Detecting and Visualising Proteins in Xenopus Embryos

DETECTING AND VISUALISING PROTEINS IN XENOPUS EMBRYOS

Maya Zita Piccinni

This thesis is submitted in partial fulfilment of the requirements for the award of the degree of Doctor of Philosophy of the University of

Portsmouth.

September 2018

Declaration

Whilst registered as a candidate for the above degree; I have not been registered for any other research award.

The results and conclusions embodied in this thesis are

the work of the named candidate and have not been

submitted for any other academic award.

WORD COUNT 62974

Acknowledgements

Is my pleasure to give my special thanks to Peter Coxhead, Darren

Gowers and Matt Guille for their support through my degree at

Portsmouth, and for believing in me all these years. To Matt Guille and Colin Sharpe, for supporting me through the doctorate years;

Colyn Crane-Robinson and Jerome Swinny for their valuable input and patience during our six annual meetings; to Sergei Sokol for allowing me to work in his lab and Keiji Itoh, Jonathan Chamberlain,

Martin Devonshire, Simon Streeter, Samantha Griffiths and Liliya

Nazlamova for their help in the practical aspects of research, to

Anna Philpott for her intellectual input and Victoria Allan for helping us define the kash5 sequence. Very special thanks to my family,

Hari, Bali, Mamá, Papá, Alex and children Eirwen, Skye and Edryd, who have been incredibly supportive and patient with me all this time.

III

Confidentiality and ethical consideration

Ethics Statement

All experiments and procedures in Xenopus are conducted under the British Home Office

PPL 70/6450 held by Professor Matthew Guille in the “European Xenopus Resource

Centre” following the Animal (Scientific Procedures) Act of 1986. The animals were regularly overseen by a Named Veterinary Surgeon.

All llama vaccination experiments are executed according to UK and EU animal welfare legislation and after approval of the local ethics committee (AWERB).

The sequence of the α-β-catenin VHH fragment kindly donated by UCB is confidential.

Abstract

There are records of animals used for experimental research more than 2000 years ago

(Hajar, 2011) and nowadays such use is controlled. UK animal research regulations date to 1876, and were most recently revised in 2013. The appropriate use of animals in biomedical research allows the elucidation of crucial information that can be used to enhance human and animal life while being humane.

Two Xenopus species (African Clawed Frogs), are major model organisms used to study

Cell and Developmental Biology. As in most animal models, mRNA levels during development are used as a proxy for protein abundance. However, recent proteomic analysis has revealed that, for at least half the genes, this assumption is wrong in vertebrate models. To visualize and assay proteins and to test their function, antibodies are needed. There have been many attempts to generate Xenopus-specific antibodies by traditional methods. These, however, had limited success and the lack of these antibodies imposes a severe research bottleneck.

This project tested different approaches to raising antibodies and compared their suitability.

Libraries of VHH (camelid single chain antibodies) raised against Xenopus protein mixtures were acquired from a collaborator and screened for Xenopus-specific VHHs; only VHHs against yolk proteins were isolated. To improve on this, antigen generation strategies were used both in vitro and in human cell cultures. Two proteins were successfully expressed in vitro, but yields were lower than expected. HEK (Human

Embryonic Kidney) cell expression was also successful. However these DNA-binding transcription factors proved challenging to solubilise and purify. Finally, commercially synthetized peptides were used to immunise llamas with a single antigen and their plasma was screened for an immune response with positive results. Experiments to test whether VHH would have advantages in Xenopus compared with conventional

antibodies were carried out with a donated VHH against human β-catenin. To understand the differences between raising VHH and conventional antibodies, a sheep was immunized with kash5 and the resulting antibody characterized. There are proteins, usually closely related members of the same family, which are difficult to tell apart using standard antibody detection. For these, a different approach to visualising the endogenous protein was undertaken. As proof of principle, the endogenous locus of

Xenopus Gata2 was tagged with HA using CRISPR cas9 to target the site, followed by homologous DNA repair to add the tag. This approach showed low incidence but the positive insertion of the HA tag into the C-terminal of Gata2.

Overall, this work has shown that Xenopus transcription factor proteins can be expressed in vitro and in vivo, however, either the quantity obtained or their solubility makes these products not suitable to raise antibodies against them. The alternative of using long peptides or MAPs (Multiple Antigenic Peptides) became a suitable option. The successful tagging of an endogenous protein was shown to also be possible approach.

This approach showed low effectivity but a few cases of HA tag insertion at the C-terminal of Gata2 were nevertheless observed.

Abbreviations

3’ Three prime 5’ Five prime 3C 3C cleavage tag AAs Amino acids AMP Ampicillin approx Approximately APS Ammonium persulphate BCA assay Bicinchoninic acid assay BLAST Basic local alignment search tool bp Base pair BSA Bovine serum albumin cDNA Coding DNA C-terminal Carboxyl-terminal CV Column volumes dH2O Distilled water DMEM Dulbecco's Modified Eagle Medium DMSO Dimethyl Sulfoxide DNA Deoxyribonucleic acid DNaseI Deoxyribonuclese I dNTP Deoxynucleoside triphosphate DTT Dithiothreitol E. coli Escherichia coli ECL Enhanced chemiluminescence EDTA Ethylene-diamine-tetraacetic acid EXRC European Xenopus Resource Centre FCS Foetal calf serum Flag3 Polypeptide protein tag g Grams gRNA Guide RNA GST Glutathione S-transferase HA Human influenza hemagglutinin HcAb Heavy- chain only camelid antibody HDR Homology-directed repair

VII

HEK cells Human Embryonic Kidney cells HEPES 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid His histidine HRP Horse- radish peroxide IPTG Isopropyl β-D- thiogalactopyranoside Kb Base x 10-3 kDa Daltons x10-3 Khis6 Lys-his-his-his-his-his-his LB Luria Broth M Molar MAP Multiple antigenic peptide MBP Maltose-binding protein MBS Modified Barth’s saline MCS Multiple coding site mg Grams x10-3

MgCl2 Magnesium chloride ml Litres x10-3 mM Molar x10-3 mmol Moles x10-3 mRFP Monomeric red fluorescent protein mRNA Messenger RNA NaAc Sodium acetate NaCl Sodium chloride Nb Nanobody NEAAs Nonessential amino acids ng Grams x10-9 nmol Moles x10-9 N-terminal Amino-terminal O/N Overnight ODC Ornithine decarboxylase OPPF Oxford Protein Production Facility ORF Open reading frame PAGE Polyacrylamide gel electrophoresis PBMC Peripheral blood mononuclear cell PBS Phosphate buffer saline

VIII

PCR Polymerase chain reaction PEG Polyethyleneglycol PEI Polyethylenimine pg Grams x 10-12 POI Point of insertion Primer Fw Forward primer Primer Rv Reverse primer RNA Ribonucleic acid RPTPmu Receptor-like protein-tyrosine phosphatase SDS Sodium dodecyl sulphate SDS-PAGE Sodium dodecyl sulphate-polyacrylamide gel electrophoresis ssDNA Single stranded DNA SS Signal sequence strepII Specifically engineered streptavidin tag TBE Tris/Borate/EDTA Buffer TBS Tris-Buffered saline TBSTw Tris-Buffered saline tween TEMED N,N,N’,N’-Tetramethylethylenediamine TF Transcription factor tRNA Transfer RNA µg Gram x10-6 µL Litres x10-6 µM Molar x10-6 µmol Moles x10-6 UTR Untranslated region VHH Nanobody V Volts v/v Volume/volume v/w Volume/weight wt Wild type xg Relative centrifuge force Xla Xenopus laevis Xtr Xenopus tropicalis

Contents

Declaration ...... II

Acknowledgements ...... III

Confidentiality and ethical consideration ...... IV

Abstract ...... V

Abbreviations ...... VII

Statement of Aims ...... XIX

List of Tables ...... XX

List of Figures ...... XXIV

1 Introduction ...... 1

1.1 General introduction ...... 1

1.2 Xenopus as a model organism ...... 4

1.3 Embryonic development of Xenopus ...... 8

1.4 Xenopus transcriptome, proteome and metabolome ...... 11

1.5 Antibodies and Research ...... 19

1.6 Heavy chain-only antibodies ...... 21

1.7 Importance and contribution of the proposed research ...... 26

1.8 Target proteins for antigen production and antibody generation ...... 27

1.8.1 Gata2 - GATA binding protein 2 ...... 28

2 Materials and Methods ...... 34

2.1 Nucleic acid techniques ...... 35

2.1.1 Restriction digests ...... 35

2.1.2 Agarose gel electrophoresis ...... 35

2.1.3 Ligation by T4 DNA ligase ...... 36

2.1.4 pGEM®-T Easy Vector Systems (Promega A1360) ...... 37

2.1.5 Transformation of E. coli DH5α ...... 38

2.1.6 Polymerase chain reaction (PCR) ...... 39

2.1.7 RNA extraction from embryos ...... 41

2.1.8 cDNA synthesis from Xenopus total RNA ...... 43

2.1.9 In vitro protein production ...... 45

2.1.10 HEK cell protein expression ...... 46

2.1.11 Cas9 RNA synthesis ...... 50

2.2 gRNA production ...... 51

2.3 Microbiological techniques ...... 53

2.1.1 Bacterial strains and phage used ...... 53

2.3.2 Expression of single-domain antibodies ...... 58

2.4 Protein techniques ...... 59

2.4.1 Total protein extract from Xenopus embryos ...... 59

2.4.2 Western blot ...... 61

2.4.3 In vitro protein expression ...... 65

2.4.4 In vivo protein expression ...... 65

2.4.5 Protein purification ...... 73

2.5 Antibody purification techniques ...... 75

2.5.1 Caprylic acid extraction ...... 75

2.5.2 Dialysis ...... 75

2.5.3 Affinity column purification ...... 76

2.5.4 G-25 desalting column: ...... 77

2.6 Embryological techniques ...... 78

2.6.1 Embryo micro-injections ...... 78

2.7 VHH libraries ...... 79

2.7.1 General considerations ...... 79

2.7.2 Llama PBMC prep on 400 ml blood samples: ...... 80

2.7.3 Library Construction ...... 81

2.7.4 Primer design ...... 82

2.8 Bacterial library construction ...... 85

2.8.1 Phage library construction ...... 86

2.8.2 VHH selection of affinity binders by phage display ...... 87

2.9 Suppliers ...... 90

3 Creation and characterization of complex libraries ...... 91

3.1 Introduction ...... 91

3.2 Results ...... 94

3.2.1 Bacterial library construction ...... 94

3.2.2 Isolation and characterisation of individual VHHs ...... 96

3.2.3 Screening libraries for the presence of VHHs ...... 97

3.3 Llama immunisation with single antigens ...... 100

3.3.1 Lymphocyte isolation ...... 101

XII

3.3.2 Testing for an immune response in llama serum ...... 101

3.4 Discussion ...... 102

4 Expression of Xenopus transcription factors ...... 104

4.1 Introduction ...... 104

4.2 Results ...... 107

4.2.1 Neurog2 ...... 107

4.2.2 Cdknx ...... 108

4.2.3 In vitro protein expression of neurog2 and cdknx ORFs ...... 110

4.2.4 HEK cell expression of target proteins ...... 115

4.2.5 A new approach, the use of vectors with secretion signals to allow the

proteins to exit the nucleus and cell ...... 122

4.2.6 Commercially synthesised and expressed proteins ...... 138

4.3 Discussion ...... 144

5 Host immunisation and antibody screen ...... 146

5.1 Introduction ...... 146

5.2 Results ...... 147

5.2.1 VHH Libraries ...... 147

5.2.2 Kash5 antibody raising and characterization ...... 153

5.3 Discussion ...... 165

6 The use of a β-catenin VHH as proof of principle ...... 167

6.1 Introduction ...... 167

6.1.1 β-catenin’s importance in development ...... 167

6.1.2 β-catenin in Xenopus ...... 169 XIII

6.1.3 Loss of function experiments ...... 170

6.1.4 A VHH against β-catenin ...... 172

6.2 Results ...... 177

6.2.1 Sub-cloning β-catenin and mutated β-catenin VHH into a vector for

expression in Xenopus ...... 177

6.2.2 Can β-catenin VHH block gene activity in the frog? ...... 179

Loss of function ...... 179

6.2.3 Resourcing a true negative control - GFPVHH ...... 184

6.2.4 Expression of β-catenin VHH as periplasmic protein and its use as

Antibody 186

6.2.5 Can single chain antibodies be used for in vivo labelling of β-catenin in

Xenopus laevis? ...... 192

6.3 Discussion ...... 200

7 HA-tagging of the endogenous Gata2 locus in Xenopus laevis ...... 201

7.1 Introduction ...... 201

7.1.1 General Introduction ...... 201

7.1.2 Epitope tagging history and pitfalls ...... 206

7.1.3 GATA Binding Protein 2 ...... 207

7.1.4 Description of the target region ...... 209

7.1.5 Experimental design ...... 211

7.1.6 T7 endonuclease assay ...... 215

7.1.7 Cas9 RNA and protein ...... 217

7.2 Results ...... 219

XIV

7.2.1 gata2 3’ region sgRNA design ...... 219

7.2.2 sgRNA efficiency determination by T7 endonuclease assay ...... 222

7.2.3 Can Cas9 protein cutting efficiency compare with Cas9 mRNA injection?

224

7.2.4 New sgRNA design and production ...... 226

7.2.5 Can sgRNA #4 successfully guide cas9 to the cut site? ...... 228

7.2.6 Identifying the mutations made by sgRNA #4 and cas9 ...... 230

7.2.7 Design of the insertion sequence for HA-tagging of gata2 ...... 233

7.2.8 Can an HA tag be inserted into the endogenous gata2 C-terminus? .... 235

7.2.9 Detecting and characterizing HA sequence ...... 237

7.3 Discussion ...... 243

8 General discussion and further work ...... 245

9 References ...... 249

10 Websites used...... 288

10.1.1 Phyre2 - Protein folding prediction tool...... 288

10.1.2 UniPort - Protein database...... 288

10.1.3 Xenbase - Xenopus model database...... 288

10.1.4 NCBI - Multiple model biotechnological databases and research journal

database...... 288

10.1.5 psipred - Protein folding prediction tool...... 288

10.1.6 Google - Search engine...... 288

10.1.7 CRISPRscan - gRNA designing tool...... 288

10.1.8 BLAST - Basic local alignment search tool...... 288

11 Appendix I ...... 289

11.1 pOPIN Vectors Maps ...... 289

11.1.1 pOPINF-his tag at the N-terminal ...... 289

11.1.2 pOPINM-MBP - MBP tag and his-tag ...... 290

11.1.3 pOPINJ-JST – JST tag and his-tag ...... 291

11.1.4 pOPINE – his-tag at the C-terminal...... 292

11.1.5 pOPINTTGneo - his Tag at the C terminal ...... 293

11.1.6 pOPINEneo-3C-2STREP - his tag and strep-tag at the C-terminal ...... 294

11.1.7 pOPINEneo-3C-3FLAG - his tag and Flag-tag at the C-terminal ...... 295

11.1.8 pOPINFS - his tag and Flag-tag at the C-terminal ...... 296

11.2 List of Primers ...... 297

11.2.1 Vector pOPINE: ...... 297

11.2.2 Vector pOPINE-Strep II (C-his tag): ...... 298

11.2.3 Vectors pOPINF - pOPINM –pOPINJ: ...... 298

11.2.4 Vector pOPINGM (ss-MBP-C-his tag): ...... 299

11.2.5 Vector pOPINTTGneo: ...... 300

11.2.6 Vector pOPINE-Flag3 (C-his tag) ...... 301

11.3 In-Fusion reaction overview ...... 302

11.4 Protein expression via transfection into HEK293T cells ...... 303

11.4.1 Cdk1 ...... 303

11.4.2 Delta1 ...... 306

11.4.3 FliI ...... 306

XVI

11.4.4 Gata2 ...... 307

11.4.5 HairyI ...... 310

11.4.6 Myt1 ...... 313

11.4.7 NeuroD ...... 314

11.4.8 Neurog2 ...... 315

11.4.9 Rarα2 ...... 315

11.4.10 Ncor2 ...... 316

11.4.11 TalI ...... 316

11.4.12 XicI ...... 317

12 Appendix II ...... 318

12.1 Insert verification by sequence BLAST (Xenbase.org) ...... 319

13 Appendix III ...... 325

13.1 Protein targets for commercial synthesis ...... 325

13.1.1 ascl1 ...... 325

13.1.2 ccna1 ...... 325

13.1.3 ccne1 ...... 325

13.1.4 cdk1 ...... 325

13.1.5 cdk2 ...... 326

13.1.6 cdknx ...... 326

13.1.7 dazl ...... 326

13.1.8 etv2 ...... 326

13.1.9 fli1...... 326

XVII

13.1.10 fus1 ...... 327

13.1.11 gata2 ...... 327

13.1.12 hes1 ...... 327

13.1.13 lis1 ...... 327

13.1.14 myt1 ...... 328

13.1.15 ncor2 ...... 328

13.1.16 ndel1 ...... 328

13.1.17 neurod1 ...... 329

13.1.18 neurog2 ...... 329

13.1.19 ngn3 ...... 329

13.1.20 ppp2r5a ...... 329

13.1.21 rara2 ...... 329

13.1.22 rarb ...... 329

13.1.23 ror2 ...... 330

13.1.24 tal1 ...... 330

13.1.25 tra2b ...... 330

13.1.26 wnt4b ...... 330

14 Literature Search techniques and sources employed ...... 332

XVIII

Statement of Aims

The primary objective of this work is to find ways to detect and visualise proteins in the

Xenopus embryo. More precisely, to develop an efficient way of raising Xenopus-specific antibodies and to tag endogenous genes.

The project can be subdivided into these different parts:

a. Development of VHHs (single chain camelid antibodies)

1. Analysis of previously generated VHH libraries (library screening)

2. Generation of Immunogens (protein expression)

3. Generation of VHH library (llama immunisation)

4. Use of β-catenin VHH as proof of principle

5. Test whether VHHs can act as neutralising antibodies for protein

knockouts in embryos

b. Tagging the endogenous gene (gata2)

1. Use of CRISPR-Cas9 technology to insert an HA-tag on the 3’ end

of gata2

2. Identification and screen of GA tadpoles

XIX

List of Tables

Chapter 1

No tables

Chapter 2

Table 2. 1. T4 DNA ligase reaction components...... 36

Table 2. 2. PCR reaction standard conditions...... 39

Table 2. 3. Standard PCR reaction (25 µl)...... 39

Table 2. 4. kash5 gene specific primers for Xenopus laevis...... 40

Table 2. 5. kash5 gene specific primers for Xenopus tropicalis...... 40

Table 2. 6. NETS buffer recipe for RNA extraction from Xenopus embryos ...... 42

Table 2. 7. cDNA synthesis from total RNA by reverse transcription...... 43

Table 2. 8. Isolation and amplification of a specific region from total cDNA ...... 43

Table 2. 9. Primer sequences used to amplify neurog2 and xicI ...... 45

Table 2. 10. . Digestion of pOPIN vectors...... 47

Table 2. 11. In-Fusion reaction components...... 50

Table 2. 12. Transcription of Cas9 by SP6 polymerase...... 50

Table 2. 13. Components and steps to synthesise double stranded DNA ...... 52

Table 2. 14. TSB (transformation and storage buffer) and 2xYT medium ...... 53

Table 2. 15. M9 minimum agar and YENB medium...... 55

Table 2. 16. Top agar, agar used to propagate phage recipe...... 56

Table 2. 17. FB1 and FB2 solutions composition to propagate E. coli DH5α cells...... 57

Table 2. 18. NETS Buffer vs Embryo Extraction Buffer (EEB) reagents...... 60

Table 2. 19. Protein loading dye composition...... 60

Table 2. 20. Recipe for 10 ml of polyacrylamide gel mix ...... 62

Table 2. 21. Western blot tris-glycine membrane transfer buffer...... 62

Table 2. 22. Blocking buffer composition and TBSTw recipie ...... 63

Table 2. 23 ECLI and ECLII buffer composition ...... 64

Table 2. 24. Complete DMEM ...... 65

Table 2. 25. DMEM + 2 mM L-Glutamine ...... 65

Table 2. 26. DMEM + 1 mM L-Glutamine + 2 % FCS ...... 65

Table 2. 27. Freezing Media ...... 66

Table 2. 28. M-PER® reagent recommended quantities per sample volume...... 71

Table 2. 29. Sonication/lysis buffer and solubilisation buffer recipes ...... 72

Table 2. 30. Binding buffer (solution A) and elution buffer (solution B) ...... 74

Table 2. 31. Barth-HEPES Saline (10 X) MBS (Modified Barth’s Saline)...... 78

Table 2. 32. Most commonly used primers to isolate the VHH ...... 83

Chapter 3

No tables

Chapter 4

Table 4. 1. OPPF vectors used to express proteins in HEK cells ...... 116

Table 4. 2. PCR amplification of target genes ...... 120

Table 4. 3. Sub-cloning successes for Xenopus proteins...... 121

Table 4. 4. Companies contacted to express the given targets and their responses. 138

Table 4. 5. Thermo Scientific proposed peptide sequences...... 143

XXI

Chapter 5

Table 5. 1. Immunization schedule for Sunny ...... 151

Table 5. 2. Immunization schedule for Valentine ...... 152

Chapter 6 No tables

Chapter 7

Table 7. 1. Commonly used protein tags (Li, 2010; Maue, 2007)...... 202

Appendix I Table I. 1. Primers designed to amplify the desired sequence and anneal to the pOPINE

vector...... 297

Table I. 2. Primers designed to amplify the desired sequence and anneal to the pOPINE-

Strep II vector...... 298

Table I. 3. Primers designed to amplify the desired sequence and anneal to the pOPINF,

pOPINM and pOPINJ vectors...... 298

Table I. 4. Primers designed to amplify the desired sequence and anneal to the

pOPINGM vector...... 299

Table I. 5. Primers designed to amplify the desired sequence and anneal to the

pOPINTTGneo vector...... 300

Table I. 6. Primers designed to amplify the desired sequence and anneal to the pOPINE-

Flag3 vector...... 301

Table I. 7. CDK1 protein expression via transfection into HEK293T cells...... 303

Table I. 8. Delta1 protein expression via transfection into HEK293T cells...... 306

Table I. 9. FliI protein expression via transfection into HEK293T cells...... 306

Table I. 10. Gata2 protein expression via transfection into HEK293T cells...... 307

XXII

Table I. 11. HairyI protein expression via transfection into HEK293T cells...... 310

Table I. 12. Myt1 protein expression via transfection into HEK293T cells...... 313

Table I. 13. NeuroD protein expression via transfection into HEK293T cells...... 314

Table I. 14. Neurog2 protein expression via transfection into HEK293T cells...... 315

Table I. 15. Rarα2 protein expression via transfection into HEK293T cells...... 315

Table I. 16. Ncor2 protein expression via transfection into HEK293T cells...... 316

Table I. 17. TalI protein expression via transfection into HEK293T cells...... 316

Table I. 18. XicI protein expression via transfection into HEK293T cells...... 317

Appendix II No tables

Appendix III No tables

Appendix IV No tables

XXIII

List of Figures

Chapter 1

Figure 1. 1. Amphibian publications per 5 years periods ...... 7

Figure 1. 2. Xenopus development...... 10

Figure 1. 3. RNA levels assayed in the early embryo...... 13

Figure 1. 4. Xenopus laevis and Xenopus tropicalis gene expression comparison ...... 14

Figure 1. 5. mRNA and protein levels in the Xenopus laevis egg...... 17

Figure 1. 6. Protein amounts in pmol per developmental stage ...... 18

Figure 1. 7. Conventional antibodies, Fab fragments and VHH structures...... 24

Figure 1. 8. VHH used to visualize mouse brain structures in vivo and ex vivo...... 25

Figure 1. 9. Gata2 expression throughout development...... 30

Figure 1. 10. Xenopus laevis Gata2 protein sequence alignment ...... 31

Figure 1. 11. Gata binding protein family sequence alignment...... 33

Chapter 2

Figure 2. 1. cDNA synthesised from Xenopus total RNA by reverse transcription...... 44

Figure 2. 2. Large-scale transient transfection of HEK293T cells...... 70

Figure 2. 3. Nested primers to isolate the heavy chain only variable region ...... 84

Chapter 3

Figure 3. 1. Purity and concentration confirmation of library elements before ligation. 95

Figure 3. 2. Screening libraries for the presence of VHHs...... 99

XXIV

Chapter 4

Figure 4. 1. Neurog2 Xenopus laevis L and Xenopus laevis S RNA-seq profiles ...... 109

Figure 4. 2. RNA-seq profile for Xenopus laevis cdknx by developmental stages. .... 109

Figure 4. 3. Protein expression via the human in vitro transcription ...... 111

Figure 4. 4. pT7CFE1-CGST-HA-his vector map...... 112

Figure 4. 5. Cdknx and Neurog2 proteins expressed in vitro ...... 114

Figure 4. 6. PCR amplification of inserts to clone into pOPIN vectors by in-fusion. ... 119

Figure 4. 7. Cloning enhancer treated PCR amplification of gata2, hairyI and cdkI ... 123

Figure 4. 8. Small-scale HEK cell transfected proteins media and cell lysate...... 126

Figure 4. 9. Small-scale HEK cell transfected proteins media and cell lysate...... 126

Figure 4. 10. Intracellular protein extraction via mPER treatment ...... 128

Figure 4. 11. Purification via his-trap column of HairyI expressed protein...... 132

Figure 4. 12. HairyI integrity and purification after storage...... 133

Figure 4. 13. Further purification of HairyI analysed by western blot...... 134

Figure 4. 14. Gata2 transfection...... 135

Figure 4. 15. Neurog2 protein transfected with pOPINTTGneo...... 136

Figure 4. 16. Western blot to visualize Gata2 expressed protein ...... 137

Figure 4. 17. Cdk1 octomeric MAP sequence suggested by Alta Bioscience...... 140

Figure 4. 18. Myt1 octomeric MAP sequence suggested by Alta Bioscience...... 141

Figure 4. 19. Kash5 octomeric MAP sequence suggested by Alta Bioscience...... 142

XXV

Chapter 5

Figure 5. 1. Schematic representation of the processes involved in the generation of

VHHs...... 149

Figure 5. 2. Immunization response assessment by western blot analysis...... 150

Figure 5. 3. Kash5 Xenopus tropicalis mRNA sequence with the tranlated Kash5. ... 155

Figure 5. 4. Xenopus tropicalis Kash5 protein sequence...... 156

Figure 5. 5. Western blot to test the immune response of the sheep ...... 159

Figure 5. 6. Western blot to test purified kash5 antibody...... 160

Figure 5. 7. Western blot to optimize purified kash5 antibody staining...... 161

Figure 5. 8. Kash5 overexpression to confirm the specificity of Kash5 antibody...... 164

Chapter 6

Figure 6. 1. Phenotype caused by the depletion of β-catenin ...... 171

Figure 6. 2. β-catenin VHH and mutated control...... 174

Figure 6. 3. Binding capacities test of LL3-scFc and LL3.CDRAAA-scFc ...... 175

Figure 6. 4. Native β-catenin VHH and control mutated form containing vector map. 176

Figure 6. 5. Sequence alignments...... 176

Figure 6. 6. Cloning of native β-catenin VHH and control mutated form into pCS2+. 178

Figure 6. 7. β-catenin loss of function experiments in Xenopus laevis embryos...... 181

Figure 6. 8. β-catenin-VHH RNA micro-injection into one-cell ...... 182

Figure 6. 9. β-catenin-VHH RNA micro-injection into four-cell ...... 183

Figure 6. 10. A true negative control, pOPINE-GFP-VHH ...... 185

Figure 6. 11. pecan126 vector map...... 188

Figure 6. 12. Sequencing and cloning of β-catenin-VHH ...... 189 XXVI

Figure 6. 13. Periplasmic expression of β-catenin-VHH...... 190

Figure 6. 14. Detection of Xenopus laevis endogenous β-catenin ...... 191

Figure 6. 15. DsRed and its derivate...... 194

Figure 6. 16. KatushkaRFP-VHH fusion in pCS2+...... 196

Figure 6. 17. An alternative approach to clone KatushkaRFP-VHH...... 197

Figure 6. 18. Strategy to clone mCherry fused to β-catenin-VHH ...... 199

Chapter 7

Figure 7. 1. Relative popularity of individual affinity tags...... 203

Figure 7. 2. Xenopus laevis gata2 gene ...... 210

Figure 7. 3. CRISPR/cas9 experimental design...... 213

Figure 7. 4. CRISPR/cas9 injection into the Xenopus laevis fertilised embryo...... 214

Figure 7. 5. T7 endonuclease assay...... 216

Figure 7. 6. CRISPR/cas9 injection methods...... 218

Figure 7. 7. sgRNA design to the 3' UTR of Xenopus laevis gata2...... 220

Figure 7. 8. sgRNA sequences are targeting the gata2 C-terminus region...... 220

Figure 7. 9. sgRNA synthesis...... 221

Figure 7. 10. None of the first 3 sgRNAs produced indels in injected embryos...... 223

Figure 7. 11. Cas9 mRNA preparation and test Vs cas9 protein (NEB)...... 225

Figure 7. 12. CRISPRscan sgRNA suggestions for gata2 3' end target sequence. ... 227

Figure 7. 13 Sequence comparison of sgRNA #3 and sgRNA #4...... 227

Figure 7. 14. T7 endonuclease assay ...... 229

Figure 7. 15. Target site amplicons ...... 231

XXVII

Figure 7. 16. Gata2 target site amplicons cloned into pGemTeasy...... 232

Figure 7. 17. Insert DNA design...... 234

Figure 7. 18. Xenopus laevis gata2 RNA-Seq expression data...... 236

Figure 7. 19. Xenopus laevis gata2 3' end ...... 238

Figure 7. 20. Cas9 protein (NEB) and sgRNA #4 injected embryos...... 239

Figure 7. 21. Sequenced co-injected Xenopus laevis embryos containing ...... 240

Figure 7. 22. Sequenced co-injected embryos, 5’ end...... 241

Figure 7. 23. Representative 5’ detailed amplification sequence ...... 242

Appendix I

Figure I. 1. pOPINF vector map...... 289

Figure I. 2. pOPINM vector map...... 290

Figure I. 3. pOPINJ vector map...... 291

Figure I. 4. pOPINF vector map...... 292

Figure I. 5. pOPINTTGneo vector map...... 293

Figure I. 6. pOPINEneo-3C-2STREP vector map...... 294

Figure I. 7. pOPINEneo-3C-3FLAG vector map...... 295

Figure I. 8. pOPINFS vector map...... 296

Figure I. 9. Outline of In-Fusion PCR Cloning System...... 302

Appendix II

Figure II. 1. pOPINF-RARα2 plasmid sequence verification...... 319

Figure II. 2.pOPINF-NeuroD plasmid sequence verification...... 320

Figure II. 3. pOPINF-FliI plasmid sequence verification...... 321

XXVIII

Figure II. 4. pOPINE-RARα2 plasmid sequence verification...... 322

Figure II. 5. pOPINE-NeuroD plasmid sequence verification...... 323

Figure II. 6. pOPINE-FliI plasmid sequence verification...... 324

Appendix III

No figures

Appendix IV

No figures

XXIX

1 Introduction

1.1 General introduction

The South African clawed frog, Xenopus laevis (Xla), has been used extensively as a model system to study development, cell biology and biochemistry; and more recently human genetic diseases (Bowes et al., 2008; Harland & Grainger, 2011). Due to the ease of egg production and manipulation, together with other advantages, Xenopus has served as a contributor to the discovery of genes and gene regulatory systems with crucial functions in development. A summary of how Xenopus evolved as a model organism is given later in this chapter.

Researchers traditionally prefer to assess the formation of RNAs as an assay for gene activity due to ease, and as a consequence, the production of Xenopus-specific antibodies has widely been put aside by the Xenopus community (Xenopus Community

White Paper 2014). This lack of antibody production was also enhanced by the fact that many of the proteins of interest to the Xenopus community are transcription factors, which are usually found in complexes, making protein production and purification for antibody raising challenging. From 1235 known transcription factors, only 295 out of 796 antibodies generated passed the ENCODE testing regime (ENCODE Project

Consortium, 2012). Moreover, companies that make human or mouse antibodies, do not see a profitable market for raising antibodies against Xenopus-specific proteins due to the relatively small market compared to other systems. The currently available Xenopus- specific antibodies can be obtained from the EXRC

(https://xenopusresource.org/about_antibodies), and a full list of antibodies that were explicitly developed against or cross-react with Xenopus proteins can be obtained from

Xenbase (http://www.xenbase.org/reagents/antibody.do).

In 2017, the National Health Institute (NIH, USA) funded the project “Production and characterization of monoclonal antibodies to Xenopus Proteins”, Project #

1R24OD021485-01A1 (University of Massachusetts Amherst, USA), to address the issue highlighted for a decade as the main reason why Xenopus is not as advanced as other model organisms in some aspects of research (according to the Xenopus community White Paper, USA, 2012).

This project focuses on testing different ways of visualising proteins in Xenopus. The use of conventional antibodies, VHHs and epitope tagging of the endogenous loci are explored with the aim of understanding the strengths and weaknesses of these approaches.

The aim is to explore two pathways to obtain antibodies against Xenopus proteins. One method follows the conventional approach of synthesising the antigen as a peptide and injecting it into a (in this case) sheep; while the other explores the expression of antigens in HEK cells and immunising a Llama glama, which produces single-chain antibodies as well as conventional antibodies. Ideally, for a direct comparison this would be done with the same protein. A limitation, however, of this study was the need to provide different antibodies to members of the Xenopus research community as part of the EXRC’s work.

This precluded the use of the same protein in the different approaches.

Conventional antibodies are usually produced in chickens, rabbits, goats, sheep or rats.

In this project, a sheep was immunised to assess the feasibility of producing antibodies in this manner. A member of the community designed the antigen used for immunisation as this project was funded by the BBSRC to generate reagents for the community, therefore complying with the grant’s service element.

In 1989, a novel type of antibody was identified, first in the sera of dromedaries and later on in various members of the Camelidae family (Hamers-Casterman et al., 1993;

Rocchetti, Hawes, & Kriechbaumer, 2014). These novel antibodies, known as nanobodies or VHHs, are very small in size (15-18 kDa) and contain the full coding sequence in a single exon (Muyldermans, 2001). They are easily produced in microorganisms and are highly soluble and robust. Their single-domain nature and strict 2

monomeric behaviour support easy cloning and fast selection from libraries. Moreover, because VHHs prefer to associate with concave-shaped epitopes (e.g., catalytic sites of enzymes), they can recognise places that are inaccessible or cryptic for conventional antibodies (De Meyer, Muyldermans, & Depicker, 2014).

The advantages of using VHHs over conventional antibodies, together with the discovery of the protein versus mRNA expression level discrepancies in Xenopus embryos

(Peshkin et al., 2015; Wühr et al., 2014), has made the generation of Xenopus-specific

VHHs and their assessment one of the priorities for the scientific community using this powerful model organism.

There are, however, proteins that even the most specific antibodies made to date could not identify. These are for example proteins of the same family which are very similar in sequence but have distinct developmental functions. In the past, an antibody raised against one would react against the other providing conflicted results, therefore, discriminating between these closely related proteins is essential.

Also for newly discovered proteins and proteins of low immunogenicity (Brizzard, 2008), where the protein cannot be purified or the purified protein fails to raise an immune response, antibodies are difficult to raise. A way of tagging the endogenous gene would allow the specific visualisation and study of such proteins.

1.2 Xenopus as a model organism

To understand why visualising frog proteins is essential, it is first necessary to see how this model organism works and what it can do.

Biomedical studies rely significantly on experimental research to elucidate the mysteries of human biology. For ethical reasons, experimental studies are very limited and extremely controlled in humans. One of the alternatives is to use animals as platforms to study the biological and biophysical aspects of life. Here we focus on the use of animals in the fields of embryology and developmental biology, and how Xenopus became the most used amphibian model.

Xenopus was first described in the early 1800s as Bufo laevis, by a French naturalist

(Daudin, 1802), and received its current name in 1827 (Gurdon & Hopwood, 2000). The

Zoological Society of London reported the first spawning of Xenopus laevis outside Africa in 1894. However, it was not until 1906 that the developmental stages of the Xenopus embryo were published, together with protocols of husbandry for raising Xenopus as a laboratory animal (Bles, 1906). By the 1920s, Xenopus was preferred over local amphibian species at schools and universities because of the number of offspring that were produced from a single animal. However, it was not until 1930 that induced ovulation of Xenopus was mentioned in a scientific paper (Gurdon & Hopwood, 2000;

Hogben & Winton, 1930).

A historically relevant fact was that in the early 1900s, pregnancy testing was invented, with the first Pregnancy Diagnostics Station opening in 1929 producing 6000 tests per year (Crew, 1936). This test was done in mice, injecting them five times a day for three days with women’s morning urine, then killing them and analysing the ovaries to see if they were enlarged; this would mean positive for pregnancy. In 1939, it was reported by the Edinburgh laboratories that the same test could be achieved in Xenopus with only one injection and results obtained 8-12 hours later (Cabrera, 1969). Tens of thousands

of tests were performed in the next two decades by this laboratory and others, and

Xenopus were acquired all over Europe and America for such tests (Gurdon & Hopwood,

2000). By the 1960s, alternative methods of pregnancy testing, which didn’t require live animals, were emerging thus replacing the Xenopus pregnancy test.

By this time, researchers worldwide had access to established Xenopus colonies, and two advantages over other species were becoming prominent: the ease of induced ovulation which simplified microsurgery timing and biochemical studies for which abundant material was needed. The development of embryonic fate maps (Dale & Slack,

1987); the elucidation of the Xenopus genome (Session et al., 2016); the development of an online database for all information related to Xenopus (Karpinka et al., 2015); the increasing use of Xenopus as a disease model (Tandon, Conlon, Furlow, & Horb, 2017); the award of two Nobel prizes for cell cycle, control (Hunt, 2001) and nuclear reprogramming (Gurdon, 2012); and the high number of published papers using

Xenopus as a model organism are all signs of the importance and impact that research carried out in this model has had, and will have in society.

Mesoderm induction, dorsalization and neural induction studies (Slack, Dale, & Smith,

1984; Tarin, 1971, 1972), cell cycle regulation studies (Aimar, Vilain, & Delarue, 1983;

Allan & Vale, 1991; Gerhart, Wu, & Kirschner, 1984; Moor, 1988; Yamano, Gannon,

Mahbubani, & Hunt, 2004), the development of the whole-mount In situ hybridization technique (Harland, 1991) and the loss of function studies that elucidated the Wnt signalling pathway (Heasman, Kofron, & Wylie, 2000; Hoppler, Brown, & Moon, 1996) are all very good examples of key findings and methods developed that allowed the growth and expansion of Xenopus as a model organism. Furthermore, the development of transgenic lines (Kroll & Amaya, 1996; Offield, Hirsch, & Grainger, 2000) re-inforced established methods such as loss of gene function by morpholino injections and established the study of late development and organogenesis (Beck & Slack, 2001; Nutt,

Bronchain, Hartley, & Amaya, 2001).

The turn of the century marked the accelerated expansion of areas where Xenopus was used as a model organism. This growth was stimulated by the establishment of the

European Xenopus Resource Centre based in the UK, and later the National Xenopus

Resource, based in USA (Pearl, Grainger, Guille, & Horb, 2012), both providing resources and intellectual knowledge to researchers using this model. The release of an online Xenopus dedicated database namely Xenbase (Bowes et al., 2009) and the creation of the Xenopus laevis isogenic line (J-line) which allowed the mapping and publication of the Xenopus laevis genome (Harland & Gilchrist, 2017; Session et al.,

2016) were of invaluable importance.

The emergence of Xenopus tropicalis of as a parallel model with a diploid genome and a shorter generation time (Hirsch, Zimmerman, & Grainger, 2002; Klein et al., 2002), also re-enforced the use of Xenopus laevis as a model organism, since now researchers could do genetic studies on a close relative of Xenopus laevis. Moreover, the techniques and tools used in Xenopus laevis were mostly transferable to Xenopus tropicalis (Khokha et al., 2002). In the early 2000s numerous high-quality papers demonstrating the usability of Xenopus as a genetic model organism were published (Abu-Daya, Khokha, &

Zimmerman, 2012; Goda et al., 2006; Khokha et al., 2009). Another considerable advantage of the early Xenopus embryo is that single cells can be isolated and analysed on a proteomic, transcriptomic and metabolomics level (Briggs et al., 2018; Onjiko,

Moody, & Nemes, 2015; Sun et al., 2016). These resources allowed Xenopus to re- emerge in this new genetic era and potentially open new paths to the understanding of development on a single cell level.

The decline in publications using Xenopus since the 1980s can probably be attributed in part to the lack of Xenopus specific reagents (Figure 1. 1).

Figure 1. 1. Amphibian publications per 5 years periods, comparing Xenopus with other amphibians used for developmental studies, based on PubMed search (April 2018). Research using Xenopus as an experimental model had its peak time in the late 1990s, according to publications per year. The drop in publications since could be attributed to a variety of factors, such as the lag in generating Xenopus specific reagents and the difficulties that this model encounter being allo-tetraploid when the genetic era began. However, the release of the Xenopus genome in 2010 (Hellsten et al., 2010), favoured this model. The continuous decline in publications in the last 15 years is considerable. However, Xenopus is still the most used amphibian model for research (according to PubMed search).

1.3 Embryonic development of Xenopus

The life cycle of Xenopus laevis has been intensely studied and described, from the life cycle (Grobman, 1958) to the fate mapping of an individual cell (Dale & Slack, 1987;

Gurdon, 1995) and the availability of gene expression patterns and quantitative mRNA measurements on Xenbase (Bowes et al., 2008). In summary, oocytes are produced and matured inside the female, laid and then fertilised with sperm. In the laboratory, this can be done in vitro or by natural mating. The fertilised egg starts dividing about 2 hour’s post-fertilisation, reaching gastrulation after 14 hours if the temperature is kept constant at 18ºC. By 30 hours, organs start to develop and 70 days later metamorphosis from a tadpole to froglet starts. It is recorded that Xenopus laevis can reach adulthood six months after metamorphosis (A. Jafkins and G. Nicholson, personal communications)

(Figure 1.2 A).

When the sperm fertilises the egg by entering the animal hemisphere (Figure 1.2 B), the female genetic material mixes with that of the male usually in equal quantities. However, the remainder of the molecular content of the embryo is inherited from the mother. The oocyte matures inside the female and when it is laid it contains all the essential proteins and material needed for the fertilised egg to develop until around stage 8 when most zygotic transcription in the embryo starts (Newport & Kirschner, 1982). Development of the fertilised egg occurs very rapidly in the first few hours with the first division occurring

90-120 minutes post fertilisation; then there is a synchronised cell division every 25-30 minutes for 7 hours until the mid-blastula stage embryo of around 5000 cells, when the embryo starts to divide asynchronously. This point is also marked by the onset of cell motility and zygotic transcription. After these 12 divisions, when the embryonic genome is activated and at the blastula stage, three distinct germ layers can be identified, ectoderm, mesoderm and endoderm. By 10 hours, the embryo is composed of 10 000 cells and undergoes gastrulation. During gastrulation, the germ layers rearrange to form the outer layer or ectoderm, the middle layer or mesoderm and the inner layer, or

endoderm. At the neurula stage (14 hours), the neural plate becomes visible in the dorsal ectoderm, this will form the central nervous system (CNS). By the tailbud stage (24–42 hours), a larva with a neural tube located between the epidermis and the notochord has formed, and organogenesis starts with the differentiation of cells into specific types. A free-swimming tadpole is formed in 4 days (Dale & Slack, 1987; De Robertis, Larraín,

Oelgeschläger, & Wessely, 2000; C. M. Jones & Smith, 2008; Méreau, Sommer,

Lerivray, Lesimple, & Hardy, 2007; Saka, Tada, & Smith, 2000).

Developmental biologists historically used assays for RNA, not only to study gene activity in the early embryo, but also to infer the presence of a protein when studying its function.

In 2014 and later in 2016, questions arose about the value of making such an assumption when it was published that the mRNA levels in the early Xenopus embryo did not correlate to the actual protein levels present at the same stage (Peshkin et al., 2015;

Wühr et al., 2014). For example, some maternal proteins stayed in the embryo longer than expected and some RNAs were not translated until much later in development.

These revolutionary papers are discussed below and highlight the importance of being able to identify, visualise, follow and isolate proteins at these early stages if one is to understand development and cell behaviour accurately.

rganogenesis

them, or they can they them, or

male lays male

stimulatedfemale (A). Development

establishedgastrulation, neurulation and o

during a natural mating as the fe the asmating natural a during

Méreau, Le Sommer, Lerivray, Lesimple, Hardy, & 2007)

(

embryo

Xenopus

tadpolegoing takes through four days, well

celltypes inthe early

swimming

primary

development.

localisation

bymacerating the sperm spreading and it overeggs that have been squeezed from the pre

Xenopus

. .

invitro

Figure 1. Figure1.

The oocytes are created and matured inside the female; the male fertilises the eggs laideggs fertilisesthe male thefemale; the maturedinside andcreated are oocytes The fertilised be egg offertilised through the free a to processes, with the

1.4 Xenopus transcriptome, proteome and metabolome

The first published Serial Analysis of Gene Expression (SAGE) dates to 1995 when

Velculescu and colleagues described how by isolating a transcribed tag, cloning it into a vector and counting the tags by sequencing, it was possible to obtain a representation of nearly every transcript in that population of cells, a transcriptome (Velculescu, Zhang,

Vogelstein, & Kinzler, 1995). It was not until 1997 that the same author published the first whole organism transcriptome, the Yeast transcriptome (Velculescu et al., 1997).

One of the first published transcriptome studies in Xenopus was related to the amphibian skin secretions and a non-invasive methodology of sampling individuals for such studies

(Chen et al., 2003); and in 2006, endoderm formation studies via transcriptomic analysis were performed which demonstrated the complexity of the process. This postulated that new studies were needed to understand the complete endoderm gene regulatory network (Sinner, 2006). Xenopus transcriptomic studies, comparing Xenopus laevis and

Xenopus tropicalis transcriptomes, also helped elucidate the link between gene expression and morphological evolution, accentuating that the critical differences between species are in the earliest embryonic stages (Yanai, Peshkin, Jorgensen, &

Kirschner, 2011).

However, it was not until much later that more relevant transcriptomic data for developmental biology emerged. Although, microarrays began to give transcriptomic information in Xenopus from the start of the century (Altmann et al., 2001) it was not until

RNA-Seq allowed a much deeper analysis of the transcriptome that major advances were made. In 2016, a study of transcriptome kinetics in Xenopus development (Owens et al., 2016) elucidated the rapid changes in gene expression during embryonic development which dictate cell fate decisions by ultra-high frequency sampling of

Xenopus embryos. Transcriptome analysis became very useful in the study of developmental pathways, and the in-depth studies of the effects that disrupting such pathways have on the embryo. An excellent example of this is the Sperman organiser

transcriptome studies performed by Ding et al. (Ding, Ploper, et al., 2017) in which they measured the effects that β-catenin, Wnt, Nodal, and Siamois alterations have on the early Xenopus embryo transcriptome. 46 RNA-Seq libraries were created and analysed, each from embryos treated with β-catenin depletion, regenerating dorsal and ventral half- embryos, and the overexpression of Wnt8, Siamois, and Cerberus mRNAs among others. These studies confirmed how the effects of β-catenin/ Wnt signalling in early development are regulated by stage-dependent mechanisms and how the dorsal and ventral regions of the embryo communicate with each other.

Such genome-wide studies were encouraged by the release of the complete Xenopus laevis genome (version 9.1) by the Joint Genome Institute (JGI) in 2017. The ease by which Xenopus embryos can be manipulated and the availability of the new genome provided the grounds to make transcriptome studies on dorsal and ventral parts of the early embryo (Figure 1. 3 A) where novel genes were discovered and identified to be expressed in a specific section of the embryo (Ding, Colozza, et al., 2017). Also in 2017, a transcriptomic insight into genetic diversity of protein-coding genes in Xenopus laevis was published by Savova et al (Savova; et al., 2017), in which they crossed the isogenic

J-strain (Harland & Gilchrist, 2017) with outbred albinos and looked at the coding variation between the offspring. This helped elucidate some of the selection pressures affecting duplicated genes in Xenopus laevis. Their findings also contributed to improving the genomic sequence, generating better sequence alignment, probe design, and proteomic analysis.

In summary, these analyses of genome-wide gene expression and mRNA levels have helped to generate a more integrated view of the relationship between the genome, gene expression and phenotypes. However, information about post-transcriptional control of gene expression, changes in protein expression levels, changes in protein synthesis and degradation rates or protein post-translational modifications cannot be achieved with these techniques (Lee, 2001).

Figure 1. 3. RNA levels assayed in the early embryo. pri-mir427 expression over the first 12 hours post-fertilisation (hpf) in rdRNA data and in situ hybridisations. Staining of the embryo correlates with RNA-seq temporal data, showing examples of most abundant transcripts for comparison (A) (Owens et al., 2016). Dorsal and ventral dissections scheme, and the experimental flowchart. Wild-type stage 10.5 Xenopus embryo (B- top), and the dissection of the dorsal and ventral lip (B-bottom). Outline of the RNA-seq analysis pipeline used to analyse gene expression on the dissected embryos (Ding, Colozza, et al., 2017).

Figure 1. 4. Xenopus laevis and Xenopus tropicalis gene expression comparison through the early stages; levels on which developmental biology can be studied.

Xenopus developmental stages from 2-cell stage embryo through to tailbud stage 33 and eight developmentally active genes were taken to illustrate the difference in expression between Xenopus laevis and Xenopus tropicalis transcription activity. These two very related species have distinct genetic activity in the early stages of development, tending to normalise as development progress (A) (Yanai et al., 2011). Advances in technology over the years made possible the molecular analysis of biological systems at multiple levels (DNA, mRNA, protein and activity and function), that in turn made it possible to start to understand the relationship between genotype and any particular phenotype in living organisms (B) (modiffied from Lee, 2001).

Historically, transcriptomics and single mRNA studies provided a dataset that established the mRNA abundance, and so it was thought in the vast majority of cases, the protein level in a given environment. To add an extra level of understanding to the developmental processes, in 2001 Kelvin Lee proposed a method by which to look at the complete protein complement of the genome. Protein studies were enlightening but highly limited by the slow technological advance (Lee, 2001).

By 2014, technology had developed and revolutionised the way scientists had to think.

Development of quantitative proteomics in early stage Xenopus laevis embryos (Sun et al., 2014) and the release of a revolutionary article, which provided evidence of major discrepancies between the mRNA levels and protein levels would be the beginning. The expression of many stored maternal transcripts is activated by cytoplasmic polyadenylation that allows for translation at specific embryonic stages (reviewed in

(Richter, 1999); but the extent to which translational regulation dominates the control of protein levels in the early vertebrate embryo remained unappreciated until proteomic analysis became possible. The lack of correlation between message and protein levels can obscure essential details of development (Sun, Champion, Huber, & Dovichi, 2015).

Figure 1. 5 A provides a graphic representation of the relative mRNA and protein levels in the Xenopus egg.

Some proteins co-exist together with their respective mRNAs (dark blue). However, some mRNAs have no proteins expressed at the studied time (orange), and some proteins have no mRNA present (light blue). These findings had a prominent effect on the scientific Xenopus community, as most developmental studies carried out in this model organism used mRNA as a proxy for protein. In the same year, thanks to the advances in mass-spectrometry, single-cell proteomics in fertilised Xenopus eggs became possible which provided a deeper understanding of cell lineages and heterogeneity during development. These and other published studies reinforced the earlier discrepancy findings (Figure 1. 5 B (Lombard-Banek, Moody, & Nemes, 2016;

Peshkin et al., 2015; Smits et al., 2014; Sun et al., 2016)). Figure 1. 5 A and Figure 1. 5

B illustrate the differences between mRNA and protein levels in some stages of development. More so, hundreds of mRNAs guiding critical tissue-specific functions are mainly regulated at the translation but not transcript level (Fujii, Shi, Zhulyn, Denans, &

Barna, 2017; Werner et al., 2015). Fujii and colleagues also highlighted the importance of translational regulation and its vital role in development. Post-translational modification, such as phosphorylation, also play a significant part in regulating the proteome (Peuchen et al., 2017).

The dominance of translational regulation in the early embryo is not limited to frogs. As technology improves and it became possible to analyse smaller embryos, the same phenomenon has been reported in mice (Gao et al., 2017).

There are 42,878 known Xenopus laevis proteins, according to UniProt (Bateman et al.,

2017; Sun et al., 2014); and the mRNA levels (transcription and degradation rates), protein levels (zygotic vs maternal (Figure 1. 6); translation and degradation rates) and protein post-translational modifications, all determine development. Therefore, to study gene activity and development, all of these aspects have to be considered. Only in this way can a complete picture of the actual state of the function that a gene has in that particular cell type, at a specific moment in time be determined.

Figure 1. 5. mRNA and protein levels in the Xenopus laevis egg. mRNA levels that have no correlated protein are represented in orange; these are mainly transcription factors and zinc finger proteins (A). Shown in blue are mRNAs that have been matched with a corresponding protein, and in light blue on Figure B, proteins that have no mRNA present. These few proteins are maternal and correspond mainly to blood plasma proteins (Wühr et al., 2014). (B) Xenopus laevis fertilised egg single cell proteome vs transcriptome analysis. Protein (measured in fentomoles) and mRNA (measured in reads per kilobase of transcript per Million mapped reads) levels correlate for some genes, however they are considerably different for others (Smits et al., 2014).

Figure 1. 6. Protein amounts in pmol per developmental stage and time in hours post fertilisation.

Maternally acquired proteins stored in the oocyte are more abundant than previously believed. Transcription of developmentally essential proteins is a slow process, and even at stage 33, only half of the total proteins in the embryo are generated by zygotic transcription (Peshkin et al., 2015).

1.5 Antibodies and Research

Antibodies have been successfully raised against some Xenopus proteins, often by immunising rabbits to generate polyclonal antibodies. These were mainly generated in research laboratories as tools to investigate specific, mostly biochemical and developmental questions (J. Maller, E. Jones and T. Hunt antibody collections). One of the largest collections known to the general research community was generated in a cell biology laboratory by Dr James Maller (University of Colorado, USA). These antibodies allowed the elucidation of key pathways in the Xenopus oocyte and egg extract systems, as they target mainly cell cycle components focusing on protein kinases that regulate mitosis (J. Gautier et al., 1990; J. Gautier, Norbury, Lohka, Nurse, & Maller, 1988; X. J.

Liu, 2006; James L. Maller, 2012; Qian, Erikson, Taieb, & Maller, 2001; Wang, Fisher,

Wahl, & Peng, 2011). Tim Hunt (Francis Crick Institute), who worked on cell cycle regulation also generated a variety of antibodies against key proteins (Kobayashi et al.,

1991; Minshull, Golsteyn, Hill, & Hunt, 1990; Strausfeld et al., 1994).

Another collection was developed by Prof. E. A. Jones (University of Warwick UK) against key organ markers for the kidney and heart, and some to the nervous system and muscle proteins (Balakrishna et al., 2011; E. A. Jones, 2003; Lyons et al., 2009;

Miller et al., 2011; Naylor, Collins, Philpott, & Jones, 2009; Simrick, Massé, & Jones,

2005; Vize, Jones, & Pfister, 1995). These were produced by injecting tissue extracts into mice, followed by isolating and characterizing monoclonal antibodies. Of the three major Xenopus collections, these are the only unlimited resource.

