DOCTOR OF PHILOSOPHY

Elucidation of the roles of PTN and ISG15 in RSV cytopathogenesis: possible biomarkers of severe disease?

Groves, Helen

Award date: 2018

Awarding institution: Queen's University Belfast

Link to publication

Terms of use All those accessing thesis content in Queen’s University Belfast Research Portal are subject to the following terms and conditions of use

• Copyright is subject to the Copyright, Designs and Patent Act 1988, or as modified by any successor legislation • Copyright and moral rights for thesis content are retained by the author and/or other copyright owners • A copy of a thesis may be downloaded for personal non-commercial research/study without the need for permission or charge • Distribution or reproduction of thesis content in any format is not permitted without the permission of the copyright holder • When citing this work, full bibliographic details should be supplied, including the author, title, awarding institution and date of thesis

Take down policy A thesis can be removed from the Research Portal if there has been a breach of copyright, or a similarly robust reason. If you believe this document breaches copyright, or there is sufficient cause to take down, please contact us, citing details. Email: [email protected]

Supplementary materials Where possible, we endeavour to provide supplementary materials to theses. This may include video, audio and other types of files. We endeavour to capture all content and upload as part of the Pure record for each thesis. Note, it may not be possible in all instances to convert analogue formats to usable digital formats for some supplementary materials. We exercise best efforts on our behalf and, in such instances, encourage the individual to consult the physical thesis for further information.

Download date: 02. Oct. 2021 Elucidation of the roles of PTN and ISG15 in RSV cytopathogenesis: Possible biomarkers of severe disease?

Helen Elizabeth Groves MB BCh BAO (Hon)

School of Medicine, Dentistry and Biomedical Sciences, The Queen’s University of Belfast

Thesis offered to The Queen’s University of Belfast for the degree of Doctor of Philosophy June 2018 Acknowledgements

I would like to thank everyone who has supported and assisted me over the past few years in the preparation of this thesis work.

I would particularly like to thank both of my supervisors, Dr Ultan Power and Prof. Mike

Shields for their patience and encouragement throughout. I thank Dr Ultan Power for teaching me the scientific method, from the basics of how to pipette, the intricacies of scientific experimental design, through to critical thinking, problem solving and the drive for innovative science.

I thank Prof. Shields for his encouragement, pragmatism and enthusiasm for the clinical application of scientific research. I commend him for his passion in teaching me the joy of robust statistical methodology and I hope one day to understand this as well as he does.

I would like to thank Dr Lindsay Broadbent, Dr Hong Guo‐Parke, Dr Lyndsey Ferguson and

Ms Andrena Millar for their training, guidance and friendship throughout my PhD work, as well as all of the other colleagues and students who supported me in their journey through the fourth floor of the Medical Biology Centre.

I am grateful to all of the parents and the infants who took part in this research. Without them this work could not have been possible. I would also like to thank the staff of the Royal

Jubilee Maternity Hospital, Belfast and the Royal Belfast Hospital for Sick Children for their assistance and support in recruiting infants.

I thank the Wellcome Trust for their backing and financial support in completing this PhD.

I am deeply grateful to my fiancé Andrew who has supported me unwaveringly during the business and challenges of this PhD work. I love him all the more for the faithfulness, kindness, patience and love he has shown me.

2 I thank my parents for their encouragement and unshakable backing throughout a lifetime of education, especially to my mother who taught me how to learn.

Finally, I praise God, who enabled me to undertake this work and has given me innumerable blessings. For from Him and through Him and for Him are all things.

3 Statement of work contribution

All work included in this thesis was conducted by myself (Dr Helen Groves). This included consenting and recruiting all participants, nasal sampling and processing of samples, as well as all laboratory work and data analysis detailed in this thesis.

Acknowledgements of other contributions

Prior to commencing this thesis work, an extensive microarray study was undertaken comparing pan‐genomic transcriptome responses in WD‐PNECs derived from cohorts of infants with histories of mild and severe RSV disease. This work was completed by Dr Hong

Guo‐Parke who kindly contributed pleiotrophin microarray data. In addition, Dr Guillermo

Lopez Campos performed analysis of the microarray data for pleiotrophin. This pleiotrophin data is presented in figure 25a. Dr Olivier Touzelet kindly contributed pleiotrophin ELISA data shown in figure 25b. qRT‐PCR analysis of cDNA from WD‐PNECs derived from cohorts of infants with histories of mild and severe RSV disease is presented in figure 39. I performed this work while being trained by Dr Lindsay Broadbent and she therefore contributed the data for analysis.

Dr Lindsay Broadbent kindly contributed data for PTN pretreatment in WD‐PBECs shown in figure 31c.

4 Abstract

Respiratory syncytial virus (RSV) infection is the most common cause of severe lower respiratory tract infection in infants under two years old. Sparking seasonal epidemics, RSV contributes to around 20,000 admissions to hospital in the UK annually and many require intensive care support. Peak incidences of severe RSV disease occur between 6 weeks and

6 months of age. Despite over 60 years of research since its discovery, no RSV vaccine or specific therapy exists. The only preventative strategy against RSV is the monoclonal antibody palivizumab, which is very expensive and, accordingly, only available to those infants at known high risk of severe RSV disease, including preterm infants and those with congenital heart conditions or bronchpulmonary dysplasia. As the majority of infants hospitalised with RSV have no known predisposing risk factor for severe illness, the benefits of palivizumab for managing the impact of RSV is very restricted.

Considerable challenges in performing research in vulnerable young infants has slowed progress in RSV research. Animal models only partially reflect RSV‐human host interactions and thus extrapolation of results from these models to human responses is of limited value.

Study of autopsy specimens from infants who succumbed to RSV infection identified that the infection is restricted primarily to the airway epithelium. As such, research aimed at detecting innate immune responses of human airway epithelium to RSV is likely to yield crucial insights into the cytopathogenesis of this infection. It has previously been demonstrated that an ex‐vivo/in‐vitro model of human airway epithelium/RSV interactions, termed well‐differentiated primary paediatric nasal epithelial cell cultures (WD‐PNECs), provides a reasonable surrogate for in vivo responses.

Because of the strong association of severe RSV disease with prematurity and young age, the first part of this thesis describes the utilisation of the WD‐PNEC model to establish and

5 characterise unique nasal samples from preterm and term infants at birth and repeated at one‐year old. This work is the first description of morphologically and physiologically authentic WD‐PNEC cultures generated from term and preterm newborn infants and as such represents an exceptional opportunity to study RSV‐human host interactions in early life.

We found that newborn term‐ and preterm‐derived WD‐PNECs were morphologically indistinguishable under light or fluorescent microscopy analysis. However, interestingly, newborn WD‐PNECs demonstrated significantly higher proportions of goblet cells compared to one‐year repeat WD‐PNECs. This finding indicates the possibility of increased mucous production in newborn infants, which may, in part, explain their susceptibility to more severe

RSV disease.

Importantly, we demonstrated nasal sampling to be a safe, minimally invasive method performed consistently with high rates of success. Furthermore, we were also able to successfully freeze, thaw, and subsequently differentiate the nasal epithelial cells. This confirmed the exciting possibility of storing newborn ‘‘naive’’ airway epithelial cells (AECs) indefinitely for use in subsequent experimentation, e.g., once clinical phenotypes, like severe RSV or asthma, have been established.

We next sought to establish if differential RSV‐induced innate immune responses of airway epithelial cells could account, at least in part, for the increased susceptibility of preterm and very young infants to severe RSV disease. To investigate this, we infected (or mock‐infected)

WD‐PNEC cultures established from term and preterm infants at birth and repeated at one‐ year‐old. No significant differences in cytopathology or viral growth kinetics were evident in

WD‐PNECs derived from any cohort following RSV infection. However, crucially, we observed significantly higher secretion of interferon 1 (IL‐29) (P<0.01), IP‐10 (CXCL‐10)

(P<0.05) and RANTES (CCL‐5) (P<0.05) following RSV infection of one‐year‐derived WD‐PNEC cultures compared to newborn‐derived cultures. These novel findings suggest airway

6 epithelium innate immune responses to RSV are less robust in newborn infants compared to older infants, which may contribute to the increased susceptibility of very young infants to severe RSV‐related disease. This is the first report of age‐related differences in airway epithelial cell innate immune responses.

Previous work performed in RSV‐infected WD‐PNECs derived from infants with histories of severe versus mild RSV disease identified differential expression in a number of genes. Two of these genes, interferon‐stimulate gene 15 (isg15) and pleiotrophin (ptn), were of particular interest based on published work identifying the anti‐viral role of ISG15 in a number of viral illnesses and the established interaction of PTN with the recently identified RSV co‐ receptor/entry factor, nucleolin. Our findings on the role of PTN are particularly exciting.

We found differential expression of PTN in infants at greater risk of severe RSV disease compared to older infants and in those with a history of severe compared to mild RSV disease. These findings suggest a potentially vital role for PTN in innate immune defence against RSV. Consistent with such a role, we demonstrated that PTN has anti‐RSV activity in vitro, mediated via interaction with cell surface nucleolin. Taken together, these novel findings suggest that relatively lower endogenous expression of PTN expression may explain, in part, the increased susceptibility of some infants to severe RSV disease.

Finally, we established that isg15 mRNA expression is increased in both newborn‐ and 1 year old infant‐derived WD‐PNECs following RSV infection, with similar levels of expression between term‐ versus preterm‐derived WD‐PNECs. We also demonstrated an anti‐viral effect of ISG15 in vitro, in which increased RSV infection was evident following isg15 knockdown in BEAS‐2B cells pre‐treated with IFN‐1 (IL 29) compared to control siRNA‐ transfected cells. In addition, we found that RSV antagonizes ISG15 expression in individual infected cells compared to surrounding non‐infected cells within WD‐PBECs. Our data suggest that ISG15 acts as an innate antiviral molecule against RSV and differential

7 expression of ISG15 in infant airways following infection may contribute to susceptibility to severe RSV disease.

In summary, this thesis reports the exciting discovery of an anti‐viral role for PTN in RSV infection and highlights the biomarker potential for both PTN and ISG15 in identifying individuals at increased risk of severe RSV disease. This work also adds to our understanding of early life innate immune responses to RSV infection and provides an innovative model for identifying new potential therapeutic targets and strategies in the management of RSV.

8 Table of Contents

Chapter 1: Introduction ...... 30 1.1 Importance of respiratory syncytial virus ...... 30 1.2 Natural history of RSV disease and known risk factors for severe disease ...... 31 1.2.1 Host‐related risk factors ...... 32 1.2.2 Virus‐related risk factors ...... 33 1.3 Molecular characteristics and structure of RSV ...... 34 1.4 RSV receptor and infectious cycle ...... 38 1.4.1 CX3CR1 as a possible receptor for RSV ...... 39 1.4.2 Nucleolin as a possible RSV receptor ...... 39 1.4.3 RSV infectious cycle ...... 41 1.5 Current progress on RSV disease treatment and vaccine development ...... 42 1.5.1 Vaccine development strategies ...... 44 1.5.2 Prophylactics under development ...... 47 1.5.3 Anti‐RSV therapeutics in development ...... 48 1.6 RSV pathogenesis ...... 53 1.6.1 Clinical manifestation of RSV disease ...... 53 1.6.2 Histopathology of RSV ...... 54 1.6.3 Innate immune responses to RSV ...... 55 1.7 The use of models in eliciting RSV pathogenesis ...... 60 1.7.1 Animal and immortalised cell line models ...... 60 1.7.2 The three‐dimensional airway epithelial cell model ...... 62 1.7.3 Nasal versus bronchial airway epithelial cell models ...... 63 1.8 Concluding remarks and project aims ...... 68 1.8.1 Summary of what is unknown regarding RSV disease ...... 68 1.8.2 Overall aim of this project ...... 69 1.8.3 Specific project objectives and plan ...... 70

Chapter 2: Materials and Methods...... 75 2.1 Cell lines culture protocols ...... 75 2.1.1 Cell lines, Media and Reagents ...... 75 2.1.2 Routine sub‐culture of cell lines ...... 76

9 2.1.3 Cell counting procedure ...... 77 2.1.4 Procedure for cell freezing of continuous cell lines ...... 79 2.1.5 Procedure for cell thawing of frozen continuous cell lines ...... 81 2.2 Well‐differentiated Primary Paediatric Airway Epithelial Cell (WD‐PAECs) culture protocols ...... 83 2.2.1 Collagen coating of tissue culture flasks and Transwells ...... 83 2.2.2 Preparation of WD‐PAEC culture media ...... 85 2.2.3 Procedure for obtaining nasal epithelial cell sampling from patients ...... 88 2.2.4 Processing of nasal sample brushes ...... 90 2.2.5 Passage of PAECs ...... 91 2.2.6 Culture of well‐differentiated PAECs (WD‐PAECs) ...... 93 2.2.7 Procedure for freezing PAECs ...... 96 2.2.8 Procedure for thawing frozen PAECs ...... 97 2.2.9 Trans‐Epithelial Electrical Resistance (TEER) measurement ...... 99 2.2.10 Fixation of WD‐PAEC Cultures ...... 100 2.3 Virus infection protocols ...... 101 2.3.1 Viruses ...... 101 2.3.2 Culture and expansion of RSV stocks ...... 101

2.3.3 RSV titration using Tissue Culture Infectious Dose (TCID50) Assay ...... 103 2.3.4 Fluorescent RSV titration using Fluorescent Forming Units (FFU) Assay ...... 106 2.3.5 RSV infection of WD‐PAECs...... 107 2.3.6 Apical washing and basolateral medium harvesting from RSV‐infected WD‐PAEC cultures ...... 109 2.3.7 Separation of Transwell membrane from Transwell insert following fixation for staining purposes ...... 110 2.3.8 General procedure for immunofluorescence staining of WD‐PAECs ...... 111 2.3.9 Cytospins and smears of apical washes ...... 114 2.3.10 Confocal Microscopy ...... 115 2.4 General Molecular Procedures ...... 116 2.4.1 RNA extraction procedure ...... 116 2.4.2 RNA Quality and Quantity analysis ...... 118 2.4.3 cDNA transcription ...... 121 2.5 Western Blot Procedure ...... 124 2.6 Statistical Analysis Methods ...... 132

10 Chapter 3: Characterisation of well‐differentiated primary paediatric nasal epithelial cell (WD‐PNEC) cultures derived from term and preterm infants at birth and one‐year‐old...... 139 3.1 Abstract ...... 139 3.2 Introduction ...... 141 3.3 Methods ...... 143 3.3.1 Subjects and study design ...... 143 3.3.2 Sampling of nasal epithelial cells ...... 143 3.3.3 Processing of nasal samples and creation of WD-PNEC cultures ...... 144 3.3.4 Measurement of trans-epithelial electrical resistances (TEER) ...... 145 3.3.5 Immunofluorescence microscopy for WD-PNEC characterisation ...... 145 3.3.6 Freezing and defrosting of harvested nasal epithelial cells ...... 147 3.3.7 Statistical analysis ...... 148 3.3.8 Ethics statement ...... 148 3.4 Results ...... 149 3.4.1 Clinical characteristics and sampling of recruited infants at birth ...... 149 3.4.2 WD-PNEC cultures from preterm and term newborn infants demonstrate similar differentiation schedules and success rates ...... 150 3.4.3 Term- and preterm-derived newborn WD-PNEC cultures are morphologically indistinguishable ...... 151 3.4.4 Clinical characteristics and repeat sampling of one year old infants ...... 154 3.4.5 Nasal AECs from one-year old infants achieve complete differentiation faster than those from newborn infants ...... 156 3.4.6 WD-PNEC cultures derived from one-year old infants demonstrate significantly reduced goblet cell content compared to their paired newborn- dervied WD-PNECs ...... 156 3.4.7 Nasal primary AECs successfully differentiate after storage in liquid nitrogen and are morpholocially indistinguishable from freshly differentiated AECs ...... 159 3.5 Discussion ...... 162

Chapter 4: In vitro modelling of respiratory syncytial virus infection of paediatric nasal epithelium derived from term and preterm infants at birth and one‐year old ...... 169 4.1 Abstract ...... 169 4.2 Introduction ...... 171 4.3 Methods ...... 176

11 4.3.1 Cell lines and virus titration ...... 176 4.3.2 Well-differentiated primary paediatric nasal epithelial cells (WD-PNECs) ...... 177 4.3.3 RSV infection of WD-PNEC cultures ...... 178 4.3.4 Measurement of transepithelial electrical resistances (TEER) ...... 178 4.3.5 Slide fixation of apical rinses ...... 179 4.3.6 Immunofluorescent staining of WD-PNEC cultures ...... 179 4.3.7 Immunofluorescence of fixed apical washes ...... 180 4.3.8 IL-29 (IFN-λ1) quantification...... 181 4.3.9 Cytokine/Chemokine quantification ...... 181 4.3.10 Ethics statement ...... 182 4.3.11 Statistical analysis ...... 183 4.4 Results ...... 184 4.4.1 Cell tropism and replication of RSV in newborn term- and preterm-derived WD-PNEC cultures ...... 184 4.4.2 Tropism and replication of RSV in one-year old-derived compared to newborn-derived WD-PNEC cultures ...... 186 4.4.3 RSV infection results in reduction of ciliated cells in newborn and one-year derived WD-PNEC cultures ...... 188 4.4.4 RSV induces goblet cell hyper/metaplasia in newborn- and one-year- derived WD-PNEC cultures ...... 190 4.4.5 Cytopathogenesis following RSV infection in newborn- and one-year- derived WD-PNEC cultures ...... 192 4.4.6 Secretion of IL-29 (Interferon- λ1) by newborn- and one-year-derived WD- PNEC cultures in response to RSV infection ...... 195 4.4.7 RSV induced cytokine/chemokine secretion in term- and preterm-derived WD-PNEC cultures at birth ...... 197 4.4.8 RSV induced cytokine/chemokine secretions in one-year- compared to newborn-derived WD-PNEC cultures ...... 200 4.5 Discussion ...... 202

Chapter 5: Identification of pleiotrophin as a novel innate anti‐viral protein in RSV infection ...... 210 5.1 Abstract ...... 210 5.2 Introduction ...... 212 5.2.1 Pleiotrophin functions and roles ...... 212

12 5.2.2 The role of PTN in inflammation ...... 214 5.2.3 Anti-viral and anti-bacterial actions of PTN ...... 215 5.3 Methods ...... 216 5.3.1 Cell lines and viruses ...... 216 5.3.2 Well-differentiated primary paediatric airway epithelial cell cultures ..... 217 5.3.3 Mild versus severe microarray analysis ...... 218 5.3.4 RNA extraction from newborn and one-year repeat derived WD-PNECs . 219 5.3.5 RNA quality analysis ...... 220 5.3.6 cDNA synthesis ...... 222 5.3.7 qRT-PCR primer design for pleiotrophin and house-keeping genes ...... 223 5.3.8 qRT-PCR Analysis ...... 224 5.3.9 Immunofluorescent staining ...... 226 5.3.10 Immunofluorescence of fixed WD-PNECs ...... 228 5.3.11 PTN ELISA ...... 228 5.3.12 Recombinant Human PTN reconstitution ...... 229 5.3.13 PTN pre-treatment experiments ...... 230 5.3.14 Anti-PTN antibody pre-treatment experiments ...... 231 5.3.15 PTN siRNA silencing experiments ...... 232 5.3.16 RSV entry experiments ...... 233 5.3.17 Ethics statement ...... 235 5.3.18 Statistical analysis ...... 235 5.4 Results ...... 237 5.4.1 Endogenous PTN is differentially expressed in WD-PNECs derived from infants with a history of mild versus severe RSV disease ...... 237 5.4.2 Endogenous PTN is differentially expressed in WD-PNECs generated from preterm and term born infants at birth, but not at one-year ...... 239 5.4.3 Endogenous PTN expression is higher in WD-PNECs generated from one- year-old compared to newborn infants ...... 242 5.4.4 ptn gene expression and PTN secretion is upregulated following RSV infection ...... 244 5.4.5 Intracellular PTN in WD-PNEC cultures is only expressed in ciliated cells and is located at the apical surface...... 246 5.4.6 PTN exerts a direct antiviral effect against RSV in an immortalised bronchial cell line and well differentiated primary paediatric bronchial epithelial cells (WD-PAECs) ...... 249

13 5.4.7 Neutralisation of endogenous PTN results in increased RSV infection in BEAS-2B cells and WD-PAEC cultures...... 252 5.4.8 Knockdown of ptn in an immortalised cell line results in increased RSV infection levels ...... 254 5.4.9 PTN exerts its anti-viral effect by inhibition of RSV entry into airway cells 256 5.4.10 PTN exerts its antiviral effect against RSV via interaction with nucleolin 258 5.4.11 Cross-linking of surface bound PTN results in redistribution of cellular nucleolin towards the cell surface ...... 259 5.5 Discussion ...... 262

Chapter 6: Antiviral role of ISG15 in RSV infection ...... 271 6.1 Abstract ...... 271 6.2 Introduction ...... 273 6.2.1 Antiviral role of interferons (IFNs) ...... 273 6.2.2 Antiviral role of interferon stimulated gene 15 (ISG15) ...... 274 6.3 Methods ...... 278 6.3.1 Cell lines and viruses ...... 278 6.3.2 Well-differentiated primary paediatric nasal epithelial cells (WD-PNECs) 278 6.3.3 Mild versus severe derived WD-PNECs qRT-PCR analysis ...... 279 6.3.4 RNA extraction and quality analysis ...... 279 6.3.5 cDNA transcription and qRT-PCR analysis ...... 280 6.3.6 Staining and immunofluorescence of fixed WD-PNECs ...... 281 6.3.7 Recombinant human ISG15 reconstitution ...... 282 6.3.8 ISG15 pre-treatment experiments ...... 283 6.3.9 ISG15 siRNA silencing ...... 283 6.3.10 Ethics statement ...... 284 6.3.11 Statistical analysis ...... 285 6.4 Results ...... 286 6.4.1 Endogenous isg15 is differentially expressed in WD-PNECs generated from infants with a history of mild versus severe RSV disease ...... 286 6.4.2 Endogenous isg15 expression in WD-PNECs generated from preterm and term born infants at birth and at one-year ...... 288

14 6.4.3 isg15 expression is upregulated following RSV infection of WD-PNECs and peaks at 72-96 hours post infection ...... 290 6.4.4 ISG15 is expressed apically following RSV infection of WD-PNECs and expression at 96 hpi is unaffected by age or prematurity ...... 291 6.4.5 Exogenous ISG15 does not exert a direct antiviral effect against RSV ...... 295 6.4.6 ISG15 knockdown in interferon lambda 1 (IFN-1)/ interleukin-29 (IL-29) stimulated BEAS-2B cells resulted in increased RSV infection levels...... 296 6.4.7 RSV antagonizes ISG15 expression in WD-PNECs ...... 299 6.5 Discussion ...... 301

Chapter 7: Conclusions and future directions ...... 308 7.1 Summary of findings ...... 308 7.1.1 Impact of gestational age on WD-PNECs and RSV-induced innate immune responses ...... 309 7.1.2 Impact of chronological age on WD-PNECs and RSV-induced innate immune responses ...... 309 7.1.3 Role of ISG15 in determining RSV disease severity ...... 310 7.1.4 Role of pleiotrophin (PTN) in RSV disease ...... 312 7.2 Study strengths and what this thesis adds to current scientific knowledge ...... 314 7.3 Study Limitations ...... 317 7.4 Future Directions ...... 319 7.5 Concluding remarks ...... 322

References ...... 323 Appendix ..………………………………………………………………………………………………………………370

15 List of Figures

Figure 1. Illustration of RSV virion structure ……………………………………………………………………36

Figure 2. Crystallographic structural arrangement of RSV F in pre‐ and post‐fusion conformations …………………………………………………………………………………………………………………37

Figure 3. A schematic representation of RSV life cycle …………..………………………………………..42

Figure 4. Immunohistochemical staining for respiratory syncytial virus (RSV) antigen in bronchiolar and alveolar tissues from infants with lower respiratory tract infection (LRTI)………………………………………………………………………………………………………………………………..55

Figure 5. Summary of project plan and chapter location of results…………………………………..72

Figure 6. Image of newborn infant undergoing nasal brushing to harvest nasal epithelial cells…………………………………………………………………………………………………………………………………144

Figure 7. Phase‐contrast light microscopy of cultured nasal epithelial cells derived from a newborn infant…………………………………………………………………………………………..…………………..151

Figure 8. Morphology and differentiation status of newborn term and preterm WD‐PNEC cultures...... 153

Figure 9. Morphology and differentiation status of birth and one‐year repeat WD‐PNEC cultures…………………………………………………………………………………………………………………………..158

Figure 10. Morphology and differentiation status of WD‐PNEC cultures derived from epithelial cells frozen at passage 3………………………………………………………………………………….161

Figure 11. RSV BT2a infection of well‐differentiated paediatric primary nasal epithelial cells (WD‐PNECs) derived from preterm and term newborns…………………………………………………185

Figure 12. RSV BT2a infection in well‐differentiated paediatric primary nasal epithelial cells (WD‐PNECs) derived from one‐year‐old infants compared to newborn infants………………187

Figure 13. RSV infection results in reduced ciliated cell proportions in WD‐PNEC cultures and this effect is similar regardless of gestation or age………..……………………………………………….189

Figure 14. RSV infection results in increased goblet cell proportions in newborn WD‐PNEC cultures………………………………………………………………………………………………………………………….191

Figure 15. RSV infection in newborn‐ and one‐year old‐derived WD‐PNECs does not cause visible cytopathology and induces modest increases in cell sloughing…..………………………194

Figure 26. IL 29 expression in newborn term‐ and preterm‐derived WD‐PNEC cultures and one‐year‐derived WD‐PNEC cultures following RSV infection………………………………………..196

Figure 37. IL 29 expression in newborn‐ and one‐year‐derived WD‐PNEC cultures following RSV infection………………………………………………………………………………………………………………….197

Figure 18. Basolateral secretion of chemokines/cytokines induced after RSV infection of newborn term‐ and preterm‐derived WD‐PNEC cultures……………………………………………….199

Figure 49. Differential basolateral secretion of chemokines/cytokines following RSV infection of newborn‐ and one‐year‐derived WD‐PNEC cultures…………………………………………………..201

16 Figure 20. Known PTN receptors and functions……………………………………………………………..213

Figure 21. Diverse functions of PTN……………………………………………………………………………….214

Figure 22. Electropherogram summary of electrophoresis analysis of RNA sample using Agilent Technologies RNA 600 nano kit………………………………………………………………………….221

Figure 23. Electropherogram summary of electrophoresis analysis of RNA sample using Agilent Technologies RNA 600 nano ………………………………………………………………………………222

Figure 24. Flow diagram of sample collection and qRT‐PCR analysis…………………….……….225

Figure 25. ptn expression and basolateral PTN secretion is endogenously lower in WD‐PNECs generated from infants with a history of severe RSV disease compared to mild………………………………………………………………………………………………………………………………..239

