Investigation of Acute and Chronic Effects of Statins on Vascular Function in Hypertensive Disorders of Pregnancy, with a Major Focus on Pre-Eclampsia

Investigation of Acute and Chronic Effects of Statins on Vascular Function in Hypertensive Disorders of Pregnancy, with a Major Focus on Pre-eclampsia.

A thesis submitted to the University of Manchester for the degree of Doctor of Philosophy in the Faculty of Biology, Medicine and Health

2019

Chinedu N. Agwu

School of Medical Sciences

List of Contents List of Figures ...... 12 List of Abbreviations ...... 16 Abstract ...... 18 Declaration ...... 20 Copyright Statement ...... 21 Presentations and Publications ...... 22 Acknowledgements ...... 23 Chapter 1: General Introduction ...... 24 1.0 Overview ...... 25 1.1 Hypertensive Disorders of Pregnancy...... 28 1.1.1 Definition ...... 28 1.1.2 Identification of High Risk Pregnancies: PE ...... 30 1.1.2.1 Diagnosis of Chronic Hypertension in Pregnancy ...... 30 1.1.2.2 Diagnosis of Fetal Growth Restriction ...... 31 1.1.3 Clinical Investigations ...... 32 1.1.4 Clinical Management of PE ...... 33 1.1.4.1 Antihypertensive Drugs ...... 33 1.1.4.2 Anticonvulsants ...... 34 1.1.4.3 Low Dose Aspirin ...... 35 1.2 Vascular Changes and Placental Development in Normal Pregnancy and Pregnancy Complicated by PE ...... 35 1.2.1 Vascular Changes in Normal Pregnancy ...... 35 1.2.2 Uteroplacental Vascular changes in pregnancies complicated be PE ...... 36 1.3 Cardiovascular adaptations in normal pregnancy and in PE ...... 37 1.4 Importance of the Placenta in Normal Pregnancy and Pregnancies Complicated by PE ...... 38 1.5 The endothelium and resistance vessels ...... 42 1.5.1. Resistance vessels ...... 42 1.5.2 Function of endothelium ...... 42 1.5.3 Endothelial-dependent and independent vasoconstrictors ...... 43 1.5.3.1 Phenylephrine ...... 43 1.5.3.2 Endothelin...... 43 1.5.3.3 Thromboxane ...... 44 1.5.3.4 Vasopressin ...... 44

1.5.3.5 Noradrenaline ...... 44 1.5.3.6 Serotonin ...... 44 1.5.3.7 Angiotensin II ...... 45 1.5.4 Endothelial-dependent and independent vasodilators ...... 45 1.5.4.1 Nitric oxide (NO) ...... 45 1.5.4.2 Prostacyclin ...... 45 1.5.4.3 Sodium nitroprusside ...... 46 1.5.4.4 Acetylcholine ...... 46 1.5.4.5 Bradykinin ...... 46 1.5.4.6 Endothelium-Derived Hyperpolarising Factor (EDHF) ...... 46 1.5.4.7 Histamine ...... 47 1.5.4.8 Substance P ...... 47 1.5.5 Methods of Measuring Vascular Reactivity ...... 48 1.5.5.1 Ex-vivo: Wire and Pressure myography ...... 48 1.6 Vascular Dysfunction in PE ...... 50 1.6.1 Altered Vascular Reactivity of Maternal and Fetal Vessels in Human PE ...... 50 1.6.2 Role of NO in normal pregnancy and in PE ...... 52 1.7 Animal Models of Pregnancy ...... 55 1.7.1 Animal Models of Pre-eclampsia ...... 57 1.7.2 Mouse vs human ...... 58 1.7.3 eNOS knockout mice ...... 60 1.8 Clinical Therapeutics for Pre-eclampsia ...... 61 1.8.1 Proposed Treatments for Pre-eclampsia ...... 61 1.8.1.1 Pre-Clinical Studies ...... 61 1.8.1.2 Clinical Trials for the treatment of PE ...... 62 1.8.1.3 Clinical Trials for Prevention of PE ...... 63 1.8.2 Statins and Known Mechanisms of Action ...... 63 1.8.3 Cholesterol levels in normotensive and PE pregnancies ...... 66 1.8.4 Pleiotropic Effects of Statins...... 67 1.10 Statins and Their Use in Pregnancy ...... 69 1.10.1 Risks of Statin Use in Pregnancy ...... 69 1.10.1.1 Pre-clinical Studies in Animals and Humans ...... 69 1.10.1.2 Clinical Studies ...... 70 1.10.2 Placental Transport of Statins...... 72

1.11 Biological Plausibility for Statins Use in Pre-eclampsia ...... 72 1.11.1 Evidence from Pre-clinical Studies ...... 72 1.11.2 Evidence of the use of statins in pregnancy from clinical studies ...... 76 1.12 Summary ...... 78 1.13 Hypotheses ...... 78 1.13.1 Human ...... 78 1.13.2 Mouse ...... 79 1.14 Aims and objectives ...... 79 1.14.1 Human vessels ...... 79 1.14.2. Mouse vessels ...... 79 Chapter 2: Materials and Methods ...... 80 2.0 Studies in human: ...... 81 2.1 Source of Chemicals ...... 81 2.2 Ethical approval ...... 81 2.3 Collection of blood samples and tissue sample preparation ...... 82 2.4. Multi Myograph System 620M ...... 83 2.5 Vasoconstrictors and Vasodilators ...... 84 2.5.1 Vasoconstrictive agents ...... 84 2.5.1.1 Vasoconstrictors...... 84 2.5.2 Vasodilatory agents ...... 85 2.5.2.1 Endothelium-dependent: Bradykinin ...... 85 2.5.2.2 Endothelium-dependent: Acetylcholine...... 85 2.5.2.3 Endothelium-independent: SNP ...... 85 2.6 Myography ...... 86 2.6.1. Choice of Statin and Dose for Experiment ...... 86 2.6.2 Chorionic Plate Artery (CPA) Protocol ...... 87 2.6.3 Omental Artery (OA) Protocol ...... 89 2.7 Data Analysis ...... 91 2.8 Studies in Mouse ...... 91 2.8.1 Ethical Considerations and Animal Husbandry ...... 91 2.9 Experiment allocation ...... 93 2.10 Sample Collection for Ex Vivo Study ...... 93 2.11 Blood vessel dissection and normalisation ...... 94 2.11.1 Maternal Uterine Artery ...... 94

2.11.2 Umbilical Artery ...... 94 2.11.3 Myography ...... 94 2.12 In Vivo Study ...... 95 2.13 Sample Collection for In Vivo Study ...... 97 2.14 Assessment of progeny at E18.5 ...... 98 2.14.1 Assessment of fetal weight, placental weight and fetal biometric measurements ...... 98 2.15 Fetal Weight Distribution Curve ...... 98 2.16 Blood vessel dissection and normalisation ...... 99 2.16.1 Wire myography ...... 99 2.16.1.1 Protocol ...... 99 2.17 Data analysis ...... 100 Chapter 3: The acute effect of statins on vascular function of chorionic plate arteries in PE and normal pregnancies ...... 101 3.0 Results ...... 102 3.1 Introduction ...... 102 3.2 Hypothesis and aims ...... 104 3.2.1 Hypothesis ...... 104 3.2.2 Aims ...... 105 3.3 Materials and methods ...... 105 3.3.1 Inclusion and Exclusion Criteria ...... 105 3.3.2. Sample collection and protocol ...... 106 3.4 Results ...... 111 3.4.1 Demographics of Women in Study ...... 111 3.4.2 Diameter of chorionic plate arteries from normal and PE pregnancies prior to statin exposure ...... 113 3.4.3 Assessment of vascular reactivity of chorionic plate arteries between normal and PE pregnancies ...... 114 3.4.4 Diameter of chorionic plate arteries from normal pregnancies prior to statin (1µM) exposure ...... 116 3.4.5 Diameter of chorionic plate arteries from normal pregnancies prior to statin (10 µM) exposure ...... 117 3.4.6 Effect of statins (1 µM) on basal tone of chorionic plate arteries from normal pregnancies118 3.4.7 Effect of statins (10 µM) on basal tone of chorionic plate arteries from normal pregnancies ...... 119 3.4.8 Effect of statins (1 µM) on contraction of chorionic plate arteries from normal pregnancies ...... 120

3.4.9 Effect of pravastatin and pitavastatin (10 µM) on contraction of chorionic plate arteries from normal pregnancies ...... 121 3.4.10 Effect of statins (1 µM) on relaxation of chorionic plate arteries from normal pregnancies following a 2 hr incubation ...... 122 3.4.11 Effect of pravastatin and pitavastatin (10 µM) on relaxation of chorionic plate arteries from normal pregnancies following a 2 hr incubation ...... 123 3.4.12 Diameter of chorionic plate arteries from PE pregnancies prior to statin exposure (1µM)124 3.4.13 Diameter of chorionic plate arteries, prior to statin exposure (10µM), from PE pregnancies ...... 124 3.4.14 Effect of statins (1 µM) on basal tone of chorionic plate arteries from pregnancies with PE ...... 125 3.4.15 Effect of pitavastatin (10 µM) on basal tone of chorionic plate arteries from pregnancies with PE ...... 126 3.4.16 Effect of statins (1 µM) on contraction of chorionic plate arteries from pregnancies with PE ...... 127 3.4.17 Effect of pitavastatin (10 µM) on contraction of chorionic plate arteries from pregnancies with PE ...... 128 3.4.18 Effect of statins (1 µM) on relaxation of chorionic plate arteries from pregnancies with PE following a 2 hr incubation ...... 129 3.4.19 Effect of pitavastatin (10 µM) on relaxation of chorionic plate arteries from pregnancies with PE following 2 hr incubation ...... 130 3.5 Discussion ...... 132 3.5.1 Myography studies of Human Chorionic Plate Artery Function ...... 133 3.5.2 Vascular Reactivity of Chorionic Plate arteries from Normal and PE Pregnancies ...... 133 3.5.2.1 Potassium- and U46619- induced contraction of chorionic plate arteries from normal and PE pregnancies ...... 133 3.5.2.2 Sodium nitroprusside-induced (endothelial-independent) relaxation of chorionic plate arteries from normal and PE pregnancies...... 136 3.5.3 Effect of short-term statin treatment on CPAs from normal pregnancies...... 139 3.5.3.1 Potassium- and U46619- induced contraction of chorionic plate arteries from normal pregnancies ...... 139 3.5.3.2 Sodium nitroprusside-induced independent relaxation of chorionic plate arteries from normal pregnancies ...... 140 3.5.4 Effect of short-term statin treatment on CPAs from PE pregnancies ...... 141 3.5.4.1 Potassium- and U46619- induced contraction of chorionic plate arteries from PE pregnancies ...... 141 3.5.4.2 Sodium nitroprusside-induced independent relaxation of chorionic plate arteries from PE pregnancies ...... 142

3.5.5 Conclusion/clinical implications ...... 144 Chapter 4: The acute effect of statins on maternal omental artery function in normal, pre-eclamptic, chronic hypertensive and super-imposed pre-eclamptic pregnancies ...... 145 4.0 Results ...... 146 4.1 Introduction ...... 146 4.2 Hypothesis and aims ...... 148 4.2.1 Hypothesis ...... 148 4.2.2 Aims ...... 148 4.3 Materials and methods ...... 149 4.3.1 Clinical Demographics ...... 149 4.3.2. Sample collection and protocol ...... 149 4.4 Results ...... 154 4.4.1 Demographics of women whose samples were used in the study...... 154 4.4.2. Inclusion and Exclusion of Samples ...... 156 4.4.3 Diameter of omental arteries from normal, PE, CHT and superimposed PE pregnancies prior to statin exposure ...... 159 4.4.4 Assessment of vascular reactivity of omental arteries between normal, PE, CHT and superimposed PE pregnancies ...... 160 4.4.5. Effect of pravastatin and pitavastatin on basal tone of omental arteries from normal pregnancies at a concentration of 1 µM following 2 hr incubation...... 162 4.4.6 Effect of pravastatin and pitavastatin on contraction of omental arteries from normal pregnancies at a concentration of 1µM following 2 hr incubation ...... 163 4.4.7 Effect of pravastatin and pitavastatin on relaxation of omental arteries from normal pregnancies at a concentration of 1 µM following 2 hr incubation ...... 166 4.4.8. Effect of pitavastatin (1 µM) on basal tone of omental arteries from PE, CHT and superimposed PE pregnancies following 2 hr incubation ...... 167 4.4.9 Effect of pitavastatin (1 µM) on contraction of omental arteries from PE, CHT and superimposed PE pregnancies following 2 hr incubation ...... 169 4.4.10 Effect of pitavastatin (1µM) on relaxation of omental arteries from PE, CHT and superimposed PE pregnancies following 2 hr incubation...... 172 4.5 Discussion ...... 174 4.5.1 Summary of Main Findings ...... 174 4.5.2 Methodological Challenges and Implications ...... 175 4.5.2.1 Rationale for the use of omental arteries (OAs) ...... 175 4.5.2.2 Desensitization to Thromboxane Agonist ...... 175 4.5.2.3. Effect of DMSO on OA Vascular Reactivity ...... 176

4.5.2.4 Self-Collapsing vessels ...... 177 4.5.3 Vascular Reactivity of Omental Arteries from Normal Pregnancies and Hypertensive Pregnancies ...... 177 4.5.3.1 Potassium- and U46619- induced contraction of omental arteries from normal and hypertensive pregnancies ...... 177 4.5.3.2 Bradykinin-induced relaxation of omental arteries from normal and hypertensive pregnancies ...... 180 4.5.4 Effect of Short-Term Statin Treatment on Omental Arteries from Normal Pregnancies and Pathological Pregnancies ...... 183 4.5.4.1 Potassium- and U46619- induced contraction of omental arteries from normal and pathological pregnancies ...... 183 4.5.4.2 Endothelium-dependent bradykinin- and endothelium-independent SNP-induced relaxation of omental arteries from normal and pathological pregnancies ...... 185 4.5.4.3 Limitations of Study...... 186 4.5.5 Conclusion ...... 187 Chapter 5: Acute and chronic effects of pitavastatin on maternal and fetal vascular function in the endothelial nitric oxide synthase knockout (eNOS-/-) mouse ...... 188 5.0 Results ...... 189 5.1 Introduction ...... 189 5.2 Hypothesis and aims ...... 191 5.2.1 Hypotheses ...... 191 5.2.2 Aims ...... 191 5.3 Materials and Methods ...... 192 5.3.1 Acute (2h) incubation of vessels with pitavastatin (ex vivo) ...... 192 5.3.1.1 Sample Collection and Protocol ...... 192 5.3.2 Blood vessel Dissection and Normalisation ...... 192 5.3.2.1 Maternal Uterine Artery ...... 192 5.3.2.2 Fetal Umbilical Artery ...... 192 5.3.3 Wire Myography ...... 193 5.3.4 In vivo Exposure to Pitavastatin in WT and eNOS-/- mice ...... 194 5.3.4.1 Experimental Allocation of Mice for In Vivo Studies ...... 194 5.3.4.2 Experimental Flowchart Illustrating Use of Animals in In Vivo Dosing Study ...... 196 5.3.5 Sample Collection and Protocol ...... 196 5.3.6 Fetal and Placental Measurements at E18.5 ...... 197 5.3.6.1 Assessment of Fetal weight, Placental weight and Fetal Biometric Measurements ..... 197 5.3.7 Blood Vessel Dissection and Normalisation ...... 197

5.3.7.1 Maternal Mesenteric Arteries ...... 197 5.3.8 Wire Myography ...... 197 5.3.9 Data Analysis ...... 198 5.4 Results: Study focused upon Acute Exposure to Vessels with 1 µM Pitavastatin (ex vivo) ...... 199 5.4.1 Maternal weight in WT and eNOS-/- mice at embryonic day 18.5 ...... 199 5.4.2 Fetal and Placental Weights in WT and eNOS-/- Mice at Embryonic Day 18.5 (ex vivo study) ...... 200 5.4.3 Diameter of uterine arteries from WT and eNOS-/- mice at embryonic day 18.5 prior to ex vivo statin exposure ...... 202 5.4.4 Effect of pitavastatin (1 µM) on basal tone of uterine arteries in WT and eNOS-/- mice .. 203 5.4.5 Effect of pitavastatin (1 µM) on contraction responses of uterine arteries in WT and eNOS-/- mice ...... 204 5.4.6 Effect of pitavastatin (1 µM) on relaxation responses of uterine arteries in WT and eNOS-/- mice ...... 206 5.4.7 Diameter of umbilical arteries from WT and eNOS-/- mice at embryonic day 18.5 prior to statin exposure ...... 208 5.4.8 Effect of pitavastatin (1 µM) on basal tone of umbilical arteries in WT and eNOS-/- mice 209 5.4.9 Effect of pitavastatin (1 µM) on contraction responses of the umbilical artery in WT and eNOS- /- mice ...... 210 5.4.10 Effect of pitavastatin (1 µM) on relaxation responses of umbilical arteries in WT and eNOS-/- mice ...... 212 5.4.11 Results: In vivo exposure to pitavastatin in WT and eNOS-/- mice ...... 214 5.4.12 Effect of maternal pitavastatin treatment on fluid intake and maternal bodyweight in WT and eNOS-/- mice ...... 214 5.4.13 Effect of maternal pitavastatin treatment on maternal organ weight in WT and eNOS-/- mice ...... 216 5.4.14 Effect of maternal pitavastatin treatment on fetal and placental weights in WT and eNOS knockout mice at embryonic day 18.5...... 218 5.4.15 Effect of maternal pitavastatin treatment on fetal biometric measurements in WT and eNOS- /- mice ...... 220 Fetal Biometric Measurements ...... 220 5.4.16 Effect of maternal pitavastatin treatment on uterine artery diameter from WT and eNOS-/- mice ...... 221 5.4.17 Effect of maternal pitavastatin treatment on contraction responses of uterine artery . 222 5.4.18 Effect of maternal pitavastatin treatment on relaxation responses of uterine artery ... 224 5.4.19 Effect of maternal pitavastatin treatment on mesenteric artery diameter from WT and eNOS- /- mice ...... 226

5.4.20 Effect of maternal pitavastatin treatment on contraction responses of maternal mesenteric artery ...... 227 5.4.21 Effect of maternal pitavastatin treatment on relaxation responses of maternal mesenteric artery ...... 229 5.4.22 Effect of maternal pitavastatin treatment on diameter of umbilical arteries from WT and eNOS knockout mice at embryonic day 18.5 ...... 231 5.4.23 Effect of maternal pitavastatin treatment on contraction responses of umbilical artery232 5.4.24 Effect of maternal pitavastatin treatment on relaxation responses of umbilical artery 234 5.5 Discussion ...... 236 5.5.1 Acute Pitavastatin Incubation Studies in eNOS-/- and WT Mice (ex vivo) ...... 236 5.5.2 In vivo Pitavastatin Treatment Studies in eNOS-/- and WT mice ...... 236 5.5.3 Confirmation of the Pregnancy Phenotype Associated with the eNOS-/- Mouse ...... 236 5.5.4 Rationale Underpinning Myography Studies of Mouse Uterine and Umbilical Artery Function ...... 238 5.5.5 Vascular Reactivity of Uterine and Umbilical Arteries Following 2 hr Pitavastatin Exposure in WT and eNOS-/- Mice ...... 238 5.5.6 Effect of Maternal Pitavastatin Treatment in vivo on Maternal and fetal Measurements in eNOS-/- and WT Mice ...... 242 5.5.7 Summary ...... 249 Chapter 6: General Discussion ...... 251 6.0 Overview ...... 252 6.1 Key Findings...... 252 6.1.1 Effect of Statin Treatment on Maternal Vessels ...... 253 6.1.2. Effect of Statin Treatment on Fetoplacental Vessels ...... 259 6.2 Application of Statins in Pregnancy and Safety Considerations ...... 261 6.3 Methodological considerations and limitations of the study ...... 264 6.4 Clinical perspective ...... 265 6.5 Conclusions ...... 266 6.6 Future work ...... 267 Chapter 7: References ...... 269 Chapter 8: Appendix ...... 302 8.1 Vascular reactivity of Uterine arteries ...... 303 8.1.1 Contraction responses of uterine artery (including water control) ...... 303 8.1.2 Relaxation responses of uterine artery (including water control) ...... 304 8.2 Vascular reactivity of mesenteric artery ...... 305

8.2.1 Contraction responses of mesenteric arteries (including water control) ...... 305 8.2.2 Relaxation responses of mesenteric artery (including water control) ...... 306 8.3 Vascular reactivity of umbilical artery ...... 307 8.3.1 Contraction responses of umbilical arteries (including water control) ...... 307 8.3.2. Relaxation responses of umbilical artery (including water control) ...... 308

Word Count: 68, 943

List of Figures Figure 1.1. Diagram outlining the pathology of PE from the pre-clinical to the clinical stage, also including the subsequent consequences of it...... 27 Figure 1.2: a) shows a normal Doppler waveform b) shows an abnormal Doppler waveform with uterine artery notching...... 31 Figure 1.3: Angiogenic imbalance due to sFlt-1, a potent circulating antagonist to PlGF, resulting in placental physiopathology in PE...... 33 Figure.1.4: Diagram outlining normal and abnormal SpA remodelling...... 37 Figure 1.5: Maternal and fetal surface of the placenta...... 40 Figure 1.6: A transverse view of a full term placenta...... 40 Figure 1.7: Arterial schematic...... 43 Figure 1.8: Illustration of vasoactivators that can cause vasodilation and vasoconstriction ...... 48 Figure 1.9: Diagram depicting vessel arrangement in a myograph bath ...... 49 Figure 1.10: Compares the placenta barrier between different animal models of pregnancy ...... 56 Figure 1.11: An illustration of how statins interact with the Mevalonate pathway...... 65 Figure 1.12: Outlines the pleiotropic effects of both lipid-dependent and lipid-independent statins...... 69 Figure 2.1: Detailed diagram of Multi Myograph System 620M ...... 84 Figure 2.2: Representation of LaPlace’s Equation ...... 87

Figure 2.3: Where EC80 is the concentration of agonist (in this case U46619) that gives an 80% response; 10-5 is used to convert to µM and 6000 is the volume of PSS in the myograph chamber (in µl)...... 88 Figure 2.4: A representative image of full myography protocol for CPAs ...... 89 Figure 2.5: A representative trace of the wire myography protocol for omental arteries ...... 90 Figure 2.6: Experimental allocation of mice for ex-vivo and in-vivo studies ...... 93 Figure 2.7: A representative myography trace for ex vivo mouse experiments...... 95 Figure 2.8: In-vivo study design...... 97 Figure 2.9: Images of fetal biometric measurements ...... 98 Figure 2.10: A representative image of protocol for in vivo mouse experiments...... 100 Figure 3.1: Flowchart of CPA samples used and exclusions ...... 109 Figure 3.2: A representative image of the full myography protocol for CPAs ...... 110 Figure 3.3: Diameter of CPAs from PE pregnancies and normal pregnancies prior to statin incubation...... 113 Figure 3.4: Assessment of vascular reactivity in CPAs from normal and PE pregnancies...... 115 Figure 3.5: Diameter of CPAs between statin and controls, prior to statin exposure, from normal pregnancies...... 116 Figure 3.6: Diameter of CPAs between statin and controls, prior to statin exposure, from normal pregnancies...... 117 Figure 3.7: Effect of 2 hr incubation with statins on basal tone of chorionic plate arteries relative to control group...... 118 Figure 3.8: Effect of 2 hr incubation with statins on basal tone relative to control group...... 119 Figure 3.9: Effect of 2 hr statin incubation on contraction of CPAs from normal pregnancies. .... 120 Figure 3.10: Effect of 2 hr statin incubation on contraction of CPAs from normal pregnancies. .. 121 Figure 3.11: Effect of 2 hr statin incubation on relaxation of CPAs from normal pregnancies. .... 122

Figure 3.12: Effect of 2 hr pravastatin and pitavastatin incubation on relaxation of CPAs from normal pregnancies...... 123 Figure 3.13: Effect of 2 hr incubation with statins on basal tone of CPAs relative to control group.125 Figure 3.14: Effect of 2 hr incubation with pitavastatin on CPA basal tone relative to control group...... 126 Figure 3.15: Effect of 2 hr statin (1 µM) incubation on contraction of CPAs from pregnancies with PE...... 127 Figure 3.16: Effect of 2 hr statin incubation on contraction of CPAs from PE pregnancies...... 128 Figure 3.17: Effect of 2 hr statin incubation on relaxation of CPAs from pregnancies with PE. .... 129 Figure 3.18: Effect of 2 hr pitavastatin incubation on relaxation of CPAs from PE pregnancies. .. 130 Figure 4.1: A myography trace illustrating a ‘self-collapsing’ vessel...... 151 Figure 4.2: Representative traces of the wire myography protocols for omental arteries assessing relaxation and contraction post-statin incubation...... 153 Figure 4.3: Flowchart of vessels used and excluded in pitavastatin and pravastatin experiments from normal and pathological pregnancies ...... 158 Figure 4.4: Diameter of OAs from PE, CHT, superimposed PE and normal pregnancies prior to treatment ...... 159 Figure 4.5: Assessment of vascular reactivity in OAs from normal, PE, CHT and superimposed PE pregnancies...... 161 Figure 4.6: Effect of 2 hr incubation with pitavastatin and pravastatin on basal tone of omental arteries relative to control group...... 162 Figure 4.7: Effect of time following 2 hr incubation with control (either DMSO or water) on contraction of OAs from normal pregnancies...... 164 Figure 4.8: Effect of 2 hr statin incubation on contraction of OAs from normal pregnancies...... 165 Figure 4.9: Effect of 2 hr statin incubation on relaxation of OAs from normal pregnancies...... 166 Figure 4.10: Effect of 2 hr incubation with pitavastatin on basal tone of OAs relative to control group...... 168 Figure 4.11: Effect of time following 2 hr incubation with control (either DMSO or water) on contraction of OAs from PE, CHT or superimposed PE pregnancies...... 170 Figure 4.12: Effect of 2 hr pitavastatin (1 µM) incubation on contraction of OAs from pregnancies with PE, CHT and superimposed PE...... 171 Figure 4.13: Effect of 2 hr pitavastatin incubation on relaxation of OAs from pregnancies with PE, CHT and superimposed PE...... 173 Figure 5.1: A representative myography trace for ex vivo mouse experiments (uterine artery). . 194 Figure 5.2: Experimental allocation of mice for in-vivo studies ...... 195 Figure 5.3: Experimental flowchart illustrating use of animals in the in vivo dosing study with pitavastatin...... 196 Figure 5.4: A representative image of the myography protocol (for uterine and mesenteric arteries) following pitavastatin treatment in vivo...... 198 Figure 5.5: Maternal weight in eNOS-/- and WT mice at E18.5 ...... 199 Figure 5.6: Fetoplacental measurements from WT and eNOS-/- mice...... 201 Figure 5.7: Uterine artery diameter from eNOS-/- and WT mice at E18.5 prior to statin exposure202 Figure 5.8: Effect of 2 hr pitavastatin incubation on basal tone in eNOS-/- and WT mice at E18.5 203 Figure 5.9: Effect of 2 hr pitavastatin incubation on contraction responses of uterine arteries from WT and eNOS-/- mice...... 205

Figure 5.10: Effect of 2 hr pitavastatin incubation on relaxation responses of uterine arteries from WT and eNOS-/- mice...... 207 Figure 5.11: Umbilical artery diameter in eNOS-/- and WT mice at E18.5 prior to statin exposure.208 Figure 5.12: Effect of 2 hr pitavastatin incubation on basal tone of umbilical arteries in eNOS-/- and WT mice at E18.5...... 209 Figure 5.13: Effect of 2 hr pitavastatin incubation on contraction responses of umbilical arteries from WT and eNOS-/- mice...... 211 Figure 5.14: Effect of 2 hr pitavastatin incubation on relaxation responses of umbilical arteries from WT and eNOS-/- mice...... 212 Figure 5.15: Comparison of fluid intake and maternal bodyweight from pitavastatin-treated and vehicle groups in eNOS-/- and WT mice...... 215 Figure 5.16: Fetal and placental weights from WT and eNOS-/- mice following pitavastatin treatment...... 219 Figure 5.17: Comparison of uterine artery diameter between pitavastatin-treated and those on vehicle in eNOS-/- and WT mice at E18.5...... 221 Figure 5.18: Contraction responses of uterine arteries to U46619 in pitavastatin-treated and vehicle groups in WT and eNOS-/- mice...... 223 Figure 5.19: Assessment of relaxation of uterine arteries from pitavastatin-treated and vehicle groups in WT and eNOS-/- mice...... 225 Figure 5.20: Comparison of mesenteric artery diameter between pitavastatin-treated and vehicle groups in eNOS-/- and WT mice at E18.5 ...... 226 Figure 5.21: Assessment of contraction of mesenteric arteries from pitavastatin-treated and vehicle groups in WT and eNOS-/- mice...... 228 Figure 5.22: Assessment of relaxation of mesenteric arteries from pitavastatin-treated and vehicle control groups in WT and eNOS-/- mice...... 230 Figure 5.23: Umbilical artery diameter in pitavastatin- treated and vehicle eNOS-/- and WT mice at E18.5 ...... 231 Figure 5.24: Assessment of contraction of umbilical arteries from pitavastatin-treated and vehicle control groups in WT and eNOS-/- mice...... 233 Figure 5.25: Assessment of relaxation of umbilical arteries from pitavastatin-treated and vehicle control groups in WT and eNOS-/- mice...... 234 Figure 8.1: Contraction responses of uterine arteries to U46619 in pitavastatin-treated, vehicle and water control groups in WT and eNOS-/- mice...... 303 Figure 8.2: Assessment of relaxation of uterine arteries from pitavastatin-treated, vehicle and water control groups in WT and eNOS-/- mice...... 304 Figure 8.3: Contraction responses of mesenteric arteries to U46619 in pitavastatin-treated, vehicle and water control groups in WT and eNOS-/- mice...... 305 Figure 8.4: Assessment of relaxation of mesenteric arteries from pitavastatin-treated, vehicle and water control groups in WT and eNOS-/- mice...... 306 Figure 8.5: Contraction responses of umbilical arteries to U46619 in pitavastatin-treated, vehicle and water control groups in WT and eNOS-/- mice...... 307 Figure 8.6: Assessment of relaxation of umbilical arteries from pitavastatin-treated, vehicle and water control groups in WT and eNOS-/- mice...... 308

List of Tables

Table 1.1: gestational information for various commonly used animal models of pregnancy...... 60 Table 1.2 Pharmacokinetics properties of statins ...... 65 Table 1.3: preclinical models using pravastatin in pregnancy ...... 74 Table 2.1: Inclusion and exclusion criteria for pregnant women ...... 82 Table 3.1: Demographic and clinical data for women whose placentas were used in the study. Data are expressed as median and range. Booking blood pressures were acquired in antenatal clinic when the women were between 7 and 15 weeks gestation and maximum blood pressure is the highest blood pressure reading prior to delivery. Mann Whitney U test was performed...... 112 Table 3.2: Summary of measurements of vascular reactivity with/without statin incubation in CPAs from normal and pre-eclamptic pregnancies...... 131 Table 4.1: Demographic and clinical data for women recruited to the study...... 155 Table 5.1 Summary of measurements of vascular reactivity with/without 2hr pitavastatin incubation in uterine and umbilical arteries from eNOS-/- and WT mice...... 213 Table 5.2 Summary table of maternal organ weights ...... 216 Table 5.3 Summary table comparing fetal biometric measurements...... 220 Table 5.4 Summary of vascular reactivity measurements comparing pitavastatin-treated and vehicle groups in uterine, mesenteric and umbilical arteries from eNOS-/- and WT mice...... 235

List of Abbreviations

ACh Acetylcholine BH4 Tetrahydro biopterin BK Bradykinin BMI Body Mass Index BP Blood Pressure C1q Complement component 1q cAMP Cyclic Adenosine Monophosphate cGMP Cyclic Guanylyl Cyclase CHT Chronic Hypertensive CMC Carboxymethylcellulose COMT Catetchol-O-methyl transferase COX Cyclo-oxygenase CPA Chorionic plate artery DAG Diacylglycerol DMSO Dimethyl sulfoxide E Embryonic EDHF Endothelium-hyperpolarizing factor EDRF Endothelium-Derived Relaxing Factor eNOS Endothelial nitric oxide synthase EPC Endothelial Progenitor cells ERK Extracellular signal-regulated kinases FGR Fetal Growth Restriction FMD Flow Mediated Dilatation g Gram HDL High Density Lipoprotein Hsd11b2-/- Corticosteroid 11-β-dehydrogenase knockout IP Intraperitoneal

IP3 Inositol triphosphate ISSHP International Society for the study of Hypertension in Pregnancy KCL Potassium Chloride KO Knockout kPa Kilo Pascal

KPSS High potassium salt solution

+ Kv Voltage-gated K channels LDL Low Density Lipoprotein LMWH Low molecular weight heparin L-NAME NG-nitro-L-arginine methyl ester MAP Mean arterial pressure NO Nitric oxide OA Omental arteries PE Pre-eclampsia PI Pulsatility index

PIP3 Phosphatidylinositol (3,4,5)-trisphosphate PlGF Placental Growth Factor PSS Physiological salt solution ROS Reactive oxygen species SBP Systolic blood pressure SC Sildenafil citrate SEM Standard error of the mean sEng Soluble Endoglin sFlt-1 Soluble fms-like tyrosine-1 SGA Small for gestational age sGC Soluble guanylate cyclase SNP Sodium nitroprusside SpA Spiral Artery StAMP Statins to Ameliorate early-onset Pre-eclampsia Superimposed PE Superimposed pre-eclampsia TNFα Tissue necrosis factor-α U46619 Thromboxane-A2 mimetic VEGF Vascular Endothelial Growth Factor WT Wild type

Abstract Pre-eclampsia (PE), which affects 3-5% of pregnancies, is defined as new-onset hypertension and proteinuria post-20 weeks of gestation. There are no effective therapies for PE, which remains a leading cause of maternal and fetal morbidity and mortality. Chronic hypertension (CHT) is a risk factor for PE and shares similar pathological similarities such as endothelial dysfunction with women with CHT often going on to develop superimposed PE. Statins have been proposed as a candidate therapy for PE due to their pleiotropic effects such as being anti-inflammatory, antioxidants and increasing NO bioavailability. However, statins are currently contraindicated in pregnancy and studies have shown them to have detrimental effects in 1st trimester placental explants. Investigations into statins’ ability to ameliorate the observed endothelial dysfunction in PE but also side effects on vascular function is lacking.

With pravastatin already being assessed in clinical trials for the treatment and prevention of PE, it is important to also investigate whether other statins are safe and efficacious for use in PE. In particular pitavastatin, a novel statin which has not be investigated in pregnancy studies and has potent pleiotropic effects. This study hypothesized that acute statin exposure would be able to improve vascular function without detrimental effects on fetal vessels in hypertensive pregnancies relative to normotensive ones. Similarly, the endothelial nitric oxide synthase knockout mouse (eNOS- /-), a model of CHT and FGR, and wildtype (WT) mice were used to assess both acute (ex-vivo) and chronic (in-vivo) statin exposure.

Human chorionic plate arteries (CPAs) from normal and PE pregnancies and omental arteries (OAs) from normal, PE, chronic hypertensive and superimposed PE pregnancies were mounted on a wire myograph. Contraction was assessed with KPSS and thromboxane-mimetic U46619. Arteries were incubated for 2h with 1µM or 10µM of a statin; time-controls in parallel. U46619 dose–response curves were repeated or dose-response curves with NO-donor SNP or endothelium-dependent bradykinin (BK; humans only) or acetylcholine (ACh; mice only) performed following U46619 pre- constriction. Uterine and umbilical arteries from eNOS-/- and WT mice were collected at embryonic day (E)18.5 and their function assessed as above. In a separate study, pitavastatin (6µg/ml) was administered, via drinking water, to eNOS-/- and WT dams from E10.5-E18.5. Uterine, mesenteric and umbilical arteries were harvested at E18.5, ex-vivo vascular function was assessed using wire myography while maternal body, hysterectomized and organ weight were recorded, as well as fetal/placental /biometric measurements.

Key findings:-

 CPAs from normal and PE pregnancies showed similar responses following exposure to the vasoconstrictive agent U46619 and the relaxatory agent SNP. Short-term exposure to pravastatin, simvastatin and pitavastatin did not cause detrimental effects on CPA reactivity.  Acute exposure of OAs from hypertensive pregnancies to pitavastatin (1 µM) does not reduce U46619-mediated contraction or enhance BK-mediated relaxation of vessels. Although endothelial function of OAs from hypertensive pregnancies showed no difference to those from normotensive pregnancies.  Endothelial dysfunction was evident in the eNOS-/- mice relative to WT mice. Short-term exposure to pitavastatin at a suprapharmacological dose had no beneficial effect on vascular function in eNOS-/- mice and a potentially harmful effect in uterine and umbilical arteries from WT mice. Chronic pitavastatin exposure resulted in an enhanced relaxation to ACh in uterine arteries from eNOS-/- treated mice. This all occurred without pitavastatin having any obvious detrimental effects on the fetus as evidenced by normal fetal/placental weight and litter size.

In conclusion, the work within this thesis has demonstrated that acute treatment with statins does not improve vascular function in vessels from humans or mice, with potentially harmful effects on

maternal and fetal vessels from WT mice. Chronic treatment with pitavastatin significantly enhanced relaxation in uterine arteries from eNOS-/- mice and was shown to be well tolerated and not detrimental to fetal wellbeing. Based on current evidence, the data does not strongly support use of pitavastatin for clinical management of PE. Future work should focus on exploring effects of alternative statins such as pravastatin or rosuvastatin on maternal and fetal vascular function in pregnancy as well as investigate maternal-fetal drug transfer.

Declaration No portion of the work referred to in the thesis has been submitted in support of an application for another degree or qualification of this or any other university or other institute of learning.

Copyright Statement i. The author of this thesis (including any appendices and/or schedules to this thesis) owns certain copyright or related rights in it (the “Copyright”) and s/he has given The University of Manchester certain rights to use such Copyright, including for administrative purposes.

ii. Copies of this thesis, either in full or in extracts and whether in hard or electronic copy, may be made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part of any such copies made.

iii. The ownership of certain Copyright, patents, designs, trademarks and other intellectual property (the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproductions.

iv. iv. Further information on the conditions under which disclosure, publication and commercialisation of this thesis, the Copyright and any Intellectual Property and/or Reproductions described in it may take place is available in the University IP Policy (see http://documents.manchester.ac.uk/DocuInfo.aspx?DocID=2442 0), in any relevant Thesis restriction declarations deposited in the University Library, The University Library’s regulations (see http://www.library.manchester.ac.uk/about/regulations/) and in The University’s policy on Presentation of Theses.

Presentations and Publications Poster presentations

1. IFPA Conference (2017), Manchester, UK

Agwu CN, Dilworth MR, Wareing M. Supplementation with Pitavastatin does not ameliorate the vascular function of chorionic plate arteries. Placenta.

2. Blair Bell Annual Meeting, (2018 and 2019) London, UK

Agwu CN, Dilworth MR, Wareing M. Pitavastatin alters vascular function of chorionic plate arteries from pregnancies complicated by pre-eclampsia.

Agwu CN, Dilworth MR, Wareing M, Myers J. Short-term exposure to pitavastatin does not ameliorate the observed endothelial dysfunction in endothelial nitric oxide synthase knockout mice.

3. SRI Conference (2019), Paris, France

Agwu CN, Dilworth MR, Myers J. Vascular reactivity following exposure to pitavastatin demonstrates sex-specific effects in the endothelial nitric oxide synthase knockout mouse. Reproductive Sciences.

Poster-Pitch Presentation

1. ISSHP Conference (2018), Amsterdam, UK

Agwu CN, Dilworth MR, Myers J, Wareing M. The effect of pitavastatin and pravastatin on omental and chorionic plate artery function in normal pregnancy and preeclampsia. Pregnancy Hypertension

Publications outside of thesis

1. Lash GE, Pitman H, Morgan HL, Innes BA, Agwu CN, Bulmer JN. Decidual macrophages: key regulators of vascular remodelling in human pregnancy. Journal for leukocyte biology. 2016 Jan; 99(2).

2. Rocha-Ferreira E, Sisa C, Bright S, Fautz T, Harris M, Contreras Riquelme I, Agwu C, Kurulday T, Mistry B, Hill D, Lange S and Hristova M. Curcumin: Novel Treatment in Neonatal Hypoxic-Ischemic Brain Injury. Frontiers in Physiology. 2019 Nov; 10:1351.

Acknowledgements Firstly, I would like to thank God for the opportunity and privilege to carry out this PhD at the University of Manchester and giving me the strength, good health, peace and motivation to start and complete it successfully.

I would also like to celebrate and thank my diligent, kind, patient and 100% dedicated supervisors Dr Mark Dilworth, Dr Mark Wareing and Dr Jenny Myers, whose practical and theoretical expertise were invaluable to me during this PhD. I have grown so much as a scientist and I will treasure the memories shared with them.

I am also thankful to my colleagues, who have continuously encouraged and supported me throughout my time at the University, providing good conversation, friendship, laughter, cake and encouragement. Dr Elizabeth Cottrell, Dr Teresa Tropea, Dr Stephanie Bosworth-Worton, Dr Sue Greenwood, Hager, Matina, Rezwana and Harry thank you so much. As well as Professor Carolyn Jones for kindly assisting me with proofreading my thesis chapters.

Additionally, I would like to thank family and friends who have been a solid and ongoing support system from the very beginning, thank you for your prayers, willingness to help and continuous encouragement. My cousin Kenechi for helping me with formatting and her general enthusiasm and support throughout my PhD journey.

Finally, I would like to thank the University of Manchester for awarding me with the Full President’s Doctoral Scholar Award and making this PhD financially feasible for me but also the additional workshops, which have helped develop me for the working world.

Chapter 1: General Introduction

1.0 Overview During pregnancy, there are a variety of physiological, and in particular cardiovascular adaptations that the gravid woman undergoes in order to support the developing fetus. These changes include an increase in plasma volume (Soma-Pillay et al., 2016) and increased cardiac output (40%) due to peripheral vasodilation leading to a decrease in systemic vascular resistance by 25-30% (Duley, 2003). Furthermore, this decrease in systemic vascular resistance results in greater perfusion of the kidneys resulting in a 50% increase in glomerular filtration rate (GFR) and 80% increase in renal plasma flow (RPF) compared to nonpregnant levels (Cheung and Lafayette, 2013; Hussein and Lafayette, 2014). In normal pregnancy, these cardiovascular adaptations help to ensure the fetus receives an adequate supply of nutrients and oxygen, through appropriate development and function of the placenta, which will be considered later in this section. However, when a woman has an underlying health condition such as diabetes or the cardiovascular adaptations to pregnancy are suboptimal, pregnancy pathologies such as pre-eclampsia (PE) may be the result, which will likely have consequences for the developing fetus as well as for the mother.

