chromatin to the nuclear envelope during mitosis. After nuclear envelope breakdown, these membrane-chromatin interactions must be broken for proper sister chromatid segregation. However, overexpression of LAP1 have been associated with a failure in nuclear membrane removal from chromatin during mitosis, causing post-mitotic nuclear abnormalities (10). As both overexpression and knockdown of LAP1 have been shown to cause similar defects in mitosis, these findings demonstrate that expression level of LAP1 needs to be delicately balanced for cell cycle progression and mitosis. 5 The most extensively studied interactor of LAP1 is the AAA+ (ATPases associated with a variety of cellular activities) chaperone torsinA (21). After the identification of its role as a cofactor of torsinA, the human gene encoding LAP1 was renamed as TOR1AIP1 (torsinA-interacting protein 1) (22). TorsinA is an ATPase shuttling between ER lumen and perinuclear space, and whose mutation causes DYT1 dystonia, a severe neurological movement disorder. At the nuclear envelope, LAP1 stimulates ATPase activity of torsinA by forming heterohexameric ring structures and providing a catalytic residue at the interaction interface (23,24). Similar activation of torsinA at the ER is complemented by LULL1 (luminal domain like LAP1) which is the LAP1 homologue located at the ER (25). Although the role of LAP1 and torsinA association is not fully understood, nuclear envelope abnormalities observed in torsinA-mutant cells suggest putative functions such as nuclear envelope remodeling or RNA vesicular export by budding of the nuclear envelope (26). Four PP1 (protein phosphatase 1) binding motifs were identified in human LAP1B and LAP1C (17,27). Function of LAP1 dephosphorylation, substrates of LAP1:PP1 complex or LAP1 kinases have not yet been identified. LAP1 was shown to interact with the inner nuclear membrane protein emerin through their nucleoplasmic domains (14). Mutations in emerin leads to Emery- Dreifuss muscular dystrophy (EDMD) presenting with progressive skeletal muscle atrophy, joint contractures and cardiomyopathy (28). While perinatal lethality is observed in constitutive knockout of LAP1 in mice (29), conditional knockout in striated muscle is associated with muscular dystrophy and cardiomyopathy accompanied by premature death (14,30). As emerin-null mice fail to phenocopy EDMD, authors suggested that higher expression levels of LAP1 in mice compared to humans might compensate for the loss of emerin in mice. Recently, LAP1 was shown to interact with TRF2 (telomeric repeat-binding factor 2) upon DNA damage (13). While being a component of the telomeric ends protecting complex shelterin, TRF2 is able to activate DNA damage response at extra- telomeric sites. Under genotoxic stress induced by double-strand breaks, LAP1:TRF2 complex co-localized with ATM (ataxia-telangiectasia mutated protein kinase) and g- H2AX (phosphorylated H2A histone X) and in micronuclei and nuclear blebs. In 6 addition, LAP1:TRF2 association was shown to be dependent on ATM/ATR-mediated LAP1 phosphorylation. Interestingly, LAP1 upregulation was observed after induction of DNA damage response, probably as a necessity to compensate the loss of LAP1 at the nuclear envelope, that is relocated to damaged chromatin sites in the nucleoplasm (13). In silico interactome analysis of experimentally confirmed LAP1 interactors collected from databases revealed post-transcriptional modification, protein synthesis and gene expression among most significant pathways (31). Another in silico study on functional prediction of the disordered region (nucleoplasmic domain) of LAP1 with PSIPred found that LAP1 might be involved in transcriptional gene expression regulation with the highest probability scores (32). 2.1.3 Human LAP1 Mutations and Related Pathologies The first human LAP1 mutation was identified in two related cases affected by progressive proximal and distal muscular dystrophy, rigid spine, joint contractures and cardiomyopathy in one patient (7). A homozygous mutation (c.186delG) in TOR1AIP1 gene caused a premature stop codon and the complete absence of LAP1B isoform in patient’s skeletal muscles. Indeed, LAP1C was not affected by the mutation as this one was located upstream of the alternative transcriptional start site specific to LAP1C. Despite this, LAP1C was only detected in endomysial cells thus did not compensate for the loss of LAP1B in myofibers. Subsequently, compound heterozygous or homozygous mutations in TOR1AIP1 were identified in patients with similar muscular dystrophy and cardiomyopathy phenotypes (9,33). While no neuromuscular synaptic involvement was reported in these previous cases, a homozygous deletion in TOR1AIP1 was found to cause congenital myasthenic syndrome presenting with limb- girdle muscular weakness accompanied by impaired transmission at the neuromuscular junction (34). The spectrum of LAP1-related diseases was broadened with the identification of biallelic nonsense mutations affecting both LAP1 isoforms in individuals presenting with early onset multisystemic, progeroid-like syndrome with neurological impairment, heart defects, hearing loss, cataracts and skin manifestations (35,36). A 7 homozygous missense mutation in TOR1AIP1 was also proposed earlier as a probable cause of severe dystonia with cardiomyopathy and cerebellar atrophy (37). These reports suggest that phenotype severity is correlated with the type and localization of the mutations, as loss of both LAP1 isoforms has been associated to more severe phenotypes while the specific loss of LAP1B lead to a milder phenotype affecting mainly striated muscle (Table 2.1). Tissue-specific pattern of diseases is characteristic of nuclear envelopathies (3). This tissue selectivity has been proposed to arise from the fact that ubiquitously expressed nuclear envelope proteins might have tissue-specific interactions and functions making those tissues more vulnerable to particular mutations compared to others. Table 2.1. Human TOR1AIP1 mutations, affected isoforms and associated phenotypes. Data not confirmed with functional experiment was specified. 8 2.2 Regulation of In Vitro Myogenic Differentiation Skeletal myogenesis is the process of generating myofibers through differentiation of myogenic progenitor cells (38). In the dermomyotome, myogenic progenitors cells give rise to proliferating myoblasts. After several rounds of proliferation, myoblasts exit the cell cycle (postmitotic myocytes) and fuse to each other to generate multinucleated myotubes and finally mature myofibers expressing myosin (Figure 2.2). At the same time, a few quiescent adult muscle stem cells (satellite cells) are maintained as a stem cell pool for muscle homeostasis and repair (39). The steps of myogenesis are efficiently reproducible in vitro from pluripotent stem cells (directed differentiation) or even from differentiated cells through direct reprogramming by overexpression of myogenic regulatory factors (MRFs) (40). MRFs are a family of muscle-specific transcription factors that activate and regulate myogenesis; namely MyoD, Myf5, myogenin and MRF4. MRFs form heterodimeric complexes with E-protein family members and bind to E-box sequences located on regulatory regions of muscle-associated genes to regulate their expression. Myogenic differentiation is tightly regulated at the transcriptional level by hierarchical and temporal expression of MRFs (Figure 2.2). Myf5 and MyoD are the early MRFs governing commitment of muscle progenitor cells to myogenic differentiation while myogenin and MRF4 are late MRFs directing fusion of myoblasts and differentiation into mature myotubes. Figure 2.2. Myogenic differentiation and temporal expression pattern of MRFs (Created on BioRender). 9 MyoD was the first master transcription factor shown to be alone sufficient for committing cells to differentiation (41). Overexpression of MyoD in non-muscle primary cells such as fibroblasts, retinal epithelial cells, smooth muscle cells and chondroblasts; as well as fibroblast, liver, fat and nerve cell lines was shown to convert these cells into functional myoblasts and myotubes (42,43). MyoD-mediated myoconversion of patient’s fibroblasts is especially useful as a source not only for modeling muscular dystrophies but also assessing therapeutic approaches (44–47). 