Dynamic surfactant metabolism in preterm infants

(1)

University of Southampton Research Repository

ePrints Soton

research or study, without prior permission or charge. This thesis cannot be

reproduced or quoted extensively from without first obtaining permission in writing from the copyright holder/s. The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the

copyright holders.

When referring to this work, full bibliographic details including the author, title, awarding institution and date of the thesis must be given e.g.

AUTHOR (year of submission) "Full thesis title", University of Southampton, name of the University School or Department, PhD Thesis, pagination

(2)

UNIVERSITY OF SOUTHAMPTON

Dynamic surfactant metabolism in

preterm infants

Kevin Colin William Goss, MB ChB MRCPCH

A thesis submitted to the

University of Southampton for the degree of

Doctor of Philosophy

Faculty of Medicine

University of Southampton

December 2012

(3)

(4)

University of Southampton

ABSTRACT

Faculty of Medicine

Doctor of Philosophy

DYNAMIC SURFACTANT METABOLISM IN PRETERM

INFANTS

by Kevin Colin William Goss

Exogenous surfactant therapy has dramatically improved survival in extremely preterm infants, however the turnover of exogenous and synthesis of endogenous surfactant components are still poorly understood in this group. Additionally there is evidence for this patient group that improving nutrition improves long-‐term outcomes in respiratory function, growth and neurodevelopment.

Phosphatidylcholine (PC) is the dominant phospholipid in both surfactant and in plasma and can be synthesised from choline by one of two pathways: the CDP-‐ choline pathway, which is present in all nucleated cells, or by three sequential methylations of phosphatidylethanolamine in the PEMT pathway, which is localised to hepatocytes and is the primary source of polyunsaturated PC species and de novo synthesis of choline. This study quantified choline phospholipid metabolism and pulmonary surfactant kinetics in preterm infants in vivo. Children aged between 23 and 28 weeks gestation and in receipt of exogenous surfactant were intravenously infused with [methyl-‐D9]choline chloride within 48 hours of

birth. Lipid extracts from sequential plasma and endotracheal aspirate samples were then analysed by electrospray ionisation tandem mass spectrometry (ESI-‐ MS/MS). Fractional incorporation into newly synthesised PC species is demonstrated rapidly in plasma samples at a higher rate than previously reported in adults, indicating a high level of hepatic activity for CDP-‐choline. Analysis of the PC species derived from the PEMT pathway shows significantly lower flux in this pathway than reported in adults. Finally incorporation into surfactant PC species is very low initially before rising slowly over several days and with the rapid changes in other acidic phospholipids suggests a rapid recycling of components of the exogenous surfactant not equilibrating with the CDP-‐choline pathway thereby providing evidence for the first time of differing rates of exogenous surfactant recycling versus de novo synthesis in the human preterm infant.

This study proves that the technique works in the clinical environment, is sensitive and rapid enough to provide data in a clinically relevant timeframe, opening the possibility for translational use to identify biomarkers for disease progression.

(5)

(6)

List of Figures

Figure 1. Pulmonary surfactant metabolism demonstrating synthesis and assembly in the alveolar type 2 cell (AT-‐II), transport to the cell surface, secretion of lamellar bodies and formation of tubular myelin, adsorption at air-‐liquid interface before degradation and recycling. [Diagram from Trapnell and Whitsett 2002] ... 15 Figure 2. The embryological development of alveoli showing absence of both surfactant synthesis and alveolar ducts capable of gas exchange prior to 23 weeks gestation. [Diagram from www.embryology.ch] ... 16 Figure 3. Structure of choline ... 19 Figure 4. A simplified schematic of tandem mass spectrometry ... 21 Figure 5. Illustration of CDP-‐choline pathway with preservation of all three methyl groups from choline through to phosphatidylcholine. Deuterium atoms have replaced hydrogen atoms in the methyl groups. ... 22 Figure 6. The synthesis of PC via the PEMT pathway. Only a single methyl group is preserved from choline in the final PC molecule. In this diagram deuterium atoms (D) have replaced hydrogen atoms (H) in the choline's methyl groups. ... 23 Figure 7. Fetal body fat accretion with gestational age during the fetal period (adapted from Widdowson,E.M. 1968 Growth and composition of the fetus and newborn. In The Biology of Gestation pp. 1-‐49. New York: Academic Press). ... 26 Figure 8. An example of a full positive ion scan spectrum. It illustrates a typical plasma lipid extract scan at recruitment (time point = 0) less than 48 hours after preterm birth. ... 51 Figure 9. An example of a precursor of mass/charge +184 (P184) scan spectrum. It illustrates a typical plasma lipid extract scan at recruitment (time point = 0) less than 48 hours after preterm birth. ... 52 Figure 10. Concentrations of phosphatidylcholine phospholipids in preterm plasma increase with time from recruitment. The error bars show the median values and interquartile ranges. Comparison of data using Mann-‐Whitney non-‐parametric significance testing shows significant increase with time to 120 hours (* p=0.0158 *** p=0.0004) ... 53