PE is a multisystemic disorder affecting 3-5% of pregnancies in developed countries and up to 7.5% in developing countries (Roberts and Cooper, 2001). PE is defined as ‘new- onset’ hypertension (blood pressure (BP) ≥140 mmHg systolic and/or ≥ 90 mmHg diastolic, based on at least 2 measurements taken at least 4 h apart) occurring in a previously normotensive pregnant woman post-20 weeks gestation, and proteinuria (defined as urinary excretion of ≥0.3 g protein in 24 h) (Chen et al., 2014). In severe cases of PE, and without clinical intervention, some women may progress to eclampsia, characterised by the onset of convulsions. Eclampsia affects up to 1% of women in developing countries compared to 0.05% of women in developed countries (Liu et al., 2009). According to the World Health organisation (WHO), hypertension is amongst the major pregnancy complications that contribute towards 75% of maternal deaths during pregnancy (PE and eclampsia). In addition, severe bleeding and infections after childbirth are contributors (Say et al., 2014). There are various clinical subtypes of PE which include early and late onset cases that present either with or without fetal growth restriction (FGR) (Cartwright et al., 2010). These subtypes will be elucidated further in the PE section (1.1). FGR is defined as a fetus that fails to achieve their genetic growth potential (Gordijn et al., 2016). Both PE and

FGR list placental dysfunction as a major cause (Brosens et al., 2011; Lean et al., 2017; Sibley, 2017), with the aetiology thought to be primarily due to inadequate trophoblast invasion in the first trimester, explained in more detail later in this section. FGR complicates ~66% of pregnancies with severe early-onset PE (Jtowicz et al., 2019; Kovo et al., 2012; Weiler et al., 2011). Figure 1.1 outlines the pre-clinical and clinical stages of PE.

Despite attempts to introduce primary and secondary prevention, PE does not currently have a curative treatment, with the only resolution being delivery of the baby to remove the placenta, often prematurely, which can also result in infant morbidity/mortality and substantial healthcare costs (Pijnenborg et al., 2006). For this reason, there is an urgent need to develop and introduce new and effective treatments which can either prevent or resolve PE (Whitley and Cartwright, 2010).

It has been suggested that PE shares a similar pathology to atherosclerotic cardiovascular disease (CVD) in regards to increased levels of inflammation and endothelial dysfunction. Statins are known to be effective in primary and secondary prevention of cardiovascular mortality and morbidity (Brugts et al., 2009; Mills et al., 2008). In addition, statins have demonstrated cholesterol-independent (pleiotropic) effects such as enhancement of endothelial function, anti-inflammatory and antioxidant properties and the prevention of platelet aggregation (Kavalipati et al., 2015). As such, statins will be a major focus of this thesis but firstly, PE, including its importance and clinical detection, will be discussed.

Figure 1.1. Diagram outlining the pathology of PE from the pre-clinical to the clinical stage, also including the subsequent consequences of it .

(Roberts and Escudero, 2012)

1.1 Hypertensive Disorders of Pregnancy

1.1.1 Definition Based on the International Society for the Study of Hypertension in Pregnancy (ISSHP) guidelines, PE is defined as new onset gestational hypertension at or after 20 weeks gestation accompanied by one or more of the following: proteinuria, liver dysfunction, neurological complications, haematological complications or complications in uteroplacental blood flow (Brown et al., 2018). PE is a multisystemic disorder which affects 3-5% of pregnancies, resulting in severe health consequences for both the mother and the fetus (Cartwright et al., 2010). PE can fall broadly into two different clinical subtypes: early onset and late onset. Early onset PE presents clinically prior to 34 weeks of gestation whilst late-onset PE occurs after 37 weeks (Nice, 2011). Cases of PE can be further broken down into mild and severe forms. Mild/moderate PE refers to a systolic BP of 140mmHg – 159mmHg systolic and/or a diastolic BP of 90-109mmHg (Nice, 2011) whereas severe PE refers to a systolic BP of ≥ 160mmHg and/or a diastolic ≥110 mmHg. Each year, 20% of the 15 million preterm births reported globally are related to PE according to WHO (Kinney et al., 2012; Liu et al., 2012). FGR complicates ~12% of total PE cases however there is a higher incidence (66%) amongst early-onset PE pregnancies (Foo et al., 2018). The clinical signs of PE do not become evident until approximately 20 weeks of gestation onwards, although the pathogenesis of the disorder, particularly in severe early-onset cases, arises in the first trimester due to inadequate spiral artery remodelling following a failed invasion process by extravillous trophoblast cells (Brosens et al., 1972; Pijnenborg et al., 1991).

A risk factor for PE is chronic hypertension, diagnosed either prior to pregnancy or <20 weeks gestation. Around 25% of chronic hypertensive women proceed to develop superimposed PE. Superimposed PE is defined as worsening hypertension, de novo proteinuria and/or the presence of other features suggestive of placental disease (Duhig et al., 2019) in a woman with a diagnosis of chronic hypertension. Chronic hypertension, alongside other hypertensive disorders of pregnancy such as PE are largely associated with risk factors such as essential hypertension along with a family history of hypertension due to renal parenchymal disorders, fibromuscular hyperplasia of renal arteries or primary hyperaldosteronism (Brown et al., 2018). Obesity is also a risk factor for both PE and chronic hypertension and is associated with an increase in sympathetic nervous system activity leading to raised arterial tone (Thorp and Schlaich, 2015), a reduction in anti-inflammatory

adiponectin levels (Matsuzawa et al., 2004) and raised insulin levels (Defronzo, 1981). Obesity can result in alterations in renal function, sodium excretion and salt-sensitivity and dysregulation of the renin-angiotensin-aldosterone system in non-pregnant individuals (Leggio et al., 2017). Together these contribute towards endothelial dysfunction, arterial stiffness and hypertension. Below is a list of risk factors for PE and other hypertensive disorders of pregnancy taken from (Burton et al., 2019): -

Chronic hypertension

Antiphospholipid antibody syndrome

Systemic lupus erythematosus

Pre-gestational diabetes

Chronic renal disease

Multifetal pregnancy

Pre-pregnancy BMI >30

Previous stillbirth

Nulliparity

Maternal age >40

Increased pre-pregnancy BMI

Long inter-pregnancy interval (>5 years)

Reduced school education

Previous pre-eclampsia

Assisted reproduction

Previous intrauterine growth restriction

Previous placental abruption

1.1.2 Identification of High-Risk Pregnancies: PE Abnormalities in uterine artery blood flow, detected via ultrasound, are suggested to be due to inadequate remodeling of spiral arteries, discussed later, which is often associated with placental malperfusion and impaired fetal growth. Measures of uterine artery blood flow velocity are commonly used for the prediction/screening of PE and FGR (Olofsson et al., 1993). Measurements of uterine artery impedance can be carried out during the first trimester (11-13 weeks) via a transabdominal or transvaginal approach (Khong et al., 2015). A high uterine artery pulsatility index (PI) is indicative of poor placental perfusion (O'gorman et al., 2016). The PI is most commonly used because of its stability and its ability to provide information on absent or reversed diastolic values as well as poor placental perfusion (Bhide et al., 2013). PI calculates the difference between systolic and diastolic velocity and normalizes that value to the mean velocity (Timor-Tritsch and Monteagudo, 2009). Once three similar consecutive waveforms are found, PI is measured, and the presence or absence of early diastolic notching is recorded (Albaiges et al., 2000). Early diastolic notching in the uterine artery is defined as a reduction in the diastolic velocity compared with that in later diastole, reflecting vessel elasticity (Gomez et al., 2008; Zhong et al., 2010), continuous early diastolic notching has been associated with abnormal maternal vascular tone (Mo et al., 1988). Uterine artery notching identifies high resistance in blood flow, which is a risk factor for later disease (Papageorghiou et al., 2001); defined as a fall of at least 50 cm/s from the maximum diastolic velocity (Khong et al., 2015). Figure 1.2 shows a Doppler waveform from a normal pregnancy and an abnormal Doppler waveform with early diastolic notching in the uterine artery (Bhide et al., 2013). Resistance index (RI) is a Doppler ultrasound index defined as peak systolic velocity minus diastolic velocity, normalised to the peak systolic velocity (Hassan As, 2018). This index provides information about resistance to blood flow within the uteroplacental circulation (Khong et al., 2015). RI values above the 95th percentile standardized for the gestational age are considered abnormal for the uterine artery (Gomez et al., 2008). These are the indices that can be utilised to help diagnose placental dysfunction in PE (Sotiriadis et al., 2019).

1.1.2.1 Diagnosis of Chronic Hypertension in Pregnancy Chronic hypertension is often diagnosed at the first antenatal appointment in the first or early second trimester, and usually confirmed within 24 hrs. Repeated hypertensive readings via ambulatory blood pressure monitoring or home blood pressure monitoring or

on two consecutive antenatal visits lead to a diagnosis of hypertension (Brown et al., 2018). Ultrasound monitoring is also used in chronic hypertensive women.

1.1.2.2 Diagnosis of Fetal Growth Restriction Measurement of umbilical artery blood flow velocity is useful in monitoring fetal wellbeing and helps inform clinicians about timing of delivery in the case of fetal compromise due to FGR (Mone et al., 2015). An umbilical artery Doppler result is considered abnormal when RI and PI values are >95th percentile standardized for gestational age (Lopez-Mendez et al., 2013). According to the recent Delphi Consensus, early-onset FGR is defined as occurring before 32 weeks with: fetal abdominal circumference < 3rd centile, fetal weight <3rd centile OR predicted fetal weight or waist circumference < 10th centile AND a pulsatility index, following Doppler blood flow velocity measures of the uterine and umbilical arteries, above the 95th centile for gestational age. Late-onset FGR occurs after 32 weeks and is defined as: fetal abdominal circumference OR estimated fetal weight < 3rd centile OR as a combination of two of the following: fetal weight OR abdominal circumference < 10th centile, reduction of > two quartiles in the growth curve and cerebroplacental association < 5th centile OR an umbilical artery pulsatility index above the 95th centile for gestational age (Nardozza et al., 2017). The umbilical artery, middle cerebral artery and precordal vein Doppler provide information regarding placental disease, level of redistribution and fetal cardiac compromise respectively.

Figure 1.2: a) shows a normal Doppler waveform b) shows an abnormal Doppler waveform with uterine artery notching. Uterine artery (Bhide et al., 2013) notch

Additional investigations can be conducted to confirm diagnosis of PE or other hypertensive disorders of pregnancy. PE is a disorder characterised by widespread endothelial dysfunction. The kidneys are particularly affected and display glomerular capillary endotheliosis and resultant proteinuria due to damage caused by the elevated BP (Spargo et al., 1959). Other signs include oliguria (secretion of abnormally low amounts of urine), cerebral or visual disturbances (due to impaired autoregulation; (Schwartz et al., 2000), pulmonary oedema, epigastric/right upper quadrant pain, impaired liver function, thrombocytopenia and edema (in face, hands and feet) (Practice, 2002; Roccella, 2000).

1.1.3 Clinical Investigations A dipstick test is performed in spot urine samples and if positive, proteinuria would be highlighted as 1+, a sample is also sent for PCR analysis. Albumin: creatinine ratio or protein: creatinine ratio can be used to help diagnose proteinuria where a value above 30 mg/mmol indicates significant proteinuria (Nice, 2019a). Additionally, maternal biochemical and haematological investigations can be conducted such as assessing a rise in creatinine (90 micromol/litre or more, 1 mg/100 ml or more) or rise in alanine transaminase (over 70 IU/litre, or twice upper limit of normal range) or fall in platelet count (under 150,000 cells/microlitre) (NICE 2019). The antiangiogenic protein sFlt-1 is a non-membrane associated and soluble form of the VEGF receptor. When levels of sFlt-1 are elevated, such as in PE, they bind to and sequester the proangiogenic VEGF and PlGF, both VEGF and PlGF are made by the placenta and circulate in high concentration during pregnancy (Karumanchi et al., 2010). More recently, quantitative determination of placental growth factor (PlGF) concentrations in maternal blood can be achieved using the Alere Triage® PlGF test (Triage® MeterPro, Inc., San Diego, USA). The test is intended for use in conjunction with other existing diagnostic tests to aid the diagnosis of PE and to assess the level of risk for delivery arising from PE within 14 days of testing (Chappell et al., 2013; Duhig et al., 2019; Frampton et al., 2016). PlGF is predominantly expressed in the placenta and is crucial in the development and maturation of the placental vascular system. In PE, urinary and serum PlGF levels decrease making it a useful biomarker for PE prediction (Chau et al., 2017). Figure 1.3 illustrates how PlGF levels vary between normal and PE pregnancy due to angiogenic imbalance between PlGF and sFlt-1.

Figure 1.3: Angiogenic imbalance due to sFlt-1, a potent circulating antagonist to

PlGF, resulting in placental physiopathology in PE.

(Lecarpentier and Tsatsaris, 2016)

1.1.4 Clinical Management of PE In severe cases of PE where there is a threat to maternal or fetal health, women are offered admission into hospital to allow clinicians to conduct regular ultrasound assessments to monitor fetal wellbeing assessing heart rate, growth and movement (Townsend et al., 2016). Following diagnosis of PE, women are offered hospital admission to conduct further assessments including the assessment of umbilical artery function via Doppler ultrasound; venous thrombus embolism risk is also assessed. Women with PE are also offered hospital admission because of the risk of maternal complication. However, if a woman with PE has stable BP and fetal wellbeing appears normal they are able to return home (Townsend et al., 2016).

1.1.4.1 Antihypertensive Drugs In order to manage hypertension, antihypertensive drugs are commonly prescribed in patients with PE (Podymow and August, 2008). Common pharmacological treatments for hypertension outside of pregnancy include ACE inhibitors, angiotensin-II receptor blockers, beta-adrenergic blockers, calcium channel blockers and thiazide diuretics (Withagen et al., 2005). A number of these antihypertensive drugs are also used to manage hypertension during pregnancy, thus also reducing the risk of cardiovascular and cerebrovascular consequences (Euser and Cipolla, 2009). The primary aim of antihypertensives during pregnancy is to stabilise or decrease maternal blood pressure. Various antihypertensives have been deemed safe to use to manage hypertension during pregnancy. These include 33

short acting oral calcium-channel blocker nifedipine, or oral/intravenous beta adrenoreceptor blocker labetalol or the vasodilator hydralazine in an acute setting (Tranquilli et al., 2014).

Women on existing antihypertensive treatment prior to pregnancy are advised to continue to keep taking their medication if it is safe to use in pregnancy, or switch to an alternative antihypertensive treatment (NICE, 2019). These antihypertensives aim to maintain blood pressure within the range of 110-140/80-85 mmHg. The CHIPS (Control of Hypertension in Pregnancy Study) trial, assessed less-tight (blood pressure that is higher than normal but not severely elevated) versus tight control (the use of antihypertensive therapy to normalize blood pressure) of hypertension on pregnancy complications. This trial provided information about optimal blood pressure ranges as their results showed that diastolic blood pressure of 85 mmHg or less decreased the chances of women developing accelerated maternal hypertension and adverse fetal outcomes (Magee et al., 2015).

1.1.4.2 Anticonvulsants Eclampsia following PE is a contributor towards maternal death, with incidences in Europe being low at 2-3/10,000 births (Subramaniam, 2007) compared to developing countries with incidences of 16-69/10,000 births (Altman et al., 2002). A number of studies have looked into the role of magnesium sulphate as an anticonvulsant treatment for eclampsia to reduce the recurrence of convulsions and reduce the number of maternal deaths as evidenced by the “Magnesium Sulfate for Prevention of Eclampsia (Magpie) trial”. This trial revealed that magnesium sulphate was capable of reducing the incidence of recurrent eclampsia by 50% as well as reducing the risk of maternal death without causing adverse effects to mother or baby (Altman et al., 2002; Duley et al., 2010a; Duley et al., 2010b).

1.1.4.3 Low Dose Aspirin Low dose aspirin (80 to 150 mg) is used for the secondary prevention of PE in high-risk patients from the 12th week of pregnancy until delivery (Podymow and August, 2008). This evidence came from the ‘Collaborative Low-dose Aspirin Study in Pregnancy’ (CLASP) Trial and further confirmed by the ‘Aspirin for Evidence-Based PREeclampsia Prevention (ASPRE) Trial (Collaborative Group., 1994 and Redman and Sargent, 2005). Aspirin works by inhibiting the production of thromboxane from platelets and hence increases the levels of prostacyclin relative to thromboxane leading to a reduction in platelet aggregation (Grigsby, 2016). Due to endothelial prostacyclin levels being unaffected, this ensures that systemic vasodilation can still take place (Sibai et al., 1989). In vitro studies in human trophoblasts have shown that under hypoxic conditions, aspirin is also capable of inhibiting sFlt-1 expression and promoting proangiogenic activity (Li et al., 2015).

In a clinical trial involving 1,776 women with a risk of preterm PE, 150 mg aspirin, administered during the first trimester, resulted in a reduction in preterm PE by 62% relative to those on placebo alone (Rolnik et al., 2017). This outcome was supported by data from other studies that involved ~ 2,200 women, and demonstrated that when aspirin was administered to women prior to 16 weeks, the incidence of preterm PE was halved in a dose-dependent manner (Roberge et al., 2018). In addition to the medications mentioned above, a majority of inpatients are prescribed low molecular weight heparin (LMWH) for thromboprophylaxis due to the increased risk of venous thromboembolism (Townsend et al., 2016).

1.2 Vascular Changes and Placental Development in Normal Pregnancy and Pregnancy Complicated by PE

1.2.1 Vascular Changes in Normal Pregnancy During normal pregnancy, and following implantation, extravillous cytotrophoblast (EVT) cells from the placenta begin to invade the decidua wall towards the myometrial spiral arteries (Vaughan and Fowden, 2016). The spiral arteries are found at the terminal end of radial arteries (Sibley et al., 1997). This invasion triggers the arteries to undergo a process called spiral artery (SpA) remodelling. The first stage is trophoblast-independent whereby the macrophages and uterine natural killer cells prime the SpA for remodelling by mediating extracellular matrix (ECM) disruption, and vascular smooth muscle cell (SMC)

loss. EVT then invade and colonise the vessels and replace the vascular cells (Cartwright et al., 2010). The decidual macrophages accumulate around the SpA and initiate phagocytosis of dead cells (Lager and Powell, 2012). The second stage involves vacuoles forming in the endothelium of the uterine spiral artery and the individual smooth muscle cells swelling, which is a decidual-associated process. Re-endothelialisation occurs after this process and subintimal thickening occurs with α-actin (Diaz et al., 2014).

The maternal decidua and the fetal trophoblasts form the feto-maternal interface and are both involved in regulating the placenta’s growth and function and the development of the embryo (Prabhudas et al., 2015). Extensive SpA remodelling helps to facilitate a greater volume of maternal blood to intervillous space but at a low rate of flow as a high flow rate such as not to damage the villi (Burton and Fowden, 2012).

1.2.2 Uteroplacental Vascular changes in pregnancies complicated be PE In PE, defective SpA remodelling results in narrow and inadequately dilated spiral arteries resulting in poor uterine perfusion and maternal blood flow entering the intervillous space in the placenta at a high velocity with the potential to cause damage to the placental villi (Burton et al., 2009b; Zhou et al., 1997). This can sometimes be detected clinically using uterine artery Doppler ultrasound. Furthermore, the intermittent arterial blood reduces oxygen exchange resulting in oxidative stress (Aouache et al., 2018). A high blood flow velocity can facilitate the dislodging of trophoblastic microparticulate debris from the villous surface and due to defective SpA remodelling, smooth muscle cells are still present around the spiral artery, increasing risk of vasoconstriction (Burton et al., 2019). Suboptimal remodelling has also been reported in decidual vessels (Lyall F, 2007) and when vessels are not remodelled appropriately, blood flow at too high a pressure can cause hydrostatic stress which ultimately damages placental villi (Burton et al., 2009b; Burton et al., 2009c), details of which are elucidated in the placenta section 1.3.

Due to suboptimal remodelling and hypoperfusion of the placenta, placental infarcts can arise at a higher frequency than that of normotensive women due to ischaemia (Young, 1914). Furthermore, Brosens et al. observed that the SpA remodelling extends into the myometrium during a normotensive pregnancy whereas in PE remodelling is restricted to the decidua. Through analysis of placental bed biopsy samples, Brosens et al. also noticed that the average diameter of spiral arteries from PE pregnancies was ~200 µM compared to 500 µM for normotensive pregnancies (Brosens et al., 1972). A low oxygen 36

tension promotes the expression of the hypoxia induced transcription factor, HIF1α in placental chorionic villi that is elevated in PE and signifies placental stress (Cindrova-Davies et al., 2015). Oxidative stress can result in damage to endoplasmic reticulum which has been observed in placentas from PE women (Fu et al., 2015; Lian et al., 2011) as well as modifications of lipids, proteins, mitochondria and apoptosis, which contribute to the clinical maternal signs of PE (Burton et al., 2009a). Figure 1.4 provides a diagrammatic representation of normal and abnormal placentation below.

EVT

Figure.1.4. Diagram outlining normal and abnormal SpA remodelling.

A) Sufficient EVT invasion and normal spiral artery remodelling leading to normal placentation B) Shallow EVT invasion resulting in defective spiral artery remodelling and abnormal placentation. Taken from (Gordijn et al., 2016; Roberts and Escudero, 2012).

1.3 Cardiovascular adaptations in normal pregnancy and in PE Cardiovascular adaptations to pregnancy are important physiological processes in order to prepare for the metabolic demands of the growing placenta and fetus (Chapman et al., 1998). These changes promote systemic vasodilation and increase perfusion of maternal organs such as the uterus (Napso et al., 2018). During normal pregnancy, there is an up- regulation in NO synthesis, prostacyclin and oestradiol, which promote vasodilation and

lead to a 25-30% decrease in systemic vascular resistance along with an increase in cardiac output (CO) by 40% (Bader et al., 1955; Soma-Pillay et al., 2016). Pregnancy-related changes also include alterations in the heart musculature such as increased left ventricle muscle mass, ventricular thickness and an increase in right atrial/ventricular diameters resulting in raised heart rate and stroke volume in addition to increased CO (Gati et al., 2014). There is also a rise in blood flow to the kidney (40-65%) and glomerular filtration rate (50-85%) (Davison and Dunlop, 1980; Soma-Pillay et al., 2016). The renin-angiotensin- aldosterone-system (RAAS) is responsible for sodium regulation and is upregulated as gestation progresses resulting in an increase in plasma volume of ~30% by the end of pregnancy (Elsheikh et al., 2001; Tkachenko et al., 2014). This rise in total blood volume occurs to meet the oxygen demands of the maternal organs and fetal demand (Chang and Streitman, 2012). The concentration of red blood cells also increases during pregnancy because of proliferation of erythroid progenitors in the spleen (Bustamante et al., 2008).

A lower cardiac index, increased total vascular resistance, impaired myocardial relaxation and abnormal heart geometry has been identified in women with early-onset PE, features also seen in late-onset PE but of reduced severity (Guy et al., 2017). A decrease in maternal CO, reduced plasma volume and raised vascular resistance is associated with FGR (Melchiorre et al., 2012). Similarly, in PE with associated FGR, CO is lower in the first trimester of pregnancy (Easterling et al., 1991). Furthermore, flow-mediated dilatation (FMD) shows the physiological response of vessels to increased demand of end organs, showing vessel dilation in response to increased blood flow. A meta-analysis has shown that FMD, measured across 18-25 weeks of pregnancy, is reduced in women who go on to develop PE (Weissgerber et al., 2016).

1.4 Importance of the Placenta in Normal Pregnancy and Pregnancies Complicated by PE The placenta is a dynamic, metabolically active organ and has a central role in facilitating a normal pregnancy (Vaughan and Fowden, 2016). The placenta is crucial for the transfer of nutrients such as glucose, fatty acids and amino acids, oxygen, ions and important micronutrients from mother to fetus as well as for the removal of waste products such as carbon dioxide and urea in the fetal-maternal direction (Sibley et al., 1997).

The placenta is also an endocrine organ. It produces human chorionic gonadotrophin (HCG) that stimulates progesterone secretion from the corpus luteum and helps to maintain the viability of the pregnancy (Gude et al., 2004). The syncytiotrophoblast also produces human placental lactogen, which reduces sensitivity to insulin and consequently increases maternal blood glucose levels (Diaz et al., 2014; Burton and Fowden, 2012). Finally, oestrogen and progesterone are also produced by the placenta. Progesterone helps to prevent uterine contractions and early onset of labour while oestrogen results in increased growth of the uterus and promotion of mammary gland development (Griffiths and Campbell, 2014). The placenta acts as a barrier as the fetus is semi-allogenic and so the syncytiotrophoblast does not express highly polymorphic MHC class 1 antigen, which creates a tolerant maternal immune environment during pregnancy, preventing rejection of the fetus (Billington and Bell, 1983).

During the formation of the early placenta, there are two important circulatory systems, vitelline and chorionic (Burton et al., 2016). The vitelline circulation connects the embryo and yolk sac, and is important for the morphogenesis of the fetal heart and materno-fetal transport during the organogenesis period (Gittenberger-De Groot et al., 2013). The yolk sac degenerates after 10 weeks of gestation (Jauniaux et al., 1991). The chorionic circulation develops in the placental villi during the 5th week of gestation and is fully established at ~12 weeks.

Two umbilical arteries, which originate from the fetal internal iliac arteries help to supply blood to the surface of the placenta, branch over the fetal surface of the placenta to produce chorionic plate arteries (CPAs), penetrating the placenta and further differentiating into arterioles and capillaries. As capillaries, they bring deoxygenated blood to the terminal villi of the placenta, an area that is in direct contact with oxygen rich maternal blood in the intervillous space, enabling gas exchange (Acharya et al., 2016). The CPAs spread across the placental cotyledons in a “pattern of disperse-type branching” with some CPAs radiating to the edge of the placenta (Wang Y, 2010). The vascular tone of the umbilical arteries is influenced by vasoactivators such as serotonin, oxytocin and angiotensin II (Benirschke K, 2006). In-vitro studies have revealed that prostaglandin and NO play an important role in umbilical vessels (Harold et al., 1988). As gestation progresses, placental vascular resistance diminishes to promote increased blood flow (Acharya et al.,

2016). Figure 1.5 shows the fetal and maternal surface of the placenta (Wang, 2010), while Figure 1.6 below shows a transverse view of a full-term placenta (Saunders, 2009).

Figure 1.5: Maternal and fetal surface of the placenta (Wang ,

2010).

Figure 1.6: A transverse view of a full-term placenta (Saunders, 2009).

In PE, there is an alteration in the vascular reactivity with umbilical artery Doppler ultrasound waveforms highlighting increased vascular resistance in the fetoplacental circulation in PE with FGR (Trudinger et al., 1985; Trudinger and Ishikawa, 1990). Additionally, studies in placental samples from PE women have also shown that in the placental stem villi there is more elastic tissue fibres in the blood vessels and the vessels are thicker, a consequence of the maternal hypertension (Wilhelm et al., 1999). Signs of hypoperfusion are evident in the placenta, with studies showing placental infarcts at a concentration of 52%, which occurs in some but not all cases of PE (Akhlaq et al., 2012). These infarcts arise due to the occlusion of spiral arteries by thrombus (blood clots), intervillous fibrinoid deposition and impairment of the fetal circulation (Predoi et al., 2015). These placental features arise because of hypoperfusion and ischaemia during pregnancy.

Placentas from PE pregnancies can be smaller, have a reduced weight, and have a reduced surface area (Burton et al., 2010) compared to placentas from normotensive women (Fitzgerald et al., 2011; Zigic et al., 2010). However, women can also have larger but less efficient placentae, especially in late-onset PE. Eccentric cord insertion is also more prevalent in PE (Dekan et al., 2012; Vinnars et al., 2011), which can lead to a sub-optimal chorionic plate vascular distribution, a poor transport gradient and lower birthweight for a given placental weight (reduced placental efficiency).

In PE, the placenta’s access to oxygen is reduced due to abnormalities in uteroplacental blood flow. However, in term PE there is low total vascular resistance and high cardiac output (Phillips et al., 2010) and it has been shown that term PE placentas are similar to normal controls, while preterm PE placentas have a more severe phenotype (Roberts and Escudero, 2012). Alterations in uteroplacental blood flow can lead to the release of antiangiogenic factors into the maternal circulation which are thought to contribute to peripheral endothelial cell injury and hence manifestation of the clinical syndrome of PE (Redman et al 2005). A key contributor to maternal endothelial dysfunction is the angiogenic imbalance between soluble fms-like tyrosine kinase-1 (sFlt-1) and placental growth factor (PlGF) (Levine et al., 2004). The antiangiogenic protein sFlt-1 is a non-membrane associated and soluble form of the VEGF receptor. When levels of sFlt-1 are elevated, such as in PE, they bind to and sequester the proangiogenic vascular endothelial growth factor (VEGF). Other studies have also observed this angiogenic imbalance (Dymara-Konopka et al., 2018; Eddy et al., 2018; Tomimatsu et al., 2017). The

anti-angiogenic factor soluble endoglin-1 also becomes elevated in PE resulting in endothelial dysfunction (Petla et al., 2013). A reduction in the bioavailability of VEGF leads to a reduction in nitric oxide (NO) levels and subsequent vasoconstriction and maternal malperfusion of organs (Burton et al., 2019; Facemire et al., 2009). As endothelial dysfunction is thought to play a key role in the pathogenesis of PE, this widespread dysfunction has a multisystemic effect on various organs in these women (Burton et al., 2019).

1.5 The endothelium and resistance vessels

1.5.1. Resistance vessels It has been demonstrated that increased peripheral vascular resistance to blood flow results in essential hypertension (O’byrne, 2009). Resistance arteries are major contributors to vascular resistance and have lumen diameters of <250 µM when in a resting state (Rucker and Dhamoon, 2019). Abnormalities in resistance artery function contribute to the pathogenesis of hypertension and its outcomes (Belfort et al., 1996; Vedernikov et al., 1995). During hypertension, the lumen diameter of resistance vessels decreases and the media thickness: lumen diameter ratio also increases. This is contributed to by eutrophic inward remodelling of vessels which involves the re-orientation of smooth muscle cells within the vessel, increasing the smooth muscle cell layers and decreasing internal/external diameters of the vessel (Heagerty et al., 2010). Poiseuille’s law says that resistance varies inversely with the fourth power of the blood vessel radius, therefore a minor decrease in the diameter of the lumen can drastically increase resistance (Wareing et al., 2002).

1.5.2 Function of endothelium The vascular wall of blood vessels is made up of three main layers, consisting of the tunica intima (inner lining), tunica media (middle lining) and tunica adventitia (outer lining) (Henderson and Byron, 2007), as depicted in Figure 1.7. Endothelial cells form the innermost cell layer lining the lumen and can be found on the tunica intima; they are responsible for the synthesis, uptake, storage and degradation of vasoactive agents and other external stimuli such as hormones and neurotransmitters, which contributes to the control of vascular function (Haschek et al., 2013; Nemenoff, 1998).

Figure 1.7: Arterial schematic. Taken from (Ng et al., 2018)

1.5.3 Endothelial-dependent and independent vasoconstrictors

1.5.3.1 Phenylephrine Phenylephrine is an endothelium-independent vasoconstrictor, which selectively targets the α1-adrenergic receptors causing a rise in blood pressure mainly by increasing systemic vascular resistance (Burnstock, 1988). The α1-adrenergic receptors are responsible for contraction and hypertrophic growth of smooth muscle cells and there are three subtypes:

α1A-adrenergic receptors, α1B-adrenergic receptors and α1D-adrenergic receptors that are all important for maintaining vascular tone. Phenylephrine activates α1-adrenergic receptors, activating Gq proteins leading to stimulation of phospholipases C, A2 and D resulting in the recruitment of Ca2+ from intracellular stores. This in turn activates mitogen- activated kinase and PI3 kinase pathways leading to vasoconstriction of smooth muscle (Sorriento et al., 2011).

1.5.3.2 Endothelin The endothelium-dependent vasoconstrictor endothelin (ET) has three isoforms: ET-1, ET- 2, and ET-3 but ET-1 is specific to endothelial cells. The endothelin converting enzyme converts the precursor Big ET-1 to the active form ET-1 (Nagatomo et al., 2004). The release of ET-1 is promoted by inflammatory mediators such as interleukins and TNF-α and downregulated by nitric oxide (NO) and prostacyclin (Kanaide et al., 2003; Purdy R.E.,

1989). Stimulation of the ET receptors leads to differential outcomes. When the receptors

2+ 2+ ETA or ETB2 are activated by ET-1, Ca channels are opened, enabling extracellular Ca to enter the cell, causing vasoconstriction of the smooth muscle in a similar manner to thromboxane A2 (Sybertz et al., 1983). However, binding of ET-1 to ET-B1 receptors results in vasodilation by promoting the release of NO and prostacyclin (Vallance et al., 1989).

1.5.3.3 Thromboxane The endothelium-independent thromboxane mimetic U46619 is one of the stable thromboxane analogues and acts as a potent agonist for the thromboxane A2 receptor, which is able to mimic thromboxane’s biological effects. Thromboxane A2 constricts smooth muscle and is released from macrophages, neutrophils and platelets (Palmer et al., 1988). When U44619 binds to the thromboxane A2 receptor, it increases intracellular calcium and activates protein kinase C by activating the beta forms of phospholipase C via

Gq coupling. Thromboxane is also important in platelet aggregation (Prast and Philippu, 2001).

1.5.3.4 Vasopressin The endothelium-independent vasopressin is able to produce smooth muscle constriction by binding to V1a receptors (Needleman and Isakson, 1997). The stimulation of the V1a receptors leads to the production of inositol 1,4,5-trisphosphate (IP3) via the phospholipase C pathway and this stimulates the release of calcium ions into the sarcoplasmic reticulum (Mcadam et al., 1999).

1.5.3.5 Noradrenaline Noradrenaline is an endothelium-independent catecholamine that exerts its constrictive actions by binding to α1-adrenergic receptors and P2X-purinoceptors on the smooth muscle (Nicosia et al., 1987). However, it also has the ability to cause vasodilation via the activation of β-receptors in the endothelium or vascular smooth muscle cells (Billington and Penn, 2003).

1.5.3.6 Serotonin Serotonin (5-HT) is an endothelium-independent monoamine neurotransmitter that stimulates contraction of smooth muscle cells by interacting with the 5-HT2A receptor (Billington and Penn, 2003). In addition, serotonin is able to stimulate α-adrenergic receptors and enhance the vasoconstrictor response to an α-adrenergic receptor agonist (Ivankovich et al., 1978). 44

1.5.3.7 Angiotensin II Active renin acts on angiotensinogen, to generate angiotensin I (Ang I). Ang I is cleaved by angiotensin-converting enzyme (ACE) resulting in physiologically active angiotensin II (Ang II); Ang II, which is endothelium-dependent, mediates its effects via type 1 Ang II receptor (Pacurari et al., 2014). In addition, Ang II binds to a lesser extent to type 2 Ang II receptor, which is believed to have an antagonistic role (Li et al., 2012). Binding of Ang II to the type 1 Ang II receptors leads to the activation of phospholipase C via the coupling with G protein

Gq. This leads to the hydrolyzation of phosphatidylinositol-4,5-bisphosphate to inositol-

2+ 1,45-trisphosphate (IP3) and diacylglycerol. IP3 then stimulates intracellular release of Ca and more Ca2+ enters via Ca2+ channels on the cell membrane. Subsequently Ca2+- calmodulin coupling occurs; this activates myosin light chain kinase, which phosphorylates key sites on myosin light chains, resulting in smooth muscle contraction (Levy, 2005).

1.5.4 Endothelial-dependent and independent vasodilators

1.5.4.1 Nitric oxide (NO) Nitric oxide is an endothelial-derived vasodilator that contributes to the vascular tone of blood vessels. The activity of eNOS is initiated through the phosphorylation of serine sites, in particular Ser1177, which increases Ca2+ sensitivity of the enzyme (Forstermann and Sessa, 2012). Oestrogen and VEGF can promote the kinase AKT to phosphorylate eNOS while insulin can promote AMPK and AKT kinases. The vasodilatory agonists acetylcholine and bradykinin stimulate phosphorylation of eNOS via Ca2+/calmodulin-dependent protein kinase II (Forstermann and Sessa, 2012). Shear stress can also stimulate eNOS phosphorylation on Ser1179 and Ser635 through the activation of protein kinase A (PKA) (Boo et al., 2002). eNOS is crucial for homeostatic regulation of vascular tone, resulting in NO production which causes vasodilation by activation of guanylyl cyclase and increasing concentrations of cGMP in smooth muscle (Rapoport et al., 1983). A further way of increasing eNOS activity is the dephosphorylation of Thr495 (Fleming and Busse, 2003).

1.5.4.2 Prostacyclin The production of endothelium-dependent prostacyclin involves the cyclooxygenase (COX) enzymes of which two isoforms exists – COX-1 and COX-2 (Cohen and Vanhoutte, 1995). COX-1 is present in endothelial cells while COX-2 is only expressed in damaged or inflamed endothelium (Edwards and Weston, 2004). Prostacyclin (PGI2) is generated by the actions of COX-2, through the conversion of arachidonic acid to prostaglandin H2, which is then

synthesized into prostacyclin by the actions of prostacyclin synthase (Edwards and Weston, 2004). Prostacyclin can bind to prostacyclin receptors on platelets and vascular smooth muscle cells (Moncada et al., 1989) and it consequently activates the enzyme adenylate cyclase which generates cyclic adenosine monophosphate (cAMP) (Spiers and Padmanabhan, 2005). cAMP stimulates protein kinase A which promotes smooth muscle relaxation (Falloon et al., 1995).

1.5.4.3 Sodium nitroprusside The endothelium-independent agonist sodium nitroprusside (SNP) is used as a NO-donor to stimulate endothelium-independent relaxation. SNP is a water-soluble sodium salt, its structure includes Fe2+ complexed with nitric oxide (NO) and 5 cyanide anions. When in the body, SNP behaves like a prodrug and reacts with sulfhydryl groups on erythrocytes, albumin and other proteins to stimulate the donation of NO (Vanbavel et al., 1990). NO activates the enzyme guanylyl cyclase to produce cyclic GMP (cGMP), which can sequester calcium and promote relaxation of the smooth muscle (Mulvany and Nyborg, 1980).

1.5.4.4 Acetylcholine The endothelium-dependent agonist acetylcholine (ACh) is a neurotransmitter found at the neuromuscular junction and in the autonomic ganglia. It is also present in non-neuronal cells such as keratinocytes, lymphocytes, placental trophoblast and endothelial cells (Mulvany and Nyborg, 1980). ACh can induce relaxation through a variety of mechanisms including activation of eNOS to stimulate NO release but also via the COX pathway to trigger prostaglandin production (Moncada et al., 1989).

1.5.4.5 Bradykinin The endothelium-dependent agonist bradykinin is a vasodilatory agonist that is able to interact potently with the B2 receptor (Flammer et al., 2012). Kinin peptides have broad roles and help regulate vascular tone and protect vessels from ischaemia-reperfusion injury (Flammer et al., 2012). Bradykinin is believed to stimulate vasodilation by activating eNOS or by stimulating endothelium-derived hyperpolarizing factor (Sandoo et al., 2010).

1.5.4.6 Endothelium-Derived Hyperpolarising Factor (EDHF) When endothelial cells are exposed to agonists such as bradykinin and acetylcholine, the endothelium-dependent Endothelium-Derived Hyperpolarising Factor (EDHF) is released (Al Qahtani, 2011). This raises [Ca2+] causing the efflux of potassium ions from the cell (Brown et al., 2018). The smooth muscle cells respond to this efflux of potassium ions by 46

also releasing potassium ions, which result in hyperpolarization of the smooth muscle (Edwards et al., 1998). This then reduces intracellular [ Ca2+] and triggers smooth muscle relaxation (Cartwright et al., 2010).

1.5.4.7 Histamine Histamine causes endothelium-dependent vasodilation by binding to H1 receptors on endothelial cells and stimulating the release of NO, which subsequently diffuses into smooth muscle cells and activates guanylate cyclase. This raises levels of cGMP, causing smooth muscle relaxation (Moncada et al 1989). Additionally the H2 receptors on smooth muscle can also be stimulated and promote endothelium-independent vasodilation through cAMP (Ebeigbe and Talabi., 2014).

1.5.4.8 Substance P Substance P is an endothelium-dependent agonist that stimulates vasodilation via NO- dependent and NO-independent mechanisms by binding to the Tachykinin receptor 1 otherwise known as NK1 receptor (Garcia-Recio and Gascon, 2015). Substance P can directly induce endogenous NO release from endothelial cells by stimulating eNOS (Karabucak et al., 2005). Alternatively, substance P can also activate adenylate cyclase in smooth muscle, producing cAMP which then activates protein kinase A. Protein kinase A is able to phosphorylate and inactivate RhoA enabling the myosin light chain phosphatase to dephosphorylate myosin light chain and promote smooth muscle (endothelium- independent) relaxation (Suvas, 2017). Figure 1.8 summarises the various vasoconstrictors and vasodilators associated with the endothelium.

Figure 1.8: Illustration of vasoactivators that can cause vasodilation and vasoconstriction

NO (nitric oxide), AII (angiotensin II receptors); Ach (acetylcholine); EDHF (endothelium-derived

hyperpolarizing factor; ET (endothelin); His (histamine); NE (norepinephrine); PE (phenylephrine); PGI2 (prostacyclin) and 5-HT (serotonin) (Oguzhan et al., 2013).

1.5.5 Methods of Measuring Vascular Reactivity

1.5.5.1 Ex-vivo: Wire and Pressure myography Wire myography is a robust ex-vivo technique that measures the functional response and vascular reactivity of small resistance arteries, and veins, to pharmacological agents. The vessel of interest is isolated, cleaned and mounted on a myograph. This involves passing wires through the lumen of the vessel and measuring tension in response to vasoconstrictors/vasodilators as well as pharmacological agents. Each vessel is then normalized to determine maximum active tension development. This allows the standardization of initial experimental conditions, an important consideration when examining pharmacological differences between vessels (Bridges et al., 2011; Spiers and Padmanabhan, 2005). The limitations are that the mounting process, where the wire is entering the lumen of the vessel, can cause endothelial injury, thus leading to a nonphysiological geometry and loading. This can in turn, affect how these vessels respond to agonists (Falloon et al., 1995; Jadeja et al., 2015). Pressure myography is used to investigate the physiological function and properties of small arteries, veins and other vessels (Shahid and Buys, 2013). The procedure involves the cannulation of isolated blood vessels, connecting them to a pressure-perfusion system that controls intraluminal pressure, and 48

then the measurement of the diameter of the artery using microscopic techniques (Lawton et al., 2019). Pressure myography is thought to have a distinct advantage over wire myography because it is considered to be more representative of in-vivo conditions. Moreover, when an artery is pressurised, within physiological limits, physiological radial distension is generated hence removing the requirement for normalising radial tension (Vanbavel et al., 1990).