These antibodies were characterised by the laboratory of origin and are generally of very limited availability to the broader research community. The only exception being the monoclonal collection from the Jones lab. Outside these collections, laboratories occasionally make one or two antibodies against Xenopus proteins they are studying.

Some examples of these polyclonal antibodies raised in rabbits are against Xcdc6

(Coleman, Carpenter, & Dunphy, 1996), Uhrf1 and Dnmt1 (E. M. Taylor, Bonsu, Price,

& Lindsay, 2013), αxSvv1, αxSvv2, αxEIAP and αxXIAP (Tsuchiya, Murai, & Yamashita,

2005); in sheep Ilf3 (Whitley et al., 2014); or in goats xVelo1 (Nijjar & Woodland, 2013).

Some of these antibodies have been deposited at the EXRC and so become available to the community (https://xenopusresource.org/).

Antibodies that work in Xenopus, such as those described above, will be critical for understanding protein function and control in the early embryo.

1.6 Heavy chain-only antibodies (HCAbs, camelid single-domain antibody, VHH, Nanobodies)

Immunoglobulin G (IgG) is one of the most abundant proteins in vertebrate serum, accounting for about 10–20 % of plasma protein. It is the major class of the five classes of immunoglobulins in human beings, IgM, IgD, IgG, IgA, and IgE (Schroeder, Cavacini,

& Cavacini, 2010). The soluble IgG glycoprotein is produced in vertebrates by B lymphocytes as part of the immune system, these cells recognise and affinity mature in vivo to target the immunogen (Schroeder et al., 2010). The IgG are usually composed of two identical heavy chains and two identical light chains (Press, Givol, Piggot, Porter, &

Wilkinson, 1966). In the early 1990s, a new type of antibody was discovered composed only of heavy chains (Hamers-Casterman et al., 1993) (Figure 1.7). These single-chain antibodies are commonly referred to as nanobodies, VHH or Nbs. They have the particular characteristic that all their coding region is in one single exon. These fragments, once isolated from the conventional variable regions, can be cloned and manipulated with traditional genetic engineering methods (Harmsen et al., 2000). Their vast range of applications, outstanding stability and ease of manipulation shows in the numerous reviews recently published about them (Beghein & Gettemans, 2017; Hu, Liu,

& Muyldermans, 2017; Leow et al., 2017; W. Liu et al., 2018). Therefore, use of VHHs may expand the range of techniques offered by conventional antibodies against Xenopus proteins.

For example, when antibodies to full-length proteins were coupled with organic dyes and used as primary antibodies, the distance between the dye and the protein usually produced linkage errors, a loss of accurate staining at the molecular level (De Meyer et al., 2014). VHHs instead, preserve their stability and fold into functional antigen-binding entities when fused to a fluorescent protein such as GFP (chromobodies) that are too small to produce linkage errors. It has been shown that after expression and binding their specific antigen, these molecules serve as tracers for in vivo intracellular target

localisation studies (Figure 1.8), avoiding the need for genetic modification of target proteins with fluorescent tags (Herce, Deng, Helma, Leonhardt, & Cardoso, 2013). In

2006, Rothbauer and colleagues showed real-time detection of dynamic changes in chromatin, nuclear lamina and the cytoskeleton in animal cells by using VHHs

(Rothbauer et al., 2006).

In another context, Rocchetti and colleagues found that they could inhibit or alter the function of molecules in plants and animals using nanobody technology (Rocchetti et al.,

2014). Historically, depleting a protein in order to study its function in a living organism was usually achieved at the gene level (genetic mutations) or at the RNA level (RNA interference and morpholinos) (Caussinus, Kanca, & Affolter, 2013). Another good example is that of the potato, where it was shown that the correct organelle could be targeted by a nanobody that would inhibit the function of the potato starch branching enzyme more efficiently than an antisense construct (Jobling et al., 2003). In 2014, a

VHH that disrupts fascin F-actin bundling was published (Van Audenhove et al., 2014), quickly followed by other VHHs from the same laboratory that acted blocking different parts of the same pathway (CORNb2 and SH3) (Bertier et al., 2017; Van Audenhove,

Debeuf, Boucherie, & Gettemans, 2015).

The ability of VHHs to interfere with protein conformation, localization or function, together with their size, specificity and stability have made VHHs become significant candidates as therapeutic and diagnostic molecules in medicine, moreover, some have already made it to clinical trials (De Meyer et al., 2014; Deckers et al., 2009; Klooster et al., 2007).

Conventional antibodies are used in chromatin immunoprecipitation (ChiP), where a protein of interest is pulled out of a chromatin preparation to determine the DNA binding sequence associated with it (Hebbes, Clayton, Thorne, & Crane-Robinson, 1994;

Tsanev, 1983). This method is a very powerful tool allowing the study of gene expression, cell differentiation, or disease by looking at the interactions of nuclear

proteins with DNA (Furey, 2012). ChiP has been successfully used to map the localization of post-translationally modified histones, histone variants, transcription factors, or chromatin modifying enzymes (Park, 2009). However practical limitations of mainly requiring large numbers of cells and a lack of antibody development, have highlighted the need for improvement of this method, in turn allowing a more complete picture of biological processes in the cell, especially transcriptional regulation (Furey,

2012).

Technological advances allowed chromatin immunoprecipitation techniques to be coupled with DNA microarray hybridisation (ChIP-chip) and high-throughput sequencing

(ChIP-Seq) for genome-wide identification of DNA–protein interactions (Buck & Lieb,

2004; Robertson et al., 2007). In 2013, VHHs were proved a novel class of affinity reagents for ChIP-chip and ChiP-Seq (Nguyen-Duc, Peeters, Muyldermans, Charlier, &

Hassanzadeh-Ghassabeh, 2013), as well as for ChiP with DNA microarrays (ChIP-on- chip), leading to the discovery of novel transcription factor binding sites (Nguyen-Duc et al., 2013).

Figure 1. 7. Conventional antibodies, Fab fragments and VHH structures.

Size comparison between conventional antibody, composed of 2 light chains and two heavy chains (150 kDa), the heavy chain only antibody (90 kDa), fab fragment (50 kDa) and VHH (15 kDa) (Mujić-Delić, De Wit, Verkaar, & Smit, 2014) (A). Gelsolin-VHH crystal structure, composed of three complementarity-determining regions (CDR1 in yellow, CDR2 in magenta and CDR3 in red), alternated with four framework regions (FRs, in green) (Beghein & Gettemans, 2017) (B).

Figure 1. 8. VHH used to visualize mouse brain structures in vivo and ex vivo. Two-photon microscopy was used to visualize the injected VHH R3VQ-S-AF488 in a one-year- old PS2APP mouse. The mouse brain structures indicated by yellow arrowheads correspond to vascular Aβ and white arrowheads parenchymal Aβ deposits, at 1 minute and 30 minutes post injection compared to T0 (A). The strongest signal was obtained 4 hours post injection (B), just before sacrifice. Immunodetection of the injected VHH on ex vivo brain section stained amyloid plaques by extrinsic R3VQ throughout the brain (C). In this study, two alpacas were immunized, one with AD phospho-tau (pTau) enriched brain extracts and the other with a mono- phosphorylated peptide derived from the C-terminus of the tau protein (pS422 peptide) coupled to the KLH protein. These two libraries were combined and screened as one, were a single clone was found to recognise the tau protein (Herce et al., 2013).

1.7 Importance and contribution of the proposed research

A comparison between the proteomes and transcriptomes of vertebrate embryos has shown that monitoring proteins rather than mRNA will make a critical combination towards understanding cellular and developmental processes. For researchers to identify and understand the genetic regulatory networks that control life, proteins’ levels, and their interaction with one another, RNA and DNA levels must all be determined. The importance of having good antibodies against proteins in the most successful model organism in which to study early development, Xenopus, is therefore clear.

The development of tools which allow the detection, isolation and visualisation of proteins in the early Xenopus embryo would have a cascade effect on the use of Xenopus as a model system. With Xenopus specific antibodies the number and quality of studies performed on this system would flourish. For example, the ability to use VHHs fused to fluorescent proteins would allow researchers to follow proteins in vivo in the early embryo. The possibility of endogenously tagging proteins is another method that could be used to differentiate between closely related proteins. Likewise, the development of these new techniques in Xenopus will encourage new routes of research, for example as important controls for protein knockouts.

1.8 Target proteins for antigen production and antibody generation

The nature of this project, the invaluable link between the EXRC and the wider community, had a key input on the targets chosen to work on. A list was generated with the “most wanted” antibodies from the community: ankrd6, armc8, ascl1, atoh1,

B56alpha, β-catenin, bicc1, ccdc55 (kash5), ccdc78, ccna1 (cyclin A1), ccne1 (cyclin

E1), cdk1 (cdc2), cdk2, cdknx (XicI), cetn1, ctnnb1, dazl, delta1, dlgap5, dvl1, dvl2, dvl3, etv2, eya1, fli1, fus , gata2, gli1, hairy1, mark2, mark3, mkln1, myt1, ncor2 (smrt), nde1, ndel1, neurod1 , ngn2 (neurog2), ngn3 (neurog3), nudc, pafah1b1 (liz1), pi4k2a, pi4k2b, pkl1, pkl2, pkl3, ppp2r5a, prickle3, rarα2, ror2, rusc2, six1, ska1, ska2, snai2, tal1, tbp2, tbx18, tcf21, tra2b, ube2i, vangl2, wnt4b, wt1.

The current knowledge about each gene, its product and function was assessed for ascl1, ctnnb1 (β-catenin), ccdc55 (kash5), cdk1 (cdc2), cdk2, cdknx (XicI), delta1, fli1, gata2, hairy1, myt1, ncor2 (smrt), ndel1, neurod1, ngn2 (neurog2), pafah1b1 (liz1) and rarα2. An example of this for gata2 is shown below. This is limited to a single gene due to space considerations. A brief description of the relevant gene/protein/target region is given at be beginning of each chapter.

1.8.1 Gata2 - GATA binding protein 2

Gata2 encodes a member of the GATA family of zinc-finger transcription factors that have two highly conserved zinc finger motifs and recognise a six-nucleotide motif

T/AGATAG/A in the promotor of target genes (Bresnick, Lee, Fujiwara, Johnson, &

Keles, 2010; Rodrigues, Tipping, Wang, & Enver, 2012). The encoded protein plays an essential role in stem cell maintenance with key roles in blood cell biosynthesis and proliferation, as well as the endocrine cell lineages (Hsu et al., 2011; Ostergaard et al.,

2011).

In Xenopus, gata2 expression is particularly strong in early development as zygotic expression starts at MBT and rapidly decreases from stage 20 onwards (Walmsley et al., 1994; Zon et al., 1991). RNA is localised to ventral blood islands, non-neural ectoderm, epidermis, hemangioblasts and foregut primordium (Watanabe et al., 2017).

The gata2 gene has been studied in Xenopus as one of the major transcription factors active during gastrulation and early development (Figure 1. 9). Its association with stem cell maintenance, hematopoietic development and proliferation have also promoted its study in mammals. However, only one polyclonal antibody was known to recognize the

Xenopus protein and was used in mouse studies before the Xenopus protein expression was characterized (Roget Patient, personal communication). The coding region of gata2 is 1392 base pairs long, the protein 464 AAs. This relatively small size, allowed the consideration of cloning and expressing the full-length protein, but primers were designed to express the N- or C- terminus of the protein avoiding the zinc finger domain conserved between the six family members. Protein folding predictions with Phyre2 and

Psipred suggest this protein does not have a complex folding structure (Figure 1. 9 D).

Thus all the sequence could be considered as an antigen.

To test whether the antibody generated against this protein would not bind, or is not likely to bind to a non-gata2 protein in the Xenopus embryo, a BLAST of the selected antigen sequences against the protein database of Xenbase was carried out (Figure 1. 10). The

outcome of this search provided in silico confirmation that an antibody generated using the chosen Gata2 sequences would interact only with the Gata2 protein in the Xenopus embryo. The limitations of using BLAST to do this check were however revealed when a manual CLUSTAL alignment was performed with the whole Xenopus gata family (Figure

1. 11). The N- terminus of the C- terminal antigen was shown to align to other members of this family. The failure of BLAST to detect this sequence identity was due to the number of EST clones in the database and demonstrates the need for considered bioinformatics approaches to antigen selection.

Gata2 protein sequences

MEVATDQPRWMAHHAVLNGQHPDSHHPGLAHNYMEPTQLLPPDEVDVFFNHLDSQ

GNPYYANSAHARARVSYSQAHARLTGSQMCRPHLLHSPGLPWLESGKTALSAAHHH

NPWTVSPFGKAPLHPAAAGGSLYPGTGSSACPSSSHSSPHLFGFPPTPPKDVSPDP

GTAASPPSSSRLEDKDSIKYQMSLSEGMKMEGGSPLRSSLAPMGTQCSTHHPIPTYP

SYVPAAHDYSSGLFHPGSLLGGPASSFTPKQRSKSRSCSEGRECVNCGATATPLWR

RDGTGHYLCNACGLYHKMNGQNRPLIKPKRRLSAARRAGTCCANCQTSTTTLWRRN

ANGDPVCNACGLYYKLHNVNRPLTMKKEGIQTRNRKMSNKSKKNKKGSECFEELSR

CMQEKSSPFSAAALASHMAPMGHLAPFSHSGHILQTPTPIHPSSSLSFGHPHHSSMV

TAMG*

Gata2

hibridizationexpressionwith

in situin

mRNAvisualized via

Gat

Expression is dormant until stage 8 and rapidly decreses after MBT. (B)

gata2. gata2.

translationalregulation of the gene(taken from Xenbase)

h h post

neural ectoderm, epidermis, hemangioblast and foregut primordium. Only 1362/3899 base pairs encode the coding region coding the encode pairs base 1362/3899 primordium.Only foregut and hemangioblast epidermis,ectoderm, neural

expressionthroughout development.

Gata2

. .

Seq Seq of the long (green) and short (pink) versions of

theseUTRs long consistent are wit

gata2,

Figure 1. Figure1.

(A) (A) RNA

expression is particularly strong is localized to lung, spleen kidney and brain in adults.(C)and spleenkidney localizedlung,strongbrain toexpressionin particularly is is non islands, blood ventral the in of

Figure 1. 10. Xenopus laevis Gata2 protein sequence alignment against the Xenopus laevis protein database (Xenbase BLAST). The Gata2 protein sequence was tested in silico using BLAST for cross-reactivity with other Xenopus proteins, especially the other 5 members of the Gata family. The sequence did not recognize any other protein than Gata2, minimizing the chances of cross-reactivity with other Xenopus proteins.

Gata5 ------0 Gata4 ------0 Gata6 MDLSEDSWSGGTRLSPENSFFPCPIKQSPGTSPPPSPCSHVGSSECLRSEPDRSQQNIST 60 Gata1 ------0 Gata2 ------0 Gata3 ------0

Gata5 ------MYPSLA 6 Gata4 ------MYQSIA 6 Gata6 EGRPSASYHPLGLSHPEDLILFRDLDHINKLVVPNREPRYSSTLPEPPSGD---MYQTLT 117 Gata1 ------MDYTTLTTQDP------DPNYTESGLASTSEDSQFLYGL-- 33 Gata2 --MEVATDQPRWMAH---HAVLNGQHPD--SHHPGLAHNYMEPTQLLPPDEVDVFFNHLD 53 Gata3 --MEVSAEQPRWVSHPHHPALLNGQHTD--SHHQ--VHNYMDPSQYPPPEDMDVLFNID- 53 ::

Gata5 LTANHAQPAYSHDTPNFLHS---TGSPPVYVPTSRMPAMLQS--LPYLQTCDTAHQGHHL 61 Gata4 MATNHGPPGYEG-AGSFMHSAT-AATSPVYVPTTRVSSMIPS--LPYLQTSGSSQQSSPV 62 Gata6 ITAAQGPLGYDPSPGTFMHS---AASSPVYVPTSRVGSMLTS--ISYLQGTGASQGTHSV 172 Gata1 -GGESSPGHYGGAVSSRAVGGF------RHSPVFQTFPLHWPETSAGI------74 Gata2 SQGN---PYYANSAHARARVSYSQAHARLTGSQMCRPHLLHSPGLPWLESGKTALSAAH- 109 Gata3 AQGNHVPSYYSNSVRTVPR--Y---PPPHHGSLVCHPSILQS----WT-DGRKSLGGPH- 102 * :: : :

Gata5 ANHPGWAQT------ANESHAFNASSPHT--PSGFSYSHSPPVGN 98 Gata4 ASHNMWAQA------GAESSAYNPGTSHPPVSPRFTFSSSPPITA 101 Gata6 N--SHWSQA------TSESSSYSSSSPHP--SSRYHYSPSPPMAN 207 Gata1 ----PSNLTAYG------RSTGTLSFYPSAASALGP--ITSPPLYSASSFLLGSAPPAER 122 Gata2 -HHNPWTVSPFGKAPLHPAAAGGSLYPGTGSSACPSSSHSSPHLF-----GFPPTPPKDV 163 Gata3 -AASAWNL-PFAKPSIHHNSPGGLSVYPSASSASLATGHSSPHLF-----TFPPTPPKDV 155 : :**

Gata5 SSARDGG------YQSPLIMGGSA--R-DQYGNTLVRTGSY------130 Gata4 ASTREVS------YSSPLAISANG--R-EQYSRGL--GATY------131 Gata6 GSTRDTG------YSSSLAVSGRD--QYAPLARPL--NGSY------238 Gata1 EGS------PKFLE-----TLKTERASPLTSDLLPLEP------149 Gata2 SPDPGTAASPPS-SSRLEDKDS--IKYQMSLSE-GMKMEGGSPLRSSLAPMGTQCSTHHP 219 Gata3 SPDPSISTPGSTSSSRHEEKECVVSKYQVSLGDPGMKLESSH-PRNSMSGIGGVPSAHHP 214 : :

Gata5 PSPY-SYVGADMPPSWATGHFEGSMLHSLQGRQ-PLPGRISSLEFLEEFHGEGRECVNCG 188 Gata4 SSPYPAYMSPDMGAAWTASPFDSSMLHNLQNRAGPVASRHPNIEFFDDF-SEGRECVNCG 190 Gata6 GSPYTPYMTPQLTSAWPAGPFDNTMLHSLQSRGAPISVRGAPGDVLDEL-PESRECVNCG 297 Gata1 ------RSPSILQVGYI------GGGGQEF------SLFQSTEDRECVNCG 182 Gata2 IPTYPSYVPAAH--DYSSGLFHPGSL-----LGGPASSFTPKQRSKSRSCSEGRECVNCG 272 Gata3 ITTYPDY------YGAGLFQPGSI-----LDGSPTHFPSKPRPKTRSSTEGRECVNCG 261 . :. : *.*******

Gata5 AMSTPLWRRDGTGHYLCNACGLYHKMNGINRPLIKPQKRLSSSRRAGLCCTNCHTSTTTL 248 Gata4 AMSTPLWRRDGTGHYLCNACGLYHKMNGINRPLIKPQRRLSASRRVGLCCANCHTTTTTL 250 Gata6 SVQTPLWRRDGTGHYLCNACGLYSKMNGLSRPLIKPQKRVPSSRRIGLACANCHTTTTTL 357 Gata1 ATVTPLWRRDMSGHYLCNACGLYHKMNGQNRPLIRPKKRLIVSKRAGTQCSNCHTSTTTL 242 Gata2 ATATPLWRRDGTGHYLCNACGLYHKMNGQNRPLIKPKRRLSAARRAGTCCANCQTSTTTL 332 Gata3 ATSTPLWRRDGTGHYLCNACGLYHKMNGQNRPLIKPKRRLSAARRAGTSCANCQTTTTTL 321 : ******* :*********** **** .****:*::*: ::* * *:**:*:****

Gata5 WRRNSEGEPVCNACGLYMKLHGVPRPLAMKKESIQTRKRKPKNIGKGKTSTGSSTSANN- 307 Gata4 WRRNAEGEPVCNACGLYMKLHGVPRPLAMKKEGIQTRKRKPKNLSKSKTSTGQSGSDSL- 309 Gata6 WRRNTEGEPVCNACGLYMKLHGVPRPLAMKKEGIQTRKRKPKNLNKSKSSSSNGNSSHHI 417 Gata1 WRRNASGDPVCNACGLYYKLHNVNRPLTMKKEGIQTRNRKVSSRSKKKKQLDNPFEPPKA 302 Gata2 WRRNANGDPVCNACGLYYKLHNVNRPLTMKKEGIQTRNRKMSNKSKKNKKGSECFEELSR 392 Gata3 WRRNANGDPVCNACGLYYKLHNINRPLTMKKEGIQTRNRKMSSKSKKSKKIHDSLEDY-- 379 ****:.*:********* ***.: ***:****.****:** .. .* ... . .

Gata5 --SP-SS-----VTNTDPTPVLKTEPNITSHYPGQAIVPVSQGQS-----QTDDLVNGSH 354 Gata4 --TP-STS--STNSTGEEMRPIKIEPGLSPPYGHSS--SISQASSLSTITSHGPSSYPMS 362 Gata6 TMTP-TST--TSSTNSDD--C--IKNGS-PS-QNTA--PVV-TSSLMSAQQTGSASPDSN 465 Gata1 GVEEPSPYPFGPLLFHGQMPP------MGHMI--NPP--H---HFLQSPRISHSAP 345 Gata2 CMQE-KSSPFSAAALASHMAP------MGHLA--PFSHSG---HILQTPTPIHPSS 436 Gata3 ---Q-KSSSFSPAALSRHMSS------LSHIS--PFSHSS---HMLTTPTPMHPPS 420 .

Gata5 ELKYIPEEYTYSPSAL---SQQSVPLRQESWCALALA-- 388 Gata4 SLKLSPQNHH-----STINSSPQASSKHDSWNNLVLA-- 394 Gata6 ILKYTGQDGLYSAVSLSSASEVAASVRQDSWCALALA-- 502 Gata1 AVSYRQAA------SG--VTPP- 359 Gata2 SLSFGHPH------HSSMVTAMG 453 Gata3 SLAFGPHH------PSSMVTAMG 437 : .

pEneoFigure-3C 1.- 112Strep. Gata-8His binding-Gata2 protein-CTer; family pEneo -sequence3C-3FLAG- alignment.Gata2-CTer ; pOPINFS-Gata2- CTer; pOPINTTGneoGata2 binding-Gata2 protein; pOPINGM sequence-Gata2 aligned- CTer to the other 5 members of the Gata family. Upon the pOPINFalignment - Gofata the2 Gata family protein sequences, regions of similarities between the sequences were found. These findings questioned the usability of BLAST as an accepted approach to test pOPINE - Gata2 protein similarities in silico. In Figure 1.11, the extended number of EST sequences present in the pOPINE-Gata2-Short; pOPINF-Gata2-Short databases, produce a large number of hits that are, in this case, all Gata2; however, the sequences that partially align are not shown. Protein sequences used in expression vectors are highlighted in 4 distinct ways, yellow, underlined, font colour blue and red.

2 Materials and Methods

Throughout this thesis a wide variety of methods were used, ranging from traditional cloning and protein expression to the use of ground breaking techniques such as generating and screening camelid single-chain antibodies and CRISPR technologies.

Consumables were acquired primarily from Sigma Aldrich, Fisher Scientific and VWR; enzymes were from NEB; antibodies from Abcam and Santa Cruz Biotechnology; primers and oligonucleotides were from Gene Tools and Thermo Fisher; plasmids were from the EXRC and Addgene.

In 2016, we noted animal welfare issues with Santa Cruz antibodies and stopped using them at that point.

2.1 Nucleic acid techniques

2.1.1 Restriction digests

Restriction digests were carried out according to the manufacturer’s instructions. In synthesis, between 1 µg and 4 µg of the plasmid were incubated at the suggested temperature for optimal efficiency of the enzyme, together with the enzyme and its corresponding buffer. The reaction was typically set up in 20 µl total volume and incubated for 2 hours, unless otherwise stated. In a 20 µl reaction, 2 µl of 10x buffer and a maximum of 2 µl of enzyme (20 units) was used.

2.1.2 DNA and RNA separation and visualization though agarose gel electrophoresis

Agarose gels of 1-4 % were used to separate and visualize nucleic acids by their size.

The gels were prepared in TBE and ethidium bromide (10 µg/ml) was included in the gel.

They were run at 40-120 V until the blue dye was ~2/3 of the way down the gel.

The samples were mixed with a loading dye before loading into the gel, to aid visualisation of the sample with the naked eye. This also contained glycerol, making the sample dense and forcing it to fall into the agarose well. The purple (6 X) loading dye from NEB was used (catalogue number B7024S).

After exposure to the current, the gels were visualized with UV light in a Gel-Doc and photographed. If the DNA/RNA had to be extracted, the appropriate bands were excised from the gel before photographing (as the excessive exposure of the DNA/RNA to UV light degrades DNA) and gel extracted using either the Qiaquick gel extraction kit or the

Gene Clean method, according to manufacturer’s instructions.

2.1.3 Ligation by T4 DNA ligase

Thermo Fisher Scientific T4 DNA ligase (EL0011) was used to ligate sequences of interest with pre-digested vectors, following the manufacturer’s instructions.

The ligation reactions typically contained a 3:1 insert to vector ratio in molar quantities, and was incubated overnight at 16ºC. If the ligation was unsuccessful, then the molar ratios and the incubation temperatures were optimized.

Table 2. 1. T4 DNA ligase reaction components.

T4 DNA ligase Vector DNA Insert DNA Nuclease free T4 DNA ligase buffer (10 x) (4 kb) (1 kb) water

50 ng 37.5 ng 3 µl To 30 µl 2 µl (0.020 pmol) (0.060 pmol)

2.1.4 pGEM®-T Easy Vector Systems (Promega A1360)

Historically, primers were designed with extra bases outside the restriction site, and amplicons digested before annealing to the pre-digested destination vector. The rate of insertion via this method was unreliable. The pGemTeasy TA cloning vector allows the direct cloning of PCR products into a vector efficiently. The amplification of the target sequence with a Taq polymerase results in amplicons containing an A overhang at the

3’ end of the molecule, facilitating the base pairing with the T overhangs on the TA cloning kit vectors. These overhangs base-pair non-directionally with the aid of T4 DNA ligase. For these reactions, between 100 ng and 500 ng of fresh PCR product was mixed with 50 ng of pGemTeasy vector and the corresponding buffer, incubated at room temperature for 1 hour or at 4ºC overnight, before transforming into E. coli DH5α competent cells. Once the pGemTeasy sub-clone was prepared, the required insert was excised by restriction digest and ligated into its final destination vector. This proved much more efficient than direct cloning of the PCR product.

2.1.5 Transformation of E. coli DH5α

E. coli DH5α competent cells were taken from - 80°C freezer and set on ice to thaw before adding 50-200 pg of DNA. The cells were incubated on ice for 30 minutes (for kanamycin resistant plasmids, is essential to have a 10 minutes recovery phase at 37°C at this point) before heat shock at 42°C for 45 seconds and placed back on the ice for 2 minutes to reduce damage to the cells. One ml of LB broth (with no antibiotic) was added and the cells incubated for 1 hour at 37°C (shaking at 200 rpm or static). 100 µl of the resulting culture were spread on LB plates (with the appropriate selective antibiotic) and grown at 37ºC. Colonies were picked 12 to 16 hours later.

2.1.6 Polymerase chain reaction (PCR)

Polymerase chain reactions were all performed in an Applied Biosystems Veriti 96 well

Thermocycler. The conditions were adjusted according to the polymerase GoTaq® G2

Hot Start Master Mixes from Promega; unless otherwise stated. Conditions are shown in

Table 2. 2 and reactions set up in Table 2. 3. Table 2.4 and Table 2.5 outline primers used in Chapter 5 for the amplification of kash5 from cDNA.

Table 2. 2. PCR reaction standard conditions.

Step Temperature Length

Initial denaturing 94ºC 2’

Denaturing 98ºC 10” 29 cycles Annealing 60ºC 30”

Extension 68ºC 2’

Final extension 68ºC 2’

Hold 4ºC

Table 2. 3. Standard PCR reaction (25 µl).

Reagent Quantity

Polymerase master mix 12 µl

Primer Fw (10 µM) 1.5 µl

Primer Rv (10 µM) 1.5 µl

DNA (50-200 ng) 1-2 µl

ddH2O 8-7 µl

Table 2. 4. kash5 gene specific primers for Xenopus laevis.

Xenopus laevis kash5 primer Sequence

Xla Kash5 EcoRI Fw gaattcggcaaagtagggctatacaatg

Xla Kash5 EcoRI Fw1 gaattcgggctatacaatgtataaacacctagg

Xla Kash5 EcoRI Fw2 gaattctcccctttaactagtaattgcatagc

Xla Kash5 XhoI Rv ctcgagcctttaaactggaggtggag

Xla Kash5 XhoI Rv1 ctcgagggaggtggagaagggtaatgc

Xla Kash5 XhoI Rv2 ctcgagaatgcaagtccagatgaggc

Table 2. 5. kash5 gene specific primers for Xenopus tropicalis.

Xenopus tropicalis kash5 primer Sequence

Xtr Kash5 EcoRI Fw gaattcactctccaggaatacatctgc

Xtr Kash5 DK fwd BamHI gttaggatcctataagctgcagggtacctg

Xtr Kash5 DK rev EcoRI atagaattcttacaactgaatacacggaacgatgctg

Xtr Kash5 DK rev HindIII ataaagcttttacaactgaatacacggaacgatgctg

Xtr Kash5 DK rev NotI atagcggccgcttacaactgaatacacggaacgatgctg

2.1.7 RNA extraction from embryos

Embryos were collected in batches of #1 = 5 embryos or #2 = 25 embryos, and frozen at -80ºC for a maximum of 12 weeks.

The embryos were mechanically homogenised in NETS buffer (Table 2. 6) by pipetting up and down (#1 = 200 µl; #2 = 400 µl) and the homogenate added to an equal amount of buffered phenol. The samples were vortexed for 30 seconds and centrifuged at 16000 xg for 4 minutes.

The aqueous top layer was carefully removed and transferred to a fresh tube containing an equal amount of 50 : 50 phenol : CHCl3, vortexed for 30 seconds and centrifuged at

13000 xg 4ºC for 4 minutes. The aqueous top layer was carefully placed in a clean tube containing 2.5 x volumes of ethanol, vortexed and incubated overnight at -20ºC.

Following overnight incubation, the sample was centrifuged at 13000 xg at 4ºC for 30 minutes. The supernatant was removed, and the pellet was left to air-dry for 10 minutes.

The dry pellet was re-suspended in 90 µl of ddH2O and aliquoted into 2 x 45 µl.

Two microliters of each sample were removed and placed on ice, to the rest 5 µl 10 x

DNase Buffer and 2 µl of DNaseI enzyme (Thermo Fisher Scientific, EN0521) were added and incubated at 37ºC for 30 minutes. Two microliters of the reaction were taken out and left on ice.

Following the DNaseI treatment, 50 µl ddH2O and 100 µl phenol were added and the solution vortexed 30 seconds followed by centrifugation at 13000 xg and 4ºC for 4 minutes. The top aqueous layer was transferred to a clean tube with 10 µl 3 M NaAc pH

5.5 and 250 µl of ethanol. The tubes were incubated at -20ºC for 2 hours to precipitate the RNA. The samples were centrifuged at 13000 xg and 4ºC for 10 minutes, the ethanol removed and the pellet air-dried before re-suspending in #1 = 10 µl and #2 = 20 µl of ddH2O.

Table 2. 6. NETS buffer recipe for RNA extraction from Xenopus embryos

NETS buffer

30 mM NaCl

1 mM EDTA

20 mM Tris, pH 7.5

1 % SDS

ddH2O

2.1.8 cDNA synthesis from Xenopus total RNA

To synthesize cDNA, Superscript 3 (Invitrogen) was used following the manufacturer’s instructions. The reactions were set up in clean microcentrifuge tubes as shown in Table

2. 5 and incubated at 50ºC for 60 minutes before freezing at -20ºC until needed. 1-2 µl of this material were used in each RT-PCR.

Table 2. 7. cDNA synthesis from total RNA by reverse transcription.

Reagent Amount used

RNA 4-5 µl

10 x reaction buffer 2 µl

100 µM random hexamers/nonamers 2 µl

2 mM dNTP mix 4 µl

Superscript 3 reverse transcriptase 0.25 µl

ddH2O to 20 µl

To check the cDNA integrity by PCR, a housekeeping gene was amplified, in this case,

ODC (forward primer 5′ GGA GCT GCA AGT TGG AGA 3′, reverse primer 5′ CTC AGT

TGC CAG TGT GGT C 3′). The expected product size was ~250 bp (Figure 2. 1).

Table 2. 8. Isolation and amplification of a specific region from total cDNA

Reagents Amount used

Polymerase (2 x green ready mix) 10 µl

Primer Fw (10 µM) 1.5 µl

Primer Rv (10 µM) 1.5 µl

cDNA 1-2 µl

ddH2O 6-5 µl

Figure 2. 1. cDNA synthesised from Xenopus total RNA by reverse transcription. To check the cDNA integrity, a housekeeping gene (ODC) was amplified. Lanes 1, 2 and 3 represent three reverse transcription reactions, and lanes 4, 5 and 6 the amplification of the ODC gene from the same 3 reactions. Reaction number 2 produces a single band of just around 250 bp when amplified with the ODC primers, correlating with the expected size for the ODC amplicon. The cDNA from reaction number 2 was used further.

2.1.9 In vitro protein production by the 1-Step Human High-Yield Mini IVT Kit (Thermo Scientific, 88885)

2.1.9.1 Cloning of target site into a suitable vector

Several vectors were considered, however, PCR2.1 TOPO was chosen to clone the insert into since this vector does not have restriction sites for either SalI or NdeI (enzymes to be used downstream).

2.1.9.2 PCR

The sequences of interest (neurog2 and xic1) were obtained from plasmids stored at the

EXRC (European Xenopus Resource Centre). These sequences were amplified with the primers below under standard PCR conditions (see Polymerase chain reaction section).

Table 2. 9. Primer sequences used to amplify neurog2 and xicI, which later were cloned into the pT7CFE1-cGST-HA-his vector to express Neurog2 and XicI protein in vitro.

Fw CATATGATGGTGCTGCTCAAGTGCG neurog2 Rv GTCGACAATGAAAGCGCTGCTGGCAG

Fw CATATGATGGCTGCTTTCCACATCGC xicI Rv GTCGACTCGAATCTTTTTCCTGGGGG

2.1.9.3 TOPO TA cloning.

The appropriate reagents were mixed and incubated for 5 minutes at room temperature

(22-23°C). Once the TOPO Cloning reaction was performed, the clones were transformed into E. coli DH5α cells following standard procedures (see Transformation section).

2.1.9.4 Sub-cloning into pT7CFE1-cGST-HA-his

The pT7CFE1-cGST-HA-his vector was included in the 1-step human high-yield mini IVT kit. This vector was digested with NdeI and SalI; the amplified inserts were ligated using

a standard T4 DNA Ligase protocol (see Ligation by T4 DNA ligase section), transformed and grown up. Correct plasmids were obtained as checked by restriction digestion of mini-preps and sequencing.

2.1.9.5 Protein expression using the 1-Step Human High-Yield Mini IVT Kit

Protein expression was performed following the manufacturer’s instructions. To measure protein production, standard western blot, silver staining and BCA protein assay methods were combined.

2.1.10 HEK cell protein expression using pOPIN plasmids and the In-Fusion system.

2.1.10.1 Destination Vectors

The OPPF (Oxford Protein Production Facility) designed a series of vectors which expressed proteins in multiple hosts: E. coli, mammalian cell lines (e.g. HEK293T cells) and insect cell lines (e.g. Sf9 cells), allowing parallel screening in multiple hosts (Bird,

2011). These vectors, in combination with the In-Fusion™ enzyme, have produced an efficient cloning procedure in which the PCR product is cloned directly into the final expression vector.

The pOPIN vectors encode a variety of tags to aid solubility and purification of the resulting proteins:

a. pOPINF-his-tag at the N-terminal (Plasmid 26042) b. pOPINM-MBP (maltose binding protein) tag + his-tag (Plasmid 26044) c. pOPINJ-GST (glutathione-S-transferase) tag + his-tag (Plasmid 26045) d. pOPINE – his-tag at the C-terminal (Plasmid 26043)

The empty backbones were acquired from Addgene (www.addgene.org).

Upon arrival, stocks were generated following transformation and growth using the

NucleoBond MidiPrep Kit (Macherey-Nagel) instructions. This was followed by the

linearization and gel purification of the vector using the MinElute Gel Extraction Kit acording to the manufacturer’s instructions (Qiagen).

Table 2. 10. . Digestion of pOPIN vectors.

O (µl) O

(µl) (µl) (µl) (µl)

x Buffer

Total

Vector

Enzyme Enzyme 2

Enzyme 1 Enzyme 2

Temp (°C)

DNA to DNA use Volume (µl)

DNA (ng/µl)DNA

Buffer to use

CutSm pOPINE NcoI 2 PmeI 2 303 7 10 79 100 37 art

NEBuff pOPINF KpnI 2 HindIII 2 1027 2 10 84 100 37 er 2.1

NEBuff pOPINM KpnI 2 HindIII 2 1736 1.5 10 84.5 100 37 er 2.1

NEBuff pOPINJ KpnI 2 HindIII 2 400 5 10 81 100 37 er 2.1

2.1.10.2 Proteins of Interest

Following input from the community, a list of proteins of interest was generated whose expression had previously been attempted in E. coli with limited success.

a. XicI b. Neurog2 c. Gata2 d. ETV2 e. FliI f. SCL g. xHairyI h. xMit1 i. xNeuroD j. xDeltaI k. RARα2 (N-terminal) l. RARβ m. Ncor2 (SMRT) As proof of principle, we decided to only work on 3 of these proteins (underlined) in the first instance with the aim of subsequently moving to a larger scale.

2.1.10.3 Primer design

Primers were designed using the SnapGene program (GSL Biotech LLC, n.d.). These were designed to amplify a fusion protein-specific sequence together with a 15-30 bp sequence at the 5’ and 3’ end homologous to termini of the digested vector. Vector maps and primers sequences can be found in Appendix I.

2.1.10.4 PCR with specific primers to isolate insert and add specific ends to enable the In-Fusion reaction

PCR reactions were carried out with 20-30 ng of plasmid containing the desired sequence and specific primers (see 2.1.6). Resulting amplicons were analysed by gel electrophoresis and used if a single band of the expected size was observed. If multiple bands were present, the correct size amplicon was gel extracted and purified using the

MinElute Gel Extraction Kit using the manufacturer’s instructions (Qiagen).

2.1.10.5 DpnI treatment

OPPF recommend that “if the PCR template has the same antibiotic resistance as your target pOPIN vector you must DpnI treat your PCR reaction” (Bird, 2012) to remove template plasmid before transformation.

The DpnI enzyme has specificity for methylated DNA and digests template without digesting the PCR product. The DpnI enzyme is active in most PCR reaction buffers and can therefore be added to the PCR product. Thus, 2.5 μl of DpnI enzyme were added to

20 μl of PCR reaction mix and incubated at 37°C for 45 minutes in the thermal cycler.

The reaction was stored at -20°C until further use.

2.1.10.6 Cloning enhancer treatment

Following PCR, the products showed a discrete single band. Clontech recommends treating products composed of a single amplicon with Cloning Enhancer, a proprietary enzyme mix for removing background plasmid DNA and PCR residue. This step eliminates the need for PCR insert purification before cloning.

For this treatment, 2 μl of Cloning Enhancer (Clontech) were added to 5 μl of PCR product and incubated at 37°C for 20 minutes; the temperature was raised to 80°C for a further 25 minutes and the treated product was used immediately or stored at -20°C until needed.

2.1.10.7 In-Fusion reaction

The In-Fusion PCR Cloning System (Clontech) allows the fusion of extended PCR fragments with a region of homology at both 3’ and 5’ ends to the homologous ends of a linearised vector. The composition of the In-Fusion reaction has not been disclosed by the company. One theory found in Researchgate (/www.researchgate.net) is that it “uses vaccinia virus DNA polymerase”, however this is not an official comment. The public knowledge about In-Fusion goes as far as knowing that is a methods consisting of a single-tube enzymatic reaction based on homolog fragment ends and it requires incubation at 50ºC.

The In-Fusion reactions components are shown in Table 2. 11. Once assembled, the reaction was incubated at 50°C for 15 minutes, followed by the transformation into E. coli

Stellar competent cells (provided with the In-Fusion kit) using standard methods.

DNA from transformants was prepared as mini-preps (Qiagen) and sent for sequencing

(Source Bioscience).

Table 2. 11. In-Fusion reaction components.

5 x In-Fusion Linearized PCR fragment dH2O

HD Enzyme (μl) Vector (μl) (μl) (μl)

pOPINF 2 2 1.5 4.5

pOPINE 2 4.5 1.5 2

+ve control 2 2 1.5 (control) 4.5

-ve Control 2 2 0 6

2.1.11 Cas9 RNA synthesis

Cas9 was transcribed from its containing plasmid (a kind gift from Dr Marko Horb) by

SP6 polymerase using the mMachine Kit (Ambion) following the manufacturer’s protocol.

Three distinct reactions were set up to generate Cas9 RNA by transcription with variations on the starting DNA template (see table below), all such reactions produced good quality RNA with yields ranging from 200-3000 ng/µl.

Table 2. 12. Transcription of Cas9 by SP6 polymerase.

Reaction 1 Reaction 2 Reaction 3

2 x NTP/CAP analog (µl) 10 10 10

10 x Reaction mix (salts, buffer, dithiothreitol, and other 2 2 2 ingredients) (µl)

1 x SP6 enzyme mix (Buffered 50% glycerol containing RNA

polymerase, RNase 2 2 2

Inhibitor, and other components) (µl)

Linear template DNA 120 ng/µl 3 4 5 (µl)

dH2O (µl) 3 2 1

2.2 gRNA production gRNAs were designed to anneal to a particular sequence in the genome, guiding Cas9 to make the double-stranded cut 3 bp away from the PAM site (Mojica, Díez-Villaseñor,

García-Martínez, & Almendros, 2009). The sequence of gRNA that anneals to the target sequence is 20-25 bp long (uppercase). To the 5’ of this sequence, the T7 polymerase promoter sequence was added. To the 3’ of the target sequence, 15 bp were added that anneal to a standard oligonucleotide, which encodes a part of the gRNA that contains a secondary structure that binds to the Cas9 protein. gRNA #1:

5’ taa tac gac tca cta taG ATA AAT AAT AAC TCT GAC GTG Ggt ttt aga gct aga a 3’ gRNA #2:

5’ taa tac gac tca cta taC CAC ACC ACT CCA GCA TGG TGA Cgt ttt aga gct aga a 3’ gRNA #3:

5’ taa tac gac tca cta taT GCT GAG AAG ACA GTG TAA Cgt ttt aga gct aga a 3’ gRNA #4:

5’ taa tac gac tca cta taG TCA CCA TGC TGG AGT GGT Ggt ttt aga gct aga a 3’

The gRNAs were synthesised as described in Nakayama et al. 2014. In summary, the designed gRNA encoding sequences were annealed to a universal single-stranded sequence and filled in using Taq polymerase to make a double-stranded DNA sequence.

Universal 3’ gRNA sequence:

5’ AAA AGC ACC GAC TCG GTG CCA CTT TTT CAA GTT GAT AAC GGA CTA GCC

TTA TTT TAA CTT GCT ATT TCT AGC TCT AAA AC 3′

Table 2. 13. Components and steps to synthesise double stranded DNA from ssDNA by touch down PCR

PCR master mix 50 µl Initial denaturing 94ºC 5’

Custom gRNA (100 µM) 2 µl Denaturing 94ºC 20”

Universal gRNA (100 µM) 2 µl Annealing 58ºC 15”

MgCl2 (25 mM) 2 µl Extension 68ºC 15”

dH2O 44 µl Final extension 68ºC 5’

Reaction total 100 µl Hold 4ºC -

When a strong band was visible from the PCR product on an agarose gel, RNA was synthesised using 6 µl of the PCR product in the Megashortscript™ T7 Transcription Kit

(Thermo Fisher AM1354) following the manufacturer’s instructions; the RNA product was then visualised on a 1.5 % agarose gel.

2.3 Microbiological techniques

2.1.1 Bacterial strains and phage used

2.3.1.1 TG1 and BL21 cells for phage infection

Cells for phage infection were obtained from the Sokol lab at Mount Sinai Hospital (New

York, USA). To grow a stock, an LB plate with no antibiotic was streaked with a sterile tip containing an aliquot from the stock vial. After overnight incubation at 37ºC, a single colony was used to inoculate 4 ml of 2xYT media containing no antibiotic. The starter culture was left to grow overnight at 37ºC shaking at 220 rpm. The starter culture was used to inoculate 400 ml of 2xYT media with no antibiotic (1/100). This culture was incubated to an OD600 of 0.3 to 0.6. Cells were centrifuged at 1000 xg for 15 minutes at

4ºC. The supernatant was discarded and the cell pellet re-suspended in 10 ml cold filtered TSB medium on ice. Cells were incubated for 10 minutes on ice and aliquoted as

500 µl before snap freezing in liquid nitrogen and storage at -80ºC until further use.

Table 2. 14. TSB (transformation and storage buffer) and 2xYT medium used to grow different strains of E. coli DH5α cells

TSB 2xYT

LB or YT 1 x Bacto-tryptone 1.6 %

NaCl 43 mM Yeast extract 1 %

PEG 10 % NaCl 85 mM

DMSO 5 % dH2O

MgCl2 10 mM pH 7.0

MgSO2 10 mM

Glycerol 10 %

dH2O

2.3.1.2 HBV88 cells – Electro-competent cells

HBV88 cells, originally a gift of A. Hayhurst to the Sokol lab (Mount Sinai Hospital, USA), contain a chloramphenicol resistance plasmid encoding a suppressor tRNA upon arabinose induction. These cells were made electro-competent (ready for electroporation) using low-temperature growth in low salt YENB media, followed by extensive washing.

Bacteria were streaked on M9/chloramphenicol/thiamine minimal agar (to select for the

F’-episome). A single colony was cultured in 2 ml of M9 medium for 8 hours, and 200 μl of the culture was added to 2 x 50 ml liquid M9 before incubating overnight at 37°C shaking at 220 rpm. 10 ml of the culture were added to six 400 ml flasks of YENB supplemented with chloramphenicol to make an OD600 of 0.05 and incubated at 30°C until the OD600 reached 0.4 - 0.6 (approximately 3 hours at 37°C or 6 - 7 hours at 25°C).

After chilling the cultures on ice for 5 minutes, cells were pelleted in 6 bottles at 4080 xg,

4°C for 20 minutes. The following procedures were performed in the cold room. Each pellet was re-suspended gently in 50 ml ice-cold deionised water and transferred into 2 x 50 ml conical tubes (approximately 25 ml each); ice-cold water was added to 50 ml.

Cells were pelleted at 4080 xg at 4°C for 15 minutes and re-suspended in 25 ml of ice- cold water. Water was added to 40 ml and the cells re-pelleted in the same conditions.

Cells were washed in 25 ml with ice-cold 15 % glycerol, pelleted again and finally re- suspended in 8.7 ml of 15 % glycerol. Competent cells were aliquoted into 48 x 270 μl in

Eppendorf tubes (pre-chilled), frozen in dry ice and stored at -80ºC. Transformation efficiency of 5x108 / μg (instead of expected 109) with pecan126 was achieved.

Table 2. 15. M9 minimum agar and YENB medium.

M9 minimum agar YENB

Sodium phosphate dibasic (anhydrous), Yeast extract 0.75 % Na2HPO4

Sodium chloride, NaCl Nutrient broth 0.8 %

Ammonium chloride, NH4Cl H2O

1 M Magnesium sulphate, MgSO4 pH 7.0 with NaCl

0.1 M Calcium chloride CaCl2

10 % (w/v) glucose (or another carbon

source)

2.3.1.3 M13K07 helper phage:

TG1 cells were grown in LB broth at 37°C with shaking (220 rpm) until an OD600 of 0.8 was reached. Helper phage (NEB catalogue #N0315S, 1 x 1011pfu/ml) were diluted 10-

4, 10-5, 10-6 and 10-7 -fold. Ten microliters of the diluted M13K07 were mixed with 100 μl of TG1 cells and poured with 3 ml top agar onto 10 cm LB plates with no antibiotic. The plates were cultured at 37°C for 14-16 hours. Several individual plaques were picked, suspended individually in 3 ml of 2xYT media supplemented with kanamycin in 15 ml round bottom tubes and cultured overnight at 37°C. The cultures were centrifuged at

13000 xg for 2 minutes, the supernatant collected and the titre measured by diluting the phage and plaking with TG1 and top agar as described above. The titre obtained was 2 x 1012 pfu/ml.

Table 2. 16. Top agar, agar used to propagate phage recipe.

Top agar

Bacto-tryptone 1.6 %

Yeast extract 1 %

NaCl 85 mM

Granulated agar 0.75 %

H2O

pH 7.0 with NaCl

2.3.1.4 Preparation of E. coli DH5α competent cells (from pre-existing stock)

Cells from an existing stock were streaked on an agar plate with no antibiotic and incubated overnight at 37°C. Four colonies were picked, incubated individually in 5 ml

LB broth with no antibiotic and incubated with shaking overnight at 37°C. These starter cultures were used to inoculate 4 x 500 ml of LB Broth that were incubated at 37°C with shaking until the OD600 reached 0.5 (~3.5 hours). The cultures were then cooled at 4ºC for 5 minutes, transferred to 1 L bottles and centrifuged using a JLA8.1000 centrifuge, at

4000 xg for 5 minutes. The supernatant was discarded and the pellet re-suspended in cold FB1 buffer (Table 2. 17). The cells were incubated on Ice for 90 minutes and centrifuged at 4000 xg for 5 minutes. The supernatant was discarded and the pelleted cells re-suspend in ice-cold FB2 (Table 2. 17). Cells were aliquoted into 500 µl and snap frozen in liquid nitrogen before being stored at -80ºC until further use.

Table 2. 17. FB1 and FB2 solutions composition to propagate E. coli DH5α cells.

FB1 (30 ml for every 100 ml culture) FB2 (4 ml for every 100 ml culture)

100 mM RbCl 10 mM MOPS

50 mM MnCl2 10 mM RbCl

30 mM Potassium Acetate 75 mM CaCl2

10 mM CaCl2 15 % Glycerol

15 % Glycerol pH 6.8 with KOH, filtered sterilise

pH 5.8, filter sterilise (No KOH)

2.3.2 Expression of single-domain antibodies and preparation of periplasmic extract

A single colony obtained by the incubation of a VHH bacterial library on ampicillin selective plates was used to inoculate 2 ml of 2xYT medium supplemented with ampicillin and incubated in a rotary shaker at 200 rpm and 30°C overnight.

For small-scale preparations, the 2 ml overnight culture was split into 2 parts. 1.4 ml was centrifuged to test for the presence of the inserts by DNA preparation and 100 μl was used to inoculate fresh 2xYT medium (2 ml).

For large-scale preparation, the pre-culture was added to 50 ml of 2xYT supplemented with ampicillin and incubated at 30°C, 200 rpm to an OD600 of approximately 0.3–0.5 as measured in a cell density meter. If master plates were to be made, one drop of the culture was placed on an LB plate supplemented with ampicillin and incubated at 37°C overnight. IPTG was added to the remaining culture to achieve a final concentration of 1 mM and incubated at 27-30°C and 200 rpm overnight or at 37°C for 4-6 hours.

The cells were pelleted for 20 minutes by centrifugation at 4°C and 3000 ×g and then re- suspended in ¼ of the initial culture volume of pre-chilled TES buffer. The cell suspension was kept on ice for at least 30 minutes before an osmotic shock was given by adding 1.5 volume of cold TES/5; it was incubated on ice for at least another 30 minutes. The mixture was centrifuged for 20 minutes at 13000 xg and 4°C forcing the proteins to leave the periplasmic space and mix with the supernatant. NaCl was added to the supernatant to 300 mM before storage, as this stabilises the proteins.

2.4 Protein techniques

2.4.1 Total protein extract from Xenopus embryos

Embryos were collected at the desired stages, medium carefully removed and embryos frozen at -80ºC (5 - 10 embryos for each protein extraction). Two different solutions were tested when extracting proteins from the embryo, NETS buffer and EEB buffer. No difference was observed in the quality of the proteins extracted, however the latter was preferred for convenience. The embryos were mechanically homogenised by pipetting up and down in 100 µl NETS buffer (Table 2. 18) or 100 µl EEB. Please note, when extracting proteins, the SDS was left out of the NETS buffer, to avoid the solubilisation of yolk proteins. To the homogenised samples, 200 µl of Freon was added, the mixture was vortexed for 20 seconds and centrifuged at 4ºC, 13000 rpm for 15 minutes. The aqueous upper phase was carefully removed, transferred to a fresh tube, mixed the same volume of 2 x SDS loading dye or 2x β-mercaptoethanol sample buffer and heated at 95ºC for 3 minutes. The denatured protein was loaded onto the SDS-PAGE gel or stored at -20ºC. Alternative methods for denaturing proteins were tested, by heating the proteins with the SDS loading dye for 20 minutes at 55ºC, and the combination of 20 minutes at 55ºC followed by 95ºC for 3 minutes. The latest was preferred as it denatured proteins to an extent where they could be visualised as well defined bands on an SDS-

PAGE.

Table 2. 18. NETS Buffer vs Embryo Extraction Buffer (EEB) reagents.

NETS EEB

30 mM NaCl 10 mM HEPES, pH 8.5

1 mM EDTA 1 mM EDTA

20 mM Tris, pH 7.5 2 mM MgCl2

1 % SDS 10 mM β-glycerophosphate

ddH2O 1 mM dithiothreitol (DTT)

EDTA-free protease inhibitor EDTA-free protease inhibitor

Table 2. 19. Protein loading dye composition.

2x SDS sample buffer 2x β-mercaptoethanol sample buffer

0.125 M Tris-HCl, pH 6.8 100 mM Tris-Cl, pH 6.8

4 % SDS 2 % SDS

0.15 M DTT 4 % β-mercaptoethanol

20 % Glycerol 20 % glycerol

0.01 % Bromophenol Blue 0.00125 % Bromophenol Blue

Stored at -20ºC Stored at room temperature

2.4.2 Western blot

2.4.2.1 SDS-PAGE (Sodium Dodecyl Sulphate Polyacrylamide Gel Electrophoresis)

The resolving gel (recipes below) was made by adding first the 30 % acrylamide solution and resolving buffer to the dH2O, followed by the 10 % APS and TEMED, since these agents produce the rapid polymerisation of the gel. The solution was poured in the plates leaving a 1.5 cm space at the top. 1 ml of water-saturated butanol was added to form a layer at the top of the resolving gel and left to stand to allow the full polymerisation of the gel. The butanol was washed off and the gel rinsed thoroughly with dH2O.

The stacking gel was made by mixing the 30 % acrylamide solution and the stacking buffer with the dH2O, before adding the 10 % APS and TEMED. This solution was poured immediately on the top of the resolving gel, and the appropriate comb was inserted. At least 10 minutes were allowed for the gel to polymerise. The gel was then placed in the electrophoresis tank together with 1 x Tris-Glycine SDS running buffer, before removing the combs. A protein marker was loaded into the first lane, followed by the samples. The gel was run at 65 V until the samples had run through the stacking gel, then the voltage was increased to 150 V to 200 V for a further 40 minutes to 1 hour.