(11)

Figure 11. Average concentrations of selected phosphatidylcholine species in preterm plasma. PC16:0/18:1 is consistently the most abundant species in this population. ... 54 Figure 12. Absolute amount of PC measured in plasma over the first 24 hours after recruitment, corrected for time elapsed since preterm birth. There is a trend towards an increase in plasma concentration with time over the first 72 hours of life as indicated by the trendline. ... 55 Figure 13. Absolute amount of PC measured in plasma over the first 24 hours after recruitment, corrected for time elapsed since preterm birth. There is no correlation between gestational age at delivery and concentration measured. ... 56 Figure 14. Composition of phosphatidylcholine species found in plasma of preterm infants at recruitment, within 48 hours of birth. Species are arranged, left to right, by increasing mass/charge and are grouped, using the colours blue, red, green and purple, by the increasing number of double bonds associated with the fatty acid on the sn-‐2 position. Error bars represent one standard deviation. ... 58 Figure 15. Relative proportions of individual PC species measured in plasma over the first 120 hours after recruitment. Species are grouped by the degree of saturation in the fatty acid esterified to the sn-‐2 position. Fully saturated and monounsaturated (blue), diunsaturated (red), arachidonoyl (green) and docosahexaenoyl (purple). ... 60 Figure 16. Relative proportions of individual PC species measured in plasma over the first 120 hours after recruitment. The 18:2 containing species increase proportionally with time as the 20:4-‐containing species fall relatively from a high starting point at recruitment. ... 61 Figure 17. Relative proportions of individual PC species measured in plasma before and after the administration of TPN in patient SO02. The proportional changes occur after the administration of TPN, marked with an asterisk (*), between the samples collected at 24 and 48 hours. ... 63 Figure 18. Relative proportions of individual PC species measured in plasma before and after the administration of TPN in patient SO03. ... 63 Figure 19. Bottom -‐ An example of a precursor of mass/charge +184 (P184) scan spectrum. It illustrates a typical plasma lipid extract scan after 72 hours. Top -‐

(12)

The precursor of mass/charge +193 (P193) scan of the same lipid extract. There is a similar pattern of peaks but offset by a mass/charge of nine higher. ... 65 Figure 20. The mean fractional incorporation of D9-‐containing PC phospholipids in plasma over the first 120 hours after recruitment. D9 enrichment has been expressed as a percentage of all the PC phospholipids measured at each time point. The error bars represent a single standard deviation from the mean. .. 66 Figure 21. The mean fractional incorporation of D9-‐containing PC phospholipids in plasma over the first 240 hours after recruitment. D9 enrichment has been expressed as a percentage of all the PC phospholipids measured at each time point. The error bars represent a single standard deviation from the mean. .. 67 Figure 22. Comparison of the proportions of individual PC species at 6 hours after recruitment. For each pair of columns the endogenous PC proportions are shown as the left-‐hand column, while the right-‐hand column is the D9-‐labelled species. ... 69 Figure 23. Comparison of the proportions of individual PC species at 48 hours after recruitment. The columns are now more similar than at 6 hours. For each pair of columns the endogenous PC proportions are shown as the left-‐hand column, while the right-‐hand column is the D9-‐labelled species. ... 70 Figure 24. Comparison of the relative proportions of selected PC species up to 120 hours after recruitment. ... 70 Figure 25. Comparison of the relative proportions of selected diunsaturated and arachidonoyl-‐containing PC species up to 120 hours after recruitment. ... 71 Figure 26. Turnover of selected enriched PC species in plasma. Higher turnover, or relative loss, of a species is reflected in a steeper slope after the peak at 12 hours. ... 72 Figure 27. Turnover of enriched PC species in plasma with a 16:0 on the sn-‐1 position. Higher turnover, or relative loss, of a species is reflected in a steeper slope after the peak at 12 hours. ... 73 Figure 28. Turnover of enriched PC species in plasma with a 18:0 on the sn-‐1 position. Higher turnover, or relative loss, of a species is reflected in a steeper slope after the peak at 12 hours. ... 74