In rat resistance arteries, normalised internal diameter has been shown to be 90% of the measured vessel diameter when fully relaxed and subjected to a transmural pressure of 100mmHg (Mulvany and Nyborg, 1980). This is therefore the definition given for describing the ‘normalised’ lumen diameter (Mulvany and Nyborg, 1980). This diameter is calculated following a mounting procedure, which includes the technique described, by Mulvany and Halpern and involving stepwise stretches of the vessel until the passive tension produced exceeds the physiological transmural pressure (Mulvany and Halpern, 1977). Figure 1.9 illustrates typical set up of vessel in a myograph bath.

Figure 1.9: Diagram depicting vessel arrangement in a myograph bath

A diagram showing a wire carefully threaded through a blood vessel. The force transducer is responsible for measuring tension of vessel while the micrometer is responsible for stepwise stretches of the vessel during normalisation. The vessel is in a physiological solution and kept at 37 °C during the experiments by a built-in heater to mimic human condition (Spiers and Padmanabhan, 2005). (Spiers and Padmanabhan, 2005)

1.6 Vascular Dysfunction in PE

1.6.1 Altered Vascular Reactivity of Maternal and Fetal Vessels in Human PE Functional studies have reported mixed findings regarding alterations in vascular reactivity in vessels from pregnancies complicated by PE versus those in normal pregnancy. Omental arteries (OAs) can be used as representatives of the maternal systemic vasculature; omental adipose tissue is harvested from a fatty layer called the greater omentum in the abdomen and artery diameters range from 100 µM-600 µM (Dong et al., 2015). OAs from PE pregnancies (without FGR) demonstrate increased contractility to potassium chloride and vasopressin compared to those from non-pregnant women and women with normotensive pregnancies (Pascoal et al., 1998). Wareing et al also demonstrated significantly increased vasoconstriction to and sensitivity to vasopressin in OAs from women with PE (early and late-onset PE with FGR) compared to those from women with normal pregnancy (Wareing et al., 2006b). Mishra et al. demonstrated that OAs from PE pregnancies show enhanced vascular reactivity to angiotensin II compared to those from normotensive pregnancies (Mishra et al., 2011). Similarly, Aalkjaer et al. showed increased responsiveness of PE OAs to angiotensin II but showed no difference in contraction to noradrenaline between PE and normotensive women (Aalkjaer et al., 1985).

Other studies using omental and subcutaneous fat resistance arteries have demonstrated no difference in vasoconstriction between PE and normotensive women, involving the vasoconstrictors endothelin-1 and U46619 but variable responses to noradrenaline with OAs from PE women being more sensitive compared to normotensive vessels (Belfort et al., 1996; Knock and Poston, 1996; Vedernikov et al., 1995). Studies using myometrial arteries have also demonstrated this trend where myometrial artery constriction to endothelin-1 from PE women has been similar to that of normotensive women (Walsh, 1985; Wolff et al., 1996). There is a single study that observed a decrease in constriction, Vedernikov et al. demonstrated that omental arteries from PE women show reduced constriction to KCL relative to normotensive women (Vedernikov et al., 1999).

Similar conflicting observations have been shown in studies investigating endothelium-dependent relaxation in vessels from PE women. Reduced endothelial- dependent relaxation to acetylcholine has been shown in OAs from PE women however, relaxation to bradykinin was similar between PE and normotensive women (Pascoal et al.,

1998). Attenuated bradykinin-induced relaxation has been observed in U46619 pre- constricted myometrial vessels from PE and FGR pregnancies (Ong et al., 2003). Additional studies using myometrial arteries have shown impaired endothelium-dependent relaxation (Ashworth et al., 1997; Kublickiene et al., 1998); as well as in subcutaneous resistance fat arteries (Knock and Poston, 1996; Mccarthy et al., 1993). There is one study, employing forearm cutaneous microvessels and iontophoretic application of acetylcholine, which did not demonstrate a difference in endothelium-dependent relaxation between normotensive and PE women (Eneroth-Grimfors et al., 1993). In the study by McCarthy et al. there was no difference identified in endothelium-independent relaxation to SNP between PE and normotensive groups.

The reasons for the varied observations in constriction and endothelium-dependent relaxation in PE versus normal pregnancy are many including the choice of method to assess vascular function- either wire myography, pressure myography or iontophoretic acetylcholine application (a transdermal delivery method), which could influence results. The vascular bed of choice could be an influencer, in that vessels from the uteroplacental circulation may respond differently to systemic vessels due to there being a receptor up- regulation or down-regulation in one vascular bed and not the other. The type of PE these women had was not specified in many studies i.e. early or late-onset, which could have influenced results. Additionally, the diameter of a vessel is a point to consider where some studies quoted diameters of 200-250 µM while others quoted 500 µM. Moreover, differences could be agonist-specific for example; some studies reported impaired relaxation to acetylcholine but not bradykinin, suggesting a defect in a G protein-mediated signal transduction pathway that is pertinent to effectiveness of that agonist. Finally, it is uncertain how the length of time a woman has had endothelial dysfunction could affect the reactivity of her vessels i.e. nulliparous women vs women with a history of PE, vascular responsiveness could be different.

Myography studies have also been performed on human placental chorionic plate arteries (CPAs) as a proxy for fetoplacental function. Functional studies have demonstrated conflicting results regarding aberrant dysfunction of CPAs from women with PE compared to CPAs from normotensive women. Ong et al. did not see a difference in constriction in CPAs between normal and PE pregnancy when CPAs were constricted using vasopressin (Ong et al., 2002). Similar to other groups PE was defined as hypertension (140/90 mmHg)

and proteinuria (> 300 mg/L in a 24-hr collection) after 20 weeks however, their samples were collected from women who had PE with and without FGR. In contrast to Ong et al., Wareing et al demonstrated a reduced maximal contraction in CPAs from pregnancies complicated by either PE or FGR compared with normal pregnancy when these vessels were constricted with U46619 (Wareing and Baker, 2004). They used the ISSHP definition of PE and collected their samples mainly at term. In a study by Benoit et al., CPAs from placentas in PE (without FGR) demonstrated significantly greater contraction to 100 mM KCL versus CPAs from normal placentas using pressure myography (Benoit et al., 2007). Bertrand et al. also demonstrated that CPAs from PE pregnancies developed greater active tension than CPAs from normal pregnancies but they used KCL and serotonin to constrict vessels (Bertrand et al., 1993). They defined PE as any delivery between 28-39 weeks after sustained hypertension and proteinuria, which is slightly vague as the key characteristic of PE is “new-onset hypertension after 20 weeks”.

Finally, in CPAs from PE pregnancies it has been observed that there is blunted vasodilation to SNP and the phosphodiesterase-inhibitor papaverine (Ong et al., 2002). In contrast to this, in FGR pregnancies, CPAs show pronounced SNP-mediated vasodilation, which is considered a compensatory response to the vascular resistance observed (Mills et al., 2005). In this study FGR was predicted fetal abdominal circumference <10th centile for gestational age.

The likely reasons for the differences in vascular reactivity in CPAs from PE and normotensive women between different studies are many. Firstly, BMI of women was not disclosed for normotensive and PE women introducing a potential confounder. Furthermore, in some studies, it was unclear whether the PE pregnancies were complicated by FGR, and whether PE cases were early or late onset, as these are likely to present with differing phenotypes. Finally, it is important to acknowledge the fact that these women were on other medications such as aspirin and antihypertensives; it is uncertain how this could have altered placental vascular tone.

1.6.2 Role of NO in normal pregnancy and in PE Nitric oxide (NO) is a soluble gas produced by 3 isoforms of nitric oxide synthase (NOS), neuronal NOS (nNOS), inducible NOS (iNOS) and endothelial NOS (eNOS). All 3 enzymes synthesise NO from L-arginine (Tousoulis et al., 2012). nNOS plays an important physiological role in learning, memory and neurogenesis (Zhou and Zhu, 2009). In the CNS, 52

NO from nNOS is involved in the regulation of synaptic transmission i.e. long-term potentiation and long-term inhibition (Izumi et al., 1992; O'dell et al., 1991). iNOS is induced in macrophages. Furthermore, NO can cause strand breaks and fragmentation by interacting with the DNA of target cells (Fehsel et al., 1993; Wink et al., 1991). These effects of NO from iNOS give it a cytotoxic role in parasitic microorganisms and certain tumour cells. Finally, eNOS promotes endothelial homeostasis and regulates essential cardiovascular functions. NO production via eNOS stimulates vasodilation of vessels by activating guanylyl cyclase and raising cyclic GMP in smooth muscle cells (Ignarro et al., 1986; Rapoport et al., 1983). eNOS is also important for regulating the expression of genes responsible for atherogenesis; NO decreases the chemoattractant protein MCP-1 (Zeiher et al., 1995). Additionally, NO can modulate leucocyte adhesion by either preventing binding ability of adhesion molecules CD11/CD18 or suppressing their expression on leucocytes (Arndt et al., 1993; Kubes et al., 1991). During a normal healthy pregnancy, NO is responsible for placental vasculogenesis and stimulates the differentiation of embryonic stem cells to the endothelial cell lineage via VEGF (Frank et al., 1999; Shizukuda et al., 1999). The vascular tone of the placenta is regulated by various vasoactivators with NO playing a crucial role (Sen et al., 2013). NO promotes vascular remodelling in the uterine circulation in preparation for pregnancy (Osol et al., 2009). Studies have shown that there is a positive correlation between raised serum NO production and increasing gestational age, and these NO levels return to nonpregnant concentrations 12 weeks after delivery (Choi et al., 2002; Hodzic et al., 2017; Jo et al., 1998; Shaamash et al., 2000). This increase in NO is believed to be important in the hemodynamic changes associated with normal pregnancies. Furthermore, NO attenuates responsiveness to vasoconstrictors (Soma-Pillay et al., 2016). A suggestion for increased NO is shear stress, which is capable of activating eNOS. In the placenta, this enzyme is able to maintain a low vascular resistance (Sprague et al., 2010); it exerts its vasodilatory effects through action of NO on soluble guanylate cyclase (Sand et al., 2002) and interaction with potassium channel (BKCa) (Sand et al., 2006). The vasodilator nitric oxide (NO) is important in pregnancy. In humans, there is a 30% decrease in the circulating asymmetrical dimethylarginine, which is an inhibitor of NO synthase (NOS) (Holden et al., 1998). In addition, when NOS is inhibited, a decrease in blood flow was observed in the forearm of gravid vs pre-gravid women (Anumba et al., 1999). This implies that with an increased bioavailability of NO, there is a reduction in peripheral

vascular resistance during pregnancy in humans. Likewise, in mice, nitrite and nitrate levels are increased in plasma and urine during pregnancy, suggesting a role of NO (Conrad et al., 1993; Sladek et al., 1997). Furthermore, when rats are treated with NOS inhibitors, the normal decrease in arterial blood pressure early in pregnancy is obstructed (Molnar et al., 1994; Zhang and Kaufman, 2000), thus preventing the rise in plasma volume and causing FGR and PE-like changes in the mother. In PE pregnancies, there have been conflicting findings regarding the levels of NO. It has been reported that there is no difference in maternal serum nitrite (NO metabolite) concentrations between PE and normotensive women (Curtis et al., 1995; Davidge et al., 1996; Lyall et al., 1995). However, other groups have demonstrated a rise in plasma or urinary nitrate levels in PE (Cameron et al., 1993; Nobunaga et al., 1996; Norris et al., 1999; Smarason et al., 1997), supposedly due to an adaptive response to the increased vascular resistance (Lyall et al., 1996; Myatt et al., 1997; Norris et al., 1999; Shaamash et al., 2001). In contrast, other groups have shown a significant decrease in NO concentrations and placental NO synthase activity in PE (Brennecke et al., 1997; Choi et al., 2002; Garmendia et al., 1997; Seligman et al., 1994), determined by conversion of [3H]L-arginine into [3H]L- citrulline or nitrate/nitrate concentrations. Studies have shown that in PE, there is a normal production of NO however, the bioavailability of the vasodilator is limited (Lopez-Jaramillo et al., 2008). It is thought that a potential reason for the reduced bioavailability of NO is the high production of oxygen free radicals such as superoxides, which scavenge NO (Sankaralingam et al., 2006). Another mechanism implicated in reduced NO production is the arginase isoform arginase II, whose mRNA is overexpressed in villous tissue (4-fold) in PE women compared to normal pregnant women (Noris et al., 2004). When arginase II is overexpressed, it can result in reduced availability of L-arginine, which is a key substrate for eNOS leading to uncoupling of eNOS and the production of superoxide, which scavenges NO to generate peroxynitrite (Kuzkaya et al., 2003; Yzydorczyk et al., 2013). Giles et al. demonstrated that NOS activity was significantly lower in placentas of women with abnormal umbilical artery flow velocity waveforms compared to placentas from women with a normal umbilical artery flow velocity waveform (Giles et al., 1997). The conflicting literature regarding NO levels in PE therefore remains controversial. Whilst methods to measure NO have developed over the years, problems still exist because NO is reactive and short lived. Furthermore, NO is sensitive to its environment (i.e., solution or sample) in which the measurement takes

place. NO can readily react with free radical species, metal-containing proteins (e.g., haemoglobin), and some methods are restricted by their ‘limit of detection’ (Hunter et al., 2013). This can lead to NO measurement results not always being reliable or reproducible. As mentioned previously, in PE proangiogenic markers VEGF and PlGF are drastically reduced which can affect endothelial NO release, and this results in vasoconstrictor imbalance and increased peripheral resistance and blood pressure (S Jurado, 2019). These changes in NO signalling and vascular reactivity (reduced vasodilation), contribute to increased uterine artery vascular resistance (Osol et al., 2009; Phipps et al., 2016).

1.7 Animal Models of Pregnancy Animals are useful pre-clinical tools to model pregnancy, thus aiding understanding of fundamental developmental processes in health and disease, and in which to test candidate therapies for their effectiveness at improving outcomes of pregnancy. Whilst there is no single perfect animal model for pregnancy, the translatability to humans based on the outcomes to be studied (e.g. placental dysfunction, cardiovascular adaptations to pregnancy) should be a key consideration driving this choice (Maria Dahl Andersen and Michael, 2018).

Non-human primates are closely related to humans and have been shown to have similar uterine and placental anatomy, gestational length and generally harbour singleton pregnancies (Grigsby, 2016). The testing of therapeutic drugs and the investigation of placental dysfunction have been explored in macaques and baboons (Krugner-Higby et al., 2009; Sunderland et al., 2011). Nevertheless, there are financial and ethical implications of using these animals and studies are restricted to specialist facilities but also there are only a limited number of non-human primates that can develop PE. Spontaneous PE has only been reported in baboon twins (Hennessy et al., 1997) and in the patas monkey, however not all of the current criteria for PE were assessed in these primates (Gille et al., 1977; Palmer et al., 1979). The sheep has also been shown to be amenable to surgical manipulation to further study the maternal-fetal interface and placental dysfunction. However, differences have been shown in the endocrinology of parturition relative to humans and the placental structure is very different to that of humans, being epitheliochorial compared to the haemochorial placenta seen in humans (Challis et al., 2000). The guinea pig is a well-established rodent model of pregnancy and has been commonly used to study placental transfer (Jansson and Persson, 1990). FGR can be 55

induced in guinea pigs by unilateral uterine artery ligation on day 30 of gestation, causing placental insufficiency and brain sparing in FGR fetuses (Carter et al., 2005; Lafeber et al., 1984). The guinea pig shares similar placentation with humans, both of which are haemochorial, with trophoblast being directly bathed in maternal blood. Finally, smaller rodents such as mice and rats also demonstrate haemochorial placentation. In addition, rodents can be genetically manipulated and practically, they are affordable and have a short-gestational length enabling quick data extraction (Grigsby, 2016). Mouse models of pregnancy are an important consideration for this thesis and will be discussed in more detail later. Figure 1.10 compares the type of placentation between different animals.

Figure 1.10: Compares the placenta barrier between different animal models of pregnancy

A) Hemomonochorial placental barrier B) Hemodichorial placenta barrier C) Hemotrichorial placental barrier and D) Epitheliochorial placenta barrier. Mono-, di- and trichorical refers to the number of trophoblast barriers between maternal and fetal circulations, while the term ‘epitheliochorial’ refers to a uterine epithelial cell directly next to a cytotrophoblast cell (Anderson et al 2018). 56

1.7.1 Animal Models of Pre-eclampsia Ideally, a model of PE would exhibit all of the relevant characteristics of PE i.e. new-onset hypertension, proteinuria, endothelial dysfunction, +/- FGR, all of which arise often as a likely consequence of inadequate trophoblast invasion and thus placental dysfunction. Nevertheless, due to the fact that PE does not develop spontaneously in animals other than primates and due to differences in placentation between rodents and humans, a model that recapitulates all facets of PE is difficult to achieve (Mccarthy et al., 2011).

Various models have been implemented in an attempt to recreate the pathogenic mechanisms seen in PE. The “reduced uterine perfusion pressure model” (RUPP) being one example. In this model, clips are placed on the aorta above the iliac bifurcation and right and left uterine ovarian artery from gestational day 14-21 in the rat, which reduces uteroplacental perfusion by 40% and mimics a hypoxic challenge (Granger et al., 2006). In this model, blood pressure is elevated and glomerular filtration rate reduced (Llinas et al., 2002). Also evident is increased peripheral resistance, endothelial dysfunction (Walsh et al., 2009) and fetal growth restriction, thus this model closely resembles severe early-onset PE (Mccarthy et al., 2011). However, the RUPP model represents a very severe model and results in early fetal loss and reduced litter size; at GD21 in the Sprague-Dawley rats the mean litter size is significantly less in RUPP group compared to SHAM group (5 vs 14 respectively) (Fushima et al., 2016; Li et al., 2012; Morton et al., 2019). Takimoto et al. mated transgenic female mice expressing angiotensinogen with transgenic males expressing renin. When these females became pregnant, they demonstrated increased blood pressure in late gestation, due to secretion of placental human renin into the maternal circulation, as well as proteinuria and myocardial hypertrophy (Takimoto et al., 1996). Kanasaki et al. reported that when catechol-O-methyltransferase (an enzyme involved in degrading catecholamines i.e. epinephrine (Phillips et al., 2016) is knocked out (COMT-/-), pregnant mice had increased blood pressure during pregnancy versus WT mice, increased urinary protein excretion and COMT-/- placentas showed acute atherosis (Kanasaki et al., 2008). Stanley et al. has also reported FGR in the COMT-/- mice (Stanley et al., 2012b). A ‘borderline hypertensive’ PE model has been reported using the mouse strain (BPH/5) which involved the inbreeding of these mice over many generations, resulting in mild blood pressure elevation throughout adult life, although aetiology of BP elevation is

unknown (Schlager G, 1989). This model demonstrated pre-pregnancy hypertension, proteinuria, reduced litter size, glomerulosclerosis, endothelial dysfunction and abnormal placental development (Davisson et al., 2002).

Pharmacological inductions have been used to create PE models through the chronic inhibition of nitric oxide enzymes using inhibitors such as nitro-L-Arg-methyl-ester (L-NAME), resulting in hypertension in pregnancy, proteinuria, reduced glomerular filtration rate and FGR. This model, along with the eNOS-/- mouse, described in the next section, have been useful in demonstrating the role of NO bioavailability in contributing to some of the clinical signs of PE (Molnar et al., 1994). An altered angiogenic profile has been implicated in PE and Maynard et al. demonstrated that when the anti-angiogenic factor s- Flt-1 is administered to pregnant rats via adenovirus-mediated gene therapy, this resulted in gestational hypertension, proteinuria, glomerular endotheliosis and a reduction in VEGF and PlGF (Maynard et al., 2003). Other PE models have been created through excessive maternal immune response (TNF-α infusion), (Lamarca et al., 2005), pre-pregnancy administration of adriamycin (Podjarny et al., 1992) and induction of chronic hyperinsulinaemia (Podjarny et al., 1998). Additionally, the STOX1 model (transgenic model created by fetoplacental overexpression of STOX1) exhibits hypertension, proteinuria, an increased plasma level of soluble anti-angiogenic factors, as well as altered kidney and placental morphology (Doridot et al., 2013). Another PE-like model includes the sENG rat model (created by intravenous injection of adenovirus carrying sENG) which exhibits increased blood pressure and proteinuria, although not to an equivalent severity as the sFlt-1 overexpression model (Venkatesha et al., 2006).

1.7.2 Mouse vs human The mouse shares common features to women in terms of their cardiovascular adaptations during pregnancy. Kulandavelu et al. showed using a C57Bl/6 (WT) mouse that increased cardiac output (CO), stroke volume, plasma volume as well as decreases in arterial pressure and haematocrit are all observed early in pregnancy (Kulandavelu et al., 2006). A study by Wong et al. also demonstrated that in early mouse gestation there is a reduction in arterial blood pressure and later on in gestation, CO increases and sensitivity to the vasoconstrictor angiotensin reduces (Wong et al., 2002). This data is in line with cardiovascular adaptations in humans. Furthermore, work by Cooke et al. has shown that during mouse pregnancy, NO and prostaglandin dependent pathways are important mediators of endothelium-

dependent vasodilation in the uterine and mesenteric arteries (Cooke and Davidge, 2003). These similar features to those found in women make mice useful models for pregnancy and they can be manipulated to display features akin to obstetrics conditions such as FGR, making mice a useful model for testing efficacy of drugs for obstetric complications. Mouse placentation has both similarities and differences to human. In mice, there is trophoblast invasion into the uterine wall following implantation (as seen in humans) hence promoting angiogenesis and vasodilation (Cross et al., 2002). However, this invasion process is shallower than in humans and the mouse invasive trophoblast subtype which enters the uterine spiral arteries are trophoblast giant cells, thought to be analogous to the human extravillous cytotrophoblast cells in humans (Burke and Karumanchi, 2013). Also, in mice/rats the placenta is supplied by only 2-3 spiral arteries whereas in non-human primates and humans there are ~100-150 in placental bed (Pijnenborg et al., 2011). In both mice and humans, there is hemochorial blood flow through the placenta, thus maternal blood is in direct contact with the syncytiotrophoblast (Furukawa et al., 2014). The mouse placenta is fully established by ~E10.5 (mid-gestation) and consists of three main compartments: the labyrinth zone, junctional zone (comprising of spongiotrophoblasts, glycogen cells and trophoblast giant cells) and the maternal decidua (Woods et al., 2018). This is not the case for placentas from humans, which do not have a zonal architecture. On a practical level, mice are relatively cost-effective, low maintenance and have a short gestation time, enabling rapid accumulation of data. A limitation of mice is multiparity; mice tend to have large litter sizes, which can make it difficult to measure and follow individual fetal and placental progress during gestation (Andersen et al., 2018). Additionally, comparing gestational periods between human and mouse, particularly near term, is complex. Mouse fetuses at birth are relatively underdeveloped compared to humans and a number of key processes e.g. organogenesis such as in the kidney, continues postnatally in mice. However, despite these differences, there is a plethora of data to suggest that mice remain a useful model in which to aid understanding of the placenta in normal and pathological situations, and for comparison with data in samples from women. Table 1.1 below provides gestational information for various commonly used animal models of pregnancy.

Table 1.1: gestational information for various commonly used animal models of pregnancy (Grigsby, 2016).

1.7.3 eNOS knockout mice Mice homozygous for deletion of eNOS (eNOS-/-) are viable and fertile and demonstrate reduced body weight compared to wildtype mice. eNOS-/- mice show pre-pregnancy hypertension (systolic and diastolic blood pressure increased by ~13 and 16 mmHg respectively) which persists during pregnancy (systolic and diastolic increased by ~15 and 18 mmHg) relative to wildtype mice (Cabrales et al., 2005; Duplain et al., 2001; Hefler et al., 2001b; Huang et al., 1995; Ortiz and Garvin, 2003; Shesely et al., 1996). eNOS-/- mice also demonstrate proteinuria, uterine artery dysfunction, FGR, reduced placental amino acid transport and placental hypoxia (Kulandavelu et al., 2012; Kusinski et al., 2012). In regards to endothelial dysfunction, Kusinski et al. demonstrated that uterine arteries from eNOS-/- mice show increased phenylephrine-induced constriction and blunted endothelium-dependent relaxation to acetylcholine compared to wildtype (WT) mice.

In the eNOS-/- mouse, CO is reduced in late gestation, which is associated with a lower stroke volume and an increased heart rate. A lower stroke volume was attributed to inadequate ventricular remodelling (Kulandavelu et al., 2006). Additionally, Kulandavelu et al. demonstrated that eNOS-/- mice exhibit reduced uteroplacental blood flow, uterine

artery diameter, spiral artery length and an increase in markers of placental hypoxia relative to normal pregnancy (Kulandavelu et al., 2012). Furthermore, eNOS−/− fetuses, which demonstrate FGR, demonstrate reduced abdominal circumference (Kusinski et al., 2012). There are differences in descriptions of placental weight in eNOS-/- mice. At E17, placentas from eNOS-/- have been shown to be 10% lighter than WT mice (Hefler et al., 2001a; Van Der Heijden et al., 2005). However, at E18.5 a significant increase in placental weight in eNOS-/- mice has been reported (Renshall et al., 2018) whilst some studies report no difference (Finn-Sell et al., 2018; Kulandavelu et al., 2012; Kusinski et al., 2012) compared with WT mice. Studies have consistently reported a reduction in the fetal: placental weight ratio, suggesting a reduced placental efficiency in eNOS-/- mice (Finn-Sell et al., 2018; Kusinski et al., 2012; Poudel et al., 2013; Renshall et al., 2018; Stanley et al., 2012a), which is supported by the previously mentioned reduction in system A amino acid transport in placentas of eNOS knockout mice (Kusinski et al., 2012). As this model does not present with new-onset hypertension as seen in PE, it is more appropriately described as a model for chronic hypertension or superimposed PE that also demonstrates FGR.

1.8 Clinical Therapeutics for Pre-eclampsia As mentioned previously, there are no current therapies for PE nor any novel drugs being developed for the treatment of PE. Rather, increased monitoring and antihypertensive medication is routinely used to help to manage the pregnancy and prolong gestation. Drugs used to treat conditions with overlapping phenotypes to PE/other hypertensive disorders of pregnancy (e.g. vascular dysfunction) are candidates for being repurposed for prevention/treatment of PE as described below.

1.8.1 Proposed Treatments for Pre-eclampsia

1.8.1.1 Pre-Clinical Studies As previous literature has presented associations between PE and reduced NO bioavailability, various therapeutics that enhance NO bioavailability have been explored, including the drug sildenafil citrate (SC). SC is a potent vasodilator via inhibition of phosphodiesterase-5, which increases NO bioavailability (Coppage et al., 2005). Work by Wareing et al. demonstrated that SC enhanced endothelial function of myometrial vessels from women diagnosed with PE (Wareing et al., 2004). In an L-NAME rat model of PE (Herraiz et al., 2012), SC (4mg/kg/day; E0-time of delivery) maintained systolic blood

pressure at similar levels to the control group, there was also improvements in proteinuria and uteroplacental/fetal perfusion post-treatment with SC and these effects could be NO- dependent. Prior to this, Ramesar et al. showed that SC administration to PE rats induced with L-NAME; significantly reduced plasma concentration of sFlt-1 and sEng (Ramesar et al., 2011). Finally, Stanley et al. demonstrated in the COMT-/- mouse modelling PE and FGR that SC (0.2 mg/mL) rescued pup growth and improved abnormal umbilical Doppler waveforms (Stanley et al., 2012b) as well as normalized the aberrant metabolomic profile in COMT−/− mice (Stanley et al., 2015).

In a study by Onda et al. the proton pump inhibitor esomeprazole has been shown in-vitro to reduce sFlt-1 and sEng secretion from trophoblast cells, placental explants and endothelial cells from PE pregnancies. Additionally, esomeprazole reduced the secretion of endothelin-1 and promoted endothelial cell migration. Ex vivo, cumulative administration of esomeprazole (1–300 μmol/L) has been shown to cause vasodilation of OAs from both normal and preterm PE pregnancies and in an in vivo PE-like sFlt-1 mouse model, it reduced maternal blood pressure (Onda et al., 2017). Another drug that has been explored for PE treatment is metformin, typically used to treat type 2 diabetes. Metformin decreases levels of sFlt-1 and sEng in vitro in endothelial cells and placental villous explants. Additionally, metformin reduced endothelial dysfunction, enhanced bradykinin-mediated vasodilation in OAs and promoted angiogenesis in OA explants (Brownfoot et al., 2016a). Nevertheless, a prospective clinical trial for metformin use in preventing PE is still required.

Co-administration of an adenoviral vector containing sFlt-1 with VEGF delivered via injection into tail vein was shown to normalise blood pressure (to control levels), caused a 70% reduction in free sFlt-1 in plasma and rescued renal function in a sFlt-1 mouse model (Bergmann et al., 2010). In the BPH/5 mouse model of PE, VEGF therapy was able to prevent the late-gestational spike in blood pressure and reduce proteinuria but had no effect on placental or fetal weights (Woods et al., 2011). Additionally, recombinant human PlGF has been shown to reduce circulating levels of free sFlt-1 and oxidative stress (evidenced by reduced expression of marker 8-isoprostane) in the RUPP rat model compared to vehicle group (Spradley et al., 2016).

1.8.1.2 Clinical Trials for the treatment of PE There have been a number of clinical trials aimed at improving outcomes in PE. In a randomised double-blind, placebo-controlled trial involving PE women between 24-33 62

weeks of gestation, when 50 mg SC was administered every 8 hrs in their 2nd trimester, the treatment group experienced a 4 day pregnancy prolongation compared to women given a placebo (Trapani et al., 2016). However, Samangaya et al. did not support the use of SC for PE as it offered no clinical benefit in a small randomised double-blinded control trial. However, there were suggestions that SC did not achieve efficacious concentrations rapidly enough to slow disease progression in this particular trial, which contained only 35 women (Samangaya et al., 2009). Finally, an open pilot study was conducted to assess the therapeutic effect of dextran sulphate cellulose column apheresis for very preterm PE women. Levels of sFlt-1 (by 18%) were reduced in these women and this was associated with reduced levels of proteinuria (Thadhani et al., 2016). In women given just a single treatment of dextran sulphate, pregnancy was prolonged by 8 days post diagnosis and 15 days in women treated multiple times compared to untreated PE controls. This was in the absence of any reported adverse maternal or fetal effects (Thadhani et al., 2016).

1.8.1.3 Clinical Trials for Prevention of PE Candidates have been proposed for the prevention of PE, which include vitamins C and E due to their anti-oxidant ability. This led to the VIP trial, a randomised, placebo-controlled trial, where women (with increased PE risk) were assigned 1000mg of Vitamin C and 400 IU of Vitamin E or a matched placebo daily, during the second trimester of pregnancy until delivery. The results however showed an increased incidence of low birth weight and no therapeutic benefit with respect to PE risk, indicating that at these doses, vitamins C and E should be contraindicated in pregnancy (Poston et al., 2006).

1.8.2 Statins and Known Mechanisms of Action It has been suggested that PE shares a similar pathophysiology to cardiovascular disease (CVD), showing similar features such as endothelial dysfunction and inflammation. In the case of PE, the spiral arteries are inadequately transformed whereas in essential hypertension the arteries and arterioles undergo “inner eutrophic vascular remodelling”. This describes an increase in the media: lumen ratio and decrease in the lumen diameter in resistance vessels of patients with hypertension without any change in the amount of wall material (Renna et al., 2013). In secondary hypertension, for example due to suboptimal kidney function, hypotrophic remodelling can occur resulting in a decrease in the lumen diameter of renal afferent arterioles as well as a reduction in the wall material of arterioles (Renna et al., 2013). Interestingly, PE is also associated with an inflammatory

response and alongside the well-known hypertensive and proteinuric complications. PE has been linked to insulin resistance and a worsened lipid profile; for this reason it can sometimes be referred to as the metabolic syndrome of pregnancy due to its multisystemic effects (Chen et al., 2014).

Similarly, in CVD frequent deposition of plaques onto the walls of arteries such as the coronary or carotid artery can result in wall stiffening and vessel narrowing leading to atherosclerosis complications such as heart attacks, stroke, chest pain and aneurysms. Furthermore, this can lead to a plethora of risk factors, which include but are not limited to hypertension, dyslipidaemia, diabetes mellitus and metabolic syndrome (Zafar, 2015). The similarities in the metabolic alterations observed in PE and cardiovascular disease suggest that a common pathophysiology may be present; oxidative stress is postulated to be important in the endothelial dysfunction seen in atherosclerosis. Similarly, it has been suggested that the secondary effect of reduced placental perfusion in PE is oxidative stress, resulting in endothelial dysfunction (Roberts, 2000).

Due to the shared pathology that underpins PE and CVD, it has been suggested that drugs used to treat CVD could be translated into use in PE. Drugs used to manage or prevent CVD include statins, which have been shown to have fast-acting pleiotropic effects such as reducing oxidative stress and inflammation and protecting endothelial function (Shaw et al., 2009) as well as offering primary and secondary prevention of CVD (Brugts et al., 2009). For the reasons stated, statins provide biological plausibility for potential use in the management or prevention of PE but also potential protection of vascular health long- term.

The development of coronary atherosclerosis has been associated with a high concentration of serum cholesterol (Larosa et al., 1990). Statins exert their cholesterol lowering effects by reversibly but competitively inhibiting the enzyme 3-hydroxyl- 3methylglutaryl Co-enzyme A (HMG-CoA) reductase, which is an enzyme responsible for converting HMG-CoA into mevalonate, a cholesterol precursor. Once bound to the active site of the enzyme, statins can alter the conformation leading to the functional structure being lost (Stancu and Sima, 2001). Consequently, statins can drastically reduce the levels of low-density lipoprotein (LDL), more so than dietary alteration (Hunninghake, 1992; Illingworth et al., 1994; Todd and Goa, 1990). This inhibition results in the sequestration of LDL, a moderate increase in LDL receptors, inhibition of LDL oxidation and inhibition of the 64

secretion of lipoproteins (Bellosta et al., 2000). This has placed statins as a first-line treatment for both primary and secondary coronary heart disease (Almuti et al., 2006). Figure 1.11 illustrates cholesterol-dependent mechanism of statins while table 1.2 provides pharmacokinetic details of different statins. Atorvastatin, fluvastatin and pravastatin are administered in their active form as acids, whereas lovastatin and simvastatin are administered in their inactive forms as lactones, which require hydrolyzation into their active form (Blumenthal, 2000). Pitavastatin is converted to its lactone form, which is its major metabolite, by a process called glucoronidation (Fujino et al., 2003).

HMG-Co-A HMG CoA Reductase

Mevalonate STATINS

Effects:-

 Increase in LDL receptors

 Increase in HDL Cholesterol synthesis  Reduction in circulating LDL

Figure 1.11: An illustration of how statins interact with the Mevalonate pathway. LDL (low-density lipoprotein) and HDL (high-density lipoprotein).

Table 1.2 Pharmacokinetics properties of statins (Ramkumar et al., 2016; Zhou et al., 2013)

Pitavastatin Atorvastatin Rosuvastatin Simvastatin Pravastatin Fluvastatin

Half-life (hrs) 12 15-30 19 4.85 1.3-2.8 0.5-2.3

Bioavailabilit 60 12 20 5 18 19-29 y (%)

Protein >99 80-90 88 94-98 43-55 >99 binding (%)

Solubility 0.00043 Practically 0.01796 7.17e-11 0.00607 0.00046 (g/L) insoluble

Lipophilicity Moderately Hydrophobic Hydrophilic Hydrophobic Hydrophilic Hydrophobic hydrophobic

Clinical doses 2-4 10-80 5-40 5-80 10-80 20-80 (mg)

1.8.3 Cholesterol levels in normotensive and PE pregnancies

In normal pregnancy, lipid levels in the plasma increase (Ogura et al., 2002), due to a change in lipid metabolism to support the nutritional demands of the growing fetus (Wild et al., 2015). It can split into two phases, the initial phases involves the accumulation of fatty acids, leading to triglycerides being deposited in the maternal adipose tissue, while the late phase involves a breakdown of adipose tissue enabling increased release of fatty acids and glycerol into the circulation, resulting in hypertriglyceridemia (Emet et al., 2013. Furthermore as estrogen levels increases during gestation, there is an increase in very low- density lipoproteins hepatic synthesis as well as the catabolism of triglyceride rich lipoproteins due to a decrease in lipoprotein lipase and hepatic lipase activities (Ghio et al., 2011) and (Nelson et al., 2010). With this in mind, these changes can subsequently result in an atherogenic phenotype characterised by small and dense LDL particles, a rise of triglycerides levels and a decline in HDL (Ghio et al., 2011). This state of hyperlipidaemia has been found to be associated with complications such as PE, where there is a greater elevation in plasma lipid (Roberts et al., 2003; Gallos et al., 2013). Although some studies have reported no change (Dekker et al., 1998; Enquobahrie et al., 2004; Gallos et al., 2013) and results remain inconsistent (Bayhan et al., 2005; Iftikhar et al., 2010 and Sahu et al.,

2009). Limitations of these studies have been small sample sizes, and the gestational age at the time of the lipid measurements being varied, making it difficult to compare findings across studies (Spracklen et al., 2014).

This atherogenic lipid profile are risk factors for the initiation and progression of atherosclerosis, subsequently leading to cardiovascular disease (Nelson et al., 2010 and Sattar et al., 2002). In PE, atherosis in the spiral arteries of the placenta has been identified in 20-40% of women in PE (Staff et al., 2013; Staff et al., 2014).For this reason, these women represent a subpopulation of PE women who could potentially benefit from statin treatment in pregnancy to not only manage dyslipidaemia but to prevent future CVD risks. Furthermore, in a safety and efficacy trial using pravastatin, it was shown that maternal total cholesterol (TC) and LDL concentrations were lower in pravastatin-treated group vs placebo group in the second trimester and in the third trimester (TC 201.7±33.5 vs 250±25.3 mg/dL, P=0.02). Nevertheless, cord blood concentrations of TC and LDL were similar between the pravastatin and placebo fetuses (Costantine et al., 2016).

1.8.4 Pleiotropic Effects of Statins Statins are mainly known for their beneficial effects in lowering lipid levels in serum. However, there is increasing evidence to suggest that they exert additional effects independent of this lipid lowering ability. It is thought that these pleiotropic effects may be, at least in part, mediated through eNOS (Liao, 2005; Takemoto and Liao, 2001). Studies involving statins have shown that endothelial function is rescued even before there are significant changes observed in serum cholesterol levels (Anderson et al., 1995; O'Driscoll et al., 1997; Treasure et al., 1995). This suggests that there could be cholesterol- independent methods of preserving endothelial function. Moreover, a short-term exposure of vessels to statins could have the potential to introduce “carry-over” protective effects, which could improve endothelial function in the long-term.

Clinical trials have supported and provided further information about these pleiotropic effects of statins, as they have shown anti-inflammatory effects through the reduction of the biomarker C-reactive protein (Kinlay et al., 2003; Ray et al., 2005). It is also believed that the anti-inflammatory effects are class-dependent, with some statins such as atorvastatin at a high dose having superior effects compared to pravastatin accounting for their differences in effectiveness. This finding is also supported by the results from the MIRACL randomized trial that showed atorvastatin 67

given 80 mg/d, initiated 24 to 96 hours after an acute coronary syndrome, reduced recurrent ischemic events in the first 16 weeks. While the PROVE-IT TIMI study (also in patients with acute coronary syndrome), compared the effectiveness of 40 mg of pravastatin daily (standard therapy) versus 80 mg of atorvastatin daily (intensive therapy). Patients who received atorvastatin were provided with greater protection against death and major cardiovascular events compared to pravastatin (Kinlay et al., 2003; Ray et al., 2005; Schwartz et al., 2001). Furthermore, neutrophils are immune cells associated with the inflammatory process and statins have been shown to prevent the migration of neutrophils across the endothelium as well as chemotaxis. This anti-inflammatory effect of statins is useful as PE is associated with inflammation due to rise in inflammatory cytokines and fall in T-helper cells, which contributes towards endothelial dysfunction.

Statins provide antioxidant effects by reducing free radical production via angiotensin-II in vascular smooth muscle cells. Statins act in this manner by inhibiting Rac1-mediated NADH oxidase activity and causing down-regulation of angiotensin II type 1 receptor expression (Wassmann et al., 2001). Statins can lower the formation of the superoxide anion in endothelial cells, through the prevention of p21 Rac protein prenylation (Wagner et al., 2000). In addition, statins have been shown to preserve the activity of endogenous antioxidant systems such as superoxide dismutase (Chen et al., 1997). All of these promote statins use in clinical conditions associated with elevated levels of oxidative stress such as PE.

Pathways implicated in the pleiotropic effects of statins include the P13/Akt pathway. Statins were shown to increase the activity of eNOS via post translational activation of this pathway (Kureishi et al., 2000). Rossini et al demonstrated that male rat mesenteric arteries, following a 2 hr incubation with simvastatin, demonstrated improved vascular function via an NO-dependent mechanism associated with AMPK and not the highly cited AKT pathway (Rossoni et al., 2011). Statins may therefore affect a number of intracellular second messenger pathways and more research is needed in this area to further delineate the important players in their short-term effects on vascular function. Figure 1.12 highlights major cholesterol-dependent and independent effects of statins.

Lowers LDL

Lipid dependent68 Raises HDL effects Raises LDL receptors Statin

Figure 1.12: Outlines the pleiotropic effects of both lipid-dependent and lipid- independent statins.

1.10 Statins and Their Use in Pregnancy

1.10.1 Risks of Statin Use in Pregnancy

1.10.1.1 Pre-clinical Studies in Animals and Humans Despite the many therapeutic benefits of statins outlined above, in terms of their use in pregnancy, a number of risks have been outlined. Kenis et al. showed that first trimester placental explants exposed to simvastatin in-vitro demonstrated reduced extravillous cytotrophoblast cell migration whilst proliferation of villous cytotrophoblasts was inhibited. Simvastatin also increased apoptosis of cytotrophoblast cells and decreased levels of progesterone (Kenis et al., 2005). Tartakover-Matalon et al. proposed that this simvastatin-induced decrease in cytotrophoblast migration could be attributed to the reduced expression of insulin growth factor (IGF)-1 and Heat shock protein (HSP) 27 as well as a lower activity of matrix metalloproteinase (MMP) 2 (Tartakover-Matalon et al., 2007).