2.2.1 Differentiation Commitment and Cell Cycle Withdrawal It has been shown that MyoD drives myogenesis mainly by interacting with chromatin remodeling factors at muscle-specific gene loci to modulate chromatin accessibility (48). Two main axes operate in parallel during early myogenic commitment: induction of cell cycle exit and activation of muscle-specific gene transcription. MyoD cooperates with the homeodomain protein Pbx1 for binding to E-box sequences. Subsequently, SWI/SNF chromatin remodeling complex and p300 histone acetyltransferase are recruited to the locus to increase chromatin accessibility (49). In turn, myogenin acts as a strong transcriptional activator in these loci with opened chromatin. Therefore, chromatin immunoprecipitation analyses showed that MyoD and myogenin share largely overlapping occupancies on genome during myogenic differentiation (50,51). Activation of myogenin leads to an irreversible engagement towards cell cycle exit with upregulation of p21 (cyclin-dependent kinase inhibitor 1A), accumulation of hypophosphorylated pRb (retinoblastoma protein) and inhibition of E2F target genes (52). Interestingly, it has been demonstrated that pRb has a dual role in differentiation: while hypophosphorylated pRb leads to cell cycle exit, independently pRb cooperates with MyoD to induce transcription of muscle-specific genes (53). 10 2.2.2 Differentiation Checkpoint To prevent the formation of aberrant cells with genetic and genomic instability, a differentiation checkpoint needs to be overcome in terminally differentiated cells after cell cycle exit but before differentiation (54). In myoblasts, genotoxic stress upregulates the levels of p53 and p21, and significantly inhibits the formation of myotubes. Moreover, MyoD-mediated myoconversion of fibroblasts is also impaired upon induction of DNA damage (54). However, this differentiation inhibition was independent of cell cycle exit, and reversible as removal of genotoxic drugs rescued myogenic differentiation. As the guardian of the genome, p53 is also involved in the myogenic differentiation checkpoint (55). In response to genotoxic stress, p53 binds to the promoter of myogenin and represses its transcription. Again, myoblasts induced to differentiate after genotoxic stress effectively withdrew from the cell cycle demonstrating that this differentiation blockade is a process independent of cell cycle exit. 2.3 Role of LAP1 in Muscle To date, the role of LAP1s in muscle was investigated on several knockout mice lines: constitutive germline knockout of Lap1, Mck-driven (muscle creatine kinase with expression starting at embryonic day 17) and Myf5-driven (myogenic factor 5 with expression starting at embryonic day 8) conditional knockouts of Lap1. Germline knockout of LAP1s resulted in perinatal lethality demonstrating that LAP1 isoforms have essential roles in mouse embryonic development (29). In situ hybridization in Lap1-null embryos at E11.5 (embryonic day 11.5) showed reduction in the number of myogenin-positive somites and reduced mean myofiber area (18). Primary myoblasts isolated from Lap1-null embryos displayed normal proliferation rates but defective in vitro myogenic differentiation. Differentiated Lap1-null myoblasts displayed shorter and thinner myotubes, reduced fusion index and downregulated MyoD, myogenin and MyHC compared to controls (18). Striated muscle-conditional knockout of Lap1 driven by Mck promoter which 11 expression starts at E17, resulted in muscular atrophy starting by week 16 and premature mortality around week 30 (14). To determine the role of LAP1 from the beginning of early embryonic myogenesis which starts at E8 in mice, a striated muscle-selective knockout of Lap1 under the control of Myf5 promoter was investigated by Shin et al. (18) . All Lap1- depleted pups died at P17.5 (postnatal day 17.5) with hypotrophic myofibers but no atrophy. Postnatal muscle hypotrophy can be due to fusion defects reflected by downregulation of key myogenic factors such as Myf5, MyoD and Mef2C in mutant mice. In addition, reduced number of satellite cells and increased expression of p21 suggested that LAP1 might play a role in differentiation and proliferation of satellite cells (18). Finally, downregulation of AKT (protein kinase K) and upregulation of the atrophy-related protein MAFbx (muscle atrophy F box) pointed to an increased protein degradation in mutants, suggesting that LAP1 might be involved in the repression of catabolic pathways (18). 2.4 Research Question and Goals of the Thesis Despite being ubiquitously expressed in a multitude of tissues, the loss of the LAP1B isoform causes muscular dystrophy in humans and LAP1B is the only LAP1 isoform detected in mature myofibers (7). In silico functional prediction suggested a transcriptional regulatory role for LAP1 (32). Moreover, LAP1 interacts with lamin A/C and emerin, which are both involved in the regulation of myogenic gene expression (14). In this thesis, it has been hypothesized that LAP1B isoform is directly or indirectly involved in the transcriptional regulation of genes acting in myogenic differentiation. The goals of this thesis were to establish an in vitro model of human LAP1B-related muscular dystrophy, perform the transcriptome profiling of cells throughout myogenic differentiation, and characterize the mutant phenotype with molecular assays in the light of transcriptomics data. 12 3. MATERIALS AND METHODS 3.1 Materials 3.1.1 Cell Culture • Human primary fibroblasts, normal, adult, female, 38 years (Lot no. 64154595, Cat. No. PCS-201-012, American Type Culture Collection) • Human primary fibroblasts, normal, adult, male, 28 years (Lot no. 63792061, Cat. No. PCS-201-012, American Type Culture Collection) • Human primary fibroblasts, with homozygous c.186delG mutation in TOR1AIP1, male, 36 years • C2C12 murine myoblast cell line (American Type Culture Collection) • 1X Trypsin-EDTA (Capricorn) • Dulbecco’s Modified Eagle’s Medium (DMEM), High glucose (Gibco) • Ca2+ and Mg2+-free 1X phosphate buffered saline (PBS), (Biowest) • Fetal bovine serum (FBS) (Capricorn) • Dimethylsulfoxide (DMSO) (Applichem) • Doxycycline (Sigma-Aldrich) • Human recombinant insulin (Sigma-Aldrich) • 1X gentamycin (Biowest) • 1X amphotericin B (Biowest) 3.1.2 Cell Transduction and Transfection • Polybrene (Sigma-Aldrich) • hTERT lentiviral vector (pTrip_PGK_puromycin_polyA_CMV_htert variant 1) (Dr. Vincent Mouly) • MyoD lentiviral vector (pSINTREMyoD_hPGKrtT_A2SM2) (Dr. Vincent Mouly) • Lipofectamine 3000 Transfection Reagent (Thermofisher Scientific) • Puromycin (Sigma-Aldrich) 13 3.1.3 CRISPR/Cas9 Genome Editing and Cloning • pSpCas9(BB)-2A-Puro (pX459) V2.0 plasmid (Cat. No. 62988, Addgene) • Single-stranded oligonucleotides (PRZ Biotech) • 1X annealing buffer (400 mM Tris pH 8.0, 200 mM MgCl2, 500 mM NaCl, 10 mM EDTA pH 8.0) • BbsI-HF restriction enzyme (NEB) • Agarose (Applichem) • Tris-acetate-EDTA buffer (40 mM Tris base, 20 mM acetic acid and 1 mM EDTA disodium salt dihydrate) • Zymoclean Gel DNA Recovery Kit (Cat. No. D4001, Zymo Research) • NEB High Efficiency 5-alpha competent E. coli (NEB) • Luria-Bertani (LB) broth (10 g Tryptone, 10 g sodium chloride and 5 g Yeast Extract, pH 7.0, for 1 Lt.) • Agar (Sigma-Aldrich) • Ampicillin (Applichem) • T4 DNA ligase, 20 U/µl (NEB) • Adenosine 5′-triphosphate (ATP) disodium salt hydrate (Sigma-Aldrich) • DTT (1,4-Dithiothreitol) (Sigma-Aldrich) 3.1.4 DNA Isolation and DNA Sequencing • Nucleospin Plasmid Mini kit (Cat. No. 740499, Macherey-Nagel) • ZymoPURE II Plasmid Midiprep kit (Zymo Research) • Nuclei lysis buffer (10 mM Tris-HCl pH 8.0, 100 mM NaCl and 25 mM EDTA pH 8.0) • Sodium dodecyl sulfate (SDS) (Carlo-Erba) • 20 mg/ml proteinase K (Applichem) • BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) • Sequencing primers, 3.2 mM (PRZ Biotech) • 3M sodium acetate, pH 5 14 • Ethanol (Carlo Erba) • HD Formamide (Applied Biosystems) • 3130xl Genetic Analyzer (Applied Biosystems) • SimpliAmp thermal cycler (Thermofisher Scientific) 3.1.5 RNA Isolation and cDNA Synthesis • RNase ZAP (Ambion) • RiboEx Total RNA Solution (GeneAll Biotechnology) • Hybrid-R total RNA isolation kit (GeneAll Biotechnology) • NanoDrop 1000 spectrophotometer (Thermofisher Scientific) • Quantitect Reverse Transcription