Figure 26. Expression of PTN is higher following RSV infection in WD‐PNECs generated from term versus preterm newborn infants……………………………………………………………………………241

Figure 27. Endogenous expression of PTN is higher in WD‐PNECs generated from one‐year old infants compared to newborn infants………………………………………………………………………243

Figure 28. PTN gene expression and protein secretion by airway epithelial cells is upregulated following RSV infection…………………………………………………………………….………..245

Figure 29. PTN is expressed in ciliated cells and confined to the apical surface of WD‐PNEC cultures………………………………………………………………………………………………………………………….246

Figure 30. PTN is expressed in ciliated cells but not goblet cells in RSV infected WD‐PNEC cultures………………………………………………………………………………………………………………………….248

Figure 31. Recombinant PTN exerts a direct anti‐viral effect against RSV at 4oC but not at 37oC...... 251

Figure 32. PTN neutralisation using anti‐PTN antibody results in increased levels of RSV infection………………………………………………………………………………………………………………..……….253

Figure 33. RSV infection levels are increased following ptn knockdown in BEAS‐2B cells………………………………………………………………………………………………………………………………..255

Figure 34. PTN exerts its antiviral effect by inhibiting RSV entry …………………………………..257

Figure 35. The anti‐RSV effect of PTN is abrogated by the addition of anti‐nucleolin neutralising antibody……….…………………………………………………………………………………………….259

Figure 36. PTN induced redistribution of nucleolin in BEAS 2B cells.……………………………..261

Figure 37. Representation of the interferon (IFN) signalling pathways.………………………….274

Figure 38. Protein ISGylation and de‐ISGylation system………………………………………………..275

Figure 39. Identification of differentially expressed genes in WD‐PNECS derived from infants with histories of severe versus mild RSV disease……………………………………………………………287

Figure 40. Expression of isg15 following RSV infection of WD‐PNECs generated from term and preterm infants at birth and one‐year old……………………………………………………………….289

17 Figure 41. isg15 gene expression in airway epithelial cells is upregulated following RSV infection………………………………………………………………………………………………………………………..291

Figure 42. ISG15 is expressed apically in RSV‐infected WD‐PNEC cultures…………………….293

Figure 43. ISG15 expression following RSV infection is similar in WD‐PNECs derived from term and preterm infants at birth and one‐year at 96 hpi…………………………………………….294

Figure 44. Addition of recombinant ISG15 does not exert a direct anti‐viral effect against RSV……………………………………………………………………………………………………………………………….295

Figure 45. RSV infection levels were increased following ISG15 knockdown in IL‐29‐ stimulated BEAS‐2B cells………………………………………………………………………………………………298

Figure 46. RSV antagonizes expression of ISG15 in infected cells within WD‐PNEC cultures………………………………………………………………………………………………………………………..300

Figure 47. Schematic representation of putative method by which PTN may inhibit RSV attachment/entry into epithelial cells…………………………………………………………………………..314

Figure 48. Schematic representation of PTN interactions and pathways which may play a role in inhibiting RSV infection of airway epithelial cells………………………………………………………321

18 List of Tables

Table 1. Complete list of all registered on clinicaltrials.gov for RSV vaccines, prophylactics and therapeutics since 2012………………………………………………………………………………………50

Table 2. Summary of studies retrieved from PubMed directly comparing nasal and bronchial epithelial cell responses……………………………………………………………………………..65

Table 3. List of cell lines, tissue of origin and growth and maintenance media used…………………………………………………………………………………………………………………………..75

Table 4. Cell culture reagents…………………………………………………………………………………….76

Table 5. Cell freezing medium……………………………………………………………………………………80

Table 6. Growth medium composition for PAEC cultures in monolayer……………………..85

Table 7. Growth medium composition for 2 x ALI for PAEC cultures differentiation…..86

Table 8. Final concentrations of supplements in working 1x ALI medium…………………..88

Table 9. Recipe for transport medium preparation…………………………………………………….88

Table 10. Volume of media and PBS recommended for culture of PAECs in Transwells………………………………………………………………………………………………………………….95

Table 11. PAEC freezing medium……………………….………………………………………………………97

Table 12. Viruses used during the preparation of this thesis work……………………………..101

Table 13. Contents of Roche High Pure RNA Extraction Kit…………………………………………117

Table 14. Contents of RNA 6000 nano kit (Agilent)……………………….……………………………119

Table 15. Contents of High Capacity cDNA RT Kit……………………………………..………………..122

Table 16. Volume of High Capacity cDNA kit components required per reaction……….122

Table 17. Thermal cycler conditions for reverse transcription………………………….…………123

Table 18. Laemmli Buffer preparation ……………………………………………………………………….125

Table 19. Lysis Buffer preparation………………………………………………………………………………126

Table 20. RIPA Buffer preparation…………………………………………….………………………………..126

Table 21. 1X Running Buffer preparation…………………………………………………………………….127

Table 22. 10X Tris Buffered Saline preparation.…………………………………………………………..127

Table 23. Resolving Gel (10%) preparation………………………………………………………………….128

Table 24. Stacking Gel (5%) preparation……………………………………………………………………..128

Table 25. Stripping Buffer preparation……………………………………………………………………….129

Table 26. List of Materials and supplier detailed in this thesis…………………………………….134

19 Table 27. List of Suppliers of reagents and kits detailed in this thesis …………………………135

Table 28. Perinatal and delivery characteristics of enrolled subjects…………………………..150

Table 29. Clinical history of infants returning for repeat sampling after first year of life……………………………………………………………………………………………………………………………….155

20 Glossary of Terms

AEC Airway epithelial cell

ALI Air‐liquid interface

AUC Area under the curve

BSA Bovine serum albumin

CO2 Carbon dioxide

cm2 Centimetre square

cDNA Complementary DNA

CI Confidence interval

oC Degrees Celsius

DAPI 4’,6‐diamidino‐2‐phenylindole

dH2O Distilled H2O

DNA Deoxyribonucleic acid

DMSO Dimethyl sulfoxide

DMEM Dulbecco’s Modified Eagle Medium

EVOM Epithelial volt‐ohm metre

EDTA Ethylenediaminetetraacetic acid

FITC Fluorescein isothiocyanate

FFU Fluorescent forming units

FBS Foetal bovine serum

eGFP Enhanced green fluorescent protein

h Hour

hpi Hours post infection

BEAS‐2B Human bronchial epithelial cells

21 HEp‐2 Human epithelial cells from laryngeal carcinoma

IFN Interferon

IP‐10/CXCL10 Interferon induced protein 10

ISG Interferon stimulated gene

ISG15 Interferon stimulated gene 15

IL‐6 Interleukin‐6

IL‐8/CXCL8 Interleukin‐8

IL‐29 Interleukin‐29

IL‐33 Interleukin‐33

MMP‐13 Matrix Metallopeptidase‐13

µL Microlitre

µm Micrometre

Mg Milligram

Mm Millimetre

mm2 millimetre square

MEM Minimum Essential Medium

Min minute/minutes

Muc5ac Mucin 5 AC

MOI Multiplicity of infection

ng Nanogram

PFA Parafprmaldehyde

Pen/Strep Penicillin/Streptomycin

PBS Phosphate buffered saline

PTN Pleiotrophin

PCR Polymerase chain reaction

22 rRSV/eGFP Recombinant RSV A2 expressing green fluorescent protein

RANTES/CCL5 Regulated upon activation, normal T cell expressed and secreted

RSV Respiratory syncytial virus

RSV BT2a Respiratory syncytial virus Belfast 2 subgroup A clinical isolate

RA Retinoic acid

qRT‐PCR Quantitative Reverse transcription real‐time PCR

RNA Ribonucleic acid

RT Room temperature

TEER Trans epithelial electrical resistance

TCID50 Tissue culture infectious dose 50%

TLR Toll‐like receptor

UV Ultraviolet

v/v Volume per volume

w/v Weight per volume

WD‐PAECs Well‐differentiated primary paediatric airway epithelial cells.

WD‐PBECs Well‐differentiated primary paediatric bronchial epithelial cells

WD‐PNECs Well‐differentiated primary paediatric nasal epithelial cells

23

24 Conference Participation

 Annual Meeting of the Association of Physicians of Great Britain and Ireland,

Manchester, 2018.

Innate immune responses to Respiratory Syncytial virus (RSV) in newborn airway

epithelial cells are decreased compared to responses in older infants (poster

presentation)

 Academy of Medical Sciences Clinician Scientist Meeting, London, 2017.

Innate immune responses of preterm and term airway epithelium to Respiratory

Syncytial virus (RSV) in well‐differentiated primary nasal epithelial cell (WD‐PNEC)

cultures (poster presentation)

 10th Respiratory Syncytial Virus World Congress Meeting, Patagonia, 2016.

Identification and characterisation of a novel anti‐viral protein in RSV infection

(Poster Presentation)

 10th Respiratory Syncytial Virus World Congress Meeting, Patagonia, 2016.

Characterisation of RSV infection in Well‐differentiated primary nasal epithelial cell

(WD‐PNEC) cultures from preterm and term newborn infants (Poster Presentation)

 Sharing Progress in Neonatology Conference, Naples, 2016.

Well‐differentiated primary nasal epithelial cell (WD‐PNEC) cultures derived from

new‐born term and preterm infants: an exciting opportunity to study airway innate

immune responses in at risk groups (Poster Presentation)

 Microbiology Society Annual Conference, Liverpool, 2016.

Elucidation of the role of an innate antiviral protein in respiratory syncytial virus

infection (Poster Presentation)

25

 Microbiology Society Host Pathogens Conference, Dublin, 2016.

Respiratory syncytial virus infection and the role of an innate anti‐viral protein

(Poster Presentation)

 Ulster Paediatric Society Junior Members Forum, Belfast, 2016.

Elucidation of the roles of innate antiviral proteins in the development of the airway

epithelial innate immune response to Respiratory Syncytial Virus (RSV) infection in

term and preterm infants (Oral Presentation)

 Irish Immunology Society Annual Meeting, Dublin, 2015.

Production of well‐differentiated primary nasal epithelial cell cultures derived from

neonates sampled within 48 hours of birth: an exciting opportunity to study airway

innate immune responses in newborn infants (Poster Presentation)

26 Prizes received for presentation of PhD thesis work

Queens University, Belfast, Centre for Experimental Medicine third year PhD symposium,

2017: First prize for oral presentation.

Colonel Davis Research Scholarship awarded by the Royal Belfast Hospital for Sick Children,

Belfast, 2017.

Queens University, Belfast, Annual Postgraduate Research Symposium, 2016: First Prize for poster presentation.

Ulster Paediatric Society, junior Members Forum Annual Meeting, 2016: First Prize for oral presentation.

Queens University, Belfast, Centre for Infection and Immunity second year PhD student symposium, 2015: second prize for oral presentation.

Queens University, Belfast, Translational Research Group Conference, 2015: first prize for poster presentation.

Queens University, Belfast, Centre for Infection and Immunity second year PhD student symposium, 2014: first prize for oral presentation.

27

28

CHAPTER 1

INTRODUCTION

Publication pertaining to introduction:

Broadbent L, Groves H, Shields MD, Power UF. Respiratory syncytial virus, an ongoing medical dilemma: an expert commentary on respiratory syncytial virus prophylactic and therapeutic pharmaceuticals currently in clinical trials. Influenza and Other Respiratory

Viruses 2015 Jul; 9 (4): 167‐178

Chapter 1: Introduction

Prelude.

Each results chapter in this thesis is written in the format of an extended journal manuscript, which includes a detailed introduction tailored to each hypotheses being investigated.

Accordingly, this introduction is an overview of the clinical impact of respiratory syncytial virus (RSV), its interaction with the airway epithelium and current progress on RSV specific drug and vaccine development. Limitations in our current knowledge of RSV pathogenesis and how this research project aims to further understanding of RSV‐airway interactions are discussed. Sections of this overview contributed directly to a published manuscript:

Broadbent L, Groves H, Shields MD, Power UF. Respiratory syncytial virus, an ongoing medical dilemma: an expert commentary on respiratory syncytial virus prophylactic and therapeutic pharmaceuticals currently in clinical trials. Influenza and Other Respiratory

Viruses 2015 Jul; 9 (4): 167‐178.

1.1 Importance of respiratory syncytial virus

Respiratory syncytial virus (RSV) is the commonest cause of severe lower respiratory tract infection in infants under two‐years worldwide.1,2 The virus manifests a range of symptoms from mild rhinorrhoea to significant lower respiratory tract infections (LRTIs), including bronchiolitis and pneumonia.3 RSV infection contributes to substantial mortality and morbidity worldwide and accounts for 20% of pneumonia cases and greater than 80% of bronchiolitis cases in infancy, respectively. Causing global seasonal epidemics in children less than five years of age, RSV is responsible for 33.8 million LRTIs, 3.4 million hospitalisations and up to 199,000 deaths annually.4–6 Of these reported deaths, greater

Chapter 1: Introduction 30 than 90% occur in low and middle‐income countries.7 The economic impact of RSV disease is considerable, with one US study estimating direct medical costs for all RSV infection‐ related hospitalisations of $394 million (US) for children under five years.8

RSV also causes significant disease burden amongst the elderly and immune‐compromised individuals. In the UK, over 14,000 hospitalisations and almost 8,000 deaths in adults aged over 65 years occur secondary to RSV‐attributable disease each season.9 In immunosuppressed individuals, RSV is recognised as a substantial problem, for example in lung transplant recipients, where it can cause severe pneumonia, respiratory failure and development of bronchiolitis obliterans.10

Despite over 60 years since RSV’s first description,11 limited progress has been made on RSV specific therapies or vaccines and the mechanisms by which RSV causes disease in humans remain largely obscure. Presently, no effective anti‐RSV drugs or vaccine exist and the only effective preventative strategy against RSV is the use of a monoclonal antibody, palivizumab.

This is prohibitively expensive and therefore its use is restricted to infants known to be at risk of severe RSV disease.12 Furthermore, the majority of infants who develop severe RSV related disease do not have any known risk factors. Therefore, greater insight into the reasons why some infants develop more severe RSV disease is urgently needed.

1.2 Natural history of RSV disease and known risk factors for severe disease

RSV infection occurs in seasonal epidemics, typically during autumnal and winter months in temperate climes and rainy season in tropical climes.13 The reasons behind this seasonal spread are not fully understood. It has been proposed that environmental temperature and

Chapter 1: Introduction 31 humidity may play a role in this variation. In a large US based study, Hartert et al found that the estimated effect of temperature and humidity on RSV bronchiolitis was most pronounced at the beginning and end of epidemics, suggesting these conditions may contribute to emergence of RSV spread and its subsequent termination.14 However many questions remain unanswered, namely, is RSV present in communities year round and only sparks an epidemic when conditions are optimal or is it introduced to communities from outside? Interestingly, a recent report documented summertime nasopharyngeal shedding of respiratory viruses, including RSV, amongst 1 in 14 adult visitors to a New York tourist attraction, many of whom were asymptomatic.15 While infectivity of this nasopharyngeal shedding is unclear, this certainly supports the concept of RSV presence in adult populations throughout the year and presents a potential mechanism by which an epidemic could be triggered, given the right seasonal conditions.

1.2.1 Host‐related risk factors

While most RSV infections in infancy result in mild symptoms, as highlighted above, large numbers of infants develop severe symptoms and the mechanisms contributing to severe disease are largely unknown. Epidemiological studies have identified a number of factors associated with severe RSV‐related illness. In a recent study of 186 Taiwanese children admitted to intensive care, 72% of cases had at least one underlying medical diagnosis.16

Prematurity is well established as a significant risk factor, with a seven times increase in the risk of acute RSV‐bronchiolitis. Furthermore, bronchopulmonary dysplasia as a result of preterm birth is correlated with more severe RSV‐related lower respiratory tract infections

(LRTI).16–25 Other comorbidities, including; congenital heart disease (CHD), chronic pulmonary disease and the presence of underlying immunodeficiency, neurological or

Chapter 1: Introduction 32 haemato‐oncological disease, are all associated with more severe RSV‐related LRTI and recent studies have identified trisomy 21 as a risk factor.22,26–31

Environmental factors including passive smoking, lower socio‐economic class, environmental pollution and high altitude contribute to severe illness, as does the presence of school/day‐care age siblings living in the same household.32–39 The presence of malnutrition or low birth weight at admission increases the likelihood of severe RSV‐related

LRTI and maternal history of asthma or atophy is also a risk factor.38,40

However, the majority of infants who develop severe RSV‐related disease have no underlying diagnosis and thus genetic factors likely play a significant role in determining severity. Certain ethnicities, for example Alaskan or Native American‐Indian, are at greater risk of more severe RSV‐related disease and an increasing number of publications highlight the association of polymorphisms in innate immune system genes and disease severity. 4,41,42

Indeed, recently an RSV transcriptional signature from whole blood gene analysis correlated the differential expression of specific innate immune genes with RSV‐disease severity.43

These studies provide further evidence for the importance of innate immune responses in determining the course of RSV infection.

1.2.2 Virus‐related risk factors

RSV has two major antigenic subgroups (A and B) which have a worldwide geographical distribution and circulate independently.44 The clinical impact of viral serotype during RSV infection remains controversial and a number of studies have investigated the relationship between disease severity and RSV subgroup. Some groups have reported increased disease severity with particular RSV strains according to subtype and genotype,45–48 while others

Chapter 1: Introduction 33 suggest equivalent severity for RSV subtypes.49–52 Simultaneous circulation of multiple strains of each subgroup, as well as seasonal variation in virus population and geographical variation in these studies may account for this discrepancy.44,53 However, the true biological significance of RSV strain variation on disease severity remains undetermined.

Several studies have also proposed correlation between higher nasal pharyngeal or tracheal viral load and RSV disease severity.54–57 However, there are limitations with these studies, such as the use of mean viral titres, which could mask the fact that certain individuals with mild disease can have very high titres of RSV and conversely, individuals with very severe

RSV disease can have very low RSV titres.53 DeVincenzo et al. have developed an elegant human challenge model of RSV disease in adults that is being used extensively for testing of novel anti‐RSV therapies.57,58 This work has demonstrated that when symptoms are mild and restricted to the upper respiratory tract, virus growth kinetics parallel disease severity kinetics, suggesting efficiency of virus replication drives symptom severity. However, care is required when extrapolating results from work performed in adults, in whom the infection is restricted to the upper airways, to infants not previously exposed to RSV infection, where the infection can affect the lower airways. Indeed, it is likely that the true determinant of the severity of RSV disease is a complex interplay of a myriad of factors including viral strain, viral load, host‐virus interaction and host immune responses. Further understanding of this complex relationship is needed to determine new strategies to prevent and treat severe RSV disease.

1.3 Molecular characteristics and structure of RSV

RSV is a member of the Pneumoviridae family, Pneumovirus genus. It consists of two antigenic subtypes as outlined above, which are distinguished primarily by genetic and

Chapter 1: Introduction 34 antigenic differences in the G gene and protein, respectively.59 Morphologically, RSV virions have been described as having two forms, spherical (≤350 nm diameter) and filamentous (≤5

µm).59,60 RSV contains a single‐stranded negative sense RNA genome, made up of 10 genes60 that encode for 11 proteins.61 Four proteins, N, P, M2‐1 and the large polymerase subunit L form the virion nucleocapsid, with L, P and M2‐1 comprising the RNA‐dependent RNA polymerase complex, which is responsible for gene transcription and genome replication.

The virion is enveloped by a lipid bilayer derived from the cell membrane, inside which is located the matrix (M) protein, which is critical for virion assembly. Three surface glycoproteins are present, including the fusion (F), the attachment protein G and small hydrophobic (SH) proteins, which are associated with viral entry and the formation of syncytia (fig. 1).62–64 RSV G is responsible for virus attachment to target cells, while F mediates fusion of the viral and cell membranes to facilitate entry of the nucleocapsid into the cytoplasm. The function of the SH protein is less clear, but evidence suggests that it may have ion‐channel forming abilities and also inhibits TNF signalling and apopotosis.65–67 Two viral non‐structural proteins (NS1 and NS2) are essential to virulence through modification of innate immune responses. This is achieved via a variety of methods, including suppression of interferon‐mediated innate immune responses by blocking IRF‐3 phosphorylation and polyubiquination of STAT2, which results in proteasomal degradation, as well as inhibition of Th subset differentiation.68–71

Chapter 1: Introduction 35

Figure 1. Illustration of RSV virion structure. RSV consists of single stranded (ss) negative sense (‐) RNA contained within a nucleocapsid (composed of nucleoprotein N, Phosphoprotein P, Matrix protein M2‐1 and the large polymerase subunit L) and enveloped by a lipid bilayer. Three glycoproteins are displayed on the surface: fusion (F), glycoprotein (G) and small hydrophobic (SH) proteins.

Further understanding of the molecular and structural composition of RSV may lead to development of better vaccine candidates and RSV‐specific drug discovery. Currently, the major targets for vaccine and drug development are the cell surface glycoproteins, especially

RSV F and G proteins.72 The RSV F protein is of particular interest due to its requirement for viral entry and high sequence conservation.73 RSV F demonstrates conformational diversity and broadly exists in two substantially different states: a metastable state prior to virus interaction with the cell (prefusion) and a stable state occurring after merging of the virus and cell membrane (postfusion) (figure 2).74

Chapter 1: Introduction 36

Figure 2. Crystallographic structural arrangement of RSV F in pre‐ and post‐fusion conformations. To mediate virus‐cell entry, the RSV F glycoprotein transitions from a metastable prefusion conformation to a stable postfusion conformation. Images at the left and right extremities display prefusion (left) and postfusion (right) trimeric structures. Inner images display a single RSV F protomer in ribbon representation, coloured as a rainbow from blue to red, N terminus of F2 to C terminus of F1, respectively. Select secondary‐structure elements are labelled Inset: Schematic of the mature RSV F protein in the RSV F(+) Fd construct. The rainbow colouring of the boxes representing the F2 and F1 subunits matches that in the structures. Glycans are shown as branches on top of the boxes, and disulfide bonds are shown as black lines under the boxes. Amino acids that move more than 5 Å in the pre‐ and postfusion conformations are indicated by black bars. From Ahmad, K. et al. Structure of RSV Fusion Glycoprotein. Science. 340, 1113–1117 (2013).75 Reprinted with permission from AAAS.

Lamb/Jardetzky et al. were the first group to stabilise a paramyxovirus F protein in the prefusion conformation using parainfluenza virus 5 F protein.76 This achievement opened the possibility of structural based vaccine design against members of the Paramyxoviridae and Pneumoviridae, including RSV. Indeed, the development of a vaccine against RSV remains frustratingly elusive. Early attempts with a formalin‐inactivated alum‐adjuvented

RSV vaccine (FI‐RSV) tragically resulted in significantly enhanced RSV disease severity and 2 deaths when infants encountered natural RSV infection. 77,78 Interestingly, recent work by

Graham et al. on the effects of formalin on RSV noted that formalin results in conformational change of the RSV F protein from pre‐F to post‐F under the conditions utilised to

Chapter 1: Introduction 37 manufacture the original formalin inactivated RSV vaccine.73 The authors argue that this change contributed to the pre‐priming of the immune response that lead to enhanced RSV disease phenotype upon natural exposure to the virus. Accordingly, a number of structure based vaccine designs have been proposed utilising stabilised prefusion F. It is hoped this will preserve neutralisation‐sensitive epitopes promoting induction of protective antibody responses rather than non‐neutralizing antibody responses, theoretically reducing subsequent immune complex formation and activation.

1.4 RSV receptor and infectious cycle

Early research suggested that RSV entry is likely to be a two‐step process, involving attachment to the cell by means of G glycoprotein interaction with cell surface proteoglycans

(electrostatic interaction), followed by fusion with the cell membrane primarily mediated by

F protein interaction.79,80 The functional RSV receptor is not yet fully determined and a large number of potential candidates have been proposed, including heparin‐like molecules,

ICAM‐1, toll‐like receptor 4 (TLR‐4) and annexin II.79,81–85 These contenders have not met all conditions for a functional receptor molecule, namely; reduced infection levels with antibody neutralization or RNA interference and increased infection following ectopic molecule expression in non‐permissive cells.79 RSV infection has been most extensively studied in immortalized cell lines, where it has been established the RSV G glycoprotein uses cell‐surface heparan sulfate (HS) as a receptor.86 However, It is also noteworthy that heparin is not expressed on the apical surface of human respiratory epithelial cells.87 Therefore, immortalized cell lines may not be the best model for the study of RSV entry, with limited relevance to certain aspects of the human airway epithelium in vivo. Furthermore, mutant

RSV lacking RSV‐G glycoprotein remain capable of infection, albeit at lower efficiency than

Chapter 1: Introduction 38 wild‐type.88 Thus, it may be that the binding of the RSV G protein to heparin is actually a demonstration of its generalised affinity for negatively charged molecules on the cell surface.81

1.4.1 CX3CR1 as a possible receptor for RSV

More recent studies have identified other possible RSV receptors, including CX3CR1 and nucleolin.81,89 Tripp et al. found the RSV‐G protein mimics the chemokine CX3CL1

(fractalkine) in its ability to bind to its receptor, CX3CR1.90 Peeples et al. have taken this further to propose CX3CR1 as a possible RSV receptor. They found that the RSV‐G protein epitope is recognised by part of the CX3C motif, determined that RSV is able to infect CHO

A745 cells (a cell line deficient in HS expression) when they transiently express CX3CR1 and that CX3CR1 knockout mice have reduced susceptibility to RSV infection than wild‐type mice.91 However, as discussed above, G protein deficient RSV is still capable of causing infection and, therefore, CX3CR1 is unlikely to represent the complete picture for an RSV receptor.

1.4.2 Nucleolin as a possible RSV receptor

Hegele et al. proposed nucleolin as a receptor for RSV. They utilised a modified western blot procedure, known as Virus Overlay Protein Binding Assay (VOPBA), where the virus substitutes for the primary antibody, to identify RSV binding proteins from various mammalian cell lines lysates. Nucleolin was consistently found in each extract and viral isolate tested.79,89 Furthermore, nucleolin was found at the cell surface, co‐precipitated only

Chapter 1: Introduction 39 with the viral F protein in every instance tested and RSV infection decreased following nucleolin neutralization and in nucleolin knockdown mice compared to wildtype.89,92 This data supports the proposal of nucleolin as an RSV receptor. However, our group have not detected nucleolin on the apical surface of well‐differentiated human primary airway epithelial cell cultures (unpublished data). This casts some doubt on the certainty of nucleolin as the RSV receptor, although the fact nucleolin actively shuttles between the nucleus and cytoplasm may account for this apparent discrepancy.93,94

Interestingly, nucleolin was implicated as a low‐affinity receptor for human immunodeficiency virus‐1 (HIV‐1) in work conducted over 20 years ago by Hovanessian et al.95 Hovanessian’s group later found that anchorage of HIV particles on lymphocyte cell lines induces nucleolin aggregation in association with lipid raft components.96 They also noted that the heparin‐binding protein pleiotrophin (PTN) inhibits HIV attachment to the surface of lymphocytes and that this effect is mediated via interaction with nucleolin.97

Therefore, given the proposal of nucleolin as a cell surface receptor for RSV, this work hints at a potential function of PTN in inhibiting RSV entry. Certainly, PTN has been established as a ligand of surface nucleolin in angiogenesis and cancer literature.98–100 PTN is a ubiquitous molecule that, as its name suggests, fulfils a vast array of functions and has roles in embryonic development, angiogenesis, neurite growth and cancer development.99,101–104 It is also increasingly recognised as playing a role in inflammation.105–107 However, other than in HIV, PTN is not known to influence responses to viral infections. Certainly, the possible role of PTN in viral infection deserves further exploration.