(13)

Figure 29. Turnover of enriched PC species in plasma with a polyunsaturated fatty acid on the sn-‐2 position. Higher turnover, or relative loss, of a species is reflected in a steeper slope after the peak at 12 hours. ... 75 Figure 30. Bottom -‐ An example of a precursor of mass/charge +184 (P184) scan spectrum. It illustrates a typical plasma lipid extract scan after 72 hours. Top -‐ The precursor of mass/charge +187 (P187) scan of the same lipid extract. There is a similar pattern of peaks but offset by a mass/charge of three higher. ... 82 Figure 31. Comparison of the proportions of individual PC species at 12 hours after recruitment. For each pair of columns the endogenous PC proportions are shown as the left-‐hand column, while the right-‐hand column is the PEMT derived D3-‐labelled species. Error bars represent one standard deviation. .... 83 Figure 32. Comparison of the relative proportions of selected D3-‐PC species, from the PEMT pathway, up to 120 hours after recruitment. ... 85 Figure 33. Comparison of the proportions of individual PC species at 72 hours after recruitment. For each pair of columns the endogenous PC proportions are shown as the left-‐hand column, while the right-‐hand column is the PEMT derived D3-‐labelled species. Error bars represent one standard deviation. .... 86 Figure 34. Concentrations of D3-‐phosphatidylcholine phospholipids in preterm plasma increase with time from recruitment to 72 hours. The error bars show the median values and interquartile ranges. Comparison of data using Mann-‐ Whitney non-‐parametric significance testing shows significant increase between times 12 and 72 hours (**** p<0.0001) and between 12 and 120 hours (** p=0.0064) ... 88 Figure 35. Average concentrations of selected D3-‐phosphatidylcholine species in preterm plasma. The PEMT pathway is associated with PUFA PC species in adults but PC16:0/18:1 is the most abundant species produced by this pathway from 48 hours in this population. ... 89 Figure 36. Average concentrations of selected D3-‐phosphatidylcholine species in preterm plasma. The PEMT pathway is associated with PUFA PC species in adults but PC16:0/18:1 is the most abundant species produced by this pathway from 48 hours in this population and continues to be most abundant until the end of the study at 240 hours. ... 90

(14)

Figure 37. Fractional incorporation of D3-‐phosphatidylcholine phospholipids in preterm plasma. The box and whisker plots show the median values, interquartile ranges and outliers. ... 91 Figure 38. The mean fractional incorporation of D3-‐containing PC phospholipids in plasma over the first 120 hours after recruitment. D3 enrichment has been expressed as a percentage of all the PC phospholipids measured at each time point. The error bars represent a single standard deviation from the mean. .. 92 Figure 39. The mean fractional incorporation of selected D3-‐containing PC phospholipids in plasma over the first 120 hours after recruitment. Proportionally more PUFA species are synthesised by the PEMT pathway. D3 enrichment has been expressed as a percentage of all the PC phospholipids measured at each time point. ... 93 Figure 40. The mean fractional incorporation of selected D3-‐containing PC phospholipids in plasma over the first 240 hours after recruitment. Proportionally more PUFA species are synthesised by the PEMT pathway and this is maintained throughout the study period. D3 enrichment has been expressed as a percentage of all the PC phospholipids measured at each time point. ... 94 Figure 41. Enrichment by S-‐Adenosyl [methyl-‐D3]methionine in preterm plasma showing median values and interquartile range plus outliers ... 96 Figure 42. Fractional synthesis by the PEMT pathway in preterm infants to 240 hours after recruitment showing median values with interquartile range and outliers. ... 97 Figure 43. Synthetic rate of flux through the PEMT pathway in preterm infants to peak at 72 hours after infusion of [methyl-‐D9]choline. Overall rate of synthetic flux is 0.06%, significantly lower than adult volunteer values of 0.53% (Pynn et al., 2011) ... 98 Figure 44. Synthetic rate of flux for individual PC species through the PEMT pathway in preterm infants up to 24 hours after infusion of [methyl-‐ D9]choline. Synthetic flux varies from 0.325% for PC18:0/22:6 to 0.027% for the surface-‐active PC16:0/16:0 ... 99 Figure 45. Concentration of choline in plasma of preterm infants over first 24 hours of study. Administration of [methyl-‐D9]choline chloride at t=0