kit (Qiagen) 3.1.6 mRNA Sequencing • SENSE mRNA-seq Library Prep Kit for Ion Torrent (Lexogen) • Ion S5 Sequencing System (Ion Torrent) 3.1.7 Quantitative Real-time PCR • 2X SensiFAST SYBR No-ROX kit (Bioline) • Reverse and forward primers, 10 mM • Rotor-Gene 6000 instrument (Corbett Life Science) 3.1.8 Protein isolation and Quantitation • 1X PBS, Ca2+ and Mg2+-free (Biowest) • 1X RIPA (Radio-Immunoprecipitation Assay) buffer (Sigma-Aldrich) • Protease Inhibitor Cocktail (Roche) • 1 mM sodium orthovanadate (Sigma-Aldrich) • Vibra-Cell probe sonicator, VCX 130, probe CV 18 (Sonics) • Bicinchoninic Acid (BCA) Protein Assay Kit (Thermo Fisher Scientific) • SpectraMax M2 Microplate Reader (Molecular Devices) 15 3.1.9 Western Blotting • 40% acrylamide/bisacrylamide (Applichem) • Tris-base (Sigma-Aldrich) • Sodium dodecyl sulfate (SDS) (Carlo Erba) • Ammonium persulfate (APS) (Sigma-Aldrich) • TEMED (N,N,N′,N′-Tetramethylethylenediamine) (Sigma-Aldrich) • Mini Protean II electrophoresis set (Bio-Rad) • 4X Laemmli loading buffer (Bio-Rad) • Spectra Multi-Color Broad Range Protein Ladder (Thermofisher Scientific) • Semi-dry transfer instrument (Bio-rad) • Nitrocellulose membrane, 0.45 μm (Thermofisher Scientific) • Whatman filter paper, 3 mm (Sigma-Aldrich) • Ponceau S (Sigma-Aldrich) • Bovine Serum Albumin powder (Capricorn) • Tris-buffered saline, TBS (50 mM Tris, 154 mM NaCl, pH 7.6) • Tween 20 (Amresco) • Nonfat dried milk (Pınar) • SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermofisher Scientific) • GeneGnome5 chemiluminescence imaging instrument (Syngene) 3.1.10 Polymerase Chain reaction • Q5 High-fidelity DNA polymerase (NEB) • SimpliAmp thermal cycler (Thermofisher Scientific) • Forward and reverse primers, 10 mM (PRZ Biotech) • dNTP mix (Roche) 3.1.11 Agarose Gel Electrophoresis • Agarose (Biomax) 16 • 1X Tris-acetate-EDTA buffer, pH 8.0 • 6X loading buffer (Thermofisher Scientific) • Ethidium bromide • DNA molecular weight markers (100 bp or 1 kb), (NEB) 3.1.12 Immunofluorescence Staining • 1X PBS (Sigma-Aldrich) • 4% paraformaldehyde (PFA) (Sigma-Aldrich) • Ethanol (Carlo Erba) • Triton X-100 (Sigma-Aldrich) • FBS (Capricorn) • Bovine serum albumin (BSA) (Sigma-Aldrich) • DAPI (4’,6-diaminodino-2-phenylindole) (Sigma-Aldrich) • Prolong® Gold antifade reagent (Molecular Probes) • DM IL Inverted Fluorescence microscope with DFC320 camera (Leica) • Axio Plan Upright Fluorescence microscope with AxioCam Erc5 5Mp camera (Carl Zeiss) 3.1.13 Antibodies • Rabbit polyclonal anti-TOR1AIP1, for human LAP1 detection (Cat. No. HPA047151, Sigma-Aldrich) • Rabbit polyclonal anti-LAP1, for mouse LAP1 detection (a kind gift from Dr. William T. Dauer) • Mouse monoclonal anti-MyoD1, Clone 5.8A IgG1 (Cat. No. M5312, Agilent DAKO) • Mouse monoclonal anti-MF20, IgG2b (Developmental Studies Hybridoma Bank) • Rabbit polyclonal anti-phospho-Rb, Serine 807/811 (Cat. No. 9308, Cell Signaling Technology) • Mouse monoclonal anti-p21, Clone 187 (Cat. No. sc-817, Santa Cruz) • Mouse monoclonal anti-H2AX (Cat. No. ab22551, Abcam) 17 • Mouse monoclonal anti-myogenin, Clone F5D (Cat. No. sc-12732, Santa Cruz) • Mouse monoclonal anti-Rb, Clone IF8 (Cat. No. sc-102, Santa Cruz) • Mouse monoclonal anti-GAPDH (Cat. No. G8795, Sigma-Aldrich) • Mouse monoclonal anti-p53 (Cat. No., Cell Signaling Technology) • Goat anti-rabbit IgG (H+L) HRP-conjugated secondary antibody (Invitrogen) • Goat anti-mouse IgG (H+L) HRP-conjugated secondary antibody (Invitrogen) • Goat anti-rabbit IgG (H+L) secondary antibody, Alexa Fluor 488 (Invitrogen) • Goat anti-mouse IgG (H+L) secondary antibody, Alexa Fluor 488 (Invitrogen) 3.1.14 Cell Cycle Assay • Muse Cell Cycle Kit (Luminex Corporation) • NovoCyte flow cytometer (Agilent) • NovoExpress software (Agilent) 3.2 Methods 3.2.1 Cell Culture Dermal primary fibroblasts from two healthy individuals were purchased from ATCC. LAP1B-mutant primary fibroblasts (harboring homozygous c.186delG mutation) were cryopreserved materials previously isolated from a 36 year old man diagnosed with autosomal recessive LAP1B-related muscular dystrophy (OMIM phenotype no. 617072) after written informed consent. The study protocol was approved by Hacettepe University Non-interventional Clinical Researches Ethics Board (Decision no. GO 16/35) (Appendix 1). Primary fibroblasts were proliferated in a proliferation medium consisting of high-glucose (4.5 g/ml) DMEM with stable L-glutamine supplemented with 10% FBS, 50 µg/ml of gentamycin and 2.5 µg/ml of amphotericin B. Cells were maintained in 5% CO2 incubator at 37ºC. When needed, cells were cryopreserved at -80ºC or in liquid nitrogen in a medium consisting of 90% FBS and 10% DMSO. 3.2.2 Lentiviral Transduction 18 Primary fibroblasts were co-transduced with two lentiviral vectors: one vector expressing the catalytic subunit of hTERT (human telomerase reverse transcriptase) under the control of a CMV (cytomegalovirus) promoter, and one vector expressing MYOD1 under the control of a doxycycline-inducible Tet-On promoter (44). When cells reached 80% confluency in 6-well plates, proliferation medium was replaced with DMEM containing viral particles (MOI [multiplicity of infection] of 5 for hTERT vector, MOI of 10 for MYOD1 vector) supplemented with 10% FBS and 4 µg/ml of polybrene but no antibiotic/antimycotic. After an overnight incubation in 5% CO2 incubator at 37ºC, transduction mixes were aspirated and cells were washed 1X PBS. After being maintained for 48 hours in proliferation medium, transduced cells were selected in proliferation medium supplemented with 1 µg/ml of puromycin during 7 days. After selection, surviving single cell clones were proliferated and maintained in proliferation medium until high confluency. 3.2.3 Induction of Myogenic Differentiation Cells were grown in proliferation medium until >90% confluency. After washing with 1X PBS, cells were maintained in serum-free differentiation medium made with 4.5 g/ml glucose DMEM with stable L-glutamine supplemented with 2 µg/ml of doxycycline, 10 µg/ml of insulin and 1% gentamicin/amphotericin B. Throughout differentiation, medium was refreshed every two days. 3.2.4 Immunofluorescence Staining Cells were cultured on coverslips, washed with 1X PBS and fixed with 4% PFA or absolute ethanol for 10 min. After two consecutive washings with 1X PBS for 5 min., cells were blocked and permeabilized in 1X PBS containing 1% FBS and 0.5% Triton X-100 for 1 hour at room temperature. Primary antibodies were diluted in 1X PBS (1:50 for anti-MYOD1, 1:20 for anti-MF20, 1:5000 for gH2AX) and cells incubated at room temperature for 2 hours. After three washings for 5 min. with 1X PBS, secondary antibodies were diluted in 1X PBS at a ratio of 1:1000 and applied to cells for 1 hour at room temperature. Cells were again washed three times with 1X PBS, counterstained with 0.1 µg/ml of DAPI for 5 min. at room temperature and 19 mounted in mounting medium after two additional washes with 1X PBS. Micrographs were taken on an Axio Plan upright fluorescence microscope (Carl Zeiss) with an attached AxioCam Erc5 5Mp camera and ZEN 2 software. 3.2.5 Determination of Fusion and Differentiation Indices Fusion and differentiation indices were used as morphological parameters for the quantitation of myogenic differentiation (56). Differentiation index represents the number of nuclei in MyHC-expressing cells divided by the total number of nuclei. Fusion index was calculated by dividing the number of nuclei in MyHC-expressing cells with at least two nuclei by the total number of nuclei. Cells were differentiated in three independent experiments and fixed on days 0, 2, 5 and 8 of differentiation. After staining with MyHC and DAPI, ten fields of each sample were photographed on an Axio Plan upright fluorescence microscope (Carl Zeiss) with an attached AxioCam Erc5 5Mp camera, and nuclei were manually counted by using the ImageJ software. 