IGF-1 is a key regulator in relation to placental cell turnover and fetal growth (Forbes and Westwood, 2008). Forbes et al. showed, in first trimester placental villous tissue, that statins could interrupt IGF-induced cytotrophoblast proliferation (Forbes et al., 2008). Furthermore, they established that the detrimental effects of statins might not be class specific, as the placenta expresses organic anion-transporting polypeptides, which may enhance statin uptake (Serrano et al., 2007). Based on these results, it was thought that statin use during early pregnancy might lead to poor pregnancy outcomes due to inadequate placental development. More recent data has revealed that the glycosylation, 69

localization and function of IGFR1 in first trimester placental explants treated with statins, and in particular, cerivastatin or pravastatin, resembled explants that had been cultured with glycosylation inhibitors. Specifically, statins are suspected to inhibit N-linked glycosylation and subsequent expression of mature IGF1R at the placental cell surface, which subsequently results in their actions being, diminished (Forbes et al., 2015).

Kane et al., in pregnant ewes maintained under hypoxic conditions, showed that pravastatin attenuated the fetal peripheral vasoconstrictor and lactic acidaemic responses. However, the cardiac chemoreflex and plasma catecholamine were unaffected (Kane et al., 2012). Both fetal peripheral vasoconstriction and lactic acidaemic responses contribute towards the fetal brain sparing response and are key defence mechanisms the fetus utilises to respond to acute hypoxia in-utero during placental insufficiency (Giussani, 2016). When this fetal defence is impaired, the fetal brain can become susceptible to injury resulting in hypoxic-ischaemic encephalopathy (Gunn and Bennet, 2009; Low et al., 1985). Moreover, in the study by Kane et al. post-pravastatin treatment, when an NO-clamp was applied, the fetal peripheral vasoconstrictor response was restored to control levels along with lactic acidaemic responses. This implied that it was the increased bioavailability of NO that resulted in a depressed response to the acute hypoxia. The research concluded that statin use in pregnancy should be viewed with caution (Kane et al., 2012).

1.10.1.2 Clinical Studies In 2004 Edison and Muenke investigated 52/178 cases of statin exposure during the first trimester of pregnancy and discovered that in 20/52 cases, fetal abnormalities were detected, including those affecting the CNS, hydrocephaly, two failures of neural tube closure and 5 cases of limb abnormalities (Edison and Muenke, 2004). However, a prospective, observational cohort study with a comparison group found that there was no statistical differences between statin exposed and control groups in regards to live births, spontaneous abortions, malformations and stillbirth. The statin group women were exposed to atorvastatin (n=46), simvastatin (n=9), pravastatin (n=6), or rosuvastatin (n=3) during the first trimester. However, in this study, pregnancy duration and birth weight were lower in the statin group (Taguchi et al., 2008).

Zarek et al. conducted a systematic review and meta-analysis on women exposed to statins during their first trimester. It revealed that there was a modest but significant increase in miscarriage risk and women using statins were twice as likely than control 70

subjects to undergo elective terminations of pregnancy. They suggested the miscarriage risk might reflect maternal morbidity rather than statin therapy risk, as women reliant on statin therapy tended to be older and have metabolic conditions that may have predisposed them to miscarriage. In addition, worthy of consideration in these studies is that the elective terminations could have resulted from health conditions not compatible with pregnancy, from pregnancies being unplanned or perceived fetal risk due to statin exposure (Zarek and Koren, 2014).

Results from a multi-centre observational prospective controlled study involved a follow up of pregnancies where the mother used statins during the first trimester and these were compared with a matched control group with 249 participants in each group. There was no statistically significant difference in birth defects detected when the two groups were compared. However, premature births were more commonly reported in statin- exposed pregnancies (16.1% versus 8.5%) (Winterfeld et al., 2013).

In 2015, a cohort study investigated the risk of congenital malformations and organ specific malformations in offspring exposed to statins during the first trimester. It was found that there was no significant increase in organ specific malformations following an assessment accounting for confounders in particular pre-existing diabetes. However, it was decided that before statin use in pregnancy is considered safe, a longitudinal study would be required to assess the long-term effect of in-utero statin exposure and that their findings would need to be replicated in large datasets (Bateman et al., 2015). That being said, with statins being more frequently used amongst women of a reproductive age it is important that data exists to provide more evidence regarding whether statins are safe to use in pregnancy or not, enabling women to make informed decisions that are evidence-based (Godfrey et al., 2012).

For the reasons listed above, the use of statins during pregnancy has been contraindicated by the U.S. Food and Drug Administration and thus statins have been placed in category X. This is because “studies in animals or humans have demonstrated fetal abnormalities or there is evidence of fetal risk based on human experience or both, and the risk of the use of the drug in pregnant women clearly outweighs any possible benefit” (FDA, 1980). Therefore, women prescribed statins are advised to stop taking them if trying to conceive, pregnant or if breastfeeding. Despite these findings questioning the safety of statins in pregnancy, recent reports suggest that statins could indeed have 71

therapeutic benefit in PE, whilst being safe for both mother and baby. This is discussed further in section 1.11. Additionally, researchers are now exploring short-term application of statins at a later stage in pregnancy. In this manner, they are targeting the earlier pleiotropic effects of statins, reducing statin exposure and hence aiming to reduce overall risks to the fetus.

1.10.2 Placental Transport of Statins A key determinant of potential adverse effects of statins on the fetus is the concentration of statin, or breakdown metabolites, that reaches the fetus and will be dependent, in part, upon the chemical properties of the statin. Simvastatin, lovastatin and atorvastatin, which are all lipophilic, have all been implicated in the inhibition of P-glycoprotein in-vitro, unlike pravastatin and fluvastatin (Holtzman et al., 2006). This is of importance because p- glycoprotein is abundantly expressed on the apical membrane of the syncytiotrophoblast and is involved with actively extruding a wide range of xenobiotics across the placenta and protecting the fetus from harmful substances (Iqbal et al., 2012; Staud et al., 2012). Breast Cancer Resistant Protein (BCRP) is also of importance (Yeboah et al., 2006).

More recently, it was shown using ex vivo dual perfusion of the human placenta, that pravastatin is able to reach the fetal circulation. However, the presence of efflux transporters within the placenta also contributes to efflux from the placenta back to the maternal circulation helping to reduce concentrations in fetal blood (Nanovskaya et al., 2013). Pravastatin is a known substrate for multidrug-resistance-associated protein 2 (Yamazaki et al., 1996) and breast cancer resistant protein (Elsby et al., 2011; Matsushima et al., 2005). These transporters within the placenta are thought to restrict drug transfer from the maternal to the fetal circulation.

1.11 Biological Plausibility for Statins Use in Pre-eclampsia

1.11.1 Evidence from Pre-clinical Studies There are a number of pre-clinical and clinical studies that have been conducted to provide supportive evidence for the use of statins in pregnancy. Studies in mice using dihydroxyheptanoic acid, a derivative of mevinolin (lovastatin) at high doses (10-47 x recommended dosage), were not reported to increase risk of congenital abnormalities (Minsker et al., 1983). This study was useful in showing how statins and their derivatives display different levels of potencies and hence difference levels of teratogenicity. 72

Cudmore et al. showed through their in vitro studies in endothelial cells and placental explants that simvastatin did increase the expression of haemoxygenase -1 (HO- 1), reducing the secretion of sFlt-1 and sEng (Cudmore et al., 2007). Cudmore et al. focused on the effects of statins in vitro and used statins with different lipophilic states, adding more breadth to their research. Furthermore, Costantine et al. demonstrated that pravastatin could attenuate phenylephrine-induced constriction in carotid arteries in a mouse model of sFlt-1-induced PE (Costantine et al., 2010). They also showed that vascular reactivity was partially sustained even in the presence of L-NAME, a non-selective inhibitor of NO synthase, suggesting that pravastatin may work through a NO-independent pathway too. Ahmed et al. showed that pravastatin could enhance circulating concentrations of VEGF and restore the angiogenic balance of VEGF/sFlt-1 in an immunological model of PE (Ahmed et al., 2010). The work conducted by Ahmed et al. yielded promising results however the model used was a CBA/J x DBA/2 mouse model of recurrent miscarriage as well as a model of immunologically-mediated PE. However, one could argue about the merit of this mouse as a model of PE. For instance, in this study they did not observe any significant changes in blood pressure in the CBA/J x DBA/2 mice. They did however see increased vascular susceptibility to vasoconstriction due to increased sensitivity to angiotensin II in their mouse model. They also administered pravastatin early from day 4- 12 of gestation, which is contradictory to studies highlighting the risks of administration during first trimester, and could make clinical translation problematic. Finally, Kumasawa and colleagues studied the effects of pravastatin (5 μg/d via i.p. administration) in a sFlt-1 model of PE with FGR. They observed that, at day E18.5, pravastatin antagonized the effects of sFlt-1 by increasing PlGF resulting in a significant reduction in blood pressure, glomerular endotheliosis, maternal proteinuria, and increased fetal weight. In over 100 mice treated, no fetal malformations were detected and the second generation of treated animals were fertile (Kumasawa et al., 2011). Kumasawa et al. produced a sFlt-1 mouse model which phenocopies human PE, in particular late-onset, showing the key characteristics such as hypertension and proteinuria, which regressed post-parturition; FGR was also observed. Furthermore, Kumasawa and his team administered pravastatin intraperitoneally, this route of administration could have influenced the results achieved and in humans, statins are routinely administered orally. Lastly, atorvastatin was mentioned in the methods section however, it was not mentioned in the objectives or discussion and there were no results pertaining to atorvastatin alone or in comparison to pravastatin, suggesting a

potential reporting bias. These studies were all critical in offering insight into the use of statins during pregnancy and the various methods that could be used to reflect a PE-like model. Table 1.3 summarises the pre-clinical mouse models used to study the effect of statins, specifically pravastatin, in pregnancy for PE or FGR.

Table 1.3: preclinical models using pravastatin in pregnancy

Preclinical model Agent Results Reference

CBA/J × DBA/2 mice Pravastatin (20 ↓ sFlt-1 ↓ Ahmed et al., 2010 ug/kg) Hypersensitivity to Ang II ↑ VEGF

CD-1 mouse injected Pravastatin (5 ↓sFlt-1 Restoration of McDonnold et al., with adenovirus mg/kg/d) glucose response in 2014 carrying sFlt-1 females

CD-1 mouse injected Pravastatin (5 Regularization of Carver et al., 2014 with adenovirus mg/kg/d) impaired vestibular carrying sFlt-1 function, balance and coordination linked with preeclampsia

CD-1 mouse injected Pravastatin (5 ↓ sFlt-1 ↓ sEng ↓ Saad et al., 2014 with adenovirus mg/kg/d) Overexpression of carrying sFlt-1 TGF-β in placenta ↓ HIF-1α

CD-1 mouse injected Pravastatin (5 ↑ eNOS in the aorta Fox et al., 2011 with adenovirus mg/kg/d) carrying sFlt-

CD-1 mouse injected Pravastatin (5 ↓ sFlt-1 ↑ PIGF ↓ Kumasawa et al., with adenovirus mg/kg/d) Hypertension ↓ 2011 carrying sFlt-1 Proteinuria

CD-1 mouse injected Pravastatin (5 ↓ sFlt-1 ↓ Contractile Costantine et al., with adenovirus mg/kg/d) response to 2010 carrying sFlt-1 phenylephrine ↑ Vasorelaxant response to ACh

C1q deficient (C1q-/-) Pravastatin (5 mg/d) ↑ VEGF ↓ sFlt-1 ↓ Singh et al., 2011 mouse Albumin creatinine ratio (ACR) ↓ STAT-8 Matrix metalloproteinase (MMP) activity Normal aortic ring response to AngII

Reduced utero- Pravastatin (1 ↓ MAP ↓ sFlt-1 ↑ Bauer et al., 2013 placental perfusion mg/kg/d) VEGF ↓ sFlt-1/VEGF pressure (RUPP) rats ratio ↓ Thiobarbituric acid reactive substances ↑ Total antioxidant capacity ↓ Endothelial tube formation No effect on HO-1 expression

11β-hydroxysteroid Pravastatin (20 ↑ placental vascular (Wyrwoll et al., 2016) dehydrogenase type μg/kg) endothelial growth 2 knockout mice factor A, vascularization,

placental fetal capillary volume, fetal weight, and cardiac function.

↓ aberrant umbilical cord velocity

1.11.2 Evidence of the use of statins in pregnancy from clinical studies The StAMP trial (“pravaStatin to Ameliorate Early Onset Pre-eclampsia) aimed to assess whether oral pravastatin (40 mg) could be used as a clinical treatment for early-onset PE given from the 2nd trimester (23-31 weeks) and until 1 month post -delivery. In particular, the primary outcome of the trial team was the effect of pravastatin on sFlt-1 at 48 hours post-randomisation and whether there were any adverse effects to the mother or baby. Over the course of the trial, some adjustments were made to the protocol such as changing their statin from simvastatin to pravastatin. These reasons were not explained but they were transparent about this change and reported it on ISRCTN registry. As well as this they halved their trial participant number from 128 to 64 and due to recruitment challenges the trial end date was extended from 2013 to July 2014 (Ahmed, 2011). The data from the publication of basic results (scientific) revealed that the primary outcome was not achieved with there being no difference in maternal serum sFlt-1 levels between pravastatin and placebo group (Ahmed et al, 2019).

From August 2012 through to February 2014, a pilot multicentre double-blind placebo-controlled randomized control trial of women at risk of PE was conducted, to collect preliminary maternal-fetal safety and pharmacokinetic data for the use of pravastatin during pregnancy. Twenty women with a history of severe PE in previous pregnancies that required delivery prior to 34 weeks gestation were recruited during 12- 16 weeks of gestation and were randomized (10 assigned to oral pravastatin 10 mg or 10 to placebo). Data suggested lower rates of PE (4 in placebo group vs 0 in pravastatin group) and indicated preterm delivery (preterm births: 5 in placebo group vs 1 in pravastatin group) and a positive pro-angiogenic profile where pravastatin increased PlGF and decreased sFlt-1 and sEng (angiogenic change was not statistically significant). Additionally, there was no significant difference in birthweight between the two groups or congenital

abnormalities maternal, fetal or neonatal deaths. This safety trial was part of a much larger randomized trial currently taking place at Eunice Kennedy Shriver National Institute of Child Health and Human Development Obstetric-Fetal Pharmacology Research Units Network, assessing the use of pravastatin for the prevention of PE in high-risk women (Costantine et al., 2016).

Brownfoot et al. demonstrated through a pilot study that when 40 mg pravastatin was administered orally to 4 women with pre-term PE (24+5 to 29+4 weeks), clinical signs of PE were improved; as demonstrated by reductions in the urine protein/creatinine ratio. When these women were admitted, their systolic blood pressure ranged between 155 and 200 mmHg and diastolic between 90-105 mmHg but following pravastatin administration mean arterial pressure stabilized or increased only slightly during the trial. These women remained hypertensive but not sufficiently to be delivered preterm (Brownfoot et al., 2015). The low N’s in this study urge caution in the interpretation of these findings. Furthermore, in another trial 11 women with antiphospholipid syndrome received 20 mg of pravastatin orally in addition to standard treatment during pregnancy from the time of PE diagnosis versus 10 women on standard treatment (low molecular weight heparin, enoxaparin or tinzaparin; 40 mg subcutaneously, once daily plus low dose aspirin; 80 mg orally, once daily). In the pravastatin-treated group, hypertension and proteinuria stabilized, uteroplacental perfusion increased as evidenced by Doppler blood flow assessment of the uterine artery. These positive signs were seen as early as 10 days post- treatment. Pravastatin treatment resulted in 100% live births for pravastatin-treated patients vs 45% for patients on conventional antithrombotic therapy (Lefkou et al., 2016).

In conclusion, there is no current clinical treatment for PE +/- FGR but statins are mooted as a potential therapy. Statins have the potential to increase NO availability, reduce oxidative stress and restore the angiogenic balance, features believed to contribute to endothelial dysfunction in PE. What remains to be explored is a thorough assessment of the effect of statins on vascular function, including maternal systemic, uteroplacental and fetoplacental vessels, in normal pregnancy and in PE including assessments of endothelial- dependent and -independent actions.

1.12 Summary PE is a serious and debilitating pregnancy complication. PE can detrimentally affect short and long-term health for both mother and fetus. Currently, PE has no cure. Chronic hypertension (CHT) is a risk factor for PE and shares similar pathological similarities such as endothelial dysfunction with women with CHT often going on to develop superimposed PE. The use of maternal OAs and placental CPAs from PE women and OAs from women with CHT and superimposed PE, present vascular beds that can be used to assess the therapeutic effects of statins on vascular function from maternal and fetoplacental vessels. These are important studies to undertake given the paucity of evidence on the effects of statins on vascular reactivity in pregnancy.

Mice are a well-characterized model of pregnancy that display similar cardiovascular physiological adaptions to pregnancy and placentation as women. The eNOS knockout mouse model exhibits chronic hypertension with FGR but represents a model in which endothelial dysfunction can be studied and statins’ effects on vascular function can be assessed. This model will not only allow us to assess statin effect on maternal and fetal vascular function but also assess teratogenic effects on fetal/placental weight and litter size. Given the suggested role of eNOS and NO in the pleiotropic effects of statins, it also affords the opportunity to study in more detail the eNOS-independent effects on vascular function. With pravastatin already being assessed in clinical trials for the treatment and prevention of PE, it is important to also investigate whether other statins are safe and efficacious for use in PE. A study assessing short and long-term effect of statins on maternal and fetal vascular function and safety profile in humans and mice would be extremely informative.

1.13 Hypotheses

1.13.1 Human Short-term (2h) exposure to statins will improve vascular reactivity of omental arteries from pre-eclamptic, chronic hypertensive, superimposed pre-eclamptic pregnancies, and have no detrimental effects on fetoplacental chorionic plate artery function.

1.13.2 Mouse  In eNOS knockout mice, short-term (2h) exposure to pitavastatin will result in improved uterine artery function and have no detrimental effects on umbilical artery function.

 In eNOS knockout mice, pitavastatin treatment from embryonic days 10.5 to 18.5 will improve maternal vascular (mesenteric artery and uterine artery) function without having detrimental effects on umbilical artery function or fetal growth.

1.14 Aims and objectives

1.14.1 Human vessels  To assess whether vascular reactivity of chorionic plate arteries is altered in placentas from PE pregnancies compared to normal pregnancies.  To assess whether vascular reactivity of omental arteries is altered in pregnancies complicated by hypertension (including preeclampsia) compared to normal pregnancies.  To assess if vascular reactivity of omental arteries and chorionic plate arteries, from hypertensive or normal pregnancies, is altered following short-term exposure to statins.

1.14.2. Mouse vessels In vitro

 To confirm whether vascular reactivity of uterine and umbilical arteries in eNOS knockout mice is altered compared to wild type mice.  To assess whether vascular reactivity of uterine and umbilical arteries in eNOS knockout mice and WT mice is altered following 2 hr pitavastatin exposure. In vivo

 To assess whether vascular reactivity of mesenteric, uterine and umbilical arteries in eNOS knockout mouse and WT mice is altered following pitavastatin treatment from E10.5 until E18.5.  To assess the effect of pitavastatin treatment on litter size, fetal weight and placental weight in eNOS knockout mice.

Chapter 2: Materials and Methods

2.0 Studies in human:

2.1 Source of Chemicals Unless otherwise stated, all chemicals were from Sigma Aldrich, UK.

2.2 Ethical approval The Research ethics committee gave approval for the work carried out herein (ethics number 15/NW/0829). Studies were carried out in accordance with the Declaration of Helsinki. Written informed consent was given from all women prior to the collection of placental and omental samples, maternal blood and/or umbilical cord blood. The inclusion and exclusion criteria for women are defined below in table 2.1.

Inclusion criteria Exclusion criteria

BMI < 35 kg/m2 Congenital abnormalities, pre-existing cardiovascular or renal disease; if women had AGE < 40 years sexually transmitted diseases; maternal diabetes; Normal Gestation: 37 + 0 to hypertension or other pathology. 42 + 0 weeks

AGA Centiles: 10th and 90th

BMI <40 kg/m2 Diabetes; hyperthyroidism; twin pregnancies or if women had sexually transmitted diseases. AGE <45 years

Pathological Gestation: 26 + 0 weeks onwards

AGA Centiles: <10th centile

Study Groups

Pre-eclampsia Pre-eclampsia was defined as de-novo hypertension before 20 weeks and a new onset of proteinuria, fetal growth restriction and/or PlGF measurement of <12 pg/ml (indicative of placental insufficiency) (Brown et al., 2018).

Chronic Chronic hypertension was defined as having a high blood pressure prior Hypertension to pregnancy or recognized <20 weeks’ gestation.

Superimposed pre- Superimposed PE was defined as worsening hypertension, new eclampsia proteinuria and/or the presence of other features suggestive of placental disease (Duhig et al., 2019) in a woman with a diagnosis of PE.

Normotensive Normal pregnancy was defined as women with an uncomplicated pregnancy with an appropriately grown fetus.

Table 2.1: Inclusion and exclusion criteria for pregnant women

This table describes the inclusion and exclusion criteria we used to recruit pregnant women in this study. The cut-off value used for BMI and age differed between normal and pathological pregnancies. This is because the % proportion of women in the pathological group were classed as obese and some were of advanced maternal age thus in order to ensure enough samples were collected to appropriately power our studies, a BMI and age cut-off of <40 kg/m2 and <45 years was utilised respectively. Clinical definitions of the study groups used are included. Hypertension was defined as systolic BP ≥140 and/or diastolic BP ≥90 mmHg” (Brown et al., 2018).

The inclusion and exclusion criteria for the pathological groups were set to increase recruitment flexibility, as these samples were scarcer. This group contained women with a higher BMI and women with advanced maternal age, as a result of this inclusion, data will need to be interpreted with caution as both BMI and maternal age influence vascular function. More detailed demographic information for the women whose samples were used in the specific studies can be found in tables 3.1 and 4.1 (chapters 3 and 4 respectively).

2.3 Collection of blood samples and tissue sample preparation All maternal venous blood samples were collected prior to delivery into vacutainers (serum gel EDTA KE monovettes, Sarstedt, Nümbrecht, Germany). For women undergoing caesarean sections, maternal venous blood was taken prior to the onset of anaesthesia. Umbilical venous blood was collected immediately following delivery (serum gel EDTA KE monovettes, Sarstedt, Nümbrecht, Germany). All blood samples were spun in a Spinchron™® centrifuge (Beckman Coulter, USA) at 3000 rpm (11180 g) for 10 mins at 4 ºC before being aliquoted into eppendorfs and stored in a -80 °C freezer for further analysis.

Maternal plasma collected in EDTA coated vacutainers and were used for the quantitative determination of PLGF using the Alere Triage® PLGF test (Triage® MeterPro, Inc., San Diego, USA). The test is intended for use in conjunction with other existing clinical diagnostic tests, to aid the diagnosis of PE and to assess the level of risk for delivery arising from PE within 14 days of testing (Duhig et al., 2019; Frampton et al., 2016).

Placentas and omental biopsies from complicated and uncomplicated pregnancies were collected within 20-30 mins of delivery following either caesarean section or vaginal delivery. Chorionic plate arteries (CPAs) were identified at the umbilical cord insertion point and traced across the placental surface. A chorionic plate biopsy was taken and placed in ice-cold physiological saline solution (PSS; 119 mM NaCl, 25 mM NaHCO3, 4.69 mM KCL 2.4 mM MgSO4, 1.6 mM CaCl2, 1.18 mM KH2PO4, 6.05 mM Glucose, 0.034mm EDTA; pH 7.4) when appropriately sized CPAs (100-500 µM) were identified. CPAs were identified using a Leica EZ40 HD stereomicroscope (Leica Microsystems, (Schweiz), Singapore) and dissected free of the surrounding connective tissue and cut into small segments (2-3 mm) ready to be mounted onto a 40 µM stainless steel wire in a chamber of a Danish Myotechnologies 610M wire myograph (DMT, Aarhus, Denmark) using dissecting scissors and forceps.

2.4. Multi Myograph System 620M The Multi Wire Myograph System - 620M is the machine used in this laboratory to carry out wire myography on small resistance arteries. The individual myograph units each have a stainless steel chamber where the supporting jaws are positioned, one jaw end is attached to the force transducer, which measures tension while the other end is attached to a micrometer which measures vessel stretch. Each myograph unit has temperature control, an individual gas inflow and suction outflow along with heating, allowing each chamber to operate under physiological conditions (37 °C with appropriate gas).

Figure 2.1: Detailed diagram of Multi Myograph System 620M

A) DMT multi-chamber 620 wire myograph, which holds 4 myograph units. B) Myograph unit with transducer and micrometer, which are connected to the corresponding jaws. C) A zoomed in view of myograph bath with a vessel inside. D) Typical set-up when mounting a vessel onto myograph unit (Veteriankey, 2016).

2.5 Vasoconstrictors and Vasodilators

2.5.1 Vasoconstrictive agents When considering the choice of vasoconstrictive agents to use, it is important that the agent of choice provides a reproducible and sustained constriction. This then affords the opportunity to assess the magnitude of vessel relaxation following addition of vasodilatory agents. Vasoconstrictive agents that could have been used include: vasopressin, KPSS, phenylephrine, endothelin-1, norepinephrine, 5-hydroxytryptamine (5-HT) or the thromboxane mimetic U46619. The purpose of the myography experiments detailed in this thesis was to assess the effect of statins on vascular function, including assessment of endothelium dependent and independent relaxation.

2.5.1.1 Vasoconstrictors Phenylephrine was used only in the ex-vivo mouse myography studies, described below in 2.12.3, to produce an acute pre-constriction of uterine arteries to assess endothelial function with the endothelium-dependent vasodilator acetylcholine (10-5 dose). The thromboxane mimetic U46619 was used to conduct a dose-response curve since it 84

produces a stable and reproducible constriction in peripheral vessels in different species (Macintyre et al., 1978). In my experiments, a U46619 dose response curve was used to assess effect of statin exposure on contraction but also to produce a stable pre-constriction prior to performing an endothelium-dependent and endothelium-independent relaxation dose response curve.

2.5.2 Vasodilatory agents A number of agonists can cause dilation of pre-constricted vessels in vitro. The choice of vasodilators in this thesis was decided based upon their ability to consistently cause reproducible relaxation in OAs and CPAs from women and in uterine and umbilical arteries from mice, described later.

2.5.2.1 Endothelium-dependent: Bradykinin Bradykinin was chosen for the endothelium-dependent relaxation experiments in the omentum due to its ability to produce consistent relaxations in vessels under the same experimental conditions. Additionally, its relative ease of use, stability under storage and replicability makes it a suitable choice. In previous studies, when bradykinin was used it was able to successfully cause relaxation of OAs pre-constricted with U46619 or endothelin (Vedernikov et al., 1999).

2.5.2.2 Endothelium-dependent: Acetylcholine Acetylcholine was used mainly for the ex-vivo and in-vivo mouse studies; acetylcholine has been the vasodilator of choice in our lab and has produced consistent and reproducible results.

2.5.2.3 Endothelium-independent: SNP In CPAs, relaxation could only be achieved by using SNP (Wareing et al., 2002). It has also been shown by other groups, that CPAs produce a minimal vasodilatory response to endothelium-dependent agonists (Mccarthy et al., 1994b; Ong et al., 2003). SNP was also used to investigate endothelium-independent effects in OAs, as per previous studies

(Suzuki et al., 2000) that showed that SNP (1 nM -10 μM) produced a concentration- dependent relaxation of OAs pre-constricted to STA2 in vessels from both normotensive pregnant and PE women (Suzuki et al., 2000).

2.6 Myography

2.6.1. Choice of Statin and Dose for Experiment Three statins chosen included pravastatin, pitavastatin and simvastatin, their different metabolic states made them useful for comparison of pleiotropic effects on vascular function. Pitavastatin was used as an example statin to explain drug preparation and treatment protocols below (preparations for the other statins was similar). When establishing this method, it was also important to determine an appropriate concentration of statin to use in the study as well as the length of exposure.

Rossini et al. assessed the effect of simvastatin on the vascular reactivity of mesenteric arteries in a non-pregnant rat at a concentration of 0.1 µM and 1 µM for 1 hr and 2 hr; they found 1 µM simvastatin attenuated U46619-induced contraction after 2 hr incubation (Rossoni et al., 2011). The outcome of this study informed our experimental approach. A dose of 10 µM was used alongside 1 µM to mimic the 10-fold safety factor commonly used in toxicology studies which accommodates for human variability and possible synergistic effect (Dorne and Renwick, 2005). The doses used in this study are pharmacological and higher than those previously measured in human serum. For pravastatin: 1 µM is 22x higher than a 40 mg dose in pregnant population and 10 µM is 44.4x higher (Costantine et al., 2016); pitavastatin: 1 µM is 4x higher than 4 mg dose in non- pregnant population and 10 µM is 40x higher (Luo et al., 2015) and simvastatin at 1 µM is 32x than 40 mg dose in non-pregnant population (Bjorkhem-Bergman et al., 2011).

Pitavastatin is a novel statin with high bioavailability in the bloodstream due to it avoiding the first pass metabolism route in the liver. Most importantly is that it has never been used in pregnancy studies and therefore an exciting therapeutic candidate to explore. Pitavastatin was dissolved in the organic solvent DMSO and a stock solution of 10-2 M (10mM) was created and aliquoted into Eppendorf tubes before being stored in a -20 freezer. When ready for use, 10 µl of pitavastatin was removed from stock solution and added to 90 µl of DMSO to create a 1 mM concentration. Then 6µl of this final volume was added to a 6ml myograph bath filled with PSS to create a final concentration of 1 µM. To control for DMSO concentration in the non-statin exposed vessels, 6 µl of the control (DMSO) was added to 6ml of PSS when pitavastatin or simvastatin was used (DMSO was diluent for both). Water was the diluent for pravastatin and so 6 µl of water would be added to 6 ml of PSS when pravastatin was used. 86

2.6.2 Chorionic Plate Artery (CPA) Protocol Following collection, CPAs (112-501µM) were mounted on a wire myograph (M610; Danish

Myotechnologies, Denmark) and normalized to 0.9L5.1kPa by incrementally stretching the vessel using the micrometer and monitoring the tension generated using the force transducer until passive tension was more than 5.1 kPa. These vessels normally would have 4-6 stretches. Data were collected and utilizing an online normalization software package (Myodata; Danish Myotech) the internal diameter was calculated, to which the vessel was subsequently adjusted (Wareing et al., 2002). Isometric tension development was continuously recorded (myodaq data acquisition system). Wareing et al. suggested that 5.1 kPa should be used for normalisation of CPAs, based upon predicted pressures found in vivo (Wareing et al., 2002). The method for this required the use of Laplace’s equation shown below figure 2.2. As well as diameter calculation, normalisation ensures the vessel is set at its effective pressure to allow optimal response to pharmacological agents (Coats and Hillier, 1999). Furthermore, normalisation ensures standardisation of all vessels at the beginning of each myography experiment, which promotes reliable comparison of results under the same conditions (Spiers and Padmanabhan, 2005).

Figure 2.2: Representation of LaPlace’s Equation The relationship between vessel wall length (ie, internal circumference, usually in mm units), vessel wall tension (usually presented as mN/ mm), and the effective pressure (Pi) is described mathematically by La Place’s Law.

Following equilibration (in PSS gassed with 5% O2 (normoxic for placenta)/5%

CO2/balance N2, at 37ºC for 20min) (Hayward et al., 2013), contractile viability was assessed with 2x high potassium PSS (KPSS; 120mM KCl, equimolar substitution for NaCl) with 10- 20min intervals in between. Vessels were then washed out with PSS to resume their passive tension. Vessels with a constriction less than 3 kPa were excluded from the study. Vessels were then exposed to the thromboxane mimetic U46619 (Merck and VWR) to conduct a constriction-dose response curve 10-10 -10-5.7, before being washed out with PSS.

Afterwards 1 µM pitavastatin was added in 6ml of PSS in the myograph bath and the control DMSO was added to a separate bath and left to incubate with vessels for 2 hr. Following incubation with statin/vehicle, a further dose response to U46619 was carried out. A dose response to the endothelium-independent vasodilator sodium nitroprusside (SNP, 10-10-10-5) was carried out with vessels in a separate bath simultaneously. At the end of the experiment, CPA’s were exposed to 1x KPSS to confirm vessels remained viable and to compare KPSS contraction relative to initial KPSS contraction. The full protocol is outlined in the representative myography trace in figure 2.4.

To calculate the effective concentration of U46619 to contract vessels to 80% of the Vmax,

(EC80 of U46619), the raw contraction data from the dose-response curve was collected and a non-linear regression curve performed to give a logECF value (Harvey Motulsky, 2004). The following method below in figure 2.3 was used to calculate the volume of

U46619 to add to each myography chamber to produce an EC80 contraction in the vessels:

* The criteria for exclusion of vessels are outlined in the relevant results chapter*

U46619 Dose-

response Curve 10-10 -10-5.7 KPSS 2X KPSS

U46619 Dose- response Curve 10-10 -10-5.7 Normalization of vessels 88 Statin incubation (1µM, 2hrs)

U46619 Dose- response Curve

10-10 -10-5.7 KPSS

2X KPSS U46619 Dose- response Curve 10-10 -10-5.7

Statin incubation (1µM, 2hrs)

Figure 2.4: A representative image of full myography protocol for CPAs Following normalisation, 2x KPSS washes and a U44619 dose response curve, vessels were randomly assigned to vehicle (DMSO) or 1µM pitavastatin, which were dissolved in 6ml of PSS and left to incubate for 2hrs in the myograph baths. Post-incubation, contraction was assessed by a U46119 dose response curve followed by a final KPSS contraction. X-axis denotes times

and Y axis denotes Tension (mN/mm). U44619 = thromboxane A2 mimetic.

2.6.3 Omental Artery (OA) Protocol Omental biopsies of ~2x2cm were excised from the greater omentum in the abdomen of women. Following collection, OAs (138-521µM) were isolated from the omentum biopsy, fat and scar tissue were removed; resistance arteries were identified and mounted. The

OAs were normalized to 0.9L13.3kPa as described previously for CPAs. These vessels were stretched 4-8 times during normalisation due to them being slightly more elastic in behaviour.

The protocol for the OAs was similar to the CPAs; differences are outlined here. A trace from a whole experiment can be seen in figure 2.5. OAs were normalised at 0.9 of

L13.3 kPa and incubated in 20% oxygen (as opposed to 5% in the experiments on CPA) which reflects the relative estimations of PO2 in maternal and fetoplacental vessels respectively. In OAs, after the first U46619 dose response, bradykinin (10-5) was added in order to assess endothelial function of these vessels. To assess the effect of relaxation after incubation with either statin or vehicle, a dose response to either bradykinin (10-10-10-5) or SNP (10-10- 89

10-5) to investigate endothelium-dependent or endothelium-independent relaxation respectively.

U46619 Dose- response Curve 10-10 -10-5.7 Single high dose KPSS BK 10-5 2X KPSS U46619 Dose- response Curve 10-10 -10-5.7

Normalization of vessels Statin incubation (1µM, 2hrs)

Figure 2.5: A representative trace of the wire myography protocol for omental arteries

Following normalisation, 2x KPSS washes and a U44619 dose response curve, endothelial KPSS -5 function was assessed (10 of bradykinin). Then vessels were randomly assigned to vehicle (DMSO) or 1µM pitavastatin, which were dissolved in 6ml of PSS and left to incubate for 2hrs in Bradykinin Dose- the myograph baths. Post-incubation, relaxation was assessed by bradykinin (10-10 – 10-5) response Curve following pre-constriction to U46619, followed by a final KPSS contraction. X-axis denotes times and Y axis denotes Tension (mN/mm). U44619 = thromboxane A2 mimetic.

At the end of each experiment, contraction data was calculated as active effective pressure Pitavastatin incubation Pre-

(1µM, 2hrs) constriction (kPa) using the equation below, it is defined as the force required to maintain the to U46619 circumference of a vessel.

푨풄풕풊풗풆 푬풇풇풆풄풕풊풗풆 푷풓풆풔풔풖풓풆 = (푾풂풍풍 푻풆풏풔풊풐풏 2π) / 푽풆풔풔풆풍 푰풏풕풆풓풏풂풍 푪풊풓풄풖풎풇풆풓풆풏풄풆 *

*The criteria for exclusion of vessels are outlined in the relevant results

chapter*

2.7 Data Analysis Following each experiment, data from the experiment was exported from Myodata and inputted into Microsoft excel. Inputted data was then used to calculate kPa and tension as a percentage of KPSS constriction (%KPSS), this data was imported into Graphpad Prism v8.0 (Graphpad, La Jolla, SD, USA). Normality tests were conducted using the Kolmogorov- Smirnov test to assess whether data fitted to a Gaussian distribution. This was used for datasets where only one comparison was made for example basal tone pre and post-statin incubation and KPSS contraction pre-and post-statin incubation; the normality test result determined whether paired parametric or non-parametric t-tests were conducted. U46619-induced constriction and SNP/Bradykinin-induced relaxation of OAs and CPAs from women with normal and pathological pregnancies post-statin incubation were assessed using repeated measures two-way ANOVA. Data was expressed as mean ± S.E.M or median ± IQR depending on the outcome of the normality tests. The number of observations (n = arteries from N patients) expressed in parenthesis. P<0.05 was considered statistically significant.

2.8 Studies in Mouse

2.8.1 Ethical Considerations and Animal Husbandry Experiments were performed in accordance with the UK Animals (Scientific Procedures) Act of 1986 under the authority of a Home Office licence (PPL P9755892D). The Animal Welfare and Local Ethical Review Board of the University of Manchester approved all protocols. The methods outlined in this project adhere to the ARRIVE guidelines (Kilkenny et al., 2012). All animals were housed in individually ventilated cages, enriched with nesting material and play tunnels, under a constant 12 h light/dark cycle at 21-23°C with full access to food (BK001 diet, Special Dietary Services, UK) and water (Hydropac, Denver, US). Female mice were communally housed whilst stud males were individually housed. eNOS−/− mice (strain B6.129P2-Nos3tm1Unc/J) were obtained from Jackson Laboratories (Bar harbour, ME, US). These mice were produced, briefly as follows; a targeting vector containing neomycin resistance and herpes simplex virus thymidine kinase genes was used to replace 129 bp of exon 12, which disrupted the calmodulin binding domain. The construct was electroporated into 129P2/OlaHsd-derived E14TG2a embryonic stem (ES) cells. Correctly targeted ES cells were injected into C57BL/6J blastocysts and the resulting chimeric males were crossed to C57BL/6J female mice. Heterozygote mice were intercrossed to generate 91

null homozygotes. The mice were subsequently backcrossed onto the C57BL/6J background for 12 generations (The Jackson Laboratory., 2019). Homozygous eNOS−/− female mice (10-16 weeks) were mated with homozygous eNOS-/- males (12-26 weeks), and the presence of a copulation plug was denoted as day 0.5 of pregnancy. Term is E19.5-20. C57Bl/6J (wild type, WT) mice, the background strain for eNOS−/− mice, were initially from Envigo (Huntingdon, UK) and later from Charles River (Harlow, UK), and were used as control mice. The reason for the change in supplier for WT mice was due to over-aggression of male mice from Envigo meaning that our facility switched suppliers to Charles River. All C57Bl/6J mice used for the ex vivo studies described below were from Envigo whilst all C57Bl/6J mice used in the in vivo study were from Charles River.

2.9 Experiment allocation

Mice

Ex vivo studies In vivo studies

Randomised once pregnancy Remained Tissue harvest at confirmed E18.5, uterine and on drinking water umbilical arteries used to assess

pitavastatin’s effect Pitavastatin Vehicle from E10.5- on vascular function. from E10.5- E18.5 E18.5

Tissue harvest at E18.5

Uterine, mesenteric and umbilical

arteries used for assessment of Maternal and fetal tissue stored for later use. vascular function with pitavastatin.

Figure 2.6: Experimental allocation of mice for ex vivo and in vivo studies The mouse work conducted in this PhD was split into 2 arms, ex vivo and in vivo. The mice in the ex vivo study were sacrificed at E18.5 and the uterine horn was harvested and uterine and umbilical arteries’ reactivity assessed, in the presence/absence of 1 µM pitavastatin for 2 hrs, by wire myography. In the in vivo arm, once mice were confirmed to be pregnant, mice were randomised to either treatment with pitavastatin via drinking water (6 µg/ml) or control in vivo for 8 days (E10.5-E18.5).

2.10 Sample Collection for Ex Vivo Study On day E18.5, pregnant mice were euthanised (cervical dislocation followed by confirmation of cessation of circulation appropriate under ASPA schedule 1) and a laparotomy and hysterectomy performed; the entire uterus was placed in ice-cold PSS to euthanise all fetuses.

2.11 Blood vessel dissection and normalisation

2.11.1 Maternal Uterine Artery The uterus, still with the fetuses therein, was carefully pinned and the uterine artery isolated using the same Leica EZ40 HD stereomicroscope as used for human tissue. Care was taken not to touch the uterine vein or overstretch the uterine artery. Once uterine arteries were isolated from the left and right side of the uterine horn, the isolated portions were cut into 4 segments resulting in 8 segments in total. 8 segments were prepared to allow for random selection of vessels and only six were used for subsequent experiments. These 6 vessels were mounted separately on two 40 µM stainless steel wires and secured in the jaws of the myograph. The vessels were left to equilibrate in PSS at 37°C (20% O2/

5% CO2/ balance N2). The normalisation process was identical to the omental arteries mentioned previously (normalised to 0.9 of L13.3kPa, section 2.6.3 in the methods).

2.11.2 Umbilical Artery Once the uterine artery had been isolated, a hysterotomy was performed on the uterus and two pups chosen at random. The umbilical artery was identified and dissected free under a stereomicroscope. Each umbilical artery was mounted separately on two 40 µM stainless steel wires and secured in the jaws of the myograph. The vessels were left to equilibrate in PSS at 37°C (5% O2/ 5% CO2/ balance N2). The normalisation process was the same as for chorionic plate arteries mentioned in section 2.6.2 (normalised to 0.9 of

L5.1kPa,). 4 pups were randomly chosen per litter (2 from each uterine horn) to acquire umbilical arteries.

2.11.3 Myography Myography was performed to assess vascular function. The thromboxane mimetic U46619 was used to elicit vasoconstriction in uterine and umbilical arteries. To assess vasorelaxation, the endothelium-dependent vasodilator acetylcholine (1x10-10-10-5M) was used for uterine arteries and the NO-donor SNP (1x10-10-10-5M) for both uterine and umbilical arteries. A schematic diagram of the full protocol can be found in figure 2.7 below.