(15)

produced a small, but statistically not significant, rise in plasma choline at 6 hours. ... 102 Figure 46. Concentration of choline in plasma of individual preterm infants over first 24 hours of study. Individual babies are arranged in the legend by gestation from 23 weeks (blue) to 28 weeks (grey). There is no measured correlation between choline level and gestational age. ... 103 Figure 47. Enrichment of D9-‐choline in plasma. The proportion of choline measured at each time point enriched with nine deuterium atoms. ... 104 Figure 48. Enrichment of D9-‐choline in plasma plotted for each patient. Individual babies are arranged in the legend by gestation from 23 weeks (blue) to 28 weeks (grey). There is no measured correlation between D9-‐choline enrichment and gestational age. ... 105 Figure 49. Concentration of betaine in plasma of preterm infants over first 24 hours of study. Administration of [methyl-‐D9]choline chloride at t=0 produced no statistically significant rise in plasma betaine over 24 hours. .. 107 Figure 50. Enrichment of D9-‐choline and D9-‐betaine in plasma. ... 108 Figure 51. Peak enrichment of D9-‐choline and D9-‐betaine at 6 hours in plasma plotted against corresponding peak enrichment value of D9-‐ phosphatidylcholine at 12 hours. The trendlines show that there is a weak correlation between a lower peak D9-‐choline or D9-‐betaine enrichment and a higher peak D9-‐PC enrichment. ... 109 Figure 52. An example of a precursor of mass/charge +184 (P184) scan spectrum. It illustrates a typical endotracheal aspirate (ETA) lipid extract scan at recruitment (time point = 0) less than 48 hours after preterm birth. ... 115 Figure 53. Relative proportions of phosphatidylcholine species measured in baseline (t=0) endotracheal aspirate samples. PC16:0/16:0 constitutes the largest percentage of PC species measured. Error bar demonstrates one standard deviation. ... 117 Figure 54. Comparison of P+184 scans from ETA and Curosurf® showing similar species and proportions of peaks. ... 118 Figure 55. Relative proportions of phosphatidylcholine species measured in baseline (t=0) endotracheal aspirate samples (left) compared with measured proportions of PC species in Curosurf® (right). The proportions are very similar with PC16:0/16:0 constituting the largest percentage of PC species

(16)

measured in both sample types. Error bars demonstrate one standard deviation for the ETA samples. ... 119 Figure 56. Relative proportions of phosphatidylcholine species in ETA measured at each time point to 240 hours. The proportions of major PC species remain steady throughout the study period. Error bars demonstrate one standard deviation for the two most abundant species shown. ... 120 Figure 57. Log transformation graph showing relative proportions of phosphatidylcholine species measured in ETA at each time point to 240 hours. The proportions of major PC species remain steady throughout the study period but PC18:1/18:2 rises proportionally from 120 hours while PC16:0/14:0 falls in the same period. Error bars demonstrate one standard deviation for the two most abundant species shown. ... 121 Figure 58. Concentration of phosphatidylcholine measured in endotracheal aspirates over the 120 hours of study. There is no significant difference between measurements. ... 122 Figure 59. Concentration of phosphatidylcholine measured in endotracheal aspirates over the 240 hours of study. There is no significant difference between measurements. ... 123 Figure 60. Concentration of phosphatidylcholine measured in endotracheal aspirates over the 240 hours of study for the 10 children who remained intubated and ventilated at 120 hours and received a second dose of [methyl-‐ D9]choline. There is no significant difference between measurements in this subgroup. ... 124 Figure 61. Average concentration of individual PC species from preterm endotracheal aspirates to 120 hours after recruitment. ... 125 Figure 62. Average concentration of PC from preterm endotracheal aspirates up to 24 hours after recruitment and corrected for time from birth. Neither the time from birth of sample collection nor gestation affect concentration of PC collected from ETA. ... 126 Figure 63. Comparison of P+184 and P+193 scans from an ETA sample (at time 72 hours) showing incorporation of deuterated choline as the displacement of nine mass units on the P+193 scan. ... 128