3.2.6 CRISPR/Cas9-mediated Tor1aip1 Mutations in C2C12 Myoblasts Murine cells express three LAP1 isoforms (LAP1A, LAP1B and LAP1C) encoded by the single Tor1aip1 gene. LAP1A and LAP1B transcripts are encoded by a common promoter while LAP1C transcription is initiated at a downstream alternative promoter (Figure 3.1) (17). However, only two bands are detected in previous immunoblot analyses, one band of approximately 70 kDa representing LAP1A/B isoforms, and a second band around 50 kDa representing LAP1C (18). LAP1A/B expressions are increased during C2C12 differentiation while expression of LAP1C is stable, demonstrating a specific role of LAP1A/B isoforms in myogenic differentiation (18). In order to test the effect of the loss-of-function of LAP1A/B on myogenic differentiation in murine cells, exon 1 of Tor1aip1 was targeted with CRISPR/Cas9 for generation of double-strand breaks inducing non-homologous end joining repair, finally leading to indel mutations. Briefly, oligonucleotides coding for specific guide RNAs (gRNAs) were cloned into pX459 plasmid expressing Cas9 nuclease. After transfection to wild type C2C12 cells and positive selection with puromycin, single- cell clones were expanded and genotyped. Expression of LAP1 isoforms and the 20 myogenic differentiation potential of Tor1aip1-mutant C2C12 clones were compared to the parental wild type C2C12 clone. Figure 3.1. Structure of mouse LAP1 transcripts. Each box is numbered according to the corresponding exon in the mouse Tor1aip1 gene. Position of the start codon (ATG) is indicated for each transcript. In LAP1C transcript, grey boxes represents exonic sequences that are transcribed as UTRs (Created on BioRender). gRNA Design Mouse Tor1aip1 gene (Ensembl Gene ID: ENSMUST00000027738.13) is located on chromosome 1 at position 156,006,781 – 156,036,480 bp. The first exon of Tor1aip1 overlaps with the Tor1aip2 sequence (chromosome 1: 156,035,403- 156,068,861 bp) which is the Tor1aip1 paralogous gene. The region to be targeted in Tor1aip1 was selected by avoiding any interference with the coding sequence of Tor1aip2. CRISPRDirect (https://crispr.dbcls.jp/) online tool was used to select guide RNAs (gRNAs) targeting the first exon of Tor1aip1 (57). Potential off-targets were screened in mouse genome assembly GRCm38/mm10 and ‘NGG’ was selected as the PAM sequence requirement. Two distinct gRNAs (gRNA-1 and gRNA-2) with high specificity and low off-targets were selected (Figure 3.2). A pair of oligonucleotides containing BbsI restriction site for cloning and one additional 5’ G nucleotide for efficient transcription by U6 RNA polymerase were designed for each gRNA (Table 3.1). In addition, genotyping primers were designed for further screening of mutant clones (Table 3.1). 21 Figure 3.2. Locations of gRNAs binding sites on exon 1 of mouse Tor1aip1 gene. Sequences overlapping between Tor1aip1 and Tor1aip2 are shown in grey. gRNA binding sequences are shown in green and yellow, PAM sites on genomic sequence are shown in red. Forward (F) and reverse (R) genotyping primers locations are indicated. Table 3.1. Sequences of oligonucleotides used for gRNA synthesis and genotyping primers. The 5’ G nucleotide shown in red was included as a requirement of U6 promoters. gRNA binding sequences are highlighted in green and yellow. Oligonucleotide 1 Oligonucleotide 2 gRNA1 5’CACCGACCCGTCGCGCCGCGGACGA 3’ 5’ AAACTCGTCCGCGGCGCGACGGGTC 3’ gRNA2 5’CACCGTGTACGGCGACTTCGAGCCC 3’ 5’ AAACGGGCTCGAAGTCGCCGTACAC 3’ Forward Reverse Primer 5’ AGGTTGGGCCATCTACGTCA 3’ 5’ GGTCGAGAGCGAAGGTTGTAA 3’ Cloning of gRNA Sequences into CRISPR/Cas9-Expressing Vector Single-stranded oligonucleotides (0.04 µM each) were annealed in 1X annealing buffer with incubation at 95 ºC for 5 min. followed by cooling to 25 ºC with a decrement of 10 ºC/min. in a thermal cycler. One-step digestion/ligation reaction was 22 performed to simultaneously digest pX459 plasmid and oligonucleotides, and ligate them as follows: pX459 plasmid 100 ng Annealed oligonucleotides (0,04 µM) 2 µl 10X CutSmart buffer 2 µl 37 °C– 5 min. DTT (10 mM) 1 µl 16 °C – 5 min. ATP (10 mM) 1 µl BbsI enzyme (20 U/µl) 0,5 µl T4 DNA Ligase (20 U/µl) 0,5 µl Brought to 20 µl with nuclease-free water. For bacterial transformation, 45 µl of NEB 5-alpha E. coli competent cells were incubated with 5 µl of ligation product for 30 min on ice. After 1 min of heat-shock at 42 ºC in a water bath, cells were incubated for 15 min on ice. Cells were transferred to 950 µl LB at 37 ºC with 250 rpm agitation for 1 hour and spread to LB agar plates containing ampicillin. Single colonies grown after overnight incubation at 37 ºC were picked with sterile tips and inoculated to 4 ml of LB supplemented with ampicillin for overnight incubation at 37 ºC with 250 rpm agitation. Plasmid DNAs were purified using NucleoSpin Plasmid Mini Kit (Macherey-Nagel) according to the manufacturer’s protocol and sequenced using the universal hU6-forward primer. The Sanger sequencing reaction was set up in a thermal cycler as follows: 5X BigDye Sequencing Buffer 2 µl BigDye Ready Reaction Mix 1 µl hU6-forward primer (3.2 µM) 0.5 µl Plasmid DNA (150 ng/µl) 1 µl Brought to 10 µl with nuclease-free water. 96 °C – 1 min. 96 °C – 10 sec. 50 °C – 5 sec. 25 cycles 60 °C – 4 min. 4 °C – ∞ 30 cycles 23 Sequencing reaction products were purified with 1/10th volume of sodium acetate (3M, pH 5.0) and precipitation with 3 volumes of absolute EtOH at -80 °C for 1 hour. After centrifugation at 16,000 g for 15 min., pellet was re-washed with 1.5 volume of 70% EtOH. Finally, the DNA pellet was resuspended in 20 µl of formamide and loaded on 3130xl Genetic Analyzer for Sanger sequencing. Chromatograms were analyzed on SnapGene Viewer (GSL Biotech LLC). Plasmid DNA samples harboring the correct gRNA sequence were re-amplified in competent cells and purified with ZymoPURE II Endotoxin-free Plasmid Midiprep kit (Zymo Research) according to the manufacturer’s protocol. Transfection of C2C12 Cells and Single-cell Cloning 5.104 wild-type C2C12 cells/well were seeded on 6-well plates. At 50-60% confluency, cells were separately transfected with each of the plasmid DNAs (gRNA- 1-pX459 or gRNA-2-pX459) using Lipofectamine 3000 Transfection Kit (Invitrogen). For each sample (well), Solution A (125 µl of DMEM, 3.5 µl of Lipofectamine 3000 Reagent) and Solution B (125 µl of DMEM, 5 µl of P3000 Reagent and 2500 ng of plasmid DNA) were mixed and vortexed for 5 sec. After an incubation of 20 min. at room temperature, the transfection mix was added to the proliferation medium (devoid of antibiotic/antimycotic) onto cells and cells were maintained in 5% CO2 incubator at 37ºC. 24 hours later, medium was replaced by fresh proliferation medium supplemented with 2.5 µg/ml of puromycin. After 72 hours of antibiotic selection, surviving cells were trypsinized for 5 min. at 37 ºC, an aliquot was stained with 0.4% of Trypan Blue and counted on a hemocytometer. Cells were plated on 100 mm petri dishes at a density of 50 cells/dish and incubated for ten days in a 5% CO2 incubator at 37 ºC. Sterile rings made from cut pipette tips were placed on single-cell clones, cells were trypsinized, collected and seeded on T-25 flasks for further expansion. After grown to sub-confluency, half of the cells were centrifuged at 2,000 rpm for 5 min. Cell pellets were dissolved in 500 µl of nuclei lysis buffer, 50 µl of 10% SDS ve 10 µl of proteinase K (20 mg/ml) and incubated overnight at 55 ºC with agitation. After addition of 1/10th volume of sodium acetate (3M, pH 5.0) and incubation for 15 min. 24 at room temperature, cell extracts were centrifuged at 13,000 rpm at 10 ºC for 20 min. 1 ml of absolute EtOH was added to the supernatant transferred to a fresh tube and DNA was precipitated for 1 hour at -20 ºC. After centrifugation at 13,000 rpm at 10 ºC for 20 min., pellets were washed with 70% EtOH, centrifuged and dissolved in 100 µl of nuclease-free water. Mutations in Tor1aip1 were determined by PCR following Sanger sequencing as described above. 3.2.7 Total mRNA Sequencing Total RNA Extraction For RNA sequencing (RNA-seq), control 1, control 2 and LAP1B-mutant hTERT-MyoD cells were seeded as triplicates on three independent 6-well plates. When cells reached >90% confluency, differentiation was induced by shifting from proliferation medium to differentiation medium (see Section 3.2.3 for protocol). Cells at day 0, day 2, day 5 and day 8 after differentiation were collected in 1 ml of RiboEx Total RNA Solution (GeneAll Biotechnology). Total RNA was extracted using Hybrid-R total RNA isolation kit (GeneAll Biotechnology) according to the manufacturer’s protocol. RNA concentrations and quality were measured on NanoDrop 1000 spectrophotometer (Table 3.2). In addition, RNA samples were run on 1.2 % agarose gel electrophoresis for confirming integrity by the detection of three distinct bands representing ribosomal RNAs (Figure 3.3). In addition, no genomic DNA contamination was observed (Figure 3.3). 