10-5ACH U46619 2X KPSS Dose- PE response curve KPSS ACh Dose- response curve

Pitavastatin incubation EC80 (1µM, 2hrs) U46619

Figure 2.7: A representative myography trace for ex vivo mouse experiments. Following normalisation, 2x KPSS washes and to assess endothelial function, uterine arteries were pre-constricted to phenylephrine and relaxed with acetylcholine (ACh, 10-5). This was followed by a U44619 dose response curve, then vessels were randomly assigned to vehicle (DMSO) or 1 µM pitavastatin, which were dissolved in 6ml of PSS and left to incubate for 2 hr in the myograph baths. Post-incubation, relaxation was assessed by pre-constricting uterine arteries to EC80 of U46619 and performing a ACh dose response curve followed by a final KPSS contraction. X-axis denotes times and Y axis denotes Tension (mN/mm). PE = phenylephrine and U44619 = thromboxane A2 mimetic.

2.12 In Vivo Study Prior to treatment, maternal weight was recorded and urine collected (if possible) at E0.5 (plug date) and again at E10.5 when treatment started. At E10.5 mice were randomised to pitavastatin, its vehicle carboxymethylcellulose or drinking water alone and drinking bottles containing pitavastatin or carboxymethylcellulose were changed daily. The mice were randomised by creating a blocked randomisation list (https://www.sealedenvelope.com/simple-randomiser/v1/lists. A dose of 6µg/ml pitavastatin was chosen as it equated to a human equivalent dose of 4 mg of pitavastatin, making our study clinically relevant. This was calculated by taking the dose of clinical dose of pitavastatin 4 mg and dividing it by 70 kg (non-pregnant human weight) which equated to 0.0571 mg/kg/d. Administration via drinking water was chosen as this method is akin to what we would see in a clinical setting if statins were administered in pregnancy, as it is 95

normally taken orally. Carboxymethylcellulose was chosen as a diluent for pitavastatin because when the organic solvent DMSO was implemented in-vivo, the mice did not tolerate the smell of it in the drinking water and as a result did not drink the water. Taking into consideration the wellbeing of the mice used in the study but also ensuring the experimental design was still preserved, an alternative diluent was chosen and carboxymethylcellose was selected based on non-pregnant literature where it has been used as a vehicle for pitavastatin. The primary outcome of this study was to observe a significant alteration in vascular function while the secondary outcome was to assess fetal viability. Study design is outlined below in figure 2.8.

Figure 2.8: In-vivo study design.

At E10.5, eNOS-/- or C57Bl/6J dams were randomised to either water, pitavastatin (6µg/ml) in 0.05% CMC or CMC vehicle (0.05%) and treated until E18.5. The mice were randomised by creating a blocked randomisation list with the treatment options. Fetal and placental weights and fetal measurements and ex-vivo wire myography were conducted on E.18.5.

2.13 Sample Collection for In Vivo Study On day E18.5, pregnant mice were euthanised (cervical dislocation appropriate under ASPA schedule 1) and a laparotomy and hysterectomy performed. Following E18.5 bodyweight being recorded, mice were sacrificed by cervical dislocation; urine was collected and maternal trunk blood collected by decapitation and snap frozen. Post-sacrifice, maternal brain, heart, liver, kidneys and spleen were harvested, weighed and snap frozen on dry ice. Following removal of the uterus from pregnant animals, a hysterectomized weight was recorded. Along with the uterine horn, the mesentery with the jejunum and ileum were 97

isolated from the abdominal cavity and surrounding tissues. Remaining procedure mimicked that mentioned in ex-vivo study.

2.14 Assessment of progeny at E18.5

2.14.1 Assessment of fetal weight, placental weight and fetal biometric measurements Fetal and placental weights were recorded approx 1-2hrs after laparotomy post-isolation of uterine and umbilical arteries but were maintained in PSS solution to avoid desiccation. All fetuses and placentas were gently blotted dry to remove any excess fluid and weighed on a Metler AC100 microbalance (Greifensee, Zurich, Switzerland). Fetal biometric measurements (crown-rump (C/R) length, abdominal (Ab) circumference and head circumference) were recorded as outlined in figure 2.9 below (A-C). All placentas, and 2-3 fetal pups per litter, were snap frozen and stored at -80 ﾟC.

C/R Measurement Ab Measurement Head Measurement A B C

Figure 2.9: Images of fetal biometric measurements

Fetal measurements were assessed through the use of a short length of thread and a ruler. Measured were the distance from the corpus callosum on the head down to where the tail meets the flank of the animal (crown-rump length, A), the abdomen just above the umbilical insertion point (abdominal circumference, B) and head circumference, C.

2.15 Fetal Weight Distribution Curve Fetal weight distribution curves were constructed from all fetal weights collected in the study and according to Dilworth et al (Dilworth et al, 2011). The 5th percentile of fetal weight was calculated using the equation below:

(-Z score x SD) + MEAN

Where Z score = -1.645 and SD = standard deviation (Dilworth et al., 2011).

2.16 Blood vessel dissection and normalisation Adult maternal uterine artery and fetal umbilical artery were dissected and normalised as performed in the ex vivo studies (section 2.11). The mesentery was pinned down and second order mesenteric arteries were dissected from the surrounded tissues. Care was taken to completely remove surrounding adipose fat tissue and not overstretch arteries. Two vessels were dissected from separate branches of the mesenteric artery and mounted on two 40 µM steel wires and secured in the jaws of the myograph. These vessels were equilibrated and normalised in the same manner as the uterine arteries.

2.16.1 Wire myography

2.16.1.1 Protocol A schematic diagram of the full protocol can be found in figure 2.10. Following equilibration (20-30 mins), vessel viability was assessed using 2x KPSS (120 mM) with 10-20 min intervals in between. If the constriction of the vessel was less than 3.0 kPa it was excluded from the study. After each KPSS exposure, the chamber was emptied and filled with PSS to restore vessel to passive tension (baseline pre-constriction). Following KPSS constrictions and PSS wash-outs, a U46619 dose-response curve was performed (10-10 - 2x10-6 at 2 min intervals).

The calculation of the EC80 of U46619 was identical to the method used in human studies.

-2 - Following calculation of EC80, endothelium-dependent relaxation to ACh (10 -2x 10 6M at 2 min intervals) was carried out only on uterine and mesenteric arteries pre- constricted to an EC80 dose of U46619. Followed by a wash with PSS, endothelial- independent relaxation to SNP (10-10 – 2x10-6M) at 2 min intervals) was investigated in uterine, mesenteric and umbilical arteries pre-constricted with an EC80 dose of U46619. Following a further wash with PSS, vessels were left for 10 mins before a final KPSS constriction to ensure vessels were still functional and constriction had not been compromised.

U46619 Dose- response ACh Dose-

curve SNP Dose- 2X KPSS response KPSS curve response curve

EC80 of EC80 of U46619 U46619

EC80 of EC80 of Figure 2.10: A representative image of protocol for in vivo mouseU46619 experiments. U46619 Following normalisation, 2x KPSS washes, uterine arteries were constricted to U44619 to conduct a dose response curve. Following a washout with PSS, they were then pre-constricted to EC80 of U44619 to perform an ACh dose response curve followed by a PSS wash out and then a SNP dose response curve. The experiment was terminated with a final KPSS contraction.EC80 of X-axis denotes times and Y axis denotes Tension (mN/mm). U44619 = thromboxane A2U46619 mimetic.

2.17 Data analysis Fetal and placental weights at E18.5 were analysed using a two-way ANOVA with Sidak’s post-hoc test, this test was used because these measures were not my primary outcome and the N’s were too small to appropriately power for fetal weight. A two-way ANOVA allowed us to take into consideration the interaction between genotype and treatment of mice. The N being tested was litter size in most cases following the calculation of mean per litter but for the fetal distribution curve N was the individual fetal weights. The diameters of the uterine, mesenteric and umbilical arteries were analysed using an unpaired t-test based on their normality test outcome. U46619-induced constriction and ACh/SNP-induced relaxation were analysed using repeated measures two-way ANOVA and expressed as mean±SEM. The EC50 and EC80 values were assessed for normal distribution and the outcome of this would determine if data was analysed with non-parametric testing due to the small dataset. The myography data analysis was the same as that performed for human vessels. Significance was determined at p<0.05 and in these experiments N = female mice used and n= number of arteries used for that experiment.

100

Chapter 3: The acute effect of statins on vascular function of chorionic plate arteries in PE and normal pregnancies

101

3.0 Results

3.1 Introduction Pre-eclampsia (PE), identified through the development of new onset hypertension and proteinuria, is associated with considerable maternal and fetal morbidity and mortality (Lorquet et al., 2010). PE is usually resolved upon delivery of the placenta, but not in all cases, (Lorquet et al., 2010). It is believed that PE consists of a pre-clinical and clinical stage; the pre-clinical, or first, stage involves poor placentation as a result of inadequate spiral artery remodelling. The clinical, or second, stage of PE is thought to be attributed to uteroplacental ischemia which contributes to the hypertension and multi-organ failure identified in PE (Rana et al., 2019). This ischemia also diminishes the levels of antioxidants such as heme oxygenase (HO-1) and raises soluble endoglin (sEng) levels (El-Sayed, 2017). Anti-angiogenic factors such as sEng and sFlt-1, secreted by the syncytiotrophoblast, are elevated in maternal blood in PE, whilst PlGF is reduced (Redman et al., 2014). The anti- angiogenic factor sFlt-1 has the ability to bind to and sequester the pro-angiogenic VEGF and PlGF (Cerdeira et al., 2018). Subsequently leading to glomerular endotheliosis (Maynard et al., 2003), this angiogenic imbalance contributes to endothelial dysfunction in maternal tissues (Maynard et al., 2003). Currently there is no cure for PE. Clinical management following diagnosis of PE is to manage hypertension through the use of anti-hypertensive medication and for increased fetal monitoring. A number of drugs have been suggested for treatment of this maternal disease, including statins (Phipps et al., 2019a). However, the use of statins during pregnancy is still controversial due to them being classed as “category X” and so they are contraindicated in women who are or may become pregnant. This is due to previous studies in animals or humans demonstrating fetal abnormalities and the risk of the use of the drug in pregnant women outweighing any possible benefit; however, there are limited data about its effects on the human placenta and fetus (Zarek et al., 2013). There have been concerns about statins’ effects on placental development and on the developing fetus should statins cross the placental barrier. Studies in human first trimester placental explants showed that 250 nM pravastatin and 50 nM cerivastatin abolished IGF-I and IGF- II stimulated cytotrophoblast proliferation, key contributors to placental development and fetal growth (Forbes and Westwood, 2008; Kenis et al., 2005). These effects have been suggested to be due to the decreasing expression of IGF-I receptor via the inhibition of N-

102

linked glycosylation and decreased MMP2 activity and HSP27 expression (Forbes et al., 2015; Tartakover-Matalon et al., 2007).

Statins are HMG-CoA reductase inhibitors and can be classed according to their hydrophobicity, being either hydrophobic, moderately hydrophobic or hydrophilic. The presence of more polar hydroxyl groups increases the hydrophilicity of a statin. Statins typically target the liver to reduce cholesterol synthesis and the more hydrophobic statins such as simvastatin passively diffuse through hepatocyte cell membranes while more hydrophilic statins like pravastatin and the moderately hydrophobic pitavastatin require active carrier-mediated transport (Schachter, 2005). Consideration of hydrophobicity is important when contemplating the use of statins in pregnancy as this will impact upon their potential to cross the placental syncytiotrophoblast into the fetal circulation.

On its surface, the placenta expresses multiple drug transporters, some of which are able to efflux exogenous and endogenous toxic compounds from the fetal side back into the maternal circulation, protecting the fetus. Nanovskaya et al. carried out dual perfusion of a placental lobule with 50 ng/mL pravastatin (serum concentration from a 40 mg dose) and showed that pravastatin was transferred to the fetal circulation. However, pravastatin was also transferred in a fetal-to-maternal direction, hypothesised to be via the action of the efflux transporters, MRP2 and BCRP, located on the MVM (Nanovskaya et al., 2013). The literature regarding drug transporter expression appears to be contradictory, with some showing a slight increase in expression of BCRP and P-gp gene expression in PE placentas (Mason et al., 2011). While others have reported a lower expression of P-gp and BCRP in placentas from women with PE compared to term placentas from uncomplicated pregnancies (Feghali et al., 2015). Currently, we are not aware if the expression of MRP2 also changes in PE like its other drug transporter counterparts. Each statin is able to cross the placental trophoblast layer but the extent to which they are eliminated (i.e. transferred back to the maternal circulation) by efflux transporters in the syncytiotrophoblast is not known for all statins, this process will help to determine the concentrations that can be reached in the fetal circulation. This will thus have consequences in terms of the potential deleterious effects of levels of statin exposure on the fetoplacental vasculature/fetus. There are trials investigating the effect of pravastatin in PE in terms of safety (Costantine et al., 2010), efficacy in early-onset PE (Ahmed, 2011) and prevention in a high-risk PE population (Costantine and Cleary, 2013).

103

Chorionic plate arteries (CPAs) branch off from the umbilical artery. They are resistance arteries and they, along with other fetoplacental small arteries, play an important role in regulating vascular tone in the fetoplacental circulation (Mulvany and Aalkjaer, 1990). CPAs are not the major determinants of vascular resistance in the fetoplacental circulation (Wareing et al., 2002) but rather the previllous arterioles within the placental cotyledon are the site of control (Myatt, 1992). In the human fetoplacental circulation, autonomic innervation is absent (Reilly and Russell, 1977; Spivack, 1943), as a result, vascular resistance in this vascular bed is controlled by humoral factors, autocrine mechanisms (Myatt, 1992) and circulating factors such as NO and changes in intrinsic smooth muscle tone within the villous resistance arteries which are further downstream of CPAs (Learmont and Poston, 1996). On the other hand, all arteries and arterioles from systemic vascular beds are innervated by the sympathetic nervous system (Mulvany and Aalkjaer, 1990). Overall the fetoplacental circulation has a much lower resistance relative to the maternal side; this is maintained by an increase in vasodilators such as NO as gestation proceeds, promoting fetal health (Myatt, 1992). The release of such vasodilators can be stimulated by physical factors such as shear stress in the vessels, hypoxia or transmural pressure (Myatt, 1992). It has been found that predominantly flow-mediated NO synthesis is important for small fetoplacental arteries, as endothelium-dependent relaxation (using acetylcholine, bradykinin etc) common in other arterial beds produces minimal relaxation in these vessels (Mccarthy et al., 1994b; Sladek et al., 1997).

So far, there have not been any studies looking at the acute effect of statins on the function of placental CPAs. Thus, the current study investigated how three statins, simvastatin (hydrophobic), pravastatin (hydrophilic) and pitavastatin (moderately hydrophobic), affect vascular function in CPAs from placentas from both normal pregnancies and those complicated by PE.

3.2 Hypothesis and aims

3.2.1 Hypothesis ‘Pravastatin, pitavastatin and simvastatin will not detrimentally affect vascular reactivity of chorionic plate arteries in placentas from normal or pre-eclamptic pregnancies.’

104

3.2.2 Aims 1. To assess whether vascular reactivity of chorionic plate arteries is altered in placentas from PE pregnancies compared to those from normal pregnancies. 2. To assess if vascular reactivity of chorionic plate arteries, from PE or normal pregnancies, is altered following short-term exposure to pravastatin, simvastatin and pitavastatin.

3.3 Materials and methods

3.3.1 Inclusion and Exclusion Criteria Normal pregnancy was defined as a women with an uncomplicated pregnancy with an appropriately grown fetus. Women with PE were included in the study if their features met the following PE definition, “…de-novo hypertension before 20 weeks and a new onset of proteinuria, fetal growth restriction and/or PlGF measurement of <12 pg/ml (indicative of placental insufficiency) (Brown et al., 2018; Duhig et al., 2019).

Placental samples were taken from women with a normal pregnancy (N=21) or a pregnancy complicated by PE (N=14). Women in the normal group delivered at term (37+0 to 42+0 weeks gestation), while women in the PE group often delivered before term (28+0 to 40+0 weeks gestation). In order to control for women in the obese category and also women of advanced maternal age seen in the PE group, a BMI and age cut-off of <35 kg/m2 and <40 years was utilised. This was slightly different from the PE group but more accurately reflects the demographics of the normotensive population. The average maternal age and BMI for all women in St Mary’s hospital is ~ 32 years and ~27 kg/M2 respectively. Fetuses of normotensive women fell between 10th - 90th centile. In order to allow sufficient sample numbers to power our studies from the PE group, a BMI and age cut-off of <40 kg/m2 and <45 years respectively was used.

Normotensive women were excluded if they presented with maternal diabetes, hypertension, or hyperthyroidism or had other existing pathology. Women with PE were excluded if they presented with diabetes, hyperthyroidism, twin pregnancies or if they had sexually transmitted diseases. Additionally, women with PE who participated in this study

105

had fetuses that were AGA, SGA or FGR. Demographic details for all the women whose placentas were used for this study are detailed in table 3.1

3.3.2. Sample collection and protocol The protocol has been discussed in detail in the methods section in chapter 2. Briefly, following collection of the placenta, CPAs were dissected out under a microscope and mounted onto a wire myograph. The optimal steady state for these vessels was determined by carrying out the normalisation process as outlined in chapter 2 section 2.6.2. Following normalisation, the CPAs were left in physiological solution (PSS) to equilibrate for 20-30 minutes to 0.9 of L5.1kPa.

For each experiment performed, a control, either DMSO or water, depending upon the vehicle in which each statin was dissolved, would be running in parallel, and these controls would be added to PSS at the beginning of the 2 hr incubation and left for the same amount of time as the statins. For example, for pitavastatin and simvastatin experiments, 4 CPAs would be isolated and allocated randomly to DMSO, 1 µM pitavastatin, 10 µM pitavastatin or 1 µM simvastatin. CPAs were tested for their viability via exposure to 2x KPSS to promote vasoconstriction, the threshold for exclusion following this test was 3 kPa. Following this, a dose response was conducted using the thromboxane mimetic U46619 (10-10 -10-5.7). In the rare cases where two out of the four vessels were excluded prior to statin incubation i.e. they failed the KPSS threshold, the remaining two vessels were randomly allocated to either treatment or control.

After the U46619 dose response, a 2 hr incubation with either pravastatin (1 µM or 10 µM), simvastatin (1 µM) or pitavastatin (1 µM or 10 µM) occurred. Following this incubation, a further dose response to U46619 was performed. In separate experiments (on a different myography machine), the effect of incubation with either statin or vehicle on endothelial-independent relaxation (SNP; 10-10- 10-5) was assessed. CPAs were pre- constricted with EC80 U46619; once vessels had constricted and plateaued with a stable constriction for 8-10mins, a SNP dose response was performed. At the end of the experiment, all statins were washed out with PSS followed by a final KPSS exposure to ensure the vessels remained viable at the end of the experiment.

CPA vessels were excluded from 22 normotensive pregnancies and 2 PE pregnancies. In experiments, vessels were excluded singly or in pairs depending on the reasons for their

106

exclusion. As long as 1 control and 1 treated vessel out of the 4 CPAs mounted met the inclusion criteria, the experiment was included. However, if 3 out of the 4 mounted vessels for the experiment did not meet inclusion criteria, the whole experiment was excluded. Depending on the size of the chorionic plate tissue, vessels were “paired” from the same arterial branch; however, in some cases vessels were “unpaired” and from different arterial branches. In the case of “paired” vessels, if exclusion of one vessel occurred in real time during the protocol, the other vessel was still used. However, if exclusions occurred during analysis, both vessels in the pair would be excluded. This occurred if U46619 pre- constriction values were highlighted as an outlier using the ROUT method. Figure 3.1 provides a schematic diagram outlining the number of experiments performed, including experiments excluded and why. Figure 3.2 shows an annotated representative image for the CPA protocol used in this study.

107

Total Experiments Performed

(PE and Normal)

(N=35, n=161)

NORMAL PREGNANCIES

Pravastatin, pitavastatin and simvastatin data combined (post-exclusions)

Total statin experiments: N=21, n=92

N=placentas, n=arteries

Pitavastatin Experiments Pravastatin Experiments Simvastatin Experiments (1 and 10µM) (1 µM) (1 and 10µM) U4 Post-incubation: U4 post-incubation: U4 post-incubation: N=10, n=20 N=7, n=8 N=11, n=22 SNP post-incubation: SNP post-incubation: SNP post-incubation: N=6, n=12 N=13, n=14 N=8, n=16

EXCLUSIONS

Pravastatin - 1µM and 10µM Pitavastatin - 1µM and 10µM Simvastatin - 1µM

Exclusions: Exclusions: Exclusions:

KPSS Constriction < 3 kPa KPSS Constriction < 3 kPa KPSS Constriction < 3 kPa

N= 4 N= 7 N= 4 n= 8 n= 12 n= 10

U46619 Pre-constriction value U46619 Pre-constriction value U46619 Pre-constriction an outlier an outlier value an outlier

N=3 N=2 N=2 n=9 n=5 n=6

108

PE PREGNANCIES

Pravastatin, pitavastatin and simvastatin data combined (post-exclusions)

Total statin experiments: N=14, n=69

N=placentas, n=arteries

Pitavastatin Experiments Pravastatin Experiments Simvastatin Experiments (1 and 10µM) (1 µM) (1 µM) U4 Post-incubation: U4 Post-incubation: U4 Post-incubation: N=10, n=21 N=6, n=7 N=6, n=12 SNP post-incubation: SNP post-incubation: SNP post-incubation: N=6, n=12 N=7, n=9 N=7, n=8

Exclusions

Pravastatin - 1µM and 10µM Pitavastatin - 1µM and 10µM Simvastatin - 1µM

Exclusions: Exclusions: Exclusions:

KPSS Constriction < 3 kPa KPSS Constriction < 3 kPa KPSS Constriction < 3 kPa N= 1 N= 1 N/A n= 3 n= 2 U46619 Pre-constriction U46619 Pre-constriction value U46619 Pre-constriction value value an outlier an outlier an outlier N/A N/A N/A

Figure 3.1: Flowchart of CPA samples used and exclusions

35 CPA experiments were performed overall. Data was excluded if the vessel contractions to KPSS were <3kPa or U46619. Pre-constriction value was an outlier using the ROUT method. N = number of placentas and n = number of arteries.

*Assessments of contraction and relaxation of CPAs post-statin incubation was not always carried out on the same placental tissue. The reasons for this included availability of myography machines on the day of experiment, the size of the placental tissue sample and the number of users of the tissue. This resulted in there being a different overall total number (N) for contraction and relaxation experiments.

109

U46619 Dose-response curve KPSS U46619 Dose- KPSS Response Curve

Normalization

Normalization Statin incubation of vessels (1µM, 2hrs)

Basal Tone calculation:

 The tension in (mN/mm) was recorded pre- and post-incubation with statin.  This value was then converted to kPa  Basal tone = post-statin tension (kPa) minus pre-statin tension (kPa)

Figure 3.2: A representative image of the full myography protocol for CPAs

Following normalisation, 2x KPSS washes and a U44619 dose response curve, vessels were randomly assigned to vehicle (DMSO) or 1 µM or 10 µM of a statin, each of which was dissolved in 6ml of PSS and left to incubate for 2 hrs in the myograph baths. Post-incubation, contraction was assessed by a U46119 dose response curve followed by a final KPSS contraction. X-axis denotes time and Y axis denotes tension (mN/mm). U44619 = thromboxane A2 mimetic.

110

3.4 Results

3.4.1 Demographics of Women in Study Following caesarean section, samples were taken from women with a normal pregnancy (N=21) or a pregnancy complicated by PE (N=14). Samples were considered ‘normal’ and included if the individualised birthweight ratio (IBR) was between the 10th-90th centile (indicating an appropriately grown fetus). An IBC <10th centile was classed as SGA and <5th defined as FGR. Those pregnancies resulting in a stillbirth were not recruited to the study. IBR was calculated using the UK BulkCentileCalculator GROW 8.04 as advised by the Perinatal institute (Perinatal., 2019). Demographic details for all the women whose placenta was used for this study are detailed in table 3.1 on the next page.

111

Table 3.1: Demographic and clinical data for women whose placentas were used in the study. Data are expressed as median and range. Booking blood pressures were acquired in antenatal clinic when the women were between 7 and 15 weeks gestation and maximum blood pressure is the highest blood pressure reading prior to delivery. Mann Whitney U test was performed. In PE cohort, 6/14 women were early-onset. No significant difference in mode of delivery between the two groups.

112

3.4.2 Diameter of chorionic plate arteries from normal and PE pregnancies prior to statin exposure The diameter of CPAs from PE and normal pregnancies was not significantly different (p=0.118) (Figure 3.3).

Figure 3.3: Diameter of CPAs from PE pregnancies and normal pregnancies prior to statin incubation.

CPA diameter from PE pregnancies was not statistically different from normal pregnancies prior to statin incubation. Data presented as average diameter per placenta for each group. The horizontal line denotes median and data are expressed as median±IQR; Mann-Whitney U test was performed.

113

3.4.3 Assessment of vascular reactivity of chorionic plate arteries between normal and PE pregnancies There was no significant difference in contraction to U46619 between CPAs from normal and PE pregnancies (Figure 3.4 A (p=0.700)). There remained no significant difference between normal and PE CPAs when data were expressed as contraction to U46619 dose as a % of the maximum KPSS contraction (Figure 3.4 B (p=0.930)). There was no significant difference in the relaxation of CPAs to SNP between normal and PE groups (Figure 3.4 C (p=0.623)). In figure 3.4, the constriction graphs include all vessels prior to exposure to statin or their relative controls. The endothelial-independent SNP dose response curves shown are for DMSO controls from normal pregnancies and DMSO controls from PE pregnancies (post-incubation). A SNP dose response curve was not conducted in PSS alone (pre-incubation).

When DMSO and water controls were separated to assess any differences between them there was no significant difference in CPA relaxation to SNP between DMSO and water controls from normal pregnancies (Figure 3.4. D (p=0.814)) or PE pregnancies (Figure 3.4. E (p=0.909)).

A B 15 Normal (N=21, n=92) 200 Normal (N=21, n=92) PE (N=14, n=69)

PE (N=14, n=69)

)

n P

10 o 150

t (

t i

K 100

s n

n 5

o T C 50

0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10 M [U46619]x 10xM

1 C

6 6

4 100

8 80

C E

o 60

i 40

t a

x 20 Normal (N=21, n=42)

a l

e PE (N=14, n=29) R

-10 -8 -6 -4 % [SNP] x 10xM

114

D E

Figure 3.4: Assessment of vascular reactivity in CPAs from normal and PE pregnancies.

Contraction to U46619 expressed as tension (kPa) (A) or as % KPSS (B) and relaxation of chorionic plate arteries to SNP (C) was not different between CPAs from normal v PE pregnancies. SNP relaxation was not different between DMSO and water control groups from normal or PE pregnancies (D and E). Graph A and B show the combined pre-incubation U46619 data for normal pregnancy (N=21) and PE (N=14) from pitavastatin, pravastatin and simvastatin experiments (at all concentrations), while graph C shows post-incubation CPA data with control groups from PE and normal pregnancy. Replicate vessels from the same pregnancy were averaged for each group. Data are expressed as mean±SEM; two-way ANOVA.

115

3.4.4 Diameter of chorionic plate arteries from normal pregnancies prior to statin (1µM) exposure The diameters of CPAs from normal pregnancies were not significantly different prior to exposure with either statin or vehicle (either water or DMSO) (Figure 3.5 A: Water vs pravastatin (p=0.684), Figure 3.5 B: DMSO vs pitavastatin (p=0.949) and Figure 3.5 C: DMSO vs simvastatin (p=0>0.999)).

A B

600 600

m 

 r

r 400 400

t e

e m

m 200 a

a 200

i D D 0 0

r M O M e  S  t 1 M 1 a D - W - n in ti t ta ta s s a a v v ta ra C i P P 600

m  400

e m

a 200

i D

O M S  1 M - D n ti ta s a v im S

Figure 3.5: Diameter of CPAs between statin and controls, prior to statin exposure, from normal pregnancies.

No difference in CPA diameter was observed between groups prior to exposure to pravastatin (1 µM), pitavastatin (1 µM) and simvastatin (1 µM) and their corresponding controls in CPAs from normal pregnancies. Data presented as average diameter per placenta for each group (not all statin experiments were from the same placenta). The horizontal line denotes median and data expressed as median±IQR; Mann -Whitney U test performed.

116

3.4.5 Diameter of chorionic plate arteries from normal pregnancies prior to statin (10 µM) exposure The diameter of CPAs from normal pregnancies were not significantly different prior to exposure with statins (Figure 3.6 A: Water vs pravastatin (p=0.931) and Figure 3.6 B: DMSO vs pitavastatin (p=0.456)).

A B

600 600 m

 

r 400

e t

m m

a 200

a 200 i

D D

0 0

r M O M te  S  a 0 0 1 M 1 W - D - n n ti ti ta ta s s a a v v a a r it P P

Figure 3.6: Diameter of CPAs between statin and controls, prior to statin exposure, from normal pregnancies.

No difference in CPA diameter was observed prior to exposure with either pravastatin (10 µM) or pitavastatin (10 µM) and their corresponding controls in normal pregnancies. Data presented as average diameter per placenta for each group (not all statin experiments were from the same placenta). The horizontal line denotes median and data are expressed as median±IQR; Mann-

Whitney U test performed.

117

3.4.6 Effect of statins (1 µM) on basal tone of chorionic plate arteries from normal pregnancies Following a 2 hr incubation with pravastatin (Figure 3.7 A: Water vs pravastatin (p=0.529), pitavastatin (Figure 3.7 B: DMSO vs pitavastatin (p=0.898)) and simvastatin, (Figure 3.7 C: DMSO vs simvastatin (p=0.336)) at 1 µM, there was no significant difference in basal tone of CPAs (0.9 of L5.1KPa) compared to their relative controls.

A B

Figure 3.7: Effect of 2 hr incubation with statins on basal tone of chorionic plate arteries relative to control group.

Following a 2 hr incubation with either pravastatin (1 µM, N=10, A), pitavastatin (1 µM, N=11, B) or simvastatin (1 µM, N=13, C), Basal tone of CPAs was not significantly different between statin- exposed group versus controls. The horizontal line denotes median and data expressed as median±IQR; Mann-Whitney U test performed.

118

3.4.7 Effect of statins (10 µM) on basal tone of chorionic plate arteries from normal pregnancies Following a 2 hr incubation with pravastatin (Figure 3.8 A: Water vs pravastatin (p>0.9999), and pitavastatin at 10 µM (Figure 3.8 B: DMSO vs pitavastatin (p=0.259)), there was no significant difference in basal tone of CPAs (0.9 of L5.1KPa) compared to their relative controls.

A B

Figure 3.8: Effect of 2 hr incubation with statins on basal tone relative to control group.

Following a 2 hr incubation with either pravastatin (10 µM, N=9, A) or pitavastatin (10 µM, N=7, B), basal tone of CPAs was not significantly different between statin-exposed group versus controls. The horizontal line denotes median and data expressed as median±IQR; Mann-Whitney U test performed.

119

3.4.8 Effect of statins (1 µM) on contraction of chorionic plate arteries from normal pregnancies Following a 2 hr incubation with statins at 1µM, there was no significant effect on CPA contraction to U46619 from normal pregnancies; (Figure 3.9 A: Water vs pravastatin (p=0.088), Figure 3.9 B: DMSO vs pitavastatin (p=0.924) and Figure 3.9 C: DMSO vs simvastatin (p=0.778)). When U46619 contraction curves were expressed as maximum KPSS contraction (%KPSS) there was again no significant difference between statin-exposed CPAs and comparable control CPAs; (Figure 3.9 D: Water vs pravastatin (p=0.974), Figure 3.9 E: DMSO vs pitavastatin (p=0.912) and Figure 3.9 F: DMSO vs simvastatin (p=0.973)).

No differences were observed in U46619 sensitivity (EC50 values shown in table 3.2).

A B C Effect of Statin (1M;2h) 20 Water (10) 20 DMSO (11) 20 DMSO (7) Pravastatin-1M (10) B Simvastatin-1M (7)

) Pitavastatin -1M (11)

) )

a 15 a

a 15 15

(

n n o 10 10

i 10

o i

T T 5 T 5 5 B 0 0 0 -10 -8 -6 -4 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10 M [U46619] x 10xM [U46619] x 10xM b

D E F

200 200 Water (10) DMSO (9) 200 DMSO (7) Pravastatin-1M (10)

Pitavastatin-1M (9) Simvastatin-1M (7)

n n

150 150 n 150

t t

100 100 t K

K 100

C C 50 50 C 50

0 0 0 -10 -8 -6 -4 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM [U46619] x 10xM

Figure 3.9: Effect of 2 hr statin incubation on contraction of CPAs from normal pregnan cies.

Following a 2 hr incubation with statins at 1µM, there was no significant effect on contraction of CPAs to U46619 from normal pregnancies expressed as tension (kPa) or as % KPSS; p>0.05. U46619 dose respo nse curves were compared using two-way ANOVA. Data are expressed as mean±SEM; number of placentas in parenthesis.

120

3.4.9 Effect of pravastatin and pitavastatin (10 µM) on contraction of chorionic plate arteries from normal pregnancies Following a 2 hr incubation with statins at 10 µM, there was no significant effect on CPA contraction to U46619 from normal pregnancies; (Figure 3.10 A: Water vs pravastatin (p=0.998) and Figure 3.10 B: DMSO vs pitavastatin (p=0.990)). When U46619 contraction curves were expressed as maximum KPSS contraction (%KPSS), there was no significant effect on contraction in 10 µM pravastatin versus water (Figure 3.10 C: Water vs pravastatin (p=0.998)) but 10 µM pitavastatin attenuated the contraction of CPAs from normal pregnancies (Figure 3.10 D: DMSO vs pitavastatin (p=0.026)).

A B

20 20 Water (9) DMSO (7)

)

Pravastatin - 10M (9) ) Pitavastatin - 10M (7) a

15 a 15

(

n 10 10

n e

e T 5

T 5

0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM C D

200 Water (9) 200 DMSO (7)

Pravastatin - 10M (9) Pitavastatin-10M (7) n

150 n 150

o i

i *

r t 100 t

K 100

o C 50 C 50

0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM

Figure 3.10: Effect of 2 hr statin incubation on contraction of CPAs from normal pregnancies.

Following a 2 hr incubation with pravastatin at 10 µM, there was no significant effect on contraction of CPAs to U46619 from normal pregnancies expressed as tension (kPa) or as % KPSS (A and C). 10 µM pitavastatin attenuated the contraction of CPAs from normal pregnancies when expressed as % KPSS only (D); p=0.026. U46619 dose response curves were compared using two-way ANOVA with Sidak’s post hoc test, where applicable (Sidak was chosen as it is more power than the Bonferroni test and assumes comparisons are independent of one another). Data are expressed as mean±SEM; number of placentas in parenthesis. *P<0.05

121

3.4.10 Effect of statins (1 µM) on relaxation of chorionic plate arteries from normal pregnancies following a 2 hr incubation Following a 2 hr incubation with pravastatin (Figure 3.11 A: Water vs pravastatin (p=0.945) and simvastatin (Figure 3.11 C: DMSO vs simvastatin (p=0.870)) there was no significant effect on CPA relaxation to SNP from normal pregnancies; 1 µM pitavastatin showed a trend towards increasing relaxation to SNP (Figure 3.11 B: DMSO vs pitavastatin (p=0.086)).

A B

1 120

9 6

1 120

6 4

U 100

100

8 8

C 80

60 60

n o

t a

40 a 40

l e

e 20 Water (4) 20 DMSO (8)

Pitavastatin-1M (8) Pravastatin-1M (4) % % 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [SNP] x 10xM [SNP] x 10 M C

1 120

6 4

U 100

0 8

C 80

o t

i t

a 40

a l

e 20 DMSO (13) R Simvastatin-1M (13) % 0 -10 -8 -6 -4

[SNP] x 10xM

Figure 3.11: Effect of 2 hr statin incubation on relaxation of CPAs from normal pregnancies.

Following a 2 hr incubation with pravastatin, pitavastatin and simvastatin at 1 µM, there was no significant effect on relaxation of CPAs to SNP from normal pregnancies; p>0.05. SNP dose response curves were compared using two-way ANOVA with Sidak’s post hoc test, where applicable (Sidak was chosen as it is more power than the Bonferroni test and assumes comparisons are independent of one another). Data are expressed as mean±SEM; number of placentas in parenthesis.

122

3.4.11 Effect of pravastatin and pitavastatin (10 µM) on relaxation of chorionic plate arteries from normal pregnancies following a 2 hr incubation Following a 2 hr incubation with statins there was no significant effect on relaxation to SNP in CPAs from normal pregnancies; (Figure 3.12 A: Water vs pravastatin (p=0.928), Figure 3.12 B: DMSO vs pitavastatin (p=0.147)).

A B

9 1 1 120

6 120

4 U

100 U 100

8 C

C 80 80

t t

60 60

i t

t 40 40

a a

x x

a DMSO (7)

a l

20 Water (6) e 20

R Pitavastatin -10M (7) Pravastatin-10M (6) 0 % 0 % -10 -8 -6 -4 -10 -8 -6 -4 [SNP] x 10xM [SNP] x 10xM

Figure 3.12: Effect of 2 hr pravastatin and pitavastatin incubation on relaxation of CPAs from normal pregnancies.

Following a 2 hr incubation with pravastatin and pitavastatin at 10 µM, there was no significant effect on relaxation of CPAs to SNP from normal pregnancies; p>0.05. SNP dose response curves were compared using two-way ANOVA with Sidak’s post hoc test, where applicable (Sidak was chosen as it is more power than the Bonferroni test and assumes comparisons are independent of one another). Data are expressed as mean±SEM; number of placentas in parenthesis.

123

3.4.12 Diameter of chorionic plate arteries from PE pregnancies prior to statin exposure (1µM) The diameter of CPAs from pregnancies with PE were not statistically different prior to treatment with statins Water 401[129-548] vs pravastatin 401[118-556] (p=0.710), DMSO 339[317-442] vs pitavastatin 290[222-430] (p=0.589) and DMSO 352[268-486] vs simvastatin 244[138-488] (p=0.456)). Data are expressed as median±IQR, Mann-Whitney U test performed.

3.4.13 Diameter of chorionic plate arteries, prior to statin exposure (10µM), from PE pregnancies The diameter of CPAs from pregnancies with PE were not statistically different prior to treatment with pitavastatin DMSO 333[187-486] vs pitavastatin 296[169-470] (p=0.971). Data are expressed as median±IQR, Mann-Whitney U test performed.

124

3.4.14 Effect of statins (1 µM) on basal tone of chorionic plate arteries from pregnancies with PE Following a 2 hr incubation with pravastatin (Figure 3.13 A: Water vs pravastatin (p=0.517)), pitavastatin (Figure 3.13 B: DMSO vs pitavastatin (p=0.342)) and simvastatin (Figure 3.13 C: DMSO vs simvastatin (p=0.383)) at 1 µM, there was no significant difference in basal tone of CPAs from pregnancies with PE (0.9 of L5.1KPa) compared to their relative controls.

A B

Figure 3.13: Effect of 2 hr incubation with statins on basal tone of CPAs relative to control group.

Following a 2 hr incubation with either pravastatin (1 µM, N=10, A), pitavastatin (1 µM, N=8, B) or simvastatin (1 µM, N=7, C), basal tone of CPAs was not significantly different between statin-exposed group versus controls. The horizontal line denotes median and data expressed as median±IQR; Mann- Whitney test performed.

125

3.4.15 Effect of pitavastatin (10 µM) on basal tone of chorionic plate arteries from pregnancies with PE Following a 2 hr incubation with pitavastatin at (10 µM), there was no significant difference in basal tone of CPAs from pregnancies with PE (0.9 of L5.1KPa) compared to its relative controls; (Figure 3.14 A (DMSO vs pitavastatin; (p=0.143)).

Figure 3.14: Effect of 2 hr incubation with pitavastatin on CPA basal tone relative to control group.

Following a 2 hr incubation with pitavastatin (10 µM, N=10), CPAs did not show any significant difference in their basal tone compared to their controls. The horizontal line denotes median and data

are expressed as median±IQR; Mann-Whitney U test performed.

126

3.4.16 Effect of statins (1 µM) on contraction of chorionic plate arteries from pregnancies with PE Following a 2 hr incubation with statins at 1 µM, there was no significant effect on contraction of CPAs to U46619 from PE pregnancies (Figure 3.15 A: Water vs pravastatin (p=0.561), Figure 3.15 B: DMSO vs pitavastatin; (p=0.971) and Figure 3.15 C: DMSO vs simvastatin (p=0.387)). There was again no difference between any groups when U46619 contraction curves were expressed as maximum KPSS contraction (%KPSS) (Figure 3.15 D (Water vs pravastatin (p=0.535), Figure 3.15 E: DMSO vs pitavastatin (p=0.891) and Figure 3.15 F: DMSO vs simvastatin (p=0.785)).

A B C 15 15 DMSO (6) Water (6) 15 DMSO (6)

Pravastatin-1M (6) Simvastatin - 1M (6)

) )

) Pitavastatin - 1M (6)

a P

P 10

P k

10 10 k

(

s s

n 5

5 5 n

e T

T T

0 0 0 -10 -8 -6 -4 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10 M [U46619]x 10xM [U46619]x 10xM

D E F

25 0 Water (6) 250 DMSO(6) 250 DMSO (6) Pravastatin-1M (6) Simvastatin - 1M (6) 200 200

n Pitavastatin - 1M (6) n

o 200

n o

S c

150 S

S c

i 150

c r

150 P

s n

100 n

% 100 n %

o 100

C C 50 C 50 50

0 0 0 -10 -8 -6 -4 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10xM [U46619]x 10 M [U46619] x 10xM

Figure 3.15: Effect of 2 hr statin (1 µM) incubation on contraction of CPAs from

pregnancies with PE.

Following a 2 hr incubation with statins at 1 µM, there was no significant effect on contraction of CPAs to U46619 from PE pregnancies expressed as tension (kPa) or as % KPSS; p>0.05. U46619

dose response curves were compared using two-way ANOVA. Data are expressed as mean±SEM; number of placentas in parenthesis. .

127

3.4.17 Effect of pitavastatin (10 µM) on contraction of chorionic plate arteries from pregnancies with PE Following a 2 hr incubation with pitavastatin at 10 µM, there was no significant effect on contraction of CPAs to U46619 from pregnancies with PE (Figure 3.16 A: (DMSO vs pitavastatin (p=0.245)). When U46619 contraction curves were expressed as % of maximum KPSS contraction (% KPSS) pitavastatin treated CPAs showed a decreased contractile response; (Figure 3.16 B: DMSO vs pitavastatin (p=0.0003)).