(17)

Figure 64. Concentration of D9-‐PC phospholipids from preterm endotracheal aspirates to 120 hours after recruitment. The rise in concentration becomes statistically significant above background levels at 48 hours. ... 129 Figure 65. Concentration of D9-‐PC phospholipids from preterm endotracheal aspirates to 240 hours after recruitment. A statistically significant rise from baseline occurs to 120 hours but variation in concentration after 120 hours is not statistically significant. ... 130 Figure 66. Concentration of D9-‐PC phospholipids from preterm endotracheal aspirates to 240 hours after recruitment in the 10 children who received a second dose of choline at 120 hours. A statistically significant rise from baseline occurs to 120 hours but variation in concentration after 120 hours is not statistically significant. ... 131 Figure 67. Average concentration of individual D9-‐PC phospholipids from preterm endotracheal aspirates to 240 hours after recruitment. Significant differences between newly synthesised species demonstrated from 24 to 48 hours with concentration of D9PC16:0/16:0 rising most rapidly. The fall in concentration after 120 hours is not statistically significant and reflects smaller patient numbers. Importantly the relative abundance of species does not change after 120 hours. ... 132 Figure 68. Average concentration of individual D9-‐PC phospholipids from preterm endotracheal aspirates to 120 hours after recruitment. Log transformation of y-‐axis to allow comparison of species back to 6 hours. ... 133 Figure 69. Average enrichment of D9-‐PC phospholipids from preterm endotracheal aspirates to 240 hours after recruitment. ... 134 Figure 70. Relative proportions of selected PC species at 12 hours after recruitment. Columns on the left are endogenous (non-‐labelled) and columns on the right are the newly-‐synthesised (D9-‐labelled) PC species. ... 136 Figure 71. Relative proportions of selected PC species at 48 hours after recruitment. Columns on the left are endogenous (non-‐labelled) and columns on the right are the newly-‐synthesised (D9-‐labelled) PC species. ... 137 Figure 72. Relative proportions of selected PC species at 120 hours after recruitment. Columns on the left are endogenous (non-‐labelled) and columns on the right are the newly-‐synthesised (D9-‐labelled) PC species. ... 138

(18)

Figure 73. Fractional incorporation of D9-‐label into individual PC species over 240 hours after recruitment to study. ... 139 Figure 74. Fractional incorporation of D9-‐label into individual PC species over the first 120 hours after recruitment to study. ... 140 Figure 75. Relative proportions of newly synthesised PC species in endotracheal aspirates from preterm infants over 240 hours after recruitment to the study. ... 141 Figure 76. A mass spectrum negative ionisation scan (MS-‐) of Curosurf® illustrating the major acidic phospholipid species. Three major phosphatidylglycerol (PG) species and three major phosphatidylinositol (PI) species are labelled. Of the acidic phospholipids, PG species are predominant in Curosurf® preparations. ... 143 Figure 77. Relative proportions of phosphatidylglycerol species and phosphatidylinositol species change from a mature to an immature ratio with time from administration of Curosurf®. This provides very early evidence of active metabolism of Curosurf® from the ET aspirates of preterm infants despite relatively slow phosphatidylcholine metabolism. ... 144

(19)

(20)

List of Tables

Table 1. The routine samples collected in TSuNaMI Study. ... 35 Table 2. Table of the liquid chromatography gradients used to study choline and metabolites ... 42 Table 3. Table of the multiple reaction monitoring transitions used to study choline and metabolites ... 43 Table 4. Table of the phosphatidylcholine species found in preterm plasma. Ordered by mass charge ratio and colour coded according to fatty acid on sn-‐2 position. ... 57 Table 5. Table of the phosphatidylcholine species found in preterm endotracheal aspirate. ... 116 Table 6. Table of the total number of patient endotracheal samples available at each time point. ... 124

(21)

(22)

Declaration of Authorship

I, Kevin Colin William Goss, declare that the thesis entitled

DYNAMIC SURFACTANT METABOLISM IN PRETERM INFANTS

and the work presented in the thesis are both my own, and have been generated by me as the result of my own original research. I confirm that:

§ this work was done wholly or mainly while in candidature for a research degree at this University;

§ where any part of this thesis has previously been submitted for a degree or any other qualification at this University or any other institution, this has been clearly stated;

§ where I have consulted the published work of others, this is always clearly attributed;

§ where I have quoted from the work of others, the source is always given. With the exception of such quotations, this thesis is entirely my own work;

§ I have acknowledged all main sources of help;

§ where the thesis is based on work done by myself jointly with others, I have made clear exactly what was done by others and what I have contributed myself; § parts of this work have been published as two abstracts:

GOSS, K. C. W., GOSS, V. M., TOWNSEND, J. P., GUNDA, R., KOSTER, G., CLARK, H. W. & POSTLE, A. D. 2012. Characterising the Metabolism of Therapeutic Exogenous Surfactant in Preterm Infants Using a Stable Isotope-‐Labelled Substrate.

Neonatology, 101, 364-‐365.

GOSS, K. C. W., GOSS, V.M., TOWNSEND, J.P. GUNDA, R., RHODES-‐KITSON, J.,

KOSTER, G., HALL, M., THWAITES, R., PAPPACHAN, J.V., CLARK, H.W., POSTLE, A.D. 2012. Dynamic lipidomic mass spectrometry of endotracheal secretions and plasma from preterm infants mechanically ventilated on the neonatal intensive care unit. Am J Respir Crit Care Med, 185:A1820

Signed: ……… Kevin C.W. Goss

Date:………..28th_{January
2013}

(23)

(24)

Acknowledgements

Firstly I would like to thank my supervisors Professor Tony Postle and Professor Howard Clark for their invaluable help, advice, encouragement, expertise and support throughout the whole TSuNaMI research project.

Thanks also to the rest of the surfactant protein research group, to Dr Jens Madsen and Dr Rosie Mackay, for their help, advice and, especially, the cakes at the lab meetings each week.

Many thanks go to Paul Townsend, Dr Grielof Koster and the rest of the BRU Mass Spectrometry Laboratory office team, for sharing their skills in the laboratory and for their encouragement.

To the rest of the senior clinical researchers with whom I have worked closely including Professor Mike Grocott, Dr John Pappachan, Dr Michael Marsh and Professor Ratko Djukanovic, thank you all for your support. Also to Dr Gary Connett, for inadvertently setting me on this journey.

Thanks to Joe Maskell, for his patience and IT skills.

To the BRU research nurses, Jane Rhodes-‐Kitson and Sibo Hakata, and to Dr Ranjit Gunda, thank you for all your patience and hard work.

To the dedicated nurses of the neonatal units who collected, labelled and refrigerated hundreds of samples for me, thank you so much.

To my patients and their parents, I will never forget the trust you have placed in this research, thank you for agreeing to participate.

To all my friends, especially Colin and Lynsey, Woolf and Sarah, Will and Becks, thank you so much for keeping me going and keeping me sane.

To my family, grandparents, parents and siblings… Sinéad, Andy and Cillian, Brendan and Georgina, Mum and Dad, for being there for me through everything. Not forgetting the four-‐legged family members Archie and Molly.

And finally, that just leaves…

My senior postdoc, my closest collaborator and my fellow researcher who has been with me for every step of this journey and who also just happens to be my

beautiful wife, Dr Victoria Goss; for everything, I thank you and I love you.

(25)

(26)

List of Abbreviations

ARDS Acute respiratory distress syndrome

AT-‐I Alveolar type 1 cell

AT-‐II Alveolar type 2 cell

ATP Adenosine triphosphate

BADH Betaine aldehyde dehydrogenase

BHMT Betaine-‐homocysteine S-‐methyltransferase

BHT Butylated hydroxytoulene

BPD Bronchopulmonary dysplasia

BRU Biomedical research unit

CDH Choline dehydrogenase

CDP-‐choline Cytidine diphosphate choline

CK Choline kinase

CPAP Continuous positive airways pressure

CRD Carbohydrate recognition domain

CRF Case report form

CPT Cholinephosphotransferase

CT Choline phosphate cytidylyltransferase

D Deuterium

DPPC Dipalmitoylphosphatidylcholine

DSPC Disaturated phosphatidylcholine

ECMO Extracorporeal membranous oxygenation

EDTA Ethylenediaminetetraacetic acid

ESI MS/MS Electrospray ionisation tandem mass spectrometry

ET(A) Endotracheal (aspirate)