Chapter 1: Introduction 40

1.4.3 RSV infectious cycle

The infectious cycle of RSV follows a similar route to that of the paramyxovirus lifecycle.

After binding of RSV via the G protein and fusion via RSV F protein, the viral genome is released into the cytoplasm,108 allowing transcription of the viral genome to commence.

Newly formed negative sense viral genomes are associated with the N protein, wrapped in a left‐handed helix to form a nucleocapsid core containing the L and P proteins.109,110

Replication of new negative sense viral genome requires an antigenome intermediate step, which is accomplished with the same polymerase complex used for mRNA transcription.110

It is not known how this switch from transcription to replication occurs, but early work highlighted the involvement of the M2‐2 protein.111 Glycoproteins are translated and trafficked to the cell apical surface where viral proteins assemble into infectious viral particles prior to separation from the cell surface membrane. The subsequent assembly and budding of RSV virions is mediated in part by the interaction of the M protein with the actin cytoskeleton, although full understanding of this process remains limited.110,112 Newly formed RSV progeny subsequently bud at the apical surface of infected cells and are released to infect further cells. Greater understanding of the RSV infectious cycle has enabled development of novel therapeutics as discussed further below.

Chapter 1: Introduction 41 Figure 3. A schematic representation of RSV life cycle.113 RSV attachment to the host cell membrane and subsequent entry allows release of the viral genome into the cytoplasm. Primary transcription of the negative‐sense RNA produces mRNAs for protein synthesis. Replication of the viral genome occurs via a two‐step process where antigenome intermediates are made from the genome template before production of progeny negative‐ sense genomes. Newly synthesised nucleoprotein associates with the genomic RNA and is transported to the plasma membrane for assembly with synthesised proteins and glycoproteins. This is followed by membrane scission and budding to release virus particles. Reproduced under the Creative Commons Attribution License from: Bawage, S. S., Tiwari, P. M., Pillai, S., Dennis, V. & Singh, S. R. Recent advances in diagnosis, prevention, and treatment of human respiratory syncytial virus. Adv. Virol. 2013. Article ID 595768.

1.5 Current progress on RSV disease treatment and vaccine development

Presently there is no specific therapy to treat RSV infection. Several meta‐analyses on available treatments for RSV bronchiolitis have been conducted in recent years, including the use of nebulised bronchodilators, adrenaline and hypertonic saline, as well as use of oral steroids. These indicate that no specific therapy is of benefit and only supportive

Chapter 1: Introduction 42 management is recommended.114–119 Ribavirin, a nucleoside analogue inhibiting viral RNA synthesis, is used in the treatment of RSV in patients requiring mechanical ventilation. Its efficacy and safety remains controversial due to lack of conclusive evidence from underpowered studies.120 As such, ribavirin use is typically confined to haematopoetic stem cell transplant recipients with severe RSV infection.121–124

The monoclonal antibody palvizumab is the only effective preventative strategy against RSV infection with proven efficacy in preterm infants, children with severe chronic lung disease and those with congenital cardiac disease.125,126 However, the significant cost of paliviumab limits its use to high risk patient groups only. Given that most infants hospitalised, and even admitted to intensive care, with RSV‐disease are previously healthy infants,56 there remains an urgent need for more effective treatment strategies or a successful vaccine.

No vaccine has been licensed for use against RSV and, as eluded to above, vaccine development has been hampered by the need to create a robust immune response without producing vaccine‐enhanced disease.127 Development is further complicated by the fact that re‐infection with RSV can occur throughout life.128

Peak incidence of severe RSV disease occurs between 6 weeks and 6 months and thus, a RSV vaccine would ideally be administered within the 1st month of life. However, immune system immaturity and presence of maternal antibodies that may negatively impact on the effectiveness of a vaccine, are difficult hurdles to overcome in infant vaccine development.

Despite these considerable barriers and historical setbacks, there is evidence to suggest that a RSV vaccine is feasible. Whilst it is known that human RSV infection occurs repeatedly throughout life, the frequency of severe disease is generally less in second and subsequent infections.129,130

Chapter 1: Introduction 43 Four potential target populations for RSV vaccines exist, namely, young infants, older children, pregnant women and those aged over 65 years. Presently, the most progressed strategies in RSV vaccine development are virus attenuation and recombinant expression of

RSV antigens using an appropriate viral vector. Other new technological advances, including nanoparticles and structural analysis of RSV proteins, are promising new strategies for RSV vaccine design.131 A comprehensive search of all clinical trials related to RSV vaccine or therapeutics development currently registered on the National Institute of Health National

Library revealed a total of 203 active studies (clinicaltrials.gov, accessed April 2018). All vaccine and therapeutic candidates listed in this database since 2012 are detailed in Table 1.

The extensive number of ongoing trials highlights the vast array of anti‐RSV strategies being pursued and the development of some of these candidates is discussed in greater detail below.

1.5.1 Vaccine development strategies

Traditionally, live attenuated RSV vaccines have been produced by a combination of serial passage and chemical mutagenesis. Cold passage, temperature sensitive (cpts) vaccines, created with the aim of restricting viral replication to nasal passages rather than warmer lower airways, have shown some promising data. Phase 1 testing of the cpts‐248/404 vaccine in infants aged 1‐2 months, for example, demonstrated that vaccine‐induced immunity is achievable in this age‐group.132 Reassuringly, this and other studies of live‐ attenuated vaccines demonstrated no evidence of FI‐RSV‐like enhanced disease133.

However, achieving the correct balance of vaccine attenuation and immunogenicity is challenging, with under‐attenuated vaccines leading to higher rates of nasal congestion, fever, LRTI, cough and otitis media132,134,135 and over‐attenuated vaccines showing greatly reduced infectivity and concomitant reduced nasal cytokine levels.136,137

Chapter 1: Introduction 44 New live attenuated RSV vaccines engineered using reverse genetics technology to modulate the RSV viral genome has enabled specific attenuating mutations to be combined, creating new vaccine candidates. A number of these are currently in phase 1/2a clinical trials. Further iterations of the cpts‐248/404 RSV vaccine with deletion of the SH gene and introduction of a temperature sensitive mutation in the L gene (N1321), include rA2cp248/404/1030/ΔSH and its derivative MEDI‐599.137,138 Clinical trials of these vaccines have shown challenges with genetic instability in attenuating mutations138 and worryingly, for MEDI‐559 an increased rate of medically attended LRTIs, including wild‐type RSV infections, in the vaccine versus placebo cohorts.139–141 Efforts have been made to produce more stable vaccine candidates engineered from MED‐559. These include RSV cps2 Lot RSV#005A with five nucleotide changes that confer genetic stability to the attenuating mutations142 and RSV

ΔNS2 Δ1313/1314L which has an amino acid change at position 1314 in the L protein and deletion of the NS2 gene and 1313 gene. Results of the phase 1 clinical trials for both candidates are awaited.

Additional gene deletion mutants as candidates for live‐attenuated vaccines, including recombinant viruses with M2‐2 gene deletion are also being explored. Preclinical studies demonstrated decreased RNA replication, attenuated virus growth and concomitant increases in F and G protein expression.111 Unfortunately, a Phase 1 study of LID ΔM2‐2 vaccine in RSV‐seronegative infants resulted in higher than desirable peak shedding titres resulting in early study termination.143 Several new iterations utilising the M2‐2 deletion such as D46cpΔM2‐2 and D46/NS2/N/ΔM2‐2‐HindIII, are presently under investigation in phase 1 trials.

A number of other vaccine strategies, including chimeric viruses and virus vector systems have also recently undergone clinical trials. MEDI‐534, for example, is a recombinant

Chapter 1: Introduction 45 chimeric bovine/human parainfluenza virus type 3 (b/hPIV3) engineered to express the RSV

F‐protein.144,145 Whilst preclinical studies of MEDI‐534 in Syrian Golden hamsters were encouraging,144 testing in RSV seronegative infants reported higher rates of coryzal symptoms in vaccinees compared to placebo. Furthermore, similar to MEDI‐599, genetic variation and mutation acquisition was noted in MEDI‐534 recovered in nasal washes.146,147

More recently, adenovirus vector‐based RSV vaccine candidates, such as GlaxoSmithKline’s

ChAd155‐RSV (GSK3389245A), VXA‐RSV‐f and Crucell Hollan BV are being tested in phase 1 and 2 trials. These systems employ chimpanzee or human adenovirus (serotype 35 and 26) engineered to encode RSV F protein. Additionally, new vaccine designs employing a prime/boost regime of simian adenovirus (PanAd3) and modified vaccinia Ankara (MVA) virus vectors expressing the RSV F protein have been trialled. This system was found to be effective in mice, cotton rats and non‐human primates148,149 and in a recently completed phase 1 trial, showed immunogenicity, with a promising safety profile. However, no current clinical trial is looking at this vaccine in seronegative infants.150

As eluded to earlier, a number of structure based and subunit designed RSV vaccine candidates are under clinical investigation. The primary target for these vaccines is the RSV‐

F protein. One such example, the Novavax RSV F nanoparticle vaccine is engineered using the baclofen/sf9 insect cell system to produce post‐fusion F protein.151 Unfortunately, despite promising initial immunogenicity data in healthy adults,151–153 a phase 3 trial in older adults failed to achieve pre‐specified vaccine efficiency objectives. Although a further study in pregnant women is ongoing, expectations for this vaccine are now more tempered.

Likewise clinical trials of Medimmune’s subunit vaccine MEDI‐7510, which contains post‐ fusion RSV F formulated with the synthetic TLR‐4 agonist glucopyranosyl lipid A adjuvant,154

Chapter 1: Introduction 46 were discontinued and further development stopped due to failure to achieve satisfactory primary outcomes.

Given the lack of efficacy seen in RSV subunit vaccines employing the post‐fusion conformation of the F protein, a number of pre‐fusion F candidates are under development.

This approach seems prudent given that most RSV neutralizing antibodies in human sera following natural RSV infection are directed towards the pre‐ rather than post‐fusion configuration.155 The earliest of these pre‐fusion vaccines was GSK3003891A, which produced significant levels of anti‐RSV neutralizing antibodies in the sera of immunized mice and protection against subsequent RSV A challenge.156 However, a recent planned study in healthy pregnant women has been withdrawn due to instability of the PreF antigen during manufacturing (ClinicalTrials.gov Identifier: NCT03191383). Other Pre‐F subunit vaccines are currently in early phase clinical trials, including DS‐Cav1, developed by B. Graham and P.

Kwong at NIAID, which has shown high levels of neutralizing antibodies and protection against subsequent RSV infection in mice and calves.157

1.5.2 Prophylactics under development

Palivizumab, as mentioned above, is the only approved preventative measure against RSV.

It is a monoclonal antibody given via monthly intramuscular injection throughout the annual

RSV season and has been shown to reduce RSV‐related hospitalisation rates in infants at high risk of severe RSV‐LRTI.158 However, the significant cost of palivizumab restricts its use to high risk infants.12 In addition, another challenge of palivizumab is the need for repeated monthly injections. Accordingly, third generation RSV‐specific monoclonal antibodies with extended half‐lives are under investigation aiming to achieve a single annual injection.159

Chapter 1: Introduction 47 One such example, REGN2222, is the only RSV‐specific drug to reach phase 3 clinical trials in recent times and results of a clinical study in preterm infants are awaited.

1.5.3 Anti‐RSV therapeutics in development

Development of RSV specific anti‐viral drugs have largely focussed on fusion inhibitors. Two promising antiviral agents include RSV fusion inhibitors GS‐5806 (Presatovir) and ALS‐008176

(Lumicitabine). Presatovir was found to efficiently neutralise a large panel of RSV clinical strains in vitro.160 Preclinical studies in cotton rats suggested a trend towards a dose‐ dependent reduction in lung viral titres following intraperitoneal administration 1 h after

RSV challenge. In an exciting development, a RSV strain Memphis 37 challenge trial in healthy adults with low serum RSV neutralisation activity showed reduced mucus production, decreased clinical severity scores, and lower mean peak viral loads in nasal washes following treatment.161 Similarly, Lumicitabine has been trialled using this challenge model, demonstrating more rapid clearance of RSV titres and reduced clinical symptoms in treated subjects compared to placebo.162 It is noteworthy, however, that in these challenge models both Presatovir and Lumicitabine were administered orally following detection of

RSV but before symptoms were evident.161,162 This experimental set up poorly reflects clinical reality, in that established symptomatic infection is likely to be evident before providing drug administration, and therefore these results should be interpreted cautiously.

More recent fusion inhibitors under investigation include AK0529, JNJ‐53718678 and RV521.

AK0529, a small molecule RSV fusion inhibitor was associated with lower RSV titres in a mouse model and was also well tolerated in a phase 1 study of healthy adult volunteers.163

JNJ‐53718678, developed by Janssen Research is another fusion protein inhibitor which has shown reduction in viral load and symptom severity in the Memphis 37 challenge model of

Chapter 1: Introduction 48 RSV infection as described above.164 ReViral, UK have also developed a fusion inhibitor,

RV521 which demonstrated favourable safety profile in phase 1 testing and is currently undergoing a phase 2a trial in adults.

An interesting new RSV therapeutic is ALX‐0171, a trivalent RSV‐F‐specific nanobody developed by Ablynx. This utilises camelid‐derived antibodies, or Nanobodies, which are antibody fragments that retain the antigen‐binding ability of the heavy chain antibody.165,166

Inhaled ALX‐0171 therapy post RSV‐challenge in a neonatal lamb model resulted in a dramatic decline in cultivatable RSV, reduced lung‐viral antigen expression and fewer lung pathological lesions.167,168 The positive effect of ALX‐0171 treatment on clinical parameters in this model was observed even after symptoms were evident. A recently completed phase

1/2a study in RSV‐infected hospitalised infants aged 5‐24 months showed no significant adverse effects and reduced time to undetectable virus in treated subjects compared to placebo. It is worth noting that the clinical impact of inhalational therapy may be limited by the presence of excess mucous once RSV symptoms are established. Therefore, successful

RSV therapeutics may need to target host inflammatory responses to reduce immune related damage at the infection site. For example, Danirixin, a reversible CXCR2 antagonist developed as an anti‐inflammatory agent for disorders associated with neutrophil accumulation,169 has been recently trialled in RSV‐infected infants. This study was completed in May 2017, and results are awaited. RV568 treatment, which also targets neutrophil responses, produced a reduction of IL‐8 levels in nasal washes relative to untreated controls in a RSV Memphis‐37 human challenge trial. However, sadly this did not translate into any effect on clinical severity.170

Chapter 1: Introduction 49 Drug Clinical trial Manufacturer/ Experimental Trial populations Outcome Name status Institution approach Vaccines Live attenuated MEDI‐559 Phase 1/2a Medimmune LLC Live‐attenuated Healthy RSV Increased rate seronegative of LRTIs in infants vaccine recipients, further study ongoing RSV LID Phase 1 NIAID Recombinant live‐ RSV‐Seronegative Vaccine ΔM2‐2 attenuated RSV infants shedding titres high, study stopped prematurely RSV cps2, Phase 1 Medimmune, NIAID Recombinant live‐ RSV‐Seronegative Completed – Lot Attenuated RSV Infants results not RSV#005A available ΔNS2/Δ13 Phase 1/2a NIAID Recombinant live‐ RSV‐Seronegative Commenced 13/I1314L attenuated RSV infants Feb 2018 – recruitment ongoing D46/NS2/ Phase 1 NIAID Recombinant live‐ RSV‐Seronegative Commenced N/ΔM2‐2‐ attenuated RSV infants April 2017 – HindIII recruitment ongoing RSV6120/ Phase 1 NIAID Recombinant live‐ RSV‐Seronegative Commenced ∆NS2/103 attenuated RSV and Seropositive October 2017 – 0s infants recruitment ongoing RSV Phase 1 NIAID Recombinant live‐ RSV‐Seronegative Commenced D46cpΔM attenuated RSV and Seropositive Nov 2015 – 2‐2 infants recruitment Vaccine ongoing Chimeric/vectored RSV 001 Phase 1 Okairos Adenovirus vector Healthy adults aged Commenced and an MVA 18 ‐ 50 Years May 2013 vector encoding Outcome RSV antigens awaited MEDI‐534 Phase 1 Medimmune LLC Chimeric/vectored RSV seronegative Genetic infants variants within vaccine detected. Ongoing research. MVA‐BN Phase 2 Bavarian Nordic Recombinant MVA Healthy adults aged Commenced RSV vector expressing >55 years April 2015, RSV F and G results awaited glycoproteins VXA‐RSV‐f Phase 1 Vaxart Adenoviral‐Vector Healthy Adults Commenced Based RSV F June 2016, Protein Vaccine ongoing GSK33892 Phase 1 GlaxoSmithKline Chimpanzee‐ RSV seropositive Commenced 45A derived infants Jan 2017, Adenovector virus recruiting vaccine SeVRSV Phase 1 NIAID Sendai‐vectored Healthy adults Commenced RSV vaccine April 2018, ongoing

Chapter 1: Introduction 50 rBCG‐N‐ Phase 1 Pontificia Recombinant Healthy adult males Commenced hRSV Universidad Catolica Mycobacterium July 2017, de Chile bovis BCG vaccine ongoing expressing hRSV Nucleoprotein (N) Ad26.RSV. Phase 2a Janssen Vaccines Human type 26 Healthy adults Commenced preF adenovirus vector Nov 2017, expressing recruiting stabilised RSV pre‐ F Ad35.RSV. Phase 1 Crucell Holland BV Human Healthy adults Completed Aug FA2 adenovirus 2017, results Regimens serotype 35 and pending Boosted 26 vector system With Ad26.RSV. FA2

Nanoparticle RSV F Phase 3 Novavax Recombinant RSV Healthy Third‐ Commenced nanoparti F protein particle trimester Pregnant Dec 2015, cle Women ongoing vaccine Subunit MEDI Phase 1a MedImmune LLC RSV sF antigen + Healthy adults aged Terminated 7510 synthetic 60 years and over early and glucopyranosyl further trials lipid A adjuvant discontinued RSV F Phase 1 Novartis vaccines F subunit vaccine Healthy adults Estimated subunit completion vaccine Sept 2016 GSK3003 Phase 2 GlaxoSmithKline Prefusion F Healthy pregnant Withdraw due 891A (GSK) Biologicals subunit vaccine women to instability of PreF DPX‐ Phase 1 ImmunoVaccine B cell epitope Healthy adults aged Commenced RSV(A) Technologies inc. peptide targeting 50 ‐ 64 years 2015, ongoing SH domain of RSV VRC‐317 Phase 1 NIAID Prefusion RSV F Healthy Adults Commenced (DS‐Cav1) Subunit Protein Feb 2017, Vaccine recruiting SynGem Phase 1 Mucosis BV Prefusion F Healthy Adults Commenced subunit vaccine Nov 2016, ongoing Prophylactics MEDI 557 Phase 1 MedImmune Recombinant Healthy adults aged Completed human 18‐45 years May 2013. monoclonal Results not antibody available MEDI Phase 2 Medimmune Human RSV Preterm infants Commenced 8897 Monoclonal Aug 2016, antibody ongoing. REGN222 Phase 3 Regeneron Human Preterm infants Completed 2 pharmaceuticals monoclonal anti‐ March 2018 – RSV F antibody results awaited Therapeutics GS‐5806 Phase 2b Gilead RSV entry Adults hospitalised Completed Nov (Presatovi inhibitor with RSV 2017, results r) awaited

Chapter 1: Introduction 51 ALN‐ Phase 2b Ablynx siRNA targeting Lung transplant Reduced RSV01 (completed the N protein recipients aged bronchiolitis May 2012) over 18 years. obliterans post RSV infection in lung transplant recipients. ALS‐ Phase 2 Janssen Research & Nucleoside analog Infants hospitalised Commenced 008176 Development, LLC with RSV Nov 2017 ‐ (Lumicitab ongoing ine) Danirixin Phase 1 GlaxoSmithKline CXCR2 antagonist. RSV‐infected Completed GSK Inhibition of children <2 years May 2017. No 1325756 neutrophil and healthy adults results activation available. MDT‐637 Phase 1 MicroDose Fusion inhibitor Healthy adults aged Commenced Therapeutx, Inc 18‐ 50 Years 2013. Data not available ALX‐0171 Phase 2 Ablynx Nanobody Infants and Completed Dec Toddlers 2016. Data hospitalised with awaited RSV AK0529 Phase 2 ARK biosciences Inc. Small molecule Hospitalised infants Commenced inhibitor or RSV with RSV Jan 2016, replication recruitment targeting F ongoing BTA‐C585 Phase 2a Vaxart Fusion inhibitor Healthy adults Completed Jan (Enzaplato 2017 – results vir) pending PC786 Phase ½ Pulmocide Ltd Inhibitor of Healthy Adults Commenced Ribonuclear Nov 2017 ‐ protein (RNP) ongoing complex activity JNJ‐ Phase 2a Janssen Research & Fusion inhibitor Non‐hospitalised Commenced 53718678 development, LLC adults infected with Dec 2017, RSV ongoing RV521 Phase 2a ReViral Ltd Fusion inhibitor Healthy adults Commenced Aug 2017, results awaited JNJ‐ Phase 1/2a Alios Biopharma Inc. non‐nucleoside Healthy adults Commenced 64417184 polymerase Jan 2018, not inhibitor yet recruiting EDP‐938 Phase 1 Enanta Non‐fusion RSV Healthy adults Commenced Pharmaceuticals inhibitor Dec 2017, recruiting Table 1. Complete list of all registered on clinicaltrials.gov for RSV vaccines, prophylactics and therapeutics since 2012.

The extensive number of RSV vaccine candidates, prophylactics and therapeutics currently under development highlights the significant attention given to discovering new strategies to manage RSV disease. However, as evidenced by the large number of candidates failing to progress beyond phase1/2a trials, it is clear that a number of challenges remain. Historically there has been a considerable disconnect between excellent preclinical efficacy in animal

Chapter 1: Introduction 52 models and limited or no clinical effectiveness for RSV vaccines and therapeutics. It is disappointing, for instance, that despite many years of research on temperature sensitive live attenuated vaccines limited progress has been made on achieving a successful vaccine.

Recent strategies, including virus‐vectored vaccines and subunit vaccines do show intriguing potential. However, there is accumulating evidence of the capacity of RSV proteins, including F, G and non‐structural proteins 1&2 , to modulate immune responses following infection of humans.68,171 Therefore, the possibility remains that these RSV vaccine strategies may result in compromised and poorly protective immune responses in humans, as is evident following natural RSV infection. New technologies for therapeutics, including fusion inhibitors and Nanobodies, are showing promising results and perhaps the answer to RSV treatment/prevention lies within these new methodologies. However, a deeper fundamental understanding of RSV disease pathogenesis and immune modulation in humans will be central to successful clinical development.

1.6 RSV pathogenesis

1.6.1 Clinical manifestation of RSV disease

Most children infected with RSV initially manifest with upper respiratory tract symptoms, incluing rhinitis, cough and coryzae. Lower respiratory tract symptoms typically follow after a period of days, characterised by dyspnoea, subcostal recession, and feeding difficulties in infants.3 Lower Respiratory tract infection (LRTI) due to RSV can manifest as bronchitis, bronchiolitis or pneumonia.53 In bronchiolitis, wheeze, crackles and prolonged phase of expiration are present, often with hyperinflation and segmental atelectasis on chest X‐ray.

When severe, bronchiolitis can result in bronchospasm, hypoxia, respiratory failure and may also manifest as apnoea in very young infants.3

Chapter 1: Introduction 53

1.6.2 Histopathology of RSV

Almost 50 years ago, Aherne et al. described the histological hallmarks RSV disease from autopsies of 22 children.172 Acute bronchiolitis was noted to begin with necrosis of the bronchiolar epithelium and destruction of ciliated cells followed by peri‐bronchiolar infiltration of lymphocytes and macrophages. Subsequently, oedema of adventitial tissues and mucous plug formation due to enhanced mucous secretion and cellular debris causes occlusion of bronchioles.173

The histopathology of fatal RSV pneumonia has been extensively described by Johnson et al.

Similar to the work by Aherne et al., they observed widespread destruction of bronchial and bronchiolar epithelium resulting in airway occlusion by cell debris. As for bronchiolitis, RSV pneumonia is characterised by peri‐bronchiolar infiltration of inflammatory cells. The resulting pneumonitis is hallmarked by diffuse alveolar damage and extensive interstitial infiltration.174 This is often accompanied by intra‐alveolar leakage of fibrin and inflammatory cells and due to poor aeration, alveolar parenchyma necrosis, haemorrhage and oedema can be seen.53,172

Experience from these histological samples of fatal RSV cases demonstrated that RSV infection is highly restricted to the airway epithelium, specifically the apical surface of the epithelium (figure 4).174 The major cellular target of RSV are ciliated cells in both the upper and lower respiratory tract which fits with the early ciliated cell destruction as described above.44

Chapter 1: Introduction 54

Figure 4. Immunohistochemical staining for respiratory syncytial virus (RSV) antigen in bronchiolar and alveolar tissues from infants with lower respiratory tract infection (LRTI). Staining of autopsy tissues obtained from fatal cases of RSV‐LRTI in human infants and normal infant lung tissue stained as a control. Brown stain indicates the presence of viral antigen on apical surface. From Welliver, T. P. et al. Severe human lower respiratory tract illness caused by respiratory syncytial virus and influenza virus is characterized by the absence of pulmonary cytotoxic lymphocyte responses. J. Infect. Dis. 195, 1126–36 (2007).175 75Reprinted with permission from Oxford University Press.

1.6.3 Innate immune responses to RSV

As discussed above, complete protective immunity to RSV is never achieved and natural re‐ infection can occur many times throughout life.176 Generally, subsequent infections are more mild, but this is not always the case for very young infants and is troublesome in the elderly population, where RSV pneumonia and/or bronchiolitis contributes in up to 10% of hospitalisations.177–179 Accordingly, innate immune responses of airway epithelial cells represent a very important line of defence. In addition to providing a protective barrier, epithelial cells actively contribute to immune protection. The initial defence is provided by surfactant which has known antimicrobial properties and indeed early work on innate

Chapter 1: Introduction 55 defence against RSV demonstrated the ability of surfactant protein A to neutralise RSV in vitro by binding to RSV‐F.180 A range of other antimicrobial peptides are secreted by lung epithelia, including lactoferrin, lysozyme, ‐defensins and phospholipase A2.181 All of these peptides are known to be important in defending against viral infections, including RSV, which will be further discussed in chapter four. In addition to the initial defence provided by these endogenous proteins, further airway epithelial cell responses are induced to defend against invading pathogens, including interferon and pro‐inflammatory responses.