When the blue dye reached to bottom of the gel, the gel was removed from the glass plates and transferred to nitrocellulose membrane for western blot or stained with

Coomassie blue.

Table 2. 20. Recipe for 10 ml of polyacrylamide gel mix; separating and stacking gels

5 ml of 6 % 10 ml of 10 % 10 ml of 12 % 10 ml of 14 %

Stacking Gel: Resolving Gel Resolving Gel: Resolving Gel:

dH2O 2.8 ml 4.2 ml 3.5 ml 2.8 ml

30 % acrylamide (37.1 : 1 1 ml 3.3 ml 4 ml 4.7 ml acrylamide : bis)

4x resolving 2.5 ml 2.5 ml 2.5 ml buffer

Stacking buffer 1.2 ml

10 % APS 100 µl 100 µl 100 µl 100 µl

TEMED 10 µl 10 µl 10 µl 10 µl

2.4.2.2 Membrane transfer

The western blot apparatus insert was opened and placed on a tray containing transfer buffer with the black side down. Inside the following items were placed: a fibre pad, 2 x

Whatman 3 mm paper pads, the gel, the membrane, 2 x Whatman 3 mm paper pads covering the membrane and another fibre pad. This “sandwich” was placed on the apparatus with the black side facing the black from the transfer apparatus (black to black), this ensures the correct orientation for transfer. The tank was filled with western transfer buffer and set to run at 30 mA overnight or 300 mA for 2 hours.

Table 2. 21. Western blot tris-glycine membrane transfer buffer.

Membrane transfer buffer

200 mm Tris Base

150 mm Glycine

0.1 % SDS

20 % Methanol

2.4.2.3 Membrane Blocking

Upon transfer, the nitrocellulose membrane was washed with ddH2O for 5 minutes at room temperature and incubated in 25 ml of blocking buffer for 1 hour at room temperature or overnight at 4⁰C. The blocking buffer was then replaced with fresh blocking buffer containing the primary antibody for 1 hour at room temperature or overnight at 4⁰C.

Table 2. 22. Blocking buffer composition and TBSTw recipie

5 % w/v non-fat dry milk (Marvel ®) 5 % w/v non-fat dry milk powder dissolved in TBStw

1 % Blocking Reagent (Roche, 1 % dissolved in maleic acid buffer 11096176001)

5 % BSA (Fisher Scientific, BPE9701) 5 % BSA powder dissolved in TBSTw

TBSTw 20 mM Tris, 150 mM NaCl, 0.1 % Tween 20

2.4.2.4 Primary Antibody Incubations

Two types of primary antibody were used, unconjugated and conjugated to the label.

Detailed procedures are given corresponding to the primary antibody used.

2.4.2.5 For Unconjugated Primary Antibodies

After blocking, the membrane was incubated in fresh blocking solution containing the primary antibody at the appropriate dilution with gentle agitation for 1 - 2 hours at room temperature or overnight at 4°C. For the his-probe (G-18 SC-804 LOT H2304) the dilution used was 1/1000 unless otherwise stated.

Following the primary antibody incubation, the membrane was washed three times for 5 minutes each with 10 – 15 ml of TBSTw and incubated with the species-appropriate

HRP-conjugated secondary antibody. A 1/2000 dilution of the Anti-Rabbit IgG-

Peroxidase conjugated (Sigma A0545) secondary antibody was incubated with the

membrane in fresh blocking buffer with gentle agitation for 1 hour at room temperature unless otherwise stated.

The membrane was washed twice for 5 minutes each with 15 ml of blocking solution, once for 5 minutes with TBSTw and once with PBS for 5 minutes before proceeding with detection.

2.4.2.6 For HRP-Conjugated Primary Antibodies

After blocking, the membrane was incubated in fresh blocking solution with the primary antibody at the appropriate dilution with gentle agitation for 1 - 2 hours at room temperature or overnight at 4°C.

The membrane was washed twice for 5 minutes each with 15 ml of blocking solution, once for 5 minutes with TBSTw and once with PBS for 5 minutes before proceeding with detection.

2.4.2.7 Protein detection

For protein detection, the membrane was incubated either for 1 minute with 10 ml of ECL solution #1 and 10 ml of ECL solution #2 (Table 2. 23), or for 5 minutes with 7 ml Clarity

Western ECL Substrate Solution A and 7 ml of Clarity Western ECL Substrate Solution

B (Bio-Rad, 170-5061). Visualization was performed using an ImageQuant LAS4000 following the manufacturer’s instructions.

Table 2. 23 ECLI and ECLII buffer composition for chemiluminescence detection of HRP- conjugated antibodies.

ECL solution #1 ECL solution #2

2.5 mM luminol 0.021 % hydrogen peroxide

0.9 mM p-coumaric acid 10 mM tris, pH 8.5

10 mM tris, pH 8.5

2.4.3 In vitro protein expression

Mini IVT Kit (Thermo Scientific, 88885)

Target proteins were cloned into the pT7CFE1-NHis-GST-CHA vector and protein synthesis was performed using a 1-Step Human High-Yield Mini IVT Kit following the manufacturers’ instructions with no modifications to the protocol.

2.4.4 In vivo protein expression

HEK cell expression

2.4.4.1 HEK293T cell maintenance

NOTE: All cell manipulations were carried out in a Class 2 laminar flow hood.

2.4.4.1.1 Media preparation

HEK293T cells were cultured in DMEM supplemented with a variety of other components depending on whether they were only growing or if they were transfected.

Table 2. 24. Complete DMEM

DMEM 500 ml

FCS 50 ml

L-glutamine 5 ml

NEAAs (100x) 5 ml

Table 2. 25. DMEM + 2 mM L-Glutamine

L-glutamine 1 ml

DMEM 99 ml

Table 2. 26. DMEM + 1 mM L-Glutamine + 2 % FCS

L-glutamine 1 ml

DMEM 195 ml

FCS 4 ml

Table 2. 27. Freezing Media

FCS 90 %

DMSO 10 %

2.4.4.1.2 Passaging to maintain and propagate cells

HEK293T cells were maintained by passaging them every 3-5 days, upon reaching 80-

90 % confluency. These cells were cultured in a variety of flasks, mainly T75 and T175.

Once the cells reached 80-90 % confluency, the old medium was removed and sufficient trypsin solution (TrypLE Express, Gibco® or equivalent) was added to cover the surface.

The flask was placed in the 37ºC incubator for 3-5 minutes until cells had dissociated.

DMEM medium supplemented with 10 % FCS was added as three times the volume of trypsin used. The added media inactivated the trypsin and protected the cells from lysing.

Cells were transferred into Falcon tubes and centrifuged at 1000 xg for 3 minutes. The supernatant was discarded and the cells re-suspended in fresh medium supplemented with 10 % FCS. Cells were counted and flasks seeded following manufacturer’s guidelines before incubating at 37°C in a 5 % CO2 and 95 % air environment.

2.4.4.1.3 Freezing cells

Cells were treated as if they were going to be passaged, with the difference that when they were re-suspended after centrifugation, freezing medium (Table 2.25) was used instead. The cells were aliquoted into 2 ml cryovials and stored at -80ºC for 1-2 days

(maximum 2 weeks). These were then transferred to liquid nitrogen storage and tested after 2-3 days on liquid nitrogen for freezing efficiency.

2.4.4.1.4 Thawing cells

DMEM with 10 % FCS (25 ml) was prepared in a T75 flask before rapidly thawing a vial of cells in a water bath at 37°C. Upon thawing, the full content of the vial was added to the T75 flask and incubated at 37°C in a 5 % CO2 and 95 % air environment until confluent (2-4 days). Cells were passaged as previously described.

2.4.4.2 Small-scale HEK293T cells transient transfection

Small-scale HEK293T transfection was achieved using the JetPRIME reagent (Polyplus,

114-01) following the manufacturer’s instructions. In summary, cells were seeded and allowed to reach 60 % to 80 % confluency at the time of transfection. Transfection was performed with a 1:2 DNA to jetPRIME® reagent ratio and in the presence of serum.

The Jet prime buffer was aliquoted into microcentrifuge tubes and the DNA added. The samples were mixed by vortex for 10 seconds and microcentrifuged for 10 seconds at full speed. The JetPrime Reagent was added to the Buffer-DNA mix, vortexed for 10 seconds and centrifuged for 10 seconds. The mixture was then incubated at room temperature for 10 minutes exactly. The transfection mix was added to the cells and incubated for 4 - 8 hours at 37°C in a 5 % CO2 and 95 % air environment. After the incubation period, the medium was changed to fresh DMEM supplemented with 2 % L- glutamine and 10 % FCS and further incubated for 24 - 48 hours.

Cells and medium were subsequently collected and stored separately at -80°C until analysed.

2.4.4.3 Medium-scale HEK293T cells transient transfection

Cells were seeded at 7.5 x 105 cells/ml in 5 ml of complete DMEM and incubated overnight at 37°C in a 5 % CO2 and 95 % air atmosphere. Once they were 80 % confluent, 87.5 μg of plasmid DNA was mixed with 2.6 ml of DMEM supplemented with

NEAAs (non-essential amino acids) and 1 mM L-glutamine. Separately, 154 μl of 1 mg/ml branched polyethylenimine (PEI) was mixed with 2.6 ml of DMEM containing 1 mM L- glutamine. The DNA-containing DMEM and the PEI-containing DMEM were mixed and incubated for 10 minutes at room temperature. The medium was removed from the T175 flask and replaced with 45 ml of fresh DMEM containing 2 % FCS, 1x NEAAs and 1 mM

L-glutamine. After 10 minutes of incubation, the DNA/PEI mixture was added to the cells and incubated at 37°C in a 5 % CO2 and 95 % air atmosphere for 3 days.

Cells and medium were subsequently collected and stored separately at -80°C until analysed.

2.4.4.4 Large-scale HEK293T cells transient transfection

HEK293T cells were transfected as previously described by Chamberlain and co- workers (Nettleship et al., 2015). In summary, four confluent T175 flasks of cells were used to inoculate four expanded surface roller bottles (Greiner Bio-One) containing 250 ml of medium each. These were incubated at 37°C in a 5 % CO2 and 95 % air environment, rotated at 0.3 rpm for the first 24 hours and at 1.0 rpm thereafter. The cells were cultured for 5-6 days until they were approximately 70 % confluent. On the day of transfection, two 500 ml DMEM bottles were supplemented with L-glutamine to 2 mM.

100 ml from each bottle were removed and placed in a T75 flask; namely Flask1 and

Flask2. To the remaining 400 ml of supplemented medium 8 ml of heat-inactivated FCS were added and left aside. To Flask1, 3.5 ml of 1 mg/ml polyethyleneimine (PEI), pH 7.0 was added. To Flask2, 2 mg of DNA was added. Flask1 and Flask2 were mixed for exactly 10 minutes. Meanwhile, the old medium from the roller bottles was taken out and replaced with 200 ml of the previously supplemented DMEM. After the 10 minutes incubation, 50 ml of the DNA/PEI mix was added to each bottle and incubated at 37°C in a 5 % CO2 and 95 % air environment rotated at 1.0 rpm for one week (Table 2. 2).

Cells and medium were subsequently collected and stored separately at -80°C until analysed.

Figure 2. 2. Large-scale transient transfection of HEK293T cells. Two bottles of DMEM were supplemented with 2 mM L-glutamine before transferring 100 ml of each bottle to a different flask. One of the remaining 400 ml DMEM bottle was supplemented with 2 % heat-inactivated FCS. To one of the flasks containing 100 ml DMEM, 1 mg/ml polyethyleneimine (PEI), pH 7.0 was added, and to the other 2 mg of purified DNA were added. Flask1 and Flask2 were mixed for exactly 10 minutes. Meanwhile, the old medium from the roller bottles was discarded and replaced with 200 ml of the previously supplemented DMEM. After the 10 minutes incubation, 50 ml of the DNA/PEI mix was added to each bottle and incubated at 37°C in a 5 % CO2 and 95 % air environment rotating at 1.0 rpm for one week. Cells and medium were subsequently collected and stored separately at -80°C until analysed.

2.4.4.5 Harvesting expressed proteins

2.4.4.5.1 Secreted proteins

The supernatant containing the protein was collected, centrifuged at 6000 xg for 15 minutes to remove any detached cells and filtered through a 0.22 μm bottle top filter before being stored at 4°C.

2.4.4.5.2 Intracellular proteins

The supernatant was removed, centrifuged at 6000 xg for 15 minutes and discarded.

The pelleted cells and T175 flask were frozen at -80°C. To lyse the cells and isolate the intracellular proteins the M-PER™ Mammalian Protein Extraction Reagent (Product No.

7850) was used supplemented with Halt™ Protease Inhibitor Cocktail Kit (Product No.

78410) according to the manufacturers’ instructions. In summary, the culture medium was removed from the adherent cells, and the cells washed with PBS. The appropriate amount of M-PER® reagent (Table 2. 28) was added and shaken gently for 5 minutes while on ice. The lysate was collected, transferred to a microcentrifuge tube and centrifuged at 14000 xg for 5-10 minutes to pellet the cell debris. The supernatant, containing the extracted protein was transferred to a new tube for further analysis.

Table 2. 28. M-PER® reagent recommended quantities per sample volume.

Plate Size/Surface Area Volume of M-PER® Reagent

100 mm2 500-1,000 µl

60 mm2 250-500 µl

6-well plate 200-400 µl per well

24-well plate 100-200 µl per well

96-well plate 50-100 µl per well

2.4.4.6 Sonication and solubilisation of HEK cell-expressed proteins

2.4.4.6.1 Cells lysis by sonication

The previously frozen cell pellets were thawed on ice and re-suspended in 20 ml sonication/lysis buffer (below). Sonication was allowed to happen on ice with the Sonic

Dismembrator, Ultrasonic Processor from Fisher Scientific. The total sonication time was

1 minute 30 seconds and a temperature cut off of 15ºC was used. Cycles of 5 seconds pulse and 9.9 seconds rest with 40 % intensity were applied until the sample appeared homogeneous.

2.4.4.6.2 Solubilisation

Deoxycholic acid was added to 1 % (w/v) to the crude lysate and stirred for 1 hour at room temperature to break down the fat contents of the sample. DNaseI was also added

(200 µl of 28.4 U/ml) with additional stirring for 1 hour at room temperature or 4 hours at

4ºC to eliminate the DNA content of the sample.

Cellular debris was removed by centrifugation at 14000 xg for 20 minutes at 4ºC; the pellet was resuspended in 20 ml of sonication/lysis buffer and further stirred for 30 minutes at room temperature. The lysed cells, were centrifuged at 14000 xg for 20 minutes at 4ºC and the pellet was resuspended in 20 ml of solubilisation buffer, stirred overnight at 4ºC, centrifuged at 14000 xg for 20 minutes at 4ºC and the supernatant retained; this contained the solubilised protein.

Table 2. 29. Sonication/lysis buffer and solubilisation buffer recipes. Filtration with a 0.45 µm filtered is required before use.

Sonication/lysis buffer Solubilisation Buffer

50 mM Tris-HCl, pH 8.0 8 M urea

1 M NaCl 130 mM NaCl

1 mM EDTA 20 mM Tris, pH 8.6

2.4.5 Protein purification

2.4.5.1 Affinity Purification using his-Spin Trap

The solutions used for this purification were obtained from GE Healthcare (#11003400).

Upon the collection and lysis of the expressing cells, 1.8 ml of supernatant containing the expressed proteins was collected. The samples were centrifuged for 10 minutes at

13000 xg to remove any cells that had been carried through. Samples were loaded into the affinity columns 600 µl at the time and purified according to the manufacturer's instructions. Two elution steps were carried out for each sample and stored at 4ºC together with the used columns.

2.4.5.2 Affinity Purification using AKTA start

All solutions that were used on the AKTA Start were filtered with a 0.22 µm filter, as this reduces the risk of tubes clogging. The AKTA Start was set on a “Manual Run” and flushed through with water at 5 ml/min for 2 minutes. All lines were flushed with water.

The column (1 ml hisTrap FF column (GE Healthcare Life Sciences, UK)) was connected, making sure no air bubbles were introduced to the system. To do this, first, the top of the column was connected and then the bottom, by clicking off and connecting to the system. The whole system was flushed with solution A at no more than 3 ml/min, then the “elution line” was flushed with solution B. The whole system was flushed a second time with solution A, avoiding the “elution line”. This step primes the system, reducing the risk of contamination.

The filtered sample was loaded, allowed to flow through the column to bind and washed with at least 5 volumes of solution A to elute the non-binding proteins. A gradient of 0 –

100 % solution B over 10 minutes was set, as the optimal imidazole concentration for elution of these proteins was unknown. The fractions collected were 1 ml each, and a

peak on the UV absorbance reflected the successful elution of the protein. The flow- through was also kept to be analysed.

Once the protein was eluted, the column was washed with solution A for 5 minutes and switched to solution B for a further 3 minutes to make sure everything was eluted. The

AKTA Start and the column were cleaned thoroughly before storing by flushing filtered

20 % ethanol through for 10 minutes. The AKTA Start was stored filled with 20 % ethanol.

Table 2. 30. Binding buffer (solution A) and elution buffer (solution B) used with the Akta Start for protein purification. Solutions were filter sterilised using a 0.22 µm filter and stored at room temperature.

Solution A Solution B

50 mM Tris-HCL, pH 7.5 50 mM Tris-HCL, pH 7.5

500 mM NaCl 500 mM NaCl

20 mM Imidazole 500 mM Imidazole

2.5 Antibody purification techniques

Antibody purification was carried out in three steps. First, the total IgGs in the serum were isolated by caprylic acid extraction. Secondly, specific IgGs were isolated by affinity purification. Lastly, the affinity-purified antibodies were desalted on a G25 column.

2.5.1 Caprylic acid extraction

Antisera from Sheep that had undergone an immunisation series of Kash5 protein was used as starting material. Two volumes of 60 mM sodium acetate, pH 4.0 (20 ml) were mixed with 10 ml of antisera. While stirring, 750 µl of 100 % caprylic acid was added

(final concentration 2.5 %). The mixture was left stirring for 20-30 minutes (until it turned white) and centrifuged at 5100 xg for 15-20 minutes. The supernatant was transferred into a fresh tube and the total IgG present in the serum assessed by UV spectroscopy.

2.5.2 Dialysis

For 10 ml starting material, 20-25 cm of pre-boiled 18/32 inch diameter visking tubing

(stored in 70 % ethanol) was rinsed well with distilled water. One end of the tubing was tied off and the cleared supernatant pipetted through the other end, before securing the top. The tubing containing the IgG was placed into a 1 L 1 x PBS and stirred gently overnight at 4°C.

The following day, the IgG concentration was evaluated using UV spectroscopy, and the sample was stored at - 20°C or taken for affinity purification.

2.5.3 Affinity column purification

2.5.3.1 Column preparation

The columns used contained peptides coupled to glass beads and were supplied with the synthesised peptide. Before first time use, the column was equilibrated with 1 x PBS and washed with 5 x bead volumes of guanidinium chloride (6 M), HCl (10 mM), ethanol

(100 %) and 1 x PBS. This intensive cleaning procedure was also used if the column started to cause problems, such as failure of the IgG to bind to the column or overloading of the column, which produce an unusually large flow-through UV reading. Columns were stored at 4°C in PBS-azide (0.02 % sodium azide / 1 x PBS) and capped at both ends to avoid drying and inhibit bacterial growth.

2.5.3.2 Antibody purification

Either after preparing the column or after storage, the column was equilibrated by washing with 3 x column volumes of 1 x PBS. The IgG was allowed to flow down the column three times, and the column rinsed with 10 bead volumes of 1 x PBS, 0.5 M NaCl in 1 x PBS and 1 x PBS again.

2.5.3.3 Antibody elution

Potassium thiocyanate (3.5 M) was used to elute the bound IgG by adding 1 ml at the time to the column, directly into the beads, for a total of 6 ml. The eluted antibody, had to be immediately de-salted by passing it through a G-25 column.

2.5.3.4 Cleaning the affinity column

Following the antibody elution, the affinity column was washed with 10 x bead volumes of HCl (10 mM), 1 x PBS and stored at 4°C in 0.02 % PBS-azide

2.5.4 G-25 desalting column:

The 1.5 x 75 cm G-25 column was stored in 0.02 % PBS-azide and had to be equilibrated by allowing at least 2 column volumes of 10 mM sodium bicarbonate to flow through before use. The eluted antibody was loaded onto the column at a flow rate of 2 ml/min, and washed with 10 mM sodium bicarbonate until a peak in UV absorbance was visible, which corresponds to the antibody being eluted. The column separates the IgG from the salt, allowing the IgG to come out first and the salt after. 8 ml fractions were collected and UV spectroscopy was used to measure the IgG concentration. Fractions that had a considerable amount of antibody were pooled and aliquoted at 100 mg per aliquot. These were snap frozen in liquid nitrogen and lyophilised overnight. The lyophilised antibody was stored at -20ºC until further use.

The G-25 column was left to run the salt off, and rinsed with 5 bead volumes of 10 mM sodium bicarbonate before flushing with two bead volumes of 0.02 % PBS-azide and stored in this solution.

2.6 Embryological techniques

2.6.1 Embryo micro-injections

Fertilized Xenopus laevis embryos were obtained from the EXRC and allowed to turn.

The animal pole migrating to the top indicates the fertilisation of the eggs. Embryos were de-jellied in 2 % cysteine in 0.1 x MBS at pH 8.0. Once de-jellied, the embryos were washed thoroughly 6 times with 1 x MBS.

Embryos were transferred into 3 % ficoll in 1 x MBS and aligned with the animal pole facing up on a grid. The liquid was partially removed to avoid the embryos moving.

Injected embryos were left in ficoll for 3-4 hours after injection and transferred to 0.1 x

MBS before gastrulation.

Injections were usually of 4 nl and a maximum of 400 – 600 ng of nucleic acid (RNA or

DNA) per embryo. Amounts were altered by altering the injection time.

For gene editing, the embryos were injected as soon as possible after fertilisation to maximise chances of cutting the DNA before replication starts. The more intact DNA molecules, the more mosaic the embryo will be.

Table 2. 31. Barth-HEPES Saline (10 X) MBS (Modified Barth’s Saline).

MBS

88 mM NaCl

1 mM KCl

2.4 mM NaHCO3

10 mM HEPES

0.82 mM MgSO4 x 7H2O

0.33 mM Ca(NO3)2 x 4H2O

0.41 mM CaCl2 x 2H2O

5 µg/ml Penicillin

5 µg/ml Streptomycin

2.7 VHH libraries

2.7.1 General considerations

The general process of VHH library production involves the generation of antigen protein, its purification, and immunisation of a camelid (in our case Llama glama), followed by the VHH library generation and screening. It was only in 1998 that it was established that dromedaries and llamas can be easily immunised (Lauwereys, et al., 1998), and that it is feasible to vaccinate an animal simultaneously with a mixture of several antigens to raise an immune response to each antigen, making the whole procedure more cost- effective and ethically preferable (Lauwereys et al., 1998; Van Der Linden et al., 2000).

There have been many studies to optimise the VHH library generation process over the years, some key points are mentioned bellow.

It was established that 20-30 ml of blood (Muyldermans, 2001) of an immunised llama contains sufficient VHH expressing B cells to clone a diverse set of the affinity-matured nanobodies with high recognition specificity for their cognate antigen; others suggest that a 100 ml sample is needed (Pardon, et al., 2014). Also, it has been found that llamas can be re-used for different VHH discovery projects by respecting a grace period of at least 6 months (Pardon, et al., 2014). According to Muyldermans (Muyldermans, 2001), the whole process from llama immunisation to the cloning, selection, expression and identification of recombinant VHH antibodies is routinely performed in less than 3 months. Muyldermand also stresses the advantages of working with libraries from llamas immunised with one or two immunogens, as the complexity of the library and the ability to screen those correlates to the number of immunogens used.

The VHH libraries used in this study were a kind gift from Sergei Sokol’s Lab (Mount

Sinai Medical Centre, New York, USA), where two llamas were immunised; one with a protein extract from gastrula-neurula stage embryos and the other with a combination of

36 proteins obtained from 12 laboratories worldwide. I was involved in the

characterisation of the gastrula-neurula stage embryo protein library, and the making of the 36 protein library.

2.7.2 Llama PBMC prep on 400 ml blood samples

Leucosep tubes (Greiner Bio-one, 227290) were prepared by adding 15 ml of mammalian lympholyte (Tebu-bio, CL5120) per tube and centrifuged for 1 minute at 1000 xg. The blood was mixed with an equal volume of sterile PBS, and 30 ml was gently layered onto each leucosep tube using a 25 ml graduated pipette. The tubes were centrifuged at 1500 xg for 30 minutes with the slowest acceleration and no brake applied for the slow-down. The buffy layer was carefully removed into previously aliquoted 25 ml of PBS per 100 ml whole blood equivalent. The buffy layer/PBS mix was centrifuged at

1500 xg to pellet cells for 20 minutes, supernatant was discarded and PBS added to re- suspend and pool into a single tube. 10 ml of PBS were used in total. Total cell count was taken by mixing 10 µl of the cells with 90 µl of 0.4 % trypan blue and counting cells with a haemocytometer.

The cells were pelleted by centrifugation at 1500 xg for 20 minutes, and re-suspended in 10 % freeze medium (FCS + 10 % DMSO) at 5 x 107 cells per ml. Lymphocytes were stored at -80ºC until further use.

2.7.3 Library Construction

To construct the VHH library, previously described methods were followed (Ladenson,

Crimmins, Landt, & Ladenson, 2006; Ryckaert, Pardon, Steyaert, & Callewaert, 2010;

Sherwood & Hayhurst, 2012). In summary, lymphocytes were isolated from the blood and the RNA extracted before synthesising the cDNA to use as template on the nested

PCR reactions that amplify the VHH. From the cDNA library, reactions were tested on a small scale to optimise conditions, then up scaled to generate the bacterial library. From the bacterial library, the phage library was generated and panned against target proteins/antigens of interest. Once these VHH were isolated, they were verified and some characterised.

2.7.4 Primer design

To generate a VHH library, the VHH specific region was amplified by nested PCR.

Several different methods of VHH amplification have been published, mainly based on a two-step nested PCR of the VHH cDNA sequence from B lymphocytes. The first set of primers amplifies the region of the second constant heavy-chain domain (CH2) that is conserved among all IgG isotypes of camelids; they anneal to the template strand of leader-signal sequences of camelid VH and VHH genes and to the coding strand of CH2- sequence of camelid IgG (for example primers CALL001 and CALL002); whereas the second pair of primers anneals in a well-conserved region of the leader signal sequence of all V elements of family III, by far the most abundant V family in camelids (such as primers SM017 and SM018) (Pardon, et al., 2014). A list of the most commonly used primers to isolate the VHH specific coding region from llama serum is shown in Table 2.

32.

Table 2. 32. Most commonly used primers to isolate the VHH specific coding region from llama serum.

Name given Sequence Paper CALL001 5’ GTC CTG GCT GCT CTT CTA CAA GG 3’ CALL002 5’ GGT ACG TGC TGT TGA ACT GTT CC 3’ 5’ CCA GCC GGC CAT GGC TCA GGT GCA GCT SM017 GGT GGA GTC TGG 3’ 5’ CCA GCC GGC CAT GGC TGA TGT GCA GCT SM018 (Rothbauer et al., 2006) GGT GGA GTC TGG 3’ 5’ CAT GCC ATG ACT CGC GGC CAC GCC GGC A4short CAT GGC 3’ Reverse 5’ GGA CTA GTG CGG CCG CTG GAG ACG GTG Primer 38 ACC TGG GT 3’ 5’ GAT GTG CAG CTG CAG GAG TCT GGR GGA VHH-Back: (Pardon, et al., 2014) GG 3’ 5’ CTA GTG CGG CCG CTG GAG ACG GTG ACC VHH-For: TGG GT 3’ (R. Zhang & Shen, 2012) 5’ CTA GTG CGG CCG CTG AGG AGA CGG TGA PMCF CCT GGG T 3’ MP57 5’ TTA TGC TTC CGG CTC GTA TG 3’ (Hassanzadeh-ghassabeh, Saerens, & Muyldermans, GIII 5’ CCA CAG ACA GCC CTC ATA G 3’ 2011) 5’ TCG CGG CCC AGC CGG CCA TGG CGC AGG SHFmu TSM ARC TGC AGG AGT CWG G 3’

5’ ATG GTG ATG ATG ATT GTG CGG CCG CGC SHRmu TGG GGT CTT CGC TGT GGT GCG 3’ (Unger et al., 2015)

5’ TCG CGG CCC AGC CGG CCA TGG CCG ATG LHFmu TGC AGC TGC AGG MGT CWG GRG GAGG 3’ 5’ ATG GTG ATG ATG ATG TGC GGC CGC TGG LHRmu TTG TGG TTT TGG TGT CTT GGG 3’ H1R 5’ GTC CTG GCT GCT CTT CTA CAA GG 3’ CH2D 5’ GGT ACG TGC TGT TGA ACT GTT CC 3’ 5’ GCC ATG GCT SAK GTB CAG CTG STG GAG H2s TCT GG 3’ H3d 5’ GCG GCT GAG GAG ACG GTG ACC WGG 3’ (Yang et al., 2014) 5’ ATT ACT GGC CCA GCC GGC CGG CTG AKG VH2SfiI TBC AGC TGG TGG AGT CTG GRG GAG GCT 3’ 5’ ATT ATT AGT GCG GCC GCT GAG GAG ACG VH3NotI GTG ACC WGG GTC C 3’

Figure 2. 3. Nested primers to isolate the heavy chain only variable region from llama’s total cDNA by PCR. The 1st set of primers (shown in black) isolates the VH specific region, followed by the 2nd set of primers (shown in blue), which targets the VHH (VHH) specific region. Adapted from (Tanha, Dubuc, Hirama, Narang, & MacKenzie, 2002).

2.8 Bacterial library construction

The cDNA made by RT of RNA isolated from lymphocytes, was subjected to nested PCR to amplify the VHH specific region. Upon the first amplification, the amplicons were separated by agarose gel electrophoresis and the correct size bands (heavy-chain only

600 bp and conventional 800 bp) gel extracted with a Qiagen gel extraction kit.

Once extracted, the purified amplicons were subjected to the second PCR and checked by gel electrophoresis. Pecan126, the vector of choice to construct the VHH library, was digested with NotI and SfiI and also checked by gel electrophoresis.

Digested vector and PCR product of large-scale, second nested PCR were gel extracted and purified using standard methods. Ligations were carried out using T4 DNA Ligase at

22°C for 20 hours. About 6 µg vector and 1.5 µg of the insert were required for a library of 2 x 109 colony forming units (cfu) using current methods. The ligated material was ethanol precipitated, and the pellets were dissolved in 60 μl H2O. A total of 5 μg of ligation material were used for transformation.

Electroporation was by a Gene Pulser (Bio-Rad, Bechhofer lab): 60 μl of ligated material and 270 μl of electrocompetent HVB88 cells were mixed, placed into 2 mm-gap cuvettes

(BioRad or Bulldog Bio) and received a 2.5 kV pulse for 5.6 to 5.8 ms. The samples were added to 2 ml of warm (37°C) SOB + 2 % glucose in a round bottom 15 ml tube and incubated on a platform shaker at 37°C for 1 hour. The cultures were centrifuged at

3000 rpm for 10 minutes. The pellets were re-suspended in 500 μl SOB, 2 % glucose and each suspension was plated onto two 15 cm 2xYT, 2 % glucose plates

(supplemented with 50 µg/ml ampicillin and 35 µg/ml chloramphenicol). The plates were incubated at 37°C overnight, with resulting colony counts in the 1000000s.

All colonies were collected from 74 15 cm plates with a scraper by adding 4 ml of 2xYT,

2 % glucose medium. A total of 180 ml of bacterial suspension was collected in four 50 ml conical tubes. The suspension was centrifuged at 3500 xg for 15 minutes to reduce

the volume to 70 ml after discarding part of the supernatant. An equal volume (70 ml) of ice-cold 30 % glycerol in terrific broth was added to the bacterial suspension and mixed well. 108 aliquots of 1.3 ml each were frozen on dry ice and stored at -80°C until needed.

2.8.1 Phage library construction

Phagemid library was made according to Sherwood and Hayhurst, 2012. 430 μl of the thawed bacterial library stock (see above) was placed into 500 ml 2xYT, 2 % glucose, supplemented with ampicillin and chloramphenicol with OD600 at 0.05, and cultured until

OD600 of 0.44 (approx. 2 hours at 37°C). The helper phage M13K07 was added at 20

MOI (multiplicity of infection) and incubated at 37°C for 30 minutes. The solution was shaken at 200 rpm for 30 minutes before the addition of arabinose (2 mg/ml final concentration), IPTG (10 μM final concentration) and Kanamycin (70 μg/ml final concentration). The culture was incubated at 30°C for 22 hours, transferred into 16 round bottom 30 ml tubes (for high-speed centrifugation) and centrifuged at 7650 xg for 1 hour at 4°C. 500 ml of supernatant was collected and 100 ml 20 % PEG8000, 2.5 M NaCl was added and stirred at 4°C overnight. 500 ml of the phage preparation was centrifuged in twelve 50 ml conical tubes at 4080 xg for 1 hour at 4°C. The pellets were re-suspended in 3.3 ml of PBS, 3.3 ml of glycerol were added and mixed well. Phage aliquots of 500

µl each were made in 13 tubes and stored at -80°C until further use.

2.8.2 VHH selection of affinity binders by phage display

Adsorption immunoassay plates (96 wells in this case) were coated with 200 μl of 10

μg/ml antigen in 1 x PBS overnight at 4ºC in a humidified container. The following morning, the lysate from the plates was collected and stored at 4°C for future use. The empty microtiter wells were washed by immersing the plate in a large container of wash buffer (1 x PBS with 0.05 % Tween™20) and agitated to release any bubbles in the wells.

The plate was removed from the container, inverted and the contents flicked out by tapping into a paper towel. This process was repeated 3 times. The wells were blocked by adding 200 μl of 1 x blocking buffer (2 % non-fat dry milk in 1 x PBS), and incubated for 1–2 hours at room temperature. In a separate, pre-blocked microcentrifuge tube, the recombinant phage antibody supernatant was diluted with an equal volume of 1 x blocking buffer and incubated for 15–30 minutes at room temperature. The blocked wells were washed three times with PBS as previously described. 200 μl of the diluted recombinant phage antibody supernatant was added to each antigen-coated well; the plate was sealed and incubated for 1 hour with shaking at room temperature and 30 minutes static at room temperature. The recombinant phage antibody supernatant was removed and stored at 4ºC for future use. The plate was washed to remove unbound phage. In the first panning, 10 washes with PBS were followed by 10 washes with

PBSTw. For the second and successive pannings 20 washes with PBS and 20 washes with PBSTw were used.

To elute the bound phage, 100 μl of 100 μM triethylamine was added to each well, the plate was covered with a sealer and incubated for 10 minutes at room temperature.

Meanwhile, 1 M Tris-HCl, pH 7.5 was aliquoted out at 250 μl per centrifuge tube. Once unbound, 5 wells of phage were collected into one tube containing Tris to neutralise the eluate and stored at 4°C until further use.

To infect the bacteria with the phage and propagate the phage, 100 μl of HBV88 cells were used to inoculate 50 ml 2xYT and incubated with shaking at 37°C for 2-3 hours,

until the OD600 of 0.4 - 0.8 was reached. These exponential phase HBV88 cells were aliquoted into 10 ml, 375 µl of the neutralised eluate were added to each 10 ml and mixed

3 times upside down before being incubated statically at 37°C for 30 minutes.

Following 30 minutes of infection, 100 μl of the cell suspension was removed for titration

(-1), and the rest pelleted by centrifugation at 3000 rpm for 10 minutes. Cells were re- suspended, spread onto 2xYT medium plates (2xYT, 2 % glucose, 50 µg/ml ampicillin and 35 µg/ml chloramphenicol) and incubated overnight at 37ºC.

The following day, the plates were scraped with 3 ml 2xYT into Falcon tubes. 200 μl of this solution was diluted into 5 ml of 2xYT, OD600 measured, and sufficient culture was added to 40 ml pre-warmed 2xYT supplemented with 2 % glucose, ampicillin and tetracycline to OD600 of 0.1. The solution was stirred for 1 hour at 37°C, and helper phage was added to an moi of 20. For the helper phage to bind, the solution was stirred gently and left static at room temperature for 30 minutes. IPTG to final 10 μM, arabinose to final

2 mg/ml and kanamycin to final 50 μg/ml were added and the solution stirred at 30ºC for at least 8 hours.

Panning was normally done for 3-4 rounds with mixed populations of VHH expressing clones. If successful enrichment was noted, the colonies from the subsequent, purified phage were used for single VHH isolation. Therefore, during this incubation, the VHH sequence was passed from the bacteria to the phage. Either the culture was centrifuged and the supernatant tested by ELISA or it was mixed with 100 µL blocking buffer and repeat processes above.

Essentially the same process was followed to purify nanobodies recognising embryo cytoskeleton except that instead of target protein being immobilised on a plate, the insoluble cytoskeleton was used as an “affinity bead” and the bound nanobodies were collected by centrifugation between washing steps.

The presence of VHH encoding sequences in the isolated phage was confirmed by digestion of isolated DNA with NdeI, which released the VHH encoding sequence.

Subsequently, the VHH itself was detected by western blotting against the his-tag encoded with the VHH using standard techniques (Robinson and Guille, 1999). To test whether the nanobodies raised against cytoskeleton recognised a single protein in embryos these were tested in a western blot against 20 µl of embryo extract using the same technique.

2.9 Suppliers

All consumables employed in this project were obtained from the following suppliers:

Abcam (http://www.abcam.com/)

Addgene (https://www.addgene.org/)

Alpha Laboratories (https://www.alphalabs.co.uk/)

Bio-Rad (http://www.bio-rad.com/)

Fisher Scientific (https://www.fishersci.co.uk/)

GE Healthcare (http://www3.gehealthcare.co.uk/)

Gene Tools (http://www.gene-tools.com/)

Grainer Bio-One (https://www.gbo.com/en_GB.html)

NEB (https://www.neb.com/)

Promega (https://www.promega.co.uk/)

Polyplus (https://www.polyplus-transfection.com)

Qiagen (https://www.qiagen.com/gb/)

Roche (https://www.roche.co.uk/)

Sigma Aldrich (https://www.sigmaaldrich.com/united-kingdom.html)

Takara (www.takarabio.com)

Thermo Fisher (http://www.thermofisher.com/uk/en/home.html)

VWR (https://uk.vwr.com/store/)

3 Creation and characterization of complex libraries

3.1 Introduction

Technological advances and the acceleration of research output over the past decades, has enhanced the need for species-specific reagents. Techniques that were established, such as the use of mRNA assays to study gene expression and protein function, were called into question by the results from new generation experiments, such as the revelation of significant discrepancies between the mRNA and protein levels in the early

Xenopus embryo (Peshkin et al., 2015). Some laboratories attempted, and some succeeded, in raising antibodies against Xenopus-specific proteins. For example, Prof

Elizabeth Jones from the University of Warwick generated a variety of antibodies against kidney, nervous system and epidermal proteins (Kyuno, Massé, & Jones, 2008; Naylor et al., 2009). Professor James Maller from the University of Colorado raised antibodies against proteins involved in the regulation of the cell cycle (Eckerdt, Yamamoto,

Lewellyn, & Maller, 2011). Prof Hugh Woodland, also from the University of Warwick, produced antibodies to aid the study of how the primordial germ cells come about and how they differentiate from cells that will become other types of cells, such as skin, muscle, nervous system and so on (Nijjar & Woodland, 2013).

In 2003, an antibody was raised in sheep against recombinant xNAP1L (Abu-daya et al.,

2005; Steer et al., 2003). The xNAP1L antibody became popular due to its cross- reactivity with the human version of the gene and aided studies of Emery-Dreifuss muscular dystrophy disease (EDMD1). This meant that what appeared to be a reasonable amount of the antibody obtained by immunizing a sheep, was quickly used up (Abu-Daya et al., 2005; Faunes et al., 2009; Friedeberg et al., 2006; Glanz, Bunse,

Wimbert, Balczun, & Kück, 2006; Gong et al., 2014; Malartre, Short, & Sharpe, 2006;

Melcon et al., 2006; Park & Luger, 2006; Puffenberger et al., 2004; Shintomi et al., 2005;

Vardabasso, Manganaro, Lusic, Marcello, & Giacca, 2008; Zlatanova, Seebart, &

Tomschik, 2007). Most of these antibodies have been raised either in sheep or rabbit, producing a limited quantity and once that stock is finished the antibody has to be raised again and entirely re-characterised. Only a small quantity, mainly from E. Jones’s collection, have been generated as hybridomas, which allows production of a specific antibody almost indefinitely.

One of the remarkable advantages that llama single-chain antibodies have over conventional antibodies, apart from the ease of storage, their size and stability, is that all their coding region is encoded in a single exon, and therefore can be cloned into a vector and stored as a conventional plasmid indefinitely. These single-chain antibodies sequences are isolated from lymphocytes and amplified, before being cloned into a phagemid vector for phage display. This technique, first described in 1985 (Smith, 1985), allows a connection of the protein to the genetic information that encodes them. Bacteria carrying the phagemid vector with the sequence of interest are infected with helper phage, this allows the phage to be produced that express the gene inserted in the phagemid. The phage then displays the protein on its coat, making the protein accessible, these can be selected by binding assays. The selected phage, in turn can re-infect fresh bacteria and transfer back the genetic material. The discovery of this selection process allowed the generation of large protein libraries that could be screened and amplified, a process called in vitro selection. There have been studies that report the successful isolation of multiple VHHs from libraries generated by immunizing llamas with multiple antigens (Conway, Sherwood, Collazo, Garza, & Hayhurst, 2010; Harmsen et al., 2013). Therefore, the potential to raise multiple single-chain antibodies in a single llama was something the Xenopus community became very interested in.

The need for a reliable and sustainable source of Xenopus-specific antibodies encouraged the scientific community that employs Xenopus as a model organism to join efforts. Twelve leading laboratories provided 36 purified proteins to immunise a Llama

glama in a project supervised by Dr Sokol at the Mount Sinai Hospital, USA. A second

Llama glama was immunised at the same time with a pool of proteins extracted from

Xenopus laevis embryos at the gastrula-neurula stage. The basics of this project were to test the feasibility of isolating VHHs from llamas immunised with multiple antigens.

The opportunity arose to collaborate with the Sokol laboratory. Upon the immunisation of the llamas, a library with the isolated VHH fragments for the gastrula-neurula stage embryos proteins was constructed, and we collaborated in the screening of this library.

The blood from the llama immunised with the 36 purified proteins was collected, and we collaborated with the construction and further screening of this library.

3.2 Results

3.2.1 Bacterial library construction

Two llamas had previously been immunised, one with a cocktail of 36 purified proteins and the other with embryo lysate from mixed gastrula-neurula stage embryos (see 2.7.2).

The llama’s blood was collected, lymphocytes were isolated and cDNA generated. The cDNA was used as a template to isolate the VHH specific region by nested PCR (see

2.7.3). Appropriate amplicons were gel purified and checked by a second gel electrophoresis (see 2.7.2). The vector used to construct the library (pecan126) was digested with NcoI and SfiI, gel extracted and purified (MinElute Gel Extraction Kit,

Qiagen).

The digested pecan126 vector and amplicons from the nested PCR had their purity and concentration confirmed via gel electrophoresis before ligation (Figure 3.1 shows the library components for the 36 protein immunization). Pecan126 digestion #2 and a mixture of the PCR products were ligated to generate the VHH library (see 2.1.1.1).

Ligation and electroporation into HBV88 cells yielded a library with an estimated total of

108 colonies and with a transformation efficiency of 2x106/μg for the ligated material. All colonies were collected from 74 15 cm plates with a scraper by adding glucose medium to each plate.

Figure 3. 1. Purity and concentration confirmation via gel electrophoresis of library elements before ligation. In order to prepare a phagemid library from the lymphocytes obtained from the llama immunized with 36 proteins, nested PCR was used to amplify the VHH encoding DNA. The vector pecan126 was double digested with NcoI and SfiI and recovered in higher concentration in reaction # 2, thus this was used further. The large-scale nested PCR products were pooled, before ligating into the pecan126 vector.

3.2.2 Isolation and characterisation of individual VHHs

VHH libraries can be screened for the presence of antigen-specific binders, either by panning or by direct colony screening (Riechmann & Muyldermans, 1999). The B-cells expressing antigen-specific VHH circulating in the blood of an immunised dromedary can quickly reach 1 % of the total B cell population. Therefore, the direct testing of 1000 individual clones of the VHH library is expected to yield perhaps 10 antigen binders

(Muyldermans, 2001). Colony screening with labelled antigen, as performed in the late

1980s (Huse et al., 1989), or screening of the induced cultures by ELISA is indeed feasible for the VHH libraries (De Meyer, et al., 2014). However, retrieving the binders by panning is to be preferred to the screening of individual colonies, because the panning will also select for those binders with the highest affinities and those that express better in bacteria. VHHs can be selected through panning of phage particles on immobilised antigen (Pardon, et al., 2014). A pH shock eluted antigen bound phage particles and these were regrown through infection of bacteria (Hoogenboom et al., 1991;

Muyldermans, 2001). Specific adsorption onto the solid surface of an ELISA plate is still the most common way to immobilise targets for selection by phage display (Pardon, et al., 2014) and this is what was used (see 2.8.2).

3.2.3 Screening libraries for the presence of VHHs

It is recommended in the literature to try different capturing or immobilisation methods when screening libraries for VHHs (Pardon et al., 2014). For example to vary the antigen concentration, use different detergents or ligands, try different washing and incubation buffers or to use different elution methods.

Two methods of immobilisation were tested in this part of the project. One method involved protein coated ELISA plates which allowed the selection of VHHs against individual antigens (see 2.8.2). After 3 rounds of panning, the VHHs were expressed, then DNA from colonies was prepared and analysed. The results obtained showed that after 3 selection rounds a considerable proportion of clones carry the correct size insert and express VHHs upon induction. An average of 70 % of the clones were positive in our hands, although Figure 3.2 A shows more. Colonies carrying the positive clones were picked from the master plate and used to express the VHHs in the bacterial periplasm as described in the methods section (see 2.3.2). The positive, expressed VHHs were tested via western blot (see 2.4.2) as VHHs made for this study are all his-tagged (Figure

3.2 B and D) and all of these showed expression of the correct sized protein (VHH + his).

The second method tested is aimed at situations when specific antigens are unavailable, and a more generic immunogen can be used. A total lysate of gastrula and neurula embryos was used to immunize a llama, resulting in a library containing a mixture of

VHHs recognising all proteins present in the embryos at this developmental time point.

To select for VHHs recognising cytoskeletal proteins, a mainly insoluble cytoskeleton preparation was used as the immobilized surface. VHHs were allowed to bind to the insoluble cytoskeleton, collected by centrifugation and unbound material was washed off, this process was repeated four times. To elute the bound phage, triethylamine was added to the cytoskeleton pellet, incubated and centrifuged. To test whether specific

VHHs had been purified, a total embryo lysate was separated by SDS-PAGE and the

VHH was in turn used as primary antibody in the western blot (see 2.4.2), followed by

the α-his antibody used for detection (Figure 3.2 C). These western blots displayed a uniform signal in all lanes at 30 kDa, which was found to be a false positive when further analysed.

These multi-antigen libraries yielded a vast number of VHHs. While screening for VHH clones both at the DNA or protein level, many positive results were obtained among the hundreds tested (above). However, the complexity of these libraries made it very difficult to find a target protein-specific VHH. Multiple tests against Xenopus embryo lysates (see

2.4.1) at a variety of developmental stages were carried out as western blots (over 30 in total), as well as ELISA tests (over 15) on purified proteins, all being negative (data not shown). These results together with the number of articles describing the use of a single antigen for immunization, contributed to the decision to immunise llamas with single antigens.

Figure 3. 2. Screening libraries for the presence of VHHs. The library made from the llama injected with 36 proteins, was used to prepare VHH DNA from separate colonies. DNA from these colonies was digested with NdeI producing a band migrating at 1.7 kb, which indicates the presence of VHH (A). Colonies that showed positive for VHH DNA were induced to produce periplasmic VHHs, which were then purified. These VHHs contain a his- tag and were detected by western blotting with α-his antibody (B). A VHH raised against mixed gastrula-neurula protein extract was purified by affinity binding to Xenopus cytoskeletal proteins in collaboration with Keiji Itho (C). Its specificity was tested by western blotting against Xenopus embryo proteins: 1) total protein stage 11; 2) total protein stage 15; 3) cytoskeletal protein. Schematic representation of VHH detection by western blot (D). Periplasmically expressed VHHs were run on SDS-PAGE, transferred into a nitrocellulose membrane and incubated with α-his antibody followed by α-mouse IgG- HRP conjugated secondary antibody and chemiluminecence detection.

3.3 Llama immunisation with single antigens

The libraries created in the USA by the Sokol lab produced libraries that were too complex for successful screening. In 2015, a scientific paper reported the use of a VHH to inhibit β-catenin-mediated Wnt signalling (Newnham et al., 2015). These VHHs were produced at the Centre for Dairy Research (CEDAR, University of Reading). Laura

Newnham introduced us to the farm procedures and provided us with a sample from the published β-catenin VHH and its mutated control (used as proof of principle in chapter

6).

Two llamas were obtained in December 2015, held and maintained by CEDAR. These animals had a period of adaptation of 6 months. One of the dairy cows from the farm contracted tuberculosis (TB), thus the farm was unable to conduct experiments until all the animals were tested and showed results clear of TB. This period extended for 20 months. In August 2017 the llamas, Sunny and Valentine, were immunised, one with a long peptide and the other with an octomeric MAP respectively (Table 5.1 and Table

5.2).

100

3.3.1 Lymphocyte isolation

Leucosep tubes were used to isolate the lymphocytes (see 2.7.2) and the resulting cells were counted in a haemocytometer. The yields were within the expected range.

a. Sunny: 3.4x107 total lymphocytes (1 aliquot of 680 µl)

b. Valentine: 5.9x107 total lymphocytes (2 aliquots of 590 µl)

The cells were pelleted by centrifugation and re-suspended in 10 % freeze medium at 5 x 107 cells per ml. Eppendorf tubes containing the lymphocytes were stored at -80ºC until

RNA can be made from them.

3.3.2 Testing for an immune response in llama serum

In order to discover whether the llamas had produced antibodies against Cdk1 and Ascl1, western blotting (see 2.4.2) was performed comparing the pre-immune and immune sera on extracts from embryos overexpressing Cdk1 or Ascl1. Additional proteins of the expected size were detected by the appropriate sera (Figure 3.3) strongly suggesting that there has been an immune response to the immunogens (Figure 5.2).

101

3.4 Discussion

The construction and characterisation of the VHH libraries created by the Sokol laboratory in the USA provided us with the knowledge and starting materials needed to transfer those techniques to Portsmouth.

VHHs from these libraries were selected by panning against embryonic cytoskeleton or immobilised proteins, and their DNA and expressed periplasmic proteins tested. While in Sokol’s lab, the worked focused on the library generated against the gastrula-neurula stage embryonic protein extract from which many VHH expressing clones were isolated.

However, they all bind yolk proteins (Itoh & Sokol, 2014). The fact that all VHHs isolated and tested from this library recognise the same embryonic yolk protein highlights the need for pure proteins and single or a reduced number of immunogens per animal.

Once in Portsmouth, the library from 36 purified proteins was created and screened against two of its immunogens, Liz1 and Ndel1. These antigens were supplied to the

Sokol lab by Prof Anna Philpott and Prof Viki Allan, who kindly donated these proteins to us.

We were able to isolate clones containing VHH sequences, confirmed by their size when digested with specific enzymes, and the proteins produced were positive on α-his western blots. However, these did not recognise any Xenopus proteins, despite having selected them against immobilized Xenopus proteins, suggesting two possibilities. First, that the selection method was ineffective, and second, that the immunogens had not been checked for purity. The immunogens could have carried contaminating bacterial proteins which would generate a far greater immune response than the proteins from

Xenopus, which, as a vertebrate, is much more closely related to llama. The level of complexity of this library will therefore be very extensive and most likely be dominated by the response to the bacterial proteins (Leenaars & Hendriksen, 2005). This made the

102

purity of the immunogen, in our experience, one of the most important points to consider when raising these types of antibodies.

These two libraries can be compared, as it is likely that yolk proteins are either at higher concentrations or have greater immunogenicity levels than other proteins in the gastrula- neurula lysate, supposedly a mixture of all proteins present in that developmental time point in the embryo. Yolk dominance stopped any antibody against the target proteins being isolated. This is similar to the lack of useful antibodies being present in the 36 antigens library, which was most likely caused by E. coli contaminants dominating the immune response. The antigens used in this study were provided by 12 different laboratories, raised and purified in different ways (likely E. coli expressed) and were not checked inhouse before administration to the llama. Therefore, the fact that there were no Xenopus proteins detected when testing the complex, 36 protein library is not surprising. Based on this experience, I recommend that bacterially produced proteins are not used for llama immunization. By this time, the knowledge about the difficulties encountered when screening VHH libraries created with multiple antigens was coming to light. It was then that the decision to immunise llamas in the UK was made.

Serum from the immunised llamas was collected and tested for its response which was positive. To complete this work which was delayed for 18 months by the TB infection, libraries will be made from the isolated lymphocytes and VHHs against Ascl1 and Cdk1 selected.

103

4 Expression of Xenopus transcription factors to be used as antigens for immunisation and tools for VHH purification

4.1 Introduction

Understanding the genetic regulatory networks involved in development and diseases requires, among other things, antibodies to transcription factors (TFs) for both system- wide and focused studies; this works well with the rapid and powerful gain and loss of function experiments possible in Xenopus (Grainger, 2012). The experience gained while working on the complex libraries described in Chapter 3, drove us to avoid bacterial expression and to look for highly purified proteins to use as antigens for llama immunization. In this chapter, we explore two methods of protein expression, a commercial in vitro kit and a cell culture method employing Human Embryonic Kidney cells. This was to understand how successful these approaches were likely to be on a larger scale, and to test the process of nanobody production and purification with a subset of these proteins.

The first step in raising antibodies in any host is to obtain purified protein in sufficient quantities for the immunisation of the host; Muyldermans et al. (2001) suggested that around 100 µg are needed for VHH raising and purification in Llama glama. However, an updated protocol from UCB (Newnham et al., 2015) suggests 500 µg per injection, thus, at least 2.5 mg per immunogen for 5 injections.

Antibodies are proteins that are produced by the body as an immune response to a specific antigen, which can be an alien protein (Janeway, 2001). Xenopus proteins, either expressed endogenously or in other organisms, can be used to immunise llamas, generate an immune response, and so antibodies specific to the injected protein can be

104

isolated from the llama’s serum. We have identified that obtaining purified proteins is the major bottleneck in the antibody production process.

Expression systems are organisms or cell extracts in which a genetically engineered protein is produced. For protein expression, one of the most commonly used expression systems is bacterial cells. These are transformed with a plasmid containing the desired protein’s coding nucleotide sequence that can be fused to other proteins for selection, expression enhancement or as a supplement to aid the protein in folding, purification and visualisation. A concise review of the type of bacterial cells used to express proteins was given by Gomes et al. (Gomes, Byregowda, Veeregowda, & Balamurugan, 2016). The main disadvantage with proteins generated in this manner for us is the potential immunogenicity that any carry-over of bacterial proteins could have on a mammalian host, since they will be recognised strongly as non-self by mammals (Leenaars &

Hendriksen, 2005).