25 Table 3.2. Concentrations and OD values of RNA samples used in RNA-seq. Sample Sample identifier Concentration (ng/ul) OD 260/280 OD 260/230 LAP1B-mutant-D0-1 1 169.3 1.87 2.02 LAP1B-mutant-D0-2 2 144.4 1.92 1.68 LAP1B-mutant-D0-3 3 190.5 1.9 2.06 Control 1-D0-1 4 164.8 1.97 2.12 Control 1-D0-2 5 232 2.05 2.23 Control 1-D0-3 6 210.2 1.99 1.76 Control 2-D0-1 7 202.6 1.95 2.12 Control 2-D0-2 8 241.6 1.95 2.07 Control 2-D0-3 9 192.1 2.02 2.16 LAP1B-mutant-D2-1 10 202 1.78 1.07 LAP1B-mutant-D2-2 11 140.2 1.94 2.2 LAP1B-mutant-D2-3 12 183.4 1.99 2.2 Control 1-D2-1 13 204.7 1.94 2.13 Control 1-D2-2 14 233.9 2.04 2.05 Control 1-D2-3 15 294.3 2.03 2.21 Control 2-D2-1 16 280.1 1.84 1.08 Control 2-D2-2 17 172.1 1.95 1.97 Control 2-D2-3 18 200 1.86 1.24 LAP1B-mutant-D5-1 19 296 1.77 0.94 LAP1B-mutant-D5-2 20 157.6 1.99 2.09 LAP1B-mutant-D5-3 21 138.3 2.03 2.17 Control 1-D5-1 22 239.3 1.94 1.85 Control 1-D5-2 23 310.2 1.98 2.18 Control 1-D5-3 24 336.8 1.95 1.84 Control 2-D5-1 25 190.6 1.96 2.15 Control 2-D5-2 26 194.1 1.98 2.07 Control 2-D5-3 27 209.3 1.97 2 LAP1B-mutant-D8-1 28 178.4 2.05 2 LAP1B-mutant-D8-2 29 188.5 1.99 2.16 LAP1B-mutant-D8-3 30 185.8 2.02 1.99 Control 1-D8-1 31 425.1 2 2.13 Control 1-D8-2 32 398.9 2.03 2.15 Control 1-D8-3 33 400.2 2.02 2.12 Control 2-D8-1 34 344 1.86 1.16 Control 2-D8-2 35 231.5 2.01 2.13 Control 2-D8-3 36 255.1 2.01 2.19 26 Figure 3.3. Electrophoresis of selected total RNA extracts used for RNA-seq. MW, molecular weight marker. Bands corresponding to 28S, 18S and 5-5.8S rRNA were indicated. Total mRNA Sequencing Hacettepe University Advanced Technologies Application and Research Center (HÜNİTEK) constructed total mRNA sequencing libraries for 36 samples using SENSE mRNA-seq Library Prep Kit for Ion Torrent (Lexogen) and ran the samples on Ion Torrent S5 Sequencing System. RNA Sequencing Raw Data Processing Quality control and data processing were performed at Medical Informatics Department, Middle East Technical University. Quality control check of reads was performed with FastQC and adaptor sequences were removed and reads trimmed with Cutadapt (58,59). Samples were aligned to the human reference transcriptome GRCh38 (ftp://ftp.ensembl.org/pub/release-94/fasta/homo_sapiens/cdna/) with Salmon and transcript and gene abundances were calculated from read counts per transcript with the tximport package (60,61). After normalization with a variance stabilizing transformation, samples were compared on a distance matrix and outliers (samples 6, 21 and 32) removed (Figure 3.4). Differential expression and statistical significances were calculated using DESeq2 (62). 27 Figure 3.4. Distance matrices of RNA-seq samples before and after removal of outliers (framed in orange). The likelihood ratio test was used for statistical significance and the Benjamini- Hochberg method was used for multiple hypothesis correction with the False Discovery Rate sets to 0.05. Corrected p-values are referred as ‘q-values’. Genes were considered to be significantly differentially expressed if the q-value is <0.05 and the expression fold change is >2 (increased or decreased). Control 1 and control 2 samples were averaged and combined as a single control group, and compared to LAP1-mutant samples. Three different comparisons were performed for calculating differential expressions (Table 3.3). First, the two groups (Control versus LAP1B-mutant) were compared at each static timepoints (condition comparison). Secondly, differential expression between consecutive timepoints within each group was compared (consecutive timepoint comparison). Finally, time course analysis of differentially expressed genes (DEGs) was performed using the Generalized Linear Model as the regression model. In time course analysis, expression of genes at each day was compared to day 0;, allowing comparison of genes that are consistently downregulated or upregulated throughout differentiation between controls and mutant. 28 Table 3.3. Comparison types used for differential expression analyses. Condition comparison Consecutive timepoints comparison Time course analysis C vs. Mut : day 0 Day 0 vs. day 2 : Control Day 0 vs. day 2 : Control C vs. Mut : day 2 Day 2 vs. day 5 : Control Day 0 vs. day 5 : Control C vs. Mut : day 5 Day 5 vs. day 8 : Control Day 0 vs. day 8 : Control C vs. Mut : day 8 Day 0 vs. day 2 : Mutant Day 0 vs. day 2 : Mutant Day 2 vs. day 5 : Mutant Day 0 vs. day 5 : Mutant Day 5 vs. day 8 : Mutant Day 0 vs. day 8 : Mutant Gene Set Enrichment Analysis Gene identifiers and associated log2(fold change) values created for condition comparison and consecutive timepoints comparison were submitted to GeneTrail 3.0 (https://genetrail.bioinf.uni-sb.de). Kolmogorov-Smirnov test for Gene Set Enrichment Analysis was used to test if a certain category of ‘Gene Ontology (GO)- Biological Process’ and ‘Kyoto Encyclopedia of Genes and Genomes (KEGG) Pathways’ annotation is significantly enriched in the data (63). Differentially expressed genes identified in time course regression analysis were annotated using Enrichr (https://maayanlab.cloud/Enrichr/) by submitting genes with q-value<0.05 and an expression fold change of at least 4 (increased or decreased). Most significant categories of GO-Biological Process and KEGG Pathways were identified. 3.2.8 cDNA Synthesis and Quantitative Real-time PCR Quantitative real-time PCR (qPCR) was used to validate gene expression fold changes generated by RNA-seq. cDNA was synthesized from total RNA isolated as above using QuantiTect Reverse Transcription Kit (Qiagen) according to the manufacturer’s protocol. cDNA samples were diluted four times with nuclease-free water before use in qPCR. qPCR reactions were set up with specific primers for MYOG, CDKN1A, PTGS2 and ANKRD1 genes that were randomly selected using SensiFast SYBR No- 29 ROX kit (Zymo Research) on Rotor-Gene 6000 instrument (Corbett Life Sciences) (Table 3.4). GAPDH was used as housekeeping gene. Table 3.4. Sequences of primers used in qPCR. Gene Forward primer sequence Reverse primer sequence GAPDH 5’GGTCACCAGGGCTGCTTTTA 3 5’TTCCCGTTCTCAGCCTTGAC 3’ CDKN1A 5’AGGTGGACCTGGAGACTCTCAG 3’ 5’TCCTCTTGGAGAAGATCAGCCG 3’ PTGS2 5’GTACTCCCGATTGAAGCCCC 3’ 5’AAGCCTAATGTGGGGACAGC 3’ ANKRD1 5’GAAAAGCGAGAAACAACGAGA 3’ 5’GGTTCCTTTACAACTGGAACTT 3’ MYOG 5’AAACTACCTGCCTGTCCACC 3’ 5’ACGGACACCGACTTCCTCTT 3’ Conditions of the amplification reaction were as follows: 2X SYBR Mix No-ROX 5 µl Forward primer (10 µM) 0.4 µl Reverse primer (10 µM) 0.4 µl cDNA 2 µl Nuclease-free water 2.2 µl 95 °C – 2 min. 95 °C – 15 sec. 60 °C – 15 sec. 40 cycles Melting condition: from 72 °C to 95 °C Experiments were performed with three biological replicates and two technical replicates. Mean Ct values were obtained from RotorGene 6000 Series Software 1.8 (Corbett Life Sciences). Relative expression fold changes were calculated by tracing a standard curve and the DDCt method, and gene expression was normalized to GAPDH. In graphics, error bars were presented as mean±SEM. 3.2.9 Protein Isolation and Quantitation Cells seeded on 6-well plates were washed one time with cold 1X PBS and collected in 30 µl of RIPA buffer supplemented with protease (1X Protease Inhibitor Cocktail) and phosphatase inhibitors (1 mM sodium orthovanadate) on days 0, 2, 5 30 and 8 of differentiation. Samples were sonicated 6 times at 50% amplitude for 20 sec. on a probe sonicator on ice and centrifuged at 14,000 rpm for 15 min. at -4 °C. Pellets containing cell debris were discarded and protein supernatants transferred to a fresh tube. Protein were quantified using BCA Protein Assay Kit (Thermofisher Scientific) according to the manufacturer’s protocol. Briefly, a standard curve was traced using protein standard samples consisting of 0.08, 0.1, 0.2, 0.4, 0.8 and 1.2 mg/ml of BSA. Protein samples to be quantified were dilutes ten times with 1% SDS. All samples were incubated with the working solution at 37 °C for 30 min. and measured on SpectraMax M2 spectrophotometer at 562 nm. 3.2.10 Western Blotting For SDS-polyacrylamide gel electrophoresis, 12% or 