Central to this innate immune response induction is the identification of pathogens by pattern recognition receptors (PRRs), which detect pathogen‐associated molecular patterns

(PAMPs).181 These receptors include Nod‐like receptors (NLRs), RIG‐1 like receptors (RLRs), toll‐like receptor (TLR) and cytosolic DNA receptors.182 TLRs are among the most studied

PRR family and a number are known to be significant in RSV‐immune interactions. TLR2‐ deficient mice, for example, display reduced TNF‐ production and decreased neutrophil migration in response to RSV compared to wild type mice.183 TLR4 is perhaps the best characterised TLR in relation to RSV and, as outlined above, has been suggested as a possible

RSV receptor.79 Furthermore, there is a significant association between certain tlr4 gene polymorphisms and severe bronchiolitis.184–186 The nature of this association remains unclear, with conflicting in vitro study results. Early descriptions of TLR4 activities in RSV infection demonstrated that F‐protein mediated IL‐6 production by macrophages was dependent on functional TLR4 and noted a lack of IL‐6 production in TLR4 knockdown mice.187 However, more recently, Marr et al showed no effect on RSV‐mediated NF‐B activation or viral entry in the absence of a functional TLR4 receptor complex, suggesting

TLR4 is not required for RSV entry or activation of innate immunity.188 Further work by

Cabarello et al. confirms the association of TLR4 in RSV disease severity, but suggests this is mediated by prior environmental TLR4 activation conditioning the immune response to

Chapter 1: Introduction 56 infection rather than direct interaction of this PRR with the virus.189 TLR3 also mediates detection of RSV by recognition of RNA present during RSV replication. Interestingly tlr3 knockout mice still effectively clear RSV, but display goblet cell hyperplasia and mucous overproduction, which could contribute to more severe disease due to narrowing of the lumens within the respiratory tract.190

1.6.3.1 Antiviral role of Interferons

Once the immune system detects RSV, PRRs activate a variety of signalling pathways, including the interferon (IFN) family, a heterogeneous family of cytokines, well‐characterized in regards to the induction of antiviral states in cells.191,192 The IFN family represent a very important aspect of the innate immune response to viral pathogens. Type I IFN receptor and

STAT1 knockout mice are much more susceptible to a number of viral infections193,194 and humans with inborn deficiencies in IFN responses are often more readily susceptible to viral illnesses.195 Of the three known classes of IFNs, I, II and III, the importance of type III IFNs is increasingly identified in relation to RSV infection and it is known the anti‐RSV innate immune response is mediated in part by the type III interferon, IFN 1 (IL‐29).196 Indeed, type III IFNs are a relatively recent discovery and have very similar signalling, induction and biological activities to those of type I IFNs.197 Both type I and III IFNs can act via the JAK‐STAT signalling pathway, which promotes activation of a large number of genes via binding of downstream mediators with IFN‐stimulated response elements of specific genes, termed interferon‐stimulated genes (ISGs).

Chapter 1: Introduction 57 1.6.3.1 The role of interferon‐stimulated genes (ISGs) in immune response

Over thirty years of research on ISGs has identified an array of functions as part of a complex network of host immune defences.198 Controlling viral, bacterial and parasite infections through targeting functions required for pathogen lifecycles, ISGs can also induce upregulation of chemokine/cytokine receptors and others perform roles as negative regulators of immune signalling to reduce inflammation and promote return to cellular homeostasis.198,199 Of the hundreds of known ISGs, only a small number have been characterised in vivo with regards to their role in antiviral immunes responses.200 This is likely due in part to the fact that the full breadth of the ISG response was only uncovered with the introduction of genome wide transcriptional profiling just over a decade ago.200

1.6.3.2 The antiviral role of ISG15

One of the earliest ISGs induced following viral infection is Interferon stimulated gene 15

(ISG15), (encoded by isg15).201–203 This small, ubiquitin‐like molecule has well characterized antiviral properties. The first description of the antiviral properties of ISG15 was in mice, where Sindbis virus engineered to express ISG15, but not control virus, protected IFN‐/ receptor knockdown mice from virus‐induced lethality.204 Since this, ISG15 has been reported to demonstrate anti‐viral activity against numerous viral pathogens, in both in vitro and in vivo mice experiments,200 including chikagunya virus, influenza A, HSV‐1, Ebola.205,206

The means by which ISG15 exerts its antiviral functions has been extensively reviewed by morales et al.207 Briefly, ISG15 can inhibit virus release, as seen with HIV‐1 and Ebola.206,208

ISG15 can also act via conjugation, in a process termed ISGylation, to modify a range of target proteins including viral proteins, such as the NS1 protein of influenza A209 and host proteins,

Chapter 1: Introduction 58 such as STAT1 and JAK1.210 Alternatively, ISG15 can act in its free form, for example as noted in its antiviral role against Chikungunya virus infection in neonatal mice.211 However, this is not the full story for the role of ISG15 and the complexity of its functions are highlighted by work demonstrating apparent pro‐viral effects, such as the in vitro promotion of hepatitis C production by ISG15.212

In regards to RSV infection, ISG15 upregulation in blood and nasopharyngeal samples of RSV‐ infected infants correlates with viral load and in vitro, ISG15 was recently identified to have a vital anti‐RSV role, mediated in part via its conjugation actions.203,213 Further work is needed to more fully elucidate the role played by ISG15 and indeed other interferon stimulate genes in RSV infection.

1.6.3.3 Role of chemokine/cytokine responses in RSV infection

In addition to interferon stimulation, activation of PRRs in response to RSV interaction results in secretion of a range of pro‐inflammatory chemokines and cytokines involved in direct or indirect antiviral effects, including chemotaxis and activation of various immune cells. For instance, bronchoalveolar and nasal lavage studies of RSV‐infected infants demonstrated secretion of a number of pro‐inflammatory cytokines, including CCL5/RANTES, CCL3/MIP‐

1, CCL2/MCP‐1, CXCL8/IL‐8, CXCL10/IP‐10, TNF, and IL‐6.214–217 The major recruited leucocyte in both the upper and lower airway in response to RSV infection are neutrophils.218,219 Whether this recruitment provides a protective or pathogenic role is not fully understood. Data from animal models of enhanced vaccine‐mediated disease have highlighted that exaggerated immune responses likely play a significant role in pathogenesis.

Lung neutrophil recruitment was shown to induce airway damage in mouse models220 and several cytokines upregulated in RSV infection, including TNF‐α, CCL3/MIP‐1α and IL‐

Chapter 1: Introduction 59 8/CXCL8, induce neutrophil activation and migration.221,222 IL‐8 is a potent inducer of neutrophil chemotaxis and previously nasopharyngeal IL‐8 concentrations have been demonstrated to correlate with disease severity in RSV bronchiolitis.223 These data suggest an important role of neutrophils in contributing to RSV disease severity.

Secretion of pro‐inflammatory chemokines results in a positive feedback loop, where recruited and activated immune cells contribute to the production of inflammatory mediators. Plasmacytoid dendritic cells (pDCs) recruited to lungs in RSV infection are an important source of interferons and this enables them to modulate the immune response to

RSV.224 Likewise, macrophages, which are one of the earliest cell types to respond to RSV infection, secrete a number of cytokines, including IL‐1, IL‐6, IL‐8, IL‐10, IL‐12 and TNF‐, which recruit lymphocytes, neutrophils, natural killer (NK) cells and eosinophils to the lungs.3

Recruited eosinophils are activated by RSV interaction with surface TLRs and can assist in viral clearance through the My‐D88 pathway.225 On the other hand, in a mouse model of

RSV disease, recruited NK cells produce significant amounts of IFN‐, which contribute to the lung injury seen in RSV disease.226 This concept fits with the generally accepted notion that exaggerated immune responses contribute to the damage seen in RSV disease.44 However, the immune response to RSV and their contribution to pathogenesis are very complex and much more work is required to extensively explore aspects of host‐virus interaction.

1.7 The use of models in eliciting RSV pathogenesis

1.7.1 Animal and immortalised cell line models

Further study of immune responses to RSV disease is limited by the available models and how closely they resemble human host‐RSV in vivo interactions. Conventionally, our

Chapter 1: Introduction 60 understanding of host immune responses to RSV has come from the study of models that include both in vivo and in vitro systems. Animal models have been extensively reviewed by

Bem et al.227 A wide range of animal models have been developed to study RSV infection, including mice, cotton rats, ferrets, nonhuman primates (e.g. macaques), cattle and lambs.227–231 Cotton rat and murine models are the most widely used, with an estimated

77% of RSV publications having been conducted in mice.227,232 For the most part, cotton rat and murine models are utilised due to extensive experience and easy breeding and handling of the animals. Furthermore, a large array of molecular reagents is available to dissect murine/RSV responses and there is an ever expanding range of accessible reagents for cotton rat research. These models have led to some seminal observations relating to RSV immunopathogenesis. Indeed, the neonatal mouse model highlighted differences in RSV‐ stimulated airway epithelium immune responses, such as increased IL‐33 secretion, in the neonatal period versus adulthood.233 However, animal models are limited by species‐ specific responses.232 Mice and cotton rats, for instance, are only semi‐permissive to RSV infection. Thus, results obtained from these models have uncertain applicability when translated to human host‐RSV interactions. The more recently developed lamb model is of interest, as lambs demonstrate similar respiratory tract morphology and development to humans, as well as susceptibility to several strains of RSV, including human strains.228,234,235

However, challenges in housing and availability of molecular tools are significant limitations in this model type.227,236

An alternative model approach is to utilise cognate virus/host pairs, such as that of bovine

RSV. Bovine RSV is the closest pneumovirus to human RSV and commonalities of bovine airway morphology to that of humans makes this a very useful model.232 Nevertheless, the overarching limitation in interpretation of these animal models is the high degree to which

RSV is adapted to its human host. Even in nonhuman primates, with the exception of chimpanzees (which are a protected species), the replication and pathogenesis of RSV poorly

Chapter 1: Introduction 61 reflects human‐RSV interactions, not to mention the prohibitive expense of working with primate species.237,238

Immortalised cell lines, such as HEp‐2, A549, BEAS‐2B cells and primary human airway epithelium cell (PAECs) monolayers are relatively straightforward to use and have been extensively utilised through the literature. However, these models poorly represent the complex morphology and physiology of airway epithelium in vivo. Accordingly, findings may not be readily translatable to in vivo responses, as highlighted above in relation to heparin sulfate, which binds RSV‐G in immortalised cells lines but is not found on the apical surface of airway epithelial cells.87

1.7.2 The three‐dimensional airway epithelial cell model

More recently, three dimensional well‐differentiated primary airway epithelial cell culture models (WD‐PAECs) have been developed which may provide novel insights into the innate immune response to RSV infection.239–241 In brief, primary airway epithelial cells are seeded onto a semi‐permeable membrane to support cells, allowing diffusion of nutrients from growth media. Once cells reach confluency, an air‐liquid interface is created by removing medium on the apical compartment, which triggers differentiation to form a polarized pseudostratified epithelium.240 These more closely mimic in‐vivo airway epithelium morphology, physiology and innate immune responses to RSV infection in infants.242 As highlighted above from histological investigations of RSV disease, airway epithelial cells are the primary targets for RSV infection and replication in vivo. As such, RSV/WD‐PAEC models offer highly relevant alternatives to animal and cell line models. They recreate several hallmarks of RSV pathogenesis in infants, which include: RSV infection restricted to ciliated and occasional non‐ciliated cells, but not goblet cells;242 infection of non‐contiguous groups

Chapter 1: Introduction 62 of apical cells; reduced number of ciliated cells; increased goblet cell numbers and occasional syncytia (figure 6).172 They also produce similar chemokine/cytokine expression to profiles to those seen in vivo, including the production of IL‐8/CXCL8, IP‐10/CXCL10, IL‐6 and

RANTES/CCL5.239,240,243 This culture model may provide the opportunity to address a number of questions relating to RSV cytopathogenesis in humans, such the identification of an RSV receptor and mechanisms of induction of innate immune responses. However, there are also challenges, such as the technical difficulty of working with these cultures and in addition, they have a limited culture window during which they may be exploited.240 As technology develops further, primary cultured human nasal epithelial cells transfected with the catalytic component of telomerase have been created244 and permit longer life span for primary cell cultures which will assist with ongoing RSV research. Other new technology, such as lung organoids, are an exciting development in the field of lung research, enabling the study of epithelial‐epithelial and epithelial‐mesenchymal cell interaction. These cultures too have a number of practical difficulties to overcome with regards to RSV work, such as how best to achieve infection and sampling from the cultures.245

1.7.3 Nasal versus bronchial airway epithelial cell models

Much of the research published regarding use of WD‐PAECs has utilised tracheal or bronchial epithelium.239,240,246–248 However, RSV is a descending infection, beginning in the nasal epithelium and there is a growing body of evidence that nasal‐derived WD‐PAECs may act as reasonable surrogates for bronchial‐derived WD‐PAECs. This is a considerably important development given that brushings to obtain nasal epithelium cells are much more accessible than bronchial cells, which require sampling under a general anaesthetic. Indeed, the use of nasal epithelial sampling also opens the possibility of sequential nasal sampling, enabling the study of innate immune responses to pathogens over time and to assess the effect of age

Chapter 1: Introduction 63 and/or environmental factors on these responses. Utilising nasal epithelial cells as a surrogate model for bronchial cells is certainly an attractive prospect, but has been controversial in some areas of lung research. Devalia et al. first compared nasal and bronchial epithelial cell cultures and found little difference in the morphology or ciliary activity but did not look at innate immune responses of the cultures.249 A review of all publications registered in PubMed using the search criteria “nasal epithelial cell compared to bronchial epithelial cell” revealed 165 studies, of which 9 studies directly compared nasal epithelium (NE) with bronchial epithelium (BE) in either monolayer or air‐liquid interface

(ALI) culture models (table 2).

Chapter 1: Introduction 64 Study Study Cell culture Outcome of comparison stimulus type Becker et al. RSV Monolayer Similar proportions of NE and BE infected following RSV infection, BE released less infectious RSV than NE.250 McDougall et Inflammation Monolayer Similar expression profiles in al. inflammatory mediator release from NE and BE following cytokine stimulation. Absolute levels did differ between cell site of origin.251 Lopez‐Souza Human ALI HRV replicates more easily in BE than et al. rhinovirus NE. Greater RANTES, IL‐8, IL‐1, IP‐ (HRV) 10 and protein kinase expression in BE than NE.252 Comer et al. COPD and ALI Similar IL‐8 responses in BE and NE to cigarette LPS but differing responses for IL‐6 smoke release and TLR‐4 expression.253 Guo‐Park et RSV ALI RSV induced similar cytopathogenesis al. and pro‐inflammatory chemokine responses in BE and NE. NE chemokine/cytokine responses quantitatively lower than in BE.254 Pringle et al. Stimulated Monolayer Unstimulated BE and NE mediator with IL‐1 release similar. Release of IL‐6 and and TNF GSCF higher in NE compared to BE following stimulation.255 Thavagnanam Asthma, ALI Increased % goblet and ciliated cells et al. IL‐13 in BE compared to NE. NE not felt to stimulated be suitable morphological surrogate for BE in asthma studies.256 Alves et al. Rhinovirus Monolayer Similar levels of IFN‐ and IFN‐ and IL‐8 following rhinovirus infection in BE and NE.257 Roberts et al. Rhinovirus ALI NE cells demonstrated similar +/‐ IL‐13 IP‐10, Mx1 and eotaxin 3 responses treatment as BE cells to rhinovirus and/or IL‐13.258 Table 2. Summary of studies retrieved from PubMed directly comparing nasal and bronchial epithelial cell responses

Early work by McDougall et al. on nasal and bronchial epithelial cell monolayers identified similar expression profiles in inflammatory mediator release following cytokine stimulation, although the absolute levels did differ between cell site of origin.251 In relation to COPD and asthma studies, some differences have been highlighted between nasal and bronchial

Chapter 1: Introduction 65 epithelial cell cultures. For instance, work by Comer et al. on COPD patients found some similarities between nasal and epithelial cell cultures, such as IL‐8 responses to P. aureginosa

LPS, but not for other responses, such as TLR‐4 expression or IL‐6 release.253 The authors concluded that not all nasal epithelial cell responses can be readily extrapolated to those of lower airways in subjects with COPD. Likewise, Thavagnanam et al. found differences in goblet and ciliated cell proportions between nasal and bronchial epithelial cell cultures and also concluded nasal were not suitable surrogates for asthma studies. However, recent work in rhinovirus257,258 and RSV infection has suggested nasal epithelial cell cultures can act as surrogates for bronchial epithelium.254 This difference may be due to the fact that viral respiratory tract infections begin in the upper respiratory tract before descending to the lower respiratory tract, whereas COPD and asthma primarily affect the lower airways. With particular reference to RSV, Guo‐Parke et al. showed that RSV infection of well‐differentiated primary nasal epithelial cell cultures (WD‐PNECs) induced remarkably similar cytopathology and pro‐inflammatory responses compared to well‐differentiated primary bronchial epithelial cell cultures (WD‐PBECs), albeit with some quantitative differences.254 This provides strong evidence for the use of nasal epithelial cells in RSV research and provides a rationale for use of these cells in the study of airway innate immune responses to RSV in early life.

As highlighted above in relation to the development of RSV therapeutics and vaccines there is a large disconnect between preclinical work and subsequent delivery of effective novel therapies. Thus, the development of models that more closely resemble human host‐virus interactions are of vital importance. Specifically, it is very difficult to overcome the ethical and practical problems of studying innate immune responses in neonates. Accordingly, we have very little knowledge of the normal early life airway innate immune response to RSV.179

In stimulation assays of cord blood versus adult blood cells using TLR ligands, Kollmann et al. reported monocytes and plasmacytoid dendritic cells produced different quantitative and

Chapter 1: Introduction 66 qualitative cytokine responses, with neonatal monocytes producing more Th17 skewing cytokines.259 TLR responses, as detailed above, play a vital role in determining the outcome of RSV infection and, therefore, age‐related differences in these responses may help explain the striking susceptibility of very young infants to severe RSV disease. A limited number of human studies have highlighted differential age‐related innate immune responses in response to a variety of stimuli including RSV,260–265 as will be discussed in detail in chapter four. It is important to note, however, that these studies are typically conducted on monocyte of dendritic cell responses using cord blood and peripheral blood. Cord blood studies are significantly limited by being far from the site of RSV infection, which, as described above, is limited to the apical pole of airway epithelium. Accordingly, we currently know remarkably little of whether or not the airway epithelial immunity matures and develops with age or what factors might influence this development.179 Indeed there are no studies on the innate immune response of airway epithelial cells to RSV in very early life and this work will represent the first study in this area.

Chapter 1: Introduction 67 1.8 Concluding remarks and project aims

1.8.1 Summary of what is unknown regarding RSV disease

This introduction provides an overview of what is currently known in relation to RSV disease, its molecular structure, pathogenicity, site of infection, host interactions and subsequent immune responses. Despite over 60 years of research since its initial discovery, understanding of the mechanisms of disease following RSV infection of humans remains limited. The reasons for diversity of symptom severity in RSV‐related infection may be driven by factors inherent to the virus itself and may be influenced by viral subtype and genotype.

However, as discussed in this introduction, the role of host factors on the disease pathology have been increasingly highlighted, particularly the importance of innate immune responses to RSV. The extensive number of phase 1 and 2 clinical trials involving RSV vaccination and therapeutic candidates is very exciting and highlights the huge importance of this virus and its clinical burden. However, the lack of successful phase 3 trials, despite massive public and industrial investment, underscores the significant challenge of developing efficacious vaccines or therapies against RSV and points to what remains unknown. The molecular structure of RSV, especially its fusion protein, have been effectively characterised in detail, but this has not yet translated into a viable vaccine, likely due in no small part to a lack of understanding of the early immune responses to RSV.

Moreover, the ethical and practical challenges of completing research in the infant population where RSV is most problematic has resulted in the need for a variety of models of RSV disease as outlined above. These models present a number of limitations, which may explain the major disconnect between preclinical trial success and clinical translational failure. It is imperative, therefore, that we develop realistic models of human‐RSV interaction to achieve greater understanding of the underlying mechanisms causing RSV

Chapter 1: Introduction 68 disease in humans and thus provide a platform for development of treatments that translate into clinical practice. Given that the target for RSV infection is the airway epithelium, the relatively new development of three‐dimensional differentiated airway epithelial cell cultures provides an exciting opportunity to simulate in vivo RSV‐host interactions.

Furthermore, the use of primary nasal epithelial cells to create these cultures, as described in this introduction, provides increased opportunity to study RSV interactions in airway epithelial cells from infant groups sampled sequentially.

1.8.2 Overall aim of this project

It is clearly established that infants who are born prematurely or who are very young when they first encounter RSV are at increased risk of severe disease. However, reasons for this are not fully understood and the overall aim of this project is to identify possible innate immune players in this increased risk. To achieve this aim we will establish an ex‐vivo/in‐ vitro WD‐PNEC model of RSV‐host interaction using samples obtained from preterm and term infants at birth and one‐year‐old. We aim to identify if differential airway epithelial innate immune responses between high risk preterm and term infants exist inherently at birth and to establish how these innate immune responses develop over the first year of life.

We also aim to identify the function of the molecules pleiotrophin and ISG15 in human airway epithelial innate immune responses to RSV disease pathogenesis and to establish the feasibility of their possible role as biomarkers of severe RSV disease.

Chapter 1: Introduction 69 1.8.3 Specific project objectives and plan

Figure 5 below provides an overview of the plan for this thesis, including location of results by chapter. In chapter 3 we describe the establishment and characterisation of human WD‐

PNEC cultures derived from newborn infants born at term and prematurely. We also describe the characterisation of WD‐PNEC cultures derived from the same infants subsequently sampled at one‐year‐old. Using these unique sequentially sampled cultures, in chapter 4 we describe the investigation of our hypothesis that the airway epithelial innate immune responses to RSV differ between term and preterm infants and between infants at birth and one‐year‐old. Specifically, chapter 4 details the observed cytopathogenic, interferon and chemokine responses to RSV infection of WD‐PNECs from preterm and term infants at birth and one‐year‐old.

Many infants with severe RSV disease have no identifiable risk factors. Prior to commencing this thesis work, an extensive microarray study was undertaken comparing pan‐genomic transcriptome responses in WD‐PNECs derived from cohorts of infants with histories of mild and severe RSV disease. Differential expression of thirty eight significantly upregulated genes were validated by qRT‐PCR and several genes were significantly differentially expressed following infection of WD‐PNECs derived from severe versus mild cohorts. One of these genes, isg15, had expression that was substantially different between the mild and severe cohorts. ISG15, as detailed above, is known to play a role in the antiviral immune response and therefore we identified it as a protein of particular interest. The microarray data also demonstrated that endogenous pleiotrophin expression levels were significantly lower in WD‐PNECs derived from the severe compared to the mild cohort. Given the known interaction of PTN with nucleolin, which was recently identified as a possible RSV receptor, we sought to investigate its role in RSV infection. Based on these preliminary data

Chapter 1: Introduction 70 inferences, which are discussed in detail in chapter 5 and 6, we hypothesised that pleiotrophin (PTN) and ISG‐15 are differentially expressed in preterm infant versus term infants and contribute to the increased risk of severe RSV‐related disease in preterm infants.

Chapter 5 of this thesis is devoted to characterising the role of PTN in RSV disease. Our first objective is to determine the differential expression and secretion of PTN following RSV infection in WD‐PNECs derived from term and preterm infants at birth and over the first year of life. Our second objective is to identify if PTN has a specific anti‐viral role in RSV infection and, if so, how it performs this function. Chapter 6 is devoted to identifying and characterising the role of ISG15 in RSV disease. Specifically, we will determine the expression profile of ISG15 in newborn and one‐year‐old infants following RSV infection and establish if

ISG15 fulfils an anti‐viral and/or immunomodulatory role in RSV infection.

In conclusion, the specific aims of this PhD project are:

1. To establish and characterise WD‐PNECs derived from preterm and term infants at

birth and one‐year old.

2. To characterise RSV‐induced cytopathology, and innate immune responses in WD‐

PNECs derived from preterm and term infants at birth and one‐year‐old.

3. To describe the expression kinetics of pleiotrophin and ISG15 following RSV infection

in WD‐PNECs derived from preterm and term infants at birth and one‐year‐old and

to identify if differential expression exists in at‐risk groups.

4. To characterise the innate immune functions of pleiotrophin and isg15 with regards

to RSV infection.

Chapter 1: Introduction 71

Figure 5. Summary of project plan and chapter location of results

Chapter 1: Introduction 72

Chapter 1: Introduction 73

CHAPTER 2

MATERIALS AND METHODS

Chapter 2: Materials and Methods

Each of the results chapters 3, 4, 5 and 6 are presented as extended manuscripts and include detailed methods relevant to experiments described in each chapter. To reduce duplication of this information and to facilitate reproduction of all experimental components of this thesis, this methods chapter includes detailed Standard Operating Procedures (SOPs) for the experimental procedures most frequently used in the preparation of this thesis.

2.1 Cell lines culture protocols

Cell lines and growth media used for this work are described below, origins of reagents are indicated in table 27 at the end of this chapter.

2.1.1 Cell lines, Media and Reagents

Cell Line Tissue of Growth Medium Maintenance Medium origin HEp‐2 Human  500 mL 1x High  500 mL 1x High laryngeal glucose (4.5 g/L) glucose (4.5 g/L) carcinoma DMEM DMEM cells  50 mL HI FBS  5 mL HI FBS  5 mL Pen/strep  5 mL Pen/strep (10,000 U/mL) (10,000 U/mL) BEAS‐2B Human  500 mL 1x Low  500 mL 1x Low bronchial glucose (1 g/L) DMEM glucose (1 g/L) DMEM epithelial  50 mL HI FBS  5 mL HI FBS cells  5 mL Pen/strep  5 mL Pen/strep (10,000 U/mL) (10,000 U/mL) Table 3. List of cell lines, tissue of origin and growth and maintenance media used

Note: Infection medium used is the same as growth medium but without FBS supplementation.

Chapter 2: Materials and Methods 75 Reagent Preparation Heat inactivated fetal bovine serum (HI FBS was inactivated in a water bath at 56oC FBS) for 30 minutes prior to use. Trypsin‐EDTA Trypsin‐EDTA was diluted 1:10 in Phosphate Buffered saline (PBS) (pH 7.2) Table 4. Cell culture reagents

2.1.2 Routine sub‐culture of cell lines

The SOP below details the procedure used for routine trypsinisation and culture of cell lines.