15 DMSO (10)

) Pitavastatin - 10M (10) a

P 10

(

i s

n 5

e T

-10 -8 -6 -4 [U46619]x 10xM

B 250 DMSO (10)

200 Pitavastatin - 10M (10) n

o ***

t S c 150

n 100

o C 50

0 -10 -8 -6 -4 [U46619] x 10xM Figure 3.16: Effect of 2 hr statin incubation on contraction of CPAs from PE pregnancies.

Following a 2 hr incubation with pitavastatin at 10 µM, there was no significant effect on contraction of CPAs to U46619 from normal pregnancies expressed as tension (kPa) or as % KPSS; p>0.05. U46619 dose response curves were compared using two-way ANOVA with Sidak’s post hoc test, where applicable (Sidak was chosen as it is more power than the Bonferroni test and assumes comparisons are independent of one another). Data are expressed as mean±SEM; number of placentas in parenthesis. ***P<0.001

. 128

3.4.18 Effect of statins (1 µM) on relaxation of chorionic plate arteries from pregnancies with PE following a 2 hr incubation Following a 2 hr incubation with pravastatin (Figure 3.17 A: Water vs pravastatin (p=0.913), pitavastatin (Figure 3.17 B: DMSO vs pitavastatin (p=0.961) and simvastatin (Figure 3.17 C: DMSO vs simvastatin (p=0.496)), there was no significant effect on CPA relaxation to SNP from pregnancies with PE.

9 A

1 B 6

9 6

120 1 6

4 120

U 4

100

0 U

100

0 C

80 8

E C

E o

t o

60 t

o t 40 i t 40

x x

a DMSO (6) a

l Water (7)

20 l 20

e e

R Pravastatin-1 M (7)

 R Pitavastatin - 1M (6)

0 0 % % -10 -8 -6 -4 -10 -8 -6 -4 x [SNP] x 10xM [SNP] x 10 M

C 9

1 120

6 6

U 100

0 8

C 80

o t

o i

t 40

a x

a l 20

e DMSO (7) R Simvastatin - 1M (7)

% 0 -10 -8 -6 -4 x [SNP] x 10 M

Figure 3.17: Effect of 2 hr statin incubation on relaxation of CPAs from

pregnancies with PE.

Following a 2 hr incubation with pravastatin, pitavastatin and simvastatin at 1 µM, there was no significant effect on relaxation of CPAs to SNP from pregnancies with PE; p>0.05. SNP dose

response curves were compared using two-way ANOVA. Data are expressed as mean±SEM; number of placentas in parenthesis.

129

3.4.19 Effect of pitavastatin (10 µM) on relaxation of chorionic plate arteries from pregnancies with PE following 2 hr incubation Following a 2 hr incubation with pitavastatin there was no significant effect on CPA relaxation to SNP from pregnancies with PE; (Figure 3.18 A: DMSO vs pitavastatin (p=0.195)).

9 1

6 120

6 4

U 100

0 8

C 80 E

o t

o i

t 40 a

x a l 20 DMSO (6) e

R Pitavastatin - 10M (6)

0 % -10 -8 -6 -4 [SNP] x 10xM

Figure 3.18: Effect of 2 hr pitavastatin incubation on relaxation of CPAs from PE pregnancies.

Following a 2 hr incubation with pitavastatin at 10 µM, there was no significant effect on relaxation of CPAs to SNP from pregnancies with PE; p>0.05. SNP dose response curves were compared using two-way ANOVA. Data are expressed as mean±SEM; number of placentas in parenthesis.

130

Table 3.2: Summary of measurements of vascular reactivity with/without statin incubation in CPAs from normal and pre-eclamptic pregnancies.

KPSS (120 mM high potassium salt solution); U46619 (2x10-6 M thromboxane-A2 mimetic). Changes in the maximal response (Rmax) and sensitivity (log EC50) to U46619 (kPa and %KPSS) in chorionic plate arteries untreated and incubated for 2 hr with statins and controls (DMSO and water) shown. Residual constriction following relaxation to SNP and sensitivity (log EC50) also shown. Basal tone of CPAs following 2 hr incubation included. Rmax data is expressed as mean±SEM and sensitivity data is expressed as mean±SEM.

CPA Maximal SNP relaxation Basal tone (kPa) condition contraction

Normal Rmax Rmax logEC50 Residual logEC50 U46619 constriction (% to (KPa) (kPa) Kpss)

Untreated 13±2.7 166±33 -7.24±0.13 29±13.0 -7.43±1.34 - .5

DMSO 12±2.5 153±30 -7.27±0.05 27±13.1 -7.27±1.55 -0.29±1.31 .9

(combined)

Water 11±2.3 164±32 -7.25±0.21 25±13.2 -6.85±0.71 -0.19±0.17 .9 (1µM)

Pravastatin 14±2.8 165±32 -7.31±0.15 25±13.7 -7.66±0.28 -0.14±0.24 statin .7 (1µM)

Pitavastati 14±2.8 167±34 -7.27±0.06 28±12.9 -7.90±1.38 -0.3±1.68 n (1µM) .1

Simvastati 14±2.7 161±31 -7.32±0.05 30±11.5 -6.54±0.52 0.058±1.21 n (1µM) .8

Pravastatin 11±2.4 174±35 -7.29±0.04 47±12 -7.80±0.23 0.0005±0.29 statin .9 (10µM)

Pitavastati 12±2.5 141±28 -7.31±0.05 23±12.7 -7.99±0.74 -0.46±0.12 n (10µM .2

131

Untreated 12±2.4 172±34 -7.18±0.14 35±11.2 -7.21±2.20 - .9

DMSO 12±2.4 160±32 -7.13±0.05 29±11.5 -6.18±3.82 -0.06±0.16 .1 (1µM)

Water 9±1.7 156±29 -6.95±0.33 33±10.3 -6.91±0.49 0.66±1.58 .9 (1µM)

Pravastatin 7±1.4 183±35 -6.91±0.38 44±9.2 -6.70±0.27 0.49±0.75 statin .3 (1µM)

Pitavastati 11±2.3 153±30 -7.12±0.08 31±11.6 -7.74±3.00 -0.30±0.31 n (1µM) .3

Simvastati 10±2.06 154±29 -7.04±0.10 19±14.2 -8.83±1.15 -0.66±0.16 n (1µM) .8

Pitavastati 11±2.1 127±24 -7.01±0.07 25±57.9 -9.12±1.32 -0.98±0.19 n (10µM) .9

3.5 Discussion This study set out to determine whether short-term statin exposure altered vascular reactivity of CPAs in placentas from normal and PE pregnancies. Contraction and endothelium-independent relaxation of CPAs from normal pregnancy was not significantly different compared to CPAs from PE pregnancy. In general, the vascular function of CPAs from normal pregnancies following statin incubation did not show altered contraction to the thromboxane A2 mimetic U46619 or relaxation via the NO donor sodium nitroprusside. However, incubation of CPAs with 10 µM pitavastatin appeared to attenuate contraction. Similarly, in CPAs from pregnancies complicated by PE, short-term statin exposure did not alter vascular reactivity to U46619 or SNP with the exception of 10 µM pitavastatin which reduced contraction of CPAs to U46619. Overall, the data from the present study have demonstrated that suprapharmacological doses of statins do not appear to overtly affect

132

vascular function of CPAs which, from a safety point of view, is the most desired outcome for clinical translation.

3.5.1 Myography studies of Human Chorionic Plate Artery Function

Wareing et al have previously normalised CPAs at 0.9 of L5.1 kPa (equivalent to ~20-30 mm Hg). These values have been informed by Doppler ultrasound sound measurements of umbilical cord blood flow in vivo; physiological blood pressure values range from 60 mmHg at the umbilical end of the cord to 25 mmHg near the placental end (Kleiner-Assaf et al., 1999) which suggested a pressure of ~20 mmHg in CPAs (Wareing et al., 2006a).This reflects physiological blood pressure in the fetoplacental circulation and thus prevents endothelial damage, allowing for reproducible contraction and relaxation (Wareing et al., 2006a) and differs from systemic vessels, which are normalised at 0.9 of L13.3kPa (equivalent to 100 mm Hg physiological blood pressure) (Kusinski et al., 2009; Wareing et al., 2002; Wareing et al., 2006a). Due to the CPAs being normalised in this way, functional responses of the CPAs to vasoconstrictors and endothelial-independent vasodilators can therefore be assessed independent of the length and diameter of the CPAs. Vessels were exposed to 5 % O2 thus reflecting the lower oxygen levels present in the feto-placental circulation compared to maternal circulation (20 % O2). Whilst wire myography is a well-established and robust method of investigating vascular function, limitations include the ability to overstretch a vessel hence distort its shape, and the possibility of damaging the endothelium whilst mounting the vessel onto the myograph machine; this can contribute to a nonphysiological geometry of the vessel (Lu and Kassab, 2011). As such, functionality of vessels is assessed prior to statin exposure. Furthermore, it should be noted that there is an absence of physiological stimuli such as shear stress during the myography protocol, which is important in modulating vascular resistance and blood flow in vessels through vasoactive molecules and endocrine signalling (Rodriguez and Gonzalez, 2014).

3.5.2 Vascular Reactivity of Chorionic Plate arteries from Normal and PE Pregnancies

3.5.2.1 Potassium- and U46619- induced contraction of chorionic plate arteries from normal and PE pregnancies We showed that there is no significant difference in contraction of CPAs between normal and PE pregnancies, whether expressed as kPa (Figure 3.4, A) or as % KPSS (Figure 3.4, B).

133

When the mean maximum KPSS responses were compared between normal and PE pregnancies, no difference was observed (p=0.528; data not shown). Furthermore, vessels from PE pregnancies were not excluded at a higher rate than those from normal pregnancies due to a lack of constriction (<3 kPa). Due to the limited sample size, it was not possible to do a sub analysis according to PE type. However, the difference observed in contractility between these two groups could be related to the smaller diameters in the PE group, this could have contributed to smaller contractions relative to CPAs from normal pregnancies (Figure 3.3 – although not significant).

Functional studies have demonstrated conflicting results regarding aberrant dysfunction of CPAs from women with PE compared to CPAs from normotensive women. Ong et al. did not see a difference in constriction in CPAs between PE pregnancy and normal when CPAs were constricted using vasopressin (Ong et al., 2002). This resembles what was observed in this study despite our use of U46619 as a constrictor. Similarly, Ong et al. used a similar PE definition to ours, used samples from a similar gestation (238 days), did not separate their PE group into early and late onset or exclude PE pregnancies complicated by FGR. However, they have used a higher proportion of vaginal deliveries in their study and have not reported the mean BMI of women in their study. In contrast, Wareing et al. demonstrated a reduced maximal contraction in CPAs from pregnancies complicated by either PE or FGR compared with normal pregnancy when these vessels were constricted to U46619 (Wareing and Baker, 2004). The different results obtained between the study by Wareing et al. and the current study could be due to differences in gestation at delivery, as Wareing et al. used PE and normal samples from women at term (mean gestation: 38 weeks). Whereas the women in this current study were a combination of early and late onset PE (28 weeks - 39 weeks) and there were more caesarean sections in the current study whereas it is unclear how many deliveries were vaginal births and caesareans in Wareing et al. study. Previous research using the isolated perfused placental cotyledon has demonstrated that placentas from pre-term pregnancies contract more to angiotensin-II compared to placentas from normal term pregnancies. Furthermore, it was shown that, in basal conditions, pre-term placentas produce 7.6x more thromboxane than normal term placentas, which may partially contribute to a potentiated response to angiotensin II (Cruz et al., 2000). Similar to Ong et al. BMI was not reported in the study by Wareing et al. so it is uncertain the role this would have played in the outcome of their results. In addition, Wareing et al. separated FGR and PE groups whereas the data here from PE also contains 134

a sub-group of FGR cases. In a study by Benoit et al. CPAs from PE placentas demonstrated significantly greater contraction to 100 mM KCL versus normal placentas using pressure myography (Benoit et al., 2007). This was not seen in the current study, most likely due to the use of a different agonist to constrict CPAs and their small PE dataset, as they used a total of 48 placentas (43 normal and 5 PE) but also they used pressure myography.

Furthermore, their vessels were exposed to a gas mixture consisting of 95 % O2 and 5 %

CO2 vs 5 % O2 and 5 % CO2 in the current study. Vessels from the Benoit et al. study were bathed in Krebs-Henseleit physiologic saline solution, which differs from the composition of the physiological salt solution used in the current study. Finally, Benoit et al. did not report the full demographics of the women used in their study so information about BMI and maternal age are missing. Bertrand et al. also demonstrated that CPAs from PE pregnancies developed greater active tension than CPAs from normal pregnancies but they used potassium chloride and serotonin to constrict vessels, which is different from the current study (Bertrand et al., 1993).

PE may occur with or without FGR and both FGR and PE are associated with an increased risk of early delivery of the baby (Grill et al., 2009). Fetal growth restriction is defined as the failure of a fetus to reach its genetic growth potential according to the race and gender of the fetus (Sharma et al., 2016) and it is believed that placental dysfunction is a useful diagnostic indicator for this condition along with fetal size and growth trajectory (Benton et al., 2016). The cause of the placental dysfunction observed in FGR is thought to be similar to PE; there is an inadequate remodelling of spiral arteries in the endometrium that ultimately supply blood to the intervillous blood space of the placenta (Brosens et al., 2011). As the PE group here also involves women who had pregnancies complicated by FGR, it is important to acknowledge CPA function in FGR. In FGR, CPAs show greater U46619-mediated contraction than CPAs from normal pregnancy (Mills et al., 2005) and enhanced thromboxane-A2 production has been observed within FGR placentas (Boura et al., 1994). This was also shown by Wareing et al. who observed increased U46619 arterial contraction in hyperoxia in CPAs from FGR pregnancies compared to CPAs from normal pregnancies (20% oxygen) (Wareing et al., 2006a). The dataset from the current study was also split into FGR and non-FGR PE pregnancies and compared to normal pregnancies (data not shown) to investigate differences in CPA contraction to U46619. No significant difference to U46619-induced contraction was observed (FGR vs Normal: p=0.752 for kPa and p=0.486 for %KPSS; N=7 and N=21 respectively; 2-way ANOVA); PE alone vs Normal: 135

p= 0.959 for kPa and p=0.988 for %KPSS; N=7 and N=21 respectively; 2-way ANOVA). When FGR was directly compared to PE pregnancies (p=0.924 for kPa and p=0.534 for %KPSS; N=7 and N=7 respectively; 2-way ANOVA). Presently, the data does not support the current literature suggesting that FGR pregnancies are more contractile and hence it may not have been a confounding factor in the study.

It is believed that in PE there is a rise in thromboxane A2 synthesis by the placenta, triggering endothelial and platelet changes locally which eventually contribute to the ischemic changes in the placenta (Liu et al., 1998). In FGR, the elevated constrictor response of placental vessels is thought to be associated with an increase in the thromboxane receptor expression or hypersensitivity to thromboxane (Walsh, 1985), reduced production of or sensitivity towards agonists that cause vasodilation such as NO and endothelial- derived hyperpolarizing factor (Molnar and Hertelendy, 1992). This being said, this study used the vasoconstrictor U46619, a thromboxane mimetic, and did not see any hyper- contractility to this agonist relative to CPAs from normal pregnancies.

3.5.2.2 Sodium nitroprusside-induced (endothelial-independent) relaxation of chorionic plate arteries from normal and PE pregnancies We showed there is no significant difference in CPA relaxation to SNP between PE and normal pregnancies (Figure 3.4, C).In the experiments assessing relaxation in CPAs, SNP was the vasoactive agent of choice since it has been demonstrated that CPAs show minimal relaxation responses to endothelium-dependent agonists such as bradykinin (Mccarthy et al., 1994b), including in women with FGR or PE (Ong et al., 2003).

In order to maintain low fetoplacental vascular resistance, nitric oxide is produced by NOS enzymes in placental endothelium, which promotes vasodilation and therefore has an important role in vascular tone within the fetoplacental circulation (Learmont and Poston, 1996). In CPAs from PE pregnancies, it has previously been described that there is blunted vasodilation to SNP and the phosphodiesterase-inhibitor papaverine (Ong et al., 2002). In contrast to this, in FGR pregnancies, CPAs show pronounced SNP-mediated vasodilation which is considered to be a compensatory response to the vascular resistance observed (Mills et al., 2005). In addition to a deficit in NO availability, ROS production is also thought to cause endothelial damage and to contribute to blunted relaxation and raised superoxide levels that have been detected in placentas from PE pregnancies (Sikkema et al., 2001). Finally, Kingdom and Kaufmann suggested that, in both PE and FGR

136

pregnancies, there is an adaptation that takes place to compensate for the hypoperfusion in the placenta and this is through an increased capacity for the smooth muscles to relax but also an angiogenic response to counteract the resistance in placental villous capillaries (Kingdom and Kaufmann, 1997).

The data from this study contrast with the results mentioned above, in that we do not see blunted or enhanced relaxation in CPAs from PE pregnancies but rather no difference from normal. The dataset from the current study, as mentioned above, was split into FGR and non-FGR PE pregnancies and compared to normal pregnancies (data not shown) to investigate differences in CPA relaxation to SNP. No significant difference to SNP- mediated relaxation was observed (FGR vs Normal: p=0.283; N=5 and N=16 respectively; 2- way ANOVA); (PE alone vs Normal: p= 0.974; N=6 and N=16 respectively; 2-way ANOVA). When FGR was directly compared to PE pregnancies, no significant difference observed (p=0.420; N=5 and N=6 respectively; 2-way ANOVA). Presently, the current data in this chapter does not support the current literature suggesting that CPAs from FGR pregnancies are significantly more responsive to SNP or that CPAs from PE pregnancies show blunted relaxation to SNP. Hence, inclusion of FGR complicated PE pregnancies may not have been a confounding factor in the current study. Additionally, the reasons for the differences seen could also be attributed to the fact that we were required to use the solvent DMSO as the control in order to dissolve pitavastatin and simvastatin. DMSO is believed to stimulate relaxation of vessels by increasing the production of cGMP as NO is released from endothelial cells (Kaneda et al., 2016). This being said, when DMSO relaxation curves were compared to water relaxation curves in our study looking at CPAs, there was no significant difference between the SNP-induced relaxation between these two controls (Figure 3.4 D and E). This outcome suggests that DMSO is unlikely to be a significant reason for the disparity between our findings and those of Mills et al. and Ong et al.

Furthermore, our cut-offs for BMI and maternal age is a factor to consider when interpreting our results; however, a significant difference in these two factors were not seen in the current placenta dataset for PE vs normotensive group. For our normotensive group, the inclusion criteria were BMI: <35 kgm2 (median of 23.4) and age <40 years (median of 33) while for the PE group, inclusion criteria were BMI: < 45 with a (median of 25.6) and age <45 years (median of 30).With regards to the studies mentioned thus far, no group appeared to report BMI values for women who donated samples for their research

137

so it is uncertain whether BMI was a controlled factor. The median age for the normotensive group is similar to the literature quoted above; mean age was 31 for Ong et al.; 30 for Wareing et al. and 32 for Mills et al. For the PE group, Ong et al. and Wareing et al. reported an average maternal age of 29; Mills et al. had a mean age 29 for their FGR group similar to Wareing et al. which once again is similar to that seen in this current study. Benoit et al. did not report the age or BMI of women in their study, nor was BMI reported for the other studies mentioned above (Benoit et al., 2007; Mills et al., 2005; Ong et al., 2002; Wareing and Baker, 2004; Wareing et al., 2006a).

The BMI cut-off point used for women with PE in the current study reflects the fact that, in this population, a few women with PE also presented with obesity. Previous studies have suggested that relaxation of vessels from obese individuals is impaired as a result of oxidised LDL and inflammation in this condition causing endothelial dysfunction (Csige et al., 2018; Doupis et al., 2011; Hayward et al., 2013). With this is mind, the present study did not show a significant difference in BMI between women with normotensive and PE pregnancies, suggesting that BMI did not contribute towards the contractile or relaxatory responses of CPAs. Similarly, a maternal age greater than 35 years is a risk factor for PE (Lean et al., 2017a) and is associated with endothelial dysfunction. This should be considered in the context of the current study as the cut off point for maternal age was 40 years for the normotensive women and 45 years for women with PE. Advanced maternal age has been associated with enhanced SNP-induced relaxation of CPAs relative to young controls (Lean et al., 2017a) as well as attenuated constrictor capacity to KCL in uterine arteries from pregnant rats (Wight et al., 2000). Furthermore, ageing has been associated with raised vascular stiffness and endothelial dysfunction (Camici et al., 2015). Nevertheless, the present study did not show a significant difference in maternal age between women with normotensive and PE pregnancies, suggesting that maternal age did not contribute towards the contractile or relaxatory responses of CPAs.

Other potential confounders which could have affected the results in the current study include the presence of smokers in PE group, as smoking is associated with the production of carbon monoxide (CO) and CO can inhibit the anti-angiogenic protein sFlt- 1, offering protection against PE (Karumanchi and Levine, 2010; Wei et al., 2015). Furthermore, factors such as the effects of analgesia (Samanta et al., 2018) and the use of anti-coagulants (Atallah et al., 2017) such as aspirin and heparin could alter vascular

138

responses measured by wire myography. Finally, it is uncertain how statins interact with antihypertensive drugs such as nifedipine which are commonly used in PE management and how this interaction could affect the vascular function of placental CPAs.

3.5.3 Effect of short-term statin treatment on CPAs from normal pregnancies

3.5.3.1 Potassium- and U46619- induced contraction of chorionic plate arteries from normal pregnancies Following a 2 hr incubation with statins, neither pravastatin, pitavastatin or simvastatin attenuated constriction when expressed as kPa or % KPSS in CPAs from normal pregnancies (figures 3.7 and 3.8). However, pitavastatin at 10 µM attenuated constriction of CPAs when data was expressed as % KPSS (figure 3.8, D). It is important to remember that this dose is suprapharmacological and 40 x higher than a useful therapeutic dose in a non-pregnant individual on treatment. The U46619-induced contraction responses of the CPAs were presented as a percentage of maximal KPSS contraction to assess any changes in vascular smooth muscle mass or function. Potassium channels are key regulators of vascular smooth muscle (VSM) contraction and growth but also influence release of Ca2+ and Ca2+

+ + sensitivity. There are 5 classes of K channels: K (BKCa) channels, intermediate-conductance

2+ + + Ca -activated K (KCa3.1) channels, multiple isoforms of voltage-gated K (KV) channels,

+ + ATP-sensitive K (KATP) channels, and inward-rectifier K (KIR) channels (Jackson, 2017).

Mistry et al. demonstrated that placentas from PE pregnancies had an upregulation of KCNQ3 and KCNE5 KV7 channels, which could influence many areas of placental function, such as vascular tone and placental proliferation (Mistry et al., 2011). Similarly, Wei et al. showed an up-regulation in mRNA levels of KCNQ3, whereas those for KCNQ4 and KCNQ5 were down regulated in CPAs from PE. They also showed that this altered expression is believed to alter Kv7 channel function in CPAs, increase U46619 constriction and basal tone leading to the increased vascular resistance seen in PE pregnancies (Wei et al., 2018). Moreover, these results imply that the impaired function of Kv7 channels and altered expression of KCNQ isoforms in CPA vessels from PE women are concomitant with the hypertensive state of preeclampsia, which contributes to the increased vascular resistance of CPAs followed by reduced fetoplacental blood flow. Pitavastatin at 10 µM could be attenuating constriction by reducing proliferation of smooth muscle cells (reduction in VSM growth) or reducing depolarization triggered by potassium channels affecting membrane

139

potential; nevertheless, 2 hrs is a short time for the smooth muscle to be modified in this way.

Studies have shown that thromboxane receptors are present in placental tissue (Hedberg et al., 1989) and are the major vasoconstrictor in the villous vascular tree (Myatt, 1992). Furthermore, work by Kwek et al. has shown that the thromboxane mimetic U46619 is effective in constricting conduit and resistance arteries from PE and normal pregnancies (Kwek et al., 2001). The vasodilator NO is important in maintaining basal vascular resistance in the fetoplacental circulation and helps to antagonize the action of vasoconstrictors (Myatt, 1992). With this mind, Wang et al. showed that 0.1 µM pitavastatin induced NO production and increased cGMP levels in HUVECS incubated with pitavastatin for 48 hrs. Furthermore, eNOS activity in endothelial cells increased following a 30 min incubation with 0.1 µM pitavastatin, evidenced by increased phosphorylation of eNOS and also AKT, (Wang et al., 2008). The data in this study suggested that 10 µM pitavastatin could be having an effect, evidenced by attenuated CPA contraction to U46619, which could be attributed to pitavastatin activating eNOS to release NO into the endothelium and promote cGMP release in the smooth muscle. However, these data need to be interpreted with caution as the dose of pitavastatin used was suprapharmacological and the effect seen was marginal, so this outcome may not be clinically relevant.

3.5.3.2 Sodium nitroprusside-induced independent relaxation of chorionic plate arteries from normal pregnancies None of the statins tested enhanced relaxation of CPAs to SNP following a 2 hr incubation, however, pitavastatin at 1 µM showed a trend towards increased relaxation. SNP is an NO- donor which promotes relaxation of smooth muscle by binding to oxyhemoglobin, which triggers the release of cyanide, methemoglobin and NO. Release of NO activates guanylate cyclase which catalyses production of cGMP, which sequesters calcium and inhibits cellular contractility (Hottinger et al., 2014). Concerning the effect of 1 µM pitavastatin, one could speculate an interaction between pitavastatin and endothelium-derived relaxing factor (EDRF) which is present in the umbilical cord and in CPAs (Chaudhuri et al., 1991). However the CPAs release less EDRF compared to the villous tree, hence this agent may not play a key role in CPAs (Chaudhuri et al., 1991). Alternatively, KV channels KV2.1 and Kv9.3 have been identified in the fetoplacental vasculature and play a role in vascular tone (Wareing et al., 2006). Statins could play a role in opening these channels and promoting relaxation

140

in CPAs. These statements are not conclusive by any means as no inhibitory experiments were conducted to confirm this; moreover, 1 µM is a dose 4 x higher than a 4 mg dose in a non-pregnant population and must be interpreted with caution.

Given that one would expect CPAs from normal pregnancy to present with normal vascular function, it is not unusual that statins are not having an enhanced effect on vascular reactivity relative to their controls. In vivo, the serum levels of statin would also be a lot lower than that used in the current study so the result for 1 µM and 10 µM pitavastatin must be interpreted with caution.

A limitation of this study is that the experimental design did not include an SNP- dose response curve prior to statin incubation or DMSO exposure. DMSO is capable of relaxing vessels, and could possibly have masked any potential relaxation the statins may have produced, especially if the effects were modest.

3.5.4 Effect of short-term statin treatment on CPAs from PE pregnancies

3.5.4.1 Potassium- and U46619- induced contraction of chorionic plate arteries from PE pregnancies Following a 2 hr incubation with each statin in CPAs from PE pregnancies, no reduction in constriction was observed when expressed as kPa or % KPSS, with the exception of pitavastatin at 10 µM which attenuated constriction of CPAs post-incubation when data was expressed as % KPSS.

It has been shown that pregnancy complications such as PE and FGR are associated with a deficiency in prostacyclin relative to thromboxane (Stuart et al., 1981). In rat pulmonary arteries, an increase in thromboxane has been associated with inhibition of Kv channels leading to depolarization, activation of L-type Ca2+ channels and vasoconstriction (Cogolludo et al., 2003). Kv channels are important in vascular tone in terms of their contribution to the resting membrane potential (Cogolludo et al., 2003) and as mentioned above are present in the fetoplacental vasculature. In PE, pitavastatin could be innervating channels to cause smooth muscle cell relaxation.

141

3.5.4.2 Sodium nitroprusside-induced independent relaxation of chorionic plate arteries from PE pregnancies SNP-induced relaxation of PE CPAs was unaltered following statin incubation, a similar result to that seen in data from normal pregnancies. As mentioned previously, DMSO was the control for pitavastatin and simvastatin and has vasodilatory effects (Kaneda et al., 2016). With this in mind, DMSO could have masked any vasodilatory effects caused by statins, especially if they were modest. A suggested reason for reduced vasodilation in PE is the reduced expression or activation of BKca channels. As a result of these changes, CPAs can be remodelled and blunted relaxation can be seen due to reduced sensitivity to vasoactive agents, therefore fetoplacental perfusion may be impaired (He et al., 2018). It is uncertain whether statins work through these channels to enhance relaxation or not as inhibitory experiments have not been performed to confirm this.

Generally statins have been shown to stimulate vasorelaxation but this has often been endothelium-dependent which was not tested in CPAs, as they generally do not relax in an endothelial-dependent manner. Thus, one would not expect statins to be effective in promoting relaxation via the endothelium in this vascular bed.

In the placental circulation, the regulation of vascular tone and blood supply is important for adequate placentation and fetal development. A change in placental blood flow resistance, fetal oxygenation or vascular reactivity can affect placental blood flow (Baschat, 2004). The placenta does not receive cholinergic or adrenergic innervation therefore depends on stimulation from local factors found in the endothelium. These factors include various vasodilators such as NO (from umbilical and chorionic endothelium) and vasoconstrictors such as the renin-angiotensin system, thromboxane and prostacyclin, endothelin and its receptors (Read et al., 1993). Therefore, the levels of vasodilators/constrictors in the maternal and placental circulation are regulated to ensure a homeostatic balance of placental vascular function.

From a safety profile perspective, it is crucial that the statins do not affect this differential vascular response of the placenta as this could affect any adaptive responses the placenta may try to initiate in response to stimuli or stresses (Myatt, 1992; Wang and Zhao, 2010). A Pilot Randomized Controlled Trial investigating safety and pharmacokinetics of 10 mg pravastatin in high-risk PE women showed no identifiable safety risks in terms of reduced cholesterol concentration. Despite maternal concentrations of total cholesterol

142

(TC) and LDL being reduced, neither the cord blood TC and LDL nor infant birthweight differed between pravastatin and placebo groups (Costantine et al., 2016). This result corresponds to what Balan et al. showed in their work, where they perfused normal placental cotyledons and explants (21 % and 1 % O2) ex vivo with 0.2 µM/L pravastatin (2x serum concentration of 40 mg dose). Following this, they did not see a change in fetoplacental vascular tone or vasoconstriction in response to the vasoconstrictor angiotensin- II following exposure to pravastatin from the maternal circulation. In cultures of placental explants treated with 0.2 µM pravastatin for 5 hrs, concentrations of sFlt-1 were not significantly reduced and levels of PlGF and sEng were below the limit of detection; increased incubation to 72 hrs did not alter this. Only under hypoxic conditions, did pravastatin decrease sFlt-1 concentrations (Balan et al., 2017).

The research covered in this chapter is of interest as it allowed us firstly to assess three statins with different hydrophobicities and at different concentrations to identify which statin to take forward into vascular reactivity studies in maternal vessels and a PE- like mouse model. Simvastatin is a lipophilic statin and it is known that lipophilic drugs can passively diffuse across the phospholipid bilayer and cross the placenta (Al-Enazy et al., 2017). Furthermore, simvastatin has been found to inhibit placental transporters such as OATP2B1 and monocarboxylate transporter 1, so although the literature has highlighted beneficial effects of simvastatin, clinicians may be less likely to choose this statin over more hydrophilic statins (Daud et al., 2017). With this in mind, the present study looks at the effect of statins on vascular function from a safety perspective. The statin that showed greatest effects in CPAs from PE pregnancies, however, was pitavastatin (at 1 µM and 10 µM) and so this was the statin taken forward to examine in maternal omental arteries to see if protective effects are seen in pathological pregnancies, at a concentration which was considered more pharmacological (1 µM). Alongside pitavastatin, pravastatin was also taken forward for assessment of short-term incubation on omental artery function as this is the statin currently in clinical trials as a potential therapy for PE (Asif Ahmed, 2011; Costantine et al., 2016; Lefkou et al., 2016).

The strengths of this study are that we used the robust in vitro wire myography technique to assess the effect of statins on vascular function in CPAs, providing novel data pertaining to their effects. CPAs, although not fetal vessels, are still associated with the fetoplacental vasculature and are modified in pregnancy. Furthermore, three different

143

statins, two of which were assessed at two different concentrations, were investigated. However, this study does have limitations in that these experiments are in vitro and are not an identical representation of how these vessels would behave in vivo. Moreover, the doses of statins used in this study were suprapharmacological and higher than what would be seen in the human plasma following a therapeutic dose in non-pregnant and pregnant individuals. Finally, a 2 hr incubation may or may not be a sufficient amount of time to fully observe the effects of statins on vascular function. Studies in the maternal omental arteries and a pre-eclampsia-like eNOS-/- mouse model will however help to address this.

3.5.5 Conclusion/clinical implications To summarise, the results presented in this chapter indicate that CPAs from normal and PE pregnancies showed similar responses following exposure to the vasoconstrictive agent U46619 and the relaxatory agent SNP. It is reassuring that, following a short-term exposure to statins, no overtly detrimental effects were seen regarding CPA reactivity, as statins will likely cross the placenta and could in theory affect vascular function in pregnancy. Furthermore, it expands the current knowledge regarding the effects of statins on vascular reactivity in CPAs from normal and PE pregnancies. From a clinical context, these data are of importance, especially if statins are considered for prophylactic treatment in high-risk PE patients.

144

Chapter 4: The acute effect of statins on maternal omental artery function in normal, pre-eclamptic, chronic hypertensive and superimposed pre-eclamptic pregnancies

145

4.0 Results

4.1 Introduction Pre-eclampsia (PE) is associated with placental dysfunction and placental ischaemia due to inadequate spiral artery remodelling and an angiogenic imbalance in the maternal circulation, which results in maternal endothelial dysfunction. These preclinical changes result in clinical signs such as hypertension, proteinuria and end organ damage (Phipps et al., 2019). A combination of pre-existing vascular dysfunction and placental factors together contribute towards the PE pathology (Myatt and Webster, 2009). The endothelium is responsible for regulating vascular tone by producing vasoconstrictors such as endothelin-1 and thromboxane, and vasodilators such as prostacyclin and nitric oxide (Mori et al., 2014).

There are no proven treatments for PE apart from preterm delivery of the fetus and placenta. Whilst this intervention resolves the clinical indicators of PE, there still remains an urgent need for maternal treatment (Phipps et al., 2019). Clinical management following diagnosis of PE is to manage hypertension using antihypertensive medication and to increase fetal monitoring. A number of drugs have been suggested for treatment of this maternal disease, including statins (Phipps et al., 2019). Brownfoot et al. demonstrated that pravastatin at 200 and 2000 μM/L reduced the secretion of sFlt-1 from isolated cytotrophoblasts and HUVECs, as well as from placental explants from women with PE (Brownfoot et al., 2016a). They also showed through a pilot study that when 40 mg pravastatin was administered to 4 women with pre-term PE (24+5 to 29+4 weeks), clinical signs of PE were improved. For example, the urine protein/creatinine ratio (UPCR) stabilized or reduced in three out of the four women and no further change in antihypertensive medication was required. When these women were admitted, their systolic blood pressure ranged between 155 and 200 mmHg and diastolic 90-105 mmHg. Upon admission, 3 out of the 4 women were given antihypertensive medication (Labetalol 200 mg). Following pravastatin administration, mean arterial pressure stabilized or increased only slightly in the trial (Brownfoot et al., 2016a).

Women with chronic hypertension (CHT) are at an increased risk of adverse events in pregnancy due to existing abnormal vascular function and a sensitised endothelium. The

146

pre-existence of vascular disease makes these women more vulnerable to triggers from the placenta. Adverse complications in pregnancies complicated by chronic hypertension include superimposed PE, fetal growth restriction (FGR) and preterm birth (Seely and Ecker, 2014). Furthermore, studies have shown that the presence of elevated brachial systolic blood pressure and central aortic pressure in women with CHT precedes the development of superimposed PE with small for gestational age fetuses (Webster et al., 2019). CHT pregnancies have an incidence of 3% (Webster et al., 2017), which could further rise due to the increased rates of women of advanced maternal age and those that present with obesity in the obstetric population (Matthews and Hamilton, 2009; Poston et al., 2016). Pregnancies complicated by CHT are frequently associated with adverse maternal and perinatal outcomes (Bramham et al., 2014). CHT can be secondary to underlying renal, cardiac or adrenal disease known as secondary hypertension, but most CHT cases do not have a definitive identifiable cause and are classed as essential or idiopathic hypertension (Beevers et al., 2001). Raised peripheral vascular resistance, vascular stiffness and sensitivity to vasoconstrictor agonists are associated with CHT (Foëx and Sear, 2004). Women with CHT are at significant risk of developing superimposed pre-eclampsia (Guedes-Martins, 2017), with studies demonstrating that 17–25 % of CHT pregnancies are complicated by superimposed PE (Sibai, 2002).

For the treatment of hypercholesterolemia, simvastatin and pravastatin were the first statins to be approved for clinical use (Endo, 2010). Pitavastatin has been developed more recently and was given FDA approval in 2009. In studies outside of pregnancy, pitavastatin has been shown to effectively improve vascular reactivity by reducing endothelial dysfunction and decreasing plasma lipids and tissue lipid peroxidation (reduction in oxidative stress) in hypercholesterolemic rabbits at doses of ≥ 0.5 mg (Almeida and Ozaki, 2014). Increased vasodilation has been observed in the brachial artery following 2 weeks of pitavastatin treatment in patients with primary hypercholesterolemia (Sakabe et al., 2008). Additionally, pitavastatin promoted endothelial-dependent flow- mediated vasodilation in chronic smokers, offering endothelial cells protection against oxidative stress (Yoshida et al., 2010).

In this study, maternal omental arteries (OAs) were used as a model of the maternal systemic vasculature to investigate the effect of statins on ex vivo endothelial function in pregnancy. So far, there have not been any studies investigating the acute effect of statins

147

on the vascular function of maternal OAs. Thus, the current study investigated how pravastatin (hydrophilic) and pitavastatin (moderately hydrophobic), affected vascular function in OAs from normal pregnancies and from pregnancies complicated by PE, CHT and superimposed PE. Pravastatin and pitavastatin were chosen to go forward with because a) clinicians still seemed tentative about using simvastatin due to its hydrophobic properties; b) there are clinical trials looking at the effectiveness of pravastatin so it was deemed a suitable choice and c) pitavastatin is novel and has not been explored in pregnancy with regards to vascular function.

4.2 Hypothesis and aims

4.2.1 Hypothesis Statins, via endothelium-dependent or independent mechanisms, will improve vasodilatation and/or vasoconstriction in omental resistance arteries from pregnancies complicated by pre-eclampsia and /or chronic hypertension and superimposed PE.

4.2.2 Aims 1. To assess whether OA vascular reactivity is altered in hypertensive pregnancies compared to normal pregnancies. 2. To assess whether OA vascular reactivity is altered in hypertensive or normal pregnancies following short-term exposure to statins.

148

4.3 Materials and methods

4.3.1 Clinical Demographics Samples were taken from women with a normal pregnancy or a pregnancy complicated by PE, CHT or superimposed PE. Normal pregnancy was defined as that of women with an uncomplicated pregnancy with an appropriately grown fetus. Women with PE were included in the study if they developed “de-novo” hypertension after 20 weeks and a new onset of proteinuria, fetal growth restriction and/or PlGF measurement of <12 pg/ml (indicative of placental insufficiency) (Brown et al., 2018; Duhig et al., 2019). Furthermore diagnosis of PE was made based on a systolic BP >140 OR a diastolic BP of >90 on two occasions. Therefore in some cases women could have had a BP of 135/95 and been diagnosed with PE. CHT was defined as having high blood pressure (≥140 mmHg systolic and/or 90 mmHg diastolic) prior to pregnancy, it was recognized before 20 weeks’ gestation or antihypertensive medication was being used. Superimposed PE was defined as worsening hypertension, new proteinuria and/or the presence of other features suggestive of placental disease (Duhig et al., 2019) in a woman with a diagnosis of PE.

Women in the normal group delivered at term (37+0 to 42+0 weeks gestation) and were included if their BMI and age were <35 kg/m2 and <40 years respectively and the birthweight of their baby was >10th customised centile (Gardosi et al., 2018). Normotensive women were excluded if they had other existing medical or obstetric complications. PE, CHT and superimposed PE women were included if their BMI and age cut-off was <40 kg/m2 and <45 years respectively and were excluded if they developed diabetes, hyperthyroidism or had a twin pregnancy.

4.3.2. Sample collection and protocol The protocol was discussed in detail in the methods section in chapter 2 of the thesis. In brief, omental biopsies ~2x2cm in size were excised from the greater omentum in the abdomen of women at the time of caesarean section. Following collection, fat and scar tissue were removed and OAs (138-521 µM) were isolated from the omental biopsy. Resistance arteries were identified and mounted under a microscope onto a wire myograph. OAs were normalised at 0.9 of L13.3kPa and incubated in 20% oxygen (as opposed to 5% in the experiments on CPA) which reflects the relative estimations of PO2 in maternal

149

and fetoplacental vessels respectively. Following normalisation, the OAs were left in physiological salt solution (PSS) to equilibrate for 20-30 minutes to 0.9 of L13.3 kPa.

For each experiment performed, at the point of the 2 hr statin incubation, controls were carried out in parallel and were left in PSS for the same amount of time as the statin. In each experiment, of the 4 isolated OAs, two were randomly allocated to the control arm of the study and two to 1 µM statin treatment.

OAs were tested for their viability via exposure to 2x KPSS to promote vasoconstriction. The threshold for exclusion following this test was 3 kPa. A dose response curve was conducted using the thromboxane mimetic U46619 (10-10 -10-5.7). After the U46619 dose response curve, a single dose of bradykinin (10-5) was added as part of the protocol to allow assessment of endothelial function prior to exposure to statin/DMSO control to provide a further comparison between vessels from normal and pathological pregnancies. In the rare cases where two out of the four vessels were excluded prior to statin incubation i.e. they failed the KPSS threshold, the remaining two vessels were randomly allocated to either treatment or control. Following this, pravastatin or pitavastatin at 1 µM was added to the OAs for 2 hr. Following this exposure, a further dose response to U46619 was performed. To terminate the experiment, all drugs were washed out with PSS followed by a final KPSS exposure to ensure the vessels remained viable at the end of the experiment.

In separate experiments (on a different myograph), the effect of incubation with either statin or vehicle on endothelial-dependent and independent relaxation was assessed. OAs were pre-constricted with a high dose of U46619 (10-5). Once vessels had constricted and plateaued with a stable constriction for 5-6mins, a dose response to either bradykinin (BK; 10-10-10-5) or Sodium Nitroprusside (SNP; 10-10-10-5) was performed to investigate endothelium-dependent or endothelium-independent relaxation respectively.