A different approach was considered, which consisted of the expression of proteins in vitro. Many companies offer kits which contain a biomolecular translation machinery extracted from cells, together with specially modified vectors and customised reagents to aid the cloning, transformation, transcription and later translation of the protein. This process is also called cell-free protein expression, as the proteins are produced in the absence of the cell itself. There have been multiple reviews of these processes in recent years (Carlson, Gan, Hodgman, & Jewet, 2014; Chong, 2008; Zemella, Thoring,

Hoffmeister, & Kubick, 2015).

As a cellular alternative, HEK293T cells, a kind gift from Simon Kolstoe, were used for transient transfection of recombinant proteins. This method, combined with the design and cloning of the protein sequence into the delivery vector via In-Fusion, has been well established by the Oxford Protein Production Facility (OPPF) and transferred to

Portsmouth via Jonathan Chamberlain (Nettleship et al., 2015). The vectors used in this study derive from a single ancestor pTriEx2 (Novagen), which was modified to produce

105

pOPINE, the first vector of the pOPIN series. From the pOPINE vector, OPPF has developed a library of vectors for protein expression with multiple qualities such as inter- host vectors, vectors with secretion signals and a wide variety of tags (Berrow et al.,

2007; Nettleship et al., 2008).

Cdknx (also known as p27Xic-1) and Neurog2 proteins were chosen as targets for in vitro expression, due to their popularity within the community. These are two key developmental proteins involved in neural differentiation (Vernon & Philpott, 2003), cell cycle control and neurogenesis (Movassagh & Philpott, 2008; Nguyen et al., 2006;

Vernon et al., 2006). Mechanisms by which these proteins regulate the cell cycle are widely studied with Xenopus oocytes and egg extract systems (Maller, 2012; Ryffel et al., 2003). In situ hybridization of the mRNA has led to the visualization of the mRNA expression in the embryo, however visualization of the protein has not been reported

(Bao, Talmage, Role, & Gautier, 2000; Nishitani et al., 2015; Sabherwal et al., 2009;

Schlosser et al., 2008; Vernon & Philpott, 2003).

For the protein expression in HEK cells, multiple proteins were studied. These were also chosen by request from the community and the approach involved discarding targets that presented difficulties, and carrying on with the ones that worked. The idea was to test whether the approach could work, then to focus on troubleshooting.

106

4.2 Results

Xenopus laevis being allotetraploid, carries duplicates of many of its genes (Session et al., 2016). They are referred as the S and L versions. Careful study of the target sequences was carried out prior to designing the experiments. Examples of these approaches are given through the thesis for a few of these targets.

4.2.1 Neurog2

Using Xenbase (James-Zorn, et al., 2013), the RNA-seq data of both Xenopus laevis L and Xenopus laevis S homeologues were compared to find out which gene pair had less mRNA expression from oocyte to pre-MBT embryo stages. By choosing the homeologue with less or no mRNA expression at this point, the maternal contribution to the later embryo was minimised. Early mRNA expression might make interpreting protein expression data from the antibody difficult, therefore misleading our judgment of antibody quality. This is due to maternally derived mRNAs having been able to express the protein in multiple sites of the embryo at different stages unrelated to mRNA expression at the tested stage (Wuhr, et al., 2014). The Xenopus laevis S Neurog2 homeologue has higher mRNA levels at the egg stage (Figure 4. 1); therefore the Xenopus laevis L sequence was chosen to work with.

The mRNA RefSeq found in Xenbase was used to identify the coding sequence (CDS from nucleotides 357-1001), and its translation was checked against the Human genome to make sure it encoded the target protein using NCBI’s BLAST (Madden, 2002). To sub- clone this sequence into the expression vector, NdeI and NotI would be used, so the sequence was checked for lack of those sites using the APE program

(http://biologylabs.utah.edu/jorgensen/wayned/ape/).

The primers used to amplify the target sequence needed to be designed so the enzyme sites used to cut the PCR fragment were in frame (NdeI and SalI), therefore resulting in

107

a maintained open reading frame when ligated into the expression vector. To design the forwards primer 20 bases from the Xenopus laevis L RefSeq sequence were selected starting at position 357, and the NdeI enzyme sequence was added at the 5’ end:

5’ cat atg atg gtg ctg ctc aag tgc g 3’.

For the reverse primer, 20 bases were selected from the Xenopus laevis L RefSeq sequence starting at nucleotide 1001. The sequence TGA was left out, as this is a stop codon and the ribosome needed to carry on reading past this site to make the protein as a GST fusion, which is the approach used in this kit. The primer was composed of the complementary sequence of these 20 base pairs plus the SalI sequence:

5’ gtc gac aat gaa agc gct gct ggc ag 3’.

4.2.2 Cdknx

The process to generate an insert with the correct primers for Neurog2 was repeated with Cdknx (Figure 4. 2). In this case, both homeologues were expressed identically at early stages, but there is no information in Xenbase other than the genomic sequence and the RNA-seq data for the S homeologue. We therefore chose to work with cdknx L.

From Xenbase’s Ref-seq, we identified CDS nucleotides 23-652 as the coding sequence.

The sequence’s translation was corroborated as previously described and primers were generated in the same manner:

Cdknx forwards (NdeI): 5’ cat atg atg gct gct ttc cac atc gc 3’

Cdknx reverse (SalI): 5’ gtc gac tcg aat ctt ttt cct ggg gg 3’

108

Figure 4. 1. Neurog2 Xenopus laevis L (green) and Xenopus laevis S (purple) RNA-seq profiles by development stages.

Note in Xenopus laevis S there is expression in the oocyte, egg and gastrula stages, dropping considerably after MBT, which indicates maternally-derived expression.

Figure 4. 2. RNA-seq profile for Xenopus laevis cdknx by developmental stages. RNA-seq data indicates expression in the blastula and gastrula stages probably reflecting some maternal imput, sequences are available only for the L version of this gene.

109

4.2.3 In vitro protein expression of neurog2 and cdknx ORFs

Target proteins were chosen from a list of genes suggested by the community. To express Cdknx and Neurog2, a commercial, coupled in vitro transcription-translation system was tested (see 2.1.9, 2.4.3). The reaction was based on HeLa cells lysate and reported to yield 100 µg/ml of protein per reaction (reactions being of 25 µl each). Cdknx

(630 bp; 210 AAs; 25.2 kDa) and neurog2 (645 bp; 215 AAs; 25.2 kDa) ORFs were sub- cloned into pT7CFE1-cGST-HA-his (Figure 4. 3 and Figure 4. 4), a plasmid supplied with the kit. This allowed the expression of the protein as a GST fusion containing HA and his-tags for ease of purification and detection of the produced protein.

Plasmids containing the complete coding region for the proteins were sourced from the

EXRC plasmid collection and the target sequence amplified by PCR. These single bands were cloned into pCR2.1-TOPO and colonies were screened. Two preps for each gene were sent for sequencing to corroborate the results; one neurog2 clone showed the correct sequence but cdknx did not. A new cdknx template was obtained. PCR amplification products (see 2.1.9.2) were ligated into 2 vectors, pGemTeasy and Topo pCR4.0, since this was a time-sensitive project and the efficiency of cloning into different vectors varies (see 2.1.3, 2.1.4). DNA from three out of 16 colonies had the correct size insert and sequence. Cdknx in pGemTeasy, neurog2 in pCR2.1 and the commercial destination vector pT7CFE1-CGST-HA-his were double digested with NdeI and SalI and the appropriate ligation performed (see 2.1.9.4). The successful insertions were assessed by double digestion (see 2.1.1) of mini preps with SalI and NdelI. Two out of

33 final clones had the correct insert, confirmed by sequencing before use.

110

Figure 4. 3. Protein expression via the human in vitro transcription – translation kit based on HeLa cell extracts (Pierce 1-Step Human Coupled IVT Kit – DNA).

Target sequences were amplified by PCR with primers containing SalI and NdeI sites, sub-cloned into pGemTeasy and digested with the mentioned enzymes. These fragments were cloned into the pre-linearized pT7CF1-CGST-HA-his vector. Proteins were expressed following manufacturer’s instructions and purified by his-trap chromatography columns.

111

Figure 4. 4. pT7CFE1-CGST-HA-his vector map. Target sequences were cloned into the pre-digested vector at the NdeI and SalI sites, to obtain GST-HA-6his tagged proteins.

112

4.2.3.1 In vitro Cdknx and Neurog2 protein expression

Proteins were expressed in vitro using the 1-step human high yield mini IVT kit following the manufacturer’s instructions. Expressed proteins were analysed via Coomassie blue stained SDS-PAGE (Figure 4.5 A; see 2.4.2.1). The SDS-PAGE confirmed that the gel had run nicely, but that insufficient exogenous protein was expressed to be detected against the endogenous HeLa proteins from the lysate.

The produced proteins were purified by affinity chromatography using his-trap columns

(see 2.4.5.1) and analysed by western blot (Figure 4. 5 B; see 2.4.2). The targets were identified as a single band very close to the expected band of 51 kDa (arrowed), showing that they have been expressed. The yield was quantified using the bicinchoninic acid assay (BCA assay) following the manufacturers’ instructions and showed that only 0.005 mg of each protein was produced (1/5 of the maximum reported yield). The reaction cost

(£400 per reaction), and the lower than expected yield showed this approach was not a time or cost effective production method, since it had already been optimised by the manufacturer but did not allow us to make close to the 2.5 mg needed to immunize a llama. Therefore, alternative methods of protein production had to be considered.

113

Figure 4. 5. Cdknx and Neurog2 proteins expressed in vitro via the high yield IVT kit. Proteins were expressed in HeLa lysate, purified by Ni column and eluted in 3 steps (1-3 above). The proteins were separated by SDS-PAGE and visualized by western blot using α-his antibody. A constant band can be observed at 49 kDa in elution 1 and 2 for both proteins, this is very close to the expected size of 51 kDa of the GST fusion protein. The third elution from the NI column for both protein is close to null, insinuating that all (or most) of the protein was eluted in steps 1 and 2.

114

4.2.4 HEK cell expression of target proteins

Protein expression via mammalian cell lines allows the production of proteins with relatively low background immunogenicity from contaminants when it comes to their use as antigens in mammals, compared to proteins produced in non-mammal mediums.

There are a variety of cell lines that can be used as a “machinery” to express proteins, such as Lung carcinoma (A549), Glioblastoma (GAMG), Embryonic kidney cells

(HEK293T) and Cervical carcinoma (Hela) among others (Bandaranayake & Almo, 2014;

Geiger, Wehner, Schaab, Cox, & Mann, 2012).

The Oxford Protein Production Facility (OPPF, (“OPPF-UK: Home,” n.d.)) has developed a standardised method which allows optimisation for producing a wide range of proteins based on Human Embryonic Kidney cells (HEK cells) and other platforms (see 2.4.4).

They have developed a system which exploits the In-fusion cloning system provided by

Takara (www.takarabio.com) and developed an innovative plasmid collection (the pOPIN vectors) with unique enzyme combinations that allow the cloning of the insert of interest with high success rates and can be used for expression in multiple systems. pOPIN destination vectors were successfully transformed, grown, digested and gel purified

(MinElute Gel Extraction Kit, Qiagen) for subsequent cloning.

115

4.2.4.1 Isolation and cloning of target sequences

Vector plasmids were a gift from Ray Owens (OPPF) through Addgene (pOPINE

(Plasmid #26043); pOPINF (Plasmid #26042); pOPINM (Plasmid #26044); pOPINJ

(Plasmid #26045)). The following plasmids were a direct donation from Ray Owens: pOPINTTGneo, pOPINE-StrepII, pOPINGM, pOPINFS, pOPINE-Flag3. (Berrow et al.,

2007). In Table 4. 1 are the cloning sites and product tags for each vector according to

OPPF.

Plasmids were digested (see 2.1.1) according to the linearization sites recommended by

OPPF (https://www.oppf.rc-harwell.ac.uk/OPPF/protocols/cloning.jsp)

Table 4. 1. OPPF vectors used to express proteins in HEK cells, their recommended cloning sites and expected products. Taken from https://www.oppf.rc-harwell.ac.uk/OPPF/protocols/cloning.jsp or given by Ray Owens (personal communications).

Vector Cloning sites Product

pOPINE NcoI - PmeI POI-KHIS6

pOPINF KpnI - HindIII HIS6-3C-POI

pOPINM KpnI - HindIII HIS6-MBP-3C-POI

pOPINJ KpnI - HindIII HIS6-GST-3C-POI

pOPINTTGneo KpnI - PmeI SS[RPTPmu]-POI-KHIS6

pOPINE-StrepII NcoI - PmeI HIS6- StrepII -3C-POI

pOPINGM KpnI - PmeI SS[MBP]-POI-KHIS6

pOPINFS KpnI - NheI HIS6-3C-POI-StrepII

pOPINE-Flag3 NcoI - PmeI HIS6- Flag3-3C-POI

The inserts of interest were obtained by PCR amplification (see 2.1.6) of the target sequences from plasmid templates (supplied by the EXRC) with specific primers. These primers contained a region homologous to the pOPIN destination vector of choice to enable the In-Fusion reaction (see 2.1.10.7). A full table with the primers used in this study can be seen in Appendix I. Fli1, gata2, hairyI and neurog2 sequences were 116

targeted by PCR. The fli1 expected sequence was not amplified; the amplicon produced was considerably larger than expected, and an alternative donor plasmid was used as template (Figure 4. 6 A) providing with the correct size amplicon. gata2, hairyI and neurog2 were amplified from sequenced plasmids and the products contained a mixture of bands of different sizes. Gata2 exhibited a smear suggesting two possible reasons; contamination of the template, or insufficient extension time, giving products of slightly different sizes. The extension time for gata2 amplification was increased from 1.15 minutes to 2 minutes due to its length; this allowed the positive amplification of the gene.

The gata2 amplicon was gel purified using Quiaquick gel extraction kit. HairyI and neurog2 were subjected to PCR (see 2.1.6) with a temperature gradient for the annealing step to find the optimal temperature to favour the amplification of a single amplicon.

HairyI with pOPINE and pOPINF primers were successfully amplified while neurog2 was not. An alternative PCR master mix was also tested with a wider temperature gradient range with no success. PCR reactions for lisI, cdknx, ndelI, deltaI, talI, mytI, neuroD- short, ndelI-short, fliI-short, gata2-short, hairyI-short and ncor2 with pOPINE and pOPINF specific primers were set up at this point (data not shown). Optimization was required to amplify gata2-Short, myt1, and talI. The optimisation of these PCR reactions yielded correct amplicons for the following inserts:

 deltaI-pOPINE  hairyI-Short-pOPINE

 cdknx-pOPINE  hairyI-Short-pOPINF

 ncor2 -pOPINE  mytI-pOPINE

 gata2-pOPINE  mytI-pOPINF

 gata2-pOPINF

 neuroD-Short-pOPINE

 neuroD-Short-pOPINF

 fliI-Short-pOPINE

 fliI-Short-pOPINF

117

These PCR products were treated with DpnI (see 2.1.10.5) which cut any carry through plasmid and then subjected to a Cloning Enhancer treatment (see 2.1.10.6) as detailed in the In Fusion protocol. Transfection of these constructs into HEK cells is detailed later in this chapter.

Upon receiving a different vector, pOPINTTGneo (a kind gift from Ray Owens), more primers were designed (Appendix I), and PCR amplification of the insert sequences was attempted. J-line Xenopus cDNA was generated (see 2.1.8) and used as starting material for some of the reactions (Table 4. 2) since no amplification was obtained from plasmids containing the inserts. PCR optimisation was attempted for the cdkI, fliI and talI clones without success (data not shown). Positive amplicons were cloned into pOPINTTGneo, sequenced and used for transfection (see 2.4.4). The unsuccessful cloning procedures were not investigated, as we used a drop-out strategy in which we carried on working with the targets that worked, the rest were left behind.

For cdkI, a different set of primers was designed, aiming to reduce the difference in annealing temperature between the primers, and so aid amplification. These primers were tested in 2 different donor plasmids; both gave good amplification with the correct size bands, 303 bp for the N-terminal and C-terminal regions, and 906 bp for the full- length version (Figure 4. 6 B).

118

Figure 4. 6. PCR amplification of inserts to clone into pOPIN vectors by in-fusion. Inserts for cloning into the pOPIN vectors were prepared using standard conditions (methods) and separated by gel electrophoresis. fliI (1362 bp) 2nd amplification with pOPINE (1) and pOPINF (2) primers (A). cdkI PCR product to be cloned into pOPINTTGneo vector successfully amplified as C-terminal, N-terminal and full-length coding region (B).

119

Table 4. 2. PCR amplification of target genes to be cloned into pOPINTTGneo vector, stating template used, primer, amplicon size and success in amplification and sub-cloning.

PCR Cloned and Template PCR primers Expected size successfully sequenced gata2- Plasmid DNA 380 bp Yes Yes pOPINTTGneo ncor2- Plasmid DNA 351 bp Yes Yes pOPINTTGneo talI-N- Plasmid DNA 422 bp No pOPINTTGneo talI-N- cDNA 422 bp Yes No pOPINTTGneo neurog2-C- Plasmid DNA 435 bp No pOPINTTGneo neurog2-C- cDNA 435 bp Yes No pOPINTTGneo neuroD- Plasmid DNA 320 bp Yes No pOPINTTGneo neuroD-C- Plasmid DNA 323 bp Yes No pOPINTTGneo talI-C- Plasmid DNA 290 bp No pOPINTTGneo talI-C- cDNA 290 bp No pOPINTTGneo fliI-C- Plasmid DNA 320 bp No pOPINTTGneo fliI-C- cDNA 320 bp No pOPINTTGneo cdkI-N- Plasmid DNA 303 bp No pOPINTTGneo cdkI-C- Plasmid DNA 303 bp No pOPINTTGneo cdkI-FL- Plasmid DNA 906 bp No pOPINTTGneo cdknx-N- Plasmid DNA 306 bp Yes Yes pOPINTTGneo cdknx-FL- Plasmid DNA 627 bp Yes Yes pOPINTTGneo

ddH2O Negative control

120

4.2.4.2 In-Fusion cloning of Xenopus proteins into pOPIN vectors for protein expression

The In-Fusion reactions were set up and incubated at 50ºC for 15 seconds (see

2.1.10.7). These reactions were then frozen at -20ºC until transformed.

Transformation of cloning reactions into Dh5α cells (see 2.1.5) produced well-formed colonies (~100-150 per plate). DNA from the positive colonies was prepared and sequenced (Table 4. 3), where the sequences corresponded to the proteins of interest

(Table 4. 3) large quantities of DNA were prepared, and they were transfected into HEK cells.

Table 4. 3. Sub-cloning successes for Xenopus proteins.

hairyI- neuroD- Vector deltaI gata2 cdknx ncor2 mytI fliI-Short Short Short

No No No Empty pOPINF OK OK OK OK colonies colonies colonies vector

No Empty Miss- pOPINE OK OK OK OK OK colonies vector matches

Empty Empty No No Empty Empty Empty Empty pOPINJ vector vector colonies colonies vector vector vector vector

For mytI, primers were designed to amplify a shorter 1065 bp region of the 3369 bp full- length mytI to go into pOPINE and pOPINF. After some optimisation of the PCR, this shorter version of mytI was successfully sub-cloned into pOPINE.

CdkI C-terminal, cdkI full-length, gata2 full-length, ncor2 and neurog2 C-terminal were successfully sub-cloned into pOPINTTGneo vector. Correct clones were isolated and sequenced for gata2, ncor2 and neurog2.

121

4.2.5 A new approach, the use of vectors with secretion signals to allow the proteins to exit the nucleus and cell

Four new vectors were subsequently obtained from OPPF, a gift from Ray Owens

(pOPINE-StrepII, pOPINE-Flag3, pOPINGM and pOPINFS).

These vectors were chosen as they all have a secretion signal, and they have different tags which might aid solubilization and secretion of the proteins. After growth of stab cultures and DNA preparation, the vectors were linearised; gel-extracted and purified

(see 2.1). The vector’s purification was assessed by running 2 µl of each gel extracted plasmid on an agarose gel (data not shown). This confirmed the quality and quantity of the purified, digested vectors before ligation.

As for the other vectors, PCR was performed on gata2, hairyI and cdkI templates with primers containing a homology region both to the template and vector. The vector’s homology was slightly adjusted to match the suggested tag on the In-Fusion protocol. A full list of primers used can be found in Appendix I. Following the In-fusion protocol, amplicons were subjected to a cloning-enhancer treatment before the In-fusion reaction into the destination vector (Figure 4. 7; see 2.1.10.4, 2.1.10.6, 2.1.10.7).

122

Figure 4. 7. Cloning enhancer treated PCR amplification of gata2, hairyI and cdkI to go into pOPIN vectors. The amplicons were all a single band of the expected sizes (+ or - 4 bp depending on primers used). For gata2-C-term the expected size was 380 bp, for hairyI-C-terminal 372 bp, cdkI-FL 905 bp and hairyI-FL 801 bp.

123

4.2.5.1 Transfection of pOPIN plasmids encoding Xenopus proteins into HEK293T cells

HEK293T cells were cultured as previously described (Nettleship et al., 2015). A 24 well culture plate was seeded with 3 different cell concentrations and left to grow overnight

(see 2.4.4.1). The wells with the healthier looking cells were chosen to be transfected using JetPrime reagent (Polyplus) using the manufacturer’s instructions (see 2.4.4.2).

All the vectors in the pOPIN collection have either a C-terminal or N-terminal his-tag.

This facilitates western blot detection of expressed products and their purification.

Transfected proteins were run on a 10 % SDS-PAGE, transferred onto a nitrocellulose membrane and analysed by western blot (see 2.4.2). Two types of antibody were used to detect the expressed proteins, an α-his HRP conjugated antibody (Abcam, ab1187), and a primary α-his antibody used with a secondary HRP conjugated antibody (Santa

Cruz Biotechnology, G-18). The α-his HRP conjugated antibody exhibited non-specific binding giving constant bands in the region of ~45 kDa (arrowed in Figure 4. 8). The

Primary α-his with secondary HRP conjugated, also exhibited a background band, however only on the cell lysate and at about ~90 kDa (arrowed in Figure 4. 8). Since this is very distinct from the size of the desired targets, we decided to use α-his primary with a separate HRP-labelled secondary for all subsequent western blot studies.

From these experiments, we observed that pE-neuroD-short and pE-hairyI-short were expressed, however, they were found in the cell lysate and not the media. The vectors sequences were examined carefully, and despite these vectors being supplied as secretory, no signal sequence was present. The presence of a secretion signal in the expression vector allows proteins to be transported from the cell out to the media

(Nettleship et al., 2015). This makes the expression and purification more efficient.

Therefore, a different set of vectors was sourced.

Following the faint expression of NeuroD present in Figure 4. 8 and the successful expression of HairyI shown both in Figure 4. 8 and Figure 4. 9, large quantities of

124

pOPINE-neuroD-short and pOPINE-hairyI-short DNA were prepared. Medium scale transfections were carried out as previously described (see 2.4.4.3). The expressed proteins were assayed by western blot of their his-tag to test expression. The lysate, the media, and the re-suspended cell pellet were analysed and successfully expressed (data not shown).

125

Figure 4. 8. Small-scale HEK cell transfected proteins media and cell lysate. Media and lysate from each transfection were separated by 10 % SDS-PAGE and transferred into a membrane. The membrane was incubated with α-his HRP conjugated antibody. HairyI- short and neuroD-short were expressed in vector pOPINE, predominantly in the cell lysate. No secretion into the media was observed.

Figure 4. 9. Small-scale HEK cell transfected proteins media and cell lysate. Media and lysate from each transfection were separated by 10 % SDS-PAGE and transferred into a membrane. The membrane was incubated with α-his primary antibody, and HRP conjugated secondary antibody. HairyI-short was expressed in vector pOPINE, predominantly in the cell lysate. No secretion into the media was observed. Bands were visible in the Rara2 and FliI lanes, however these proteins run very low on the gel, not correlating to their actual size (arrowed in Figure).

126

4.2.5.1.1 Purification of medium scale transfected HairyI and NeuroD proteins

Cells from the pE-hairyI-Short and pE-neuroD-Short medium scale transfections (see

2.4.4.3) were lysed by mPER which contains a mild, non-denaturing detergent that dissolves cell membranes to extract and solubilize total protein from most cellular compartments (see 2.4.4.5.2), purified with his SpinTrap columns (see 2.4.5.1) and loaded into a 10 % SDS-PAGE. HairyI was highly expressed but the subcellular nature and insolubility of hairyI decreased the efficiency of the mPER treatment. For NeuroD, only a small quantity of expressed protein was observed (Figure 4. 10), therefore only

HairyI was taken further.

HairyI expression plasmid was transfected into four T175 flasks (see 2.4.4.3), following

4 days of incubation the cells were harvested, and intracellular proteins were extracted using the mPER method (see 2.4.4.5.2). Despite the potential of this method, the proteins were still in the cell pellets after the extraction (Figure 4. 10).

A summary table of transfections can be found in Appendix I.

127

Figure 4. 10. Intracellular protein extraction via mPER treatment of four T175 flasks of HEK293T cells transfected with the pE-hairyI-short construct.

HEK293T cells were transfected with pE-hairyI plasmid, cells were collected, treated with mPER and analysed by Coomassie stain. The successfully expressed short version of hairyI is present very strongly in the cell pellet compared with the mPER lysate, suggesting the insolubility of the protein. Two bands can be prominently detected, one at the expected size of 14.8 kDa and one slightly lower. These could be two different isoforms of the same protein.

128

4.2.5.1.2 HairyI storage and integrity check after storing and other transfections

Due to external factors, the successfully expressed HairyI protein had to be stored for over two months. To maximise the chances of recovering the protein after storage, multiple transfected samples were pooled into one volume and purified through an his- trap column (Figure 4. 11; see 2.4.5.1). Fractions 5, 6 and 7 were found to have HairyI protein, they were collected and pooled again into a single volume. The pure protein was then split into two equal quantities, one mixed with its same volume of 20 % glycerol and the other left in its native form, and stored at -80ºC.

Following storage, the proteins were thawed and tested on a western blot (Figure 4.12

A; see 2.4.2). The integrity of the protein was unaffected by the storage, more so, the signal levels between the protein stored with glycerol and the protein stored in its native state were the same, suggesting both storing methods worked well.

Both batches were pooled and dialysed against solution A from the AKTA start overnight at 4ºC, to remove glycerol that could affect further purification (see 2.5.2). The cleaned proteins were purified on an AKTA start with an his-trap FF Crude (1 ml) column (Figure

4.12 B and C; see 2.4.5.2). Fractions 10 and 11 of elution 1 and fractions 3-6 of elution

2 were collected and analysed on a 14 % SDS-PAGE (Figure 4.13) together with the load material. The protein was detected before and after dialysis, and fractions 3 and 5 of the second elution.

Meanwhile, large-scale transfections in 4x roller bottles containing a total of 1L of cultured cells were set with hairyI and gata2 (see 2.4.4.4). Upon culturing cells for 5 days, the media was extracted and the cell pellets lysed by the mPER method (see 2.4.4.5.2) and purified with a his-trap FF-crude column on an AKTA start (see 2.4.5.2). Fractions were collected and analysed via western blot (see 2.4.2). The expressed proteins were present only in the cell pellets. The sequences were re-checked to corroborate the correct arrangement and in frame location of the insert and signal sequence, which were

129

correct. Two theories arose. One was the probability of the proteins being secreted and degraded quickly thus; they were not present in the media at the end of the transfection course. The other suggested that the proteins were not efficiently secreted from the cell.

A small-scale transfection was carried out and a time course was taken in which the medium was analysed to assess the secretion of the transfected proteins (see 2.4.4.2).

The medium from the time course and the pelleted cells were analyzed by western blot and Coomassie staining (Figure 4.14 A and B). Gata2 was clearly present in the cell pellet in both analyses. The constant band present at ~28 kDa in the western blot gave doubts as there was no sign of the protein on the Coomassie-stained gel. A separate western was run to confirm these results (Figure 4.14 C), where there was a clear expression in the cell pellet, however, no proteins in the media at any time point. The first western false positives for the media might have been sample leakage when loading the gel.

These results show that the proteins were not secreted at any point from the cell.

Transcription factors are often DNA binding proteins which are found mainly in the nucleus. The strong protein-DNA interactions might be one of the causes they were not successfully secreted.

The vector sequences were checked to corroborate that the secretion signal was present, were it came to our attention that this was missing from the plasmids we used, despite their original description.

A new vector was tested, pOPINTTGneo, which contained a secretion signal. Small- scale transfections of Gata2, Ncor2 and Neurog2 were carried out to test the transfection efficiencies of the pOPINTTGneo vector. Neurog2 was detected when the cell pellet was analysed in a western blot (showed in Figure 4.15), however this was unsuccessfully up- scaled and purified.

Large-scale transfected cells with the gata2-pOPINTTGneo, was solubilized by sonication and 8 M urea treatment (see 2.4.4.4, 2.4.4.6), and purified through an his-trap 130

column in the AKTA start (Figure 4.16; see 2.4.5.2), where the pellet gave a signal with the α-his antibody, while all the fractions from and around the elution’s peaks were negative (data not shown). The flow-through was clear, suggesting that the protein possibly became stuck in the column. The column material was boiled in SDS-PAGE loading buffer and tested on a Coomassie stained gel, where no appropriately sized bands were detected (Figure 4.16).

By this point, the cost of protein production via HEK cell expression had become very high due to the complications that arose due to the proteins being insoluble and the production time-frame was greatly underestimated, thus commercial companies were contacted to either synthesise or express proteins.

131

Figure 4. 11. Purification via his-trap column of HairyI expressed protein. Fractions 5, 6 and 7 were collected, pooled and stored at -80ºC. Western blot of his-tag protein eluted in 9 fractions after being purified with the FF-crude his-trap column (A). UV absorbance was detected between fractions 4 and 6, indicating the elution of the protein, confirmed by the western blot (B).

132

Figure 4. 12. HairyI integrity and purification after storage.

The stored proteins were thawed and tested on a western blot (A), both batches were then pooled to make a total volume of 3 ml, which was dialysed into solution A to separate the protein from the glycerol purified through a his-trap column in the AKTA start. The UV peak corresponding to the protein elution (B, C enlarged) was very low.

133

Figure 4. 13. Further purification of HairyI analysed by western blot. HairyI sample before and after dialysis, together with a number of fractions collected from the AKTA start purification. HairyI is present before and after dialysis, however, it was eluted in the second 0-100 % imidazole gradient, most prominently on fraction 5.

134

Figure 4. 14. Gata2 transfection.

The stability of the protein produced by the transfection of HEK293T cells with Gata2 in pOPINTTGneo, was tested by taking a time course of the media to assess protein secretion from the cells to the media and its potential degradation. Analysis by Coomassie stain (A) and western blot (B), shows the presence prominent band at around the expected size of 20 kDa. Western blots are more sensitive than Coomassie stained gels, therefore the band is visible in all the samples, however there is no visible protein in the Coomassie stained gel. A repeat western (C), demonstrates that the band present in the first western was an artefact from over-filled wells, as no protein was visible on the time course media. Gata2 was present in the pelleted cells, however it was not secreted to the media in any of the tested time periods.

135

Figure 4. 15. Neurog2 protein transfected with pOPINTTGneo.

Small-scale transfection of Neurog2 by HEK293T cell transfection using the pOPINTTGneo vector. A selection of DNA preparations were used to maximize the chances of expression, where 4/5 DNA preparations produced a band at the expected protein size of 17 kDa (arrowed). The cell pellet containing expressed Gata2 was loaded as positive control (arrowed). Despite multiple attempts to up-scale and solubilize Neurog2, this was unsuccessful, and due to time restrictions, not pursued.

136

Figure 4. 16. Western blot to visualize Gata2 expressed protein before and after purification. Gata2 protein made by HEK293T cell transfection, was collected and solubilized with 8 M urea before loading into an his-trap column for purification. Samples were taken before loading the column, in 1 ml elution volumes and the flow through was also tested. Column samples were boiled in SDS loading buffer, to extract any bound protein. These samples were tested both by western blot and Coomassie stained SDS-PAGE. The Gata2 protein is present at the expected size (20 kDa) in the pre-loaded sample, but not in any elution or even the flow-through or column itself. A separate column, used to purify HairyI was also tested, without success. These findings suggested the loss of the protein either by precipitation or degradation. This was not pursued.

137

4.2.6 Commercially synthesised and expressed proteins

Although proteins were expressed successfully via transfection into HEK cells, these showed a high level of insolubility and proved innefficient and costly to purify.

Commercial companies were contacted to express 25 potential targets in a mammalian system. These targets were mainly early Xenopus transcription factors. Four out of eight companies were not willing to try to express these in mammalian systems due to the difficulty in the purification process for nuclear proteins. The remaining companies would carry out the expression and purification, at high price, however, they could not assure us there was going to be a product at the end of the procedure (Table 4. 4). A list of the suggested targets can be found in Appendix III.

Table 4. 4. Companies contacted to express the given targets and their responses. Thermo Scientific Synthesis of 9/25 proteins (£700-£4000 small scale) Bio Basic Yes - (> £ 7000 per mg) Neo Biolabs Yes (> £ 5000 small scale) Biorbyt No Lonza No Abm No GeneScript No Evotec Yes (>£16000 each w/ IMAC, cleavage, sub IMAC, SEC)

Due to the unfeasibility of using mammalian systems to produce an antigen in-house or commercially, we explored the option of synthesising peptides. Two companies were considered, Alta Bioscience and Thermo Scientific.

Alta Bioscience provides the option of synthesising peptides as octameric multi antigenic peptides (MAPs). MAPs are an effective method of raising antibodies using synthetic peptides. They consist of eight identical peptide chains which are synthesised joined to a common polylysine core. The molecular weight of these peptides, (10 kDa-15 kDa) is sufficient for them to be highly immunogenic without the need to couple to a carrier protein. This technique gives very high titers and works for any epitope except the

138

extreme C-terminus of a protein, according to Alta Bioscience and in the experience of our colleagues (A. Thorne, F. Myers and C. Crane-Robinson, personal communications).

Sequences of 3 proteins were sent to Alta Bioscience. Their analysis identified the most immunogenic section of the protein. These suggested sequences were checked against the Xenopus genome to avoid generating antibodies that recognise multiple Xenopus proteins. For CDKI a 42 AA 8-branched sequence was suggested (Figure 4.17) and for

MytI a 33 AA 8-branched MAP was also suggested (Figure 4.18).

The third antigen supplied by Alta Bioscience, Kash5 (Figure 4.19), was generated and used as directed by a collaborator. This study aimed to validate the use of octameric

MAPs to raise antibodies in sheep and test it as an alternative route to antibody generation in llamas. The development of this antibody is described in the “Kash5 antibody” section of this thesis (Chapter 5).

Thermo Scientific provided us with a different approach, the option of generating long peptides of up to 120 AAs long. These peptides are long enough to provide an immune response without the need to couple them to a carrier protein. Out of the 26 targets that were sent for analysis, only peptides to 9 were suggested as shown in Table 4.5. This was due to the extensive analysis needed to identify suitable peptide sequences for the other proteins. The purity of the peptides produced decreases with the peptide length.

Also, the quantity obtained is about 1/5 of that obtained with the Alta Bioscience. Despite these disadvantages, we obtained two antigens from Thermo Scientific, asclI, and rarα2, so that we could directly compare the MAPs and linear peptides as antigens. The five sequences suggested by both companies were checked by BLAST to confirm no cross- reactivity with other Xenopus proteins.

139

green). The sequence suggested by Alta bioscience Alta by suggested sequence The green).

1 octomeric 1 MAP sequence suggestedby Bioscience. Alta

Cdk1 coding region (dark pink) with structures predicted by psipred (light green) and Phyre2 (dark Phyre2 and green) (light psipred by predicted structures with pink) (dark region coding Cdk1 highlighted in orange. Figure 4.

140

sted by Alta

sequencesuggested by Bioscience. Alta

Myt1octomeric MAP

Figure 4. Figure4.

Myt1 coding region (dark pink) with structures predicted by psipred biosciencehighlighted (light in orange. green) and Phyre2 (dark green). The sequence sugge

141

ctures predicted by psipred (light green) and Phyre2 (dark green). The sequence suggested by Alta bioscience

octomericMAP sequence suggested by Bioscience. Alta

. . Kash5

Figure 4. Figure4.

highlighted in highlighted orange. in Kash5 coding region (grey) with stru

142

Table 4. 5. Thermo Scientific proposed peptide sequences for 9/26 proteins that were sent for analysis. The amount and purity are estimates from the company, corresponding to the length and nature of the sequence.

Peptide Peptide Sequence Amount Purity Name Length

AAVARRNERE RNRVKLVNLG FATLREHVPN ASCLI GAANKKMSKV ETLRSAVEYI RALQQLLDEH 70 1-4mg >90 % DAVSAAFQSG

SGETVLKIKK TRRVKANNRE RNRMHNLNSA NEUROG2 LDSLREVLPS LPEDAKLTKI ETLRFAYNYI 79 1-4mg >90 % WALSETLRLG DPVHRSAST

VKQRRNRRVK ANDRERNRMH NLNSALDALR NGN3 SILPTFPDDA KLTKIETLRF AHNYIWALSE 70 1-4mg >90 % TLRMADQSLL

MYENVDVSPT HYHMMDFYSH NRQCLWPEKR RARA2 48 1-4mg >90 % INPYGTPLGT QHWSSSNH

HIGNRANPDP NCCLGVFGLS LYTTERDLRE VFSKYGPLSD VSIVYDQQSR RSRGFSFVYF TRA2B 100 1-4mg >80 % ENVDDAKEAK ERANGMELDG RRLRVDFSIT KRPHTPTPGI

MAAFHIALQE EMISAPAVLP RLSAGTGRGA CRNLFGPIDH DEMRSELKRQ LKEIQASDCQ CDKNX 100 1-4mg >80 % RWNFDFETGT PLKGIFCWEP VESKDMPSFY SQNRSIAANT

DKKPLPTASS GPIQLWQFLL ELLQDSSCQK LISWTGNGWE FKLSDPNEVA RRWGRRKNKP ETV2 119 1-4mg >80 % RMNYEKLSRG LRYYYHKNII HKTGGQRYVY RFVCDLQDLL SRPTQTSQPP TKTQRSRIQ

MSGKEESSNY AATAEEEAVN QGFVLPEGEI MPNTVFVGGI DITMDEIEIR DFFTRFGNVK DAZL 120 1-4mg >80 % EVKIITDRTG VSKGYGFISF SDEVDVQKIV KSQISFHGKK LKLGPAIRKI CTYVQPRPVV

EIEEILDVIE PTQFKKIQEP LFKQISKCVS SSHFQVAERA LYFWNNEYIL SLIEENIDKI PPP2R5A 120 1-4mg >80 % LPIMFGSLYK ISKEHWNPTI VALVYNVLKT LMEMNGKLFD ELTGSYKAER QREKKKELER

143

4.3 Discussion

The expression of these proteins in bacteria was not considered, to minimize the background on downstream analysis that was not an option.

The in vitro transcription of Neurog2 and Cdknx via cell-free extracts was successful, showing that this is definitely a powerful tool to express DNA binding and insoluble proteins. The size of the proteins detected in the western correlated to the size of the fusion formed by the expressed protein and the MBT tag. However, the low yield produced by this method could have been investigated, but due to the expense and time restrictions, this process was left aside. The price of this method has decreased significantly since 2012, becoming more accessible.

Protein production via HEK cells was an established method at our facility, where the expression of high quantities of soluble material easily purified was routinely done. A table with all the transfection plasmids created and the transfection results can be found in Appendix 1. Transfections were carried out in 6 or 12 well plates at first, and once a plasmid was found to express the desired protein, it was transfected into T175 flasks or roller bottles.

We worked with either full length or short versions of CdkI, DeltaI, FliI, Gata2, HairyI,

MytI, NeuroD, Neurog2, RARα2, Ncor2, TalI and Cdknx. A variety of nine vectors were tried, all derived from the original pOPINE vector. Protein expression was achieved for

CDKI (pOPINE-StrepII-FL; pOPINE-Flag3-FL), Neurog2 (pOPINTTGneo-C-term),

Gata2 (pOPINTTGneo-C-term; pOPINE-StrepII-C-term) and HairyI (pOPINE-C-term; pOPINE-StrepII-C-term; pOPINE-Flag3-C-term; pOPINFS-C-term).

All the target proteins we worked with were found to be insoluble and the inconvenience that this caused was underestimated until the commercial companies themselves refused, or overpriced the non-guaranteed expression of these proteins in mammalian hosts.

144

The decision to use peptides was taken, and even from the target sequences we supplied to various commercial companies, there were some that not even peptides were suggested to us, because of the complexity of their structures. Alta Bioscience supplied us with 3 octomeric MAPs, for MytI, CDKI and Kash5. ThermoFisher provided with pricing for 9 peptides, from which we asked them to synthesise AsclI and Rarα2.

CdkI and AsclI were later used to immunize llamas, to generate VHHs; and Kash5 was used to immunize a sheep, to provide a conventional antibody. The success of these approaches is described in Chapter 5.

From my wealth of experience, if I was to do this again or if somebody else was trying to do it, what I would recommend for the raising of Xenopus-specific antibodies against transcription factors, is to avoid bacterial expressed proteins (due to the potential carry over contamination), avoid in vitro expression (as the cost of producing sufficient antigen to immunize a large mammal is excessive), and also to avoid HEK cell expression (as the purification of the proteins is nearly impossible). Therefore the investment of time and effort to design a suitable peptide is recommended.

145

5 Host immunisation and antibody screen

5.1 Introduction

This chapter explores two methods used to obtain antibodies. Kash5 is used in the most conventional of the ways, as an immunogen in sheep. This immunization generates the most widely used antibodies, conventional light and heavy chain Abs. Ascl1 and Cdk1 were used to immunize 2 llamas and generate single chain antibodies.

Since VHHs were first discovered, their uses have been broadened considerably, going beyond the conventional applications of antibodies. VHHs are produced as coding region libraries which allow them to be selected, provided in unlimited quantities and engineered to have a range of properties, such as enhanced tissue penetration and lower immunogenicity (Lauwereys et al., 1998; Muyldermans, 2001). They have become key players in multiple applications, including fundamental research, diagnostics, and therapeutics (De Meyer et al., 2014; Ma, Sun, Kang, & Wan, 2014).

A brief review of the potential that VHHs offer is given in the main introduction (Arias et al., 2014; Hagihara & Saerens, 2014; Kim, Hussack, Kandalaft, & Tanha, 2014;

Siontorou, 2013). More importantly, if VHHs as tools are introduced into the Xenopus community, backed up by published findings, they are expected to have significant impact on the type, ease, cost and time-scales of experiments, hence improving the usefulness of Xenopus as a model.

146

5.2 Results

5.2.1 VHH Libraries

VHHs were generated via the immunization of two camelids (Llama glama) with an antigen each. After 6-weeks the serum was extracted and a cDNA library was generated from the lymphocytes which hold sequences encoding all the expressed antibodies present in the llama’s serum (see 2.7). These single-chain antibodies can either be screened directly or alternatively a series of selection rounds (panning) can be run. In panning, the proteins encoded by the library are bound to the target protein, and then washed off, eventually purifying E. coli containing a phagemid encoding the specific, high affinity VHH (Figure 5.1). Commercially synthesised peptides were used as immunogens for this study as the in-house protein expression did generate proteins, however these were insoluble and could not be purified (Chapter 4).

Two types of antigens were used, a long peptide (50 AAs) synthesised by Thermo Fisher for AsclI, and an octomeric MAP (42 AAs) made by Alta Bioscience from CdkI. AsclI is a transcription factor that plays a key role in neuronal differentiation and since it was first described in Xenopus in 1993 (Ferreiro, Skoglund, Bailey, Dorsky, & Harris, 1993) became a popular target when studying mesoderm induction and neural differentiation among other developmental processes (Min et al., 2016; K. M. Taylor & LaBonne, 2007;

Wylie, Hardwick, Papkovskaia, Thiele, & Philpott, 2015).

CdkI in the other hand, is a protein involved in the control of the eukaryotic cell cycle.

This protein is one of the most studied in Xenopus, with over 800 papers associated with its activity and pathways in frog development (taken from Xenbase). It was first described in Xenopus 1975 (Drury & Schorderet-Slatkine, 1975) when describing the maturation promoting factor (MPF) in Xenopus laevis oocytes. Since then is been associated with spindle formation, mitogenic signalling and protein phosphorylation between others

(Lohka & Maller, 1985; J L Maller, 1986). Both these proteins have been extensively

147

studied and antibodies against them are high on the priority list for the Xenopus community.

Llama immunization followed procedures first established and recommended by Dr

Laura Newnham (UCB). In summary, two llamas were acquired by the EXRC and kept under contract at “The centre for dairy research” (part of the University of Reading). The contract included the EXRC’s right to use the llamas for research purposes, and the centre for dairy research, to house, look after and follow the immunizations procedures when instructed. Immunizations followed the recommended timings and concentrations

(Table 5.1 and Table 5.2). Three bleeds were collected, one before the immunization course started (pre-immune), one after 3 immunization were carried out (1st bleed) and the third, after 5 doses of the antigen were given.

The bleeds were composed of 300 ml of blood and 5 ml of serum from each llama. The first bleed was treated by centrifugation to isolate the plasma by standard methods.

The serum was used as primary antibody on western blot (see 2.4.2) with α-llama IgG goat-HRP conjugated as secondary antibody (Figure 5.2), and the immune response of the animals was confirmed with bands of the expected size present in the immune serum but not in the pre-immune.

Following immunization, the llamas can be “re-used” and there is no need for them to be culled. There is a recommended resting period of 6 months between immunizations, which allows the immunogen to leave the blood stream, and the llama to recover from the treatment. At the end of their use llamas are employed as guard animals by farmers

(The centre for dairy research, personal communications), making their use ethically preferable to animals normally killed like rabbits and sheep.

148

Immunogen Llama glama

Collect blood

Express as soluble VHHs Isolate lymphocytes

mRNA isolation and RT-PCR Rinse Bind Express as soluble VHHs PCR with nested primers Panning

Identify clone with Elute antigen-specific VHHs Ligation to phage Coat display vector

Screen 1000 colonies VHH phage display Transform in E. coli and generate library

Figure 5. 1. Schematic representation of the processes involved in the generation of VHHs. Two llamas were immunized with an antigen each and after 6 weeks 100 ml of the blood was collected. A series of processes take the raw llama blood, isolates the lymphocytes, from these white blood cells the mRNA is extracted, and a pool of all the VHH specific cDNA is generated by a nested PCR. The VHH specific regions were cloned into a phagemid vector which produced the bacterial libraries. These libraries could be infected with helper-phage and screened directly or undergo selection rounds to isolate the high affinity binders to the chosen target, a process called panning. The identified VHHs are finally expressed and characterized.

149

Figure 5. 2. Immunization response assessment by western blot analysis of post-injected llama serum, compared to pre-immune.

Two llamas were immunised with one antigen each. Sunny (llama #1) was injected with the Ascl1 long peptide generated by ThermoFisher, while Valentine (llama #2) was injected with the Cdk1 octomeric MAP provided by Alta Bioscience. Immunization procedures followed previously stablished protocols outlined in Tables 5.1 and 5.2. Total protein extracts from embryos injected with cdk1 and ascl1 to over-express such proteins, were separated in a 10 % SDS-PAGE and detected by using a 1/250 dilution of the llama serum in blocking buffer. The llama proteins were visualized with 1/2000 dilution of HRP goat α-llama antibody in blocking buffer. Ascl1 displays a protein of the correct size (~22 kDa), while in the Cdk1 western blot two distinct bands are observed, one of the correct size (~33 kDa) and the other lower (~18 kDa). The nature of the second band would be clarified once the library is created and screened.

150

Table 5. 1. Immunization schedule for Sunny. A 100 ml pre-bleed was taken before starting the course of injections. Following three doses of 0.5 mg Ascl1 protein mixed with the adjuvant GERBU, a first bleed was taken and the llama immunized twice more the same antigen. A second, final bleed was then taken.

Schedule

Day Date Notes Action No

100 ml Pre-bleed - Whole blood in sodium 0 07th August 2017 heparin + 1 ml serum

1 08th August 2017 Dose A 0.5 mg Ascl1 protein + GERBU LQ

05th September 28 Dose B 0.5 mg Ascl1 protein + GERBU LQ 2017

03rd October 56 Dose C 0.5 mg Ascl1 protein + GERBU LQ 2017

Decision point: are Test (or final) bleed - 400 ml whole blood in 66 13th October 2017 further doses are sodium heparin + 1 ml serum required?

84 31st October 2017 Dose D 0.5 mg Ascl1 protein + GERBU LQ

28th November 112 Dose E 0.5 mg Ascl1 protein + GERBU LQ 2017

08th December Final bleed - 500 ml whole blood in sodium 122 2017 heparin + 1 ml serum

151

Table 5. 2. Immunization schedule for Valentine. A 100 ml pre-bleed was taken before starting the course of injections. Following three doses of 0.5 mg Cdk1 protein mixed with the adjuvant GERBU, a first bleed was taken and the llama immunized twice more the same antigen. A second, final bleed was then taken.

Schedule

Day Date Notes Action No

100 ml Pre-bleed - whole blood in sodium 0 07th August 2017 heparin + 1 ml serum

1 08th August 2017 Dose A 0.5 mg Cdk1 protein + GERBU LQ

05th September 28 Dose B 0.5 mg Cdk1 protein + GERBU LQ 2017

03rd October 56 Dose C 0.5 mg Cdk1 protein + GERBU LQ 2017

Decision point: are 13th October Test (or Final) Bleed – 400 ml whole blood 66 further doses are 2017 in sodium heparin + 1 ml serum required?

31st October 84 Dose D 0.5 mg Cdk1 protein + GERBU LQ 2017

28th November 112 Dose E 0.5 mg Cdk1 protein + GERBU LQ 2017

08th December Final Bleed - 500 ml whole blood in sodium 122 2017 heparin + 1 ml serum

152

5.2.2 Kash5 antibody raising and characterization

The tight link between this reaserch and the community through the EXRC encouraged the immunization of sheep with an octomeric map which provides comparison terms for the raising and testing of antibodies agains VHHs.

The coiled-coiled domain-containing protein 155 (CCDC155), more commonly known as

Kash5, is a nuclear protein that is involved in the connection between the nuclear lamina and the cytoskeleton (Houben, Ramaekers, Snoeckx, & Broers, 2007). Kash5 is required for homolog pairing during meiotic prophase (Zeng et al., 2018) and an antibody against this would be a useful tool for cell biologists studying a variety of cell processes.

This protein was not straight forward to work with bioinformatically as it is not annotated in the Xenopus genome. The protein sequence was previously identified and provided to us by Prof Viki Allan. The Xenopus tropicalis sequence was the starting point, and the sequence identity when compared to Xenopus laevis is surprisingly low (Figure 5.4).

There is however likely to be sufficient identity with Xenopus laevis for the epitope to be recognised in both species. A search of the most recent version (9.1) of the Xenopus tropicalis genome, made possible the identification of the gene. It is situated in chromosome 7, scaffold_707 between bases 104521509 and 104540066, and consists of 16 exons. This annotation has been submitted to Xenbase (Figure 5.3)

The antigen was generated by Alta Bioscience as an octameric MAP (Figure 5.3) of a section of the Xenopus tropicalis version of the protein. The 20 AAs long sequence was compared to that of Xenopus laevis and found to be very similar, only 2 amino acids different (Figure 5.4). All tests for this antibody were carried out using Xenopus laevis animals, unless otherwise stated.

Two milligrams of octameric Kash5 antigen were shipped to “Origen Antibodies Limited” to immunize one sheep under their standard protocol. In synthesis, this protocol involved taking a pre-immune bleed, followed by the immunization with 0.5 mg of the Kash5 antigen. After a period of 4 weeks, a bleed was taken and the sheep immunized again 153

with a second dose of 0.5 mg of the antigen. This was done 3 times with an outcome of one pre-immune and 3 immune-bleeds with accumulated dosages of the antigen.

The serum was tested by western blot on egg extracts and embryo extracts. As positive control, an MBP tagged Xenopus Kash5 protein was used, a kind gift from Prof Victoria

Allan (University of Manchester).

The Kash5 protein is 59.159 kDa in weight and the MBP tag 42.5 kDa, the fused protein is ~90 kDa in weight.

154

. .

of this protein. Exon arrangements are also displayed, some exon

sequencewith the Kash5 tranlated

mRNA

sequence and its translation (grey). Highlighted in orange is the sequence used to generate the octomeric MAP, proposed MAP, octomericthe generate to used sequence the is orange in Highlighted (grey). translation its and sequence

Xenopustropicalis

Kash5

3 Xenopus tropicalis Xenopus

Kash5 by Alta Bioscience personel as the most immunogenic and unique fragment sequences overlapdue to the need determine to exon boundaries,collaborative a processXenbase. with

Figure5. 155

Comparison of Xtr KASH5 vs Xla PREDICTED: lymphoid-restricted membrane protein-like [Xenopus laevis] Sequence ID: XP_018083668.1 Length: 572 Number of Matches: 1

Query 1 YKLQGTCWNPE--SEEGQLSDSKNATQGCLNGVPNSTSEEQLLDTAFEACDTEGRGEVTV 58 YK GT NPE SE+G LSDS NAT+ CL PNS+SEEQLLDTAF+ACDTEG+GEVTV Sbjct 2 YKHLGTFLNPEEPSEDGLLSDSANATEDCLGDFPNSSSEEQLLDTAFQACDTEGKGEVTV 61

Query 59 FQIIDYLKSVTDQPCDGDRLQHLCKMLDPEEQGRSLDLPTFRKVMSEWIASCCSDSDITE 118 +QIIDYLKSVTDQ CDGDRLQHLCKMLDPE+QGRSLDLPTFR++MSEWIASCC+D +TE Sbjct 62 YQIIDYLKSVTDQTCDGDRLQHLCKMLDPEDQGRSLDLPTFRRLMSEWIASCCNDRYVTE 121

Query 119 TVENIKSEKGSSLPCG-VLEREAEHEGYGG-DVNKFMGEHADFISKIRDLSFANKKLMDQ 176 VE+IKSE GS LPC V E AEHEGYGG DVNKF+GEHA F+SKIRDL+FANKKLMDQ Sbjct 122 KVEDIKSESGSYLPCANVFEHGAEHEGYGGGDVNKFLGEHAYFVSKIRDLNFANKKLMDQ 181

Query 177 KMKLQKSLETADETNSHLLEEISEVKSKLKISQQAVNHSRSLSNELEDLKAYTKTLEDKI 236 K+KLQKSLETADETNSHL+EEISEVKSKLKISQQAV+HSRSLSNELEDLKAYTK+LEDKI Sbjct 182 KIKLQKSLETADETNSHLIEEISEVKSKLKISQQAVSHSRSLSNELEDLKAYTKSLEDKI 241

Query 237 AAFGSQKEELEKDNLSLCNQRQMLQEEMYNLLIDKENLKGNIDSLSAEKDNMGLQICEYE 296 +AFGSQKEELEKDNLSL NQR+MLQEE+YNLLI+K+ LKGNIDSL+AEK MG QI EY+ Sbjct 242 SAFGSQKEELEKDNLSLSNQREMLQEEIYNLLIEKDKLKGNIDSLTAEKGKMGHQISEYK 301

Query 297 SLISQKEMLLEEEKVKSQQLRSIVDEHRLQVQVLQREKRLLHEQLLQPNKENMPLNQSVS 356 LISQKEMLLE+ V+S +L+SI++EHRLQVQVL+REKRLLHEQLLQPNK+NM L+QSVS Sbjct 302 RLISQKEMLLEDRNVQSHRLQSILEEHRLQVQVLKREKRLLHEQLLQPNKDNMFLHQSVS 361

Query 357 VKQLPLPVQSVHRELEEIQEKIR-VKSELLSPICGIPQASGSNVLCPIQDTDMTSDDAWA 415 VKQLPLPVQSVHRELEEIQE+ + K+ELLSPICGI QA+GS++LCPIQD DMTSDDAWA Sbjct 362 VKQLPLPVQSVHRELEEIQEQKKCFKTELLSPICGISQATGSHMLCPIQDADMTSDDAWA 421

Query 416 VTELLLVQLKQGTEALIRSVHEMTSSDQSVSYLYEQSGHFLLELNYLLELKNVWERQWGR 475 +TELLLVQLKQGTEALI SVHEM SSD SVSYLYE+S HFL ELNYLLELKN+WERQ G Sbjct 422 ITELLLVQLKQGTEALITSVHEMASSDHSVSYLYEKSSHFLQELNYLLELKNIWERQSGS 481

Query 476 LGRAVQLYTGKEAKEPDSEELPATEWLCTDTLHTSIVPCIQLQNKSSLWLTLILLGTLMF 535 L A+QLYTGKE ++PD+ E PAT+W+CTD L T VP +QLQNKSS+WLT++LLGTLMF Sbjct 482 LSGALQLYTGKETRKPDTHEFPATKWICTDMLDTPTVPYMQLQNKSSIWLTVLLLGTLMF 541

Query 536 CPAWAGEHIWTVLKAEIWPHLDLHYHSPPPV 566 CPAWAGEHIWTVLK E+WPHLDLHY SPPPV Sbjct 542 CPAWAGEHIWTVLKGELWPHLDLHYPSPPPV 572

Figure 5. 4. Xenopus tropicalis Kash5 protein sequence aligned to Xenopus laevis Kash5 protein sequence.