Materials

 Cell monolayers in 175 cm2 (90% confluent) tissue culture flasks

 175 cm2 tissue culture flasks

 Growth medium as detailed above e.g. HEp‐2 growth medium: DMEM high glucose

(Gibco) (4.5 g/L) + 5% FBS (Gibco) + 1% pen/strep (Gibco)

 Phosphate buffered saline (PBS) (Gibco)

 Trypsin‐ethylenediaminetetraacetic acid (EDTA) (Gibco) (diluted 1:10 as described

above)

 Serological pipettes

 1% (w/v) Virkon solution (Du Pont, USA)

 Class II laminar flow safety cabinet

o  Cell culture incubator 37 C, 5% CO2, 80% relative humidity (RH)

Method

1. Check cell monolayer for confluency, health and contamination by eye and by

microscopic examination. Pour old growth medium into 1% Virkon

2. Wash the monolayer twice with 10 mL PBS (for a 175 cm2 flask).

Chapter 2: Materials and Methods 76 3. Pipette 3 mL trypsin‐EDTA and distribute over whole monolayer with gentle

rocking.

4. Place flask in 37°C for 2 to 5 min. Check by eye and stop the incubation when all

cells have detached. If all cells are not detached by 5 min, tap the side of the flask

to dislodge the cells remaining attached.

5. Inactivate the trypsin by adding 7 mL of growth medium. Carefully pipette up and

down several times to break cell clumps.

6. If necessary, perform a cell count (See 2.1.3). If a cell count is not needed continue

to the distribution of the cell suspension.

7. Dilute cell suspension accordingly and distribute in tissue culture flasks. Add

growth medium (final volume for 175 cm2 flask = 20 mL).

8. Incubate at 37°C.

9. Check cells daily for health, growth, and contamination.

2.1.3 Cell counting procedure

Cell counting was a necessary prerequisite for many experiments undertaken during this thesis work. This method is based on the fact that Trypan blue only stains dead cells, enabling an accurate count of living cells.

Materials

 Trypan blue 0.4% solution (Sigma, UK)

 Glass coverslip

 Haemocytometer (Neubauer, UK)

 Micropipette (P10)

 Micropipette tips

Chapter 2: Materials and Methods 77  Tissue culture flask with confluent cells

 Inverted microscope (Nikon Eclipse T100)

Method

1. Place the cover slip across chamber of haemocytometer.

2. Following trypsinisation, take a 10 µL aliquot of cell suspension and add to 10 µL of

trypan blue solution in a 1:1 dilution. Pipette up and down to mix well.

3. Slowly pipette the trypan blue/cell suspension mixture into the haemocytometer

chamber so it is completely filled without air bubbles. Ensure all of the grid is

covered.

4. Examine the haemocytometer under the microscope at x10 magnification.

5. Only unstained, viable cells within, or touching 2 of the 4 perimeter lines of each of

the four 1 mm2 grids are counted. Note; when applying the 2‐side rule cells touching

e.g., the top and right outside lines are counted in, while cells touching the bottom

and left outside lines are not counted. (represented below).

6. Calculate the average number of living cells for the 4 quadrants. Multiply this by 2

(in view of the 1:1 dilution of cell suspension with trypan blue) and due to the volume

of the haemocytometer chamber the cell concentration is always this number

x104/mL:

Cell concentration = average cell count for four quadrants x 2 x 104 cells/mL

7. Multiply this by the total volume of cell suspension to obtain the total cell count.

Chapter 2: Materials and Methods 78 Representative diagram of haemocytometer quadrant. Trypan blue stains dead cells dark blue as shown and are not included in the count of live cells. All cells overlapping the upper and right grid lines are included in the count. All cells overlapping the left and bottom grid lines are not included in the cell count.

2.1.4 Procedure for cell freezing of continuous cell lines

Materials

 Cell monolayer in 175 cm2 tissue culture flask, 70% confluence

 Growth medium appropriate to the cell line as in section 2.1.1 eg. HEp‐2 growth

medium: DMEM high glucose (4.5 g/L) + 5% FBS) + 1% pen/strep

 Phosphate buffered saline (PBS) (Gibco)

 Trypsin‐EDTA (Gibco)

Chapter 2: Materials and Methods 79  Serological pipettes (Starstedt, UK)

 Foetal bovine serum (FBS)

 Dimethyl sulfoxide (DMSO) (Sigma‐Aldrich)

 Micropipettes (P1000, P100)

 Micropipette tips

 Labelled cryovials

 Isoproponal cell freezing apparatus (i.e. Nalgene, Mr. Frosty) to achieve a rate of

cooling of approximately ‐1oC/min

 Low speed centrifuge

 ‐80oC Freezer

 Liquid nitrogen storage tank

Method

1. A healthy monolayer in a 175 cm2 flask is trypsinised and cell count performed as

described in sections 2.1.2 and 2.1.3.

2. The cell suspension is transferred into a 15 mL centrifuge tube and centrifuge at 129

x g for 5 min.

3. The supernatant is discarded and the pellet resuspended in cell freezing medium

(table 5) to produce a final concentration of 1x106 cells/mL.

Reagent 10 % Heat inactivated FBS 10 % DMSO 80% cell suspension Table 5. Cell freezing medium

4. Add 1 mL of the prepared cell suspension/freezing medium mixture per labelled

cryovial

Chapter 2: Materials and Methods 80 5. Next place cryovials into an isopropanol cell freezing apparatus (Nalgene, Mr Frosty),

ensuring that the contents of the apparatus are at RT.

6. Freeze cell freezing apparatus with the cryovials as per manufacturer’s instructions

at ‐80oC.

7. After 24 h, transfer the cryovials to the gaseous phase of the liquid nitrogen storage

tank.

2.1.5 Procedure for cell thawing of frozen continuous cell lines

Materials

 Cryovial containing frozen cells to be thawed

 Water bath at 37oC

 Growth medium (appropriate for cell line as described above in section 2.1.1)

 Serological pipettes (10 mL, 5 mL)

 15 mL Centrifuge tubes (Sarstedt)

 75 cm2 tissue culture flask

 Low speed centrifuge

o  Humidified cell culture incubator at 37 C, 5% CO2

Method

1. Remove cryovial of cells from liquid nitrogen storage tank and allow to stand at RT

for 1 min.

2. Rapidly transfer cryovial to the water bath at 37oC and wait until the cell suspension

is entirely thawed. (Remove cell suspension when just defrosted, do not leave for

longer than required.)

Chapter 2: Materials and Methods 81 3. Gently Transfer cells into a 15 mL centrifuge tube and centrifuge at 129 x g for 5 min.

4. Pipette off supernatant and gently resuspend cell pellet in 10 mL of appropriate

growth medium (as detailed in section 2.1.1).

5. Seed cell suspension directly into a 75 cm2 flask and place in a humidified cell culture

o incubator at 37 C, 5% CO2

Chapter 2: Materials and Methods 82 2.2 Well‐differentiated Primary Paediatric Airway Epithelial Cell (WD‐PAECs) culture protocols

2.2.1 Collagen coating of tissue culture flasks and Transwells

Materials

 PureCol Bovine collagen stock solution (3 mg/mL) (Advanced Biomatrix)

 Sterile double distilled (dd) H2O

 Tissue culture flasks: 25 cm2 and 75 cm2

 Transwells (6.5 mm diameter)

 Micropipettes (P1000, P100)

 Micropipette tips

 Serological pipettes

o  Humidified cell culture incubator 37 C, 5% CO2

Method

Tissue Culture flasks:

1. Dilute the PureCol stock solution 1:30 with sterile ddH2O.

2. For PAECs obtained fresh from donors, a 25 cm2 placed on its side is coated to create

a 10 cm2 surface, this should be coated using 1 mL of collagen solution. For 25 cm2 2

mL of diluted collagen is required and for 75 cm2 flask, 3 mL of collagen solution

should be added.

3. Ensure collagen solution completely covers the surface to be coated by gently

rocking the flask from side to side.

Chapter 2: Materials and Methods 83 4. Put flasks in a humidified cell culture incubator at 37oC for at least 24 h before use.

Note: coated flasks can be kept in the incubator until use and do not require “re‐

coating” once coated.

5. Prior to using the flasks ensure excess collagen solution is removed by aspiration.

Transwells:

1. Dilute the collagen stock solution 1:100 with sterile ddH2O.

2. Add 100 µl to the upper surface of each 6.5 mm diameter Transwell. Take care not

to damage the Transwell membrane while adding the collagen.

3. Place coated Transwells in a humidified cell culture incubator at 37oC for at least 24

h before use. Note: coated Transwells can be left in the incubator until use and do

not require “re‐coating” once coated.

4. Ensure that excess collagen solution is carefully aspirated prior to cell seeding and

Transwell plates should be sterilised using UV light for 10 min before use.

Chapter 2: Materials and Methods 84 2.2.2 Preparation of WD‐PAEC culture media

Two types of media were used for PAEC culture: monolayer growth medium and growth medium for differentiation.

PAECs monolayer growth medium

Materials

 Airway Epithelial Cell Growth Medium Kit (PromoCell):

Component Volume Final Concentration Airway Epithelial Cell Basal Medium (C‐21260) 500 mL Bovine Pituitary Extract 2 mL 0.4% (v/v) Epidermal Growth Factor (recombinant human) 500 μL 10 ng/mL Insulin (recombinant human) 500 μL 5 μg/mL Hydrocortisone 500 μL 0.5 μg/mL Epinephrine 500 μL 0.5 μg/mL Triiodo‐L‐thyronine 500 μL 6.7 ng/mL Transferrin (human) 500 μL 10 μg/mL Retinoic Acid 500 μL 0.1 ng/mL Table 6. Growth medium composition for PAEC cultures in monolayer

 Penicillin/ Streptomycin 10000 U/mL (Pen/Strep) 5 mL

 Micropipette (P1000)

 Micropipette tips

 Water bath at 37oC

Method

1. Using water bath defrost supplement pack and carefully pipette all pack contents

into the 500 mL bottle of Airway Epithelial Cell Basolateral Medium.

Chapter 2: Materials and Methods 85 2. Add 5 mL Pen/Strep to give 1% (v/v)

3. Mix gently before use

PAECs air‐liquid interface (ALI) growth medium

Materials

 Airway Epithelial Cell Basal Medium (C‐21260) 500 mL (PromoCell)

 2 X Airway Epithelial Cell Growth Medium Supplement Pack:

Component Volume Final Concentration Bovine Pituitary Extract 4 mL 0.8% (v/v) Epidermal Growth Factor (recombinant human) 1 mL 20 ng/mL Insulin (recombinant human) 1 mL 10 μg/mL Hydrocortisone 1 mL 1 μg/mL Epinephrine 1 mL 1 μg/mL Transferrin (human) 1 mL 20 μg/mL Table 7. Growth medium composition for 2 x ALI for PAEC cultures differentiation

Note: discard triiodothyronine and retinoic acid kit contents.

 2 X Pen/Strep 10000 U/mL 5 mL

 DMEM Low glucose (1 g/L) 500 mL

 All trans‐Retinoic Acid (50 mg)

 Bovine serum albumin (BSA) (30% in Dulbecco`s Phosphate Buffered Saline)

 50 mL sterile centrifuge tubes

 Tinfoil

 Micropipettes (P1000, P100, P10)

 Micropipette tips

 Serological pipettes

Chapter 2: Materials and Methods 86  ‐20oC freezer

Method

1. Bovine serum albumin (BSA) aliquots preparation :

BSA aliquots are prepared in a laminar flow hood. One mL of BSA solution (Sigma,

UK) is mixed 1:1 (v/v) with 1 mL of Airway Epithelial Cell Basal Medium to give a BSA

solution of concentration 150 mg/mL. Next 10 µL of this solution is further diluted

in 990 µL airway epithelial basolateral medium to give a solution of 1.5 mg/mL BSA.

This is repeated to create a stock of 1 mL aliquots.

2. Retinoic acid (RA) aliquot preparation:

RA is extremely light sensitive and accordingly, dilution and aliquoting should be

performed rapidly under the darkest conditions possible and kept on ice throughout

the procedure. Dissolve 50 mg of RA in 16.67 mL of sterile DMSO to give 3 mg/mL

solution. This solution is aliquoted into 110 µL volumes and stored at ‐20oC. To form

the working concentration, 100 µL of the 3 mg/mL RA solution is diluted in 20 mL

sterile DMSO and aliquoted into 60 µL volumes. Aliquots are stored at ‐20oC.

3. To prepare the 2x ALI medium, two Airway Epithelial Cell Growth Medium

supplement packs without triiodothyronine and retinoic acid, as detailed in table 7

above, are added to 500 mL Airway Epithelial Cell Basal Medium. Next add 5 mL

Pen/Strep, 1% (v/v) and 1 mL BSA 1.5 mg/mL (prepared as detailed in point 1 above).

4. Next prepare a 500 mL bottle of DMEM low glucose (1 g/L) and add 5 mL Pen/Strep

to give 1% (v/v).

5. To prepare working 1x ALI medium, pipette 25 mL of DMEM low glucose (1 g/L) into

a 50 mL centrifuge tube and 25 mL of 2x ALI medium as described above in point 3.

Next add 50 µL of the working concentration of RA, prepared as described in point

Chapter 2: Materials and Methods 87 2 above. Rapidly cover the 50 mL centrifuge tube with tinfoil to protect the working

medium from light.

6. Store prepared 1x ALI at 4oC for up to 4 weeks.

Component Final Concentration Bovine Pituitary Extract 52 μg /mL Epidermal Growth Factor (recombinant human) 10 ng/mL Insulin (recombinant human) 5 μg /mL Hydrocortisone 0.5 μg /mL Epinephrine 0.5 μg /mL Transferrin (human) 10 μg /mL BSA 1.5 μg /mL Retinoic Acid 15 ng/mL Penicillin 100 U/mL Streptomycin 100 μg /mL Table 8. Final concentrations of supplements in working 1x ALI medium

2.2.3 Procedure for obtaining nasal epithelial cell sampling from patients

Materials

 Brushes used to obtain brushings: Interdental brush 2.7mm brushes (Dent‐O‐Care,

London, England) for neonatal sampling and Cepillo cell sampler brush (Deltlab SLU,

Barcelona, Spain) for sampling from older infants.

Note: all brushes are autoclaved to ensure sterility prior to use.

 Sterile 0.9% Sodium Chloride (NaCl)

 Transport medium

Component Volume Final Concentration DMEM (Gibco) low glucose (1 g/L) 50 mL Penicillin/Streptomycin 10000 U/mL 50 μL 10 U/mL Table 9. Recipe for transport medium preparation

Chapter 2: Materials and Methods 88  15 mL centrifuge tubes

 Sterile Phosphate Buffered Saline (PBS) pH 7.4.

 Re‐sealable polythene bags

Method

1. Infant identified following discussion with clinical team.

2. Parents approached, informed of study and given patient information leaflet.

3. Informed consent obtained using written consent form after period of thinking time.

4. Infant prepared for sampling procedure: If neonate, by wrapping in a blanket

(option of parents to nurse infant or place in cot), if older infant, to be held in

parent’s arms.

5. Each nostril is brushed only once and second nostril brushing should not be

attempted if the infant is very upset by the procedure.

6. First dip the brush into the sterile 0.9% NaCl solution.

7. Stand to the right of infant’s head, gently insert the brush approximately 1 cm or

until resistance is felt. Perform gentle agitation aiming posteriorly and turn brush 2‐

3 times to obtain nasal basal cells.

8. After removing the brush, immediately insert into a sterile 15 ml centrifuge tube

containing 5 mL of PBS (pH 7.4, room temperature).

9. Repeat the procedure in the other nostril if tolerated using a second interdental

brush and place in the same centrifuge tube.

10. Cut off the excess length of the stem of the cytology/interdental brush (non‐brush

end).

11. Immediately add 5 mL transport medium (see above) to the centrifuge tube

containing the nasal brushes in PBS. Seal and label the transport centrifuge tube.

Chapter 2: Materials and Methods 89 12. Place tube in a sealed sterile plastic bag and transport sample to the tissue culture

laboratory for processing (see section 2.2.4).

2.2.4 Processing of nasal sample brushes

Materials

 Sterile tweezers

 Micropipette (P1000)

 Micropipette tips

 Serological pipettes (10 mL, 5 mL)

 PAEC monolayer medium (see section 2.2.2)

 Low speed centrifuge

 Sharps box labelled as containing biological material

 25 cm2 flask pre‐coated with collagen on its side to create a 10 cm2 area for cellular

adherence, referred to as a T‐10 flask. (as described in section 2.2.1 above)

Method

1. Grip the non‐brush end of the cytology brush with sterile tweezers.

2. Gently agitate brush upwards and downwards in medium contained in transport

centrifuge tube.

3. Using a P1000 micropipette gently expel monolayer medium over the brush bristles

into the transport centrifuge tube containing the brushes. Do this several times over

Chapter 2: Materials and Methods 90 all of the bristles to ensure all cellular material is removed from the brushes into the

transport tube.

4. Discard the brushes in a sharps box labelled as containing biological material.

5. Centrifuge the 15 mL transport tube containing the transport medium and the

monolayer medium used to dislodge cells at 129 x g for 5 min at RT.

6. Carefully pipette off all but the last 0.5 mL of supernatant and discard.

7. Gently re‐suspend the cell pellet in 2 mL of monolayer medium and transfer into the

25 cm2 flask pre‐coated on its side (T‐10 flask). This smaller surface area is

recommended for initial culture due to the low cell numbers likely to be obtained

from nasal brushing samples.

o 8. Place flask in a humidified cell culture incubator at 37 C, 5% CO2 and leave cells to

settle for 48‐72 h. Observe daily for cell adhesion, health and contamination.

9. To change PAEC monolayer medium, gently pipette off the old medium from the

flask. Place this old medium in a second labelled T‐10 flask and leave for a further

48 h to allow additional time for remaining cells to adhere. Add 2 mL fresh PAEC

monolayer medium to the first flask.

10. Change PAEC monolayer medium with a fresh 2 mL of medium every 48 to 72 h

11. Once cells reach approximately 80% confluency, the cells are passaged into a

collagen‐coated 75 cm2 flask using the procedure described below in section 2.2.5.

2.2.5 Passage of PAECs

Materials

 PAECs in tissue culture flask at 70‐80% confluency

 0.05% Trypsin‐EDTA prepared as detailed above in table 4

Chapter 2: Materials and Methods 91  DMEM low glucose (1 g/L) supplemented with 5% foetal bovine serum (FBS)

 PAEC Monolayer medium prepared as described in section 2.2.2

 Collagen‐coated flasks

 Low speed centrifuge

 15 mL sterile centrifuge tubes

 Micropipettes (P1000, P100)

 Micropipette tips

 Serological pipettes (10 mL, 5 mL)

 Inverted microscope (e.g., Nikon Eclipse T100)

o  Humidified cell culture incubator at 37 C, 5% CO2

Method

1. Gently pipette off old medium from the flask containing PAECs for passage.

2. Rinse cells gently twice with 2‐3 mL PBS (pH 7.4) and discard.

3. Add 3 mL trypsin‐EDTA and incubate the flask at 37oC for 3‐5 min until all cells have

detached and become singular. Confirm cell detachment using an inverted

microscope.

4. Next add 7 mL low glucose DMEM supplemented with 5% FBS in order to

inactivate trypsin‐EDTA and, after gently mixing, transfer the detached cell

suspension to 15 mL sterile centrifuge tube.

5. Centrifuge the tube at 129 x g for 5 min at RT.

6. Gently pipette off the supernatant being careful not to disturb the cell pellet.

7. Re‐suspend the cell pellet in 10 mL monolayer medium and distribute to 75 cm2

flask.

Chapter 2: Materials and Methods 92 o 8. Place flask in a humidified cell culture incubator at 37 C, 5% CO2. Change

monolayer medium every 2‐3 days to encourage cell growth.

2.2.6 Culture of well‐differentiated PAECs (WD‐PAECs)

Following attainment of 70‐80% confluence in 75 cm2 collagen coated flasks, PAECs can be seeded onto semi‐permeable membranes as part of the Transwell culture system allowing differentiation into a columnar, pseudo‐stratified epithelium to occur.

Materials

 PAECs in tissue culture flask at 70‐80% confluence

 0.05% Trypsin‐EDTA prepared as detailed above in table 4

 Phosphate Buffered Saline (PBS) (pH7.4)

 DMEM low glucose (1 g/L) supplemented with 5% foetal bovine serum (FBS)

 PAEC 1x ALI medium prepared as described in section 2.2.2

 Collagen‐coated Transwells (6.5 mm) as described in section 2.2.1

 Low speed centrifuge

 15 mL sterile centrifuge tubes

 Micropipettes (P1000, P100)

 Micropipette tips

 Serological pipettes (10 mL, 5 mL)

 Trypan blue 0.4 % (Sigma, UK)

 Haemocytometer

 UV steriliser (Syngene UV Crosslinker)

 Inverted phase contrast microscope (Nikon Eclipse T100)

Chapter 2: Materials and Methods 93 o  Humidified cell culture incubator at 37 C, 5% CO2

Methods

1. Passage PAECs from the confluent (70‐80%) monolayer in the 75 cm2 flask as

described above in section 2.2.5.

2. After centrifuging the cells and removal of supernatant, re‐suspend cells in 2 mL of

PAEC 1x ALI medium.

3. Perform cell count as described in section 2.1.3.

4. Prepare Transwells by carefully aspirating off excess collagen and then sterilising the

plate using a UV steriliser at 184 mW/m2 for 10 min.

5. Seed cells to the apical compartment at 1x105 density for 6.5 mm diameter

Transwells. Note: cells should be added apically in a total volume of 250 µL of PAEC

1x ALI medium. Take care not to allow any cells into the basolateral compartment.

6. Next add 750 µL PAEC 1x ALI medium to the basolateral compartment.

7. Label the Transwell plate and place in a humidified cell culture incubator at 37oC, 5%

CO2.

8. Leave Transwells undisturbed for 48 h to allow sufficient time for adherence of cells.

9. After 48 h replace the apical and basolateral media with fresh PAEC 1x ALI medium

every 48 h (see table 10 below for quantities). Be very careful when aspirating apical

compartment not to touch cells as this will damage the monolayer.

10. Monitor cell health and appearance every 24 h under a phase contrast microscope.

Note: to enable better visualisation of cells, remove apical compartment media prior

to viewing cells.

Chapter 2: Materials and Methods 94 11. PAECs should be maintained under submersion until they achieve 100 % confluence.

(This should take between 4 and 8 days, during which cells are kept in a humidified

o cell culture incubator at 37 C, 5% CO2.)

12. Once fully confluent, with no holes evident, gently aspirate the apical medium to

create an Air‐liquid interface (ALI). Replace the basolateral compartment medium

with 350 μL fresh medium.

Culture procedure for ALI:

13. Continue to replace the basolateral compartment with 350 μL fresh PAEC 1x ALI

medium every 48 h. Once mucous production begins gently wash the apical surface

once weekly with 200 μL low glucose DMEM (1 g/L). Ensure complete aspiration of

apical wash to return to ALI and be careful not to scratch the cell surface.

14. Signs of differentiation should appear from day 7 in ALI. However, PAEC cultures are

grown until they become fully differentiated with visualization of widespread

ciliated cells in all four quadrants and production of mucous. At this point they are

referred at as Well‐Differentiated (WD) PAECs. PAECs should be left for a minimum

of 21 days in ALI before use. However, it may take longer for full differentiation to

occur, as the time required may vary from donor to donor.

Volume PAEC 1x ALI in apical compartment before confluence 250 μL PAEC 1x ALI in basal compartment before confluence 750 μL PAEC 1x ALI in apical compartment after confluence and None formation of air‐liquid interface PAEC 1x ALI in basal compartment after confluence and 350 μL formation of air‐liquid interface DMEM low glucose (1 g/L) wash in apical compartment 200 μL after confluence and formation of air‐liquid interface Table 10. Volume of media and PBS recommended for culture of PAECs in Transwells

Chapter 2: Materials and Methods 95

2.2.7 Procedure for freezing PAECs

Materials

 PAEC monolayer in 75 cm2 tissue culture flask, 70‐80% confluence

 PAEC monolayer medium prepared as described in section 2.2.2

 Phosphate buffered saline (PBS) (Gibco)

 Trypsin‐EDTA (Gibco)

 Serological pipettes (Starstedt, UK)

 HI Foetal bovine serum (FBS)

 Dimethyl sulfoxide (DMSO) (Sigma‐Aldrich)

 Micropipettes (P1000, P100)

 Micropipette tips

 Labelled cryovials

 Isoproponal cell freezing apparatus (i.e. Nalgene, Mr. Frosty) to achieve a rate of

cooling of approximately ‐1oC/min

 Low speed centrifuge

 ‐80oC Freezer

 Liquid nitrogen storage tank

Method

This procedure follows on after trypsinisation of a healthy monolayer, with cell count performed as described above in sections 2.2.5 and 2.2.6.

Chapter 2: Materials and Methods 96 1. Once the cell count is performed, the cell suspension should be diluted to a final

concentration of 1 x 106 cells/mL. Table 11 shows the composition of the required

freezing medium:

2. Gently mix the freezing medium and add 1 mL per labelled cryovial

Reagent

10 % Heat inactivated FBS

10 % DMSO

80% cell suspension in PAEC monolayer medium

Table 11. PAEC freezing medium

3. Next place cryovials into an isopropanol cell freezing apparatus (Nalgene, Mr Frosty),

ensuring that the contents of the apparatus are at RT.

4. Place the cell freezing apparatus containing the labelled cryovials at ‐80oC, according

to manufacturer’s instructions.

5. After 24 h, transfer the cryovials to the gaseous phase of the liquid nitrogen storage

tank.

2.2.8 Procedure for thawing frozen PAECs

Materials

 Cryovial containing frozen cells to be thawed

 Water bath at 37oC

 PAEC monolayer medium (prepared as described in section 2.2.2)

 Serological pipettes (10 mL, 5 mL)

 15 mL centrifuge tubes (Sarstedt)

 Collagen‐coated 75 cm2 tissue culture flask (coated as described in section 2.2.1)

Chapter 2: Materials and Methods 97  Low speed centrifuge at RT

o  Humidified cell culture incubator at 37 C, 5% CO2

Method

1. Remove cryovial of cells from liquid nitrogen storage tank and allow to stand at RT

for 1 min.

2. Rapidly transfer cryovial to the water bath at 37oC and wait until the cell suspension

is entirely thawed. (Remove cell suspension when just defrosted, do not leave for

longer than required.)

3. Gently transfer cells into a 15 mL centrifuge tube and then rinse the cryovial with 1

mL of PAEC monolayer medium and subsequently transfer this to the centrifuge

tube.

4. Centrifuge the cell suspension at RT at 129 x g for 5 min.

5. Pipette off supernatant and gently re‐suspend cell pellet in 1 mL of PAEC monolayer

medium.

6. Seed cell suspension directly into a collagen‐coated 75 cm2 flask (after having

removed excess collagen) and place in a humidified cell culture incubator at 37oC,

5% CO2

7. Leave cells for 48 h to attach to the flask before changing medium, as described

above in section 2.2.4.