14% resolving gels and 4% stacking gels were prepared as shown in Table 3.5. Table 3.5. Composition of gels used in SDS-PAGE. 12% resolving gel 14% resolving gel 4% stacking gel Distilled water 3.4 ml 3ml 3.1 ml 40% acrylamide/bisacrylamide 2.4 ml 2.8 ml 0.5ml 1.5 M Tris pH 8.8 2 ml 2 ml - 0.5 M Tris pH 6.8 - - 1.25ml 10% SDS 0.08 ml 0.08 ml 0.05 ml 10% APS 0.08 ml 0.08 ml 0.05 ml Temed 0.008 ml 0.008 ml 0.005 ml 30 µg of protein extracts were mixed with 4X Laemmli buffer, denaturated at 100 °C for 4 min. and loaded on SDS-PAGE gels with protein molecular weight marker. Electrophoresis was performed for 3 hours at 100 V. For wet transfer, gel and nitrocellulose membrane were sandwiched between four Whatman filter papers and two sponges in the transfer cassette. Wet transfer was performed at 120 V for 1.5 hours at 4 °C. For checking transfer efficiency, the nitrocellulose membrane was reversibly stained with Ponceau S before blocking. Membranes were blocked with either 5% nonfat dried milk or 5% BSA in 0.1% Tween 20 in TBS (TBS-T) for 1 hour at room 31 temperature. Primary antibodies were diluted in 0.1% TBS-T (1:500 for anti-MyoD, 1:2000 for anti-TOR1AIP1, 1:5000 for anti-GAPDH, 1:200 for anti-p21, 1:200 for anti-myogenin, 1:200 for anti-Rb, 1:1000 for anti-phosphoRb, 1:1000 for anti-p53) and membrane was incubated with antibodies overnight at 4 °C. After three washes for 10 min. with 0.1% TBS-T, membranes were incubated with HRP-conjugated secondary antibodies diluted 1:3000 in 0.1% TBS-T for 1 hour at room temperature. After three washes for 10 min. with 0.1% TBS-T, membranes were stained with SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermofisher Scientific) according to the manufacturer’s protocol. Chemiluminescent imaging was performed on GeneGnome 5 instrument (Syngene). 3.2.11 Cell Cycle Assay Cells seeded on three independent 6-well plates were washed two times with 1X PBS, harvested by trypsinization and days 0, 1 and 3 of differentiation when all cells are still mononucleated. After centrifugation at 300 g for 5 min., cell pellets were washed once with 1X PBS and fixed in 1 ml cold absolute EtOH added dropwise when vortexing. After an overnight incubation at -20 °C, fixed cells were centrifuged at 14,000 rpm for 8 min. and incubated with Muse Cell Cycle Reagent (Luminex Corporation) for propidium iodide staining for 30 min. at room temperature. Results were acquired on NovoCyte flow cytometer from at least 10,000 events per sample. Gating and analyses were performed with NovoExpress software. Primary gate (P1) was on forward scatter height (FSC-H) against side scatter height (SSC-H), secondary gate (P2) was on FSC-H against FCS-area and tertiary gate (P3) was on PE-area against width. Cells were selected by excluding cell debris. The gating strategy used for ‘control-D0-replicate1’ sample was exemplified in Figure 3.5. 32 Figure 3.5. Gating strategy used for data analysis in flow cytometry. Total cells were first gated (P1) on forward scatter height (FSC-H) against side scatter height (SSC-H). Secondary gate (P2) was on FSC-H against FCS-area and tertiary gate (R3) was on PE-area against width. Population frequencies at each phase of the cell cycle (Freq G1, Freq S, Freq G2) was expressed as percent of the total gated population. 3.2.12 Micronuclei Count Control and LAP1B-mutant cells were fixed at differentiation days 0, 5 and 10, nuclei stained with 0.1 µg/ml of DAPI for 5 min. at room temperature and mounted in mounting medium after two washes with 1X PBS. Micrographs were taken on an Axio Plan upright fluorescence microscope (Carl Zeiss) with an attached AxioCam Erc5 5Mp camera and ZEN 2 software. 33 4. RESULTS 4.1 Immortalization and Myoconversion of Primary Fibroblasts Control 1, control 2 and LAP1B-mutant primary fibroblasts were immortalized with hTERT and myoconverted with stable genomic integration of an inducible MYOD gene. Before proceeding to subsequent experiments, expression of both LAP1 isoforms in control immortalized fibroblasts, and absence of LAP1B isoform in fibroblasts isolated from the patient were confirmed by western blotting (Figure 4.1). Figure 4.1. Expression of LAP1 isoforms (LAP1B and LAP1C) in control 1 immortalized (indicated as +hTERT) fibroblasts, LAP1B-mutant primary and immortalized fibroblasts as shown by western blotting. GAPDH was used as loading control. 4.1.1 Determination of Transduction Efficiency After transduction of primary cells with hTERT and MyoD lentiviral vectors, MyoD expression was induced by incubating cells in differentiation medium supplemented with doxycycline. After 18 hours of induction, cells were fixed and stained with anti-MyoD antibody and DAPI (Figure 4.2). 34 Figure 4.2. MyoD expression in controls and LAP1B-mutant immortalized myoconverted fibroblasts induced with doxycycline (+Dox) for 18 hours as shown by immunofluorescence staining. Nuclei were stained with DAPI. Scale bar: 20 µm. The number of nuclei positive for MyoD was compared to the total number of nuclei for assessing transduction efficiency. Percentages of MyoD-positive nuclei in Control 1, Control 2 and LAP1B-mutant cells were respectively 83%, 87% and 83% (Figure 4.3A). In addition, no illegitimate MyoD expression was detected in fibroblasts incubated in differentiation medium devoid of doxycycline for 18 hours (Figure 4.3B). Expression of MyoD at the transcript level throughout differentiation has been confirmed by RNA-seq and at the protein level by western blotting (Figure 4.3C-D). These results showed that transduction efficiencies were comparable between all samples and MyoD expression was specifically induced upon doxycycline addition. 35 Figure 4.3. (A) Percentages of MyoD-positive nuclei in control and LAP1B-mutant cells after 18 hours of doxycycline induction. N represents the number of counted nuclei. (B) MyoD expression in LAP1B-mutant cells incubated in differentiation medium without doxycycline (-Dox) for 18 hours. Nuclei were stained wit DAPI. Scale bar: 20 µm. (C). MyoD mRNA expression levels throughout differentiation as assessed by RNA-seq. (D) MyoD protein levels in control and LAP1B-mutant cells at days 0 and 2 as shown by western blotting. GAPDH was used as loading control. 4.1.2 Myoconversion of Fibroblasts To investigate their myogenic differentiation potential, Control 1 and LAP1B- mutant fibroblasts were differentiated for 8 days and stained with anti-MF20 antibody which recognizes myosin heavy chain isoforms (MyHC, muscle markers) at days 0, 2, 5 and 8 after differentiation (Figure 4.4). At day 0, as expected, no MyHC expression was detected in both control and LAP1B-mutant fibroblasts. At day 2, a few mononucleated cells expressing MyHC were detected in control, indicating that cells were committed to myogenic differentiation and started to express muscle markers. However, no MyHC expression was detected in mutant cells at this stage. At day 5, control cells containing multiple nuclei and displaying increased MyHC expression indicated the beginning of myogenic fusion, while a few mononucleated LAP1B- 36 mutant cells just started to express MyHC. Finally, at day 8, large multinucleated myotubes were visible in control cells (Figure 4.5) while the number of MyHC- expressing LAP1B-mutant cells was still very low. Figure 4.4. Monitoring of myogenic differentiation of myoconverted fibroblasts by MyHC expression at day 0 (D0), D2, D5 and D8 of differentiation. Nuclei were stained with DAPI. Scale bar: 20 µm. 37 Figure 4.5. MyHC expression in control multinucleated myotubes at day 8. Nuclei were stained with DAPI. Scale bar: 10 µm. To test whether myogenic differentiation was delayed in LAP1B-mutant cells, cells were further cultured in differentiation medium for a total of 17 days (Figure 4.6). At day 10, cells started to detach from the plate and die. At day 17, >80% of cells were dead with only 28% of surviving cells expressing MyHC. Figure 4.6. Morphology of LAP1B-mutant cells differentiated for 17 days. Micrographs were taken in brightfield microscope at D0, D10 and D17 of differentiation. The fourth panel shows MyHC expressing cells (stained with anti-MF20) at D17, nuclei were stained with DAPI. Scale bar: 10 µm. These results show that control cells were efficiently committed to myogenic differentiation and were able to form mature myotubes within 10 days, whereas LAP1B-mutant cells, despite efficient induction of MyoD expression, failed to fully differentiate and eventually died starting from day 10. 