-5 A high dose of U46619 (10 ) was chosen instead of an EC80 dose, because OAs lost their sensitivity to U46619 over time and required a more concentrated dose to stimulate constriction. To terminate the experiment, all drugs were washed out with PSS followed by a final KPSS exposure to ensure the vessels remained viable at the end of the experiment. Figure 4.1 illustrates what the trace from a self-collapsing vessel looked like relative to a trace from a vessel with a sustained constriction; collapsing vessels were excluded during the experimental protocol. 150

Assessments of contraction and relaxation of OAs post-statin incubation was not always carried out on the same omental biopsy. The reasons for this included availability of myography machines on the day of experiment, the size of the omental biopsy sample and the number of users of the biopsy. This resulted in there being a different overall total number (N) for contraction and relaxation experiments. Figure 4.2 shows a representative image of the myography protocol for OAs in all experiments assessing relaxation and contraction post-statin incubation. Vessels were excluded in real time using the following criteria: contraction <3 kPa following KPSS or self-collapsing vessels (i.e. fall in tension within 4-6 minutes). During data analysis, further vessels were excluded if the pre- constriction U46619 value was found to be an outlier using the ROUT method.

A B

Figure 4.1: A myography trace illustrating a ‘self-collapsing’ vessel

In the case where a U46619 pre-constricted vessel failed to hold its tension and self-collapsed prior to a bradykinin dose response curve, it was excluded. Similarly, if vessels were stable prior to incubation but then self-collapsed after incubation, they were excluded. A) A self-collapsed vessel B) U46619 pre- constricted vessel holding tension. X-axis denotes times and Y axis denotes tension (mN/mm). U44619

= thromboxane A2 mimetic.

151

function test constriction to U46619

Normalization

U46619 Dose- response Curve High Dose bradykinin 1X KPSS (10-5) Bradykinin Dose- 2X KPSS response Curve

Pre- constriction to U46619

Normalization Pitavastatin OR pravastatin incubation (1 µM, 2hrs)

Basal Tone calculation:

1. The tension in (mN/mm) was recorded pre- and post-incubation with statin.

2. This value was then converted to kPa

3. Basal tone = post-statin tension (kPa) minus pre-statin tension (kPa)

152

U46619 Dose- response Curve High dose bradykinin 1X KPSS (10-5) U46619 Dose- 2X KPSS response Curve

Pitavastatin OR pravastatin incubation (1 µM, 2hrs) Normalization

Figure 4.2: Representative traces of the wire myography protocols for omental arteries assessing relaxation and contraction post-statin incubation

A) Following normalisation, 2x KPSS washes and a U44619 dose response curve, a single -5 high dose of bradykinin was added (10 ) to vessels to allow assessment of endothelial function prior to exposure to statin/DMSO to provide a further comparison between vessels from normal and pathological pregnancies. Vessels were then randomly

assigned to vehicle (DMSO or water) or 1 µM statin, which were dissolved in 6 ml of PSS and left to incubate for 2 hrs in the myograph baths. Post-incubation, relaxation -10 -5 was assessed by bradykinin (10 – 10 ) following pre-constriction to high dose

U46619, followed by a final KPSS contraction. B) For the protocol assessing contraction, the first half was identical to the relaxation experiment except after statin incubation, a U46619 dose response was repeated. X- axis denotes times and Y axis denotes tension (mN/mm). U44619 = thromboxane A 2 mimetic.

153

4.4 Results

4.4.1 Demographics of women whose samples were used in the study Following caesarean section, samples were taken from women with a normal pregnancy (N=34) or a pregnancy complicated by PE (N=9), CHT (N=10) or superimposed PE (N=5). Samples were considered ‘normal’ and included if fetal weight was within 10th-100th centile for IBR indicating an appropriately grown fetus. If below the 10th percentile, these babies were likely to be SGA or FGR; this was confirmed by calculation of the individualised birthweight ratio (IBR) using the UK BulkCentileCalculator GROW 8.04 (Perinatal, 2019). Macrosomic fetuses (>90th centile; (Pasupathy et al., 2012) were included where maternal characteristics were deemed normal, there were 7/34. Demographic details for all the women whose omentum was used for this study are detailed in table 4.1 on the next page.

154

Table 4.1

Table 4.1: Demographic and clinical data for women recruited to the study.

Data are expressed as median and range. Booking blood pressures were acquired in antenatal clinic when the women were between 7 and 15 weeks gestation and max blood pressure is the highest blood pressure reading prior to delivery. From this data, there were significant differences in fetal weight (<0.0001 PE vs normal,<0.01 SI-PE vs Normal), placental weight (<0.01 PE vs normal, <0.05 SI-PE vs Normal) , IBR (<0.05 PE vs normal,<0.05 SI-PE vs Normal), gestation (<0.001 PE vs normal,<0.01 SI-PE vs Normal), parity (<0.05 PE vs CHT), booking systolic pressure (<0.05 CHT vs Normal, <0.05 SI-PE vs Normal), booking diastolic pressure and UPCR (<0.05 PE vs CHT) , which are highlighted in the table. Confirmed by nonparametric Kruskal-Wallis test and Dunn’s multiple comparison’s test. In the case where UPCR levels were low i.e. 15 mg/mmol in the PE group yet new-onset hypertension was existent, placental insufficiency evidenced by a low PlGF (<100pg/ml) measurement or another sign of PE was used to confirm diagnosis. (e.g. concurrent FGR). Women who had a PlGF test: 6/9 women with PE, 5/10 women with CHT and 2/5 women with Superimposed PE. In PE cohort, 4/9 women were early-onset.

155

4.4.2. Inclusion and Exclusion of Samples OA vessels were excluded from 23 normotensive pregnancies, 5 PE pregnancies, 1 CHT and 1 superimposed PE pregnancy. In experiments, vessels were excluded singly or in pairs depending on the reasons for their exclusion. As long as 1 control and 1 treated vessel out of the 4 OAs mounted met the inclusion criteria, the experiment was included. However, if both control vessels for the experiment did not meet inclusion criteria, the whole experiment was excluded. Depending on the size of the omental biopsy, vessels were “paired” so from the same arterial branch however in some cases vessels were “unpaired” and from a different arterial branch. In the case of “paired” vessels, if exclusion of one vessel occurred in real time during the protocol, the other vessel was still used. However, if exclusions occurred during analysis, both vessels in the pair would be excluded. This occurred if U46619 pre-constriction values were highlighted as an outlier using the ROUT method. The control for pitavastatin was DMSO as this was the vehicle used to dissolve pitavastatin. DMSO was added to the myograph baths at a final concentration of 0.1% during the 2 hr incubation for every experiment performed. For studies using pravastatin, the appropriate control was water.

After exclusions (see figure 4.3 A), 18 omental biopsies and 68 arteries (n=34 for DMSO and pitavastatin each) were included for the experiments using pitavastatin. For the experiments using pravastatin, 16 omental biopsies and 55 arteries (n=27 for water and n=28 for pravastatin) were included in the analysis. For experiments using biopsies from pathological pregnancies, 9 biopsies from women with PE (n=39 for pitavastatin and DMSO combined), 10 biopsies from women with CHT (n=30 for pitavastatin and DMSO each) and 5 biopsies from women with superimposed PE (n=24 for pitavastatin and DMSO each) were included.

156

Allocation of vessels in

experiments

Normal Pregnancies

Pravastatin Experiments Pitavastatin Experiments Total N=24, n=87 Total N=33, n=128

U4 (pre- BK (post- U4 (pre- BK (post- incubation): incubation): incubation): incubation):

N=16 N= 9 N=18 N=9 n=55 n= 26 n=68 n= 31

U4 (post- U4 (post- SNP (post- incubation): incubation): incubation): N=11 N=8 N= 9 n= 29 n=18 n= 19

Exclusions: Exclusions:

Self-collapsing Self-collapsing

N=3 N= 6

n=14 n=29

KPSS Constriction < 3 kPa KPSS Constriction < 3 kPa

N= 1 N=3

n= 4 n=6 Pre-constriction value an outlier Pre-constriction value an outlier

N=4 N=6

n=14 n=25

Pravastatin and pitavastatin data combined (post-exclusions :-)

U4: N=34, n=123

BK: N=18, n=57

157

Pathological Pregnancies

(Pitavastatin only)

PE Experiments: CHT Experiments: Superimposed PE Experiments: Total N=14, n=57 Total N=11, n=34 Total N= 6, n=28

U4 (pre-incubation): U4 (pre-incubation): U4 (pre-incubation):

N=9 N=10 N=5 n=39 n=30 n=24 U4 (post-incubation): U4 (post-incubation): U4 (post-incubation):

N=7 N=7 N=5

n=20 n=21 n=15

BK (post-incubation): BK (post-incubation): BK (post-incubation):

N=6, n=19 N=3, n=9 N=4, n=9

Exclusions: Exclusions: Exclusions: Self -collapsing Self -collapsing Self -collapsing N=1, n= 2 N/A N/ A KPSS Constriction < 3 kPa KPSS Constriction < 3 kPa KPSS Constriction <3 kPa

N=1, n=4 N/A N/A

Pre-constriction value an Pre-constriction value an Pre-constriction value an outlier outlier outlier

N=3, n=12 N=1, n=4 N=1, n=4

Figure 4.3: Flowchart of vessels used and excluded in pitavastatin and pravastatin experiments from normal and pathological pregnancies 88 OA experiments were performed overall. Data were excluded if vessels self-collapsed, contracted < 3kPa at the start of experiment or U46619 pre-constriction was identified as an outlier using the ROUT method. N = number of omental biopsies and n = number of arteries.

158

4.4.3 Diameter of omental arteries from normal, PE, CHT and superimposed PE pregnancies prior to statin exposure

The diameters of OAs from PE, CHT, superimposed PE and normal pregnancies were not significantly different. However, there was a trend towards smaller diameters in OAs from CHT pregnancies relative to normal, PE and superimposed PE pregnancies (Figure 4.4).

600

m 

e t

e 400

m a

i D

200 n

e Mean Diameter MeanDiameter M 0

L E T E A P H P M C I- R S O N

Figure 4.4: Diameter of OAs from PE, CHT, superimposed PE and normal pregnancies prior to treatment

OA diameters from PE, CHT, superimposed PE (SI-PE) and normal pregnancies prior to statin incubation were not statistically different from each other prior to statin incubation but a trend towards a smaller diameter in CHT pregnancies was observed relative to normal, PE and superimposed PE pregnancies (SI-PE) (p=0.088). Data presented as average diameter per omental biopsy for each group. The horizontal line denotes median and data expressed as median±IQR; Kruskal Wallis test performed.

159

4.4.4 Assessment of vascular reactivity of omental arteries between normal, PE, CHT and superimposed PE pregnancies

There was a trend towards increased contraction to U46619 in OAs from superimposed PE pregnancies relative to normal, PE and CHT pregnancies; (Figure 4.5 A; (p=0.068)). There were many fewer exclusions based on self-collapsing vessels amongst the pathological pregnancies compared to the normal vessels. Furthermore, prior to statin exposure, OAs from PE pregnancies contracted more to U46619 relative to OAs from CHT, superimposed PE and normal pregnancies when data was expressed as contraction to U46619 dose as a % of the maximum KPSS contraction (Figure 4.5 B; (p=0.030)). When endothelial function of OAs was assessed with a single high dose of bradykinin following a U46619 dose response curve (pre-incubation), no significant difference in the relaxation at 5 minutes was observed between normal, CHT and superimposed PE groups, however a trend towards reduced relaxation in the PE group was observed (figure 4.5 C (p=0.097). The data reflects the enormous biological variability in response to a single dose of an endothelial- dependent agonist. Endothelial function was also compared between groups following 2 hr incubation with the control DMSO. There was no significant difference in the relaxation of control OAs to bradykinin between normal, PE, CHT and superimposed PE groups (Figure 4.5 D (p=0.993)).

In figure 4.5, the constriction graphs include vessels prior to exposure to statin or their relative controls. The endothelial-dependent dose response curves shown are for DMSO controls from normal pregnancies and DMSO controls from pathological pregnancies (post- incubation). A bradykinin dose response was not conducted in PSS alone (pre-incubation).

160

A B

25 NORMAL (N=34, n=123) 180 NORMAL (N=34, n=123) PE (N=9, n=39) ) 20 160 PE (N=9, n=39)

a * n

P CHT (N=10,n=30) 140 o

k CHT (N=10, n=30)

(

S t

15 120 S

SI-PE (N=5, n=24) c n

i SI-PE (N=5, n=24)

P 100

s s 10 80

% e

o 60 T 5 C 40 20 0 0 1 2 3 4 5 6 1 2 3 4 5 6 x C [U46619] x 10xM [U46619] x 10 M D

i g o 140

r n

20 t

d i

120

s b 100

t 80

- a

s NORMAL (N=9, x

60 e

o r

a 60 n=27)

e h

9 PE (N=6, n=19)

R 40

g 80

% 6

h CHT (N=3, n=9)

% 6 100 4 SI-PE (N=4, n=9)

U 0 Normal PE CHT SI-PE -10 -8 -6 -4 [BK] x 10xM

Figure 4.5: Assessment of vascular reactivity in OAs from normal, PE, CHT and superimposed PE pregnancies.

There was a trend towards elevated contraction to U46619 in OAs from superimposed PE pregnancies compared to the other groups when expressed as tension (kPa) (A), however, when

U46619 data was expressed as %KPSS, OAs from PE pregnancies showed elevated contraction to U46619 compared to the other groups (B). A high dose of U46619 (10-5) was used to pre-constrict vessels prior to relaxation. When assessing endothelial function pre-incubation, there was no significant difference in OAs from normal, CHT or superimposed PE pregnancies, but a trend towards endothelial dysfunction in the PE group was observed (C). Relaxation of omental arteries to BK was not different between OAs from normal, PE, CHT or superimposed PE (SI-PE) pregnancies following the 2 hr incubation with DMSO. (The relaxation curves were performed and normotensive pregnancies were compared to pathological pregnancies) (D). Replicate vessels from the same pregnancy were averaged for each group. Graphs A and B show the combined pre- incubation U46619 data for normal pregnancy (N=34) from pitavastatin (N=18) and pravastatin (N=16) experiments. Data are expressed as mean±SEM; two-way ANOVA (A, B and D) and as median±IQR with Kruskal-Wallis test for graph C. *p<0.05

161

4.4.5. Effect of pravastatin and pitavastatin on basal tone of omental arteries from normal pregnancies at a concentration of 1 µM following 2 hr incubation.

Following a 2 hr incubation with pitavastatin (Figure 4.6 A: DMSO vs pitavastatin (p=0.832)), or pravastatin (Figure 4.6 B: Water vs pravastatin (p=0.413)), there was no significant difference in basal tone of OAs (0.9 of L13.3KPa) compared to their relative

controls. A )

a 2

(

n o

i 1

e T 0

e g

n -1

h C -2

O M S  -1 M n D ti ta B s a v a it

P )

a 2

(

i 1

e T

e g

a -1 h C -2

R M E  T 1 - A n W ti ta s a v ra P

Figure 4.6: Effect of 2 hr incubation with pitavastatin and pravastatin on basal tone of omental arteries relative to control group. Following a 2 hr incubation with either pravastatin (1 µM, N=16, A) or pitavastatin (1 µM, N=18, B), the basal tone of OAs was not significantly different between the statin-exposed group versus controls. The horizontal line denotes median and data ar expressed as median±IQR; Mann-Whitney U test performed. 162

4.4.6 Effect of pravastatin and pitavastatin on contraction of omental arteries from normal pregnancies at a concentration of 1µM following 2 hr incubation Following a 2 hr incubation, OAs incubated with the control DMSO showed reduced maximal contraction to U46619 (Figure 4.7 A: DMSO 1 vs DMSO 2 (p=0.042)). This observation continued when data were expressed as % of maximum KPSS contraction (Figure 4.7 C: DMSO 1 vs DMSO 2 (p<0.0001)). Similarly, following incubation with water alone, OAs showed reduced maximal contraction to U46619 (Figure 4.7 B: Water 1 vs Water 2 (p=0.0005)). This observation continued when data were expressed as % of maximum KPSS contraction (Figure 4.7 D: Water 1 vs Water 2 (p=0.0003)). A trend towards reduced sensitivity to U46619 was also observed in water controls (p=0.076).

Following a 2 hr incubation with pravastatin or pitavastatin at 1 µM, there was no significant effect on OA contraction to U46619 from normal pregnancies (Figure 4.8 A: DMSO vs pitavastatin (p=0.987) and Figure 4.8 B: Water vs pravastatin (p>0.999)). When U46619 contraction curves were expressed as maximum KPSS contraction (%KPSS) there was again no significant difference between statin-exposed OAs and comparable control OAs (Figure 4.8 C: DMSO vs pitavastatin (p=0.755) and Figure 4.8 D: Water vs pravastatin (p=0.343)).

163

A B 15 15 DMSO WATER (pre-incubation)

(pre-incubation)

)

) a a * WATER DMSO P *** P 10 (post-incubation)

10 k k

(post-incubation) (

(

i s

5 n 5

e T T 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM C D 180 180 WATER DMSO (pre-incubation) 160 (pre-incubation) 160

n 140 n 140 WATER

DMSO o o

**** i ***

S i

t (post-incubation) S (post-incubation) 120

t 120

i P

r 100 P

r 100

t K

80 80

% o

o 60 60 C C 40 40 20 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM

Figure 4.7: Effect of time following 2 hr incubation with control (either DMSO or water) on contraction of OAs from normal pregnancies.

There was reduced responsiveness of control OAs to U46619 following 2 hr incubation with DMSO (A-C) or water (B-D) expressed as kPa and %KPSS. U46619 dose response curves were compared using two-way ANOVA and U46619 sensitivity was determined using EC50 values and performing a Mann-Whitney U test. Data are expressed as mean±SEM; number of omental biopsies in parenthesis.*P<0.05; ***p<0.001; ****p<0.0001.

164

A B

15 DMSO 15 Water

Pitavastatin-1M Pravastatin - 1M

)

) a

(N=8, n=18) a (N=11, n=29) P 10 P 10

(

s s

5 n 5

T T

0 0 -10 -8 -6 -4 -10 -8 -6 -4 x C [U46619] x 10xM D [U46619] x 10 M 140 DMSO 140 Water

120 Pitavastatin-1M 120 Pravastatin 1M

n o

o 100 100

i i S (N=11, n=29)

S (N=8, n=18)

c S

P r

P 80 80

n 60 60

o C 40 C 40 20 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM

Figure 4.8: Effect of 2 hr statin incubation on contraction of OAs from normal pregnancies.

Following a 2 hr incubation with pravastatin or pitavastatin at 1µM, there was no significant effect on contraction of OAs to U46619 from normal pregnancies expressed as tension (kPa) (A- B) or as % KPSS; (C-D). U46619 dose response curves were compared using 2-way ANOVA. Data are expressed as mean±SEM; number of omental biopsies in parenthesis.

165

4.4.7 Effect of pravastatin and pitavastatin on relaxation of omental arteries from normal pregnancies at a concentration of 1 µM following 2 hr incubation

Following a 2 hr incubation with pitavastatin (Figure 4.9 A: DMSO vs pitavastatin (p=0.690)) and pravastatin (Figure 4.9 B: Water vs pravastatin (p=0.401)), there was no significant effect on OA relaxation to bradykinin from normal pregnancies. Furthermore, when endothelium-independent relaxation to SNP was assessed with 1 µM pitavastatin, there was no significant difference between pitavastatin treated and control groups (Figure 4.9 C: DMSO vs pitavastatin (p=0.688)).

A B n

n 120

120 o

i t

i 100 f

100 r

n n

n 80 80

- a a 60

60 e

e 40 9

9 DMSO 40 Water

6 % % Pravastatin 1M

6 Pitavastatin-1 M

20  6 20 4

(N=9, n=31) 4 U U (N=9, n=26) 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [BK] x 10 M [BK] x 10xM

n o

i 120

i f

r 100

s n

n 80

c -

a 60

l p

e DMSO

R 1 Pitavastatin-1M

6 20

% 6

4 (N=9, n=19) 0 U -10 -8 -6 -4 [SNP] x 10xM

Figure 4.9: Effect of 2 hr statin incubation on relaxation of OAs from normal pregnancies. Following a 2 hr incubation with pravastatin or pitavastatin at 1 µM, there was no significant effect on relaxation of OAs to BK from normal pregnancies. Likewise, no significant effect on relaxation of OAs to SNP following pitavastatin (only) incubation; BK and SNP dose response curves were compared using two-way ANOVA. Data are expressed as mean±SEM; number of omental biopsies in parenthesis.

166

4.4.8. Effect of pitavastatin (1 µM) on basal tone of omental arteries from PE, CHT and superimposed PE pregnancies following 2 hr incubation

Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant difference in basal tone of OAs from pregnancies with PE (Figure 4.10 A: DMSO vs pitavastatin (p=0.297)) with CHT (Figure 4.10 B: DMSO vs pitavastatin (p=0.138)) or with superimposed

PE (Figure 4.10 C: DMSO vs pitavastatin (p=0.310)) (0.9 of L13.3KPa) compared to their relative controls.

167

A )

a PE Pregnancies

P k

( 2

i s

n 0

e T

e -2

g n

a h

C M O  S -1 M n D ti ta s a v a it

P ) a

B P 2

(

n CHT Pregnancies o

i s

n 0

e T

e -2

g n

a h

C M O  S -1 M n D ti ta s a v a it

P )

C a P 2

k Superimposed PE Pregnancies

(

o i

n 0

n i

e -2

n a

C O M S  -1 M n D ti ta s a v a it P

Figure 4.10: Effect of 2 hr incubation with pitavastatin on basal tone of OAs relative to control group.

Following a 2 hr incubation with pitavastatin (1 µM) in PE pregnancies (N=9), CHT pregnancies (N=10) or superimposed PE pregnancies (N=5) basal tone of OAs was not significantly different between statin-exposed group versus controls. The horizontal line denotes median and data expressed as median±IQR; Mann-Whitney U test performed.

168

4.4.9 Effect of pitavastatin (1 µM) on contraction of omental arteries from PE, CHT and superimposed PE pregnancies following 2 hr incubation Following a 2 hr incubation with DMSO control only, OAs from PE pregnancies showed reduced maximal contraction to U46619 (Figure 4.11 A: DMSO 1 vs DMSO 2 (p=0.003)). This observation continued when data was expressed as a % of maximum KPSS contraction (Figure 4.11 B: DMSO 1 vs DMSO 2 (p<0.0001)), a trend towards reduced U46619 sensitivity was noticed also (p=0.097). A similar trend in reduced contraction to U46619 was seen in OAs from CHT pregnancies (Figure 4.11 C: DMSO 1 vs DMSO 2 (p=0.001)); this observation remained when data were expressed as % KPSS (Figure 4.11 D: DMSO 1 vs DMSO 2 (p=0.0006)). A decreased U46619 sensitivity was also observed (p=0.018). Finally, in OAs from superimposed PE pregnancies, a smaller contractile response to U46619 was observed (Figure 4.11 E: DMSO 1 vs DMSO 2 (p=0.046)). This observation remained when data were expressed as a % KPSS (Figure 4.11 F: DMSO 1 vs DMSO 2 (p=0.020)). A trend towards reduced U46619 sensitivity was also observed (p=0.056).

Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant effect on contraction of OAs to U46619 from PE pregnancies; (Figure 4.12 A: DMSO vs pitavastatin (p=0.796.)), CHT pregnancies (Figure 4.12 C: DMSO vs pitavastatin (p=0.996)) or superimposed PE pregnancies (Figure 4.12 E: DMSO vs pitavastatin (p=0.265)). There was again no difference between any groups when U46619 contraction curves were expressed as maximum KPSS contraction (% KPSS); PE pregnancies (Figure 4.12 B: DMSO vs pitavastatin (p=0.592)), CHT pregnancies (Figure 4.12 D: DMSO vs pitavastatin (p=0.951)) or superimposed PE pregnancies (Figure 4.12 F: DMSO vs pitavastatin (p=0.343)).

169

A B 25 PE+DMSO 180 (pre-incubation) PE+DMSO 160 (pre-incubation) ) 20 a PE+ DMSO PE+DMSO n 140

P ****

(post-incubation) o (post-incubation)

S t ( 120

i n

P 100

s 10 *** 80

% o

e 60 T 5 C 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM

C D

25 CHT+DMSO (pre-incubation) 180 CHT+DMSO

) 20 160 (pre-incubation)

a CHT+DMSO P

n 140 k

(post-incubation) o CHT+DMSO 2

(

i S 15

t 120 (post-incubation) ***

n c

** i

P o

r 100

s s

10 80

% e

o 60 T 5 C 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM

E F

25 SI-PE+DMSO 180 SI-PE+DMSO (pre-incubation) (pre-incubation)

) 20 160 a

SI-PE+DMSO n

P 140

o SI-PE+DMSO 2 k

(post-incubation) i S

( * t 120

15 (post-incubation) *

i P

r 100

s 10 80 n

% o

e 60 T 5 C 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10xM [U46619] x 10 M

Figure 4.11: Effect of time following 2 hr incubation with control (either DMSO or water) on contraction of OAs from PE, CHT or superimposed PE pregnancies.

There was reduced maximal constriction of control OAs to U46619 following 2 hr incubation with DMSO when expressed as kPa (A, C, E) and %KPSS (B, D, F). A reduced U46619 sensitivity was also seen. U46619 dose response curves were compared using two-way ANOVA and U46619 sensitivity was determined using EC50 values and performing a Mann-Whitney U test. Data are expressed as mean±SEM; number of omental biopsies in parenthesis.*P<0.05;**p<0.01 ***p<0.001; ****p<0.0001. .

170

A B

140 PE+DMSO 20 PE+DMSO PE+Pitavastatin 1M

PE+Pitavastatin 1M 120

) n a 15 (N=7, n=20)

o 100

P (N=7, n=20)

(

i 80

r n

10 t

n e 5 C 40 T 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 C [U46619] x 10xM D [U46619] x 10xM 20 CHT+DMSO 140 CHT+DMSO CHT+Pitavastatin 1M

120 CHT+Pitavastatin 1M )

a 15

(N=7, n=21) n P

o 100 (N=7, n=21)

(

c i

n 80 P

10 r

s 60

% o e

T 40

5 C 20

0 0 -10 -8 -6 -4 -10 -8 -6 -4 E [U46619] x 10xM F [U46619] x 10xM

20 SI-PE+DMSO 140 SI-PE+DMSO SI-PE+Pitavastatin 1M

120 SI-PE+Pitavastatin 1M

) n

a 15 (N=5, n=15) o

P 100 (N=5, n=15)

(

S c

i 80

P r

10 t

s 60

n o

C 40 T 5 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10xM [U46619] x 10 M

Figure 4.12: Effect of 2 hr pitavastatin (1 µM) incubation on contraction of OAs from pregnancies with PE, CHT and superimposed PE. Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant effect on contraction of OAs to U46619 from PE, CHT or superimposed PE pregnancies expressed as tension kPa (A, C, E) or as % KPSS (B,D,F). U46619 dose response curves were compared using 2-way ANOVA Data are expressed as mean±SEM; number of omental biopsies in parenthesis.

171

4.4.10 Effect of pitavastatin (1µM) on relaxation of omental arteries from PE, CHT and superimposed PE pregnancies following 2 hr incubation.

Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant effect on relaxation of OAs to bradykinin from PE pregnancies; (Figure 4.13 A: DMSO vs pitavastatin (p=0.963)), CHT pregnancies (Figure 4.13 B: DMSO vs pitavastatin (p=0.980)) or superimposed PE pregnancies (Figure 4.13 C: DMSO vs pitavastatin (p=0.232)).

172

n 140 o

i t

r 120

s 100

i c

t 80

e x r

a 60

9 1

R 40 PE+DMSO

6 6 % PE+Pitavastatin 1M

4 20 U

(N=6, n=19) 0 -10 -8 -6 -4 x B [BK] x 10 M

n 140

t e

c 120 t

s 100

o o i

t 80

x r

a 60

9 CHT+DMSO

R 40

6 CHT+Pitavastatin 1M % 6 20 4 (N=3, n=9) U 0 -10 -8 -6 -4

C [BK] x 10xM 140

i r

t 120

r t

100

o o

i 80

c -

x r

a 60 SI-PE+DMSO

9 R

1 40

SI-PE+

6 % 6 Pitavastatin 1M

4 20

U (N=4, n=9) 0 -10 -8 -6 -4 [BK] x 10xM

Figure 4.13: Effect of 2 hr pitavastatin incubation on relaxation of OAs from pregnancies with PE, CHT and superimposed PE.

Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant effect on relaxation of OAs to bradykinin from pregnancies with PE, CHT or superimposed PE; p>0.05. BK dose response curves were compared using two-way ANOVA. Data are expressed as mean±SEM; number of omental biopsies in parenthesis.

173

4.5 Discussion

4.5.1 Summary of Main Findings This study set out to determine whether short-term statin exposure altered vascular reactivity of OAs from normal, pre-eclamptic (PE), chronic hypertensive (CHT) and superimposed pre-eclamptic (SI-PE) pregnancies. OAs from PE pregnancies showed a larger contractile response to U46619 (expressed as %KPSS) compared to OAs from normal, CHT and superimposed PE pregnancies. However, there was no significant difference in bradykinin-mediated relaxation of OAs between any of the four groups. The assessment of endothelium function following the first U46619 dose response curve revealed that there was no difference between normal, CHT and superimposed PE pregnancies, however, there was a trend towards endothelial dysfunction in the PE group. Furthermore, since a bradykinin dose response curve without DMSO exposure (i.e. in PSS alone) was not performed post-incubation, it was not possible to formally compare endothelial function between groups. In general, the vascular function of OAs from normal pregnancies following pravastatin and pitavastatin incubation did not show altered contraction to the thromboxane A2 mimetic U46619 or relaxation via endothelium-dependent bradykinin (BK) or the NO-donor sodium nitroprusside (SNP). Similarly, following incubation with pitavastatin at 1 µM only, OAs from pregnancies complicated by PE, CHT or superimposed PE showed no difference in vascular reactivity to U46619 or BK. Overall, the data from the present study has demonstrated that acute exposure to suprapharmacological doses of pravastatin and pitavastatin does not appear to overtly affect vascular function of OAs from normal pregnancies. This finding was expected considering the endothelium was healthy. However, in hypertensive pregnancies where the endothelium is potentially damaged, pitavastatin did not appear to promote relaxation or attenuate constriction. Nevertheless, it is likely that the constrictive and relaxatory functions of all vessels in these experiments were affected by exposure to DMSO (figures 4.4.9-4.4.10) which was necessary to dissolve pitavastatin and therefore the findings presented need to be interpreted with a note of caution. Pitavastatin was chosen for use in pathological maternal systemic vessels firstly due to the fact that pitavastatin has not been used in pregnancy studies in either mice or humans previously. Secondly, data from the results chapter assessing statins and chorionic plate artery function, suggested pitavastatin could attenuate U46619-mediated constriction in PE vessels. Whilst it would have been more insightful to focus upon a 174

number of statins, the relative scarcity of samples from pathological pregnancies meant that, logistically, only one statin could be investigated in detail over the course of this study.

4.5.2 Methodological Challenges and Implications

4.5.2.1 Rationale for the use of omental arteries (OAs) Omental arteries (OAs) were chosen for this study as they are representative of the maternal systemic vasculature and are accessible at the time of caesarean section (all omental biopsies were either from elective or emergency caesarean sections). These vessels can be used as a human vascular tissue model for both normal and pathological pregnancies (Dong et al., 2015). OAs are systemic vessels and are useful to study in pregnancy as different vascular beds can show heterogeneity in their vascular function. Studies have shown that BK-induced relaxation is greater in OAs compared to myometrial arteries when isolated from non-pregnant and pregnant women (Ashworth et al., 1996). However, studies in vessels from pregnant sheep have not shown a difference in endothelium-dependent relaxation to acetylcholine or bradykinin; instead, they showed a reduced sensitivity and maximal response to U46619 in myometrial arteries compared to OAs (Anwar et al., 1999).

A unique vascular phenotype specific to OAs is that they show both tonic and phasic contractions of their smooth muscle. Phasic contractions of smooth muscle relates to the presence of a rhythmic oscillatory contractile motion whereas tonic smooth muscle relates to continuous contractions at low energy. Phasic contraction is often associated with resistance arteries due to their role in vascular function as they help to regulate pressure and flow (Fisher, 2010). Whilst these OAs can show continuous tonic contractions, we noticed that they were not able to hold their U46619 pre-constriction for longer than 6-8 minutes.

4.5.2.2 Desensitization to Thromboxane Agonist The ex vivo myography experiments using OAs revealed that there was significant drift over the time course of the experiment. Maximal constriction as well as sensitivity to U46619 diminished after the 2 hr incubation with DMSO and water controls, in both normal and pathological pregnancies. It is uncertain what caused this; however, one could postulate a decline in responsiveness to free Ca2+ concentration, affecting the frequency of Ca2+ dependent activation of myosin light chain kinase and subsequent phosphorylation of myosin light chain, resulting in a reduced contractile force (Itoh et al., 1989). Furthermore, 175

as this observation was also found with the thromboxane mimetic U46619, the effect of time could be attributed to a change in the affinity, number or activity of thromboxane receptors leading to an attenuated constriction of smooth muscle. Reduced RhoA/Rho kinase signalling could also play a role (Webb, 2003). Moreover, as the L-type calcium channels are believed to play a large role in modulating intracellular calcium levels and determining the contractile state of smooth muscles, it could be the case that expression and activity of these channels diminished over time in OAs (Knot and Nelson, 1998).

4.5.2.3. Effect of DMSO on OA Vascular Reactivity A challenge associated with the experimental protocol used in this study was the interpretation of vascular function assessments following exposure to DMSO, the solvent used to dissolve pitavastatin, which appeared to have an effect on constriction and relaxation of OAs in normal and pathological pregnancies. The concentration of DMSO used for the incubation period was equivalent to the final concentration of DMSO (0.1%, equivalent to 12.8 mM) in which pitavastatin was dissolved in the myograph bath. Other groups have reported different findings to those in the current study. Using pressure myography on OAs from elective caesarean sections at term, Brownfoot et al. showed no effect of incremental doses of DMSO for ~22.5 mins (with a final concentration of 0.1% which is equivalent to our study) on relaxation following preconstriction to U46619 when compared to their drug of interest, sulfasalazine (N=5) (Brownfoot et al., 2019). However, other groups used DMSO at 1.74 µM in OAs from patients awaiting gastric surgery (aged 50-74 yrs) (Kinoshita et al., 2017) and 2 µM in male rat aorta (Haba et al., 2009) with no effect seen, either. In the study by Kinoshita and Haba et al. DMSO (following 20-30min exposure), was not shown to have an effect on vasomotor function or produce vasodilatory effects in preconstricted vessels (Haba et al., 2009; Kinoshita et al., 2017). The present study used 0.1% DMSO and OAs were exposed to DMSO for 2 hrs which is significantly longer than the exposure time of the studies above. Kaneda et al. proposed that DMSO worked through the endothelium to release NO to trigger cGMP production in the smooth muscle to promote relaxation and decrease Ca2+ sensitivity, partially through inhibition of Rho-kinase to attenuate constriction to phenylephrine, confirmed through myography experiments using male rat aorta (Kaneda et al., 2016). Effects were attained in a dose- dependent manner (0.1-3%) and were seen in the presence and absence of endothelium (Kaneda et al., 2016); this could explain why effects of DMSO were still seen in OAs from

176

pathological pregnancies. The outcome of DMSO observed by Kaneda et al. was similar to ours and they saw effects of DMSO from as low as 0.1% (same as in the current study) up to 3%, suggesting that relaxatory effects of DMSO are seen with increasing concentration but also length of exposure.

4.5.2.4 Self-Collapsing vessels Finally, another methodological issue that was experienced during the myography protocol was self-collapsing vessels, where vessels would spontaneously lose their tension following pre-constriction to U46619, meaning these vessels were not used to conduct a dose- response curve to bradykinin relaxation curves following post-incubation. The literature has suggested that arteries as well as veins can lose their mechanical stability under mechanical loads (Han et al., 2013); the stability of a vessel is crucial to its physiological function, however vessels, and in particular OAs, may be less stable ex-vivo. Furthermore, it has been shown that vessels become unstable when the lumen pressure exceeds a critical value, or the axial tension is reduced (Han, 2009a; Han, 2009b). It is uncertain why this continuously happened but it may be a characteristic of this vessel type. To overcome this issue, experiments (where possible) were carried out in duplicate vessels.

4.5.3 Vascular Reactivity of Omental Arteries from Normal Pregnancies and Hypertensive Pregnancies

4.5.3.1 Potassium- and U46619- induced contraction of omental arteries from normal and hypertensive pregnancies This study has demonstrated a significant difference in contraction of OAs between normal and PE pregnancies and no difference in comparison to CHT pregnancies. There was a trend, however, towards increased contraction in OAs from superimposed PE pregnancies. When the mean maximum KPSS responses were compared between normal, PE, CHT and superimposed PE pregnancies, no significant difference was observed in contraction to KPSS (p=0.264; Kruskal-Wallis test; data not shown).

Mishra et al. using pressure myography, have previously demonstrated in samples from 40 normotensive women and 9 PE women, that OAs from PE women showed enhanced reactivity to angiotensin II as evidenced by a change in vessel diameter (Mishra et al., 2011). Furthermore, Pascoal et al. showed that contractile responses to KCL and vasopressin were larger in OAs from PE pregnancies compared to those from non-pregnant

177

and normotensive pregnant women. Additionally, more phasic activity was observed in vessels from PE women (Pascoal et al., 1998). This trend was also observed by Wareing et al. who demonstrated significantly increased vasoconstriction and sensitivity to vasopressin (effective active pressure) in OAs from women with PE compared to normotensive women (Wareing et al., 2006b). This increased contractile response in OAs from PE women is further supported by Aalkjaer et al. who demonstrated that OAs from PE women have increased wall thickness: luminal diameter ratios compared to women with normal pregnancies. The response of PE vessels to angiotensin II was similar to vessels from non-pregnant women, indicating the absence of vascular remodelling in PE compared to normal pregnancy; PE OAs were more responsive to angiotensin II but there was no difference in contraction to noradrenaline between the groups (Aalkjaer et al., 1985).

Other studies using omental and subcutaneous fat resistance arteries have demonstrated no difference in vasoconstriction between PE and normotensive women, as measured via reactivity to endothelin-1 and U46619, but have shown that OAs from PE women were more sensitive to noradrenaline compared to normotensive vessels (Belfort et al., 1996; Knock and Poston, 1996; Vedernikov et al., 1995). Studies using myometrial arteries have also demonstrated this trend where myometrial artery constriction to endothelin-1 from PE women has been similar to that of normotensive women (Walsh, 1985; Wolff et al., 1996). There was a single study that observed a decrease in OA constriction in PE. Vedernikov et al. demonstrated that OAs from PE women show reduced constriction to KCL relative to normotensive women and they did not see a significant difference in the contractile response to U46619 and endothelin-1 between PE and normotensive women (expressed as % KPSS) (Vedernikov et al., 1999).

In a study by Khammy et al. who isolated arcuate and femoral arteries from hypertensive and normotensive rabbits, they demonstrated that the sensitivity of arcuate arteries to noradrenaline and methoxamine was unaffected, showing minimal effects of hypertension. In this study, femoral arteries showed an increased sensitivity to noradrenaline and methoxamine (Khammy et al., 2016). In a prospective controlled observational study by Hibbard et al. comparing the arterial system of normotensive women with those of women with PE and superimposed PE, M-mode echocardiograms and continuous wave Doppler velocity waveforms provided data about flow and pressure. An increased total vascular resistance and reduced vascular compliance was observed in PE.

178

This response was exaggerated in women with superimposed PE pregnancies compared to PE women who showed a loss of normal adaptive vascular mechanisms of pregnancy. This may be due to the chronic hypertension aspect in superimposed PE pregnancies, as these women have had vascular dysfunction over a prolonged period. Increased vascular tone and vascular stiffness was observed in both peripheral arterioles and conduit vessels (Hibbard et al., 2005). The stroke prone spontaneously hypertensive rat has been shown to display similar features to superimposed PE during pregnancy. In this model, impaired uterine artery remodelling (assessed using pressure myography), increased arterial stiffness and reduced uteroplacental perfusion (assessed using Uterine artery Doppler waveform) was evident (Small et al., 2016). In the eNOS knockout mouse, a mouse that shows similar pathological features to superimposed PE pregnancies i.e. chronic hypertension and proteinuria, uterine arteries show an enhanced U46619-mediated contractile response compared to wildtype mice (when data was expressed as % KPSS) (Kusinski et al., 2012).

The data from the present study showed that OAs from superimposed PE pregnancies produced a bigger contractile response to U46619 than OAs from CHT, PE or normotensive pregnancies; this trend was seen when the data were expressed as effective active pressure (kPa). Whilst there have not been functional vascular studies using vessels from human superimposed PE pregnancies, this observation was in agreement with what has been observed in humans in vivo and pregnant superimposed PE-like animal models. However, when contraction to U46619 was expressed as % KPSS, OAs from PE pregnancies produced a greater contractile response relative to superimposed PE, CHT and normal pregnancies and this observation was in agreement with other ex-vivo studies assessing vascular function of vessels from PE pregnancies. Data expressed as % KPSS takes into account structural alterations in smooth muscle secondary to hypertrophy or hyperplasia. However, this change in smooth muscle size could mask changes in vascular response alone (based on vessel length and lumen circumference). A large contractile response of OAs from PE pregnancies was not observed when the data were expressed as kPa. This highlights the importance of expressing data as both active effective pressure (kPa) and % KPSS; active effective pressure takes into account initial circumference and initial vasoconstriction independent of structural changes of a vessel, each providing different but complementary information.

179

In chronic hypertension, sclerotic changes in vessels may reduce contractility, thus making vessels less responsive to vasoconstrictor agents such as U46619. In the present study, when data were expressed as kPa and % KPSS, the change in constriction to U46619 for the CHT dose response curve was similar to that of the normotensive group except at the higher doses. However, one potential limitation of this present study is that the chronic hypertension group was not split into essential (primary) and renal (secondary) hypertension; it is uncertain whether these two types of hypertension have differences in their pathological phenotype in pregnancy. The sample size was too small to investigate a difference between primary and secondary hypertension however this was not deemed to be relevant to the pathophysiology of endothelial dysfunction. One important consideration for women with CHT is the length of time a woman has had hypertension and other cardiovascular co-morbidities, as the extent of damage to the endothelium and hence vascular functionality is dependent on time. In the animal study by Khammy et al. they noticed that the hypertension induced within the rabbit produced a heterogeneic response in vessels from different vascular beds (Khammy et al., 2016). This differential response infers that hypertension may not manifest as general vascular dysfunction in all vascular beds. The choice of agonist is another factor to consider. This current study used U46619, which binds to the thromboxane A2 receptor and works through the phospholipase C pathway to generate IP3 and DAG to stimulate contraction by increasing intracellular calcium. Other studies have used other vasoconstrictive agents such as endothelin-1, which binds to endothelin receptor-A and endothelin receptor-B2 and angiotensin-II that operates through angiontensin-1 receptors and noradrenaline and methoxamine, which stimulate α1-adrenoreceptors. Hence, the increased vasoconstriction mentioned in other studies may be due to receptor-specific responses. In the present study, thromboxane receptor expression, agonist affinity and signal transduction post- receptor activation may not have been enhanced, resulting in no difference in contraction between normotensive and CHT OAs.