Chapter 2: Materials and Methods 98 2.2.9 Trans‐Epithelial Electrical Resistance (TEER) measurement

Materials

 DMEM low glucose (1 g/L) (Gibco, UK) (no additives)

 Sterile forceps

 End Ohm tissue resistance measurement chambers for 6 mm tissue culture cups

(World Precision Instruments, UK)

 Epithelial Voltohmmeter (EVOM2) (World Precision Instruments, UK)

 70% Ethanol (Sigma, UK)

Methods

1. Immerse EVOM2 electrodes and chamber in 70% ethanol for 2 minutes and allow to

air dry in laminar flow hood.

2. Gently wash apical side of Transwell insert with DMEM low glucose to remove excess

mucous which could interfere with TEER measurements.

3. EVOM2 meter is tested and calibrated as per manufacturer’s instructions. A blank

reading is taken using an empty Transwell submerged in DMEM low glucose (250 μL

apically and enough DMEM low glucose in the basolateral compartment to reach the

same level as inside the Transwell).

4. Next place Transwell of interest in the EVOM2 meter chamber, filling the apical and

basolateral compartments with DMEM low glucose, as described above, and take a

reading.

5. The blank value should be subtracted from this reading and the result multiplied by

the surface area of the insert to obtain the actual resistance in .cm‐2

Chapter 2: Materials and Methods 99 2.2.10 Fixation of WD‐PAEC Cultures

Materials

 Freshly prepared (or frozen at ‐20 oC) 4% paraformaldehyde (PFA)

 Phosphate Buffered Saline (PBS) (pH 7.4) at RT

 Class II Laminar Flow Safety Cabinet

 Micropipettes (P1000, P100)

 Micropipette tips

 70% ethanol

 Parafilm

Methods

1. Add 250 μL and 500 μL 4% PFA, respectivel,y to the apical and basolateral

compartments of WD‐PAEC Transwell for fixation.

2. Incubate Transwell at RT for 45 min before carefully removing PFA.

3. Gently rinse Transwell x2 with PBS to remove residual PFA

4. Add 300 μL and 1 mL 70% ethanol, respectively, to the apical and basolateral

compartments of WD‐PAEC Transwell and wrap Transwell plate in parafilm.

5. Store the Transwell plate at 4oC. The fixed cultures will be stable for about one

month. For longer term storage the 70% ethanol will need replaced every few weeks

to ensure the Transwells do not dry out.

Chapter 2: Materials and Methods 100 2.3 Virus infection protocols

2.3.1 Viruses

Virus Source Titration method

RSV BT2a Clinical isolate from an infant hospitalised with TCID50 bronchiolitis in Belfast in 2005. Used at passage 4. rRSV A2/eGFP Kind gift from Prof. Ralph Tripp (University of TCID50 and Georgia) and Prof. Michael Teng (University of fluorescent South Florida) titration rRSV A2/mkate2 Rescued in Dr. Power’s laboratory from an TCID50 and infectious clone and helper plasmids kindly fluorescent provided by Dr Martin Moore (Emory University, titration Atlanta, USA) Table 12. Viruses used during the preparation of this thesis work

2.3.2 Culture and expansion of RSV stocks

Materials

 HEp‐2 cells (seeded at 8 x 106 in 175 cm2 tissue culture flasks the day before virus

infection)

 Frozen cryovial containing RSV stock for titration

 HEp‐2 infection medium (prepared as described section 2.1.1 above)

 HEp‐2 maintenance medium (prepared as described in table 3 above)

 Phosphate Buffered Saline (pH 7.4) (at 37oC)

 Water bath at 37oC

 Serological pipettes (5 mL, 10 mL and 25 mL)

 50 mL centrifuge tubes

 Sterile cell scrapers

 Pre‐labelled cryovials (2 mL)

 Sonicator bath (CREST)(Timer:10 min, Heater: 17, Gas Power: 9)

Chapter 2: Materials and Methods 101  Low speed centrifuge

 Inverted phase contrast microscope (Nikon Eclipse T100)

o  Humidified cell culture incubator at 37 C, 5% CO2

Method

1. Prior to commencing the RSV stock production, examine the HEp‐2 monolayers to

check cell health and confluence.

2. Defrost the RSV virus stock rapidly in a water bath at 37oC.

3. Pipette off the growth medium from the HEp‐2 monolayers.

4. Infect the HEp‐2 monolayers at an MOI of 0.1‐0.5 in a total volume of 5 mL of

infection medium per 175 cm2 tissue culture flask.

5. Incubate infected HEp‐2 monolayers at 37oC for 2 h. Rock the flasks every 15 min to

achieve even distribution of the inoculum throughout the 2 h period and to ensure

the monolayer does not dry out.

6. After 2 h remove the inoculum and gently rinse the infected monolayer once with 5

mL of warm PBS (pH7.4). Add 20 mL of maintenance medium per 175 cm2 flask.

7. Incubate the infected HEp‐2 monolayers for 24 h at 37oC.

8. Remove the maintenance medium and rinse the monolayers twice with 5 mL warm

PBS.

9. Add 20 mL of maintenance medium per flask and place in a humidified cell culture

o incubator at 37 C, 5% CO2.

10. Observe the infected HEp‐2 monolayers daily for signs of cytopathic effect (CPE).

11. When extensive CPE is evident, including large clumps of cells floating in the medium

and large multinucleated syncytia, harvest the virus.

Chapter 2: Materials and Methods 102 12. To harvest the virus, scrape all attached cells into the maintenance medium and

collect into 50 mL centrifuge tubes on ice.

13. Sonicate the tubes for 10 min under the conditions stated above.

14. Following sonication vortex the centrifuge tubes thoroughly and then centrifuge at

460 x g for 15 min at 4oC

15. Aliquot the supernatant into labelled cryovials on ice, snap freeze and store in liquid

nitrogen.

2.3.3 RSV titration using Tissue Culture Infectious Dose (TCID50) Assay

Materials

 HEp‐2 cells seeded at 5x104/well in two 24 well plates (per titration) 24 h prior to

titration

 Sterile 24 well plates

 Sterile 48 well plates or 1.5 mL Eppendorf tubes

 Infection medium prepared as detailed in section 2.1.1

 Maintenance medium prepared as detailed in section 2.1.1

 P1000 and P200 micropipettes

 Micropipette tips

 Serological pipettes

 Vacusafe (Integra)

 Water bath at 37oC

 Ice bucket and ice

 Class II laminar flow safety cabinet

 Inverted phase contrast microscope (Nikon Eclipse T100)

Chapter 2: Materials and Methods 103 o  Humidified cell culture incubator 37 C, 5% CO2

Method

HEp‐2 cell preparation

1. HEp‐2 cells are grown in 175 cm2 flasks until 80‐90% confluent.

2. Trypsinise HEp‐2 cells and perform cell count as detailed in sections 2.1.2 and 2.1.3.

3. Seed into 24 well plates at a concentration of 5 x 104 cells per well in cell growth

medium. Be careful to tap the side of each 24 well plate firmly to ensure even

distribution of HEp‐2 cells in well.

RSV Titration

4. Place the virus stock at RT for 1 min and then defrost quickly at 37oC. For all

subsequent steps, the virus must be kept on ice.

5. Next prepare a 10‐fold serial dilution of the virus: Add 900 µL infection medium to

8 wells of a 48‐well plate or 8 sterile Eppendorf tubes labelling this ‐1 to ‐8. Add 100

µL of the defrosted virus to the first tube/well (labelled ‐1) and mix well by pipetting

up and down. Transfer 100 µL of this inoculum to the next tube/well (labelled ‐2)

and mix well by gently pipetting up and down. Repeat the serial dilution until all 8

tubes/wells are included. Discard 100 µL from the final tube. The dilutions must be

kept on ice throughout.

6. Using the vacusafe device, aspirate the medium from a column of 4 wells of HEp‐2

cells on the 24 well plate and add 200 µL of the ‐8 dilution of inoculation to 4 wells

of the 24‐well plate. Repeat this for each subsequent dilution of inoculum moving

from highest to lowest dilution (ie. from ‐8 to ‐1). It is imperative not to move from

a lower dilution to a higher dilution of virus stock using the same pipette tip. Be

Chapter 2: Materials and Methods 104 careful to leave a small volume of medium in each well when aspirating to ensure

that the cells do not dry out.

o 7. Incubate the HEp‐2 cells and inoculum in a cell culture incubator at 37 C, 5% CO2,

for 1.5 h, gently rocking the 24 well plates every 30 min to ensure even distribution

of the virus and so that the wells do not dry out.

8. After 1.5 h aspirate the inoculum from the HEp‐2 cells starting at the highest dilution

(‐8) and moving to the lowest dilution (‐1). Add 1 mL of maintenance medium to

each well. Be careful to perform this quickly to avoid cells drying out and becoming

damaged.

o 9. Incubate the plate in a humidified incubator at 37 C, 5% CO2, for 7 days.

10. After 7 days, observe each well under phase contrast microscopy and mark wells

which are positive for cytopathic effect (CPE).

11. Calculate virus titres according to the Kärber equation as described below :

‐Log10 TCID50 = ‐Δ ‐ δ(S‐0.5)

Δ = Log10 of the last dilution demonstrating 100% CPE, i.e., all wells of the dilution

are positive for viral CPE.

δ = Log10 of the dilution factor (in this case the dilution factor is 10)

S = the sum of the fraction of all the wells inoculated for each dilution, including and

beyond the last dilution with 100% CPE, that are positive for viral CPE (as 4 wells are

inoculated, the fraction beyond the 100% CPE row will be 0.25, 0.5, 0.75).

Units used are Tissue Culture Infectious Dose 50 (TCID50),

Chapter 2: Materials and Methods 105 2.3.4 Fluorescent RSV titration using Fluorescent Forming Units (FFU) Assay

Materials

 HEp‐2 cells seeded at 5x104/well in two 24 well plates (per titration) 24 h prior to

titration

 Sterile 24 well plates

 Sterile 48 well plates or 1.5 mL Eppendorf tubes

 Infection medium prepared as detailed in section 2.1.1

 Maintenance medium prepared as detailed in section 2.1.1

 P1000 and P200 micropipettes

 Micropipette tips

 Serological pipettes

 Vacusafe (Integra)

 Water bath at 37oC

 Ice bucket and ice

 UV Microscope with appropriate fluorescent filter

 Class II laminar flow safety cabinet

o  Humidified cell culture incubator 37 C, 5% CO2

Method

1. The protocol 2.3.3 above for virus titration by TCID50 is followed until point 8.

2. Incubate the infected HEp‐2 cells for 24 h at 37oC.

3. At 24 h post infection observe infected wells using microscope with fluorescent

filter. Identify the dilution which has between 20‐200 fluorescent foci evident.

Chapter 2: Materials and Methods 106 Count and record the total number of fluorescent foci in each of the 4 wells in this

dilution column.

4. Calculate the FFU using the following formula:

5 x (Average number of foci of the 4 wells) x 10x

X = ‐Log10 [dilution at which the foci were counted]

Worked example: an average number of 100 foci are counted at the fourth 1 in 10

dilution i.e., the 1 in 10,000 dilution (0.0001) dilution. Therefore, the FFU is equal to

5 x 100(average number of foci) x 104 FFU/mL = 5 x 106 FFU/mL.

Note: for all fluorescent forming viruses, titration was also performed using TCID50 to confirm the result obtained using the FFU method.

2.3.5 RSV infection of WD‐PAECs.

Materials

 WD‐PAEC cultures in Transwells to be used in the infection

 Inverted microscope (Nikon Eclipse T100)

 End Ohm tissue resistance measurement chambers for 6 mm tissue culture cups

(World Precision Instruments, UK)

 Epithelial Voltohmmeter (EVOM2) (World Precision Instruments, UK)

 1.5 mL Eppendorf tubes

 Trypan blue 0.4% solution (Sigma, UK)

 Glass coverslip

 Haemocytometer (Neubauer, UK)

Chapter 2: Materials and Methods 107  Trypsin‐EDTA prepared as detailed above in table 4

 DMEM low glucose (1 g/L) supplemented with 5% foetal bovine serum (FBS)

(Gibco)

 RSV stock to be used in infection

 Micropipettes (P1000, P100, P10)

 Micropipette tips

 DMEM low glucose (1 g/mL) (Gibco)

o  Humidified cell culture incubator 37 C, 5% CO2

 Pre‐labelled Cryovials

 Liquid nitrogen

 ‐80oC Freezer

Method

1. Observe WD‐PAEC Transwells under light microscopy to ensure satisfactory ciliary

coverage, mucus, and cell health prior to use in experiment.

2. Gently wash apical surface of WD‐PAECs once with 200 µL DMEM to remove surface

mucus and replace basolateral medium with 300 µL fresh PAEC 1x ALI medium.

3. Choose one representative Transwell to sacrifice. Perform TEER measurement as

described in section 2.2.9. Next add 200 µL Trypsin‐EDTA to apical and basolateral

compartment of Transwell and place in humidified cell culture incubator at 37oC, 5%

CO2 for 20 min to detach cells from Transwell membrane.

4. Firmly tap side of Transwell plate to ensure all cells are detached and singular and

transfer cell suspension to 1.5 mL Eppendorf tube. Add 500 µL DMEM low glucose

(1 g/L) supplemented with 5% foetal bovine serum (FBS) to inactivate Trypsin and

perform cell count as described in section 2.1.3. This will provide a specific cell count

Chapter 2: Materials and Methods 108 for use with Transwells in the experiment which should have been seeded at the

same density and time as the representative Transwell.

5. Calculate volume of virus stock required to achieve desired MOI and dilute in DMEM

low glucose (1 g/L) as needed.

6. Defrost virus stock as described in section 2.3.2 and add 100 µL of diluted virus stock

apically to WD‐PAEC Transwells. The volume in the apical compartment should not

exceed 250 µL (6.5 mm Transwells).

7. Incubate virus‐inoculated Transwells in humidified cell culture incubator at 37oC, 5%

CO2 for 2 h.

8. Gently wash apical surface 3 times with 250 µL DMEM low glucose (1 g/L).

9. Wash a 4th time, but retain this sample as the 2 h post‐infection (hpi) time point and

completely return WD‐PAEC cultures to ALI.

10. Immediately place the 2 hpi sample in a pre‐labelled cryovial and place on ice.

o 11. Return WD‐PAEC cultures to a humidified cell culture incubator at 37 C, 5% CO2.

12. Snap freeze the 2 hpi sample in liquid nitrogen as soon as possible thereafter and

store at ‐80oC or below until used.

2.3.6 Apical washing and basolateral medium harvesting from RSV‐ infected WD‐PAEC cultures

Materials

 WD‐PAECs infected with RSV as described in section 2.3.5.

 Micropipettes (P1000)

 Micropipette tips

 Pre‐labelled cryovials

 DMEM low glucose (1 g/mL) with no added supplements

Chapter 2: Materials and Methods 109  PAEC 1x ALI medium at RT

 Liquid nitrogen

 ‐80oC Freezer

Method

1. Apical washes and basolateral medium should be harvested at desired time points

post infection. Typically this is performed on a daily basis.

2. To perform apical washes, add 250 µL DMEM to the apical surface of the cultures.

Gently pipette the wash up and down three times and remove the apical wash

entirely, pipetting it into a pre‐labelled cryovial on ice.

3. As soon as possible thereafter, snap freeze apical washes in liquid nitrogen

4. Basolateral medium (300 µL) is removed and placed into a cryovial on ice. As for

apical washes, as soon as possible thereafter, snap freeze in liquid nitrogen.

5. Replace basolateral medium with 300 µL fresh PAEC 1x ALI medium.

6. Store apical washes and basolateral medium at ‐80oC until used.

2.3.7 Separation of Transwell membrane from Transwell insert following fixation for staining purposes

Materials

 WD‐PAECs fixed and stored in 70% ethanol as described in section 2.2.10

 Phosphate Buffered Saline (PBS) pH 7.4 at RT

 Scalpel (must be very sharp)

 Pasteur pipette

Chapter 2: Materials and Methods 110  Glass slide

Method

1. Carefully remove the 70% ethanol from the apical and basolateral compartments

using a Pasteur pipette.

2. Perform two gentle washes by adding 200 µL PBS to apical surface and 300 µL to

basolateral compartment and removing using Pasteur pipette.

3. Lift the transwell and turn it upside down.

4. Using a very sharp scalpel, cut around the outside edge of the membrane. Be careful

not to cut any of the hard plastic of the Transwell insert. Do not cut the entire way

around the Transwell membrane, leave a small 2‐3 mm area attached to the insert.

5. Add a small drop of PBS to the surface of the glass slide.

6. To ensure that the cell side of the membrane is facing upwards when placing on a

slide, turn the Transwell the correct way up and use the hydrostatic pressure of the

PBS drop to direct the Transwell membrane onto the slide. Gently tilt the Transwell

insert to a 45o angle and using the scalpel, detach the remaining section of the

membrane from the insert.

2.3.8 General procedure for immunofluorescence staining of WD‐PAECs

Throughout this thesis work a number of different antigens were targeted for immunofluorescence staining and the specific protocols and antibodies utilised are detailed in the methods section of each relevant chapter. Below is a description of the general protocol for applying antibody staining in WD‐PAECs and all detailed protocols are a variation of this basic method.

Chapter 2: Materials and Methods 111 Materials

 Phosphate Buffered Saline (PBS) (pH 7.4) (Gibco)

 Triton X‐100 (Sigma, UK)

 Bovine serum albumin (BSA) flakes (Sigma, UK)

 Sterile Pasteur pipettes

 Micropipette (P200) (P20) (P10)

 Micropipette tips

 Rocker

 15 mL sterile centrifuge tubes

 Appropriate primary and secondary antibodies

 DAPI mounting medium (Vectashield)

 Glass Slides (VWR international)

 Glass Cover slips (Premier Scientific)

 1.5 ml Eppendorf tubes

 Nail polish

o  Incubator at 37 C, 5% CO2, 80%

 Slide boxes

 4oC refrigerator

Method

1. When staining an entire Transwell membrane the antibodies can be added whilst

the membrane is attached in the Transwell insert and then subsequently separated

as described in section 2.3.7. If only a section of the Transwell membrane is planned

for staining the membrane can first be separated and then staining performed.

Chapter 2: Materials and Methods 112 2. Aspirate 70% ethanol from Transwell of interest.

3. Wash the Transwell membrane twice with 200 µL PBS (pH 7.4) at RT.

4. Prepare a 0.2% solution of Triton X‐100 in PBS. Triton X‐100 is used to permeabilise

the cell membranes. If you are not staining intracellularly skip steps 4‐6 and 8.

5. Add 250 µL 0.2% Triton X‐100 to cover the apical surface of the Transwell.

6. Leave for 2 h at RT.

7. Dissolve Bovine Serum Albumin flakes in PBS to give a final concentration of 0.5%

(w/v).

8. Remove Triton X‐100 from Transwells.

9. Add 250 µL 0.5% BSA to cover the surface of the Transwell.

10. Leave for 30 min at RT

11. Prepare the primary antibody by diluting in 0.5% BSA to obtain a concentration

specified by the manufacturer’s instructions or by prior optimisation.

12. Add 200 µL of the primary antibody apically. If the primary antibody is conjugated

to a fluorophore, protect the Transwells from light as much as is practicable from

this step forwards.

13. Incubate at 37oC for 1 h or at 4oC overnight.

14. Remove primary antibody and add PBS (pH7.4) to Transwell to cover cells. Incubate

for 5 min at RT on gentle rocking before removing the PBS. Repeat this washing step

twice more. If a secondary antibody is not required (fluorophore‐conjugated primary

antibody used), proceed to step 23.

15. If a secondary antibody is necessary, dilute it in 0.5% BSA to obtain a concentration

specified by the manufacturer’s instructions or by prior optimisation.

16. Add 200 µL of the secondary antibody apically. Protect the Transwells from light as

much as is practicable from this point on.

17. Incubate at 37oC for 1 h.

Chapter 2: Materials and Methods 113 18. Remove secondary antibody and wash 3 times with PBS as in 14 above.

19. If staining for multiple proteins, ensure that the sequence of addition of primary

and/or secondary antibodies is appropriate to prevent inappropriate cross‐reaction

of the antibodies with off‐target proteins.

20. Once staining is complete, cut out the Transwell membrane onto a glass slide as

described in section 2.3.7.

21. Add a drop of DAPI mounting medium to the apical surface of the Transwell

membrane.

22. Gently cover with a coverslip, being careful to avoid any air bubbles.

23. Seal the outer edge of the coverslip with nail varnish and store in the dark at 4oC

until used for fluorescent microscopy.

2.3.9 Cytospins and smears of apical washes

To determine the effects of RSV infection of WD‐PAECs on epithelial cell sloughing, the apical surface of dedicated infected and control WD‐PAEC Tranwells were gently washed and cytospins or apical smears performed.

Materials

 DMEM low glucose (1 g/L) (Gibco)

 EZ single CYTOFUNNEL with White filter cards (ThermoFisher Scientific, UK)

 Cytocentrifuge, Cytospin 4 (Shandon, UK)

 Glass slides (VWR International)

 PAP Pen, hydrophobic barrier pen

 Freshly prepared (or frozen at ‐20 oC) 4% paraformaldehyde (PFA)

 Phosphate Buffered Saline (PBS) (pH 7.4) at RT

Chapter 2: Materials and Methods 114  Class II Laminar Flow Safety Cabinet

Method

1. Apical rinses, obtained as described in section 2.3.6 were harvested and 250 μL wash

was added to a cytospin chamber or dropped onto the surface of a glass slide

prepared previously using a PAP pen to draw a large circle to contain the apical wash

(these were termed apical smears).

2. Cytospin chambers were centrifuged at 1000 rpm for 4 min in a Cytospin 4 centrifuge

and the sample slide subsequently gently removed.

3. Slides were air dried at RT for 1 h.

4. Dried cells were fixed by the addition of 4% PFA for 15 min at RT and subsequently

rinsed x2 with PBS.

5. Fixed slides were placed in slide boxes and stored at ‐20 oC until used.

2.3.10 Confocal Microscopy

The LEICA SP5 is a scanned‐laser confocal microscope that has five high sensitivity detectors for a range of accurately defined wavelengths, permitting the emissions from fluorophores to be clearly distinguished. Orthogonal sections were undertaken using the Z‐stack process available on the microscope. Live pictures were taken with a resolution of 1024 x 1024 pixels and were all exported as .tiff files to ensure low compression rate and conserve image quality.

Chapter 2: Materials and Methods 115 2.4 General Molecular Procedures

2.4.1 RNA extraction procedure

RNA extraction occurred after trypsinisation and centrifugation of cells as described in section 2.1.2 for monolayer cells and section 2.3.5 for WD‐PAEC cultures. RNA extraction proceeded using a High Pure RNA extraction kit (Roche, UK) and below is a description of this procedure. All RNA extraction was performed in a dedicated room separate from other cell and virus work. Reagents and materials in the RNA extraction room were kept strictly separate from the area for preparation of qRT‐PCR and PCR reagents and materials. All materials used in RNA extraction were certified as RNAse free.

Materials

 RNAse free Microfuge Tubes (Thermofisher Scientific)

 Molecular Biology Grade RNAse free PBS (Sigma, UK)

 RNAse Zap (Ambion)

 Tabletop centrifuge

 ART Barrier Pipette tips (ThermoFisher Scientific)

 Pipettes (P1000, P200, P10)

 Molecular Biology Grade absolute ethanol (Sigma, UK)

 High Pure RNA extraction kit (Roche, UK)

Chapter 2: Materials and Methods 116 Component Function Volume Lysis/Binding Buffer 4.5 M Guanidine‐HCL, rom M Tris‐HCl, 25 mL 30% Triton X‐100 (w/v), pH 6.6 (25oC) DNase 1, recombinant 10 KU lyophilized DNase 1 Re‐suspend in lyophilizate 0.55 mL Elution Buffer DNase incubation 1 M NaCl, 20 mM Tris‐HCl and 10 mM 10 mL o Buffer MnCl2, pH 7.0 (25 C) Wash Buffer 1 5 M guanidine hydrochloride and 20 mM 33 mL (add 20 mL Tris‐HCl, pH 6.8 (25oC); final absolute ethanol concentrations after addition of 40 mL before first use) absolute ethanol Wash Buffer II 20 mM NaCl, 2 mM Tris‐HCl, pH 10 mL (add 40 mL 7.5 (25°C); final concentrations absolute ethanol after addition of 40 ml absolute before first use) ethanol Elution Buffer Water, PCR Grade 30 mL High Pure Filter Tubes 50 polypropylene tubes with two layers of glass fiber fleece Collection Tubes 50 polypropylene tubes Table 13. Contents of Roche High Pure RNA Extraction Kit

Method as detailed in Roche High Pure RNA extraction Kit protocol

1. Re‐suspend cell pellet in 200 μL RNAse free PBS.

2. Add 400 μL Lysis/Binding Buffer and vortex for 15 s.

3. Insert one High Pure Filter Tube into one Collection tube and pipette entire sample

into the upper reservoir of the Filter Tube

4. Insert assembled High Pure Filter Tube into a standard table‐top centrifuge.

Centrifuge the tube assembly 15 s at 8,000 x g.

5. After centrifugation remove the Filter Tube from the Collection Tube; discard the

flow through, and again combine the Filter Tube and used Collection Tube.

6. After re‐inserting the Filter Tube for each sample, pipette 90 μl DNase I Incubation

Buffer into a sterile reaction tube, add 10 μl DNase I, mix, and pipette the solution

onto the glass fiber fleece in the upper reservoir of the Filter Tube. Incubate for 15

min at +15 to +25°C.

Chapter 2: Materials and Methods 117 7. Add 500 μl Wash Buffer I to the upper reservoir of the Filter Tube assembly and

centrifuge 15 s at 8,000 × g. Discard the flow through and combine the Filter Tube

with the used Collection Tube.

8. Add 500 μl Wash Buffer II to the upper reservoir of the Filter Tube assembly and

centrifuge 15 s at 8,000 × g. Discard the flow through and combine the Filter Tube

with the used Collection Tube.

9. Add 200 μl Wash Buffer II to the upper reservoir of the Filter Tube assembly and

centrifuge for 2 min at maximum speed (approx. 13,000 × g) to remove any residual

Wash Buffer. (The extra centrifugation time ensures removal of residual Wash

Buffer.)

10. Discard the Collection Tube and insert the Filter Tube into a clean, sterile 1.5 ml

RNAse free microcentrifuge tube.

11. To elute the RNA add 50 – 100 μl Elution Buffer to the upper reservoir of the Filter

Tube. Centrifuge the tube assembly for 1 min at 8,000 × g. The microcentrifuge

tube contains the eluted, purified RNA, which can be used directly in RT‐PCR or

stored at –80°C for later analysis.

2.4.2 RNA Quality and Quantity analysis

All RNA samples were analysed for quality and quantity prior to use using RNA 6000 nano kit according to the following method.