38 4.1.3 Determination of Differentiation and Fusion Indices To compare differentiation potential in control and LAP1B-mutant cells, differentiation and fusion indices were calculated by MyHC and DAPI stainings at days 5 and 10 of differentiation as described in Section 3.2.3 (Figure 4.7). At day 5 and day 10, both differentiation and fusion indices were significantly lower in LAP1B- mutant cells compared to control cells (** indicates p<0.01, *** p<0.001, Mann- Whitney U, n = 3). At day 10, 16% of LAP1B-mutant cells expressed MyHC but only 2% of these cells contained more than one nucleus, indicating a major fusion defect. Figure 4.7. (A) Differentiation and (B) fusion indices of control and LAP1B-mutant cells at days 5 and 10 of differentiation. (** indicates p<0.01, *** p<0.001, Mann-Whitney U, n = 3). 4.2 CRISPR/Cas9-mediated Tor1aip1 Mutations in C2C12 Myoblasts To support that the impaired myogenic differentiation phenotype observed was specific to the absence of LAP1B and did not reflect any other intrinsic variability between fibroblasts isolated from controls and patient, two C2C12 murine myoblast cell lines harboring mutations in Tor1aip1 were created by CRISPR/Cas9. Murine LAP1A and LAP1B are encoded by a common promoter whereas LAP1C possesses an alternative transcriptional start site (17). LAP1A is the longest transcript and LAP1B differs only by a skipped exon. In previous work, LAP1A/B were detected as a single protein band in western blotting (18). For these reasons, it has not been possible to target a region affecting only LAP1B transcript in Tor1aip1 gene. Therefore, exon 1 encoding LAP1A/B transcripts was targeted to generate a loss-of- 39 function mutation. For this purpose, two gRNAs were designed and individually cloned into CRISPR/Cas9-expressing vector pX495. Wild type C2C12 cells were separately transfected with CRISPR/Cas9-expressing vectors and selected with puromycin. Single-cell clones were further expanded, genotyped and monitored for their differentiation capacity. 4.2.1 Generation of CRISPR/Cas9-expressing Vectors After transformation of competent bacteria, the presence of the cloned gRNA fragment in plasmids of 10 bacterial colonies were tested by colony-PCR (Figure 4.8A). A band at the expected size (110 bp) corresponding to the cloned gRNA fragment was detected in all tested colonies. One plasmid DNA for each gRNA was sequenced and no variation was detected in gRNA-coding sequences (Figure 4.8B, C). Figure 4.8. Cloning of gRNA-coding sequences into CRISPR/Cas9-expressing vectors. (A) Electrophoresis results of colony-PCR of plasmids (samples 1-5 are gRNA1-cloned vectors, 6-10 are gRNA2-cloned vectors). MW, molecular weight marker; Neg., negative control (colony-PCR performed with empty vector). Chromatograms showing the correct insertion of (B) gRNA1- and (C) gRNA2-coding sequences between U6 promoter and gRNA scaffold sequences within the backbone vector. 40 4.2.2 Genotyping of Mutant Clones and Determination of LAP1 Expression Levels After transfection and selection, two surviving C2C12 myoblast clones were isolated and expanded for genotyping. By sequencing, a heterozygous frameshift was detected starting from position c.192A in Tor1aip1 gene in mutant clone 1 (mutClone1) (Figure 4.9). Unfortunately, it has not been possible to resolve the exact sequence of alternate alleles in this clone, probably because of the polyploid nature of some C2C12 clones causing mutations in more than one alternate alleles for the same locus. In mutant clone 2 (mutClone2), a biallelic 28 bp deletion and 22 bp insertion at position c.214C in Tor1aip1 was detected (Figure 4.9). This mutation has been predicted to lead to the inframe deletion of 10 amino acids (PRAAKERSPG) and the insertion of 8 amino acids (FSGRRTGR) at the protein level. Figure 4.9. Chromatograms showing mutations in exon 1 of Tor1aip1 gene in mutant C2C12 clones. Wild type sequence is shown in the upper panel. In mutant clone 1, heterozygous frameshift mutation was observed. In mutant clone 2, homozygous 28 bp deletion and 22 bp insertion was detected. 41 Expression levels of LAP1 isoforms in mutant C2C12 myoblasts clones were compared to wild type C2C12 cells at days 0 and 5 after differentiation (Figure 4.10). In wild type cells, LAP1C was the predominant isoform at day 0 corresponding to proliferating myoblasts. After 5 days of differentiation, an increase in the expression of LAP1A/B was observed in wild type cells compared to day 0. In both Tor1aip1- mutant clones, LAP1A/B expression was significantly decreased compared to wild type at days 0 and 5. At day 5, LAP1A/B was downregulated by 11-fold in mutClone1 and by 8.3-fold in mutClone2. In mutant clone 1, expression of LAP1C was also slightly downregulated by approximately 2-fold (* indicates p<0.05, *** represents p<0.001, two-way ANOVA, Bonferroni’s post-hoc test, n=3). In conclusion, CRISPR/Cas9-mediated gene editing of Tor1aip1 in C2C12 myoblasts effectively lead to a significant downregulation of LAP1A/B isoforms accompanied by a smaller but significant decrease in LAP1C expression in mutClone1. Figure 4.10. LAP1 isoforms expression in Tor1aip1-mutant C2C12 clones at days 0 and 5 of differentiation as assessed by western blotting. Expression levels of LAP1A/B and LAP1C isoforms were compared between wild type and mutant clones. LAP1s expression was normalized to GAPDH. (* indicates p<0.05, *** represents p<0.001, two-way ANOVA, Bonferroni’s post-hoc test, n = 2). 42 4.2.3 Myogenic Differentiation Potential of Tor1aip1-mutant C2C12 Myoblasts Wild type and mutant C2C12 clones were differentiated and stained for monitoring MyHC expression at day 5 of differentiation (Figure 4.11A). Differentiation and fusion indices were compared between wild type and mutant clones (Figure 4.11B). Differentiation and fusion indices were significantly decreased in mutant clones compared to wild type (p=0.0357 for WT vs. mutClone1, p=0.0159 for WT vs. mutClone2, Mann-Whitney U, n=3). Figure 4.11. (A) Micrographs of WT and Tor1aip1-mutant C2C12 clones differentiated for 5 days and stained with MF20 antibody for MyHC expression. Nuclei were stained with DAPI and images merged. Scale bar: 20 µm. (B) Differentiation and fusion indices of WT and Tor1aip1-mutant C2C12 myoblasts differentiated for 5 days (* indicates p<0.05, Mann-Whitney U test, n=3). As a result, the impairment of myogenic differentiation observed in LAP1B- mutant myoconverted fibroblasts has been phenocopied by LAP1A/B knockdown in C2C12 myoblasts. 43 4.3 Transcriptome Profiling For transcriptome profiling of control and LAP1B-mutant cells throughout myogenic differentiation, total RNA was extracted at four timepoints representing key stages of myogenic differentiation: day 0 for proliferating fibroblasts, day 2 for commitment to differentiation, day 5 for fusion and day 8 for mature myotubes. To begin, an overall analysis of the whole transcriptome data was performed by comparing the number of differentially expressed genes at different stages. After validation of expression fold changes of selected genes by qPCR, enriched pathways represented by differentially expressed genes were identified by gene set enrichment analyses. In addition, expression fold changes of MyoD target genes in LAP1B-mutant cells were analyzed to gain more insight into transcriptional changes underlying their defective myogenic differentiation. 4.3.1 Statistics of Differential Expression Analysis The number of differentially expressed genes (DEG) in LAP1B-mutant cells compared to controls was determined at each day of differentiation (Figure 4.12A). The average number of genes mapped and analyzed for statistical significance was 22,607 ± 196 per day. Genes were considered to be significantly differentially expressed when p<0.05 with a minimum of 2-fold increase or decrease. In LAP1B- mutant proliferating fibroblasts at day 0, a total of 1,731 genes were found to be differentially expressed compared to controls; of these, 1,094 were downregulated and 637 upregulated. At day 2, 3,419 genes were differentially expressed in LAP1B- mutant cells committed to differentiation, with 1,224 downregulated genes and 2,195 upregulated genes. At day 5, 1,519 genes were differentially expressed with 1,087 downregulated and 432 upregulated genes. Finally, at day 8, 2,979 genes were differentially expressed with 1,726 of them being downregulated and 1,253 upregulated. 