These orthologues have regions with considerable differences between species, however, the region selected to generate the octomeric MAP (20 AAs shown in yellow) is more conserved with only two amino acids difference (shown in green) and 15 amino acids of consecutive alignment, which is sufficient for the antibody to recognize the site. These sequences and alignments were provided by Prof Viki Allan.

156

Upon the end of the immunization course, the three bleeds and the pre-immune sample were tested via western blot against the purified Kash5/MBP protein and wild-type

Xenopus laevis embryo extracts (Figure 5.5; see 2.4.2).

Embryos were collected at stage 6, 11 and 21 and kept at -80ºC until treated. Proteins were extracted as previously described (see 2.4.1). The SDS-PAGE used to characterize these antibodies were 10 %, unless otherwise stated. Two different protein ladders were used, since the preliminary results had unclear sizing and the second marker was used as size confirmation.

Two concentrations of the purified protein were run, together with 3 different quantities of the mixed embryo extract. The serum was used at a 1/1000 dilution in blocking buffer

(5 % milk in TBSTw), followed by 1/2000 Donkey α-sheep HRP conjugated secondary.

All three bleeds showed specific binding to the purified kash5/MBP protein which runs at the expected size corresponding to the Kash5/MBP fusion (59.159 kDa + 42.5 kDa).

Also, stronger bands around the 55-60 kDa can be seen, corresponding the expected size of endogenous kash5 protein. However, these western blots using crude serum show considerable background (Figure 5.5), thus the antibodies were purified.

The sheep’s pre-immune serum was treated with caprylic acid and dialyzed, as previously described (see 2.5.1, 2.5.2). Alta Bioscience provides pre-packed columns of the antigen peptides covalently bound to controlled pore glass, (Peptide-CPG) for use as affinity columns to purify antibodies raised by the MAP version of the same peptide.

Serum from Kash5 injected sheep was treated with caprylic acid, dialysed and affinity purified through the supplied column before being desalted in a G-25 column (see 2.5).

The purification yield for bleed 1 was 4.48 mg, bleed 2 was 10.32 mg and bleed 3 was

1.12 mg. These were aliquoted into 0.1 or 0.2 mg lots and lyophilised before storing at -

20ºC until needed. An aliquot of each bleed was re-suspended in PBS and stored at 4ºC to be tested by western blot (Figure 5.6). The purified serum was tested at 1 µg/ml.

157

To test the purified antibody a 10 % SDS-PAGE was prepared and run with kash5/MBP positive control, embryo extract and egg extract were also included. The egg extract was included as kash5 is a key protein in the early egg/embryo development as it associated with chromosome pairing and synapsis among other key developmental processes

(Zeng et al., 2018).

The background in the western blot was reduced considerably after purification. Even the pre-immune control was much clearer after the caprylic acid and dialysis treatment

(Figure 5.6).

158

Figure 5. 5. Western blot to test the immune response of the sheep against the Kash5 octomeric MAP immunisations. Two size ladders were used in these western blots to confirm the size marks from the SeeBlue plus2 ladder, which were not clear in previous runs. Proteins of two sizes have been arrowed in green for the Precision plus protein marker. The Kash5 protein is 59 kDa in weight and fused to the MBP tag is 90 kDa. The four gels were run and treated under the same conditions, the only variable being the primary antibody used. Pre-immune bleed (A); first bleed (B); second bleed (C); third bleed (D). The positive control, an MBP-Kash5 protein, was clearly detected at ~95 kDa on the 50 ng well as a duplex (orange), which is possibly due to proteolysis that may occur and change the protein mass. Protein from embryos at stage 6, 11 and 21 were isolated and run at 0.5, 1 and 2 embryos per lane. Despite the high background levels, there were two prominent bands absent in the pre-immune at 20 kDa and 55 kDa (blue arrows). To further assess the antibody, this was purified.

159

Figure 5. 6. Western blot to test purified kash5 antibody. The Kash5 protein is 59 kDa in weight and fused to the MBP tag is 90 kDa. The four gels were run and treated under the same conditions, the only variable being the primary antibody used. pre-immune bleed (A); first bleed (B); second bleed (C); third bleed (D). The positive control, an MBP-Kash5 protein, was clearly detected at ~95 kDa in the 3 bleeds (orange), being negative on the pre-immune. Protein from embryos at stage 6, 11 and 21 were isolated and run at 0.5 and 1 embryos per lane. Egg extracts were included at 1 µl and 3 µl per lane. The high background levels observed before purification of the antibody were minimized, even on the pre-immune that was only treated with caprylic acid and dialyzed gave a clearer signal. The previously seen bands at 20 kDa and 60 kDa were present on bleed 1 and bleed 3 (blue arrows). The high molecular weight bands could be aggregated proteins producing background.

160

Figure 5. 7. Western blot to optimize purified kash5 antibody staining. The Kash5 protein is 59 kDa in weight and fused to the MBP tag is 90 kDa. The four gels were run and treated under the same conditions, the only variable being the primary antibody used. pre-immune bleed (A); first bleed (B); second bleed (C); third bleed (D). The positive control, an MBP-Kash5 protein, was clearly detected at ~95 kDa in the 3 bleeds (orange), being negative on the pre-immune. Different denaturation of the protein extracts were tested, altering the temperature and length of incubation, with no clear difference between them. Protein from embryos at stage 6, 11 and 21 were isolated and run at one embryo per lane and egg extracts at 1 µl per lane. The previously detected band at 60 kDa was particularly prominent on bleed one, while undetected with bleed 3. The band at 20 kDa was present mainly on bleed 3, and there was an extra band detected at about 40 kDa present on bleed 1 and bleed 3. To further test this antibody, embryos with over-expressed Kash5 need to be considered.

161

Following transfer, the membrane was incubated with 1/1000 dilution of purified serum in blocking buffer, followed by 1/2000 donkey α-sheep HRP conjugated secondary also in blocking buffer. Bleed 1 shows the strongest bands. Bleed 2 gave no visible bands with the embryo or egg extracts, and bleed 3 gave very faint bands.

The proteins did not penetrate the SDS-PAGE very well, thus a variety of denaturing methods were used improve protein displacement through the gel. At this stage, the loading dye also was adjusted and instead of using a DTT based loading dye, we used a mercaptoethanol based loading dye. Bleed 2 and 3 did not show the expected response (Figure 5.7).

By comparing at the amount of antibody collected in bleed 2 (10.32 mg) and bleed 3

(1.12 mg), with the lack of detection on the western blot, the assumption was made that the affinity column had failed to purify the antibody from the second and third bleeds.

The purification of antibodies from these bleeds was repeated and obtained 3.2 mg and

1.76 mg respectively.

To aid the identification of the correct bands on western blot, and confirm the specificity of the antibodies, embryos were micro-injected with kash5 mRNA to overexpress the protein and give a stronger kash5 signal compared to WT were tested (see 2.6).

To generate the mRNA, total RNA was extracted from embryos as previously described

(see 2.1.7) and cDNA synthesized following standard protocols (see 2.1.8). kash5 was targeted with gene specific primers (Table 2.4 and Table 2.5). Multiple primers for

Xenopus laevis and Xenopus tropicalis were tested to optimize amplification of kash5 from wild type cDNA with no luck. The ODC housekeeping gene was used as control to test the integrity of the cDNA with positive amplifications in all reactions (data not shown).

The kash5 sequence is A-rich and repetitive, thus PCR was not possible, despite optimization.

162

Kash5 was then synthetized by Genewiz (https://www.genewiz.com/) straight into a pCS2+ vector, from which mRNA was translated and injected into embryos. No overexpression was visible when tested with the Kash5 antibody (Figure 5.8).

163

Figure 5. 8. Kash5 overexpression to confirm the specificity of Kash5 antibody. Kash5 mRNA injection into one-cell stage Xenopus laevis embryos, allowed to reach st16 and total protein extracted from them. Un-injected embryos of the same stage were used as control. The 1st bleed of Kash5 antibody was used to detect endogenous Kash5, and visualized with his- antibody. A single band can be seen just below the 50 kDa marker, however this do not correlate to the previously detected band or Kash5 expected size of 59.1 kDa. The stability of this antibody came into question, therefore further experiments have to be considered to stablish if the loss of recognition of the 60 kDa was due to the antibody losing its affinity for the target.

164

5.3 Discussion

Kash5 is one of many key proteins required in early development, it forms part of the linker of nucleoskeleton and cytoskeleton during mitotic cell division (Zeng et al., 2018).

Work carried out with this protein was particularly challenging due to the lack of information available about it in Xenopus. Kash5 has not been mapped to the Xenopus genome and its expression together with EST or mRNA data is not available (to our knowledge). The information used to work with was provided by Prof Viki Allan, which allowed us to elucidate and propose new data for the community, currently being assessed by Xenbase.

The kash5 mRNA sequence is very repetitive and with multiple sections of consecutive

4 and 5 of the same nucleotide. The translated sequence also displayed repetitive amino acid motifs. These characteristics made Kash5 challenging to clone. Multiple attempts to isolate the kash5 sequence from genomic DNA were unsuccessful. Therefore the production of an octomeric MAP by Alta Bioscience was considered.

Compared to llamas, the sheep immunization was a straight-forwards process. Sheep immunization for antibody generation is relatively common and the procedures are widely available and easily accessible.

The response of the sheep to the antigen was clear from the un-purified, first bleed identifying the positive control, an MBP-tagged Kash5, and extra band of the expected size compared to the pre-immune. The detection of high molecular weight proteins, suggested that the embryo lysate was not completely denatured. To reduce background, the antibodies were purified, and the embryo proteins were tested with different denaturing approaches. The purification of the antibody was successful recognizing a clear band of the expected size with reduced background.

We used Kash5 in this study, because it was part of the agreement with the EXRC to try this protein, however there are several features of Kash5 which made it not the ideal

165

protein for these analysis, mainly its sequence repetitive nature. If we would have had the chance to choose, we would have chosen a well characterized protein, with known sequence and function. Also, we would have used the same protein to immunize the llama and the sheep, as this would have provided with a direct comparison of the benefits and drawbacks of both methods.

166

6 The use of a β-catenin VHH as proof of principle

6.1 Introduction

6.1.1 β-catenin’s importance in development

Vertebrate development is dictated by numerous pathways and processes by which the fertilised egg develops into an adult (Horner & Wolfner, 2008). Early development is influenced by a combination of the genetic background and the maternal signal carried by the egg (Tadros & Lipshitz, 2009). Finding the localization and functions of these maternal proteins combined with understanding the regulation of the transcription, mRNA turnover and later translation of new proteins are key to developmental biology studies.

One of the best studied, key pathways in development is that of Wnt signalling. Since the elucidation of the first members of this family in 1982 (Nusse & Varmus, 1982), there have been over 25 000 research papers published (taken from a PubMed search, April

2018) focused on different aspects of the Wnt signalling members, mechanisms and function. Wnt signalling occurs in all phyla (Loh, van Amerongen, & Nusse, 2016) and is well conserved. Nusse and Varmus (Nusse & Varmus, 2012) present a fantastic review of the Wnt pathway and its historical development, in which they highlight β-catenin as one of the key players in such processes. β-catenin is a protein that is maternally inherited and is key to many developmental processes in the early embryo such as the neurogenic niche for midbrain dopamine neurons (Tang, Miyamoto, & Huang, 2009) and cell fate (Huang & Niehrs, 2014) among others. β-catenin is localized in 3 distinct cellular compartments: the plasma membrane, the cytoplasm and nucleus (MacDonald, Tamai,

& He, 2009), and its study requires a diverse set of approaches, as for example conventional antibodies might bind the cytoplasmic protein, but cannot penetrate the nuclear envelope to bind the nuclear proteins under normal conditions (Marschall,

167

Frenzel, Schirrmann, Schüngel, & Dübel, 2011). β-catenin is an excellent example of a

“dual-function” protein, as it is involved in cell-cell adhesion and as later discovered, in signalling as a transcription factor.

168

6.1.2 β-catenin in Xenopus

β-catenin was first described in 1989, when cDNA studies in human HeLa and avian fibroblasts linked β-catenin with cell adhesion properties (Ozawa, Baribault, & Kemler,

1989). It wasn’t until 1991 that the same molecule was associated with cell to cell adhesion in Xenopus (McCrea & Gumbiner, 1991). Through the 1990s there were many studies around β-catenin and its effects on early development, such as the elucidation of its expression patterns (DeMarais & Moon, 1992). Its involvement in axis induction was first postulated using antibody injections (McCrea, Brieher, & Gumbiner, 1993), and confirmed by maternal knockdowns (Heasman et al., 1994). It was also implicated in mesoderm induction (Schohl & Fagotto, 2003), cell migration and histogenesis (Broders

& Thiery, 1995) among others. In 1997, a revolutionary methodology was published called “Morpholino” oligos, which allowed the knockdown of gene expression

(Summerton & Weller, 1997). These were morpholino ring based “oligonucleotide” fragments which bind to the endogenous RNA and block translation and splicing. These worked well in Xenopus; Heasman et al showed it working with β-catenin (Heasman et al., 2000). Gene function studies bloomed with this approach, although careful controls are needed (Eisen & Smith, 2008; Tandon, Showell, Christine, & Conlon, 2012) and some off-target effects appear difficult to avoid (Gentsch et al., 2018).

169

6.1.3 Loss of function experiments

Morpholino injections became the popular way of inhibiting gene function, particularly in

Xenopus and zebrafish. The injection of an antisense RNA created a phenotype representative of the embryo state when such protein is depleted, or knocked-down.

More so, the phenotype, can sometimes be rescued by the injection of the sense RNA, that, given time, restores the embryo to its original healthy state. Some of the first overexpression and knockdown β-catenin studies by sense and antisense RNA injections were performed by Janet Heasman in 1994 (Heasman et al., 1994) in which she found that knocking down β-catenin inhibits dorsal mesoderm induction in the early embryo. The key function that β-catenin has in the Wnt signalling pathway is highlighted when the gene is knocked-down and a variety of distinct phenotypes arise (Figure 6. 1).

Some embryos develop without dorsal structures, including the notochord, somites, and neural tube, embryos also lack heads and tails (Heasman et al., 2000). Figure 6. 1 illustrates phenotypes that occur when β-catenin is depleted from the embryo at 4-cell stage and 8-cell stage. These results demonstrate a dorsal signalling pathway dependent on nuclearly localised β-catenin. However, they also raise some new questions, such as why the depletion of β-catenin affects both the dorso-ventral and anterior-posterior axes, suggesting that β-catenin could be involved in other pathways

(Heasman et al., 2000).

The study of β-catenin and the Wnt signalling pathway have been the focus of many developmental biology laboratories were Xenopus serves as a model organism. This, combined with the possibility of acquiring a β-catenin VHH, made this gene a perfect candidate for proof of principle studies for understanding how having a VHH to a frog protein could be exploited.

170

Figure 6. 1. Phenotype caused by the depletion of β-catenin in the early Xenopus embryo. Control un-injected embryos (A); injections into two dorsal at the 4-cell stage (B); injection into two dorsal animal cells at the 8-cell stage (C); injection into two dorsal vegetal cells at the 8-cell stage (D); injection into two dorsal animal and two dorsal vegetal cells at the 8-cell stage embryo (E). Injection into two dorsal cells at the four-cell stage caused complete ventralisation in all cases (Heasman et al., 2000).

171

6.1.4 A VHH against β-catenin

An unpredicted meeting with previous users of the Reading facility for VHH production, made it possible for us to obtain a tested and published human β-catenin VHH

(Newnham et al., 2015). This VHH is commercially sensitive and can be used only under certain restrictions. A brief description of the procedures carried out by Laura Newnham to obtain, isolate and characterize the β-catenin VHH are describe below.

The β-catenin VHH was obtained by the immunization of 2 llamas with full-length, recombinant, human β-catenin protein. A phage library was synthesised and 2 x 108 clones screened. Seventy eight percent of the clones contained in-frame VHH domains and each VHH subfamily was present as proposed by Harmsen et al (Harmsen et al.,

2000).

The immune phage library was screened against biotinylated β-catenin by panning and monoclonal identification of positive clones was carried out by enzyme-linked immunosorbant assay (ELISA), in which 43 to 100 % were positive. Over 50 different variable region sequences that bound β-catenin were identified; 31 could be grouped into 12 distinct families, with 19 being significantly different from any family.

The functional inhibition of β-catenin activity was tested by a luciferase-based Wnt signalling HEK293T reporter bioassay, in which 4 active intracellular antibodies inhibited

β-catenin’s function. LL3 was chosen for further work as it did not display toxicity to the cell as did other active VHHs. In some cases, the penetration of the VHH into the cells caused the cell culture to perish or reduce growth. Three amino acids were mutated in the LL3 sequence to serve as negative control (Figure 6. 2 A). To assess the efficiency of the negative control, an ELISA assay was performed by capturing biotinylated β- catenin on to streptavidin coated plates, adding VHH-scFc constructs and revealed with

α-mouse Fc-HRP. As a negative control in this binding assay, an empty vector was used.

Results show that as the concentration of the VHH-scFc increases, so does the binding to the immobilised β-catenin; however, the mutated VHH-scFc does not bind (Figure 6.

172

2 B). The transfection efficiency of the LL3VHH expression clone and its mutated form was tested by analysing post-transfection cell lysis via western blot. A single band of the correct size was observed in the LL3 and LL3.CDRAAA, while no band was present in cells transfected with an empty vector (Figure 6. 2 C). The relative activity of intracellular antibodies LL3 and LL3.CDRAAA in a transfected HEK293T assay was analysed, with differences between the inhibition potential of the LL3 compared to LL3.CDRAAA and empty vector (Figure 6. 2 D).

LL3-scFc and LL3.CDRAAA-scFc coated beads were used for immunoprecipitation of cellular β-catenin from HEK293T bioassay cellular lysate, and specifically-bound proteins were eluted. The un-transfected cell lysate was included as a positive control for the detection of endogenous β-catenin and negative control for the VHH. Western blots for (A) β-catenin, detecting endogenous β-catenin and LL3, while negative for the mutated LL3; (B) VHH-scFc, positive for LL3 and its mutated form, while no detected in the cell lysate (Figure 6.3).

This human β-catenin VHH was received from Laura Newnham, UCB; propagated and sequenced upon receipt (Figure 6. 4 and Figure 6.5).

173

Figure 6. 2. β-catenin VHH and mutated control.

The relative expression and binding capacities of the β-catenin VHH and its mutated vertion were analysed against a vector control. LL3VHH and its mutated, non-binding form (A). ELISA test to assess the binding capacity of the negative control (B). Testing the transfection efficiency of the LL3VHH and it is mutated by analysing post-transfection cell lysis via western blot (C). The relative activity of intracellular antibodies LL3 and LL3.CDRAAA in the transfected HEK293T assay was analysed, by the inhibition of a wnt reporter gene with clear differences between the inhibition potential of the LL3 compared to LL3.CDRAAA and empty vector (D) (Newnham et al., 2015) .

174

Figure 6. 3. Binding capacities test of LL3-scFc and LL3.CDRAAA-scFc by immunoprecipitation of cellular β-catenin from HEK293T cells.

LL3-scFc and LL3.CDRAAA-scFc coated beads were used for immunoprecipitation of cellular β- catenin from HEK293T bioassay cellular lysate, and specifically bound protein was eluted. The un-transfected cell lysate was included as a positive control for the detection of endogenous β- catenin and negative control for the VHH. Western blots for β-catenin, detected endogenous β- catenin and LL3, while negative for the mutated LL3 (A); VHH-scFc, positive for the LL3 and its mutated form, while no detected in the cell lysate (B (Newnham et al., 2015).

175

Figure 6. 4. Native β-catenin VHH and control mutated form containing vector map. Vector map received from Laura Newnham, containing the human β-catenin VHH and control. The full vector sequence cannot be publicised for confidentiality reasons.

Figure 6. 5. Sequence alignments of native β-catenin VHH and control mutated form. Human β-catenin VHH sequence upon propagation, sequences were identical except the eight nucleotide changes that correspond to the amino acid changes deliberately made in the mutant protein.

176

6.2 Results

6.2.1 Sub-cloning β-catenin and mutated β-catenin VHH into a vector for expression in Xenopus

One of the advantages of VHHs, is that all of the coding region is encoded into a single exon, allowing the easy manipulation of the sequence. This way, the sequence can be sub-cloned into an expression vector which would allow it to be transcribed into RNA and this RNA later injected into embryos, which translate the protein from the inserted RNA

(see 2.6).

RNA transcription and later translation of the target protein requires sections of the vector to be specific. In this case, the pCS2+ vector’s backbone includes an SP6 promoter, which drives the transcription reaction; an SV40 late polyadenylation site and two polylinkers, between which the target sequence is inserted.

Primers were designed to amplify the β-catenin VHH sequence and add a 5’ BamHI site and a 3’ EcoRI site. Amplification of β-catenin and mutated β-catenin VHH were performed by PCR (Figure 6. 6 A; see 2.1.6) and purified on an agarose gel for ligation into pGemTeasy (see 2.1.4). DNA from single colonies was prepared and subjected to digestion with EcoRI to release the insert (see 2.1.1). Positive clones were sequenced

(Figure 6. 6 B). Two clones containing β-catenin VHH and two clones containing mutated

β-catenin VHH were double digested with BamHI and EcoRI, gel purified and then ligated into the pre-digested pCS2+ vector (see 2.1.3). Five colonies from each of the inserts were prepared and double digested with BamHI and EcoRI to release the insert; positive colonies were sent for sequencing (Figure 6. 6 C). Two β-catenin and three mutated β- catenin VHH sequences were correct. These clones were stored for further use.

177

Figure 6. 6. Cloning of native β-catenin VHH and control mutated form into pCS2+. Primers were designed to amplify the β-catenin VHH sequence and add a 5’ BamHI site and a 3’ EcoRI site. Amplification of β-catenin and mutated β-catenin VHH were performed by PCR (A), heated to 68ºC and loaded onto an agarose gel. The fragments were gel purified and ligated into pGemTeasy. DNA was extracted from single colonies, prepared and digested with EcoRI to release the insert (B). Positive clones were sequenced and double digested with EcoRI and BamHI before T4 DNA ligate into a pre-digested pCS2+ vector. The final plasmids were double digested with EcoRI and BamHI to release the sequence (C), and the five positive clones (arrowed) were sent for sequencing.

178

6.2.2 Can β-catenin VHH block gene activity in the frog?

Loss of function

The ability to block gene activity in the early embryo allows the elucidation of the importance of such genes in development. However, traditional methods have been challenged by recent developments in technology, such as single-cell sequencing and the ever-growing “omics” data pool (Blighe et al., 2018; Getwan & Lienkamp, 2017;

Packer & Trapnell, 2018), therefore new avenues need to be investigated.

Native β-catenin-VHH and mutated β-catenin-VHH were transcribed from the carrier pCS2+ vector with SP6 and the RNA analysed by agarose gel electrophoresis; four reactions were performed for each Figure 6.7 A). One-cell Xenopus embryos were microinjected with 200 pg, 500 pg and one ng of the RNA. These were allowed to develop to organogenesis (St 26) and analysed; 61 % of the defective embryos were expressing the native form of β-catenin-VHH. However, 37 % of embryos microinjected with RNA encoding the mutated β-catenin-VHH were also defective. The control, un-injected embryos were mostly normal, with only 2 % showing defects (Figure 6. 8).

These findings started us questioning whether the mutated β-catenin-VHH acts as a true non-binding negative control in Xenopus.

Previous research has suggested that targeted microinjections of β-catenin either into the dorsal or the ventral part of the 4-cell Xenopus embryo gave rise to different phenotypes (Itoh & Sokol, 1999) since the developmental pathways affected were different. Microinjection of two ventro-vegetal (VV) blastomeres or two dorso-vegetal

(DV) blastomeres at the 4-cell stage embryo (Figure 6.7 B) with 200 pg per injection of native β-catenin-VHH made the majority of the embryos dissociate (Figure 6.7 C).

Injections with the mutated control β-catenin-VHH caused numerous premature deaths and some similar phenotypes to that of the produced by the native version. Again, the effect observed by the microinjection of the mutated, control β-catenin-VHH were similar to those observed with the binding, non-mutated form rather than to the un-injected

179

control. Therefore, a further control experiment was designed to normalise the expression of the mutated control to that of the binding protein (Figure 6. 9). Injections were performed in batches of 25 embryos, and repeated 4 times to obtain this data.

180

Figure 6. 7. β-catenin loss of function experiments in Xenopus laevis embryos. Mutated and native β-catenin-VHH mRNA were synthesised (A) and microinjected into one-cell Xenopus embryos which were allowed to develop to stage 26 and screened. A considerable proportion of the embryos injected with the mutated, non-binding form of β-catenin-VHH also displayed abnormalities, while non-injected controls behaved as expected. Microinjection of two ventro-vegetal (VV) blastomears or two dorso-vegetal (DV) blastomears at the 4-cell stage embryo (C) with 200 pg per injection. Again, the effects observed from the microinjection of the mutated, control β-catenin-VHH were similar to those observed with the binding, non-mutated form rather than to the un-injected control. Injections were performed in batches of 25 embryos, and repeated 4 times to obtain this data. 181

Figure 6. 8. β-catenin-VHH RNA micro-injection into one-cell Xenopus laevis embryos. One-cell Xenopus embryos were microinjected with 200 pg, 500 pg and 1 ng of native β-catenin- VHH and mutated, non-binding form RNA. These were allowed to develop to tailbud stage (st 26) and analysed. Embryos expressing the native form of β-catenin-VHH had defects, however, embryos microinjected with RNA encoding the mutated β-catenin-VHH were also defective. The control, un-injected embryos were mostly normal.

182

Figure 6. 9. β-catenin-VHH RNA micro-injection into four-cell Xenopus laevis embryos. Xenopus laevis embryos microinjected with control mutated β-catenin and native β-catenin. 200 pg were injected into each of the two ventro-vegetal (VV) blastomeres or two dorso-vegetal (DV) blastomeres. Embryos injected with native β-catenin displayed cell dissociation and early death, as well as a considerable proportion of the embryos injected with the mutated non-binding form of β-catenin. Non-injected control embryos developed as expected.

183

6.2.3 Resourcing a true negative control - GFPVHH

The ability of the mutated β-catenin VHH to interfere in Xenopus development made validating the use of these VHHs as a way of blocking gene activity by RNA injection very difficult. The difference in severity and penetrance of native and mutant forms of β- catenin VHH makes it highly likely that the VHH can act as a functional blocking reagent but the effect of the mutant VHH suggests that it too can bind β-catenin.

A different VHH with no relation to β-catenin and no molecular effect in the wild-type early Xenopus embryo would serve as a true negative control. We chose the pOPINE-

GFP-VHH (Addgene #05162; (Kubala, Kovtun, Alexandrov, & Collins, 2010)) which was propagated and sequenced upon receipt (see 2.1.5). PCR primers were designed to amplify the nucleotide sequence encoding the VHH, and add specific restriction enzyme sites at the 3’ and 5’ ends Figure 6.10 A). The amplicons produced by PCR (Figure 6.10

B) were TA sub-cloned into pGemTeasy, single colonies DNA prepared, digested to release the insert and positive clone sequenced (Figure 6.10 C; see 2.1.1, 2.1.4, 2.1.6).

Only 1/22 tested colonies had an inserted sequence, far less than for other PCR products cloned this way previously. The target sequence was then separated from the backbone by double-digest and T4 DNA ligated into a pre-digested pCS2+ plasmid (see 2.1.3). The ligation was repeated multiple times, with different optimisation procedures such as time, concentration, temperature and cells used to transform with no positive outcome. To solve this cloning issue, primers for the In-fusion cloning kit were designed and will be tested as further development to this project (see 2.1.10).

184

Figure 6. 10. A true negative control, pOPINE-GFP-VHH, Addgene #05162A. PCR primers were designed to amplify the nucleotide sequence encoding the pOPINE-GFP-VHH and add specific restriction sites at the 3’, and 5’ ends (A). The amplicons produced by PCR (B) were TA sub-cloned into pGemTeasy, single colonies DNA prepared, digested to release the insert and positive clone sequenced (1/22 – C). The target sequence was then separated from the backbone by double-digest and T4 DNA ligated into a pre-digested pCS2+ plasmid. The ligation was repeated multiple times, with different optimisation procedures such as time, concentration, temperature and cells used to transform with no positive outcome (D). To solve this cloning issue, primers for the In-fusion cloning kit were designed and will be tested.

185

6.2.4 Expression of β-catenin VHH as periplasmic protein and its use as Antibody

One of the multiple advantages that heavy-chain-only antibodies have is that they can be sub-cloned into a vector and be expressed as periplasmic protein in bacterial cultures such as E. coli BL21 cells. The vector used in this study, pecan126, was a gift from Serge

Muylderman (Figure 6. 11, unpublished vector), which allows the translation of the encoded protein and adds a his-tag to its C-terminal. It contains the lacZ gene sequence, allowing easy identification of vectors carrying the insert by blue-white test, and the pelB sequence that allows the secretion of the protein into the bacterial periplasm.

Control, mutated β-catenin-VHH and native β-catenin-VHH were amplified by PCR with specific primers and sub-cloned into pGemTeasy (data not shown; see 2.1.4, 2.1.6).

Positive, sequenced clones were double digested with EcoRI and NotI (see 2.1.1), and ligated into a pre-digested pecan126 vector (see 2.1.3). Single colonies were DNA prepared and screened by double digestion with the same enzymes (Figure 6. 12 B). As the vector sequence was unknown, we designed primers from the insert into the vector

(Figure 6. 12 A), these sequences later allowed the design of primers from the vector into the inserted sequence.

The pecan126 vector containing the mutated, control and native forms of β-catenin-VHH sequence, were transformed into E. coli BL21 cells and plated on ampicillin selective plates (see 2.1.5, 2.3.1.1). In synthesis, single colonies were used to inoculate ampicillin supplemented 2xYT media and left to grow overnight at 37ºC. This overnight culture was used as a starter culture for large scale 2xYT medium and incubated until the OD600 was approximately 0.3-0.5. IPTG was added to a concentration of 1 mM and the culture incubated at 30ºC overnight to induce protein production. After the incubation period, cells were centrifuged, supernatant removed and the pellet collected. The cells were re- suspended in TES buffer and kept on ice for one hour. This incubation induces the release of proteins from the cell, followed by an osmotic shock to boost the proteins into

186

the media and incubated further. The protein was collected by centrifuging the cells and collecting the supernatant, then periplasmic proteins were stabilised by the addition of salt and stored at 4ºC for up to 1 month, or -20ºC for the longer term (see 2.3.2).

Periplasmic expression of the mutated control and binding forms of β-catenin-VHH was tested by an α-his-tag western blot (Figure 6. 13 A; see 2.4.2). Six colonies were chosen for expression from each sequence, and all 12 successfully expressed the his-tagged periplasmic protein. Large-scale preparation of one colony each was prepared and tested by western blot and Coomassie staining (Figure 6. 13 B). In the α-his-tag western blot, a single band can be observed of 18 kDa, corresponding to the expected size of the protein. While on the Coomassie stain, multiple bands are observed, but no significant background. Suggesting the protein is clean enough to use for staining in either western blot or immunohistochemistry.

The periplasmic produced VHHs were used as antibodies to detect endogenous proteins in Xenopus laevis embryo extracts via western blot. Whole embryo protein extractions

(see 2.4.1) were taken from embryos from a developmental course (st 2, st 6, st 10, st

15, st 21/22, st 33/34, st 37 and st 43) and separated through an SDS-PAGE. These were transferred to a membrane and one blotted with the native β-catenin-VHH, and the other with the mutated form of β-catenin-VHH (Figure 6.14). These were detected by using α-llama-HRP antibody and revealed a constant band at around 55 kDa which did not correlate to the size of β-catenin (93.8 kDa). This band could correlate to the endogenous β-catenin that might run at a different size, or to the mecp2 protein which the his antibody can sometimes bind to. The need for a true negative control became imminent.

187

Figure 6. 11. pecan126 vector map. The pecan126 vector allows the expression of the periplasmic protein in bacterial cultures such as E. coli BL21 cell. It contains a lacZ gene sequence, allowing easy identification of vectors carrying the insert by blue-white test, and the pelB sequence which allows the secretion of the protein into the bacterial periplasm. When the protein is translated, this vector allows the insertion of a his-tag to the protein’s C-terminal. A kind gift from Serge Muylderman (unpublished vector).

188

Figure 6. 12. Sequencing and cloning of native and control β-catenin-VHH into the pecan126 vector. Native β-catenin and control VHH sequences were amplified by PCR and sub-cloned into pGemTeasy (data not shown). Positive, sequenced clones were double digested with EcoRI and NotI, and ligated into a pre-digested pecan126 vector. Single colonies were DNA prepared and screened by double digestion with the same enzymes (B). As the vector sequence was unknown, we designed primers from the insert into the vector (A), these sequences later allowed the design of primers from the vector into the inserted sequence.

189

Figure 6. 13. Periplasmic expression of the mutated control and native forms of β-catenin- VHH. Six colonies from each construct were grown, and all 12 successfully expressed the his-tagged periplasmic protein. An α-his-tag western blot (A) illustrates two of the expressed clones that were up-scaled to large-scale preparation and tested by western blot and Coomassie staining (Figure 6. 13 B). In the α-his-tag western blot, a single band can be observed of 18 kDa, corresponding to the expected size of the protein. While on the Coomassie stain, multiple bands are observed, but no significant background. Suggesting the protein is clean enough to use for staining in either western blot or immunohistochemistry.

190

Figure 6. 14. Detection of Xenopus laevis endogenous β-catenin by antibody staining.

Proteins from wild-type embryos at stages 2-43 were extracted and separated by SDS-PAGE. These proteins were transferred into a membrane that was subjected to either non-binding control β-catenin or the native, binding form. An α-his antibody was used to locate the bound VHH and visualised by HRP detection. The continuous band detected at ~55 kDa does not correlate to the size of β-catenin (93.8 kDa). This band could correlate to mecp2 (467 AAs, 56 kDa) protein which the his antibody can sometimes bind to. The need for a true negative control became imminent.

191

6.2.5 Can single chain antibodies be used for in vivo labelling of β-catenin in Xenopus laevis?

Another advantage of single chain antibodies, is that they can be used to label proteins in vivo, as their reduced size and high specify allow them to penetrate and bind proteins within the cell. To investigate the possibility of using VHHs as in vivo imaging agents in

Xenopus, the β-catenin VHH and its mutated form were designed as fusions with fluorescent proteins.

Fluorescent proteins were first described in jellyfish in the 1950s (Davenport & Nicol,

1955). However it was not until much later that these proteins were cloned into an expression vector for the preparation of non-fluorescent apoGFP (Prasher, Eckenrode,

Ward, Prendergast, & Cormier, 1992). Since then, GFP has been used extensively in

Xenopus and other species to label many specific aspects of the anatomy (A. Gautier &

Tebo, 2018; Kuroda, Wessely, & De Robertis, 2004; Mishin, Belousov, Solntsev, &

Lukyanov, 2015; Preston, Aras, & Zaidi, 2018; Takagi et al., 2013). The main disadvantage with using GFP in Xenopus is that the yolk from the embryo auto fluoresces at the same wavelength that GFP is detectable, thus producing background in the imaging. New fluorescent proteins became available at the end of the 1990s, the Red fluorescent proteins (DsRed) (Matz et al., 1999). This protein was isolated from nonbioluminescent reef corals and led to the discovery of many new fluorescent proteins and chromoproteins. In 2001, it was published that proteins tagged with DsRed were often found to aggregate within the cell (Lauf, Lopez, & Falk, 2001) and by 2004 a whole range of monomeric units derived from the original DsRed were published (Figure 6.15

(Shaner et al., 2004)). The protein of choice in this study is on the red spectrum, katushka

RFP, which was discovered in 2007 by Andrey Zaraisky and colleagues (Shcherbo et al., 2007), and released by Evrogen together with a fluorescent protein colour palette which covers the entire visible range from blue to far-red (Day & Davidson, 2009). Upon encountering drawbacks in the cloning and experience with the fluorescent signal

192

emitted by katushka RFP as a fusion protein (personal comments from A. Abu-Daya and

M. J. Guille), a different derivate from DsRed, mCherry (Figure 6.15), was used in parallel

(Shaner et al., 2004). This approach enhanced the chances of success, and if both worked would allow a direct comparison of expression between these proteins.

193

Figure 6. 15. DsRed, one of six fluorescent proteins isolated by Matz and colleagues from non-bioluminescent reef corals in 1999 (Matz et al., 1999) and its derivate. Matz and colleagues used mRNA microinjection assays in Xenopus embryos and expression in mammalian cell culture to demostrate these proteins’ usefulness for in vivo labelling. However, proteins tagged with DsRed were often found to aggregate within the cells (Lauf et al., 2001) and efforts were made to generate derivatives of the DsRed protein that would have less or no side effect when fusing to a recombinant protein. Shaner et al., published in 2004 a variety of enhanced derivatives from DsRed with absorbance at different wavelengths and less or no effect when fused to another protein. GenBank accession numbers are given under the name of each (Shaner et al., 2004).

194

6.2.5.1 Cloning β-catenin VHH with KatushkaRFP in pCS2+

To create a KatushkaRFP-VHH fusion in pCS2+ (Figure 6.16 A), Katushka was extracted from a donor vector (Love et al., 2011) by amplification with primers containing an added sequence corresponding to EcoRI and SfiI (Figure 6.16 B; see 2.1.6). The amplicons obtained were cloned into pGemTeasy (see 2.1.4) and digested with EcoRI and SfiI (see

2.1.1). Bands of the expected size were sliced from the agarose gel and purified

(MinElute Gel Extraction Kit, Qiagen). These bands were ligated into a pre-digested pCS2+-VHH vector containing the mutated and non-mutated forms of β-catenin-VHH

(see 2.1.3). Single colonies were analysed by EcoRI-SfiI digestion of extracted DNA, but no positive clones were observed (Figure 6.16 C). The same DNA fragments were tested by amplification of the insert region when bands were observed for some of the clones

(Figure 6.16 D). The clones showing bands upon amplification by PCR were sequenced, and for ¾ of the sequenced clones, the VHH sequence was intact. However there was no KatushkaRFP sequence (Figure 6.16 D).

As result, a different approach was taken (Figure 6.17 A), in which the VHH encoding sequences and the katushkaRFP sequence were amplified by PCR (Figure 6.17 B; see

2.1.6), co-ligated into pGemTeasy (Figure 6.17 C; see 2.1.4) and, once the VHH- katuskaRFP sequence was linked, excised from the pGemTeasy vector (see 2.1.1) and ligated into a pre-digested pCS2+ vector (see 2.1.3). Single colonies were tested by

EcoRI digest, designed to yield two products, the backbone of 4110 bp and the whole

VHH-katushkaRFP cassette of 1209 bp. Upon analysing the EcoRI digested DNA, a band of only 700 bp could be seen (Figure 6.17 D), suggesting that only one insert was cloned into the vector.

195

500 1000

Bp 500 1000

Bp 500 Figure 6. 16. KatushkaRFP-VHH fusion in pCS2+. Cloning design (A) where the amplified katushkaRFP (B) was cloned into pGemTeasy, digested and purified before ligation into pre-digested pCS2+-VHH vectors. Single colonies were analysed, by preping the DNA from bacterial colonies and digesting them to release the insert, with no visible positive clones (C). These colonies were further analysed by amplification and plasmids displaying bands indicating the present of the insert (D) were sequenced. These plasmids contained the VHH sequence, but the KatushkaRFP was not present (E).

196

Figure 6. 17. An alternative approach to clone KatushkaRFP-VHH. Cloning design (A) were the VHH encoding sequences, and the katushkaRFP sequence was amplified (B), ligated into pGemTeasy (C), digested and ligated into a pre-digested pCS2+ vector. Single colonies were tested by EcoRI digest, with only a 700 bp band could be seen (D), suggesting that only one insert was incorporated to the backbone.

197

6.2.5.2 Cloning VHH with mCherry in pCS2+

The strategy to clone the mCherry sequence fused to the native and mutated β-catenin-

VHHs into a pCS2+ vector (Figure 6. 18) involved the amplification of such sequences with specific primers, and separately TA cloned them into pGemTeasy. The VHH encoding pGemTeasy vectors were digested (see 2.1.1), gel purified (MinElute Gel

Extraction Kit, Qiagen) and T4 DNA ligated into a pre-linearized mCherry-pGemTeasy vector (see 2.1.3). Once the native and mutated β-catenin-VHH-mCherry cassette sequence was confirmed in pGemTeasy, it was digested out and clone into a linearised pCS2+ vector.

The outcome of this cloning strategy was similar to that of the katushkaRFP cloning. Only the mCherry or the VHH sequences were present at the end of the cloning in the destination vector. To optimise this cloning, primers for the In-fusion cloning kit were designed and will be tested (see 2.1.10).

198

Figure 6. 18. Strategy to clone mCherry fused to native and mutated β-catenin-VHH into the pCS2+ vector. The VHHs and mCherry sequences were amplified with specific primers, and separately TA cloned into pGemTeasy. The VHH sequences were digested and purified before T4 DNA cloned into a pre-linearized mCherry-pGemTeasy vector. Once the entire cassette sequence was confirmed in pGemTeasy, it was digested out and clone into a linearised pCS2+ vector.

199

6.3 Discussion

The possibility to work with a tested and published VHH as a proof of principle provided a wide variety of opportunities that were not achievable with the libraries generated in

USA (discussed in chapter 3).

The differences between the human and Xenopus laevis β-catenin sequences is minimal, with high chances of VHHs raised against the human form cross-reacting with the Xenopus protein. These strictly controlled and commercially sensitive VHHs were sequenced and confirmed upon receipt. The cloning into pCS2+ and mRNA expression was successful and created a phenotype when injected. Such phenotypes can be directly linked to previously published β-catenin knock-downs and Wnt signalling pathways disruptions (as previously discussed in the Introduction).

The mutated version of β-catenin-VHH caused a higher than expected proportion of the embryos to develop abnormally compared to the un-injected. This suggests that this does bind to the endogenous β-catenin with lesser affinity than the native β-catenin-VHH.

To find a true-negative control, the GFPVHH was sourced from Addgene. The propagation of the original clone was successful, but all attempts to clone this sequence into an expression vector were unsuccessful, for unknown reasons. An In-Fusion reaction will be tried as an alternative cloning strategy.

The expression of the native and mutated forms of β-catenin provided a high, reproducible yield of relatively pure protein. These proteins were used as antibodies and did recognise an endogenous Xenopus protein. However, the size it runs did not correlate to the β-catenin protein and is almost certainly Mscp2.

The potential use of these VHHs as visualisation agents were not tested as the cloning of the VHHs with the KatushkaRFP and mCherry proteins were unsuccessful. However, strategies for further work have been established.

200

7 HA-tagging of the endogenous Gata2 locus in Xenopus laevis

7.1 Introduction

7.1.1 General Introduction

Protein detection via antibody binding is a powerful tool used broadly in science.

Nevertheless, it has its limitations. For example, proteins that are closely related (such as H2AZ1and H2AZ2 or H3.1 and H3.3) are sometimes difficult to differentiate and identify (Wen et al., 2014). The specificity barrier, poor accessibility of the protein to the antibody’s binding site or cross-reactivity of the antibody can decrease the use of conventional antibodies. Additionally, antibodies may not be the best approach when studying poorly characterised proteins (because exposed regions cannot be identified) or proteins with low immunogenicity. To follow these proteins, a different approach has to be taken; epitope tagging. This technique is particularly useful when discriminating between individual members of closely related proteins families or the identification of in vitro-mutagenized variants in the context of the endogenous wild-type protein (Fritze &

Anderson, 2000).

Epitope tagging refers to fusing a short amino acid sequence recognised by a pre- existing antibody with a known epitope (Brizzard, 2008). It was first developed as a method of visualising proteins, and the main advantage that this procedure offers is that the time and expense associated with generating and characterising antibodies against multiple proteins are obviated (Fritze & Anderson, 2000; Munro & Pelham, 1987). From the first tag ever reported, the Evans Blue T-1824 (Rawson, 1943), many different tags with a variety of properties have been used, usually with commercially available antibodies against them. A significant discovery was made when fluorescent tags were reported that allow the visualisation of proteins in tissue culture without the need of a secondary antibody (Tsien, 1998), and this has revolutionised many aspects of cell

201

biology (Ellenberg, Lippincott-Schwartz, & Presley, 1999). The tag chosen in this study is the HA tag, first described in 1983 (Niman et al., 1983), a sequence from the influenza virus hemagglutinin epitope (YPYDVPDYA). A table of the commonly used protein tags

(Li, 2010) is given in Table 7.1 and a graphic representation of the relative popularity of individual affinity tags according to a data search of PubMed Central (May 2018) in

Figure 7.1.

Table 7. 1. Commonly used protein tags (Li, 2010; Maue, 2007).

Affinity Tag Sequence or size (aa)

Protein A IgG-binding domain (aas)

his tag HHHHHH (aas)

FLAG DYKDDDDK (aas)

HA YPYDVPDYA (aas)

c-myc EQKLISEEDL (aas)

V5 GKPIPNPLLGLDST (aas)

StrepII WSHPQFEK (aas)

SBP MDEKTTGWRGGHVVEGLAGELEQLRARLEH HPQGQREP (aas)

S-peptide KETAAAKFERQHMDS (aas)

CBD TNPGVSAWQVNTAYTAGQLVTYNGKTYKCL QPHTSLAGWEPSNVPALWQLQ (aas)

GST 220 (aas)

MBP 396 (aas)

202

Figure 7. 1. Graphic representation of the relative popularity of individual affinity tags according to a data search of PubMed Central (May 2018).

203

A variety of approaches were used over the years to achieve the epitope tag labelling of proteins. Scientific studies evolved considerably from the first transgenic mouse, in which simian virus 40 (SV40) containing an origin of DNA replication and a gene product

(thymidine kinase (TK)) was inserted into the mouse genome in 1980 by Gordon et al.

(Gordon et al., 1980), to the numerous transgenic or mutant Xenopus lines today available (EXRC, 2016). The advances in technology and the elucidation of the Xenopus laevis genome have made it possible to perform the latest of gene editing techniques.

Although an artificial change to the genome was first reported in 1927 (Muller, 1927) when it was shown that the rate of mutagenesis could be enhanced with radiation or chemical treatment, the first targeted genomic changes were produced much later in

1979 (Scherer & Davis, 1979). Scherer and Davis proposed a system by which foreign sequences or deletions, constructed in vitro, could be introduced into a chromosome and be used to replace the wild-type chromosomal copy. In the early 1980s mouse embryonic stem cells (ES cells) became hugely popular due to the realisation that they could be used as a vehicle for introducing targeted genetic modifications into the germline by homologous recombination (reviewed by Solter, 2006), producing gene knock-out

(complete loss of function). The development of programmable site-specific nucleases, including zinc-finger nucleases (ZFNs) (Young & Harland, 2012), transcription activator- like effector nucleases (TALENs) (Lei et al., 2012), meganucleases (MNs) (Pan, Chen,

Loeber, Henningfeld, & Pieler, 2006), and most recently, the clustered, regularly interspaced, short palindromic repeats (CRISPR) associated proteins (including cas9)

(Nakayama et al., 2013) have greatly enabled and accelerated genome editing (Guha &

Edgell, 2017).

The CRISPR/cas9 technology was used in this study to tag a protein endogenously. This aimed to allow researchers to visualise and identify the protein by altering the genetic background of the frog whilst affecting the protein’s structure and function minimally. This approach was first described in 2016 (Armstrong et al., 2016) where a knock-in line was

204

generated by homology-directed repair (HDR) in the tardbp and fus genes in zebrafish

(Danio rerio), genes associated with amyotrophic lateral sclerosis (ALS). In summary, sgRNAs were designed to bind to the target sequence, guiding cas9 to make a double strand break and allowing the insertion of a donor single-stranded DNA molecule containing defined point mutations within the gRNA target site allowing the mimicking of human diseases, such as ALS.

The Xenopus embryo genome can be manipulated by gene editing (Lei et al., 2012;

Nakayama et al., 2013; Pan et al., 2006; Young & Harland, 2012), thus allowing researchers to insert, mutate or delete genes. One of the qualities of Xenopus laevis as a model organism is, that being allo-tetraploid, one copy of a particular gene can be manipulated without affecting the animal’s survival since the copy that remains generally allows the healthy development of the embryo. One advantage of having this copy number, is that the genes can be manipulated without compromising their function, as long as the animal is heterozygous for that gene. This is most useful in cell biology and basic function studies. A disadvantage to this system is that it challenges the feasibility of producing true disease models, as the analysis of function in disease would be shadowed by the extra copy of the gene in question.

205

7.1.2 Epitope tagging history and pitfalls The beginnings of epitope tagging the endogenous locus are found in the birth of biotechnology in the last century, which revolutionised biological sciences. These discoveries range from the discovery of zymotechnology and industrial fermentation

(Bud & Cantley, 1994) in early 1900, and the manipulation of an individual’s genetic structure to gene therapy (first attempted in 1980 by Martin Cline (Ermak, 2015)).

Particularly relevant to this approach are the accidental discovery of the underlying biology underpinning the ground-breaking technology CRISPR/CAS by Francisco Mojica in 1993 (Mojica et al., 2005). The knowledge of this bacterial “immune system” grew until the early 2000s when the sgRNA (Makarova, Grishin, Shabalina, Wolf, & Koonin, 2006) and CRISPR/Cas9 (Josiane E. Garneau, 2010) systems were adequately described.

Since 2013, the method developed from this biology has grown incredibly in popularity and can be applied to epitope tagging. Gene therapy human trials using this genome- editing technique have occurred since 2016 (Reardon, 2016), and with this, new variants of this approach are emerging.

Tagging the endogenous locus requires the insertion of a foreign sequence in the genome, providing a way of identifying and isolating proteins, but also jeopardising the ability of the altered protein to function in vivo. The alterations to the mRNA sequence can produce changes at the protein level such as the disruption of the functional domains or blockage of protein interaction surfaces, making the tagged gene non-functional.

Furthermore, the expression level and stability of the protein might deviate from the endogenous expression level, making the expression levels of the tag non-physiological

(Kanca, Bellen, & Schnorrer, 2017; Kolodziej & Young, 1991). Hence, functionality of the tagged proteins has to be tested.

The aim in this section of the project was to add an HA-tag directly, in frame, at the end of the gata2 gene using CRISPR-cas9 so that the expression of the gata2 gene can be followed in fixed sections and whole mount using an anti HA antibody.

206

7.1.3 GATA Binding Protein 2

In vertebrates, Gata binding protein 2, commonly known as Gata2, forms part of the gata family, six transcription factors involved in many aspects of embryonic development but associated primarily with hematopoietic development (de Pater et al., 2013). It binds as a multiprotein complex to regulatory elements (SCL/LYL1/LMO2/GATA2), and it also interacts with Runx1 (Wilson et al., 2010). There is a particularly prominent expression in early blood progenitors, as well as in megakaryocytes and mast cell lineages.

Furthermore, it is crucial for the proliferation and maintenance of hematopoietic stem cells/early progenitor cells (Vicente, Conchillo, García-Sánchez, & Odero, 2012). Gata2 germline deficiency results in embryonic lethality in mouse embryos and anaemic phenotype in Gata-/- stem cells (Tsai et al., 1994). In Xenopus, gata2 is expressed as a maternal transcript in the early embryo and abundantly from stage 11 onwards (Zon et al., 1991). As well as regulating the proliferation and maintenance of hematopoietic stem and progenitor cells (Moriguchi & Yamamoto, 2014), gata2 is involved in the Wnt signalling pathway by inhibiting the cannonical route and permiting progenitors to exit the cell cycle and commit to a hematopoietic fate. Gata2 has been associated with several human dissorders such as Early-Onset Coronary Artery Disease (Connelly et al.,

2006), Chronic Myeloid Leukemia (S.-J. Zhang et al., 2008), Immunodeficiency 21 (Hsu et al., 2011), Primary Lymphedema with Myelodysplasia (Ostergaard et al., 2011) and

Predisposition to Myelodysplastic Syndrome and Acute Myeloid Leukemia (Hahn et al.,

2011).

The most common gata2 mutations in humans occur within the highly conserved C- terminal coding region. Hsu et al. (2011) stated that the 480-amino acid Gata2 protein contains two zinc finger domains and a nuclear localisation signal in its C-terminal half.

The mutations can be divided into two main groups, those that allow the transcription of the mRNA into a non-functional protein (missense changes and deletions) or mutations which cause nonsense-mediated decay (NMD) of the mRNA (full gene deletions,

207

frameshift or early stop mutations) (Hsu et al., 2013). However, on fewer occasions mutations within the conserved noncoding and intronic regions have also been reported in patients, suggesting haploinsufficiency as the mechanism of Gata2 deficient disease

(Hsu et al., 2013).

The identification and visualisation of Gata2 in the early Xenopus embryo are of vital importance to validate and expand the knowledge obtained in Xenopus and mouse models.

208

7.1.4 Description of the target region

In this study, the 3’ end of the Xenopus gata2 gene was targeted (Figure 7. 2). Gata2 has been extensively studied in the Xenopus embryo as one of the major players in hematopoiesis (Walmsley et al., 1994) and other vital developmental processes (Zon et al., 1991). Antibodies against this Xenopus protein have not been reported (according to

Xenbase) despite numerous attempts to produce them by a variety of labs.