FiO2 Fraction of inspired oxygen

GC-‐C-‐IRMS Gas chromatography combustion isotope ratio mass

spectrometry

LB Lamellar body

LC-‐MRM-‐MS/MS Liquid chromatography multiple reaction monitoring

tandem mass spectrometry

m/z Mass-‐charge ratio

(27)

nCLD Neonatal chronic lung disease

NG Nasogastric

NICU Neonatal intensive care unit

NIHR National institute for health research

NNU Neonatal unit

P184 Precursor scan of mass-‐charge ratio 184

PAH Princess Anne Hospital, Southampton

PC Phosphatidylcholine

PDA Patent ductus arteriosus

PE Phosphatidylethanolamine

PEMT Phosphatidylethanolamine-‐N-‐methyltransferase

PG Phosphatidylglycerol

PI Phosphatidylinositol

PUFA Polyunsaturated fatty acid

RDS Respiratory distress syndrome

SAMe S-‐Adenosyl methionine

SP-‐A Surfactant protein type A

SP-‐B Surfactant protein type B

SP-‐C Surfactant protein type C

SP-‐D Surfactant protein type D

TPN Total Parenteral Nutrition

TSuNaMI The surfactant, nutrition and microorganisms

interaction study in babies born at risk of developing neonatal chronic lung disease

UHS University Hospital Southampton

UoS University of Southampton

VLDL Very low density lipoprotein

(28)

Chapter 1. Introduction

1.1 Overview

There are many reasons why a newborn infant may need to be treated in a neonatal intensive care unit (NICU) but one of the more frequently encountered and complex is prematurity. The last 20 years have seen significant improvements in the care of the premature infant, with survival at 28 weeks gestation now seen as almost routine. However, while children born at the limits of viability have improved survival chances they are still at significant risk of developing long-‐term problems directly as a result of being born prematurely (Costeloe et al., 2012). Survival of babies born at less than 23 weeks gestation is very rarely seen and then long term severe neurodevelopmental disability is near universal. Babies delivered at 23 weeks still have severely limited chances of survival and most have severe respiratory and neurodevelopmental problems at discharge home (Costeloe et al., 2012). The chances of survival increase from 24 weeks until about 30-‐32 weeks gestation when most babies are expected to survive with few long-‐term problems. A determinant for the development of long-‐term problems in preterm survivors is their clinical course in the first hours and days of life. The introduction of exogenous surfactant has dramatically reduced the severity of the initial respiratory distress symptoms experienced by preterm newborn children (Schwartz et al., 1994) and now it is routinely used as prophylaxis in any baby born in the UK at less than 29 weeks gestation. However, little is known about how this exogenous surfactant affects the infants own surfactant production and metabolism in the clinical setting.

(29)

Another potential area for intervention in the first days of preterm life has sought an improvement in the nutrition for preterm infants. Recent studies in the UK and abroad have shown that, even with well-‐recognised recommendations for nutritional intake, many preterm infants show severe growth restriction and stunting by the time of discharge from the NICU (Wood et al., 2003, Costeloe et al., 2000). As these children fail to meet their macronutrient targets it is likely that they also fail to meet minimum micronutrient recommendations.

Of particular interest in Southampton is the role of phospholipid metabolism in health and disease (Todd et al., 2009, Burdge et al., 1994, Ashton et al., 1992). Much work has looked at the metabolism of phosphatidylcholine (PC) phospholipids in both human and animal studies. PC is a fundamentally important phospholipid building block for lipid membranes, both in cell membranes and monolayers. The disaturated dipalmitoylphosphatidylcholine (DPPC) is the predominant surface-‐active phospholipid in lung surfactant – the complex surface tension lowering substance that is known to be deficient in the preterm infant lung (Clements, 1997). The building blocks for PC are derived from nutritional sources, namely choline and fatty acid intake. Choline is an essential nutrient (Zeisel and da Costa, 2009) and while preterm infants will receive some during their stay in NICU their nutritional requirement has not been firmly established. Choline is metabolised by two distinct pathways to produce PC. The first pathway, known as the CDP-‐choline pathway, is in most cells in the body, including the lungs (Vance and Vance, 1985), so if it were possible to measure PC in the preterm infant lung it would be possible to describe its metabolism, not only in terms of the exogenous surfactant that all of these babies receive, but also in terms of a baby’s ability to synthesise new PC molecules thereby acting a marker for lung maturity.