44 Figure 4.12. (A) Number of DEGs in LAP1B-mutant cells compared to control at each day of differentiation. (B) Number of DEGs between consecutive days throughout differentiation within control or LAP1B-mutant cells. To gain more insight into the differences underlying the course of myogenic differentiation between control and LAP1B-mutant cells, DEGs between consecutive timepoints were separately determined within control and LAP1B-mutant cells (time points comparison) (Figure 4.12B). In control cells, 3,246 genes were differentially expressed between D0 and D2, corresponding to withdrawal from the cell cycle and differentiation commitment. During the transition between D2 and D5 corresponding to the stage of fusion, 562 genes were differentially expressed while only 153 differentially expressed genes were found between D5 and D8, when mature myotubes were formed. As expected, the most extensive transcriptional reprogramming occurred during the first transition when cells were committed to differentiation. During subsequent transitions (D2 vs. D5 and D5 vs. D8), the number of DEGs was sharply reduced reflecting the completion of the differentiation program at the transcriptomic level. A similar pattern was observed in LAP1B-mutant cells, except that the number of DEGs at each transition was higher compared to controls. In mutant cells, the numbers of DEGs were 4,867; 2,650 and 1,128 at D0 vs. D2, D2 vs. D5 and D5 vs. D8 transitions, respectively. Finally, the time course regression analysis identified 798 genes that were significantly differentially expressed between control and LAP1B-mutant cells 45 throughout differentiation (Figure 4.13). Although some DEGs were shared between control and mutant, a cluster heat map generated from expression fold changes of DEGs revealed two clusters containing genes displaying opposite expression patterns in control and LAP1B-mutant cells. In Cluster 1, some of the genes that were upregulated throughout differentiation in control cells, remained unchanged or were downregulated in LAP1B-mutant cells (Figure 4.13, Cluster 1). Conversely, Cluster 2 contained some genes that were upregulated in LAP1B-mutant cells upon induction of differentiation, but remained mainly unchanged or were downregulated in controls (Figure 4.13, Cluster 2). These aberrantly expressed genes might reflect defects underlying the impaired myogenic differentiation of LAP1B-mutant cells. Figure 4.13. Cluster heat map of DEGs in control and LAP1B-mutant cells determined by time course regression analysis. Expression fold changes at each day were compared to day 0 (D0). Downregulated genes are shown in blue, upregulated genes are shown in yellow. 46 4.3.2 Validation of RNA-seq Data by qPCR To validate differential expression data generated by RNA sequencing, three genes (PTGS2, ANKRD1 and MYOG) were randomly selected at different days of differentiation and analyzed by qPCR (Figure 4.14). By RNA-seq, PTGS2 at day 2 was 3.58-fold upregulated, while ANKRD1 at day 5, MYOG at day 2 and MYOG at day 5 were respectively 15.5-fold, 130-fold and 14.7-fold downregulated in LAP1B- mutant cells compared to controls. qPCR results for PTGS2, ANKRD1, MYOG-D2 and MYOG-D5 showed 1.68-fold increase, 30.53-fold decrease, 31-fold decrease and 3.4- fold decrease, respectively. Figure 4.14. Validation of RNA-seq data by qPCR. Log2 values of expression fold changes obtained from RNA-seq and qPCR of selected genes in LAP1B- mutant cells compared to controls were plotted. Fold change of each selected gene was normalized to GAPDH (n=3). As a result, expression fold changes of selected DEGs obtained from RNA-seq and qPCR were concordant in terms of upregulation or downregulation, with a variation in the magnitude of the fold change which is often observed in RNA-seq validation experiments (64). PT GS 2-D 2 AN KR D1 -D 5 MY OG -D 2 MY OG -D 5 -8 -6 -4 -2 0 2 lo g 2 (fo ld c ha ng e) in L A P 1B -m ut an t/c on tro ls RNA-Seq qPCR 47 4.3.3 Gene Set Enrichment Analysis Gene Set Enrichment Analysis (GSEA) was used to determine which Gene Ontology-Biological Process and KEGG annotation categories were significantly (adjusted p<0.05) enriched in the ranked gene lists. To begin, DEGs generated by the comparison of LAP1B-mutant and control cells at each day of differentiation were analyzed. However, pathways identified in this comparison were not informative in terms of the identification of potential pathways involved myogenic differentiation. Analysis of DEGs during consecutive days of differentiation within control and mutant cells, and time course data has been more informative as this approach revealed pathways that were uniquely enriched in each group during their own differentiation course. Condition Comparison at Static Timepoints First, DEGs in LAP1B-mutant cells compared to controls at each day of differentiation were analyzed (condition comparison at static timepoints) (Figure 4.15 and Appendix 2). DEGs at day 0 reflected transcriptomic differences at the basal level between mutant and control proliferating fibroblasts before induction of differentiation. At day 0, most significantly enriched categories in mutant cells compared to control were related to viral response, probably arising from variation in viral transduction rates (Figure 4.15, Day 0). At day 2, upon induction to myogenic differentiation, genes associated with RNA processing, protein modifications and ubiquitination were enriched in LAP1B-mutant cells compared to control (Figure 4.15, Day 2). At day 5, generation of precursors metabolites and energy was the only significantly enriched category in LAP1B-mutant cells compared to control (Figure 4.15, Day 5). Finally, at day 8, genes involved in metabolites generation and oxidative phosphorylation were enriched in LAP1B-mutant cells compared to controls (Figure 4.15, Day 8). 48 Figure 4.15. Top ten most significantly enriched categories containing DEGs in LAP1B-mutant cells compared to controls at day 0, day 2, day 5 and day 8 of differentiation. GO BP, Gene Ontology Biological Process; KEGG, Kyoto Encyclopedia of Genes and Genomes. Pathway analysis of DEGs at static timepoints did not reveal potential pathways that might contribute to intrinsic differences between control and LAP1B- mutant cells at the basal level, at day 0. Pathways enriched in LAP1B-mutant cells at subsequent days indicated differences in RNA processing, protein degradation and energy metabolism compared to control. However, as these biological processes cover a wide range of pathways and mechanisms, it has been difficult to interpret these results in terms of deregulation of myogenic differentiation. Consecutive Timepoints Comparison for Each Condition To gain more insight into the transcriptomic changes underlying the impaired myogenic differentiation observed in LAP1B-mutant cells, it has been reasoned that analyzing genes differentially expressed throughout differentiation within control and LAP1B-mutant cells separately, and then comparing associated enriched pathways would be more informative (Figure 4.16, Appendix 2). 49 Figure 4.16. Top ten most significantly enriched categories containing DEGs between consecutive timepoints (Day 0 vs. Day 2, Day 2 vs. Day 5 and Day 5 vs. D8) within control or LAP1B-mutant cells. In control cells, upon induction of differentiation (D0 vs. D2), genes associated with muscle function (muscle contraction, filament sliding) and cell division (cell cycle phase transition, chromosome organization) were enriched. Subsequently, DEGs during D2 vs. D5 transition of control cells were enriched for muscle contraction and actomyosin filament sliding, indicating synthesis of sarcomeric proteins. No significantly enriched category was found by GSEA of the 153 genes differentially expressed during the final stage of differentiation in control (D5 vs. D8); therefore, these genes were still annotated by using overrepresentation analysis. DEGs in controls at D5 vs. D8 were associated with extracellular matrix. During the first ph