Gata2 is a developmentally essential protein, and it has been shown that on some occasions tagging can induce changes in protein surface expression or function, as well as its cellular localisation (Brothers, Janovick, & Conn, 2003). Xenopus laevis is allotetraploid, thus having four copies of many genes (Session et al., 2016) such as gata2 which allows the modification of one pair and the frog can still develop with the “healthy” copy of gata2. N-terminal endogenous tagging is not recommended as proteins tagged this way are prone to be overexpressed (Lafontaine & Tollervey, 1996) or mis-localised due to interference with the signal sequence (Gauss, Trautwein, Sommer, & Spang,

2005), although this is not true when looking at transcription factors.

209

Figure 7. 2. Xenopus laevis gata2 gene (long chromosome), where the target site, situated at the 3' end of the coding sequence, has been highlighted by an arrow. sgRNA were designed around the target site to guide cas9 protein to make a double strand break and allow the insertion of the single-stranded oligonucleotide containing the HA tag.

210

7.1.5 Experimental design

The aim of this chapter was to insert an in-frame HA tag at the end of Gata2 by CRISPR cas9 and homologous recombination, this will enable it to be followed through development.

To generate an HA-tag knock-in, it was first necessary to find a sgRNA that cuts the target site well. The efficiency with which the genome gets edited by CRISPR/cas9 depends on multiple variables, including the distance from the desired mutation point to the PAM site (Ward, 2015) and the guide sequence (Doench et al., 2016). In our hands, we saw that the sgRNA design has significant implications for the successful production of insertions and deletions (indels) in Xenopus embryos. The ease of production and manipulation of Xenopus laevis eggs allows making multiple injections at one point in time, providing the opportunity to test multiple sgRNAs in one set of injections and having sibling embryos. It has been established that starting with three different sgRNAs gives a good working sgRNA most of the time (M. Guille personal comments).

Upon micro-injection (see 2.6.1) with cas9 protein and each sgRNA, the embryos were left to develop to stage 12-16 and tested by T7 endonuclease assay to assess the cutting efficiency (see 7.1.6), which detects indels resulting from CRISPR-mediated non homologous end joining (NHEJ). The best cutting sgRNA was chosen and co-injected with cas9 and the ssDNA containing the HA tag insertion. Embryos were left to develop to stage 41 when tail-snips were taken, DNA was prepared and the presence of insert tested by PCR (see 2.1.6). Most of the injected embryos did not survive to this stage.

The sgRNAs were designed to affect only the gata2 L gene, having some sequence differences to the gata2 S gene. It is known that the alteration of gata2 is lethal (de Pater et al., 2013), thus if the embryos were affected in both alloalleles, these would not grow past gastrulation. Only unaffected or heterozygous embryos, with the insertion not causing significant disruptions to the protein, would make it past this stage.

211

The insert positive tadpoles were kept and grown to adults. Once they were identifiable by their markings, they were photographed and toe-clipped. This second test was performed to identify the individual frogs that have the insertion in frame with the Gata2 protein.

In this case, the vast natural selection via gata2 functionality and the ease of testing tadpoles in large numbers keeps the need for tanks low. As the proportion of embryos that, in our case, grew to adulthood was 0.4 %.

A flowchart of the methodology and graphic description of what happens in the embryo can be found in Figure 7. 3 and Figure 7. 4.

212

ssOligonucleotide co-microinjection of sgRNA Injection to designed according sgRNA design sgRNA and test if it cuts to the most efficient ssOligonucleotide sgRNA

use primer pair clone and does it go in? – if test batches of 5 going upstream sequence 10 of no, redesign. if yes, embryos at st15 from HA to identify each - to confirm micro-inject +ve pools in-frame insertions

make DNA and tail snip individuals PCR with primer grow embryos pool individuals to (x50) pair as above 10 lots of 5

as mature frogs, individualize +ve clone and keep +ve tadpoles photograph and toe pools and re-PCR sequence clipe

PCR individuals, clone and western blot blood test outcross sequence protein prep to outcross embryos by PCR of (to identify in-frame assess expression tail snips insertions)

if numbers of embryos are good, cull some for western blot and IHC to confirm successful and efficient tagging

Figure 7. 3. CRISPR/cas9 experimental design. From this experiment it was possible to design an effective strategy for this approach. A target region was selected according to previous knowledge about the protein of interest and multiple sgRNAs were designed to target this region. After the selection of the best cutting sgRNA, the ssDNA was designed to contain the sgRNA sequence (mutated to alter the nucleotide sequence, but not the amino acid sequence), complementary regions to the genome at both sides, and the inserted tag. The sgRNA, ssDNA and cas9 proteins were co-injected into one-cell embryos. Tadpoles were grown to stage 42 and tail clipped. Pools of the five tadpoles were tested, the positive pools then isolated and individual PCRs performed. The positive tadpoles were kept and grown to adulthood. As adults, the frogs were photographed, and toe clipped, DNA was prepared, and PCR was performed to identify the individuals with in-frame insertions.

213

Figure 7. 4. CRISPR/cas9 injection into the Xenopus laevis fertilised embryo. Cas9 protein, sgRNA, and ssDNA for insertion were injected into one-cell embryos and allowed to develop. The sgRNA guides cas9 to make a double strand break in a specific site in the genome, allowing the ssDNA containing homology arms at either side of the cut to bind and be incorporated the genome by the cell’s repair mechanisms; it cannot be re-cut since the ssDNA contains silent mutations in the sgRNA sequence (orange). The rate at which cas9 cuts differs from the rate to which the ssDNA gets “repaired” or incorporated into the genome, inducing mosaicism. Mosaic embryos have some cells with incorporated ssDNA, some cut by cas9 but without the ssDNA (probably displaying a variety of indels), and some not cut.

214

7.1.6 T7 endonuclease assay

A simple method has been developed to test if the cas9/sgRNA micro-injections successfully altered the target gene in co-injected embryos. In summary, an amplification of the target site from genomic DNA extracted from injected and control embryos are denatured and allowed to reanneal. Once annealed, T7 endonuclease is introduced into the mix and further incubated (Figure 7.5). The reaction products are then visualised via gel electrophoresis; homoduplexes are not cut by the enzyme and heteroduplexes are, a difference readily detected by gel electrophoresis.

215

Cas9 + sgRNA Denatured at 95C Microinjected F primer INDEL

R primer Genomic DNA PCR Amplification of target region Extracted Annealed

Homoduplex

Heteroduplex T7 Endonuclease cleaved Homoduplex heteroduplexes. Possible re-annealed products

Figure 7. 5. T7 endonuclease assay. Genomic DNA extracted from injected and control, un-injected embryos, was subjected to PCR with primers designed to amplify a 322bp fragment surrounding the target site. The amplified DNA was denatured and re-annealed allowing the native DNA to form heteroduplexes with indels when cutting with a T7 endonuclease to detect mismatches. Upon running the samples through an agarose gel, differences in size were seen between the wild-type DNA and the manipulated DNA. Higher proportions of cut fragments reflect more indels in the target site (Nakayama et al., 2013).

216

7.1.7 Cas9 RNA and protein

CRISPR/cas9 technology requires the co-injection of the cas9 enzyme with the sgRNA that targets it to the site to be cut. A variety of methods have been reported to work

(Figure 7. 6). The injection of cas9 and the sgRNA both as DNA, the injection of cas9 and sgRNA both as RNA, and more recently, the injection of sgRNA as RNA and cas9 as the stable protein (Thurtle-Schmidt & Lo, 2018). DNA injection is inefficient in embryos, and at the start of this project, the cas9 protein was not readily available.

Therefore we generated cas9 RNA.

217

Figure 7. 6. CRISPR/cas9 injection methods. sgRNA and cas9 have been reported to work when injected in different formats, DNA, RNA and protein (Thurtle-Schmidt & Lo, 2018). sgRNA and cas9 protein can be injected into the embryo as circular DNA plasmids (A), as RNA (B) or a combination of sgRNA as RNA and cas9 as a protein (C).

218

7.2 Results

7.2.1 gata2 3’ region sgRNA design

Three sgRNAs were designed to anneal to a particular sequence in the genome, in this case, the end of the gata2 gene. There was no precise target site on the 3’ UTR as the

NHEJ would repair the sequence and introduce the tag in frame. The cas9 was guided to make a double-stranded cut 3 bp away from the PAM site. The sequence of sgRNA that anneals to the target sequence are 20-25 bp long (uppercase), to the 5’ of this a promoter sequence was added (T7 in this case) to enable RNA synthesis. To the 3’ is an overlap to the remainder of the sgRNA encoding the external sequence that allows cas9 to bind (lowercase) (Figure 7. 7 and Figure 7. 8).

The sgRNAs were synthesised as described in Nakayama et al. 2014. In summary, the designed sgRNAs were annealed to a standard single-stranded sequence and extended with Taq polymerase to make a double-stranded DNA (dsDNA) template which was then transcribed with T7 RNA polymerase and visualised by agarose gel electrophoresis

(Figure 7. 9) (Takuya et al., 2014).

219

Figure 7. 7. sgRNA design to the 3' UTR of Xenopus laevis gata2. Sequences were comprised of the T7 promoter sequence, the target specific sequence, and an overlap sequence with a second oligonucleotide including the cas9 binding region.

Figure 7. 8. sgRNA sequences are targeting the gata2 C-terminus region. Three sgRNAs were designed (orange, pink and blue) to target the 3’ end of gata2 (dark red) or the 3’UTR (grey). Testing primers (purple), used in the T7 endonuclease assay were designed to amplify the target region (yellow). This region was also the target sequence subsequently used in CRISPRscan,

220

Figure 7. 9. sgRNA synthesis. The specific sgRNA encoding oligonucleotides were annealed to the universal ssOligonucleotide and extended with Taq polymerase to produce a dsDNA molecule (A) which was transcribed into RNA using T7 RNA polymerase (B). Two transcription reactions were performed for each sgRNA.

221

7.2.2 sgRNA efficiency determination by T7 endonuclease assay

To test the efficiency of the sgRNAs designed to guide cas9 to the 3’ end of gata2 (see

2.2), the region surrounding the cut site was amplified by PCR (Figure 7. 8; see 2.1.6) and subjected to digestion by the T7 endonuclease. One-cell Xenopus laevis embryos were co-injected with sgRNA and cas9 protein (NEB) (see 2.6.1). Three sgRNAs were tested, and cas9 was also injected on its own as a negative control.

Pools of 10 embryos were collected at stage 16, DNA was prepared, amplified and tested with T7 endonuclease (Figure 7. 10) as described in chapter 2. Amplicons subjected to

T7 endonuclease were visualised by agarose gel electrophoresis, but only one band of correct size for un-cut DNA was present. The absence of a second band corresponds to the absence of cutting site. Thus no genetic modification was achieved in the samples, the cas9 control behaved as expected. Single embryos were tested, and faint bands were observed when using sgRNA#3 (data not shown), suggesting that the assay was working. The incision rate was low, and not visible in a pool of 10 embryos (Figure 7.

10).

222

Figure 7. 10. None of the first 3 sgRNAs produced indels in injected embryos. One-cell Xenopus laevis embryos were co-injected with sgRNA and cas9 RNA. Three sgRNAs were tested, and cas9 was also injected on its own as a negative control. Embryos were grown to stage 16, gDNA was prepared and the target region amplified by PCR. The amplicons were denatured, reannealed and incubated with a T7 endonuclease to revealed mismatches. The products were separated by agarose gel electrophoresis and visualised by ethidium bromide staining under UV light. No miss-maches were observed with any of the sgRNAs.

223

7.2.3 Can Cas9 protein cutting efficiency compare with Cas9 mRNA injection?

The cas9 containing plasmid (a kind gift from Dr Marko Horb) was transformed and prepared upon receipt (see 2.1.5). The plasmid was linearised and transcribed with SP6

(Figure 7. 11 A and B; see 2.1.11). The use of cas9 mRNA worked well for the whole group before cas9 protein was available (data not shown). However the drawback of this method was the risks in RNA synthesis consistency. RNA has limited stability and we observed differences in performance from batch to batch. The accelerated growth of

CRISPR/cas9’s popularity allowed this inconvenience to be overcome by sourcing commercially available cas9 protein (NEB). Cas9 protein is stable at -20ºC and does not degrade as readily as RNA, moreover, the commercial quality assurance allowed the reproducibility of results in a more cost-effective manner. Nonetheless, we needed to compare the efficiency of the mRNA and protein directly.

A duplicate experiment was carried out where the commercial protein was tested alongside cas9 RNA injection (see 2.6). The cas9 RNA was co-injected with three different sgRNAs; embryos were incubated until they reached stage 16, DNA was extracted from them and tested by T7 endonuclease assay (see 7.1.6). This showed that the cas9 RNA used in the previous experiment worked well and so the problem was with the guide RNA design (Figure 7. 11 C).

224

Figure 7. 11. Cas9 mRNA preparation and test Vs cas9 protein (NEB). The cas9 containing a plasmid (a kind gift from Dr Marko Horb) was transformed into DH5α cells and prepared upon receipt (A). The plasmid was linearized with NotI, cleaned by chromatography and transcribed with SP6 polymerase (B). Two reactions were performed producing good quality RNA. Reaction #1 showed a higher band which corresponded to leftover DNA that was not successfully eliminated despite the treatment with DNAaseI. To test the comparative efficiency of this cas9 mRNA with the protein commertially available, a test experiment was performed (C). One-cell Xenopus embryos were injected with a control sgRNA (known to work), control cas9 mRNA (a gift from Dr Abu-Daya) and, the cas9 mRNA described here (A and B) and commercially available cas9 protein. There was no difference in the efficiency between the control cas9 mRNA, the cas9 mRNA made for these experiments and the commercial Cas9 protein (C).

225

7.2.4 New sgRNA design and production

It was therefore decided that a further sgRNA was needed to cut the gata2 target region.

Due to the increasing popularity of CRISPR/cas9 as a genetic engineering tool, and thus the investments aimed at making this research model more efficient and of widespread use, new tools or variants on this approach are regularly published. One of these advances was the development of an online predictive sgRNA-scoring algorithm that allowed the identification of potential sgRNAs that might be used. CRISPRscan was designed using data from Xenopus considering the molecular features that influence sgRNA stability, activity, and loading into cas9 in vivo (Moreno-Mateos et al., 2015). A fourth sgRNA (sgRNA #4) was thus designed utilising CRISPRscan (Figure 7. 12).

At this point, it is interesting to note that sgRNA #3, designed by hand, and sgRNA #4, designed in silico, cover the same sequence. However, sgRNA #4 was 3 bp shorter and had its second nucleotide changed from a T to a G (Figure 7. 13). sgRNA #3 injected embryos previously showed very faint bands when subjected to T7 endonuclease assay (data not shown). Therefore, sgRNA#3 was transcribed as a new, fresh batch when sgRNA#4 was made to minimize the differences between them during subsequent testing.

226

Figure 7. 12. CRISPRscan sgRNA suggestions for gata2 3' end target sequence. The 197 bp target sequence, composed of the 3’ end of the gata2 ORF and the beginning of the 3’UTR of gata2 was submitted to CRISPRscan as a template. Four sgRNAs were suggested. The first suggested sequence in the list was chosen.

Figure 7. 13 Sequence comparison of sgRNA #3 designed by hand against sgRNA #4 designed by CRISPRscan. Note the sequences cover the same region. However, CRISPRscan alters one base pair (from T to G) and is slightly shorter. The differences are shown in red above.

227

7.2.5 Can sgRNA #4 successfully guide cas9 to the cut site?

DNA from one-cell embryos micro-injected with cas9 protein and sgRNA #4 was extracted (Qiagen DNeasy Blood & Tissue Kit), amplified with target-specific primers

(see 2.1.6) and subjected to T7 endonuclease assay as previously described (see 7.1.6).

Embryos were injected with cas9 protein on its own as a negative control.

The sgRNA#3 and sgRNA#4 exhibit consistent cutting. However, sgRNA#4 was chosen to work with since the stronger lower band on the gel suggests a more efficient indel formation at the target DNA (Figure 7. 14).

228

Figure 7. 14. T7 endonuclease assay of embryos co-injected with cas9 protein and sgRNA #3 and sgRNA#4. Embryos were co-injected at the one-cell stage with the gRNA shown and Ca9 protein, and allowed to grow to stage 16. Embryos were collected, DNA extracted, and the target region was amplified and reannealed. T7 endonuclease was added to the amplified target sequence and incubated before running on a 1 % agarose gel. Cas9 protein was injected on its own as a negative control. Both sgRNA #3 and sgRNA #4 cut the amplified sequence, sgRNA#4 doing so in a significantly more efficient manner. Each lane represents a single embryo.

229

7.2.6 Identifying the mutations made by sgRNA #4 and cas9

To assess the efficiency and specificity of the mutagenesis by CRISP/cas9, the target

PCR products were sub-cloned into pGemTeasy and sequenced (see 2.1.4, 2.1.6).

Eighty-one percent of the sequences obtained showed deletions around the desired cut site. It also showed some random insertions in 6 % of the sequenced clones. All observed changes are shown in Figure 7. 15, while Figure 7. 16 shows the deletions in more detail.

230

Figure 7. 15. Target site amplicons were sub-cloned into pGemTeasy and sequenced. One-cell Xenopus laevis embryos were microinjected with cas9 protein and sgRNA#4. Embryos were allowed to develop to stage 16, DNA extracted from them and PCR amplified with specific primers. The PCR products were sub-cloned into pGemTeasy and sequenced. Deletions and random insertions were observed around the targeted cut site (enlarged in Figure 7.16). The large insertion on the 5th sequence was identified by BLAST against the Xenopus genome as a section of the gata2 gene outside the target region.

231

Figure 7. 16. Gata2 target site amplicons cloned into pGemTeasy. Three distinct populations can be observed after the co-injection with cas9 and sgRNA; one with 5 deletions, one with 11 bp deletions and one with 12 bp deletions. In only one of the sequences were 2 inserted nucleotides observed. There are 3/16 uncut sequences.

232

7.2.7 Design of the insertion sequence for HA-tagging of gata2

Once the sgRNA to be used had been identified, a single-stranded oligonucleotide

(ssOligonucleotide) was designed to insert an HA epitope tag into the 3’ UTR of gata2 in

Xenopus laevis. The design consisted of 20 base complementary sequence at the 3’ and

5’ ends of the oligonucleotide, to guide the correct targeting of the ssOligonucleotide.

The sequence corresponding to sgRNA sequence in the ssOligonucleotide was mutated, so the nucleotide sequence changed, but the amino acid sequence was the same. These nucleotide changes allowed the sgRNA to make an incision on the wild-type DNA, but once the oligonucleotide was inserted, the sgRNA would not recognise the mutated sequence as a target any longer. Following the sgRNA sequence, an HA tag sequence was designed in frame with the gata2 ORF. Straight after the HA tag, a new stop codon was inserted, followed by a BamHI site created for the identification of embryos with the successful insert (Figure 7.17).

233

n HA tag

ends ends of the

(blue)

so the sgRNA would notsgRNA the would so

till till be the same. A

(purple)

altered

DNA DNA at the 3’ and 5’

was

endogenous

the nucleotide sequencethenucleotide

complementary to the

encoded was not changed, so the protein would s

sequence

sequence(orange) where

with 20

sgRNA

the amino acid sequence

(orange)

DNA design

and and BamHI sequence

Insert

(pink)

DNA design, comprised of the mutated the ofcompriseddesign, DNA

cut the inserted ssOligonucleotide, but

Figure 7. Figure7.

Insert re sequence oligonucleotide.

234

7.2.8 Can an HA tag be inserted into the endogenous gata2 C-terminus?

Embryos were co-injected with cas9 protein, sgRNA#4 and the HA tag-containing DNA

(see 2.6.1). Three amounts of the insert DNA were tested: 50 pg, 100 pg and 200 pg.

The majority of embryos (~90 %) injected with 200 pg of the insertion DNA did not develop after gastrulation and were mostly deformed. When injected with 100 pg and 50 pg of insert DNA, embryos had a higher rate of successful gastrulation (~20 %), developing less frequent deformities. Most of the deaths occurred between stages 8 and

12, correlating with the onset expression of gata2 in the embryo (Figure 7. 18). Therefore, injections were performed with 50 pg and 100 pg of insert DNA thereafter. From over

6000 injected embryos, only 168 grew to the tail-bud stage from which 27 tested positive for the insertion (16 %). Tadpoles were tested by PCR of the extracted genomic DNA

(Qiagen DNeasy Blood & Tissue Kit) and in some cases cloned into pGemTeasy for sequencing (see 2.1.4, 2.1.6). The sequences showed a variety of in-frame and out-of- frame insertions. The tadpoles at these stages (between stage 30 and 42) could not be kept separately, due to welfare and practical considerations, for long enough to be able to extract the DNA, PCR, sub-clone and sequence. Thus they were mixed and left to grow to adults. Only 15 frogs developed to adulthood, which could then be identified by photographing the patterns on their backs allowing for the subsequent identification of the in-frame mutation-containing animals.

235

Figure 7. 18. Xenopus laevis gata2 RNA-Seq expression data. Expression of gata2 starts at stage 8-9, with the highest steady-state RNA levels at stage 12. Over 90 % of embryos co-injected with sgRNA and cas9 did not survive past gastrulation (stages 10.5-12). This could be due to different reasons. It is good practice to account for the possibility of DNA toxicity and its effects on the embryos. Therefore control embryos were injected with either the sgRNA or the ssOligonucleotide alone. These embryos developed normally, as did the un- injected controls. When co-injecting cas9 and sgRNA, without the ssOligonucleotide, embryos exhibit a higher rate of mortality compared to uninjected controls. This may indicate that the incisions made by the activity of cas9 can produce off-target lethal effects and that the gata2 protein is mutated and becomes non-functional.

236

7.2.9 Detecting and characterizing HA sequence insertions at the gata2 locus

Embryos injected with sgRNA, Cas9 protein and the HA tag insertion sequence were allowed to develop to stage 16-20, tail snipped, DNA extracted (Qiagen DNeasy Blood

& Tissue Kit) and tested for the insertion of the HA tag. Testing was done by PCR (see

2.1.6) and amplicon digestion with BamHI (see 2.1.1), a site designed following the HA tag on the insertion DNA (highlighted in orange in Figure 7.19).

Amplicons produced by amplification of the insertion site and surrounding genomic DNA subjected to BamHI digestion were expected to yield three products. An un-cut fragment corresponding to the wild-type, non-mutated DNA, and two smaller fragment corresponding to the digested inserted DNA fragment. This approach was unsuccessful

(data not shown). This may be due to the mosaicism between cells from which the DNA was extracted and suspected low insertion rate of the DNA in the fast-dividing early embryo or due to a complete failure of insertion.

To bias the amplification of the insertion positive cells, two sets of primers were designed that promoted the amplification of DNA samples containing the inserted DNA. These primers amplified two fragments. One PCR reaction amplified the 5’ end of the insert and the other set the 3’ end of the sequence. The primers were designed so the products would also contain the junction between the insert and the genomic DNA, allowing the identification of frogs containing an in-frame insertion and those in which the insertion is out of frame (Figure 7.19 and Figure 7.20).

Upon positive PCR amplification, samples were cloned into pGemTeasy as described in the methods section (see 2.1.4, 2.1.6). Multiple colonies from each plate were analysed, since the mosaicism of the embryo would be represented in the amplified DNA. The insertion rate was 11.3 %, with some fragments inserted correctly and some fragments out-of-frame (Figures 7.21, 7.22 and 7.23).

237

the

rom the

Amplification of the target

regions (light blue) and primers used to amplify the

sequence.

This approach was not successful. We hypothesise that

target

complementary

by by agarose gel electrophoresis.

primer, subjected to digestion with BamHI would cut the target sequence containing the insertion intoinsertion containingthe sequence target the cut BamHIwould with digestion to subjectedprimer,

, displaying , primers, insert and the

visuali

3' 3' end

3' 3' end, indicating the gRNA #4 (orange), the HA tag (pink),

Xenopuslaevis gata2

F0 animals subjected to PCR reduced the possibility of successfully visualising the digested product. Two primers extending f

. .

Figure 7. Figure7.

Xenopus laevis gata2 insert (purple). Upstream of the HA tag, a BamHI site was introduced to allow the identification of positive clones (orange). Reverse 3’ primer and Forward 5’ the with region two smaller bands which could later be mosaicism in tag HA weredesigned to selectively amplify the HA insertion (5’ Reverse and 3’ Forwards).

238

Figure 7. 20. Cas9 protein (NEB) and sgRNA #4 injected embryos. Embryos were co-injected at the one-cell stage with sgRNA #4 and cas9 protein, and allowed to develop to stage 40. DNA was extracted from tail clips, target sequence amplified by PCR and visualised by agarose gel electrophoresis. Cas9 protein was injected on its own as a no DNA and no sgRNA negative control. Each lane represents a single embryo were 7/10 tested positive for an insertion. The PCRs were designed to enrich for the positive cells (Figure 7.19), as the primers were designed to amplify the insert from the HA tag to the genomic sequence. The cas9 protein injection gave no signal, while the positive control (a previously analysed sample) also behaved as expected.

239

are are

gata2

' end of the insert insert the ofend '

20, tail clipped and DNA

were grown to stage 16

natureof these outbred animals(B).

frame HA HA frame insertionon the end of 3’ the endogenous

tothe polymorphic

When analysingclosely sequence the of fourfrogs, be can positive noted it that

containingin

(A)

beattributed

embryos

cloned, sequenced and a representativesequencedcloned,andalignment a sequencesfrom of the

in two in of sequenc the

Xenopuslaevis

injected with cas9 protein, ssOligonucleotide and sgRNA#

injected

be observed be

cell embryos co

one

Sequenced co

Mutations can

here.

e e ofsequences these showed point 2 mutations that can

Xenopus laevis regiontargetsub was amplicon extracted.the An of shown on Figure 7. Figure7.

240

20,tail clipped and DNA

in two of the sequences,

be observed

weregrown to stage 16

Point mutations can

and sgRNA# and

h h oforiginal the amplicon, probably dueto degradation of the amplifiedproduct

ssOligonucleotide

cloned,sequenced anda representative alignment of sequences from 5' the end of the insert

injected cas9 with protein,

morphisms the between outbredanimals.

injectedembryos, 5’ end.

topoly

cell embryosco

one

. . Sequenced co

here.Some sequences of do these not extend the lengt

beattributed

Xenopus laevis extracted. Anamplicon of the target region sub was shown are before cloning or to the repetitive nature of the sequence that challenges sequencing. can which Figure 7. Figure7.

241

cloning or the

be be attributed

e e

intwo of the sequences, can which

be be observed

shownto contain the tag HA

frogs

amplificationsequence of five

detailed

ntative 5’

Represe

Figure 7. Figure7.

Only Only 2/5 sequences show the whole extent of the PCR amplification, probably due to degradation of the amplified product befor repetitive natureof the sequence that challenges sequencing. mutations Point can polymorphismsbetween the outbred animals

242

7.3 Discussion

This study aimed to endogenously tag the gata2 locus so that the Gata2 protein could be visualised and isolated. To do this, the 3’ end of the endogenous Xenopus laevis gata2 was successfully tagged; an HA tag was inserted via CRISPR/cas9 technology and tested by PCR and sequencing. We do not know exactly how efficient this process was, although from one experiment of around 10000 injections, only 106 froglets survived, and 12 displayed the correct size band when analysed by T7 endonuclease.

This is an 11.3% insertion rate, from which only 8 grown to adulthood. The in frame insertion and germ line transmission efficiency of these frogs still has to be tested.

The method employed in this study required very high numbers of micro-injections, DNA extractions, PCR and sequencing, and due to husbandry requirements tadpoles could not be in isolation for a long enough time to distinguish individuals with in-frame insertions from those out-of-frame. Nevertheless, we hypothesise that biology has acted to favour this study by naturally selecting heterozygous embryos with non-lethal genetic modifications to survive. Also, the percentage of embryos that survived suggest that the combination of cas9 and sgRNA induced either very efficient molecular changes in the gata2 gene, making most of these embryos homozygous for mutated gata2 (lethal to the embryos if non-functional), or that cas9 and sgRNA modified other parts of the genome causing lethality. More so, there are controversies regarding epitope tagging of the endogenous locus, as is believed that the use of such tags might not accurately reflect the cellular localization of modified proteins (Brothers et al., 2003), if true in this study, it might contribute to the high mortality rate at the early stages.

Towards the end of this study, a potentially revolutionary paper was published by Aslan et al. (Aslan, Tadjuidje, Zorn, & Cha, 2017) which describes, as the title of the article suggests, a “High-efficiency non-mosaic CRISPR-mediated knock-in and indel mutation in F0 Xenopus”. This report described the success of a method by which oocytes were injected with the cas9, sgRNA, and ssDNA and incubated with DNA ligase IV inhibitor to

243

promote homology-directed repair (HDR) over nonhomologous end joining (NHEJ). This method is advantageous as the rate of insertion is increased, the chances of re-editing the already inserted change are decreased, and in consequence, the embryos were not mosaic. The oocytes were incubated for three days after CRISPR/cas9 co-injection, which gives time for the cas9 and sgRNA to degrade, thus when the oocyte is matured and fertilised, the embryo can develop without the risk of cas9 altering the genome. The main drawbacks of the method were technical as the rate of survival after sperm nuclear transfer are very low. These type of techniques, to encourage HDR-induced point mutations are quickly evolving also in zebrafish (Y. Zhang, Zhang, & Ge, 2018) where the focus also was to increase the germline transmission and establish lines.

In the experiments described in this thesis, fifteen positive frogs grew to adulthood.

These frogs need to be tested to discover if the insertion is in-frame. Following this, the frogs need to be outbred to test germline transmission. Blood from the frogs can also be used in western blots to determine whether the tagging has been useful in providing a model in which to study the Gata2 protein, something which has not so far been possible.

These frogs are being held at the EXRC, and studies are underway to analyse them, however, breeding and analysis of the offspring will take over a year.

244

8 General discussion and further work

Scientific knowledge is an accumulative process, where new studies either confirm, enhance or reform previously obtained knowledge. Often, scientists re-visit and new studies occur as new techniques develop and inter-disciplinary strategies are incorporated and associated with a research area. In recent years has become evident that many aspects of biology, such as developmental, genetic and biochemical studies require the understanding of proteins, their function, structure and localisation, and for some studies, at a specific time point in development.

By stage 33 a Xenopus tadpole has formed blood, heart tube and muscles, the kidney is differentiating, the eye is partially formed and it has the ability to swim away from stimuli.

Despite all this, at stage 33 fewer than half of its proteins present in the Xenopus embryo were generated from zygotic transcripts (Peshkin et al., 2015) and not from embryonic translation. The critical need to be able to study the levels, control and function of proteins, in addition to RNA in the early embryo is therefore clear.

With limited number of commercially available Xenopus-specific reagents, the poor cross-reactivity of reagents raised against other models with Xenopus, and the revelling knowledge about the proteins’ origin, the importance of obtaining a well characterized pool of Xenopus-specific reagents became clear. To address this growing issue, the

EXRC was founded by BBSRC to generate and supply the community with Xenopus- specific reagents, specially plasmids and antibodies. During this project, we worked closely with the community through the EXRC’s network to obtain a list of target proteins of significant interest and we had invaluable support and experience provided by the

EXRC’s Strategy Board team, a team of 10 – 12 experience Xenopus researchers.

From the experience shared by the Strategy board members, which included Prof E

Jones, and the literature at the time, the production of single chain antibodies was suggested as a good way forwards. Importantly, VHH libraries have shown to have a

245

very profound impact on the fields where they have been applied. The popularity of VHHs and the chance to collaborate with the Sokols’ laboratory which facilitated the training and knowledge to work with VHHs made this approach possible. Upon screening the libraries obtain from the Sokols’ lab, it was found that the complexity of the libraries had a significant effect on the ease of screening. These libraries generated many positive results when screened against VHHs. However, there were no Xenopus-specific VHHs found in our hands. At Sokols’ lab, a VHH against a Xenopus yolk protein was successfully isolated, highlighting, even more, the importance of the purity of the immunogens used.

The generation of pure proteins for immunisation became our priority. By the use of the1-

Step Human High-Yield Mini IVT Kit, proteins were successfully expressed. However the small yield and price per reaction at the time, suggested we looked at other approaches.

Since the time this method was used (2014), the price of this experiments has been reduced in such way that if this approach could now be re-visited.

A more efficient way of protein production is the established method of protein production via transfection of a plasmid into HEK cells, which became a suitable option. This method provides with large amounts of proteins that easily purify when secreted to the cell’s media. By the use of the In-Fusion system, we were able to express proteins which emitted a strong signal when tested by western blot and Coomassie stain. This method was promising. However, the proteins we worked with were mainly transcription factors and DNA binding proteins, which were found to be extremely difficult to solubilise and purify. We therefore decided to investigate if commercial immunogen production would be an option.

Twenty-six protein sequences were sent to commercial companies for expression or synthesis. No companies were willing to express any of the suggested targets in mammalian systems; they suggested bacterial expression, however that was not an option in this project as any carryover bacterial residue would be very immunogenic in a

246

mammalian host, potentially becoming the prominent VHH found in the library. Other systems were also available such as insect cell expression, but we did not consider any non-mammalian systems.

The synthesis of peptides was investigated, were two companies were chosen to work with. Alta Bioscience provided with octomeric MAPs, which are recognised as a good option when choosing immunogens that are not coupled to any carrier protein. The other company was ThermoFisher, which provided us with long peptides of up to 120 AAs.

The decision to don’t couple the proteins to a carrier was also made considering the possibility of the carrier protein producing an immune response.

Once the targets were chosen, and the most immunogenic but unique peptide sequence suggested (by the commercial companies), two llamas were successfully immunised.

Valentine was immunized with a Cdk1 octomeric MAP and the Sunny with an Ascl1 long peptide. A pre-bleed and two immune-bleeds were taken, which show an excellent immune response. The library making and VHH isolation from the stored lymphocytes remains to be done due to an 18-month delay caused by TB at the llama’s farm.

For a direct comparison of the protocols to make llama and conventional antibodies, a sheep was immunized with the Kash5 octomeric MAP. Ideally we would have used the same immunogen in a llama and a ship, but due to the direct link between this project and the needs of the community, a Kash5 antibody became a priority. The sheep immunization was successful, where a clear response is present from the first bleed when tested against an MBP-tag kash5 protein, apparently decreasing in bleed 2 and 3.

These results need to be confirmed with further analysis of this conventional antibody, the detection of the endogenous Xenopus protein has to be optimised, as the detection is not easily reproducible.

Lastly, the successful introduction of an HA-tag into the endogenous Xenopus laevis gata2 provides a different way of visualising proteins in the early embryo. The fact that the tagging is carried out at a genetic level, allows the targeting of otherwise challenging 247

to differentiate proteins, such as close members of the same family. The efficiency by which this tag was inserted was low, however the number of possitive animals needed to establish a line is one. New techniques are being developed that potentially allow the insertion of this sort of tag more efficiently. These techniques have been published, but could not be replicated in-house to date, despite more than a year’s work on them.

F1 embryos produced from crossing F0 gata2-HA with wild-type Xenopus laevis are needed to corroborate the germline transition of the insertion over generations, and these will be available in 2019. These embryos then have to be tested via immunohistochemistry to confirm the visualisation of the HA-tag on fixed embryos.

This project contributed to the Xenopus community’s general knowledge about antibody- making and protein visualisation and detection in embryos. The work involved analysing the mixed protein VHH libraries and the methodologies for protein production, although it did not provide positive results to date, it still provided knowledge about these processes and how to approach them in further experiments.

The experience obtained while on this doctorate course, encouraged the evaluation of the methods used. Primarily, if I was to do this again (and not be restricted by grant related expectations), I will choose proteins that have been previously characterized. I will also reduce the number of proteins I worked with, and have the same protein raised in different media and used to immunize a variety of hosts, as this provides true direct comparisons. I will also take extra time purifying and checking the quality of the antigen, and when raising the antigen is not feasible, to compare the octomeric MAP and long peptide efficiency as immunogen side by side.

Even with all the refinement that these methods need, we achieved the successful immunisation of llamas and sheep, and also provided with material for further studies and the successful insertion of the HA-tag in the endogenous gata2 was something without precedent.

248

9 References

Abu-Daya, A., Khokha, M. K., & Zimmerman, L. B. (2012). The Hitchhiker’s guide to

Xenopus genetics. Genesis, 50(3), 164–175. http://doi.org/10.1002/dvg.22007

Abu-Daya, A., Nishimoto, S., Fairclough, L., Mohun, T. J., Logan, M. P. O., &

Zimmerman, L. B. (2011). The secreted integrin ligand nephronectin is necessary

for forelimb formation in Xenopus tropicalis. Developmental Biology, 349(2), 204–

212. http://doi.org/10.1016/j.ydbio.2010.10.015

Abu-Daya, A., Steer, W. M., Trollope, A. F., Friedeberg, C. E., Patient, R. K., Thorne, A.

W., & Guille, M. J. (2005). Zygotic nucleosome assembly protein-like 1 has a

specific, non-cell autonomous role in hematopoiesis. Blood, 106(2), 514–520.

http://doi.org/10.1182/blood-2005-02-0598

Aimar, C., Vilain, C., & Delarue, M. (1983). Regulation of cell cycle by cytoplasmic

components in the amphibian eggs. Cell Differentiation, 13(4), 293–300.

http://doi.org/10.1016/0045-6039(83)90039-8

Allan, V. J., & Vale, R. D. (1991). Cell cycle control of microtubule-based membrane

transport and tubule formation in vitro. Journal of Cell Biology, 113(2), 347–359.

http://doi.org/10.1083/jcb.113.2.347

Altmann, C. R., Bell, E., Sczyrba, A., Pun, J., Bekiranov, S., Gaasterland, T., & Brivanlou,

A. H. (2001). Microarray-based analysis of early development in Xenopus laevis.

Developmental Biology, 236(1), 64–75. http://doi.org/10.1006/dbio.2001.0298

Altschul, S. F., Gish, W., Miller, W., Myers, E. W., & Lipman, D. J. (1990). Basic local

alignment search tool. Journal of Molecular Biology, 215(3), 403–410.

http://doi.org/10.1016/S0022-2836(05)80360-2

Arias, J. L., Unciti-Broceta, J. D., Maceira, J., Del Castillo, T., Hernandez-Quero, J.,

Magez, S., … Garcia-Salcedo, J. A. (2014). Nanobody conjugated PLGA

249

nanoparticles for active targeting of African Trypanosomiasis. Journal of Controlled

Release, 197, 190–198. http://doi.org/10.1016/j.jconrel.2014.11.002

Armstrong, G. A. B., Liao, M., You, Z., Lissouba, A., Chen, B. E., & Drapeau, P. (2016).

Homology directed knockin of point mutations in the zebrafish tardbp and fus genes

in ALS using the CRISPR/Cas9 system. PLoS ONE, 11(3), 1–10.

http://doi.org/10.1371/journal.pone.0150188

Aslan, Y., Tadjuidje, E., Zorn, A. M., & Cha, S.-W. (2017). High efficiency non-mosaic

CRISPR mediated knock-in and mutations in F0 Xenopus. Development, 144(15),

dev.152967. http://doi.org/10.1242/dev.152967

Balakrishna, S., Saravia, J., Thevenot, P., Ahlert, T., Lominiki, S., Dellinger, B., &

Cormier, S. A. (2011). Environmentally persistent free radicals induce airway

hyperresponsiveness in neonatal rat lungs. Particle and Fibre Toxicology, 8(7),

1693–1704. http://doi.org/10.1186/1743-8977-8-11

Bandaranayake, A. D., & Almo, S. C. (2014). Recent advances in mammalian protein

production. FEBS Letters, 588(2), 253–260.

http://doi.org/10.1016/j.febslet.2013.11.035

Bao, J., Talmage, D. A., Role, L. W., & Gautier, J. (2000). Regulation of neurogenesis

by interactions between HEN1 and neuronal LMO proteins. Development

(Cambridge, England), 127, 425–435.

Bateman, A., Martin, M. J., O’Donovan, C., Magrane, M., Alpi, E., Antunes, R., … Zhang,

J. (2017). UniProt: The universal protein knowledgebase. Nucleic Acids Research,

45(D1), D158–D169. http://doi.org/10.1093/nar/gkw1099

Beck, C. W., & Slack, J. M. W. (2001). An amphibian with ambition: a new role for

Xenopus in the 21st century. Genome Biology, 2(10), REVIEWS1029.

http://doi.org/10.1186/gb-2001-2-10-reviews1029

Beghein, E., & Gettemans, J. (2017). Nanobody technology: A versatile toolkit for

250

microscopic imaging, protein-protein interaction analysis, and protein function

exploration. Frontiers in Immunology, 8(JUL), 1–14.

http://doi.org/10.3389/fimmu.2017.00771

Berrow, N. S., Alderton, D., Sainsbury, S., Nettleship, J. E., Assenberg, R., Rahman, N.,

… Owens, R. J. (2007). A versatile ligation-independent cloning method suitable for

high-throughput expression screening applications. Nucleic Acids Research, 35(6).

http://doi.org/10.1093/nar/gkm047

Bertier, L., Boucherie, C., Zwaenepoel, O., Vanloo, B., Van Troys, M., Van Audenhove,

I., & Gettemans, J. (2017). Inhibitory cortactin nanobodies delineate the role of NTA-

and SH3-domain–specific functions during invadopodium formation and cancer cell

invasion. The FASEB Journal, 31(6), 2460–2476.

http://doi.org/10.1096/fj.201600810RR

Bles, E. J. (1906). The Life-History of Xenopus lævis, Daud. Transactions of the Royal

Society of Edinburgh, 41(03), 789–821.

http://doi.org/10.1017/S0080456800035584

Blighe, K., DeDionisio, L., Christie, K. A., Chawes, B., Shareef, S., Kakouli-Duarte, T.,

… Moore, C. B. T. (2018). Gene editing in the context of an increasingly complex

genome. BMC Genomics, 19(1), 595. http://doi.org/10.1186/s12864-018-4963-8

Bowes, J. B., Snyder, K. A., Segerdell, E., Gibb, R., Jarabek, C., Noumen, E., … Vize,

P. D. (2008). Xenbase: A xenopus biology and genomics resource. Nucleic Acids

Research, 36(SUPPL. 1), 761–767. http://doi.org/10.1093/nar/gkm826

Bowes, J. B., Snyder, K. A., Segerdell, E., Jarabek, C. J., Azam, K., Zorn, A. M., & Vize,

P. D. (2009). Xenbase: Gene expression and improved integration. Nucleic Acids

Research, 38(SUPPL.1), 607–612. http://doi.org/10.1093/nar/gkp953

Bresnick, E. H., Lee, H. Y., Fujiwara, T., Johnson, K. D., & Keles, S. (2010, October 8).

GATA switches as developmental drivers. Journal of Biological Chemistry.

251

http://doi.org/10.1074/jbc.R110.159079

Briggs, J. A., Weinreb, C., Wagner, D. E., Megason, S., Peshkin, L., Kirschner, M. W., &

Klein, A. M. (2018). The dynamics of gene expression in vertebrate embryogenesis

at single-cell resolution. Science, pp. 1–16. http://doi.org/10.1126/science.aar5780

Brizzard, B. (2008). Epitope tagging. BioTechniques, 44(5), 693–695.

http://doi.org/10.2144/000112841

Broders, F., & Thiery, J. P. (1995). Contribution of cadherins to directional cell migration

and histogenesis in Xenopus embryos. Cell Adhesion and Communication, 3(5),

419–40. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/8640379

Brothers, S. P., Janovick, J. A., & Conn, P. M. (2003). Unexpected Effects of Epitope

and Chimeric Tags on Gonadotropin-Releasing Hormone Receptors: Implications

for Understanding the Molecular Etiology of Hypogonadotropic Hypogonadism.

Journal of Clinical Endocrinology and Metabolism, 88(12), 6107–6112.

http://doi.org/10.1210/jc.2003-031047

Buck, M. J., & Lieb, J. D. (2004). ChIP-chip: considerations for the design, analysis, and

application of genome-wide chromatin immunoprecipitation experiments.

Genomics, 83(3), 349–60. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/14986705

Cabrera, H. A. (1969). A comprehensive evaluation of pregnancy tests. American

Journal of Obstetrics and Gynecology, 103(1), 32–38.

http://doi.org/10.1016/S0002-9378(16)34336-8

Carlson, E. D., Gan, R., Hodgman, E. C., & Jewet, M. C. (2014). Cell-Free Protein

Synthesis : Applications Come of Age, 30(5), 1185–1194.

http://doi.org/10.1016/j.biotechadv.2011.09.016.Cell-Free

Caussinus, E., Kanca, O., & Affolter, M. (2013). Protein knockouts in living eukaryotes

using degradFP and green fluorescent protein fusion targets. Current Protocols in

252

Protein Science, (August), 1–13. http://doi.org/10.1002/0471140864.ps3002s73

Chen, T., Farragher, S., Bjourson, A. J., Orr, D. F., Rao, P., & Shaw, C. (2003). Granular

gland transcriptomes in stimulated amphibian skin secretions. The Biochemical

Journal, 371(Pt 1), 125–130. http://doi.org/10.1042/BJ20021343

Chong, S. (2008). Overview of Cell-Free Protein Synthesis: Historic Landmarks,

Commercial Systems, and Expanding Applications. Molecular Biology, 6(9), 2166–

2171. http://doi.org/10.1021/nl061786n.Core-Shell

Clements, D., & Woodland, H. R. (2003). VegT induces endoderm by a self-limiting

mechanism and by changing the competence of cells to respond to TGF-?? signals.

Developmental Biology, 258(2), 454–463. http://doi.org/10.1016/S0012-

1606(03)00124-6

Coleman, T. R., Carpenter, P. B., & Dunphy, W. G. (1996). The Xenopus Cdc6 protein

is essential for the initiation of a single round of DNA replication in cell-free extracts.

Cell, 87(1), 53–63. http://doi.org/10.1016/S0092-8674(00)81322-7

Connelly, J. J., Wang, T., Cox, J. E., Haynes, C., Wang, L., Shah, S. H., … Gregory, S.

G. (2006). GATA2 is associated with familial early-onset coronary artery disease.

PLoS Genetics, 2(8), 1265–1273. http://doi.org/10.1371/journal.pgen.0020139

Conway, J. O., Sherwood, L. J., Collazo, M. T., Garza, J. A., & Hayhurst, A. (2010).

Llama single domain antibodies specific for the 7 botulinum neurotoxin serotypes

as heptaplex immunoreagents. PLoS ONE, 5(1), e8818.

http://doi.org/10.1371/journal.pone.0008818

Crew, F. A. E. (1936). Notes from a pregnancy diagnosis laboratory (1935). British

Medical Journal, 1(3932), 993–996. http://doi.org/10.1136/bmj.1.3932.993

Dale, L., & Slack, J. M. W. (1987). Fate map for the 32-cell stage of Xenopus laevis.

Development (Cambridge, England), 99(4), 527–51.

253

Daudin, F. M. (1802). Histoire naturelle des rainettes, des grenouilles et des crapauds :

dédiée à B.G.E.L. Lacépède / par F.M. Daudin ... A Paris : De l’Imprimerie de

Bertrandet ... : http://doi.org/10.5962/bhl.title.5054

Davenport, D., & Nicol, J. A. C. (1955). Luminescence in Hydromedusae. Proceedings

of the Royal Society B: Biological Sciences, 144(916), 399–411.

http://doi.org/10.1098/rspb.1955.0066

Day, R. N., & Davidson, M. (2009). The fluorescent protein palette: tools for cellular

imaging. Chem Soc Rev, 38(10), 2887–2921. http://doi.org/10.1039/b901966a.The

De Meyer, T., Muyldermans, S., & Depicker, A. (2014). Nanobody-based products as

research and diagnostic tools. Trends in Biotechnology, 32(5), 263–270.

http://doi.org/10.1016/j.tibtech.2014.03.001 de Pater, E., Kaimakis, P., Vink, C. S., Yokomizo, T., Yamada-Inagawa, T., van der

Linden, R., … Dzierzak, E. (2013). Gata2 is required for HSC generation and

survival. The Journal of Experimental Medicine, 210(13), 2843–2850.

http://doi.org/10.1084/jem.20130751

De Robertis, E. M., Larraín, J., Oelgeschläger, M., & Wessely, O. (2000). The

establishment of Spemann’s organizer and patterning of the vertebrate embryo.

Nature Reviews. Genetics, 1(3), 171–181. http://doi.org/10.1038/35042039

Deckers, N., Saerens, D., Kanobana, K., Conrath, K., Victor, B., Wernery, U., … Dorny,

P. (2009). Nanobodies, a promising tool for species-specific diagnosis of Taenia

solium cysticercosis. International Journal for Parasitology, 39(5), 625–633.

http://doi.org/10.1016/j.ijpara.2008.10.012

DeMarais, A. A., & Moon, R. T. (1992). The armadillo homologs β-catenin and

plakoglobin are differentially expressed during early development of Xenopus

laevis. Developmental Biology, 153(2), 337–346. http://doi.org/10.1016/0012-

1606(92)90118-Z

254

Ding, Y., Colozza, G., Zhang, K., Moriyama, Y., Ploper, D., Sosa, E. A., … De Robertis,

E. M. (2017). Genome-wide analysis of dorsal and ventral transcriptomes of the

Xenopus laevis gastrula. Developmental Biology, 426(2), 176–187.

http://doi.org/10.1016/j.ydbio.2016.02.032

Ding, Y., Ploper, D., Sosa, E. A., Colozza, G., Moriyama, Y., Benitez, M. D. J., … De

Robertis, E. M. (2017). Spemann organizer transcriptome induction by early beta-

catenin, Wnt, Nodal, and Siamois signals in Xenopus laevis. Proceedings of the

National Academy of Sciences, 114(15), E3081–E3090.

http://doi.org/10.1073/pnas.1700766114

Doench, J. G., Fusi, N., Sullender, M., Hegde, M., Vaimberg, E. W., Donovan, K. F., …

Root, D. E. (2016). Optimized sgRNA design to maximize activity and minimize off-

target effects of CRISPR-Cas9. Nature Biotechnology, 34, 184. Retrieved from

http://dx.doi.org/10.1038/nbt.3437

Drury, K. C., & Schorderet-Slatkine, S. (1975). Effects of cycloheximide on the

“autocatalytic” nature of the maturation promoting factor (MPF) in oocytes of

Xenopus laevis. Cell, 4(3), 269–274. http://doi.org/10.1016/0092-8674(75)90175-0

Eckerdt, F., Yamamoto, T. M., Lewellyn, A. L., & Maller, J. L. (2011). Identification of a

polo-like kinase 4-dependent pathway for de novo centriole formation. Current

Biology : CB, 21(5), 428–32. http://doi.org/10.1016/j.cub.2011.01.072

Eisen, J. S., & Smith, J. C. (2008). Controlling morpholino experiments: don’t stop

making antisense. Development, 135(10), 1735–1743.

http://doi.org/10.1242/dev.001115

Ellenberg, J., Lippincott-Schwartz, J., & Presley, J. F. (1999). Dual-colour imaging with

GFP variants. Trends in Cell Biology, 9(2), 52–56. http://doi.org/10.1016/S0962-

8924(98)01420-2

ENCODE Project Consortium, T. E. P. (2012). An integrated encyclopedia of DNA

255

elements in the human genome. Nature, 489(7414), 57–74.

http://doi.org/10.1038/nature11247

Faunes, F., Sánchez, N., Castellanos, J., Vergara, I. A., Melo, F., & Larraín, J. (2009).

Identification of novel transcripts with differential dorso-ventral expression in

Xenopus gastrula using serial analysis of gene expression. Genome Biology, 10(2),

R15. http://doi.org/10.1186/gb-2009-10-2-r15

Ferreiro, B., Skoglund, P., Bailey, A., Dorsky, R., & Harris, W. A. (1993). XASH1, a

Xenopus homolog of achaete-scute: a proneural gene in anterior regions of the

vertebrate CNS. Mechanisms of Development, 40(1–2), 25–36.

http://doi.org/10.1016/0925-4773(93)90085-C

Friedeberg, C. E., Scarlett, G., McGeeghan, J., Abu-daya, A., Guille, M. J., & Kneale, G.

(2006). Identification of a structural and functional domain in xNAP1 involved in

protein-protein interactions. Nucleic Acids Research, 34(17), 4893–4899.

http://doi.org/10.1093/nar/gkl434

Fritze, C. E., & Anderson, T. R. Epitope tagging: General method for tracking

recombinant proteins, 327 Methods in Enzymology 3–16 (2000). Academic Press.

http://doi.org/10.1016/S0076-6879(00)27263-7

Fujii, K., Shi, Z., Zhulyn, O., Denans, N., & Barna, M. (2017). Pervasive translational

regulation of the cell signalling circuitry underlies mammalian development. Nature

Communications, 8, 1–13. http://doi.org/10.1038/ncomms14443

Furey, T. S. (2012). ChIP–seq and beyond: new and improved methodologies to detect

and characterize protein–DNA interactions. Nature Reviews Genetics,

13(December), 840–852. http://doi.org/10.1038/nrg3306

Gao, Y., Liu, X., Tang, B., Li, C., Kou, Z., Li, L., … Gao, S. (2017). Protein Expression

Landscape of Mouse Embryos during Pre-implantation Development. Cell Reports,

21(13), 3957–3969. http://doi.org/10.1016/j.celrep.2017.11.111

256

Gauss, R., Trautwein, M., Sommer, T., & Spang, A. (2005). New modules for the

repeated internal and N-terminal epitope tagging of genes in Saccharomyces

cerevisiae, 1–12. http://doi.org/10.1002/yea.1187

Gautier, A., & Tebo, A. G. (2018). Fluorogenic Protein-Based Strategies for Detection,

Actuation, and Sensing. BioEssays. http://doi.org/10.1002/bies.201800118

Gautier, J., Minshull, J., Lohka, M., Glotzer, M., Hunt, T., & Maller, J. L. (1990). Cyclin is

a component of maturation-promoting factor from Xenopus. Cell, 60(3), 487–494.

http://doi.org/10.1016/0092-8674(90)90599-A

Gautier, J., Norbury, C., Lohka, M. I., Nurse, P., & Maller, J. L. (1988). Purified

maturation-promoting factor contains the product of a Xenopus homolog of the

fission yeast cell cycle control gene cdc2+. Cell, 54(3), 433–439.

http://doi.org/10.1016/0092-8674(88)90206-1

Geiger, T., Wehner, A., Schaab, C., Cox, J., & Mann, M. (2012). Comparative Proteomic

Analysis of Eleven Common Cell Lines Reveals Ubiquitous but Varying Expression

of Most Proteins. Molecular & Cellular Proteomics, 11(3), M111.014050.

http://doi.org/10.1074/mcp.M111.014050

Gentsch, G. E., Spruce, T., Monteiro, R. S., Owens, N. D. L., Martin, S. R., & Smith, J.

C. (2018). Innate Immune Response and Off-Target Mis-splicing Are Common

Morpholino-Induced Side Effects in Xenopus. Developmental Cell, 44(5), 597–

610.e10. http://doi.org/10.1016/j.devcel.2018.01.022

Gerhart, J., Wu, M., & Kirschner, M. (1984). Cell cycle dynamics of an M-phase-specific

cytoplasmic factor in Xenopus laevis oocytes and eggs. Journal of Cell Biology,

98(4), 1247–1255. http://doi.org/10.1083/jcb.98.4.1247

Getwan, M., & Lienkamp, S. S. (2017). Toolbox in a tadpole: Xenopus for kidney

research. Cell and Tissue Research, 369(1), 143–157.

http://doi.org/10.1007/s00441-017-2611-2

257

Glanz, S., Bunse, A., Wimbert, A., Balczun, C., & Kück, U. (2006). A nucleosome

assembly protein-like polypeptide binds to chloroplast group II intron RNA in

Chlamydomonas reinhardtii. Nucleic Acids Research, 34(18), 5337–5351.

http://doi.org/10.1093/nar/gkl611

Goda, T., Abu-Daya, A., Carruthers, S., Clark, M. D., Stemple, D. L., & Zimmerman, L.