(30)

The second pathway, known as the PEMT pathway is specific to hepatocytes and acts through methyl transfer to produce PC from another phospholipid, phosphatidylethanolamine (PE) (Vance et al., 1997). Investigation and monitoring of this system would provide further information on how these babies handle choline from nutritional sources and may provide information on how the methyl transfer system in general is working. Methylation is an essential component of epigenetic gene expression and it’s function in early life and development is theorised to impact directly on future health (Menon et al., 2012, Barnes and Ozanne, 2011, Thornburg et al., 2010, Burdge and Lillycrop, 2010, Mathers and McKay, 2009).

In summary, a better understanding of choline metabolism in these vulnerable and under researched infants will provide important insights to both nutritional and respiratory metabolic function and may offer biomarkers for future disease and inform future monitoring of disease progression in a clinical setting. A better understanding of surfactant turnover and of nutritional metabolism can be obtained through the investigation of choline metabolism, which in turn can lead to improved treatment for individual infants on the NICU and ultimately better outcomes in the short and long term.

To clarify the subsequent steps taken in this research project it is necessary to review the background to the investigation of surfactant in this age group and to establish the development of normal in utero nutrition.

1.2 History of pulmonary surfactant research

“Surfactant” is a contraction of surface-‐active agent and, generically, surfactants are defined as substances that spread over the surface of a liquid and, by adsorbing at the liquid-‐gas interface, lower the surface tension.

(31)

The alveolar-‐capillary unit is the site of gas exchange in the lung and to reduce surface tension in the alveolus and small airways the mature mammalian lung produces pulmonary surfactant, a lipoprotein complex formed by the pulmonary alveolar type II (AT-‐II) cells. In human embryological development, full term gestation is 40 weeks with initial production of pulmonary surfactant occurring between 24 and 34 weeks gestation. When combined with immature lung microanatomy, absence of this complex surface-‐active compound means that babies born before 24 weeks rarely survive (Costeloe et al., 2012, Allen et al., 1993). In the last two decades major advances in the provision of improved neonatal intensive care, including the replacement of pulmonary surfactant with exogenous material (Merenstein et al., 1991), have resulted in a marked improvement in the outcome and survival of the extreme preterm infants born between 24 and 28 weeks gestation (Goldenberg et al., 2006).

The theory of surface forces had been elaborated at the end of the 19th_{century
by}

Lord Rayleigh, who determined molecular size by preparing monomolecular films (Rayleigh, 1890) and then obtained the Nobel prize in 1904 for this work. In their comprehensive reviews of the early years of surfactant research both Halliday (Halliday, 2008) and Wrobel (Wrobel and Clements, 2004) cite the work of Von Neergaard (Von Neergaard, 1929) in the 1920’s as the starting point in the story of the development of surfactant as a treatment for respiratory distress syndrome (RDS). Only in the 1950’s was this pioneering work revisited when Clements, using a homemade dynamic tensiometer measured the surface tension of films from mammalian lung preparations (Clements, 1957). In his famous paper, originally published in 1957, Clements confirmed that surface tension is low in the tissue extracts containing the lining of the lung. Building on this work, he discovered that

Dynamic surfactant metabolism in preterm infants

University of Southampton Research Repository

ePrints Soton

UNIVERSITY OF SOUTHAMPTON

Dynamic surfactant metabolism in

preterm infants

Kevin Colin William Goss, MB ChB MRCPCH

A thesis submitted to the

University of Southampton for the degree of

Doctor of Philosophy

Faculty of Medicine

University of Southampton

December 2012

University of Southampton

ABSTRACT

Faculty of Medicine

Doctor of Philosophy

DYNAMIC SURFACTANT METABOLISM IN PRETERM

INFANTS

by Kevin Colin William Goss

Table of Contents

List of Figures

List of Tables

Declaration of Authorship

Acknowledgements

List of Abbreviations

Chapter 1. Introduction

1.1 Overview

1.2 History of pulmonary surfactant research