Materials

 0.5 mL and 1.5 mL RNAse free Microfuge Tubes (Thermofisher Scientific)

 Molecular Biology Grade RNAse free PBS (Sigma, UK)

 RNAse Zap (Ambion)

Chapter 2: Materials and Methods 118  RNAse‐free water (Ambion)

 Tabletop Microcentrifuge

 Heating Block

 ART Barrier Pipette tips (ThermoFisher Scientific)

 Pipettes (P200, P10)

 Bioanalyzer 2100 (Agilent)

 Chip Priming Station (Agilent)

 IKA vortexer

 RNA 6000 nano kit (Agilent)

Component Quantity RNA Nano Chip 25 Electrode Cleaner 2 Syringe Kit 1 syringe PCR clean Eppendorf tubes for Gel‐Dye Mix 30 RNA 6000 NanoLadder – when unopened stored at ‐20 oC 1 vial RNA 6000 Nano Dye concentrate – keep in dark at 4oC 1 vial RNA 6000 Nano Gel Matrix – Store at 4oC 2 vials RNA 6000 Nano Marker – Store at 4oC 2 vials Spin Filters 4 Table 14. Contents of RNA 6000 nano kit (Agilent)

Method

1. All reagents are allowed to equibrilate to RT for 30 min before use.

2. Preparing the RNA Ladder:

Spin down briefly and pipette into RNA‐free vial. Heat denature the ladder for 2 min

at 70oC and then immediately cool the vial on ice. Aliquot 1.2 µL denatured ladder

in 0.5 mL RNase‐free vials and store at ‐70oC.

3. Preparing the Gel:

Chapter 2: Materials and Methods 119 Pipette 550 µL of RNA gel matrix into a spin filter and centrifuge at 1500 x g for 10

min at RT. Aliquot 65 µL of filtered gel into 0.5 mL RNase‐free microcentrifuge tubes

and store at 4oC. Use filtered gel within 4 weeks.

4. Preparing the Gel‐Dye Mix:

Equilibrate RNA dye concentrate to RT for 30 min (protect from light). Vortex RNA

dye concentrate for 10 s, spin down briefly and add 1 µL of dye into a 65 µL aliquot

of filtered gel (prepared as detailed in point 3). Vortex solution for 10 s and spin

tube at 13,000 x g for 10 min at RT (protect from light). Use prepared gel‐dye mix

within one day.

5. Loading the Gel‐Dye Mix:

Place a new RNA chip on the chip priming station and pipette 9 µL gel‐dye mix in the

well marked.

Attach the syringe and make sure the plunger is positioned at 1 mL and then close

the chip priming station. Press the plunger until it is held by the clip, wait for 30 s

and then release the clip. Wait for 5 s and then slowly return the plunger to the 1

mL position. Open the chip priming station and pipette 9 µL gel‐dye mix in the wells

marked.

6. Loading the Marker:

Pipette 5 μL of RNA nano marker in all 12 sample wells and in the ladder well as

shown:

Chapter 2: Materials and Methods 120

7. Next load the RNA ladder and samples for RNA analysis. Prior to adding the samples

and RNA ladder, heat denature them for 2 min at 70oC and then keep on ice. Pipette

1 μL of prepared ladder in to the ladder well and 1 μL of each sample into the 12

sample wells. (Note: pipette 1 μL of RNA Marker into any unused sample wells.)

Place prepared chip in the IKA vortexer and vortex for 1 min at 2400 rpm. Run

completed chip in the Agilent 2100 Bioanalyzer instrument within 5 min.

To minimise secondary structure, the RNA Nano kit RNA ladder and RNA samples were heat denatured (70oC, 2 min) before loading 1 μL into marked wells as shown.

After vortexing for 60 s at 2400 rpm using the IKA vortex (Agilent Technologies) the RNA

Nano chip was inserted into the Bioanalyser (Agilent 2100 Expert) for analysis.

2.4.3 cDNA transcription

Transcription of cDNA was carried out in a dedicated area, separate from qRT‐PCR experimentation and all mastermix solutions were produced in a PCR hood dedicated to this purpose.

Chapter 2: Materials and Methods 121 Materials

 Molecular Grade PCR Tubes

 Molecular Biology Grade RNAse free PBS (Sigma, UK)

 Molecular Grade H2O (Ambion)

 Tabletop Microcentrifuge

 ART Barrier Pipette tips (ThermoFisher Scientific)

 Pipettes (P200, P10)

 High‐Capacity cDNA Reverse Transcription Kit (Applied Biosystems, UK)

 Thermal Cycler (Bio rad)

Component Quantity 10X RT Buffer, 1.0 mL 2 Tubes 10X RT Random Primers, 1.0 mL 2 tubes 25X dNTP Mix (100mM) 1 Tube, 1.0 mL MultiScribe Reverse Transcriptase, 50 U/μL 1 Tube, 1.0 mL Table 15. Contents of High Capacity cDNA RT Kit

Method

1. Prepare 2X RT mater mix:

Allow the kit components to thaw on ice. Calculate the volume of components

needed to prepare the required number of reactions and prepare the RT master mix

on ice as follows:

Component Volume/Reaction (μL)

10X RT buffer 2.0 25X dNTP Mix (100mM) 0.8 10X RT Random Primers 2.0 MultiScribeTM Reverse Transcriptase 1.0

Nuclease‐Free H2O 4.2 Total per Reaction 10.0 Table 16. Volume of High Capacity cDNA kit components required per reaction

Chapter 2: Materials and Methods 122 2. Mix gently on ice and pipette 10 μL of 2X RT mastermix into individual PCR reaction

tube. Pipette 150 ng of sample RNA diluted in PCR grade H2O to a final volume of 10

μL (total volume of 20 μL per PCR reaction tube).

3. Briefly centrifuge the PCR reaction tubes to spin down contents and eliminate any

air bubbles.

4. Place tubes on ice until you are ready to load the thermal cycler.

5. Program the thermal cycler using the following conditions:

Step 1 Step 2 Step 3 Step 4

o Temperature ( C) 25 37 85 4

Time 10 min 120 min 5 min ∞

Table 17. Thermal cycler conditions for reverse transcription

6. Once samples have completed thermal cycling, prepared cDNA is stored at ‐20oC for

later analysis.

Chapter 2: Materials and Methods 123 2.5 Western Blot Procedure

Below details the western blot procedure used for detection of ISG15 protein expression in

BEAS 2B cells (see chapter 6).

Western Blot Reagents

Buffer preparations:

1 M Tris‐Cl

Dissolve 121.1 g of Tris (Sigma‐Aldrich) in 800 mL of distilled H2O. Adjust pH to desired value by adding concentrated HCl dropwise. Allow solution to cool to room temperature before making final adjustments to the pH. Adjust volume of final solution to 1 L with distilled H2O.

1.5 M Tris‐Cl pH 8.8

Dissolve 90.83 g Tris (Sigma‐Aldrich) in 400 mL of distilled H2O. Adjust pH to 8.8 using HCl added dropwise. Allow solution to cool to room temperature before making final adjustments to pH. Adjust final volume to 500 mL with distilled H2O.

0.5 M Tris‐Cl

Dissolve 30.28 g of Tris (Sigma‐Aldrich) in 400 mL of distilled H2O. Adjust pH to 6.8 using HCl added dropwise. Allow solution to cool to room temperature before making final adjustments to pH. Adjust final volume to 500 mL with distilled H2O.

SDS (10%) stock solution

Dissolve 10 g of Sodium dodecyl sulfate (SDS) (Sigma‐Aldrich) in 80 mL of distilled H2O and once fully dissolved add distilled H2O to final volume of 100 mL.

Chapter 2: Materials and Methods 124 2X Laemmli Buffer

Reagent Volume Final concentration

10% SDS solution 4 mL 4%

Glycerol (Sigma‐Aldrich) 2 mL 20%

1M Tris‐Cl (pH 6.8) 1.2 mL 120 mM

Distilled H2O 2.8 mL ‐

Total 10 mL ‐

Table 18. Laemmli Buffer preparation

Once prepared add bromophenol blue (Sigma‐Aldrich) to a final concentration of 0.02%

(w/v). Store the 2X Laemmli buffer at room temperature.

Before using Laemmli buffer add 2‐Mercaptoethanol (Sigma‐Aldrich) at ratio 1:19 (v/v) to give 20% of final volume (e.g. 20 μL 2‐Mercaptoethanol added to 380 μL 2X Laemmli buffer).

5 M NaCl

Add 146.1 g of NaCl (Sigma‐Aldrich) to 450 ml of distilled H2O by stirring and next add distilled

H2O until final volume is 500 ml.

0.5 M Ethylenediaminetetraacetic acid (EDTA) pH 8.0

Add 18.61 g EDTA disodium salt (Sigma‐Aldrich) to 80 mL distilled H2O. Adjust the pH using sodium hydroxide (NaOH) (Sigma‐Aldrich) until it is 8.0. Next make the solution up to 100 mL using distilled H2O.

Chapter 2: Materials and Methods 125 Lysis buffer

Reagent Volume RIPA buffer 970 μL Halt Phosphatase inhibitor cocktail (ThermoFisher Scientific) 100X, 10 μL vortexed before use 0.1 M Phenylmethanesulfonyl fluoride solution (in ethanol) (Sigma 1 μL Aldrich, UK) Halt Protease Inhibitor cocktail (ThermoFisher Scientific) 100X, vortexed 10 μL before use Total volume of prepared lysis buffer 1000 μL Table 19. Lysis Buffer preparation

RIPA buffer

Reagent Volume

1 M Tris‐Cl (pH 7.4) 10 mL

5 M NaCl 15 mL

0.5 M EDTA pH 8.0 1 mL

10 % SDS 5 mL

Triton X‐100 solution ((1%) ThermoFisher Scientific) 5 mL

Distilled H2O 464 mL

Total 500 mL

Table 20. RIPA Buffer preparation

Chapter 2: Materials and Methods 126 1X Running Buffer

Reagent Quantity

Glycine (Sigma‐Aldrich) 14.4 g

Tris (Sigma‐Aldrich) 3 g

SDS (Sigma‐Aldrich) (final concentration 0.1%) 1 g

Distilled H2O added to give final volume 1000 mL 1000 mL

Table 21. 1X Running Buffer preparation

Check pH and adjust to 8.3

1X Transfer Buffer

Reagent Quantity

Glycine (Sigma‐Aldrich) 14.4 g

Tris (Sigma‐Aldrich) 3 g

Distilled H2O 800 mL

Table 21. 1X Transfer Buffer preparation

Methanol (Sigma‐Aldrich) added fresh to 1X Transfer buffer at 20% (v/v) of final volume i.e.,

200 mL to above solution.

10X Tris Buffered Saline (TBS)

Reagent Quantity

Tris (Sigma‐Aldrich) 24.23 g

NaCl (Sigma‐Aldrich) 60 g

Distilled H2O 1000 mL

Table 22. 10X Tris Buffered Saline preparation pH adjusted by adding HCl dropwise to achieve pH 7.6.

Chapter 2: Materials and Methods 127 1X TBS‐Tween (TBST) pH 7.6

Add 100 mL of 10X TBS to 900 mL of distilled H2O to give 1X TBS. Next add 1 mL of Tween

20 (Sigma‐Aldrich) and store at room temperature. pH adjusted to pH 7.6.

Resolving Gel (10%)

Reagent Volume

Distilled H2O 4 mL

Acrylamide 30% solution (Sigma‐Aldrich) 3.3 mL

1.5 M Tris‐Cl (pH 8.8) 2.5 mL

10% SDS 0.1 mL

10 % Ammonium persulfate solution prepared fresh (w/v) in distilled H2O 0.1 mL

(Sigma‐Aldrich)

TEMED (ThermoFisher Scientific) 0.004 mL

Total 10 mL

Table 23. Resolving Gel (10%) preparation

Stacking Gel (5%)

Reagent Volume

Distilled H2O 3.4 mL

Acrylamide 30% solution (Sigma‐Aldrich) 0.83 mL

0.5 M Tris‐Cl (pH 6.8) 0.63 mL

10% SDS 0.05 mL

10 % Ammonium persulfate solution prepared fresh (w/v) in distilled H2O 0.05 mL

(Sigma‐Aldrich)

TEMED (ThermoFisher Scientific) 0.005 mL

Total 5 mL

Table 24. Stacking Gel (5%) preparation

Chapter 2: Materials and Methods 128 Stripping buffer

Glycine (Sigma‐Aldrich) 30 g

Sodium dodecyl sulfate (SDS) (Sigma‐Aldrich) 2 g

Tween 20 (Sigma‐Aldrich) 20 mL

Table 25. Stripping Buffer preparation

Above added to distilled H2O to give final volume of 1000 mL and pH adjusted to pH 2.2 with

HCl added dropwise.

Method

Cell lysis procedure

1. Cells are lysed directly on ice using 50 μL lysis buffer prepared as above.

2. Incubate on ice for 15 min and then remove lysis buffer using pipette into an

Eppendorf microfuge tube.

3. Spin samples at 16,099 x g for 10 min at 4oC (Eppendorf 5430 microcentrifuge).

4. Once centrifugation is complete, carefully remove the lysate and pipette into a

o cryovial before snap freezing in liquid N2. Samples were stored at ‐80 C for

subsequent analysis.

Western Blot Procedure

5. Casting SDS‐PAGE gel

Protein gel plates (Bio‐rad) are cleaned with distilled H2O and with 70% ethanol solution.

Next glass plates and spacers are assembled according to manufacturer’s instructions before pouring 10% resolving gel solution into the mould between the glass plates. Subsequently,

300 μL of isopropanol (Sigma‐Aldrich) is gently pipetted on top of the resolving gel to create

Chapter 2: Materials and Methods 129 a flat edge. Having allowed the gel to set at room temperature for 15‐30 min the isopropanol is then carefully poured off and the gel washed gently x3 with distilled H2O. Stacking gel solution (5%) is subsequently added on top of the hardened resolving gel and a comb placed into the mould. Care should be taken to remove all air bubbles from under and around the comb teeth. The gel is then allowed to set for 10‐15 min at room temperature.

6. Loading protein samples

Protein samples should be defrosted on ice and 2X Laemmli buffer added at a ratio of 1:1

(v/v). Samples are then boiled for 5 min at 95oC and immediately placed on ice. The gel is assembled into the Mini‐PROTEAN Tetra Cell (Bio‐rad) for handcast gels. Twenty μL of each protein sample and a molecular weight marker (10 μL) (Thermofisher Scientific) are then loaded into the SDS‐PAGE gel (10% as described above). Once all samples are loaded, 1X

Running Buffer is added to the top and bottom compartments of the Tetra cell (Bio‐rad) and the gel was run at 100 V for 1‐2 h to facilitate protein separation.

7. Transfer of proteins from gel to membrane

Following electrophoresis, the glass plates are gently separated and the gel placed in 1X

Transfer Buffer. Subsequently, a blotting pad pre‐soaked in 1X Transfer Buffer is placed on the semi‐dry western blot protein transfer apparatus, one piece of 3M filter paper pre‐ soaked in 1X Transfer Buffer is laid on top, followed by nitrocellulose membrane (Hybond‐c super, 0.45 micron, Amersham biosciences) cut to fit the gel size. The gel is placed on top of this nitrocellulose membrane, taking care to remove all air bubbles, followed by another piece of 3M filter paper and a blotting pad both pre‐soaked in 1X Transfer Buffer. Protein transfer is then performed using a semi‐dry protein transfer apparatus (Bio‐Rad Trans‐Blot

SD Cell) at 20 V for 2 h.

Chapter 2: Materials and Methods 130 8. Antibody staining of membrane

After transfer, the membrane is washed with TBST pH 7.6 and blocked for 1 h in TBST + 5 %

(v/v) BSA (Filtered) at RT. Following a further TBST wash, primary anti‐ISG15 antibody (Santa

Cruz, sc 50366, rabbit polyclonal IgG) is added at 1:3000 dilution (v/v) in TBST + 3% (v/v) BSA and incubated while rocking at 40C overnight. The membrane is then washed 3 x 10 min with

TBST whilst rocking before adding horseradish peroxidase‐conjugated goat anti‐rabbit IgG secondary antibody (Abcam, ab97051) at 1:5000 dilution (v/v) in TBST + 3% (v/v) BSA for 1 h at RT. Following 3 x 10 min washes using TBST, Pierce enhanced chemiluminescence (ECL)

Western Blotting Substrate (ThermoFisher Scientific) is added, according to manufacturer’s instructions. Proteins are visualised by chemiluminiscence using the G:BOX Chemi gel doc

Imaging System Instrument (Syngene).

For visualisation of ‐tubulin (loading control), the blot is stripped using stripping buffer applied for 10 min x 2 at RT. Subsequently, the blot is washed in PBS (pH 7.4) twice for 10 min each at RT and twice with TBST for 10 min each. Following washing, the blot is blocked for 1 h in TBST + 5 % (v/v) BSA (Filtered) at RT before addition of primary ‐tubulin antibody

(Abcam, rabbit polyclonal IgG) at 1:3000 dilution (v/v) in TBST + 3% (v/v) BSA at 4oC overnight. Washes, secondary antibody staining, and protein visualisation proceed as described above, using G:BOX Chemi gel doc Imaging System Instrument (Syngene). Images were analysed using Image J software.

Chapter 2: Materials and Methods 131 2.6 Statistical Analysis Methods

All statistical analysis described in this thesis were conducted using GraphPad Prism version

5.0 for Windows, GraphPad Software, La Jolla California USA, www.graphpad.com.

Data were entered into GraphPad Prism and reviewed to identify outliers. Outliers were checked to confirm accuracy of data entered and experimental conditions reviewed to ensure consistency of the scientific method. Having confirmed accuracy, all true outliers were included in analyses.

Descriptive statistics were applied to the data to identify their distribution and normality established by visual inspection or using D’Agostino and Pearson omnibus normality test and

Kolmogorov‐Smirnov tests. If normally distributed, data were described using mean, standard deviation and range, if non‐normally distributed, data were described using median and interquartile range.

Data were compared using t‐test or one‐way ANOVA of variance for normally distributed data sets and reported as mean difference, 95% confidence intervals and p values unless otherwise stated. For non‐normally distributed data sets, comparison was performed using non‐parametric equivalent as specified (eg. Mann‐Whitney test).

For analysis of serial measurement data sets, it is recommended that summary measures are used to describe data profiles.266 We used at least two summary measures to describe serial data, including slope obtained from linear regression and area under the curve analysis. Comparisons between data sets were conducted as detailed above. Statistical significance for all comparisons was defined as a p value <0.05.

Image analysis was conducted using ImageJ software, version 1.49 for Windows, ImageJ,

National Institutes of Health, USA, http://rsbweb.nih.gov/ij/. The threshold function of

Chapter 2: Materials and Methods 132 ImageJ was used to evaluate the percentage fluorescence and the average of at least 5 fields calculated as detailed in each chapter. All images were analysed by one observer to establish consistency. A randomly selected sample of images were analysed a second time by the first observer after a period of several months, and by a second independent observer In order to ensure reproducibility and validity of the technique for analysis.

Chapter 2: Materials and Methods 133 Material Supplier

ART PCR barrier pipette tips (10 μL) Molecular Bioproducts, ThermoFisher Scientific ART PCR barrier pipette tips (200 μL) Molecular Bioproducts, ThermoFisher Scientific ART PCR barrier pipette tips (1000 μL) Molecular Bioproducts, ThermoFisher Scientific EZ single CYTOFUNNEL with White filter ThermoFisher Scientific cards Glass slides VWR International

Glass Cover Slips Premier Scientific

Haemocytometer Neubauer, Germany

Parafilm M Sigma, UK

Pasteur Pipettes Deltalab, UK

PCR tubes (200 μL) Invitrogen, UK

RNAse free microfuge tubes (2 mL) ThermoFisher Scientific

Serological pipette 5 mL Starstedt, UK

Serological pipette 10 mL Starstedt, UK

Serological pipette 25 mL Starstedt, UK

Tissue culture plates, 24 well Corning, UK

Tissue culture plates, 48 well Corning, UK

Tissue culture plates, 96 well Corning, UK

Tissue culture flasks, 25 cm2 Davidson & Hardy, UK

Tissue culture flasks, 75 cm2 Davidson & Hardy, UK

Tissue culture flasks, 175 cm2 Davidson & Hardy, UK

Table 26. List of Materials and supplier detailed in this thesis

Chapter 2: Materials and Methods 134 Reagents and Kits Supplier Absolut Ethanol (molecular biology grade) Sigma, UK Acrylamide 30% solution Sigma‐Aldrich, UK Airway Epithelial Cell Basal Medium PromoCell, Germany Airway Epithelial Cell Basal Medium Supplement pack PromoCell, Germany 10% Ammonium Persulphate Sigma‐Aldrich, UK Bovine Serum Albumin flakes Sigma, UK Bovine Collagen, PureCol® Advanced Biomatrix, USA DAPI mounting solution VectorShield, UK Dimethyl Sulphoxide, DMSO Sigma, UK Dulbecco's Modified Eagle Medium, DMEM Gibco, UK EDTA disodium salt Sigma‐Aldrich, UK Fetal Bovine Serum Gibco, UK Glycine Sigma‐Aldrich, UK Glycerol Sigma‐Aldrich, UK Halt Phosphatase inhibitor cocktail 100X ThermoFisher Scientific Halt Protease Inhibitor cocktail 100X ThermoFisher Scientific High Capacity cDNA Reverse Transcription Kit Applied Biosystems, USA High Pure RNA extraction kit Roche Life Science, UK Lipofectamine RNAiMax Invitrogen, UK Methanol Sigma‐Aldrich, UK 2‐Mercaptoethanol Sigma‐Aldrich, UK NaCL Sigma, UK NaOH Sigma, UK Opti‐MEM Medium ThermoFisher Scientific, UK Paraformaldehyde Qiagen, UK Penicillin/Streptomycin Gibco, UK Phenylmethanesulfonyl fluoride solution 0.1 M Sigma Aldrich, UK Phosphate Buffered Saline (PBS) solution Gibco, UK Pierce enhanced chemiluminescence (ECL) Western ThermoFisher Scientific Blotting Substrate RealTime Ready qPCR assay Roche Life Science, UK RNA 6000 Nano Kit Agilent Technologies, USA

Chapter 2: Materials and Methods 135 RNAse free PBS Sigma, UK RNAse Zap Ambion, UK

RNAse free H2O Ambion, UK siRNA silencer select Ambion, UK Sodium dodecyl sulfate (SDS) Sigma‐Aldrich, UK TEMED ThermoFisher Scientific, UK Tris Sigma‐Aldrich, UK Triton‐X Sigma, UK Trypan blue Sigma, UK Trypsin‐EDTA (10X) Gibco, UK Tween 20 Sigma‐Aldrich, UK Virkon Du Pont, USA Table 27. List of Suppliers of reagents and kits detailed in this thesis

Chapter 2: Materials and Methods 136

Chapter 2: Materials and Methods 137

CHAPTER 3

CHARACTERISATION OF WELL‐DIFFERENTIATED PRIMARY PAEDIATRIC NASAL EPITHELIAL CELL (WD‐PNEC) CULTURES DERIVED FROM TERM AND PRETERM INFANTS AT BIRTH AND ONE‐ YEAR‐OLD.

Chapter 3: Characterisation of well‐differentiated primary paediatric nasal epithelial cell (WD‐PNEC) cultures derived from term and preterm infants at birth and one‐year‐old.

3.1 Abstract

Background

Innate immune responses of airway epithelium are important defences against respiratory pathogens and allergens. Little is known regarding human neonatal airway epithelium and whether age‐related morphological and/or innate immune changes contribute to the development of airway disease.

Methods

We collected nasal epithelial cells from 41 newborn infants (23 term, 18 preterm) within 5 days of birth. Repeat sampling was achieved for 24 infants (13 term, 11 preterm) at a median age of 12.5 months. Morphologically and physiologically authentic well‐differentiated primary paediatric nasal epithelial cell (WD‐PNEC) cultures were generated and characterised using light microscopy and immunofluorescence.

Results

WD‐PNEC cultures were established for 15/23 (65%) term and 13/18 (72%) preterm samples at birth, and 9/13 (69%) term and 8/11 (73%) preterm samples at one‐year. Newborn and infant WD‐PNEC cultures demonstrated extensive cilia coverage, mucous production and tight junction integrity. Newborn WD‐PNECs took significantly longer to reach full differentiation and had greater proportions of goblet cells compared to one‐year repeat WD‐

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 139 PNECs. No differences were evident in ciliated/goblet cell proportions between term‐ and preterm‐derived WD‐PNECs.

Conclusion

WD‐PNEC culture generation from newborn infants is feasible and represents a powerful opportunity to study differential innate immune responses in human airway epithelium very early in life.

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 140 3.2 Introduction

The airway epithelium plays a crucial role in initiating airway innate immune response mechanisms in humans. It facilitates this initial response by providing a mechanical barrier to pathogen entry and releasing antimicrobial and inflammatory peptides in response to innate immune receptor stimulation by pathogens.267 Certain respiratory disorders, including asthma and cystic fibrosis, are associated with altered airway epithelial cell (AEC) immune responses and impaired barrier function of the epithelium.268,269 There is increasing evidence that asthma and other chronic respiratory disorders begin in early life and it is possible that airway innate immune responses undergo maturation in parallel with postnatal lung growth, differentiation, microbiome colonisation, and external infectious and non‐ infectious insults.268,270 Furthermore, young infants, especially those born prematurely, have increased susceptibility to severe respiratory disease following infections, such as respiratory syncytial virus.271,272 Very limited insights into the development of early AEC immune responses exist and a greater understanding of these responses is likely to yield novel insights into susceptibility and mechanisms behind childhood airway disease.

In adults and children AECs can be obtained from brushings of the nasal or bronchial tracts and prior studies have demonstrated the successful use of both bronchial and nasal AEC cultures in investigating early life respiratory disorders.242,254,273,274 However, undertaking bronchial brushings in very young infants is impractical, as acquiring samples could only be ethically conducted opportunistically when infants are intubated. As an alternative, the nasal passage provides an easily accessible source of AECs that is much less invasive. Recent publications have highlighted the potential benefit of this technique in investigating early changes in immune function in cystic fibrosis (CF) infants, in the study of the nasal transcriptome of infants, and in the innate immune responses of the airway epithelium to

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 141 allergens as part of asthma pathogenesis research.273,275,276 Mosler et al described the culture of nasal epithelial cells from CF infants, some of whom were as young as 1 month old, but to date, Miller et al is the only publication describing monolayer culture of nasal epithelial cells from newborn infants.273

One of the major drawbacks of studying human infant primary AECs is the challenge of reduced proliferation after a small number of cell passages. As a possible strategy to deal with this challenge, Wolf et al described the generation of conditionally reprogrammed cells

(CRC) from harvested infant AECs with enhanced proliferative and survival capacity.277

However, generating CRCs requires alteration of primary AECs and it remains unclear what impact this might have on AEC innate immune responses.