B. (2006). Genetic screens for mutations affecting development of Xenopus

tropicalis. PLoS Genetics, 2(6), 0811–0825.

http://doi.org/10.1371/journal.pgen.0020091

Gong, H., Yan, Y., Fang, B., Xue, Y., Yin, P., Li, L., … Zou, Y. (2014). Knockdown of

Nucleosome Assembly Protein 1-Like 1 Induces Mesoderm Formation and

Cardiomyogenesis Via Notch Signaling in Murine-Induced Pluripotent Stem Cells.

STEM CELLS, 32(7), 1759–1773. http://doi.org/10.1002/stem.1702

Gordon, J. W., Scangos, G. A., Plotkin, D. J., Barbosa, J. A., & Ruddle, F. H. (1980).

Genetic transformation of mouse embryos by microinjection of purified DNA.

Proceedings of the National Academy of Sciences, 77(12), 7380–7384.

http://doi.org/10.1073/pnas.77.12.7380

Grainger, R. M. Xenopus tropicalis as a Model Organism for Genetics and Genomics:

Past, Present, and Future, 917 Methods in molecular biology (Clifton, N.J.) 3–15

(2012). Humana Press, Totowa, NJ. http://doi.org/10.1007/978-1-61779-992-1_1

Grobman, A. B. (1958). Normal Table of Xenopus Laevis (Daudin). A Systematical and

Chronological Survey of the Development from the Fertilized Egg Till the End of

Metamorphosis. P. D. Nieuwkoop , J. Faber. The Quarterly Review of Biology,

33(1), 85–85. http://doi.org/10.1086/402265

GSL Biotech LLC. (n.d.). SnapGene | Software for everyday molecular biology. Retrieved

July 6, 2018, from http://www.snapgene.com/

Guha, T. K., & Edgell, D. R. (2017). Applications of alternative nucleases in the age of

258

CRISPR/Cas9. International Journal of Molecular Sciences, 18(12).

http://doi.org/10.3390/ijms18122565

Gurdon, J. B. (1995). Normal table of Xenopus laevis (Daudin). Trends in Genetics,

11(10), 418. http://doi.org/10.1016/S0168-9525(00)89129-5

Gurdon, J. B., & Hopwood, N. (2000). The introduction of Xenopus laevis into

developmental biology. International Journal of Developmental Biology, 50, 1–8.

http://doi.org/10.1387/ijdb.10761846

Hagihara, Y., & Saerens, D. (2014). Engineering disulfide bonds within an antibody.

Biochimica et Biophysica Acta - Proteins and Proteomics, 3–10.

http://doi.org/10.1016/j.bbapap.2014.07.005

Hahn, C. N., Chong, C.-E., Carmichael, C. L., Wilkins, E. J., Brautigan, P. J., Li, X.-C.,

… Scott, H. S. (2011). Heritable GATA2 mutations associated with familial

myelodysplastic syndrome and acute myeloid leukemia. Nature Genetics, 43(10),

1012–1017. http://doi.org/10.1038/ng.913

Hajar, R. (2011). Animal Testing and Medicine. Heart Views : The Official Journal of the

Gulf Heart Association, 12(1), 42. http://doi.org/10.4103/1995-705X.81548

Hamers-Casterman, C., Atarhouch, T., Muyldermans, S., Robinson, G., Hammers, C.,

Songa, E. B., … Hammers, R. (1993). Naturally occurring antibodies devoid of light

chains. Nature, 363(6428), 446–448. http://doi.org/10.1038/363446a0

Harland, R. M. (1991). Appendix G: In Situ Hybridization: An Improved Whole-Mount

Method for Xenopus Embryos. In Methods in cell biology (Vol. 36, pp. 685–695).

http://doi.org/10.1016/S0091-679X(08)60307-6

Harland, R. M., & Gilchrist, M. J. (2017). Editorial: The Xenopus laevis genome.

Developmental Biology, 426(2), 139–142.

http://doi.org/10.1016/j.ydbio.2017.04.016

259

Harland, R. M., & Grainger, R. M. (2011). Xenopus research: Metamorphosed by

genetics and genomics. Trends in Genetics, 27(12), 507–515.

http://doi.org/10.1016/j.tig.2011.08.003

Harmsen, M. M., Blokker, J. C., Pritz-Verschuren, S. B., Bartelink, W., Van Der Burg, H.,

& Koch, G. (2013). Isolation of Panels of Llama Single-Domain Antibody Fragments

Binding All Nine Neuraminidase Subtypes of Influenza A Virus. Antibodies, 2, 168–

192. http://doi.org/10.3390/antib2020168

Harmsen, M. M., Ruuls, R. C., Nijman, I. J., Niewold, T. a., Frenken, L. G. J., & De Geus,

B. (2000). Llama heavy-chain V regions consist of at least four distinct subfamilies

revealing novel sequence features. Molecular Immunology, 37(10), 579–590.

http://doi.org/10.1016/S0161-5890(00)00081-X

Hassanzadeh-ghassabeh, G., Saerens, D., & Muyldermans, S. (2011). Nanoproteomics.

Methods, 790, 239–259. http://doi.org/10.1007/978-1-61779-319-6

Heasman, J., Crawford, A., Goldstone, K., Garner-Hamrick, P., Gumbiner, B. M.,

McCrea, P., … Wylie, C. (1994). Overexpression of cadherins and underexpression

of β-catenin inhibit dorsal mesoderm induction in early Xenopus embryos. Cell,

79(5), 791–803. http://doi.org/10.1016/0092-8674(94)90069-8

Heasman, J., Kofron, M., & Wylie, C. (2000). βCatenin Signaling Activity Dissected in

the Early Xenopus Embryo: A Novel Antisense Approach. Developmental Biology,

222(1), 124–134. http://doi.org/10.1006/dbio.2000.9720

Hebbes, T. R., Clayton, A. L., Thorne, A. W., & Crane-Robinson, C. (1994). Core histone

hyperacetylation co-maps with generalized DNase I sensitivity in the chicken beta-

globin chromosomal domain. The EMBO Journal, 13(8), 1823–30. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/8168481

Hellsten, U., Harland, R. M., Gilchrist, M. J., Hendrix, D., Jurka, J., Kapitonov, V., …

Rokhsar, D. S. (2010). The Genome of the Western Clawed Frog Xenopus

260

tropicalis. Science, 328(5978), 633–636. http://doi.org/328/5978/633

[pii]\r10.1126/science.1183670

Herce, H. D., Deng, W., Helma, J., Leonhardt, H., & Cardoso, M. C. (2013). Visualization

and targeted disruption of protein interactions in living cells. Nature

Communications, 4(May), 2660. http://doi.org/10.1038/ncomms3660

Hirsch, N., Zimmerman, L. B., & Grainger, R. M. (2002). Xenopus, the next generation:

X. tropicalis genetics and genomics. Developmental Dynamics, 225(4), 422–433.

http://doi.org/10.1002/dvdy.10178

Hogben, L. T., & Winton, F. R. (1930). Studies on the Pituitary. VII: The separate identity

of the pressor and melanophore principles. Biochem J, 16(5), 619–630.

Hoogenboom, H. R., Griffiths, a D., Johnson, K. S., Chiswell, D. J., Hudson, P., & Winter,

G. (1991). Multi-subunit proteins on the surface of filamentous phage:

methodologies for displaying antibody (Fab) heavy and light chains. Nucleic Acids

Research, 19(15), 4133–4137. http://doi.org/10.1093/nar/19.15.4133

Hoppler, S., Brown, J. D., & Moon, R. T. (1996). Expression of a dominant-negative wnt

blocks induction of MyoD in Xenopus embryos. Genes and Development, 10(21),

2805–2817. http://doi.org/10.1101/gad.10.21.2805

Horner, V. L., & Wolfner, M. F. (2008). Transitioning from egg to embryo: Triggers and

mechanisms of egg activation. Developmental Dynamics.

http://doi.org/10.1002/dvdy.21454

Houben, F., Ramaekers, F. C. S., Snoeckx, L. H. E. H., & Broers, J. L. V. (2007). Role

of nuclear lamina-cytoskeleton interactions in the maintenance of cellular strength.

Biochimica et Biophysica Acta - Molecular Cell Research.

http://doi.org/10.1016/j.bbamcr.2006.09.018

Hsu, A. P., Johnson, K. D., Falcone, E. L., Sanalkumar, R., Sanchez, L., Hickstein, D.

D., … Holland, S. M. (2013). GATA2 haploinsufficiency caused by mutations in a

261

conserved intronic element leads to MonoMAC syndrome. Blood, 121(19), 3830–

3837. http://doi.org/10.1182/blood-2012-08-452763

Hsu, A. P., Sampaio, E. P., Khan, J., Calvo, K. R., Lemieux, J. E., Patel, S. Y., … Holland,

S. M. (2011). Mutations in GATA2 are associated with the autosomal dominant and

sporadic monocytopenia and mycobacterial infection (MonoMAC) syndrome.

Blood, 118(10), 2653–2655. http://doi.org/10.1182/blood-2011-05-356352

Hu, Y., Liu, C., & Muyldermans, S. (2017). Nanobody-Based Delivery Systems for

Diagnosis and Targeted Tumor Therapy. Frontiers in Immunology, 8, 1442.

http://doi.org/10.3389/fimmu.2017.01442

Huang, Y.-L., & Niehrs, C. (2014). Polarized Wnt Signaling Regulates Ectodermal Cell

Fate in Xenopus. Developmental Cell, 29(2), 250–257.

http://doi.org/10.1016/j.devcel.2014.03.015

Huse, W. D., Sastry, L., Iverson, S. a, Kang, a S., Alting-Mees, M., Burton, D. R., …

Lerner, R. a. (1989). Generation of a large combinatorial library of the

immunoglobulin repertoire in phage lambda. Science (New York, N.Y.), 246(1984),

1275–1281. http://doi.org/10.1126/science.2531466

Itoh, K., & Sokol, S. Y. (1999). Axis determination by inhibition of Wnt signaling in

Xenopus. Genes and Development, 13(17), 2328–2336.

http://doi.org/10.1101/gad.13.17.2328

Itoh, K., & Sokol, S. Y. (2014). Expression cloning of camelid nanobodies specific for

xenopus embryonic antigens. PLoS ONE, 9(10), 1–7.

http://doi.org/10.1371/journal.pone.0107521

Janeway, C. (2001). Immunobiology 5 : the immune system in health and disease.

Garland Pub.

Jobling, S. a, Jarman, C., Teh, M.-M., Holmberg, N., Blake, C., & Verhoeyen, M. E.

(2003). Immunomodulation of enzyme function in plants by single-domain antibody

262

fragments. Nature Biotechnology, 21(January), 77–80.

http://doi.org/10.1038/nbt772

Jones, C. M., & Smith, J. C. (2008). An overview of Xenopus development. (P. Sharpe;

I. Mason, Ed.)Methods in Molecular Biology. New York, NY: Humana Press, New

York, NY. http://doi.org/10.1007/978-1-60327-483-8_28

Jones, E. A. (2003). Molecular Control of Pronephric Development. An Overview. In The

Kidney: From Normal Development to Congenital Disease (pp. 93–118). Elsevier.

http://doi.org/10.1016/B978-012722441-1/50010-5

Kanca, O., Bellen, H. J., & Schnorrer, F. (2017). Gene tagging strategies to assess

protein expression, localization, and function in Drosophila. Genetics, 207(2), 389–

412. http://doi.org/10.1534/genetics.117.199968

Karpinka, J. B., Fortriede, J. D., Burns, K. A., James-Zorn, C., Ponferrada, V. G., Lee,

J., … Vize, P. D. (2015). Xenbase, the Xenopusmodel organism database; new

virtualized system, data types and genomes. Nucleic Acids Research, 43(D1),

D756–D763. http://doi.org/10.1093/nar/gku956

Kelly, L. A., Mezulis, S., Yates, C., Wass, M., & Sternberg, M. (2015). The Phyre2 web

portal for protein modelling, prediction, and analysis. Nature Protocols, 10(6), 845–

858. http://doi.org/10.1038/nprot.2015-053

Khokha, M. K., Chung, C., Bustamante, E. L., Gaw, L. W. K., Trott, K. A., Yeh, J., …

Grammer, T. C. (2002). Techniques and probes for the study of Xenopus tropicalis

development. Developmental Dynamics, 225(4), 499–510.

http://doi.org/10.1002/dvdy.10184

Khokha, M. K., Krylov, V., Reilly, M. J., Gall, J. G., Bhattacharya, D., Cheung, C. Y. J.,

… Zimmerman, L. B. (2009). Rapid gynogenetic mapping of Xenopus tropicalis

mutations to chromosomes. Developmental Dynamics, 238(6), 1398–1406.

http://doi.org/10.1002/dvdy.21965

263

Kim, D. Y., Hussack, G., Kandalaft, H., & Tanha, J. (2014). Mutational approaches to

improve the biophysical properties of human single-domain antibodies. Biochimica

et Biophysica Acta - Proteins and Proteomics.

http://doi.org/10.1016/j.bbapap.2014.07.008

Klein, S. L., Strausberg, R. L., Wagner, L., Pontius, J., Clifton, S. W., & Richardson, P.

(2002, December). Genetic and genomic tools for Xenopus research: The NIH

Xenopus initiative. Developmental Dynamics. http://doi.org/10.1002/dvdy.10174

Klooster, R., Maassen, B. T. H., Stam, J. C., Hermans, P. W., ten Haaft, M. R., Detmers,

F. J. M., … Theo Verrips, C. (2007). Improved anti-IgG and HSA affinity ligands:

Clinical application of VHH antibody technology. Journal of Immunological Methods,

324, 1–12. http://doi.org/10.1016/j.jim.2007.04.005

Kobayashi, H., Minshull, J., Ford, C., Golsteyn, R., Poon, R., & Hunt, T. (1991). On the

synthesis and destruction of A- and B-type cyclins during oogenesis and meiotic

maturation in Xenopus laevis. Journal of Cell Biology, 114(4), 755–765.

http://doi.org/10.1083/jcb.114.4.755

Kolodziej, P. A., & Young, R. A. (1991). Epitope tagging and protein surveillance.

Methods in Enzymology, 194, 508–19. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/1706460

Kroll, K. L., & Amaya, E. (1996). Transgenic Xenopus embryos from sperm nuclear

transplantations reveal FGF signaling requirements during gastrulation.

Development (Cambridge, England), 122(10), 3173–83. Retrieved from

http://dev.biologists.org/content/122/10/3173.abstract

Kubala, M. H., Kovtun, O., Alexandrov, K., & Collins, B. M. (2010). Structural and

thermodynamic analysis of the GFP:GFP-nanobody complex. Protein Science,

19(12), 2389–2401. http://doi.org/10.1002/pro.519

Kuroda, H., Wessely, O., & De Robertis, E. M. (2004). Neural induction in Xenopus:

264

Requirement for ectodermal and endomesodermal signals via Chordin, Noggin, β-

Catenin, and Cerberus. PLoS Biology, 2(5).

http://doi.org/10.1371/journal.pbio.0020092

Kyuno, J. ichi, Massé, K., & Jones, E. A. (2008). A functional screen for genes involved

in Xenopus pronephros development. Mechanisms of Development, 125(7), 571–

586. http://doi.org/10.1016/j.mod.2008.03.001

Ladenson, R. C., Crimmins, D. L., Landt, Y., & Ladenson, J. H. (2006). Isolation and

characterization of a thermally stable recombinant anti-caffeine heavy-chain

antibody fragment. Analytical Chemistry, 78(13), 4501–4508.

http://doi.org/10.1021/ac058044j

Lafontaine, D., & Tollervey, D. (1996). One-step PCR mediated strategy for the

construction of conditionally expressed and epitope tagged yeast proteins, 24(17),

3469–3471.

Lauf, U., Lopez, P., & Falk, M. M. (2001). Expression of fluorescently tagged connexins:

a novel approach to rescue function of oligomeric DsRed-tagged proteins. FEBS

Letters, 498(1), 11–5. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/11389889

Lauwereys, M., Ghahroudi, M. A., Desmyter, A., Kinne, J., Hölzer, W., De Genst, E., …

Muyldermans, S. (1998). Potent enzyme inhibitors derived from dromedary heavy-

chain antibodies. EMBO Journal, 17(13), 3512–3520.

http://doi.org/10.1093/emboj/17.13.3512

Lee, K. H. (2001). Proteomics: a technology-driven and technology-limited discovery

science. Trends in Biotechnology, 19(6), 217–22. http://doi.org/10.1016/S0167-

7799(01)01639-0

Leenaars, M., & Hendriksen, C. F. M. (2005). Critical steps in the production of polyclonal

and monoclonal antibodies: Evaluation and recommendations. ILAR Journal, 46(3),

265

269–279. http://doi.org/10.1093/ilar.46.3.269

Lei, Y., Guo, X., Liu, Y., Cao, Y., Deng, Y., Chen, X., … Zhao, H. (2012). Efficient

targeted gene disruption in Xenopus embryos using engineered transcription

activator-like effector nucleases (TALENs). Proceedings of the National Academy

of Sciences, 109(43), 17484–17489. http://doi.org/10.1073/pnas.1215421109

Leow, C., Fischer, K., Leow, C., Cheng, Q., Chuah, C., & McCarthy, J. (2017). Single

Domain Antibodies as New Biomarker Detectors. Diagnostics, 7(4), 52.

http://doi.org/10.3390/diagnostics7040052

Li, Y. (2010). Commonly used tag combinations for tandem affinity purification.

Biotechnology and Applied Biochemistry, 55(2), 73–83.

http://doi.org/10.1042/BA20090273

Liu, W., Song, H., Chen, Q., Yu, J., Xian, M., Nian, R., & Feng, D. (2018). Recent

advances in the selection and identification of antigen-specific nanobodies.

Molecular Immunology, 96, 37–47. http://doi.org/10.1016/j.molimm.2018.02.012

Liu, X. J. (2006). Xenopus protocols : cell biology and signal transduction. Humana

Press. Retrieved from

https://books.google.co.uk/books?id=AKAEqcQz9F4C&pg=PA391&lpg=PA391&d

q=protein+depletion+in+extracts+xenopus&source=bl&ots=TIZdMV6B4I&sig=9aI6

T9ytHK-r24JehdToVajGVwA&hl=en&sa=X&ved=0ahUKEwjwnMrKn8PZAhXGC-

wKHcI1COIQ6AEIVjAE#v=onepage&q=protein depletion in extracts

xenopus&f=false

Loh, K. M., van Amerongen, R., & Nusse, R. (2016). Generating Cellular Diversity and

Spatial Form: Wnt Signaling and the Evolution of Multicellular Animals.

Developmental Cell, 38(6), 643–655. http://doi.org/10.1016/j.devcel.2016.08.011

Lohka, M. I., & Maller, J. L. (1985). Induction of nuclear envelope breakdown,

chromosome condensation, and spindle formation in cell-free extracts. Journal of

266

Cell Biology, 101(2), 518–523. http://doi.org/10.1083/jcb.101.2.518

Lombard-Banek, C., Moody, S. A., & Nemes, P. (2016). High-Sensitivity Mass

Spectrometry for Probing Gene Translation in Single Embryonic Cells in the Early

Frog (Xenopus) Embryo. Frontiers in Cell and Developmental Biology, 4(October),

1–11. http://doi.org/10.3389/fcell.2016.00100

Love, N. R., Thuret, R., Chen, Y., Ishibashi, S., Sabherwal, N., Paredes, R., … Amaya,

E. (2011). pTransgenesis: a cross-species, modular transgenesis resource.

Development, 138(24), 5451–5458. http://doi.org/10.1242/dev.066498

Lyons, J. P., Miller, R. K., Zhou, X., Weidinger, G., Deroo, T., Denayer, T., … McCrea,

P. D. (2009). Requirement of Wnt/β-catenin signaling in pronephric kidney

development. Mechanisms of Development, 126(3–4), 142–159.

http://doi.org/10.1016/j.mod.2008.11.007

Ma, L., Sun, Y., Kang, X., & Wan, Y. (2014). Development of nanobody-based flow

injection chemiluminescence immunoassay for sensitive detection of human

prealbumin. Biosensors and Bioelectronics, 61, 165–171.

http://doi.org/10.1016/j.bios.2014.04.026

MacDonald, B. T., Tamai, K., & He, X. (2009). Wnt/β-catenin signaling: components,

mechanisms, and diseases. Developmental C, 17(1), 9–26.

http://doi.org/10.1016/j.devcel.2009.06.016.Wnt/

Malartre, M., Short, S., & Sharpe, C. (2006). Xenopus embryos lacking specific isoforms

of the corepressor SMRT develop abnormal heads. Developmental Biology, 292(2),

333–343. http://doi.org/10.1016/j.ydbio.2006.01.007

Maller, J. L. (1986). Mitogenic signalling and protein phosphorylation in Xenopus

oocytes. J Cyclic Nucleotide Protein Phosphor Res, 11(7), 543–555. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/2831261

Maller, J. L. (2012). Pioneering the xenopus oocyte and egg extract system. Journal of

267

Biological Chemistry, 287(26), 21640–21653.

http://doi.org/10.1074/jbc.X112.371161

Marschall, A. L. J., Frenzel, A., Schirrmann, T., Schüngel, M., & Dübel, S. (2011).

Targeting antibodies to the cytoplasm. MAbs, 3(1), 3–16.

http://doi.org/10.4161/MABS.3.1.14110

Matz, M. V., Fradkov, A. F., Labas, Y. A., Savitsky, A. P., Zaraisky, A. G., Markelov, M.

L., & Lukyanov, S. A. (1999). Fluorescent proteins from nonbioluminescent

Anthozoa species. Nature Biotechnology, 17(10), 969–973.

http://doi.org/10.1038/13657

Maue, R. A. (2007). Understanding Ion Channel Biology Using Epitope Tags: Progress,

Pitfalls, and Promise. J. Cell. Physiol (Vol. 213). Retrieved from

http://shaker.umh.es/congresos/struc.biol.neurosci.meeting/abstracts/paper.08_06

.pdf

McCrea, P. D., Brieher, W. M., & Gumbiner, B. M. (1993). Induction of a secondary body

axis in Xenopus by antibodies to β-catenin. Journal of Cell Biology, 123(2), 477–

484. http://doi.org/10.1083/jcb.123.2.477

McCrea, P. D., & Gumbiner, B. M. (1991). Purification of a 92-kDa cytoplasmic protein

tightly associated with the cell-cell adhesion molecule E-cadherin (uvomorulin):

Characterization and extractability of the protein complex from the cell

cytostructure. Journal of Biological Chemistry, 266(7), 4514–4520. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/1999432

McGuffin, L. J., Bryson, K., & Jones, D. T. (2000). The PSIPRED protein structure

prediction server. Bioinformatics (Oxford, England), 16(4), 404–5. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/10869041

Melcon, G., Koslov, S., Cutler, D. A., Sullivan, T., Hernandez, L., Zhao, P., … Stewart,

C. L. (2006). Loss of emerin at the nuclear envelope disrupts the Rb1/E2F and

268

MyoD pathways during muscle regeneration. Human Molecular Genetics, 15(4),

637–651. http://doi.org/10.1093/hmg/ddi479

Méreau, A., Le Sommer, C., Lerivray, H., Lesimple, M., & Hardy, S. (2007). Xenopus as

a model to study alternative splicing in vivo. Biology of the Cell / under the Auspices

of the European Cell Biology Organization, 99(1), 55–65.

http://doi.org/10.1042/BC20060073

Méreau, A., Sommer, C., Lerivray, H., Lesimple, M., & Hardy, S. (2007). Xenopus as a

model to study alternative splicing in vivo. Biology of the Cell, 99(1), 55–65.

http://doi.org/10.1042/BC20060073

Miller, R. K., Gomez de la Torre Canny, S., Jang, C.-W., Cho, K., Ji, H., Wagner, D. S.,

… McCrea, P. D. (2011). Pronephric Tubulogenesis Requires Daam1-Mediated

Planar Cell Polarity Signaling. Journal of the American Society of Nephrology,

22(9), 1654–1664. http://doi.org/10.1681/ASN.2010101086

Min, Z., Lin, H., Zhu, X., Gao, L., Khand, A. A., & Tao, Q. (2016). Ascl1 represses the

mesendoderm induction in Xenopus. Acta Biochimica et Biophysica Sinica, 48(11),

1006–1015. http://doi.org/10.1093/abbs/gmw092

Minshull, J., Golsteyn, R., Hill, C. S., & Hunt, T. (1990). The A- and B-type cyclin

associated cdc2 kinases in Xenopus turn on and off at different times in the cell

cycle. The EMBO Journal, 9(9), 2865–75. http://doi.org/10.1002/J.1460-

2075.1990.TB07476.X

Mishin, A. S., Belousov, V. V, Solntsev, K. M., & Lukyanov, K. A. (2015). Novel uses of

fluorescent proteins. Current Opinion in Chemical Biology, 27, 1–9.

http://doi.org/10.1016/j.cbpa.2015.05.002

Mojica, F. J. M., Díez-Villaseñor, C., García-Martínez, J., & Almendros, C. (2009). Short

motif sequences determine the targets of the prokaryotic CRISPR defence system.

Microbiology, 155(3), 733–740. http://doi.org/10.1099/mic.0.023960-0

269

Moor, R. M. (1988). Regulation of the Meiotic Cycle in Oocytes of Domestic Mammals.

Annals of the New York Academy of Sciences, 541(1), 248–258.

http://doi.org/10.1111/j.1749-6632.1988.tb22262.x

Moreno-Mateos, M. A., Vejnar, C. E., Beaudoin, J.-D., Fernandez, J. P., Mis, E. K.,

Khokha, M. K., & Giraldez, A. J. (2015). CRISPRscan: designing highly efficient

sgRNAs for CRISPR-Cas9 targeting in vivo. Nature Methods, 12, 982. Retrieved

from http://dx.doi.org/10.1038/nmeth.3543

Moriguchi, T., & Yamamoto, M. (2014). A regulatory network governing Gata1 and Gata2

gene transcription orchestrates erythroid lineage differentiation. International

Journal of Hematology, 100(5), 417–424. http://doi.org/10.1007/s12185-014-1568-

Movassagh, M., & Philpott, A. (2008). Cardiac differentiation in Xenopus requires the

cyclin-dependent kinase inhibitor, p27Xic1. Cardiovascular Research, 79(3), 436–

447. http://doi.org/10.1093/cvr/cvn105

Mujić-Delić, A., De Wit, R. H., Verkaar, F., & Smit, M. J. (2014). GPCR-targeting

nanobodies: Attractive research tools, diagnostics, and therapeutics. Trends in

Pharmacological Sciences, 35(5), 247–255.

http://doi.org/10.1016/j.tips.2014.03.003

Muller, H. J. (1927). ARTIFICIAL TRANSMUTATION OF THE GENE. Science, 66(1699),

84 LP-87. Retrieved from

http://science.sciencemag.org/content/66/1699/84.abstract

Munro, S., & Pelham, H. R. B. (1987). A C-terminal signal prevents secretion of luminal

ER proteins. Cell, 48(5), 899–907. http://doi.org/10.1016/0092-8674(87)90086-9

Muyldermans, S. (2001). Single domain camel antibodies: Current status. Reviews in

Molecular Biotechnology, 74(4), 277–302. http://doi.org/10.1016/S1389-

0352(01)00021-6

270

Nakayama, T., Fish, M. B., Fisher, M., Oomen-Hajagos, J., Thomsen, G. H., & Grainger,

R. M. (2013). Simple and efficient CRISPR/Cas9-mediated targeted mutagenesis

in Xenopus tropicalis. Genesis, 51(12), 835–843. http://doi.org/10.1002/dvg.22720

Naylor, R. W., Collins, R. J., Philpott, A., & Jones, E. A. (2009). Normal levels of p27 are

necessary for somite segmentation and determining pronephric organ size.

Organogenesis, 5(4), 201–10. Retrieved from

http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2878748&tool=pmcentr

ez&rendertype=abstract

Nettleship, J. E., Ren, J., Rahman, N., Berrow, N. S., Hatherley, D., Neil Barclay, a., &

Owens, R. J. (2008). A pipeline for the production of antibody fragments for

structural studies using transient expression in HEK 293T cells. Protein Expression

and Purification, 62, 83–89. http://doi.org/10.1016/j.pep.2008.06.017

Nettleship, J. E., Watson, P. J., Rahman-Huq, N., Fairall, L., Posner, M. G., Upadhyay,

A., … Owens, R. J. (2015). Transient Expression in HEK 293 Cells: An Alternative

to E. coli for the Production of Secreted and Intracellular Mammalian Proteins. In E.

García-Fruitós (Ed.), (pp. 209–222). New York, NY: Springer New York.

http://doi.org/10.1007/978-1-4939-2205-5_11

Newnham, L. E., Wright, M. J., Holdsworth, G., Kostarelos, K., Martyn, K., Rabbitts, T.

H., & Lawson, A. D. (2015). Functional inhibition of beta-catenin-mediated Wnt

singaling by intracellular VHH antibodies. MAbs, 7(1), 180–191.

http://doi.org/10.4161/19420862.2015.989023

Newport, J., & Kirschner, M. (1982). A major developmental transition in early xenopus

embryos: I. characterization and timing of cellular changes at the midblastula stage.

Cell, 30(3), 675–686. http://doi.org/10.1016/0092-8674(82)90272-0

Nguyen-Duc, T., Peeters, E., Muyldermans, S., Charlier, D., & Hassanzadeh-

Ghassabeh, G. (2013). Nanobody(R)-based chromatin immunoprecipitation/micro-

271

array analysis for genome-wide identification of transcription factor DNA binding

sites. Nucleic Acids Research, 41(5), e59. http://doi.org/10.1093/nar/gks1342

Nguyen, L., Besson, A., Heng, J. I., Schuurmans, C., Teboul, L., Parras, C., … Guillemot,

F. (2006). p27kip1 independently promotes neuronal differentiation and migration

in the cerebral cortex. Genes and Development, 1511–1524.

http://doi.org/10.1101/gad.377106.Cyclin-dependent

Nijjar, S., & Woodland, H. R. (2013). Protein interactions in Xenopus germ plasm RNP

particles. PLoS One, 8(11), e80077. http://doi.org/10.1371/journal.pone.0080077

Niman, H. L., Houghten, R. A., Walker, L. E., Reisfeld, R. A., Wilson, I. A., Hogle, J. M.,

& Lerner, R. A. (1983). Generation of protein-reactive antibodies by short peptides

is an event of high frequency: implications for the structural basis of immune

recognition. Proceedings of the National Academy of Sciences of the United States

of America, 80(16), 4949–53. Retrieved from

http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=384165&tool=pmcentre

z&rendertype=abstract

Nishitani, E., Li, C., Lee, J., Hotta, H., Katayama, Y., Yamaguchi, M., & Kinoshita, T.

(2015). Pou5f3.2-induced proliferative state of embryonic cells during gastrulation

of Xenopus laevis embryo. Development Growth and Differentiation, 57(9), 591–

600. http://doi.org/10.1111/dgd.12246

Nusse, R., & Varmus, H. (2012). Three decades of Wnts: A personal perspective on how

a scientific field developed. EMBO Journal, 31(12), 2670–2684.

http://doi.org/10.1038/emboj.2012.146

Nusse, R., & Varmus, H. E. (1982). Many tumors induced by the mouse mammary tumor

virus contain a provirus integrated in the same region of the host genome. Cell,

31(1), 99–109. http://doi.org/10.1016/0092-8674(82)90409-3

Nutt, S. L., Bronchain, O. J., Hartley, K. O., & Amaya, E. (2001). Comparison of

272

Morpholino based translational inhibition during the development of Xenopus laevis

and Xenopus tropicalis. Genesis, 30(3), 110–113. http://doi.org/10.1002/gene.1042

Offield, M. F., Hirsch, N., & Grainger, R. M. (2000). The development of Xenopus

tropicalis transgenic lines and their use in studying lens developmental timing in

living embryos. Development (Cambridge, England), 127(9), 1789–1797.

Onjiko, R. M., Moody, S. A., & Nemes, P. (2015). Single-cell mass spectrometry reveals

small molecules that affect cell fates in the 16-cell embryo. Proceedings of the

National Academy of Sciences, 112(21), 6545–6550.

http://doi.org/10.1073/pnas.1423682112

OPPF-UK: Home. (n.d.). Retrieved May 22, 2018, from https://www.oppf.rc-

harwell.ac.uk/OPPF/

Ostergaard, P., Simpson, M. A., Connell, F. C., Steward, C. G., Brice, G., Woollard, W.

J., … Mansour, S. (2011). Mutations in GATA2 cause primary lymphedema

associated with a predisposition to acute myeloid leukemia (Emberger syndrome).

Nature Genetics, 43(10), 929–931. http://doi.org/10.1038/ng.923

Owens, N. D. L., Blitz, I. L., Lane, M. A., Patrushev, I., Overton, J. D., Gilchrist, M. J., …

Khokha, M. K. (2016). Measuring Absolute RNA Copy Numbers at High Temporal

Resolution Reveals Transcriptome Kinetics In Development. Cell Reports, 14(12),

632–647. http://doi.org/10.1097/OGX.0000000000000256.Prenatal

Ozawa, M., Baribault, H., & Kemler, R. (1989). The cytoplasmic domain of the cell

adhesion molecule uvomorulin associates with three independent proteins

structurally related in different species. The EMBO Journal, 8(6), 1711–7. Retrieved

from

http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=401013&tool=pmcentre

z&rendertype=abstract

Packer, J., & Trapnell, C. (2018). Single-Cell Multi-omics: An Engine for New

273

Quantitative Models of Gene Regulation. Trends in Genetics, 34(9), 653–665.

http://doi.org/10.1016/j.tig.2018.06.001

Pan, F. C., Chen, Y., Loeber, J., Henningfeld, K., & Pieler, T. (2006). I-SceI

meganuclease-mediated transgenesis inXenopus. Developmental Dynamics,

235(1), 247–252. http://doi.org/10.1002/dvdy.20608

Pardon, E., Laeremans, T., Triest, S., Rasmussen, S. G. F., Wohlkönig, A., Ruf, A., …

Steyaert, J. (2014). A general protocol for the generation of Nanobodies for

structural biology. Nature Protocols, 9(Step 26), 674–93.

http://doi.org/10.1038/nprot.2014.039

Park, Y.-J., & Luger, K. (2006). Structure and function of nucleosome assembly

proteinsThis paper is one of a selection of papers published in this Special Issue,

entitled 27th International West Coast Chromatin and Chromosome Conference,

and has undergone the Journal’s usual peer review process. Biochemistry and Cell

Biology, 84(4), 549–549. http://doi.org/10.1139/o06-088

Pearl, E. J., Grainger, R. M., Guille, M. J., & Horb, M. E. (2012). Development of xenopus

resource centers: The National Xenopus Resource and the European Xenopus

Resource Center. Genesis, 50(3), 155–163. http://doi.org/10.1002/dvg.22013

Peshkin, L., Wühr, M., Pearl, E. J., Haas, W., Freeman, R. M., Gerhart, J. C., …

Kirschner, M. W. (2015). On the Relationship of Protein and mRNA Dynamics in

Vertebrate Embryonic Development. Developmental Cell, 35(3), 383–394.

http://doi.org/10.1016/j.devcel.2015.10.010

Peuchen, E. H., Cox, O. F., Sun, L., Hebert, A. S., Coon, J. J., Champion, M. M., …

Huber, P. W. (2017). Phosphorylation Dynamics Dominate the Regulated Proteome

during Early Xenopus Development. Scientific Reports, 7(1), 1–14.

http://doi.org/10.1038/s41598-017-15936-y

Prasher, D. C., Eckenrode, V. K., Ward, W. W., Prendergast, F. G., & Cormier, M. J.

274

(1992). Primary structure of the Aequorea victoria green-fluorescent protein. Gene,

111(2), 229–233. http://doi.org/10.1016/0378-1119(92)90691-H

Press, E. M., Givol, D., Piggot, P. J., Porter, R. R., & Wilkinson, J. M. (1966). The

chemical structure of the heavy chains of rabbit and human immunoglobulin G

(IgG). Proceedings of the Royal Society of London. Series B, Biological Sciences,

166(1003), 150–8. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/4167013

Preston, S., Aras, S., & Zaidi, M. (2018). Spatiotemporal Labeling of Melanocytes in

Mice. International Journal of Molecular Sciences, 19(5), 1469.

http://doi.org/10.3390/ijms19051469

Puffenberger, E. G., Hu-Lince, D., Parod, J. M., Craig, D. W., Dobrin, S. E., Conway, A.

R., … Stephan, D. A. (2004). Mapping of sudden infant death with dysgenesis of

the testes syndrome (SIDDT) by a SNP genome scan and identification of TSPYL

loss of function. Proceedings of the National Academy of Sciences of the United

States of America, 101(32), 11689–11694.

http://doi.org/10.1073/pnas.0401194101

Qian, Y.-W., Erikson, E., Taieb, F. E., & Maller, J. L. (2001). The Polo-like Kinase Plx1

Is Required for Activation of the Phosphatase Cdc25C and Cyclin B-Cdc2 in

Xenopus Oocytes. Molecular Biology of the Cell, 12(6), 1791–1799.

http://doi.org/10.1091/mbc.12.6.1791

Rawson, R. A. (1943). The binding of T-1824 and structurally related diazo dyes by the

plasma protein. American Jurnal of Physiology, 138, 708.

Reardon, S. (2016). US government issues historic $ 3 . 5 • million fine over animal

welfare. Nature, 4–7. http://doi.org/10.1038/nature.2016.19958

Richter, J. D. (1999). Cytoplasmic polyadenylation in development and beyond.

Microbiology and Molecular Biology Reviews : MMBR, 63(2), 446–56. Retrieved

from http://www.ncbi.nlm.nih.gov/pubmed/10357857

275

Riechmann, L., & Muyldermans, S. (1999). Single domain antibodies: comparison of

camel VH and camelised human VH domains. Journal of Immunological Methods,

231, 25–38. http://doi.org/10.1016/S0022-1759(99)00138-6

Robertson, G., Hirst, M., Bainbridge, M., Bilenky, M., Zhao, Y., Zeng, T., … Jones, S.

(2007). Genome-wide profiles of STAT1 DNA association using chromatin

immunoprecipitation and massively parallel sequencing. Nature Methods, 4(8),

651–657. http://doi.org/10.1038/nmeth1068

Rocchetti, A., Hawes, C., & Kriechbaumer, V. (2014). Fluorescent labelling of the actin

cytoskeleton in plants using a cameloid antibody. Plant Methods, 10(1), 12.

http://doi.org/10.1186/1746-4811-10-12

Rodrigues, N. P., Tipping, A. J., Wang, Z., & Enver, T. (2012). GATA-2 mediated

regulation of normal hematopoietic stem/progenitor cell function, myelodysplasia

and myeloid leukemia. The International Journal of Biochemistry & Cell Biology,

44(3), 457–60. http://doi.org/10.1016/j.biocel.2011.12.004

Rothbauer, U., Zolghadr, K., Tillib, S. V., Nowak, D., Schermelleh, L., Gahl, A., …

Leonhardt, H. (2006). Targeting and tracing antigens in live cells with fluorescent

nanobodies. Nature Methods, 3(11), 887–889. http://doi.org/10.1038/nmeth953

Ryckaert, S., Pardon, E., Steyaert, J., & Callewaert, N. (2010). Isolation of antigen-

binding camelid heavy chain antibody fragments (nanobodies) from an immune

library displayed on the surface of Pichia pastoris. Journal of Biotechnology, 145,

93–98. http://doi.org/10.1016/j.jbiotec.2009.10.010

Ryffel, G. U., Werdien, D., Turan, G., Gerhards, A., Goosses, S., & Senkel, S. (2003).

Tagging muscle cell lineages in development and tail regeneration using Cre

recombinase in transgenic Xenopus. Nucleic Acids Research, 31(8), e44.

http://doi.org/10.1093/nar/gng044

Sabherwal, N., Tsutsui, A., Hodge, S., Wei, J., Chalmers, A. D., & Papalopulu, N. (2009).

276

The apicobasal polarity kinase aPKC functions as a nuclear determinant and

regulates cell proliferation and fate during Xenopus primary neurogenesis.

Development, 136(16), 2767–2777. http://doi.org/10.1242/dev.034454

Saka, Y., Tada, M., & Smith, J. C. (2000). A screen for targets of the Xenopus T-box

gene Xbra. Mechanisms of Development, 93(1–2), 27–39.

http://doi.org/10.1016/S0925-4773(00)00260-4

Savova;, V., Pearl;, E. J., Boke;, E., Nag;, A., Adzhubei, I., Marko E. Horb;, & Peshkin,

L. (2017). Transcriptomic Insights into Genetic Diversity of Protein-Coding Genes

in X. laevis. Dev Biol, 424(2), 181–188.

http://doi.org/10.1002/adma.201403943.Evaluating

Scherer, S., & Davis, R. W. (1979). Replacement of chromosome segments with altered

DNA sequences constructed in vitro. Proceedings of the National Academy of

Sciences of the United States of America, 76(10), 4951–5.

http://doi.org/10.1073/pnas.76.10.4951

Schlosser, G., Awtry, T., Brugmann, S. A., Jensen, E. D., Neilson, K., Ruan, G., …

Moody, S. A. (2008). Eya1 and Six1 promote neurogenesis in the cranial placodes

in a SoxB1-dependent fashion. Developmental Biology, 320(1), 199–214.

http://doi.org/10.1016/j.ydbio.2008.05.523

Schohl, A., & Fagotto, F. (2003). A role for maternal β-catenin in early mesoderm

induction in Xenopus. EMBO Journal, 22(13), 3303–3313.

http://doi.org/10.1093/emboj/cdg328

Schroeder, H. W., Cavacini, L., & Cavacini, L. (2010). Structure and function of

immunoglobulins. The Journal of Allergy and Clinical Immunology, 125(2 Suppl 2),

S41-52. http://doi.org/10.1016/j.jaci.2009.09.046

Session, A. M., Uno, Y., Kwon, T., Chapman, J. A., Toyoda, A., Takahashi, S., …

Veenstra, G. J. C. (2016). Genome evolution in the allotetraploid frog Xenopus

277

laevis. Nature, 538(7625), 1–15. http://doi.org/10.1038/nature19840

Shaner, N. C., Campbell, R. E., Steinbach, P. A., Giepmans, B. N. G., Palmer, A. E., &

Tsien, R. Y. (2004). Improved monomeric red, orange and yellow fluorescent

proteins derived from Discosoma sp. red fluorescent protein. Nature Biotechnology,

22(12), 1567–1572. http://doi.org/10.1038/nbt1037

Shcherbo, D., Merzlyak, E. M., Chepurnykh, T. V., Fradkov, A. F., Ermakova, G. V.,

Solovieva, E. A., … Chudakov, D. M. (2007). Bright far-red fluorescent protein for

whole-body imaging. Nature Methods, 4(9), 741–746.

http://doi.org/10.1038/nmeth1083

Sherwood, L. J., & Hayhurst, A. (2012). Hapten mediated display and pairing of

recombinant antibodies accelerates assay assembly for biothreat

countermeasures. Scientific Reports, 2, 807. http://doi.org/10.1038/srep00807

Shintomi, K., Iwabuchi, M., Saeki, H., Ura, K., Kishimoto, T., & Ohsumi, K. (2005).

Nucleosome assembly protein-1 is a linker histone chaperone in Xenopus eggs.

Proceedings of the National Academy of Sciences, 102(23), 8210–8215.

http://doi.org/10.1073/pnas.0500822102

Sievers, F., Wilm, A., Dineen, D., Gibson, T. J., Karplus, K., Li, W., … Higgins, D. G.

(2011). Fast, scalable generation of high-quality protein multiple sequence

alignments using Clustal Omega. Molecular Systems Biology, 7(1), 539–539.

http://doi.org/10.1038/msb.2011.75

Simrick, S., Massé, K., & Jones, E. A. (2005). Developmental expression of Pod 1 in

Xenopus laevis. International Journal of Developmental Biology, 49(1), 59–63.

http://doi.org/10.1387/ijdb.051982ss

Sinner, D. (2006). Global analysis of the transcriptional network controlling Xenopus

endoderm formation. Development, 133(10), 1955–1966.

http://doi.org/10.1242/dev.02358

278

Siontorou, C. G. (2013). Nanobodies as novel agents for disease diagnosis and therapy.

International Journal of Nanomedicine, 8, 4215–4227.

http://doi.org/10.2147/IJN.S39428

Slack, J. M. W., Dale, L., & Smith, J. C. (1984). Analysis of embryonic induction by using

cell lineage markers. Philosophical Transactions of the Royal Society of London.

Series B, Biological Sciences, 307(1132), 331–336.

http://doi.org/10.1098/rstb.1984.0135

Smith, G. P. (1985). Filamentous fusion phage: Novel expression vectors that display

cloned antigens on the virion surface. Science, 228(4705), 1315–1317.

http://doi.org/10.1126/science.4001944

Smits, A. H., Lindeboom, R. G. H., Perino, M., Van Heeringen, S. J., Veenstra, G. J. C.,

& Vermeulen, M. (2014). Global absolute quantification reveals tight regulation of

protein expression in single Xenopuseggs. Nucleic Acids Research, 42(15), 9880–

9891. http://doi.org/10.1093/nar/gku661

Solter, D. (2006). From teratocarcinomas to embryonic stem cells and beyond: A history

of embryonic stem cell research. Nature Reviews Genetics, 7(4), 319–327.

http://doi.org/10.1038/nrg1827

Steer, W. M., Abu-Daya, A., Brickwood, S. J., Mumford, K. L., Jordanaires, N., Mitchell,

J., … Guille, M. J. (2003). Xenopus nucleosome assembly protein becomes tissue-

restricted during development and can alter the expression of specific genes.

Mechanisms of Development, 120, 1045–1057. http://doi.org/10.1016/S0925-

4773(03)00176-X

Strausfeld, U. P., Howell, M., Rempel, R., Maller, J. L. L., Hunt, T., & Blow, J. J. (1994).

Cip1 blocks the initiation of DNA replication in Xenopus extracts by inhibition of

cyclin-dependent kinases. Current Biology : CB, 4(10), 876–83. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/7850420

279

Summerton, J., & Weller, D. (1997). Morpholino Antisense Oligomers: Design,

Preparation, and Properties. Antisense and Nucleic Acid Drug Development, 7(3),

187–195. http://doi.org/10.1089/oli.1.1997.7.187

Sun, L., Bertke, M. M., Champion, M. M., Zhu, G., Huber, P. W., & Dovichi, N. J. (2014).

Quantitative proteomics of Xenopus laevis embryos: expression kinetics of nearly

4000 proteins during early development. Scientific Reports, 4, 4365.

http://doi.org/10.1038/srep04365

Sun, L., Champion, M. M., Huber, P. W., & Dovichi, N. J. (2015). Proteomics of Xenopus

development. Molecular Human Reproduction, 22(3), 193–199.

http://doi.org/10.1093/molehr/gav052

Sun, L., Dubiak, K. M., Peuchen, E. H., Zhang, Z., Zhu, G., Huber, P. W., & Dovichi, N.

J. (2016). Single cell proteomics using frog (Xenopus laevis) blastomeres isolated

from early stage embryos, which form a geometric progression in protein content.

Analytical Chemistry, 88(13), 6653–6657.

http://doi.org/10.1021/acs.analchem.6b01921

Tadros, W., & Lipshitz, H. D. (2009). The maternal-to-zygotic transition: a play in two

acts. Development (Cambridge, England), 136(18), 3033–42.

http://doi.org/10.1242/dev.033183

Takagi, C., Sakamaki, K., Morita, H., Hara, Y., Suzuki, M., Kinoshita, N., & Ueno, N.

(2013). Transgenic Xenopus laevis for live imaging in cell and developmental

biology. Development, Growth & Differentiation, 55(4), 422–433.

http://doi.org/10.1111/dgd.12042

Takuya, N., Blitz, I. L., Fish, M. B., Odeleye, A. O., Manohar, S., Cho, K. W. Y., &

Grainger, R. M. (2014). Cas9-Based Genome Editing in Xenopus tropicalis.

Methods Enzymol, 546(10), 355–375. http://doi.org/10.3174/ajnr.A1256.Functional

Tandon, P., Conlon, F., Furlow, J. D., & Horb, M. E. (2017). Expanding the genetic toolkit

280

in Xenopus: Approaches and Opportunities for Human Disease Modeling.

Developmental Biology, 38(3), 1–22.

http://doi.org/10.1177/0164027515620239.Perceived

Tandon, P., Showell, C., Christine, K., & Conlon, F. L. (2012). Morpholino injection in

Xenopus. Methods in Molecular Biology (Clifton, N.J.), 843, 29–46.

http://doi.org/10.1007/978-1-61779-523-7_4

Tang, M., Miyamoto, Y., & Huang, E. J. (2009). Multiple roles of -catenin in controlling

the neurogenic niche for midbrain dopamine neurons. Development, 136(12),

2027–2038. http://doi.org/10.1242/dev.034330

Tanha, J., Dubuc, G., Hirama, T., Narang, S. a., & MacKenzie, C. R. (2002). Selection

by phage display of llama conventional VH fragments with heavy chain antibody

VHH properties. Journal of Immunological Methods, 263, 97–109.

http://doi.org/10.1016/S0022-1759(02)00027-3

Tarin, D. (1971). Histological features of neural induction in Xenopus laevis. Journal of

Embryology and Experimental Morphology, 26(3), 543–70. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/5146319

Tarin, D. (1972). Ultrastructural features of neural induction in Xenopus laevis. Journal

of Anatomy, 111(Pt 1), 1–28. Retrieved from

http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1271111&tool=pmcentr

ez&rendertype=abstract

Taylor, E. M., Bonsu, N. M., Price, R. J., & Lindsay, H. D. (2013). Depletion of Uhrf1

inhibits chromosomal DNA replication in Xenopus egg extracts. Nucleic Acids

Research, 41(16), 7725–7737. http://doi.org/10.1093/nar/gkt549

Taylor, K. M., & LaBonne, C. (2007, August). Modulating the activity of neural crest

regulatory factors. Current Opinion in Genetics and Development.

http://doi.org/10.1016/j.gde.2007.05.012

281

Thurtle-Schmidt, D. M., & Lo, T.-W. (2018). Molecular biology at the cutting edge: A

review on CRISPR/CAS9 gene editing for undergraduates. Biochemistry and

Molecular Biology Education, 1–11. http://doi.org/10.1002/bmb.21108

Tsai, F. Y., Keller, G., Kuo, F. C., Weiss, M., Chen, J., Rosenblatt, M., … Orkin, S. H.

(1994). An early haematopoietic defect in mice lacking the transcription factor

GATA-2. Nature, 371(6494), 221–226. http://doi.org/10.1038/371221a0

Tsanev, R. (1983, May). Transcriptionally active chromatin. Molecular Biology Reports.

European Molecular Biology Organization. http://doi.org/10.1007/BF00777468

Tsien, R. Y. (1998). the Green Fluorescent. Annual Review of Biochemistry, 199(2), 293–

306. http://doi.org/10.1146/annurev.biochem.67.1.509

Tsuchiya, Y., Murai, S., & Yamashita, S. (2005). Apoptosis-inhibiting activities of BIR

family proteins in Xenopus egg extracts. FEBS Journal, 272(9), 2237–2250.

http://doi.org/10.1111/j.1742-4658.2005.04648.x

Unger, M., Eichhoff, A. M., Schumacher, L., Strysio, M., Menzel, S., Schwan, C., …

Koch-Nolte, F. (2015). Selection of Nanobodies that Block the Enzymatic and

Cytotoxic Activities of the Binary Clostridium Difficile Toxin CDT. Scientific Reports,

5, 7850. http://doi.org/10.1038/srep07850

Van Audenhove, I., Boucherie, C., Pieters, L., Zwaenepoel, O., Vanloo, B., Martens, E.,

… Gettemans, J. (2014). Stratifying fascin and cortactin function in invadopodium

formation using inhibitory nanobodies and targeted subcellular delocalization. The

FASEB Journal, 28(4), 1805–1818. http://doi.org/10.1096/fj.13-242537

Van Audenhove, I., Debeuf, N., Boucherie, C., & Gettemans, J. (2015). Fascin actin

bundling controls podosome turnover and disassembly while cortactin is involved in

podosome assembly by its SH3 domain in THP-1 macrophages and dendritic cells.

Biochimica et Biophysica Acta - Molecular Cell Research, 1853(5), 940–952.

http://doi.org/10.1016/j.bbamcr.2015.01.003

282

Van Der Linden, R., De Geus, B., Stok, W., Bos, W., Van Wassenaar, D., Verrips, T., &

Frenken, L. (2000). Induction of immune responses and molecular cloning of the

heavy chain antibody repertoire of Lama glama. Journal of Immunological Methods,

240, 185–195. http://doi.org/10.1016/S0022-1759(00)00188-5

Vardabasso, C., Manganaro, L., Lusic, M., Marcello, A., & Giacca, M. (2008). The histone

chaperone protein Nucleosome Assembly Protein-1 (hNAP-1) binds HIV-1 Tat and

promotes viral transcription. Retrovirology, 5(1), 8. http://doi.org/10.1186/1742-

4690-5-8

Velculescu, V. E., Zhang, L., Vogelstein, B., & Kinzler, K. W. (1995). Serial Analysis of

Gene Expression. Science (New York, N.Y.), 270(October), 484–488.

http://doi.org/10.1161/01.RES.0000036018.76903.18

Velculescu, V. E., Zhang, L., Zhou, W., Vogelstein, J., Basrai, M. A., Bassett, D. E., …

Kinzler, K. W. (1997). Characterization of the yeast transcriptome. Cell, 88(2), 243–

251. http://doi.org/10.1016/S0092-8674(00)81845-0

Vernon, A. E., Movassagh, M., Horan, I., Wise, H., Ohnuma, S., & Philpott, A. (2006).

Notch targets the Cdk inhibitor Xic1 to regulate differentiation but not the cell cycle

in neurons. EMBO Reports, 7(6), 643–8. http://doi.org/10.1038/sj.embor.7400691

Vernon, A. E., & Philpott, A. (2003). A single cdk inhibitor, p27Xic1, functions beyond cell

cycle regulation to promote muscle differentiation in Xenopus. Development

(Cambridge, England), 130(1), 71–83. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/12441292

Vicente, C., Conchillo, A., García-Sánchez, M. A., & Odero, M. D. (2012, April 1). The

role of the GATA2 transcription factor in normal and malignant hematopoiesis.

Critical Reviews in Oncology/Hematology. Elsevier.

http://doi.org/10.1016/j.critrevonc.2011.04.007

Vize, P. D., Jones, E. A., & Pfister, R. (1995). Development of the Xenopus pronephric

283

system. Developmental Biology. http://doi.org/10.1006/dbio.1995.1302

Walmsley, M. E., Guille, M. J., Bertwistle, D., Smith, J. C., Pizzey, J. A., & Patient, R. K.

(1994). Negative control of Xenopus GATA-2 by activin and noggin with eventual

expression in precursors of the ventral blood islands. Development (Cambridge,

England), 120(9), 2519–2529.

Wang, L., Fisher, L. A., Wahl, J. K., & Peng, A. (2011). Monoclonal Antibodies Against

Xenopus Greatwall Kinase. Hybridoma, 30(5), 469–474.

http://doi.org/10.1089/hyb.2011.0051

Ward, J. D. (2015). Rapid and precise engineering of the caenorhabditis elegans

genome with lethal mutation co-conversion and inactivation of NHEJ repair.

Genetics, 199(2), 363–377. http://doi.org/10.1534/genetics.114.172361

Watanabe, M., Yasuoka, Y., Mawaribuchi, S., Kuretani, A., Ito, M., Kondo, M., …

Kinoshita, T. (2017). Conservatism and variability of gene expression profiles

among homeologous transcription factors in Xenopus laevis. Developmental

Biology, 426(2), 301–324. http://doi.org/10.1016/j.ydbio.2016.09.017

Wen, D., Noh, K. M., Goldberg, A. D., Allis, C. D., Rosenwaks, Z., Rafii, S., &

Banaszynski, L. A. (2014). Genome editing a mouse locus encoding a variant

histone, H3.3B, to report on its expression in live animals. Genesis, 52(12), 959–

966. http://doi.org/10.1002/dvg.22827

Werner, A., Iwasaki, S., McGourty, C. A., Medina-Ruiz, S., Teerikorpi, N., Fedrigo, I., …

Rape, M. (2015). Cell-fate determination by ubiquitin-dependent regulation of

translation. Nature, 525(7570), 523–527. http://doi.org/10.1038/nature14978

Whitley, D. C., Runfola, V., Cary, P., Nazlamova, L., Guille, M. J., & Scarlett, G. (2014).

APTE: Identification of indirect read-out A-DNA promoter elements in genomes.

BMC Bioinformatics, 15(1), 1–8. http://doi.org/10.1186/1471-2105-15-288

Wilson, N. K., Foster, S. D., Wang, X., Knezevic, K., Schütte, J., Kaimakis, P., …

284

Göttgens, B. (2010). Combinatorial transcriptional control in blood stem/progenitor

cells: Genome-wide analysis of ten major transcriptional regulators. Cell Stem Cell,

7(4), 532–544. http://doi.org/10.1016/j.stem.2010.07.016

Wühr, M., Freeman, R. M., Presler, M., Horb, M. E., Peshkin, L., Gygi, S. P., & Kirschner,

M. W. (2014). Deep proteomics of the Xenopus laevis egg using an mRNA-derived

reference database. Current Biology : CB, 24(13), 1467–75.

http://doi.org/10.1016/j.cub.2014.05.044

Wylie, L. A., Hardwick, L. J. A., Papkovskaia, T. D., Thiele, C. J., & Philpott, A. (2015).

Ascl1 phospho-status regulates neuronal differentiation in a Xenopus

developmental model of neuroblastoma. Disease Models & Mechanisms, 8(5),

429–441. http://doi.org/10.1242/dmm.018630

Yamano, H., Gannon, J., Mahbubani, H., & Hunt, T. (2004). Cell Cycle-Regulated

Recognition of the Destruction Box of Cyclin B by the APC/C in Xenopus Egg

Extracts. Molecular Cell, 13(1), 137–147. http://doi.org/10.1016/S1097-

2765(03)00480-5

Yanai, I., Peshkin, L., Jorgensen, P., & Kirschner, M. W. (2011). Mapping Gene

Expression in Two Xenopus Species: Evolutionary Constraints and Developmental

Flexibility. Developmental Cell, 20(4), 483–496.

http://doi.org/10.1016/j.devcel.2011.03.015

Yang, S., Shang, Y., Yin, S., Tian, H., Chen, Y., Sun, S., … Liu, X. (2014). Selection and

identification of single-domain antibody fragment against capsid protein of porcine

circovirus type 2 (PCV2) from C. bactrianus. Veterinary Immunology and

Immunopathology, 160(1–2), 12–19. http://doi.org/10.1016/j.vetimm.2014.03.004

Young, J. J., & Harland, R. M. (2012). Targeted Gene Disruption with Engineered Zinc-

Finger Nucleases (ZFNs). In Methods in molecular biology (Clifton, N.J.) (Vol. 917,

pp. 129–141). http://doi.org/10.1007/978-1-61779-992-1_7

285

Zemella, A., Thoring, L., Hoffmeister, C., & Kubick, S. (2015). Cell-Free Protein

Synthesis: Pros and Cons of Prokaryotic and Eukaryotic Systems. ChemBioChem,

16(17), 2420–2431. http://doi.org/10.1002/cbic.201500340

Zeng, X., Li, K., Yuan, R., Gao, H., Luo, J., Liu, F., … Yan, X. (2018). Nuclear Envelope-

Associated Chromosome Dynamics during Meiotic Prophase I. Frontiers in Cell and

Developmental Biology, 5, 121. http://doi.org/10.3389/fcell.2017.00121

Zhang, R., & Shen, W. (2012). Monoclonal Antibody Expression in Mammalian Cells.

Antibody Engineering. http://doi.org/10.1007/978-1-61779-974-7

Zhang, S.-J., Ma, L.-Y., Huang, Q.-H., Li, G., Gu, B.-W., Gao, X.-D., … Chen, S.-J.

(2008). Gain-of-function mutation of GATA-2 in acute myeloid transformation of

chronic myeloid leukemia. Proceedings of the National Academy of Sciences,

105(6), 2076–2081. http://doi.org/10.1073/pnas.0711824105

Zhang, Y., Zhang, Z., & Ge, W. (2018). An efficient platform for generating somatic point

mutations with germline transmission in the zebrafish by CRISPR/Cas9-mediated

gene editing, 1–23. http://doi.org/10.1074/jbc.RA117.001080

Zlatanova, J., Seebart, C., & Tomschik, M. (2007). Nap1: taking a closer look at a juggler

protein of extraordinary skills. The FASEB Journal, 21(7), 1294–1310.

http://doi.org/10.1096/fj.06-7199rev

Zon, L. I., Mather, Ch., Burgess, S., Bolce, M. E., Harland, R. M., & Orkin, S. H. (1991).

Expression of GATA-binding proteins during embryonic development in Xenopus

laevis. Proceedings of the National Academy of Sciences of the United States of

America, 88(23), 10642–6. http://doi.org/10.1073/PNAS.88.23.10642

286

287

10 Websites used

10.1.1 Phyre2 - Protein folding prediction tool (Kelly, Mezulis, Yates, Wass, & Sternberg, 2015). http://www.sbg.bio.ic.ac.uk/~phyre2/html/page.cgi?id=index

10.1.2 UniPort - Protein database. http://www.uniprot.org/

10.1.3 Xenbase - Xenopus model database. http://www.xenbase.org/entry/

10.1.4 NCBI - Multiple model biotechnological databases and research journal database. https://www.ncbi.nlm.nih.gov/

10.1.5 psipred - Protein folding prediction tool (McGuffin, Bryson, & Jones, 2000). http://bioinf.cs.ucl.ac.uk/psipred/

10.1.6 Google - Search engine. https://www.google.co.uk/

10.1.7 CRISPRscan - gRNA designing tool.

http://www.crisprscan.org/

Clustal Omega - Multiple sequence alignment tool (Sievers et al., 2011).

http://www.clustal.org/omega/

10.1.8 BLAST - Basic local alignment search tool (Altschul, Gish, Miller, Myers, & Lipman, 1990).

https://blast.ncbi.nlm.nih.gov/Blast.cgi

288

11 Appendix I

In-Fusion® HD Cloning Kit User Manual (Clontech® Laboratories, Inc.)

11.1 pOPIN Vectors Maps

11.1.1 pOPINF-his tag at the N-terminal (Addgene Plasmid 26042)

Figure I. 1. pOPINF vector map. Vector containing a his-tag at the N-terminal.

289

11.1.2 pOPINM-MBP - MBP tag and his-tag (Addgene Plasmid 26044)

Figure I. 2. pOPINM vector map. Vector containing an MBP tag, which might enable/ease solubility and purification of the target protein.

290

11.1.3 pOPINJ-JST – JST tag and his-tag (Addgene Plasmid 26045)

Figure I. 3. pOPINJ vector map. Vector containing a GST-tag, which might enable/ease solubility and purification of the target protein.

291

11.1.4 pOPINE – his-tag at the C-terminal (Addgene Plasmid 26043)

Figure I. 4. pOPINF vector map. A vector containing an his-tag at the C-terminal.

292

11.1.5 pOPINTTGneo - his Tag at the C terminal (A kind gift from Ray Owens)

Figure I. 5. pOPINTTGneo vector map. A vector containing an his-tag at the N-terminal and a secretion signal.

293

11.1.6 pOPINEneo-3C-2STREP - his tag and strep-tag at the C-terminal (A kind gift from Ray Owens)

Figure I. 6. pOPINEneo-3C-2STREP vector map. Vector containing 8 x his-tags and 2 x Strep-tags at the C-terminal and a secretion signal.

294

11.1.7 pOPINEneo-3C-3FLAG - his tag and Flag-tag at the C-terminal (A kind gift from Ray Owens)

Figure I. 7. pOPINEneo-3C-3FLAG vector map. Vector containing 8 x his-tags and 3 x Flag-tags at the C-terminal and a secretion signal.

295

11.1.8 pOPINFS - his tag and Flag-tag at the C-terminal (A kind gift from Ray Owens)

Figure I. 8. pOPINFS vector map. Vector containing a StrepII-tag and a partial N-terminal his-tags as well as a secretion signal.

296

11.2 List of Primers

11.2.1 Vector pOPINE: Table I. 1. Primers designed to amplify the desired sequence and anneal to the pOPINE vector. Insert Length Direction Sequence Fw aggagatataccatgatgggacagcagcgca Delta1 FL Rv gtgatggtgatgttttgtatttttttgtttgatcataagcttaccttgg Fw aggagatataccatgatggacggaaccattaaggaagc FL Rv gtgatggtgatgtttgtaaaaaccacctaagtgggactg FliI Fw same as fl version C-term Rv gtgatggtgatgaacctgagaacgttgctgatcccc Fw aggagatataccatgatggaagtggccactgaccag FL Rv gtgatggtgatgttttcccatagccgtcaccatgc Gata2 Fw same as fl version C-term Rv gtgatggtgatgaacgcaggatctgctcttgctcc Fw aggagatataccatgatgccggctgatgtgatgg FL Rv gtgatggtgatgtttccagggcctccaaacagaat HairyI Fw aggagatataccatggccaactgcatgaaccagatca C-term Rv same as fl version Fw aggagatataccatgatggtgctgtcacaaagacaaagag Lis1 FL Rv gtgatggtgatgtttacggcactcccacact Fw aggagatataccatgatggaaaacgaggaaatccctgag FL Rv gtgatggtgatgttttacactaagcggaagcattccc Ndel1-a Fw aggagatataccatgggagagacagacggatggtacaagg C-term Rv same as fl version Fw aggagatataccatgatgaccaaatcgtatggagagaatgg FL Rv gtgatggtgatgtttatcatgaaagatggcatttagctggg NeuroD Fw aggagatataccatggagcccagacctggtgt C-term Rv same as fl version Fw aggagatataccatgatggtgctgctcaagtgcg Neurog2 FL Rv gtgatggtgatgtttaatgaaagcgctgctggca Fw aggagatataccatgatgtatgagaatgtggacgtcagc RARα2 FL Rv gtgatggtgatgtttgtggttggaactgctccagtg Fw aggagatataccatgacaggcagtggaaaagccaag Ncor2 FL Rv gtgatggtgatgtttctcactgtctgacagagtctcatattgg Fw aggagatataccatgatgtccctaaagatgatggagaggc TalI FL Rv gtgatggtgatgtttccgctgcccactgcc Myt1 FL Fw aggagatataccatggatgaatgtagacaatgttaacaa

297

Rv gtgatggtgatgtttaaaccacaccttgatccctttg Fw aggagatataccatggaagatcgcaaataccccggaga C-term Rv gatggtgatgtttaaccaccttgatccctttgactgcttg Fw aggagatataccatgatggctgctttccacatcgc XicI FL Rv gtgatggtgatgttttcgaatctttttcctgggggtctg

11.2.2 Vector pOPINE-Strep II (C-his tag): Table I. 2. Primers designed to amplify the desired sequence and anneal to the pOPINE- Strep II vector. Insert Length Direction Sequence AGGAGATATACCATGGGCCTGTGGTCTCTATCACAAAAT Fw Gata2 C-term GAAT Rv CAGAACTTCCAGTTTAAACGCCAAGTGACCCATAGGAGC Fw aggagatataccatggccaactgcatgaaccagatca HairyI C-term Rev cagaacttccagtttccagggcctccaaacagaatca Fw aggagatataccatggatggacgagtacactaaaatagagaagattgg cdk1 FL Rv cagaacttccagtttaaactttctaatctgattggcgggaaggct

11.2.3 Vectors pOPINF - pOPINM –pOPINJ: Table I. 3. Primers designed to amplify the desired sequence and anneal to the pOPINF, pOPINM and pOPINJ vectors. Insert Length Direction Sequence Fw aagttctgtttcagggcccgatgggacagcagcgcat Delta1 FL Rv ctggtctagaaagctttatgtatttttttgtttgatcataagcttaccttggttaca Fw aagttctgtttcagggcccgatggacggaaccattaaggaagctc FL FliI Rv ctggtctagaaagctctagtaaaaaccacctaagtgggactgc Fw same as fl version C-term Rv ctggtctagaaagcttctgagaacgttgctgatcccc Fw aagttctgtttcagggcccgatggaagtggccactgacca FL Rv ctggtctagaaagctttatcccatagccgtcaccatgc Gata2 Fw same as fl version C-term Rv ctggtctagaaagcttggatctgctcttgctcctctgt Fw aagttctgtttcagggcccgatgccggctgatgtgatgg FL Rv ctggtctagaaagctttaccagggcctccaaacagaat HairyI Fw aagttctgtttcagggtaccgccaactgcatgaaccagat C-term Rv same as fl version

298

Fw aagttctgtttcagggcccgatggtgctgtcacaaagac Lis1 FL Rv ctggtctagaaagcttcaacggcactcccacact Fw gttctgtttcagggcccgatggaaaacgaggaaatccct FL Rv ctggtctagaaagcttcatacactaagcggaagcattccc Ndel1-a Fw aggagatataccatgggagagacagacggatggtacaagg C-term Rv same as fl version Fw aagttctgtttcagggcccgatgaccaaatcgtatggagagaatgg FL Rv ctggtctagaaagctttaatcatgaaagatggcatttagctggg NeuroD Fw aagttctgtttcagggtaccaaaagcccagacctggtgt C-term Rv same as fl version Fw aagttctgtttcagggcccgatggtgctgctcaagtgc Neurog2 FL Rv ctggtctagaaagcttcaaatgaaagcgctgctggc Fw aagttctgtttcagggcccgatgtatgagaatgtggacgtcagc RARα2 FL Rv ctggtctagaaagcttcagtggttggaactgctccagtg Fw aagttctgtttcagggcccgacaggcagtggaaaagccaag Ncor2 FL Rv ctggtctagaaagctctcactgtctgacagagtctcatattgg Fw aagttctgtttcagggcccgatgtccctaaagatgatggagaggc TalI FL Rv ctggtctagaaagcttcaccgctgcccactgcc Fw gttctgtttcagggtaccatgaatgtagacaatgttaaca FL Myt1 Rv tggtctagaaagctttcacaccttgatccctttga Fw ttctgtttcagggtaccgaagatcgcaaataccccggaga C-term Rv ctggtctagaaagcttcaccttgatccctttgactgcttg Fw aagttctgtttcagggcccgatggctgctttccacatcgc XicI FL Rv ttagtctagaaagcttcgaatctttttcctgggggtct

11.2.4 Vector pOPINGM (ss-MBP-C-his tag): Table I. 4. Primers designed to amplify the desired sequence and anneal to the pOPINGM vector. Insert Length Direction Sequence Fw aagttctgtttcagggtacccctgtggtctctatcacaaaatgaatgg Gata2 C-term Rv gtgatggtgatgtttaaacgccaagtgacccataggagc Fw aagttctgtttcagggtaccgccaactgcatgaaccagatcaat HairyI C-term Rv gtgatggtgatgtttaaaccagggcctccaaac Fw aagttctgtttcagggtacccatggacgagtacactaaaatagagaagat cdk1 FL Rv gtgatggtgatgtttaaacttaatttctaatctgattggcgggaaggc pOPINGM tgcgtactgcggtgatcaacg Seq

299

pOPINGM atgtccgctttctggtatgccg Seq2

11.2.5 Vector pOPINTTGneo: Table I. 5. Primers designed to amplify the desired sequence and anneal to the pOPINTTGneo vector. Insert Length Direction Sequence Fw gcgtagctgaaaccggtacctgtggtctctatcacaaaat Gata2 C-term Rv gtgatggtgatgttctgccaagtgacccataggagccat Fw gcgtagctgaaaccggtaccacaggcagtggaaaagccaa Ncor2 C-term Rv gtgatggtgatgtttaaactcactgtctgacagagtctca Fw tagctgaaaccggtaccatgtccctaaagatgatggagag N-term Rv gtgatggtgatgtttaaaccagaattcagcagcaggggtt TalI Fw tagctgaaaccggtaccgaagtggaaatctctgaaggtcc C-term Rv gtgatggtgatgtttaaacatgtcttgaagcagttcttgc Fw tagctgaaaccggtaccaaaaagaaaaaaatgacgaaagc N-term Rv gtgatggtgatgtttaaacctggggttcagctgcagac NeuroD Fw gcgtagctgaaaccggtacccctcactcctatggggcgg C-term Rv gtgatggtgatgtttaaactcatgaaagatggcatttagc Fw gcgtagctgaaaccggtaccaagagcggagaaactgtgtt Neurog2 C-term Rv gtgatggtgatgtttaaacatgaaagcgctgctggca Fw gcgtagctgaaaccggtaccagtggccaaattcagctgtg FliI C-term Rv gatggtgatgtttaaactcagaaggatatttgtacattga Fw gcgtagctgaaaccggtaccatggctgctttccacatcg N-term Rv atggtgatggtgatgtaatggcgttgtgttggcagct XicI Fw gcgtagctgaaaccggtaccatggctgctttccacatcg Fl Rv gtgatggtgatgtttaatcgaatctttttcctgggggtct Fw gcgtagctgaaaccggtacccccctcttccacggtgact C-term Rv gtgatggtgatgaacatttctaatctgattggcgggaagg Fw gctgaaaccggtaccatggacgagtacactaaaatagaga cdk1 N-term Rv tggtgatggtgatgaacatcgatatactggccgct Rv-New tggtgatggtgatgaacatactggccgctgggta Fw gctgaaaccggtaccatggacgagtacactaaaatagaga Fl Rv gtgatggtgatgaacatttctaatctgattggcgggaagg

Sequencing Fw cgctgctctccctcgtgagcct primer Rv cggccgcctgcacctgaggttaat

Sequencing Fw acaatcaaaggagatataccatggggatccttccc primer 2 Rv tgaaaataaatttcctttattagccagaggtcgaggtcgg

300

pTT-fwd TCC ACA GGT GTC CAC TCC

11.2.6 Vector pOPINE-Flag3 (C-his tag) Table I. 6. Primers designed to amplify the desired sequence and anneal to the pOPINE- Flag3 vector. Insert Length Direction Sequence Fw aggagatataccatggcctgtggtctctatcacaaaatgaat Gata2 C-term Rv cagaacttccagttttgccaagtgacccataggagc Fw aggagatataccatgatgccggctgatgtgatgg HairyI FL Rv cagaacttccagtttccagggcctccaaacagaatca Fw aggagatataccatgatggacgagtacactaaaatagagaagattgg cdk1 FL Rv cagaacttccagtttatttctaatctgattggcgggaagg Fw aggagatataccatggccaactgcatg HairyI C-term Rv cagaacttccagtttccagggcctccaaacag

301

11.3 In-Fusion reaction overview

Figure I. 9. Outline of In-Fusion PCR Cloning System. Linear vector and PCR products with extended regions are cloned and finally transformed into competent cells.

302

11.4 Protein expression via transfection into HEK293T cells

11.4.1 Cdk1 Table I. 7. CDK1 protein expression via transfection into HEK293T cells. Target Sequenced Insert Vector PCR Expression protein clones no Cdk1 N-Term pOPINTTGneo amplification no Cdk1 C-Term pOPINTTGneo amplification no Cdk1 FL pOPINTTGneo amplification Cdk1 C-Term pOPINTTGneo OK Miss-matches Cdk1 N-Term pOPINTTGneo OK Miss-matches Cdk1 FL pOPINTTGneo OK Miss-matches Cdk1 C-Term pOPINTTGneo OK Miss-matches Cdk1 FL pOPINTTGneo OK Miss-matches Cdk1 FL pOPINGM OK OK no expression Cdk1 FL pOPINFS OK OK no expression

Cdk1 FL pOPINE-StrepII OK OK no expression

Cdk1 FL pOPINE-Flag3 OK OK no expression Cdk1 FL pOPINGM OK Miss-matches Cdk1 FL pOPINGM OK Miss-matches Cdk1 FL pOPINGM OK 1 changes in AAs Cdk1 FL pOPINGM OK OK no expression Cdk1 FL pOPINGM OK OK no expression Cdk1 FL pOPINGM OK Miss-matches Cdk1 FL pOPINGM OK Miss-matches Cdk1 FL pOPINGM OK Miss-matches Cdk1 FL pOPINGM OK Miss-matches Cdk1 FL pOPINGM OK Miss-matches Cdk1 FL pOPINFS OK OK no expression Cdk1 FL pOPINFS OK Miss-matches Cdk1 FL pOPINFS OK Miss-matches Cdk1 FL pOPINFS OK OK no expression Cdk1 FL pOPINFS OK OK no expression

303

Cdk1 FL pOPINFS OK OK no expression

Cdk1 FL pOPINFS OK OK Band Cdk1 FL pOPINFS OK Miss-matches

Cdk1 FL pOPINFS OK OK Band

Cdk1 FL pOPINE-Flag3 OK OK no expression

1 point mutations Cdk1 FL pOPINE-Flag3 OK no expression no changes in AAs Cdk1 FL pOPINE-Flag3 OK Empty vector Cdk1 FL pOPINE-Flag3 OK Miss-matches Cdk1 FL pOPINE-Flag3 OK OK Band

Cdk1 FL pOPINE-Flag3 OK 1 changes in AAs no expression

Cdk1 FL pOPINE-Flag3 OK Empty vector Cdk1 FL pOPINE-Flag3 OK OK no expression Cdk1 FL pOPINE-Flag3 OK OK no expression Cdk1 FL pOPINE-Flag3 OK Empty vector

Cdk1 FL pOPINE-StrepII OK Miss-matches

1 point mutations Cdk1 FL pOPINE-StrepII OK no expression no changes in AAs

Cdk1 FL pOPINE-StrepII OK Miss-matches

1 point mutations Cdk1 FL pOPINE-StrepII OK no expression no changes in AAs 1 point mutations Cdk1 FL pOPINE-StrepII OK no expression no changes in AAs

Cdk1 FL pOPINE-StrepII OK Miss-matches

Cdk1 FL pOPINE-StrepII OK OK no expression

Cdk1 FL pOPINE-StrepII OK Miss-matches

304

Cdk1 FL pOPINFS OK OK no expression Cdk1 FL pOPINFS OK OK no expression Cdk1 FL pOPINE-Flag3 OK OK no expression Cdk1 FL pOPINE-Flag3 OK OK no expression

Cdk1 FL pOPINE-StrepII OK OK no expression

Cdk1 FL pOPINFS OK OK no expression Cdk1 FL pOPINFS OK OK no expression Cdk1 FL pOPINFS OK Miss-matches Cdk1 FL pOPINFS OK Miss-matches Cdk1 FL pOPINE-Flag3 OK OK no expression Cdk1 FL pOPINE-Flag3 OK OK no expression Cdk1 FL pOPINFS OK Miss-matches Cdk1 FL pOPINFS OK Miss-matches Cdk1 FL pOPINFS OK Miss-matches Cdk1 FL pOPINFS OK OK Cdk1 FL pOPINFS OK OK Cdk1 FL pOPINFS OK OK Cdk1 FL pOPINE-Flag3 OK OK Cdk1 FL pOPINE-Flag3 OK OK Cdk1 FL pOPINE-Flag3 OK OK Cdk1 FL pOPINFS OK OK Cdk1 FL pOPINFS OK OK Cdk1 FL pOPINFS OK OK

Cdk1 FL pOPINE-StrepII OK OK

Cdk1 FL pOPINE-Flag3 OK OK Cdk1 FL pOPINE-Flag3 OK OK Cdk1 FL pOPINE-Flag3 OK OK

305

11.4.2 Delta1 Table I. 8. Delta1 protein expression via transfection into HEK293T cells.

Target Insert Vector PCR Sequenced clones Expression Protein

2 bands, OK match with Delta 1 FL pOPINF 900 bp and different region of 1800 bp DNA

Delta 1 FL pOPINE 400 bp band

2 bands, Delta 1 FL pOPINJ 900 bp and 1800 bp

11.4.3 FliI Table I. 9. FliI protein expression via transfection into HEK293T cells.

Target Sequenced Insert Vector PCR Expression Protein clones no FliI FL pOPINE amplification no FliI FL pOPINF amplification 1 mismatch OK - new FliI FL pOPINE generated one stop no expression template codon

OK - new mega prep too low FliI FL pOPINF template yield

FliI C-Term pOPINF OK OK no expression

FliI C-Term pOPINE OK mismatches FliI C-Term pOPINJ OK mismatches no FliI C-Term pOPINTTGneo amplification Temp FliI C-Term pOPINTTGneo gradient - no amplification no FliI C-Term pOPINTTGneo amplification no FliI C-Term pOPINTTGneo amplification

306

11.4.4 Gata2 Table I. 10. Gata2 protein expression via transfection into HEK293T cells. Target Sequenced Insert Vector PCR Expression Protein clones increase extension to 2'/gel extract Gata2 FL pOPINE the band/design short version increase extension to 2'/gel extract Gata2 FL pOPINF the band/design short version Gata2 FL pOPINF OK OK no expression Gata2 FL pOPINE OK OK no expression no Gata2 FL pOPINF amplification no Gata2 FL pOPINE amplification Gata2 FL pOPINJ OK empty vector Gata2 C-Term pOPINTTGneo OK miss matches very good Gata2 C-Term pOPINTTGneo OK OK expression Gata2 C-Term pOPINGM OK OK no expression Gata2 C-Term pOPINFS OK OK no expression OK Gata2 C-Term pOPINE-StrepII OK no expression

Gata2 C-Term pOPINE-Flag3 OK OK no expression Reads through Gata2 C-Term pOPINGM OK AMPr

Gata2 C-Term pOPINGM OK 2 changes in AAs

Reads through Gata2 C-Term pOPINGM OK AMPr

307

Reads through Gata2 C-Term pOPINGM OK AMPr Gata2 C-Term pOPINGM OK 2 changes in AAs Reads through Gata2 C-Term pOPINGM OK AMPr Reads through Gata2 C-Term pOPINGM OK AMPr Reads through Gata2 C-Term pOPINGM OK AMPr Reads through Gata2 C-Term pOPINGM OK AMPr Gata2 C-Term pOPINGM OK empty vector Gata2 C-Term pOPINFS OK Miss-matches

Gata2 C-Term pOPINFS OK 4 changes in AAs

Gata2 C-Term pOPINFS OK Miss-matches Gata2 C-Term pOPINFS OK Miss-matches Gata2 C-Term pOPINFS OK 2 changes in AAs Gata2 C-Term pOPINFS OK Miss-matches Not Gata2 C-Term pOPINFS OK OK transfected Gata2 C-Term pOPINFS OK Miss-matches Gata2 C-Term pOPINFS OK Miss-matches

Gata2 C-Term pOPINFS OK OK no expression

Gata2 C-Term pOPINE-Flag3 OK Miss-matches Gata2 C-Term pOPINE-Flag3 OK Miss-matches Gata2 C-Term pOPINE-Flag3 OK Miss-matches

Gata2 C-Term pOPINE-Flag3 OK 2 changes in AAs

Gata2 C-Term pOPINE-Flag3 OK Miss-matches Gata2 C-Term pOPINE-Flag3 OK Very low yield Gata2 C-Term pOPINE-Flag3 OK Very low yield Gata2 C-Term pOPINE-Flag3 OK Very low yield Gata2 C-Term pOPINE-Flag3 OK Very low yield

308

Gata2 C-Term pOPINE-StrepII OK Miss-matches

Gata2 C-Term pOPINE-StrepII OK Very low yield

Gata2 C-Term pOPINE-StrepII OK Miss-matches

Gata2 C-Term pOPINE-StrepII OK Very low yield

Gata2 C-Term pOPINE-StrepII OK 2 changes in AAs

1 changes in AAs Gata2 C-Term pOPINE-StrepII OK

Gata2 C-Term pOPINE-StrepII OK 1 changes in AAs Gata2 C-Term pOPINFS OK OK no expression Gata2 C-Term pOPINFS OK OK no expression

Gata2 C-Term pOPINE-StrepII OK 2 changes in AAs

1 changes in AAs Gata2 C-Term pOPINE-StrepII OK

Gata2 C-Term pOPINE-StrepII OK 2 changes in AAs

309

Gata2 C-Term pOPINE-StrepII OK 2 changes in AAs

Not Gata2 C-Term pOPINFS OK OK transfected Not Gata2 C-Term pOPINFS OK OK transfected Not Gata2 C-Term pOPINFS OK OK transfected 1 changes in AAs Gata2 C-Term pOPINE-StrepII OK

1 changes in AAs Gata2 C-Term pOPINE-StrepII OK

11.4.5 HairyI Table I. 11. HairyI protein expression via transfection into HEK293T cells. Target Insert Vector PCR Sequenced clones Expression Protein need temp gradient/not good/ HairyI FL pOPINE designed short version need temp gradient/not good/ HairyI FL pOPINF designed short version HairyI C-Term pOPINF OK OK no expression

310

Good expression - HairyI C-Term pOPINE OK OK faint in lysate but most stuck in cells HairyI C-Term pOPINJ OK miss-matches HairyI C-Term pOPINGM OK OK no expression HairyI C-Term pOPINFS OK OK no expression HairyI C-Term pOPINE-StrepII OK OK no expression HairyI C-Term pOPINE-Flag3 OK OK no expression band wrong HairyI FL pOPINE-Flag3 size HairyI C-Term pOPINGM OK miss-matches HairyI C-Term pOPINGM OK miss-matches HairyI C-Term pOPINGM OK OK no expression HairyI C-Term pOPINGM OK miss-matches HairyI C-Term pOPINGM OK OK no expression HairyI C-Term pOPINGM OK miss-matches HairyI C-Term pOPINGM OK miss-matches HairyI C-Term pOPINGM OK miss-matches HairyI C-Term pOPINGM OK miss-matches HairyI C-Term pOPINGM OK miss-matches HairyI C-Term pOPINFS OK miss-matches HairyI C-Term pOPINFS OK miss-matches HairyI C-Term pOPINFS OK miss-matches HairyI C-Term pOPINFS OK miss-matches

HairyI C-Term pOPINFS OK miss-matches

HairyI C-Term pOPINFS OK 1 changes in AAs

HairyI C-Term pOPINFS OK OK no expression

HairyI C-Term pOPINFS OK Miss-matches

HairyI C-Term pOPINE-StrepII OK 1 changes in AAs

HairyI C-Term pOPINE-StrepII OK OK no expression

311

HairyI C-Term pOPINE-StrepII OK OK Good Band

HairyI C-Term pOPINE-StrepII OK OK Band

HairyI C-Term pOPINE-StrepII OK OK no expression HairyI C-Term pOPINE-StrepII OK Miss-matches good HairyI C-Term pOPINE-StrepII OK OK expression good HairyI C-Term pOPINE-StrepII OK OK expression HairyI C-Term pOPINE-StrepII OK OK no expression good HairyI C-Term pOPINE-StrepII OK OK expression HairyI C-Term pOPINE-Flag3 OK OK no expression HairyI C-Term pOPINE-Flag3 OK miss-matches HairyI C-Term pOPINE-Flag3 OK miss-matches HairyI C-Term pOPINE-Flag3 OK miss-matches HairyI C-Term pOPINE-Flag3 OK miss-matches

HairyI C-Term pOPINE-Flag3 OK 1 changes in AAs

good HairyI C-Term pOPINE-Flag3 OK OK expression good HairyI C-Term pOPINE-Flag3 OK OK expression good HairyI C-Term pOPINE-Flag3 OK OK expression HairyI C-Term pOPINE-Flag3 OK Miss-matches HairyI C-Term pOPINFS OK OK no expression HairyI C-Term pOPINFS OK OK no expression HairyI C-Term pOPINFS OK OK no expression HairyI C-Term pOPINFS OK OK no expression HairyI C-Term pOPINFS OK OK no expression HairyI C-Term pOPINFS OK OK no expression HairyI C-Term pOPINFS OK OK no expression HairyI C-Term pOPINFS OK OK no expression good HairyI C-Term pOPINE-StrepII OK OK expression good HairyI C-Term pOPINE-StrepII OK OK expression

312

good HairyI C-Term pOPINE-StrepII OK OK expression

HairyI C-Term pOPINFS OK miss-matches HairyI C-Term pOPINFS OK miss-matches HairyI C-Term pOPINFS OK miss-matches HairyI C-Term pOPINFS OK miss-matches HairyI C-Term pOPINFS OK miss-matches HairyI C-Term pOPINFS OK miss-matches HairyI C-Term pOPINE-StrepII OK OK HairyI C-Term pOPINE-StrepII OK OK HairyI C-Term pOPINE-StrepII OK OK HairyI C-Term pOPINFS OK 1 changes in AAs HairyI C-Term pOPINFS OK 1 changes in AAs HairyI C-Term pOPINFS OK 1 changes in AAs HairyI C-Term pOPINFS OK OK HairyI C-Term pOPINFS OK OK HairyI C-Term pOPINFS OK OK

11.4.6 Myt1 Table I. 12. Myt1 protein expression via transfection into HEK293T cells. Target Insert Vector PCR Sequenced clones Expression Protein no MytI FL pOPINF amplification no MytI FL pOPINE amplification need to increase MytI FL pOPINF empty vector extension to 3' need to increase MytI FL pOPINE empty vector extension to 3' MytI FL pOPINJ OK empty vector

313

tried 2 pOPINF - different MytI C-Term pOPINM - templates- pOPINJ none work tried 2 different MytI C-Term pOPINE templates- none work different pOPINF - polymerase MytI C-Term pOPINM - - did not pOPINJ amplify Tried different MytI C-Term pOPINE polymerase - did not amplify Temp pOPINF - gradient - MytI C-Term pOPINM - did not pOPINJ amplify Temp gradient - MytI C-Term pOPINE did not amplify

11.4.7 NeuroD Table I. 13. NeuroD protein expression via transfection into HEK293T cells. Target Insert Vector PCR Sequenced clones Expression Protein NeuroD FL pOPINE OK NeuroD FL pOPINF OK faint NeuroD Short pOPINE OK OK expression on Lysate faint NeuroD Short pOPINF OK OK expression on Lysate

314

NeuroD Short pOPINJ OK miss-matches NeuroD N-Term pOPINTTGneo OK NeuroD C-Term pOPINTTGneo OK

11.4.8 Neurog2 Table I. 14. Neurog2 protein expression via transfection into HEK293T cells. Target Sequenced Insert Vector PCR Expression Protein clones no Neurog2 C-Term pOPINTTGneo amplification temp gradient - Neurog2 C-Term pOPINTTGneo did not amplify

no Neurog2 FL pOPINE amplification

no Neurog2 FL pOPINF amplification

no Neurog2 C-Term pOPINTTGneo amplification good Neurog2 C-Term pOPINTTGneo OK OK expression no Neurog2 C-Term pOPINTTGneo amplification Neurog2 C-Term pOPINTTGneo OK no colonies Neurog2 C-Term pOPINTTGneo OK no colonies

11.4.9 Rarα2 Table I. 15. Rarα2 protein expression via transfection into HEK293T cells. Target Insert Vector PCR Sequenced clones Expression Protein

Rarα2 Short pOPINE OK OK no expression

Rarα2 Short pOPINF OK OK no expression

315

11.4.10 Ncor2 Table I. 16. Ncor2 protein expression via transfection into HEK293T cells. Target Sequenced Insert Vector PCR Expression Protein clones

Ncor2 Short pOPINE OK OK no expression

no Ncor2 Short pOPINF amplification Ncor2 Short pOPINTTGneo OK miss-matches

Ncor2 Short pOPINTTGneo OK OK no expression

11.4.11 TalI Table I. 17. TalI protein expression via transfection into HEK293T cells. Protein Sequenced Insert Vector PCR Expression Target clones faint shorter band than TalI FL pOPINE expected - tried 2 templates faint shorter band than TalI FL pOPINF expected - tried 2 templates multiple bands - not TalI FL pOPINE the correct size multiple bands - not TalI FL pOPINF the correct size no TalI N-Term pOPINTTGneo amplification

316

no TalI C-Term pOPINTTGneo amplification Temp gradient - TalI N-Term pOPINTTGneo did not amplify Temp gradient - TalI C-Term pOPINTTGneo did not amplify no TalI N-Term pOPINTTGneo amplification no TalI C-Term pOPINTTGneo amplification TalI N-Term pOPINTTGneo OK no colonies no TalI C-Term pOPINTTGneo amplification

11.4.12 XicI Table I. 18. XicI protein expression via transfection into HEK293T cells. Target Insert Vector PCR Sequenced clones Expression Protein no XicI FL pOPINE OK OK amplification wrong size XicI FL pOPINF band XicI N-Term pOPINTTGneo OK miss-matches XicI FL pOPINTTGneo OK miss-matches

no XicI N-Term pOPINTTGneo OK OK amplification

317

12 Appendix II

Some of the most common primers used to amplify VHH specific sequences for the

development of libraries:

CALL001: 5’ GTC CTG GCT GCT CTT CTA CAA GG 3’ (Rothbauer, et al., 2006);

(Pardon, et al., 2014)

CALL002: 5’ GGT ACG TGC TGT TGA ACT GTT CC 3’ (Rothbauer, et al., 2006);

(Pardon, et al., 2014)

VHH-Back: 5’ GAT GTG CAG ctg cag GAG TCT GGR GGA GG 3’ (PstI) (Pardon, et al.,

2014)

VHH-For: 5’ CTA GTG CGG CCG CTG GAG ACg gtg acc TGG GT 3’ (Eco91I) (Pardon,

et al., 2014)

MP57: TTA TGC TTC CGG CTC GTA TG 3’ (Pardon, et al., 2014)

GIII: 5’ CCA CAG ACA GCC CTC ATA G 3’ (Pardon, et al., 2014)

SM017: 5’ CCA GCC GGC CAT GGC TCA GGT GCA GCT GGT GGA GTC TGG 3’

(Rothbauer, et al., 2006)

SM018: 5’ CCA GCC GGC CAT GGC TGA TGT GCA GCT GGT GGA GTC TGG 3’

(Rothbauer, et al., 2006)

318

12.1 Insert verification by sequence BLAST (Xenbase.org)

The sequence of interest was isolated and amplified from the original vectors by PCR, which in turn added ends to a product which is homologous to the destination vector.

These were cloned by In-Fusion reaction and verified by plasmid prep followed by sequencing. The sequences were confirmed using the BLAST tool from Xenbase.

12.1.1.1 pOPINF-RARα2

Figure II. 1. pOPINF-RARα2 plasmid sequence verification.

319

12.1.1.2 pOPINF-NeuroD

Figure II. 2.pOPINF-NeuroD plasmid sequence verification.

320

12.1.1.3 pOPINF-FliI

Figure II. 3. pOPINF-FliI plasmid sequence verification.

321

12.1.1.4 pOPINE-RARα2

Figure II. 4. pOPINE-RARα2 plasmid sequence verification.

322

12.1.1.5 pOPINE-NeuroD

Figure II. 5. pOPINE-NeuroD plasmid sequence verification.

323

12.1.1.6 pOPINE-FliI

Figure II. 6. pOPINE-FliI plasmid sequence verification.

324

407081 13 Appendix III

13.1 Protein targets for commercial synthesis

13.1.1 ascl1 achaete-scute homolog 1 [Xenopus laevis]

NCBI Reference Sequence: NP_001079247.1

aavarrnere rnrvklvnlg fatlrehvpn gaankkmskv etlrsaveyi ralqqlldeh davsaafqsg

13.1.2 ccna1 cyclin A1 L homeolog [Xenopus laevis]

NCBI Reference Sequence: NP_001087670.1

mrrsmvssgh ilttssvvga ssafqnrcla kvevqpnlpq rtvlgvisdn eqhrrsisrg gapakclpgi envlpfpgkf lsanpapvvp kpsftvyvde peqptetysi evdcpsldev dsnlvkqnih llldisaasp mvvdasfqts qeddsitdpd avavseyide ihqylreael knrpkayymr kqpditsamr tilvdwliev geeyklrtet lylavnyldr flscmsvlrg klqlvgtaai llaskyeeiy ppgvdefvyi tddtyskkql lrmehlllkv lafdltvptt sqfllqylqr havsvktehl amylaeltlf evepflkyvp sltaaaaycl anyalnkvfw petleaftgy tlseiapcls dmhqaclhap yqaqqairek yktpkymqvs llempatlpl n

13.1.3 ccne1 G1/S-specific cyclin-E3 [Xenopus laevis]

NCBI Reference Sequence: NP_001081446.1

hepnptsyph faslrfspvs asplprlgwa nqddvwrnml nkdriylrdk nffekhpqlq pnmrailldw lmevcevykl hretfylaqd ffdrfmatqk nviksrlqli gitslfiaak meeiyppklh qfafitdcac tedeitsmel iimkdldwcl spmtmvswfn vflqvayire lqhflrpqfp qevyiqivql ldlcvldicc ldypygvlaa salyhfscpe lmekvsgfkl telqgcikwl vpfamaikdg gksklkffkg vdiedvhniq thtgclelme

13.1.4 cdk1 cyclin-dependent kinase 1-B [Xenopus laevis]

NCBI Reference Sequence: NP_001080093.1

mdeytkieki gegtygvvyk grhkatgqvv amkkirlene eegvpstair eisllkelqh pnivclldvl mqdsrlylif eflsmdlkky ldsipsgqyi dtmlvksyly qilqgivfch srrvlhrdlk pqnllidnkg vikladfgla rafgipvrvy

325

407081 thevvtlwyr asevllgsvr ystpvdvwsv gtifaeiatk kplfhgdsei dqlfrifrsl gtpnnevwpe veslqdyknt fpkwkggsls snvknidedg ldllskmlvy dpakrisark amlhpyfddl dksslpanqi rn

13.1.5 cdk2 cyclin-dependent kinase 2 [Xenopus laevis]

NCBI Reference Sequence: NP_001084120.1 menfqkveki gegtygvvyk arnretgeiv alkkirldte tegvpstair eisllkelnh pnivklldvi htenklylvf eflnqdlkkf mdrsnisgis lalvksylfq llqglafchs hrvlhrdlkp qnllinsdga ikladfglar afgvpvrtft hevvtlwyra peillgckfy stavdiwslg cifaemitrr alfpgdseid qlfrifrtlg tpdevswpgv ttmpdykstf pkwirqdfsk vvppldedgr dllaqmlqyd snkrisakva lthpffrdvs rptphli

13.1.6 cdknx cyclin-dependent kinase inhibitor XicI S homeolog [Xenopus laevis]

NCBI Reference Sequence: NP_001108275.1 maafhialqe emisapavlp rlsagtgrga crnlfgpidh demrselkrq lkeiqasdcq rwnfdfetgt plkgifcwep veskdmpsfy sqnrsiaant

13.1.7 dazl deleted in azoospermia-like-A [Xenopus laevis]

NCBI Reference Sequence: NP_001081772.1 msgkeessny aataeeeavn qgfvlpegei mpntvfvggi ditmdeieir dfftrfgnvk evkiitdrtg vskgygfisf sdevdvqkiv ksqisfhgkk lklgpairki ctyvqprpvv

13.1.8 etv2 ets variant 2 L homeolog [Xenopus laevis]

NCBI Reference Sequence: NP_001089600.1 dkkplptass gpiqlwqfll ellqdsscqk liswtgngwe fklsdpneva rrwgrrknkp rmnyeklsrg lryyyhknii hktggqryvy rfvcdlqdll srptqtsqpp tktqrsriq

13.1.9 fli1 retroviral integration site protein Fli-1 homolog [Xenopus laevis]

326

407081 NCBI Reference Sequence: NP_001084371.1 vdcsinkcsk liggsegnam tytymdekng ppppnmttne rrvivpadpa lwsqdhvrqw lewaikeygl veidcslfqn idgkelckms kedflrstsi yntevllshl nylrdssssl gyntqahtdq ssrltakedp syeavrrsgw gnsmsspvtk sppmggtqnv nksgdqqrsq pdpyqilgpt ssrlanpgsg qiqlwqflle llsdssnasc itwegtngef kmtdpdevar rwgerkskpn mnydklsral ryyydksimt kvhgkryayk fdfhgiaqal qphpt

13.1.10 fus1 FUS RNA binding protein L homeolog [Xenopus laevis]

NCBI Reference Sequence: NP_001080383.1 gfskfggphe gggpprnema eqdnsdnnti fvqglgenvt vesvadyfkq igiikinkkt gqpminlytd retgklkgea tvsfddppsa kaaidwfdgk efsgnpikvs fatrradfns rgggngrgrg rggpmgrggf ggppggsssr ggsggypsgg ggqqragdwk cpnpgcenmn fswrnecnqc kapkpegpgg

13.1.11 gata2 GATA-binding factor 2 [Xenopus laevis]

NCBI Reference Sequence: NP_001084043.1

Hdyssglfhp gsllggpass ftpkqrsksr scsegrecvn cgatatplwr rdgtghylcn acglyhkmng qnrplikpkr rlsaarragt ccancqtstt tlwrrnangd pvcnacglyy klhnvnrplt mkkegiqtrn rkmsnkskkn

13.1.12 hes1 transcription factor HES-1-A [Xenopus laevis]

NCBI Reference Sequence: NP_001081396.1 mpadvmekns sspvaatpas vsntpdkpkt asehrksskp imekrrrari neslgqlktl ildalkkdss rhsklekadi lemtvkhlrn lqrvqmsaal stdpsvlgky ragfsecmne vtrflstceg vntdvrtrll ghlancmnqi

13.1.13 lis1 lissencephaly-1 homolog [Xenopus laevis]

NBI Reference Sequence: NP_001083934.1 mvlsqrqrde lnraiadylr sngyeeaysv fkkeaeldmn eeldikyagl lekkwtsvir lqkkvmeles klneakeeft sggpigqkrd pkewiprppe

327

407081 13.1.14 myt1 myelin transcription factor 1 [Xenopus laevis]

NCBI Reference Sequence: NP_001081661.1 rpidtmrksf ydagrpekrd ikcptpgcdg tghvtglyph hrslsgcphk drippeilam henvlkcptp gctgqghvns nrnthrslsg cpiaaaeklt rshekqqqpg dlsksssnsd rilrpmcfvk qleipqygsy rpnmapatpr anlakeleky skvtfdyasf daqvfgkrll apkipssets pkafkskpfp kasspchsps ssyikstsss sssgfdythd aeaahmaata ilnlstrcwe mpenlstkqq dtpsksseie vdengtldls mnkhrkrest fpsssscsss psmkspdqsq rqnctsatss nmtsphssqt srqddwdgpi dytkpnrqre eepeemepaa asfassevde qemqemqemq emqeesyedr kypgdvtltn fklkflskds kkellscptp gcdgsghitg nyashrslsg cpladkslrn lmaahsadlk cptpgcdgsg hitgnyashr slsgcprakk sglkitptkd dkddpdlmkc pvpgcdglgh isgkyashrs asgcplaarr qkegalngsa fswkslkteg pscptpgcdg sghangsflt hrslsgcpra sfagkkgkis gdellgtnfk tsdvlendee ikqlnkeine lnesnsemea dmvnlqsqit tmeknlknie eenkvieeqn ealfvelsgl sqalirsltn

13.1.15 ncor2 nuclear receptor corepressor 2 S homeolog [Xenopus laevis]

NCBI Reference Sequence: NP_001084492.1 aevkrprlei lhdplirhsp litqnqhatg eelnkdhrma gklepvspss ppladgdldl lparlskeel iqnmdrvdre iimveqqisk lkkkqqqlee eaakplepek pvspppvesk hrslvqiiyd enrkkaeaah rileglgphv elplynqpsd tkqyheniki nqtmrkklil yfkrrnhark qweqklcqry dqlmeawekk veriennprr rakeskirey yekqfpeirk qrelqermqr vgqrgsgltl saarsehevs eiidglseqe nlerqmrqla vippmlfdae qqrirfinmn glmddpmkvy kerqvmnmws eqekdtfrek fiqqpknfvf iasflerktv sdcvlfyylt kknenyknli rrnyrrrgkn qqmprtnqee kdekekegek eeykqdldne kdepgkeknd dtsgdenddk eavmakgrkt ansqgrrkgr itrsmanean eestapqhyt dlatmemnes srwteeemet akkglfeygr nwsaiarmvg sktvsqcknf yfnykkrqnl dgilqqhklk mekernnrrk kkknpqmeee tpyppalede emeasgmsgn eeemadegda

13.1.16 ndel1 nuclear distribution protein nudE-like 1-A [Xenopus laevis]

NCBI Reference Sequence: NP_001086332.1 meneeipefl spkeeivywr elskrlkqsy qeardelief qegsreleae letqlvqaeq rnrdllsdnq rlkceveslk eklehqyaqs ykqvsllede lararsikdq lhkyvreleq anddlerakr ativsledfe qrlnqaiern afleseldek

328

407081 esllvsvqrl kdeardlrqe lavrerqtdg trksapsspt ldcdktdsav qaslslpatp vgkicdnsft spkgipngfg ttpltpsari salnivgdll rkvgaleskl aacrnfakdq asrksytpvn lnsnssssvl

13.1.17 neurod1 neurogenic differentiation factor 1 [Xenopus laevis]

NCBI Reference Sequence: NP_001079263.1 kvrrmkanar ernrmhglnd aldslrkvvp cysktqklsk ietlrlakny iwalseilrs gkspdlvsfv qtlckglsqp ttnlvagclq lnprtflpeq sqdiqshmqt asssfplqgy pyqspglpsp pygtmdsshv fhvkphsyga alepffdsst vtectspsfd gplspplsvn gnftfkhehs eydknytftm

13.1.18 neurog2 neurogenin 2 L homeolog [Xenopus laevis]

NCBI Reference Sequence: NP_001081802.1 sgetvlkikk trrvkannre rnrmhnlnsa ldslrevlps lpedakltki etlrfaynyi walsetlrlg dpvhrsast

13.1.19 ngn3 neurogenin 3 S homeolog [Xenopus laevis]

NCBI Reference Sequence: NP_001128257.1 vkqrrnrrvk andrernrmh nlnsaldalr silptfpdda kltkietlrf ahnyiwalse tlrmadqsll

13.1.20 ppp2r5a protein phosphatase 2 regulatory subunit B', alpha [Xenopus laevis]

NCBI Reference Sequence: NP_001108316.1 eieeildvie ptqfkkiqep lfkqiskcvs sshfqvaera lyfwnneyil slieenidki lpimfgslyk iskehwnpti valvynvlkt lmemngklfd eltgsykaer qrekkkeler

13.1.21 rara2 retinoic acid receptor, alpha L homeolog [Xenopus laevis]

NCBI Reference Sequence: NP_001083723.1 myenvdvspt hyhmmdfysh nrqclwpekr inpygtplgt qhwsssnh

13.1.22 rarb retinoic acid receptor beta, partial [Xenopus laevis]

329

407081 GenBank: ABE96637.1 pppppriykp cfvcqdkssg yhygvsaceg ckgffrrsiq knmvytchre kncvinkvtr nrcqycrlqr cfevgmskes vrndrnkkkk epskqecies yemtaelddl tekirkahqe tfpslcqlgk yttnssadhr vrldlgfwdk fselatkcii kivefakrlp gftsltiadq itllkaacld ililrictrf tpeqdtmtfs dgltlnrtqm hnagfgpltd lvftfanqfl plemddtetg llsaiclice drqdleepak vdklqeplle alkiyirkrr pnkphmfpki lmkit

13.1.23 ror2 receptor tyrosine kinase-like orphan receptor 2 L homeolog precursor [Xenopus laevis]

NCBI Reference Sequence: NP_001082312.1

Lgatprgyyl tflepvnnit ivqgqaatlh ckvsgnplpn vkwlkndapv vqepkritir ktdygsrlri qdldttdtgy yqcvatngik titatgvlfv rlgptnspnp niqddyhkdg fcqpyrgiac arfignrtiy vdslqmqgei enritaaftm igtsthlsdq csqfaipsfc hfvfplcddq tskprelcrd ecevlendlc rqeyniarsn plilmqlhlp nceelplpes peaancmrig ipveklnryq qcyngtgtdy rgsvsvtksg hqcqpwshqv phshslsnad ypeiggghsy crnpggqmeg pwcftqnknv rmelcdipac rtrdntkmei

13.1.24 tal1 T-cell acute lymphocytic leukemia protein 1 [Xenopus laevis]

NCBI Reference Sequence: NP_001081746.1 pertvelsgv kegaapnspp ravpviellr rgeglgnika reqelrlqni rttelcratl tpatelcrap ltpttelcra pltpttelcr apltpttelc rppltpaaef crasltpase lcrapssvtg psltattelc rppiplptps tgppaeqave armvqlspta slplqaagrt mlyglnqpla sdnsgyfgdp dtfpmytsns rakrrpgpie veisegpqpk vvrriftnsr erwrqqnvng afaelrklip thppdkklsk neilrlamky inflaklldd qeeegnqrnk

13.1.25 tra2b Splicing factor, arginine/serine-rich 10 [Xenopus laevis]

NCBI Reference Sequence: NP_001082402.1

Hignranpdp ncclgvfgls lytterdlre vfskygplsd vsivydqqsr rsrgfsfvyf envddakeak erangmeldg rrlrvdfsit krphtptpgi

13.1.26 wnt4b Protein Wnt-4 isoform 2 precursors [Xenopus laevis]

NCBI Reference Sequence: NP_001239014.1 mtpeyflrsl lmmilavfsa nasnwlylak lssvgsisee etceklkgpi qrqvqmckrn levmdsvrrg aqlaieecqy qfrnrrwncs tldtlpvfgk vvtqgtreaa fvyaissagv afavtracss gdlekcgcdr tvhgvspqgf 330

407081 qwsgcsdnil ygvafsqsfv dvrerskggs ssralmnlhn neagrkailn nmrveckchg vsgscevktc wkamptfrkv gnvlkekfdg ateveqkkig stkvlvpkns qfkphtdedl vyldsspdfc dhdlkngvlg ttgrqcnkts kaidgcelmc cgrgfhteev eivercsckf hwccfvkckq chkvvemhtc r

331

407081 14 Literature Search techniques and sources employed

The main source of information has been gathered searching the NCBI database

(http://www.ncbi.nlm.nih.gov/) and the Xenopus model organism database Xenbase

(http://www.xenbase.org/). Once a paper of interest was found, occasionally references

within that text were followed, to archive an insight into some detail. Scientific articles not

freely available were requested as inter-library loans at the university.

332