A further development in AEC culture has been the creation of differentiated epithelial cell cultures via formation of an air‐liquid interface. Indeed, we previously described the formation of well‐differentiated paediatric nasal airway epithelial cell cultures (WD‐PNECs) from older infants and demonstrated this model reproduces many of the hallmarks of respiratory syncytial virus (RSV) cytopathogenesis seen in vivo.254 Here we describe the first successful generation of WD‐PNECs from infants at birth. Our work, therefore, presents an exciting opportunity to study ‘‘naive’’ human airway epithelial cells in early life and to investigate the developmental immunobiology of the airway epithelium over the first year of life.

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 142 3.3 Methods

3.3.1 Subjects and study design

Healthy newborn term infants, (37‐42 weeks gestation) and preterm infants (28‐34 weeks gestation) underwent a nasal brushing procedure up to 5 days after birth (median age 2 days, range 6 hours to 5 days) at the Royal Jubilee Maternity Hospital, Belfast. Infants with known severe congenital anomaly of the airway, immunodeficiency, or congenital heart disease at the time of recruitment were excluded. A repeat nasal brushing sample was taken from a subset of infants at one‐year‐old and a medical questionnaire recording previous episodes of upper/lower respiratory tract symptoms and/or bronchiolitis was completed based on parent recall (see appendix).

3.3.2 Sampling of nasal epithelial cells

Nasal airway epithelial cells were harvested from healthy, non‐sedated neonates at the earliest opportunity following delivery using the method described in chapter 2, section

2.2.3. Nasal sampling was performed either with the neonate lying in a parent’s arms or in a cot using a technique as described by Miller et al.273 In brief, the infant’s head was gently secured using one hand and a 2.7mm diameter interdental brush (DentoCare Professional,

London, UK) was introduced into each nostril in turn to obtain cells from the medial aspect of the inferior turbinate. Brushes were dipped in sterile 0.9% (w/v) saline solution prior to sampling and a separate brush was used for each nostril (Figure 6). For preterm infants, nasal brushing was performed only if the supervising medical team confirmed the neonate was clinically stable and if the infant no longer required respiratory ventilatory support via an intranasal endotracheal tube.

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 143

Figure 6. Image of newborn infant undergoing nasal brushing to harvest nasal epithelial cells. Reproduced under the Creative Commons Attribution License with permission from Miller D, Turner SW, Spiteri‐Cornish D, et al. Culture of airway epithelial cells from neonates sampled within 48‐hours of birth. PLoS One. 2013;8(11):e78321. 273

Nasal AECs from infants at one‐year old were collected with the infant held in a parent’s arms. A cepillo cell sampler brush (Deltlab SLU, Barcelona, Spain) was introduced into each nostril in turn using the same technique as described above.

3.3.3 Processing of nasal samples and creation of WD‐PNEC cultures

Processing of nasal brushes and subsequent culture of primary nasal epithelial cells (PNECs) proceeded as described in chapter 2, sections 2.2.4‐2.2.6. In brief, each brush, with attached cells, was placed in sterile phosphate buffered saline (PBS) mixed with transport medium

(DMEM, 0.1% Penicillin/streptomycin (1:1 V/V)). Paediatric nasal cells were removed from the brushes and expanded in airway epithelial cell basal medium supplemented with an

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 144 airway epithelial cell growth medium supplement pack (Promocell, Heidelberg, Germany) using established methods.242,243 Upon reaching 70‐80% confluency, the cells were seeded at passage 3 onto collagen‐coated 6 mm diameter Transwells (Corning, Tewksbury,

Massachusetts) at a density of 1x105 cells/Transwell. After reaching confluency, air‐liquid interface (ALI) conditions were established and maintained until complete differentiation occurred. Following complete differentiation, which was defined by extensive cilia coverage and mucus production under light microscopy, these cultures were designated well‐ differentiated primary nasal epithelial cells (WD‐PNEC) cultures.

3.3.4 Measurement of trans‐epithelial electrical resistances (TEER)

To ensure the formation and integrity of tight junctions between cells in the epithelium, trans‐epithelial electrical resistance (TEER) measurements were taken using an EVOM meter

(World Precision Instruments, FL) as described in chapter 2, section 2.2.9. Measurements were performed three times per well and a mean resistance calculated. Values were then corrected for the blank resistance value and surface area.

3.3.5 Immunofluorescence microscopy for WD‐PNEC characterisation

For WD‐PNEC cultures, representative Transwells were fixed in 4% (v/v) paraformaldehyde

(PFA) for 30 min at room temperature (RT). Fixed Transwells were stored in 70% ethanol at

+4oC until used. To prepare for immunofluorescence staining, ethanol was removed and cells rinsed twice with 200‐300 μL PBS added to the apical surface (pH 7.4). Cells were permeabilised using 0.2% (v/v) Triton X‐100 (Sigma‐Aldrich, ST. Louis, Missouri) in PBS for 1 h at RT and subsequently blocked with 0.4% (w/v) bovine serum albumin (BSA) (Sigma‐

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 145 Aldrich, ST. Louis, Missouri) in PBS for 1 h at RT. Cultures were next stained for β‐tubulin,

Muc5Ac and nuclei (DAPI). Briefly, 200 μL rabbit anti‐Muc5Ac antibody (ab78660, Abcam,

Cambridge, UK) (1:100 dilution in 0.4% (w/v) BSA (Sigma‐Aldrich, ST. Louis, Missouri) in PBS) was added and incubated overnight at 4oC. Cultures were washed 3 times with 200‐300μL

PBS added to the apical surface (pH 7.4) for 5 min at RT. Two hundred μL anti‐rabbit secondary antibody (A11011, Alexa‐Fluor 568, Invitrogen, Waltham, Massachusetts) was added (1:500 dilution in 0.4% (w/v) BSA (Sigma‐Aldrich, ST. Louis, Missouri) in PBS) and incubated at 37oC for 1 h. Cultures were further washed 3 times with 200‐300μL PBS added to the apical surface (pH 7.4), and 200 μL Cy3‐conjugated mouse anti‐β‐tubulin antibody

(ab11309, Abcam, Cambridge, UK) added (1:200 dilution in 0.4% (w/v) BSA (Sigma‐Aldrich,

ST. Louis, Missouri) in PBS) and incubated at 37oC for 1 h. Following further 3 x 200‐300μL

PBS (pH 7.4) washes added to the apical surface, nuclei were stained using DAPI‐mounting medium (Vectashield, Vector Laboratories, Burlingame, California). Fluorescence was detected by confocal laser microscopy (TCS SP5 Leica, Germany) or by UV microscopy (Nikon

Eclipse 90i, Nikon, Surrey, UK).

Cells from one representative Transwell for each culture were trypsinised and the cell suspension was either smeared onto two microscope slides or 200‐250 μL of cell suspension was added to a cytofunnel (EZ single cytofunnel, Thermo Fisher Scientific, Waltham,

Massachusetts) and spun at 100 rpm (Using Thermo Shandon Cytospin 4 Cytocentrifuge) for

4 min onto a microscope slide. Smeared and cytospun slides were then fixed in 4% (v/v) PFA for 15 min at RT. Fixed slides were stored in the dark at ‐20oC until immunofluorescence was performed. Slides were stained either for ciliated cells (anti‐β‐tubulin) or goblet cells (anti‐

Muc5Ac). In brief, cells were permeabilised using 0.2% (v/v) Triton X‐100 (Sigma‐Aldrich, ST.

Louis, Missouri) in PBS (pH 7.4) for 30 min at RT and subsequently blocked with 0.4% (w/v) bovine serum albumin (BSA) (Sigma‐Aldrich, ST. Louis, Missouri) in PBS (pH 7.4) for 30 min

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 146 at RT. One slide was stained for Muc5Ac by addition of 100 μL rabbit anti‐Muc5Ac antibody

(ab78660, abcam) (1:100 dilution in 0.4% (w/v) BSA (Sigma‐Aldrich, ST. Louis, Missouri) in

PBS) incubated overnight at 4oC, followed by 3 x 200 μL PBS (pH 7.4) washes for 5 min at RT before addition of 100 μL anti‐rabbit secondary antibody (A11011, Alexa‐Fluor 568,

Invitrogen, Waltham, Massachusetts) (1:500 dilution in 0.4% (w/v) BSA (Sigma‐Aldrich, ST.

Louis, Missouri) in PBS) at 37oC for 1 h. A second slide was stained for β‐tubulin by addition of 100 μL Cy3‐conjugated mouse anti‐β‐tubulin antibody (ab11309, Abcam, Cambridge, UK)

(1:200 dilution in 0.4% (w/v) BSA (Sigma‐Aldrich, ST. Louis, Missouri) in PBS) incubated at

37oC for 1 h. Both slides next underwent 3 x 200 μL PBS (pH 7.4) washes and nuclei were stained using DAPI‐mounting medium (Vectashield, Vector Laboratories, Burlingame,

California). Quantification of ciliated, goblet and total DAPI+ cell numbers was undertaken for each slide by counting under fluorescent microscopy (Nikon Eclipse 90i; Nikon, Surrey,

UK) and the proportion of ciliated and goblet cells relative to total DAPI+ cell numbers was determined.

3.3.6 Freezing and defrosting of harvested nasal epithelial cells

Expanded nasal cells were frozen at passage three and defrosted as described in chapter 2, sections 2.2.7 and 2.2.8. In brief, expanded paediatric nasal cells were trypsinised at passage

3, added to low glucose DMEM containing 5% (v/v) foetal bovine serum (FBS) and centrifuged for 5 min at 129 x g. Following supernatant removal, the resulting cell pellet was re‐suspended in monolayer medium (epithelial cell basal growth medium, Promocell,

Heidelberg, Germany) containing 10% FBS and 10% dimethylsulfoxide (DMSO) (Sigma‐

Aldrich, ST. Louis, Missouri) to give a final concentration of 1 x 106 cells/mL. One mL aliquots of the cell suspension in cryovials were placed into an isopropanol cell freezing apparatus

(Mr Frosty, Nalgene, Thermo Fisher Scientific, Waltham, Massachusetts) at RT and

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 147 transferred to ‐80oC for 24 h, before being stored long‐term in the gaseous phase of a liquid

Nitrogen (N2) tank.

To resuscitate frozen primary nasal epithelial cells, vials were removed from the liquid N2, defrosted rapidly in a water bath at 37oC and the contents centrifuged at 129 x g for 5 min.

The resulting cell pellet was re‐suspended in monolayer medium (epithelial cell basal growth medium, Promocell, Heidelberg, Germany), transferred to a collagen‐coated 75 cm2 flask

o (Thermo Fisher Scientific, Waltham, Massachusetts) and incubated at 37 C in 5% CO2. The generation of WD‐PNEC cultures then proceeded as described above.

3.3.7 Statistical analysis

Data are presented as means ± standard deviation and median, interquartile range (IQR) and range for skewed data. Statistical analysis was performed by a student’s paired or unpaired t‐test unless otherwise stated. Statistical significance was set at a p‐value less than 0.05 (* p < 0.05 or **p <0.01). Data were analysed using GraphPad® Prism 5.0 (GraphPad Software,

Inc., La Jolla, CA).

3.3.8 Ethics statement

Written informed consent was obtained from parents using standardised information leaflets and consent forms (see appendix) and basic demographic data was recorded at recruitment. The study was approved by the Office for Research Ethics Committee Northern

Ireland (ORECNI), (REC reference 14/NI/0056).

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 148 3.4 Results

3.4.1 Clinical characteristics and sampling of recruited infants at birth

Nasal AECs were collected from 41 newborn infants (23 term, 18 preterm) within the first 5 days of life (median age 2 days, range 6 hours to 5 days) (Table 28). All nasal samples from both preterm and term newborns were carried out unassisted by the same trained research doctor. The procedure lasted less than 5 seconds in duration and was well tolerated by all neonates with no adverse events or overt bleeding noted during the brushing. No respiratory adverse events were observed for any of the preterm neonates who underwent the procedure. Consistent with that reported by Miller et al 273 some infants cried during the procedure but stopped immediately upon being nursed by a parent/carer.

Three term (13%) and 9 preterm (50%) infants were delivered by Caesarean section and maternal smoking during pregnancy was present for 17% of enrolled infants. All mothers of preterm infants enrolled received antenatal steroids compared with only one mother in the term cohort, which is consistent with standard clinical practice for the obstetric management of preterm delivery. Perinatal clinical characteristics of enrolled infants are summarised in table 28.

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 149

Term (23) Preterm (18) Male (%) 9 (39%) 10 (56%) Mean gestational age (range) 279 days (260 to 291) 229 days (207 to 244) Mean birth weight (range) 3396 g (2200 to 4210) 1855 g (820 to 2540) Caesarean section delivery 3 (13%) 9 (50%) Siblings at home 0 11 (48%) 8 (44%) 1 10 (43%) 6 (33%) 2 2 (9%) 1 (6%) >2 0 (0%) 3 (17%) Maternal smoking antenatally (%) 3 (13%) 4 (22%) FH of asthma/atopy (%) 11 (48%) 7 (39%) Antenatal steroids received (%) 1 (4%) 18 (100%) Table 28. Perinatal and delivery characteristics of enrolled subjects. FH = Family history

3.4.2 WD‐PNEC cultures from preterm and term newborn infants

demonstrate similar differentiation schedules and success rates

Primary monolayers were successfully established in 18 term and 15 preterm infants (80%).

One sample failed to grow due to fungal contamination during culture. For the remaining samples, culture failure was due to insufficient cells harvested following brushing. Following establishment of primary monolayer culture (as shown in figure 7), the subsequent success rate of differentiation into WD‐PNECs was 28/33 (91%). Of the three cultures failing to achieve differentiation, no obvious cause was evident, but it may be due to low sample yield resulting in cell division beyond the upper limit at which successful cell differentiation can occur. Although numbers were low, there was no correlation evident between maternal antenatal smoking and cell culture failure. Overall, the rate of success for complete differentiation from initial sample acquisition was 15/23 (65%) for recruited term neonates and 13/18 (72%) for recruited preterm neonates. WD‐PNEC differentiation was established

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 150 by the formation of extensive cilial coverage in all quadrants of each Transwell with clear mucous production and no holes evident, as observed under light microscopy. No significant difference was noted in the median duration of time required to achieve differentiation in term (77.5 days; IQR: 74 to 94, range: 64 to 97 days) and preterm (80 days; IQR: 70 to 88, range 61 to 133 days) cultures.

Figure 7. Phase‐contrast light microscopy of cultured nasal epithelial cells derived from a newborn infant. (A) monolayer showing typical ‘‘cobblestone’’ morphology of epithelial cells and in (B) well‐differentiated epithelial cell cultures (magnification X 20)

3.4.3 Term‐ and preterm‐derived newborn WD‐PNEC cultures are

morphologically indistinguishable

Newborn term and preterm WD‐PNEC cultures were indistinguishable under phase‐contrast light microscopy (figure 8a). Fluorescent microscopy of fixed and stained cultures confirmed the formation of multi‐layered pseudostratified cultures containing ciliated and goblet cells

(figure 8b). The mean proportions of ciliated cells in fixed and stained Transwells were similar in term‐ and preterm‐derived WD‐PNEC cultures (29.2% and 34.2%, respectively), as were the goblet cell proportions (26.7% and 23%, respectively) (figure 8c). These proportions were comparable to mean proportions of ciliated and goblet cells observed in

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 151 representative trypsinised Transwell smears and cytospins (figure 8d). Tight junction integrity, as demonstrated by robust transepithelial electrical resistance (TEERs) of ≥300

Ω.cm‐2, was evident for most Transwells in both term and preterm derived WD‐PNEC cultures, with no significant difference between cohorts (figure 8e). Three term WD‐PNEC cultures had TEERs ≤300 Ω.cm‐2; however these cultures demonstrated extensive apical cilia coverage and obvious mucous production with no holes evident under phase‐contrast light microscopy.

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 152

A. B. β‐tubulin Muc5Ac

Term

Preterm

C. D. E.

term Term

preterm Preterm 4000 50 50

-2 40 3000 40 * cm * 30 30

 2000

20 20 1000 % Total cells % Total cells 10 10 TEER

0 0 0 Ciliated Cells Goblet Cells Ciliated Cells Goblet Cells Term Preterm

Figure 8. Morphology and differentiation status of newborn term and preterm WD‐PNEC cultures. Cultures were monitored by (A) phase‐contrast microscopy (magnification x20) or (B) confocal microscopy after staining for β‐tubulin (ciliated cell marker) (red), Muc5Ac (goblet cell marker) (green), or nuclei (DAPI) (blue). For (B), square panels represent en face images, whereas rectangular panels represent orthogonal sections, with the apical side at the top (magnification x63, with x1.5 digital zoom, bar = 20 μm). (C) Transwell cultures from term and preterm donors (n=4 each) were fixed and stained for ciliated or goblet cells. Images from 5 fields/Transwell were taken at magnification x60 and ciliated, goblet cell and total DAPI+ cell numbers were counted using fluorescent microscopy. The % ciliated and goblet cells were determined relative to total DAPI+ cell numbers. Data are presented as mean (±SD). (D) Transwells cultures were trypsinized, contents fixed onto slides by smearing or use of cytospin funnels, as described, and stained for ciliated (anti–β‐tubulin) and goblet (anti‐Muc5Ac) and total DAPI+ cells (n=9 term, n=10 preterm). Ciliated and goblet cell

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 153 numbers were expressed as a percentage of the total DAPI+ cell numbers. Data presented as mean (±SD). (E) Transepithelial electrical resistances (TEER) of neonatal term‐ and preterm‐derived WD‐PNECs. Measured by EVOM epithelial voltometer and presented as mean (±SD) Ohm.cm‐2

3.4.4 Clinical characteristics and repeat sampling of one year old infants

All infants sampled at birth were invited by both letter and telephone to return for repeat sampling at one year of age. Repeat sampling was achieved for 24 infants (13 term, 11 preterm) at a median age of 12.5 months (IQR: 12‐14.75 months, range: 12‐22 months)

(table 2). Follow‐up was not achieved in 17 infants either because parents declined re‐ attendance or did not respond to re‐attendance invitation. No infant’s parent requested total withdrawal from the study. Interestingly, infants born to mothers who smoked were twice as likely to fail to re‐attend compared to non‐smokers (relative risk 2.02, 95%CI: 1.05 to 3.89). As for sampling at birth, all nasal samples for recalled infants were carried out unassisted by the same trained research doctor and the nasal brushing procedure was well tolerated by all infants with no adverse events. Consistent with previous work, some infants cried during the procedure but stopped almost immediately upon being nursed by a parent/carer.

At one year follow‐up, only one term infant was reported to have confirmed RSV bronchiolitis on PCR testing of nasopharyngeal aspirate and required attendance at hospital for management of symptoms (admission <1 day). Two preterm infants had reported bronchiolitis, only one of which was confirmed as RSV positive and neither infant required hospital admission. There was no significant difference in hospital attendance/admission between cohorts for other conditions. Perinatal clinical characteristics of infants at one‐ year‐old are summarised in table 29.

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 154

Subject Age Sex Recurrent URTIs Bronchiolitis Atopy Medications (M) FT 16 F Y N N None FT 18 F N N N None FT 22 F Y N Y IC, β2 agonist FT 13 F Y N N None FT 15 M N N N None FT 15 M Y Y (RSV) Y None FT 13 F N N N None FT 15 F N N N None FT 13 F N N N None FT 14 M N N Y None FT 12 F N N Y None FT 14 F N N Y None FT 12 M N N Y None PM (33wks, 12 F Y Y (RSV) N None GA) PM (29wks, 12 M N N N None GA) PM (33wks, 12 M N N N None GA) PM (33wks, 12 M N N N None GA) PM (34 13 F N N N None wks, GA) PM (29 12 M N N N H2 antagonist wks, GA) PM (32 12 M Y N N None wks, GA) PM (33 12 F N N Y None wks, GA) PM (33 12 F N N Y None wks, GA) PM (33 12 F N N Y None wks, GA) PM (33 12 F N Y Y None wks, GA) Table 29. Clinical history of infants returning for repeat sampling after first year of life.

FT = Full term, PM = Preterm, GA = gestational age, M= Months, RSV = Respiratory Syncytial Virus, URTI= Upper respiratory tract infection, IC= inhaled corticosteroids

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 155 3.4.5 Nasal AECs from one‐year old infants achieve complete differentiation

faster than those from newborn infants

Primary monolayers were successfully established in 10/13 (77%) term and 9/11 (82%) preterm of repeat nasal brushing samples. Following establishment of primary monolayer culture the subsequent success rate of differentiation into WD‐PNECs was 17/19 (89%). Of the two cultures failing to achieve differentiation no obvious cause was evident. Overall, the rate of success for differentiation from initial sample acquisition was 9/13 (69%) for one‐ year term infants and 8/11 (73%) for one‐year preterm infants. Importantly, the time to achieve complete culture differentiation was significantly shorter for one‐year cohort samples (median 63.5 days; IQR 49 to 72 days) than for birth cohort samples (median 80 days; IQR 72 to 133 days), p=0.0001 (Mann‐Whitney U = 48.50, n1 16, n2 = 23, P =0.0001 two‐ tailed).

Adequate tight junction formation, as demonstrated by robust transepithelial electrical resistance (TEERs) of ≥300 Ohm.cm‐2, was evident for all Transwells in all one‐year repeat

WD‐PNEC cultures.

3.4.6 WD‐PNEC cultures derived from one‐year old infants demonstrate

significantly reduced goblet cell content compared to their paired

newborn‐dervied WD‐PNECs

One‐year WD‐PNEC cultures were indistinguishable from newborn‐derived cultures under phase‐contrast light microscopy. As for newborn‐derived WD‐PNECs, there was no difference observed between the proportions of ciliated and goblet cells in term and preterm‐derived cultures at one year (figure 9). However, we observed a significant

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 156 decrease in mean goblet cell proportions at one‐year compared to newborn‐derived WD‐

PNECs for both term (11.7% vs 25.5%; p=0.0003) and preterm cohorts (8.9% vs 33.1%;

(p<0.0001) when trypsinised Transwell cell smear proportions were examined (figure 9d).

We noted the total number of cells per Transwell culture was higher for the one‐year versus newborn cohort (Figure 9e). Therefore, to determine if the observed differences in goblet cell proportions could be explained by differences in total cell numbers within pseudostratified cultures we re‐analysed the data as total goblet cell numbers for each donor. Consistent with the proportion data, combined total goblet cell count for all newborn‐derived WD‐PNECs (mean= 93,797 cells; SD 39427) was double that of one‐year derived‐WD‐PNECs (mean= 46,969; SD 28607: t‐test of difference in means, p=0.0014)

(figure 9f). These data suggest the observed decrease in goblet cell proportions in one‐year‐ derived WD‐PNECs is not simply the result of increased total WD‐PNEC culture cell numbers in the one‐year cohort.

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 157

A. B. β‐tubulin Muc5Ac

Newborn

1‐year repeat

C. Newborn D. Term 1 yr Preterm 50 50 *** *** 40 40

30 30

20 20 % total cells % total cells 10 10

0 0 Ciliated Cells Goblet Cells Ciliated Cells Goblet cells Goblet cells Ciliated cells Newborn 1‐year repeat E. F. *** 600000 ** 100000

400000

50000 200000

Total count cell Total Goblet Cell Count 0 0 newborn 1-year repeat newborn 1-year repeat

Figure 9. Morphology and differentiation status of birth and one‐year repeat WD‐PNEC cultures. Cultures were monitored by (A) phase‐contrast microscopy (magnification x20) or (B) confocal microscopy after staining for β‐tubulin (ciliated cell marker) (red), Muc5Ac (goblet cell marker) (green), or nuclei (DAPI) (blue). For (B), square panels represent en face

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 158 images, whereas rectangular panels represent orthogonal sections, with the apical side at the top (magnification x63, with x1.5 digital zoom, bar = 20 μm). (C) Transwell cultures from birth and 1‐year donors (n=8 each) were fixed and stained for ciliated, goblet and total DAPI+ cells. Images from 5 fields/Transwell were taken at x60 magnification and ciliated, goblet and total DAPI+ cell numbers were counted using fluorescent microscopy. The % ciliated and goblet cells were determined relative to total DAPI+ cell numbers. (D) Representative preterm and term Transwell cultures (newborn term n=9, preterm n=10, one‐year term n=9, preterm n=6 donors) were trypsinised, contents fixed onto slides by smearing or use of cytospin funnels, as described, and stained for ciliated (anti–β‐tubulin), goblet (anti‐ Muc5Ac), and total DAPI+ cells. Ciliated and goblet cell numbers were expressed as mean (±SD) % proportion of total DAPI+ cell numbers for each cohort. *** unpaired t‐test, p<0.0001. (E) Transwell cultures were trypsinized and total cell count performed by trypan blue exclusion using a haemocytometer. Total cell numbers are presented as mean (±SD) for newborn (n=17) and 1‐year cohorts (n=15), p= 0.0087 (unpaired t‐test of diff in means). (F) Combined total goblet cell numbers were determined for trypsinised representative Transwell cultures. Data plotted for newborn (n=17) and 1‐year (n=15) cohorts as mean (±SD) for each culture cohort. ***unpaired t‐test of difference in means, p=0.0005.

3.4.7 Nasal primary AECs successfully differentiate after storage in liquid nitrogen and are morpholocially indistinguishable from freshly differentiated AECs

We next sought to determine if nasal AECs could be successfully frozen and stored, as this vastly increases the versatility of this model for future research into the origins of airway disease. Newborn and one‐year old nasal samples stored in liquid N2 for approximately one year were defrosted as described and expanded to enable Transwell seeding and differentiation.

Both samples successfully differentiated in a similar fashion, requiring one additional week to achieve complete differentiation compared to fresh counterparts. Differentiation was evidenced by the formation of extensive cilial coverage in all quadrants of each Transwell with clear mucous production under light microscopy (figure 10a). The extra week was due to additional time needed to enable cell expansion in monolayers prior to Transwell seeding.

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 159 Transepithelial electrical resistances (TEERs) >300 Ohm.cm‐2 were recorded for Transwells in both cultures indicating robust epithelial cell tight junction formation. As described above for freshly processed nasal epithelial cells, fluorescent microscopy of fixed and stained WD‐

PNEC cultures derived from frozen cells confirmed extensive ciliated and goblet cell contents

(figure 10b). Importantly, proportions of goblet and ciliated cells in frozen samples were very similar to those from the same donor cultured without freezing, with the unusually high goblet cell content from this donor being reproduced (figure 10c). While this was limited to cells from a single donor, our data demonstrate a proof of principle for the use of freezing newborn nasal epithelial cells for subsequent culture, differentiation and experimentation.

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 160 A. B. β‐tubulin Muc5Ac

C. Fresh sample 50 Frozen sample 40 30 20

% Total cells 10 0 Goblet cells Ciliated cells

Chapter 3: Characterisation of well-differentiated primary paediatric nasal epithelial cell (WD-PNEC) cultures derived from term and preterm infants at birth and one-year-old. 161