4.5.3.2 Bradykinin-induced relaxation of omental arteries from normal and hypertensive pregnancies In the relatively small sample size for these studies, we did not observe a significant difference in OA relaxation to bradykinin (BK) between PE, CHT, superimposed PE and normal pregnancies. However, all bradykinin dose response curves were performed post- incubation and involved the endothelium being exposed to DMSO, hence we cannot 180

confirm whether the endothelium was or was not functional in these diseased groups. The solvent DMSO was the control used in order to dissolve pitavastatin. The initial assessment of endothelial function involving the addition of a single high dose of bradykinin following the first U46619 dose response curve revealed no difference in endothelial function between normal, CHT and superimposed PE vessels but a trend towards endothelial dysfunction in PE vessels. These different groups displayed significant heterogeneity in their endothelial responses to the vasodilator bradykinin.

Whilst the present study did not confirm endothelial dysfunction in PE, CHT or superimposed PE OAs, other studies have demonstrated attenuated endothelium- dependent relaxation in vessels from hypertensive pregnancies compared to normotensive pregnancies. Myometrial vessels were shown to have diminished endothelium-dependent relaxation to BK compared to vessels from normal pregnancies following pre-constriction to a suprapharmacological dose of vasopressin (Ashworth et al., 1997). This is supported by other studies that have shown blunted endothelial-dependent relaxation to vasodilatory agents BK and acetylcholine (ACh) in myometrial and subcutaneous resistance fat arteries from women with PE (Knock and Poston, 1996; Kublickiene et al., 1997; Mccarthy et al., 1994a). However, Pascoal et al. demonstrated reduced endothelial-dependent relaxation to ACh in OAs from PE women; relaxation to BK was similar between PE and normotensive women (Pascoal et al., 1998). In an in-vitro comparative study, which assessed endothelial- dependent relaxation of myometrial arteries following incubation with normal pregnant plasma, plasma from women with PE, or without plasma, reduced BK-induced relaxation was seen in vessels exposed to PE plasma (Hayman et al., 2000). Using wire myography, Wareing et al showed that omental resistance arteries from PE pregnancies show reduced relaxation to BK compared to normal pregnancies following pre-constriction to vasopressin (Wareing et al., 2006b).

In patients with essential hypertension, impaired endothelium-mediated dilation to ACh has previously been observed as evidenced by reduced blood flow and vascular resistance compared to normotensive control groups, believed to be associated with a defect in endothelium-derived NO. These findings were confirmed using strain-gauge plethysmography to measure forearm vascular responses of brachial artery in vivo (Panza et al., 1993; Panza et al., 1990). Finally, Mori et al. noticed that FMD was reduced in PE and CHT women compared to normotensive women post-delivery. FMD normalised in the PE

181

women while FMD in the CHT women improved but remained lower than normotensive women after delivery (Mori et al., 2014). When this same group split the PE women into early and late onset PE, the outcome differed slightly. When pregnant women with early or late onset PE were compared to CHT women or normotensive women, it was found that FMD was significantly reduced in mild/severe PE and CHT women relative to normotensive women. In contrast, FMD increased in the late-onset PE group and normalized 3 months post-delivery, whereas early-onset PE and CHT women still had reduced FMD versus normotensive controls at this postnatal time point.

It is difficult to confirm whether the present study supported the findings above because of the relatively small sample size and the fact that endothelial function was not assessed formally using a dose-response prior to statin incubation. As mentioned in the methodological section 4.5.2, the solvent DMSO was used as the control for pitavastatin experiments and the results suggested that following a 2 hr incubation with this control, BK-mediated relaxation was similar between normal and pathological groups. 0.1% DMSO was used in the myograph baths in the present study, as in the study by Brownfoot et al. however, this dose of DMSO was able to cause maximal relaxation of vessels from normal, PE, CHT and superimposed PE pregnancies, potentially masking endothelial dysfunction. This being said, the assessment of endothelial function conducted pre-incubation, at a one- off dose, revealed that the BK-mediated relaxation was not different between normal, CHT and superimposed PE vessels, with only PE vessels showing a trend towards blunted relaxation. However, numbers are too small to be certain of this outcome. This assessment of endothelial function highlighted the intra- and inter- heterogeneity in groups as they each produced different relaxatory responses to the single dose of bradykinin.

The current study showed that normal, PE and CHT control vessels relaxed by ~80% to BK (0.1–1000 nM/L) following the 2 hr incubation while superimposed PE vessels relaxed by ~90%. Other groups have shown varying magnitudes of relaxation of PE vessels. In PE myometrial vessels, Ashworth et al. showed that they relaxed only by 5% to BK (10-10 M/L to 10-6 M/L) relative to normal vessels. However, Knock and Poston demonstrated that PE subcutaneous fat arteries relaxed by ~80% to BK (10-9 to 10-5 M/L) but were less sensitive to this agent. Similarly, Wareing et al. showed that OAs from PE pregnancies relaxed by ~80% to BK (10−10 to 10−6 M) but full dilation only occurred at the higher doses, indicating reduced sensitivity. Using OAs from PE pregnancies, Pascoal et al. showed that OAs could

182

only relax by 10% to ACh but showed normal vasodilation to BK (BK, 0.01–10 µM) relative to normal vessels. The doses of bradykinin used in these studies were similar to those of the present study except in the case of Pascoal et al. who used slightly higher BK concentrations. Factors contributing towards these differences include the vascular bed of choice; vessels from the uteroplacental circulation i.e. myometrial vessels, may respond differently to systemic vessels due to there being a receptor up-regulation or down- regulation in one vascular bed and not the other. Moreover, differences could be agonist- specific. For example, some studies reported impaired relaxation to ACh but not BK, suggesting a defect in a G protein-mediated signal transduction pathway that is pertinent to effectiveness of that agonist, suggesting selective endothelial dysfunction.

Further reasons for differences in contraction/relaxation results relative to the existing literature could include whether these studies included FGR/SGA fetuses in their study as was done for the hypertensive cohorts in the present study. Of the studies quoted above, only 4 groups (Mishra et al., Wareing et al., Pascoal et al. and Hayman et al.) reported fetal weight or IBR values for normotensive and PE groups while other groups reported neither and focused on maternal characteristics only. It is uncertain how FGR/SGA complicated hypertensive pregnancies could have affected vascular reactivity of OAs and hence could have confounding effects.

4.5.4 Effect of Short-Term Statin Treatment on Omental Arteries from Normal Pregnancies and Pathological Pregnancies

4.5.4.1 Potassium- and U46619- induced contraction of omental arteries from normal and pathological pregnancies Following a 2 hr incubation with pravastatin or pitavastatin, neither were shown to alter contraction to U46619 when expressed as kPa or % KPSS in OAs from normal pregnancies. In pathological pregnancies, a 2 hr incubation with pitavastatin did not alter U46619- mediated contraction when expressed as kPa or % KPSS in OAs from PE, CHT or superimposed PE pregnancies.

Independent of their cholesterol-lowering ability, statins have been shown to increase the bioavailability of NO through mechanisms including, but not limited to, raised expression of eNOS and activation of the eNOS enzyme, as well as decreasing oxidative stress (Takemoto et al., 2001). Tesfamariam et al. demonstrated that when aortic vessels from male Sprague-Dawley rats were incubated with simvastatin, atorvastatin, or 183

pravastatin (0.5, 5, or 50 µM) or vehicle (DMEM) for 24 hrs, phenylephrine-induced contraction was reduced by simvastatin and atorvastatin but not pravastatin (Tesfamariam et al 1999). A similar effect was observed by Rossini et al. who showed that 1 µM simvastatin attenuated U44619 contraction following a 2 hr incubation in male mesenteric rat arteries (Rossini et al., 2011). Finally, in an ex-vivo study using paired internal mammary arteries, a 6 hr incubation with 5 μmol/L atorvastatin was able to rapidly reduce the

·− generation of free radicals O2 , suggested to occur through the improved coupling of eNOS (Antoniades et al., 2011).

The present study did not demonstrate an added benefit of statin incubation on U46619-mediated constriction. Firstly, it is important to note that in the present study, as discussed in section 4.5.2, there were significant differences in contraction pre- and post- incubation with both the DMSO and water control OAs. This observation was a time- dependent effect involving the desensitization of OAs to the vasoconstrictor agonist U46619, which could have directly influenced the effects of statins on OA constriction by masking any potential effects, especially if the vascular effects of statins were modest. Another reason why our results may differ from the literature mentioned above, includes the choice of agonist as Tesfamariam et al. included phenylephrine, which stimulates α1- adrenergic receptors on smooth muscle, while U46619 stimulates thromboxane A2 receptors and both then stimulate contraction through the phospholipase C pathway. Statins do not appear to modify this pathway directly or influence the downstream secondary messenger pathway to reduce constriction in OAs from normal pregnancies. Furthermore, the study by Tesfamariam et al. involved a 24 hr statin incubation while the present study only conducted a 2 hr incubation. Potentially if the incubation time was increased, this may have altered vascular function as previous in vitro studies involving statins have quoted 2 hrs however the use of DMSO was a confounding factor for the current study. Differences could also be species-specific as some of the studies discussed above involved animals from non-pregnant physiological states while our study involved human OAs from pregnancy. Finally, the present study used DMSO, which is believed to have relaxatory effects of its own and could have weakened the vascular response of OA to U46619.

184

4.5.4.2 Endothelium-dependent bradykinin- and endothelium- independent SNP-induced relaxation of omental arteries from normal and pathological pregnancies Following a 2 hr incubation with pravastatin or pitavastatin, there was no effect on BK or SNP-mediated relaxation of OAs from normal pregnancies. In OAs from PE, CHT or superimposed PE pregnancies, blunted endothelium-dependent relaxation was not observed, and a 2 hr incubation with pitavastatin did not alter BK-induced relaxation.

In male Wistar rats, when the aorta and superior mesenteric arteries (SMA) were pre-constricted to EC80 noradrenaline, the direct application of simvastatin produced relaxation in a concentration-dependent manner (simvastatin 10-6-3x10-4 for aorta; simvastatin 10-9-10-4 in SMA) in both the presence and absence of a functional endothelium. These effects were attributed to the release of both NO and vasodilator

− products from cyclooxygenase by a mechanism sensitive to O2 scavengers, superoxide dismutase (SOD) (Alvarez De Sotomayor et al., 2000). It was suggested that this relaxation is associated with both Ca2+ release from intracellular stores and blockade of extracellular Ca2+entry (Alvarez De Sotomayor et al., 2001). Finally, Mukai et al. demonstrated that, following a 2 hr incubation with the hydrophobic cerivastatin (1 µM), ACh-induced endothelium-dependent relaxation was enhanced in the rat aorta via the PI3 kinase/Akt pathway and also endothelium-independent relaxation via Kv channel-mediated smooth muscle hyperpolarizations (dose-dependent: 1-100 µM). Direct application of fluvastatin (a moderately hydrophilic statin) was also used in this study and was able to stimulate relaxation of vascular smooth muscle in the aorta and mesenteric arteries in a dose- dependent manner (1-300 µM); vessels were pre-constricted to prostaglandin F2 alpha (Mukai et al., 2003).

The present study did not show an additional effect of statins in OAs from normal or pathological pregnancies. An important issue to highlight is that OAs from the pathological pregnancies achieved a relaxation equivalent to vessels from normotensive pregnancies and therefore there was no improvement to be made with statins. The use of DMSO may have contributed to this, which could have altered endothelial function of OAs post-incubation. Furthermore, different doses of statins were used in other studies and there were differences in experimental design, as direct application of statins in the form of a dose-response curve was not performed in the present study. Additionally, the current 185

study dealt with whole resistance vessels in a closed system in the absence of flow and other physiological factors from the body and hence 2 hrs may not have been sufficient time to observe an effect on the endothelium. Clinically, in a high-risk subpopulation, women may be exposed to statins during pregnancy in order to prevent the onset of PE and so maternal vessels could be exposed to statins despite the absence of disease. Thus, this study suggests that we did not see any detrimental effects of pravastatin or pitavastatin on vascular reactivity of vessels from normal pregnancies.

4.5.4.3 Limitations of Study The caveat of the study was that no BK dose response curve was performed with a positive control (vessel alone in PSS pre-statin incubation). It is also important to acknowledge that the women used in the present study were on antihypertensive medication at various points within their pregnancy and this may have contributed to vascular function ex vivo. Nifedipine and labetalol are commonly used as the first-line treatment for hypertension in pregnancy. The antihypertensive labetalol works by inhibiting α- and β-adrenergic receptors and has been found to preserve uteroplacental blood flow and has a rapid onset of action with effects seen within 2 hrs; adverse effects include peripheral vasoconstriction (Podymow and August, 2008). The calcium-channel blocker nifedipine can also be used to manage hypertension and inhibits influx of Ca2+ ions into the smooth muscle, leading to vasodilation (Magee, 2001). A crucial point is that women with CHT will have potentially been exposed to antihypertensive medications for longer than women with PE and consequently this could have affected ex vivo vascular function of OAs. Alternatively, the vascular bed of women with CHT may have adjusted to the effects of antihypertensive medication over time while the vascular bed of women with PE may be more sensitized to the effects of antihypertensive medications reflecting in the ex vivo vascular function. An unanswered question is: how do statins affect vascular reactivity of OAs pre-exposed to antihypertensive medication? It is uncertain how these drugs may interact with one another and hence affect vascular function. Finally, low-dose aspirin is used in a high-risk population to prevent PE onset i.e. in chronically hypertensive women (Braunthal and Brateanu, 2019). Aspirin exerts its effects by inhibiting thromboxane production and promoting prostacyclin release to attenuate vasoconstriction (Wallenburg et al., 1986). Once again, it is uncertain how statins may affect the vascular function of vessels pre- exposed to aspirin.

186

4.5.5 Conclusion Hypertensive disorders of pregnancy such as PE, CHT and superimposed PE pose clinical challenges due to their pathological complexity and the associated maternal and perinatal morbidity and mortality. This study hypothesized that pitavastatin would improve vascular reactivity of OAs from hypertensive pregnancies. This study has shown for the first time that acute exposure (2 hr) of OAs from hypertensive pregnancies to pravastatin and pitavastatin (1 µM) does not reduce U46619-mediated contraction or enhance BK- mediated relaxation of vessels. A caveat of the study was the small sample size of hypertensive pregnancies and the lack of confirmation of endothelial dysfunction in the absence of a BK dose-response prior to statin exposure in OAs from hypertensive pregnancies, making direct comparisons to the literature difficult. The effects of chronic pitavastatin exposure (at a clinical dose) in an in vivo eNOS-/- mouse model, which displays chronic hypertension, endothelial dysfunction and FGR, will be discussed in the next chapter and a different vehicle control was implemented to avoid confounding effects. A positive observation of this study was that acute statin treatment of OAs from normal or hypertensive pregnancies did not cause any detrimental effects on vascular reactivity.

187

Chapter 5: Acute and chronic effects of pitavastatin on maternal and fetal vascular function in the endothelial nitric oxide synthase knockout (eNOS-/-) mouse

188

5.0 Results

5.1 Introduction The studies in this chapter were carried out to examine short (ex vivo) and longer-term (in vivo) effects of pitavastatin on vascular function in the endothelial nitric oxide synthase knockout (eNOS-/-) mouse. This model displays pre-pregnancy hypertension, which persists throughout pregnancy (Hefler et al., 2001b). The eNOS-/- mouse has previously been described as a PE-like model, but is best defined as a model of chronic hypertension, also presenting with proteinuria, placental dysfunction and FGR (Kusinski et al., 2012). Studies have shown that, as a result of s-Flt’s antagonistic effect on VEGF in PE, there is a decrease in the phosphorylation of Ser 1177 of the endothelial nitric oxide synthase (eNOS) enzyme, resulting in a reduction in activity (Gelinas et al., 2002). This is important because the eNOS enzyme significantly contributes to nitric oxide (NO) bioavailability in the endothelium, which plays a key role in modulating vascular tone.

The vasodilator nitric oxide (NO) plays a significant role in cardiovascular adaptations in pregnancy and is important in the observed increases in uterine and fetoplacental blood flow, the maintenance of low vascular resistance in the fetoplacental circulation and the reduction in peripheral vascular resistance (Veerareddy et al., 2002). In the pregnant eNOS-/- mouse, Kusinski et al. demonstrated that these mice exhibit dysregulated vascular adaptations to pregnancy, and FGR. Furthermore, uterine artery constriction is significantly increased while endothelium-dependent relaxation is significantly blunted in eNOS-/- versus wildtype (WT) mice (Kusinski et al., 2012). Additionally, a study by Kulandavelu et al. showed that, in eNOS mice, there is a reduction in uteroplacental perfusion, with a reduced uterine artery diameter leading to increased vascular resistance, reduced spiral artery length and hypoxia in the junctional zone of the placenta (Kulandavelu et al., 2012).

Studies in PE-like rodents involving the administration of the statin pravastatin during pregnancy have shown encouraging results and have demonstrated an improvement in angiogenic balance, ameliorated glomerular injury, attenuation of vascular constriction and lowered blood pressure. These effects occurred without altering maternal cholesterol levels and without an increase in pup resorption or birth weights (Ahmed et al., 2010; Costantine et al., 2010; Fox et al., 2011; Kumasawa et al., 2011; Singh et al., 2011).

189

Animal studies such as these have promoted the initiation of trials such as “Pravastatin for the prevention of pre-eclampsia, a double blinded placebo-controlled pilot trial to collect preliminary maternal-fetal safety data for pravastatin during pregnancy, and evaluate its pharmacokinetics when used as a prophylactic daily treatment in high-risk pregnant women.” In this trial, women with a high risk of developing PE were recruited and administered 10 mg, 20 mg or 40 mg of pravastatin prophylactically to assess the safety and efficacy of pravastatin in reducing PE incidence (Costantine and Cleary, 2013).

In the absence of being able to assess the effects of statins, administered in vivo, on the contractility and relaxation of maternal systemic vessels from women, we assessed this in the eNOS-/- mouse. Pitavastatin was chosen due to its novelty in pregnancy studies but also as pitavastatin decreased U46619 contraction in chorionic plate arteries from normal and PE pregnancies and hinted towards an effect in omental arteries from PE pregnancies; sample size was, however, a limiting factor (as discussed in previous chapters).

Statins have been found to exert their protective effects on the endothelium through the enzyme eNOS to promote NO release and subsequent vasodilation of vessels. By employing the eNOS-/- mouse, this model will shed light on potential mechanisms that may underpin any effects of pitavastatin on vascular function including any eNOS- independent effects, whilst also informing whether pitavastatin has any deleterious effects on litter size and fetal weight. Pre-clinical studies have shown that pitavastatin is able to activate the P13/AKT and AMPK pathways and exert its pleiotropic effects through eNOS- independent mechanisms which may enhance vascular function (Davignon, 2012; Mitsuhashi et al., 2018; Yagi et al., 2008).

The acute (2 hr treatment ex vivo) and chronic (8 day treatment in vivo) effect of pitavastatin on maternal and fetal vascular function in the eNOS-/- mouse was investigated. The rationale behind the short-term approach was because the pleiotropic effects of statins are believed to be fast acting and evident within hours to days, so we wanted to explore whether a short exposure to statin would be sufficient to improve vascular reactivity in this mouse model. Chronic exposure was explored to see whether the pleiotropic effects of statins on vascular function persisted but also assess safety by investigating whether long- term exposure had detrimental effects on fetal development.

190

5.2 Hypothesis and aims

5.2.1 Hypotheses

‘In eNOS knockout mice, short-term (2h) exposure of vessels to pitavastatin will result in improved uterine artery function and have no detrimental effects on umbilical artery function.’

‘In eNOS knockout mice, pitavastatin treatment from embryonic days 10.5 to 18.5 will improve maternal vascular (mesenteric artery and uterine artery) function without having detrimental effects on umbilical artery function or fetal growth.’

5.2.2 Aims 1. To confirm whether vascular reactivity of uterine and umbilical arteries in eNOS knockout mice is altered compared to wild type mice.

2. To assess whether vascular reactivity of uterine and umbilical arteries is altered in eNOS knockout and WT mice following a 2 hr pitavastatin exposure.

3. To assess whether vascular reactivity of mesenteric, uterine and umbilical arteries in eNOS knockout and WT mice is altered following pitavastatin treatment from E10.5 until E18.5.

4. To assess the effect of pitavastatin treatment on fetal and placental weight in WT and eNOS knockout mice.

191

5.3 Materials and Methods

5.3.1 Acute (2h) incubation of vessels with pitavastatin (ex vivo)

5.3.1.1 Sample Collection and Protocol Experiments were performed in accordance with the UK Animals (Scientific Procedures) Act of 1986 under the authority of a Home Office licence (PPL P9755892D). The Animal Welfare and Local Ethical Review Board of the University of Manchester approved all protocols. The methods outlined in this project adhere to the ARRIVE guidelines (Kilkenny et al., 2012). Mating regimes for mice were described in chapter 2 (Materials and Methods section).

On day E18.5, WT (N=10) and eNOS-/- (N=16) pregnant mice were euthanased (cervical dislocation followed by confirmation of cessation of circulation appropriate under ASPA schedule 1) and a laparotomy and hysterectomy performed; the entire uterus was placed in ice-cold PSS. Fetuses were then killed by a schedule 1 approved method (decapitation). Fetal and placental weights were recorded. If the litter size was <4, animals were excluded from the study. Litters (N=3 from WT and N=3 from eNOS-/- mice) were excluded due to incorrect dates of vaginal plugs being recorded by technical staff, resulting in a final number of N=10 WT dams and N=16 eNOS-/- dams.

5.3.2 Blood vessel Dissection and Normalisation

5.3.2.1 Maternal Uterine Artery The methods for the dissection of the uterine artery are fully detailed in the methods chapter. Once uterine arteries were isolated from the left and right sides of the uterine horn, the isolated portions were cut into 4 segments resulting in 8 segments in total. Eight segments were prepared but, to allow for random selection of vessels, only 6 were chosen and used for subsequent experiments. These 6 vessels were each mounted separately on two 40 µM stainless steel wires and secured in the jaws of the myograph. The vessels were left to equilibrate in PSS at 37°C (20% O2/ 5% CO2/ balance N2). The normalisation process was identical to that mentioned previously in methods chapter (normalised to 0.9 of L13.3kPa, section 2.11 in the methods).

5.3.2.2 Fetal Umbilical Artery 4 fetuses were randomly chosen per litter (2 from each uterine horn) and umbilical arteries from these fetuses were identified and dissected free under a stereomicroscope. Each umbilical artery was mounted separately on two 40 µM stainless steel wires and secured 192

in the jaws of the myograph. The vessels were left to equilibrate in PSS at 37°C (5% O2/ 5%

CO2/ balance N2). The normalisation process was the same as for chorionic plate arteries described in methods chapter 2 (normalised to 0.9 of L5.1kPa, section 2.11).

5.3.3 Wire Myography Wire myography was performed to assess vascular function. To assess viability, vessels were exposed to 2x high dose potassium washes (KPSS, 120 mM high potassium salt solution) before the thromboxane mimetic U46619 was used to elicit vasoconstriction in uterine and umbilical arteries in the form of a dose response curve. Vessels were then exposed to 1 µM pitavastatin for 2 hrs. To assess vasorelaxation, the endothelium- dependent vasodilator acetylcholine (ACh; 1x10-10-10-5M) was used for uterine arteries and the NO-donor sodium nitroprusside (SNP, 1x10-10-10-5M) for both uterine and umbilical arteries. Prior to carrying out an ACh dose response curve, endothelial function was assessed by preconstricting vessels with the vasoconstrictor phenylephrine (PE; 10-5), once constriction stabilised a single high dose of ACh (10-5) was added. Furthermore, if the vessel contractions were <3kPa (because of damage during mounting or the normalization stage), or vessels self-collapsed prior to a relaxation dose response curve, these vessels would be excluded from the dataset. The experiment ended with a final KPSS contraction. A schematic diagram of the full protocol can be found in figure 5.3 below.

The exclusion process meant that the number of vessels isolated were not all used for a full experiment hence contraction and relaxation experiments were allocated to remaining functional vessels available. Experimentally, this meant that the total number of contraction, endothelium-dependent (ACh) and endothelium-independent (SNP) experiments (N) in eNOS-/- and WT mice were different.

193

Tensio

Time

Basal Tone calculation:

 The tension in (mN/mm) was recorded pre and post incubation with statin.

 This value was then converted to kPa  Basal tone = post-statin tension (kPa) minus pre-statin tension (kPa)

Figure 5.1: A representative myography trace for ex vivo mouse experiments (uterine artery). Following normalisation, 2x KPSS washes were performed. To assess endothelial

function, uterine arteries were preconstricted to phenylephrine and relaxed with -5 acetylcholine (ACh, 10 ). This was followed by a U44619 dose response curve, then vessels were randomly assigned to vehicle (DMSO) or 1µM pitavastatin, which

were dissolved in 6ml of PSS and left to incubate for 2 hrs in the myograph baths. Post-incubation, relaxation was assessed by pre-constricting uterine arteries to EC80 of U46619 and performing an ACh dose response curve followed by a final KPSS contraction. X-axis denotes times and Y axis denotes Tension (mN/mm). PE = phenylephrine and U44619 = thromboxane A mimetic. 2

5.3.4 In vivo Exposure to Pitavastatin in WT and eNOS-/- mice

5.3.4.1 Experimental Allocation of Mice for In Vivo Studies In the chronic pitavastatin (in vivo) study, once mice were confirmed to be pregnant at E10.5, eNOS-/- or C57Bl/6J (WT) dams were randomised to either pitavastatin (6 µg/ml) in 0.05% carboxymethylcellulose (CMC) or CMC vehicle (0.05%) in drinking water and treated until E18.5. The dose of pitavastatin was clinically relevant and equivalent to an oral dose

194

of 4 mg in humans. Mice were randomised by creating a blocked randomisation list (via https://www.sealedenvelope.com/simple-randomiser/v1/lists) according to treatment. Assessments of litter size, fetal and placental weight, fetal biometric measurements, and ex vivo wire myography of uterine, mesenteric and umbilical arteries were conducted at E18.5.

WT or eNOS-/- Dams

Randomisation at E10.5

6 µg/ml pitavastatin + 0.05% CMC in 0.05% CMC in drinking drinking water water Tissue Harvest at E18.5 water

Urine and plasma Myography performed Litter size recorded and collected. Maternal organs on uterine, mesenteric fetuses and placentas (brain, heart, kidney, and and umbilical arteries were weighed. spleen, liver, left and right to assess vascular kidney) harvested. function.

Fetal biometric measurements taken. Uterine horn and

mesentery were also harvested. All placentas and 2-3 pups/litter were snap Figure 5.2: Experimental allocation of mice for in-vivo studiesfrozen.

In the chronic pitavastatin (in vivo) study, mice were confirmed to be pregnant and randomised to either pitavastatin (6 µg/ml) in 0.05% carboxymethylcellulose (CMC) or CMC vehicle (0.05%) in drinking water and treated until E18.5. Fetal and placental weights and fetal measurements and ex vivo wire myography of uterine, mesenteric and umbilical arteries were conducted at E18.5.

195

5.3.4.2 Experimental Flowchart Illustrating Use of Animals in In Vivo Dosing Study Litters were excluded due to incorrect dates of vaginal plugs being recorded by technical staff (N=1 WT, N=3 from eNOS-/- mice). One WT mouse (treatment group) littered prematurely (day E17.5) and so was excluded from the study.

Exclusions:

3 WT Litter

4 eNOS-/- Litter

8 WT Fetuses

22 eNOS-/- Fetuses

CMC (0.05%) control mice (Final Pitavastatin (6µg/ml) treated mice numbers): (Final numbers): 6 WT Litter 7 WT Litter 6 eNOS-/- Litter 8 eNOS-/- Litter 58 WT pups 57 WT pups

49 eNOS-/- pups 53 eNOS-/- pups Fetal and placental measurements Fetal and placental measurements taken taken

Figure 5.3: Experimental flowchart illustrating use of animals in the in vivo

dosing study with pitavastatin.

Number of litters and corresponding number of pups for each treatment arm are shown in the primary tier. All measurements were taken at E18.5. Reasons for exclusion of dams were either

that the female mouse delivered early or animal plug dates were incorrectly noted.

5.3.5 Sample Collection and Protocol On gestational day E18.5, pregnant mice were euthanased (cervical dislocation followed by confirmation of cessation of circulation appropriate under ASPA schedule 1) and a laparotomy and hysterectomy performed. Following removal of the uterus from pregnant animals, a hysterectomised weight was recorded. The mesentery with the jejunum and ileum still attached was isolated from the abdominal cavity and surrounding tissues for later dissection of the mesenteric artery. Maternal brain, heart, liver, kidneys and spleen

196

were harvested, weighed and snap frozen on dry ice. Fetuses were killed by a schedule 1 approved method (immersion in ice-cold solution and confirmed by decapitation).

5.3.6 Fetal and Placental Measurements at E18.5

5.3.6.1 Assessment of Fetal weight, Placental weight and Fetal Biometric Measurements Fetal and placental weights were recorded approximately 1-2 hrs after laparotomy and post-isolation of uterine and umbilical arteries, but were maintained in PSS solution to avoid desiccation. All fetuses and placentas were gently blotted dry to remove any excess fluid and weighed. Fetal biometric measurements (crown-rump (C/R) length, abdominal (Ab) circumference and head circumference) were measured using cotton thread and a ruler in a systematic manner and measurements recorded.

5.3.7 Blood Vessel Dissection and Normalisation Uterine and umbilical arteries were dissected and normalised as performed in the ex vivo studies (section 5.3.2.1 and 5.3.2.2).

5.3.7.1 Maternal Mesenteric Arteries Second order mesenteric arteries were dissected from the mesentery. Care was taken to completely remove surrounding adipose fat tissue and not to overstretch arteries. Two vessels were dissected from separate branches of the mesenteric artery and mounted on two 40 µM steel wires and secured in the jaws of the myograph. These vessels were equilibrated and normalised in the same manner as the uterine arteries (5.3.2.1).

5.3.8 Wire Myography A schematic diagram of the full protocol can be found in figure 5.4 below. Following equilibration (20-30 mins), vessel viability was assessed via 2x KPSS (120 mM) exposures with 10-20 mins intervals in between. If the constriction of the vessel was less than 3.0 kPa or a vessel self-collapsed following pre-constriction, it was excluded from the study (1 of each vessel type, uterine, mesenteric and umbilical arteries, was excluded across 3 litters in both genotypes, eNOS-/- (n=3) and WT mice (n=3). After each KPSS exposure, the chamber was emptied and filled with PSS to restore the vessel to passive tension (baseline pre-constriction). Following KPSS constrictions and PSS wash-outs, a U46619 dose- response curve was performed (10-10 - 2x10-6 at 2 min intervals).

197

-2 -6 Following the calculation of EC80, endothelium-dependent relaxation to ACh (10 -2x 10 M at 2 min intervals) was carried out on uterine and mesenteric arteries only (not umbilical arteries) pre-constricted to an EC80 dose of U46619. Following a wash with PSS, endothelial- independent relaxation to SNP (10-10 – 2x10-6M at 2 min intervals) was investigated in uterine, mesenteric and umbilical arteries pre-constricted with an EC80 dose of U46619. After a further wash with PSS, vessels were left for 10 minutes before a final KPSS constriction to ensure vessels were still functional and constriction had not been compromised.

Tensio

Time

EC80 EC80 U46619 U46619 Normalisation

Figure 5.4: A representative image of the myography protocol (for uterine and mesenteric arteries) following pitavastatin treatment in vivo.

Following normalisation, 2x KPSS washes, uterine arteries were constricted to U44619 to

conduct a dose response curve. Following a washout with PSS, they were then pre- constricted to EC80 of U44619 to perform an ACh dose response curve followed by a PSS wash out and then a SNP dose response curve. The experiment was terminated with a final KPSS contraction. X-axis denotes times and Y axis denotes Tension (mN/mm). U44619 =

thromboxane A2 mimetic.

5.3.9 Data Analysis Vascular function was the primary outcome of these studies and thus studies were powered according to data from similar studies within this laboratory, based on 80% power and a 5% significance level (Renshall et al, 2018 and Costantine et al., 2010). U46619- induced constriction and ACh and SNP-induced relaxation were analysed using repeated measures two-way ANOVA and expressed as mean±SEM. For myography studies in the in vivo dosing study, a mean per litter was taken when more than one vessel was used, and 198

this is the value represented in the graphs. Fetal and placental weights at E18.5 were analysed using two-way ANOVA, in order to assess the effect of both genotype and treatment and their interaction, with Sidak’s post-hoc test. Normality tests were conducted using the Kolmogorov-Smirnov test to assess whether data fitted to a Gaussian distribution; the normality test result determined whether paired parametric or non-parametric t-tests were conducted.

5.4 Results: Study focused upon Acute Exposure to Vessels with 1 µM Pitavastatin (ex vivo)

5.4.1 Maternal weight in WT and eNOS-/- mice at embryonic day 18.5 Maternal weight of eNOS-/- mice was significantly lower than WT mice at E18.5 (Figure 5.5; p= 0.0099).

Figure 5.5: Maternal weight in eNOS - /- and WT mice at E18.5

The maternal weight of eNOS-/- mice (N=16) at embryonic day 18.5 was significantly lower than WT mice (N=10). Horizontal line denotes median. Data are median± IQR. Mann- Whitney U test. **P<0.01.

199

5.4.2 Fetal and Placental Weights in WT and eNOS-/- Mice at Embryonic Day 18.5 (ex vivo study) There was no significant difference observed in litter size between eNOS-/- and WT mice (figure 5.6 A; p=0.344). Fetal weight in eNOS-/- mice was, however, significantly lower at E18.5 compared with WT mice (figure 5.6 B; p<0.001). When plotted as a frequency distribution curve, the 5th percentile (vertical dashed line) on the WT curve was 0.942g and demonstrated that 27% of eNOS-/- fetuses fell below this (figure 5.6 C; p<0.0001 chi- squared test). No significant difference was observed in placental weight between eNOS-/- and WT mice (figure 5.6 D; p=0.856). There was a significantly reduced fetal: placental weight ratio in eNOS-/- vs WT mice (figure 5.6 E; p=0.0003). Across all WT litters (N=10), there were 3 resorptions compared to 9 resorptions in eNOS-/- litters (N=16).

200

A B

15 1.3 ****

)

g e

( 1.2

t i h 1.1

s 10

i r

e 1.0

l 0.9 L

5 a t

e 0.8 F 0 - -/ 0 T - S W -/ T O S W N O e N e

C 5th percentile 40 -/-

n eNOS

i (n=116) t

a 30

l C57 u

p (n=78)

o 20

f o

% 0 0.0 0.5 1.0 1.5 2.0 Fetal weight (g)

D E

t h

) ***

g i g 16

( 0.12

0.11 w h

g i

0.10 a t e 12

0.09 n

e l

0.08 c 10

t a

l n 0.07

p 8

: l

c 0.06

0 F 0 - /- -/ T - T S W S W O O N N e e

Figure 5.6: Fetoplacental measurements from WT and eNOS - /- mice.

There was no difference in litter size (A), fetal weight of eNOS-/- fetuses was significantly

lower than WT fetuses (B), a significant proportion of fetuses were below the 5th centile in the fetal weight frequency distribution curve (C). There was no difference in placental weight between eNOS-/- and WT mice (D). The fetal: placental weight ratio in eNOS-/- was significantly less than that of the WT mice (E) at embryonic day 18.5 (N=16 for eNOS-/- and N=10 for WT). For A, B, D and E, each symbol corresponds to an average (mean) per litter.

Horizontal line denotes median ± IQR. In C, vertical black dashed line denotes the 5th centile of WT fetal weights (0.942g). Individual pup n’s: WT n = 78, eNOS-/- n = 116. Mann-Whitney U test * p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001.

201

5.4.3 Diameter of uterine arteries from WT and eNOS-/- mice at embryonic day 18.5 prior to ex vivo statin exposure The diameter of uterine arteries from WT and eNOS-/- mice was not significantly different between the two groups (Figure 5.7 (eNOS-/- vs WT; p=0.113)).

Figure 5.7: Uterine artery diameter from eNOS - /- and WT mice at E18.5 prior to statin exposure

-/- Uterine artery diameter from eNOS mice was not significantly smaller than that from WT mice prior to statin incubation. Data presented as the average diameter of uterine artery segments per mouse for each group (N=10 for eNOS-/- mice and N=9 for WT mice). The horizontal line denotes median and data are expressed as median±IQR. Mann-Whitney U test. * p<0.05

202

5.4.4 Effect of pitavastatin (1 µM) on basal tone of uterine arteries in WT and eNOS-/- mice Following a 2 hr incubation with 1 µM pitavastatin, there was no significant difference in

-/- the basal tone of uterine arteries (0.9 of L13.3KPa) compared to DMSO controls in eNOS mice (Figure 5.8 A (DMSO vs pitavastatin; p=0.670)). However, in uterine arteries from WT mice, basal tone was significantly higher following pitavastatin exposure compared to the DMSO control (Figure 5.8 B (DMSO vs pitavastatin; p=0.011)).

A eNOS-/- B WT

) )

a 6 a 6

P MICE

P *

k k

(

n n

o 4 4

i s

e e

2 T 2 T

e e

0 g 0

h h

C -2 C -2

O M O M S  S  M -1 M 1 D n D - ti n a ti t ta s s a a v v ta a i it P P

Figure 5.8: Effect of 2 hr pitavastatin incubation on basal tone in eNOS - /- and WT mice at E18.5

Following a 2 hr incubation with pitavastatin (1 µM), the basal tone of uterine arteries was not significantly different between the statin-exposed group versus control in eNOS-/- but pitavastatin increased basal tone in WT mice; (N=10 for eNOS-/- and N=9 for WT). The horizontal line denotes median and data expressed as median±IQR; Mann-Whitney U test.

*P<0.05.

203

5.4.5 Effect of pitavastatin (1 µM) on contraction responses of uterine arteries in WT and eNOS-/- mice There was no significant difference in contraction to U46619 between uterine arteries from WT and eNOS-/- mice; (Figure 5.9 A; p=0.952)). However, when data was expressed as contraction to U46619 dose as a % of the maximum KPSS contraction, uterine arteries from eNOS-/- mice were significantly more contractile (Figure 5.9 B; p<0.0001)).

Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant effect on uterine artery contraction to U46619 in WT mice; Figure 5.9 C (DMSO vs pitavastatin; p=0.362). When U46619 contraction curves were expressed as maximum KPSS contraction there was again no significant difference between pitavastatin-exposed uterine arteries and comparable control; (Figure 5.9 D (DMSO vs pitavastatin; p=0.898)). However, a trend towards increased U46619 sensitivity was seen in the pitavastatin treated uterine arteries relative to the DMSO group (DMSO vs pitavastatin; p=0.093).

Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant effect on uterine artery contraction to U46619 in eNOS-/- mice; Figure 5.9 E (DMSO vs pitavastatin; p=0.954) and when U46619 contraction curves were expressed as maximum KPSS contraction there was again no significant difference between pitavastatin-exposed uterine arteries and a comparable control; Figure 5.9 F (DMSO vs pitavastatin; p=0.583)). No differences were observed in U46619 sensitivity (EC50 values shown in table 5.1).

204

A B

160 -/- 20 eNOS-/- (10) eNOS (10) 140 WT (9) ****

) WT (9) a

15 n 120

S t k

( 100

r n

10 t 80

n s

% 60

n C e 5 40 T 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10xM [U46619] x 10 M C D

20 WT+DMSO (6) 1 6 0 W T + D M S O (6 )

WT+Pitavastatin (6) 1 4 0 W T + P ita v a s ta tin (6 )

) a

15 n 1 2 0

o i

t (

1 0 0

P r

10 t

o 8 0

K i

n %

n 6 0

o e 5 C

T 4 0

2 0 0 0 -10 -8 -6 -4 -1 0 -8 -6 -4 x x [U46619] x 10 M [U 4 6 6 1 9 ] x 1 0 M E F

20 eNOS+DMSO (4) 160 eNOS+DMSO (4) eNOS+Pitavastatin (4) ) 140 eNOS+Pitavastatin (4)

a 15 P

n 120

i (

S t

100

i P

10 r

t i

K 80 s

% n

o 60 e

5 C T 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM x [U46619] x 10 M Figure 5.9: Effect of 2 hr pitavastatin incubation on contraction responses of uterine arteries from WT and eNOS -/- mice.

Contraction of uterine artery to U46619 expressed as tension (kPa) (A) or as % KPSS (B) in WT (N=9, n=18) and eNOS-/- mice (N=10, n=20). Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant effect on uterine artery contraction in WT mice (C-D) or eNOS-/- mice (E-F) expressed as kPa and %KPSS. U46619 dose response curves were compared using 2-way ANOVA with Sidak’s post hoc test, where applicable (Sidak was chosen as it is more power than the Bonferroni test and assumes comparisons are independent of one another). Data are expressed as mean±SEM; N= number of litters, n=uterine arteries. ****p<0.0001 205

5.4.6 Effect of pitavastatin (1 µM) on relaxation responses of uterine arteries in WT and eNOS-/- mice Blunted uterine artery relaxation to ACh was observed in eNOS-/- compared to WT mice (Figure 5.10 A; p<0.0001). However, there was no significant difference in uterine artery relaxation to SNP between eNOS-/- and WT mice; (Figure 5.10 B; p=0.168).

Following a 2 hr incubation with pitavastatin at 1µM, there was no significant effect on uterine artery relaxation to ACh in WT mice; Figure 5.10 C (DMSO vs pitavastatin; p=0.368). Similarly, uterine artery relaxation to SNP was not significantly different between pitavastatin and vehicle groups in WT mice; (Figure 5.10 D (DMSO vs pitavastatin; p=0.629).

Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant effect on uterine artery relaxation to ACh in eNOS-/- mice; Figure 5.10 E (DMSO vs pitavastatin; p=0.927). Furthermore, uterine relaxation to SNP showed no significant difference between pitavastatin-exposed versus control uterine in eNOS-/- mice; Figure 5.10 F (DMSO vs pitavastatin; p=0.956).

206

A B

1 9

6 120

1 6

6 120 4

6 4

U 100

100

0 C 80 8

t 60

o 60

t n

o n i 40

t ****

o 40

a t

x WT (7) 20 a

a WT (9) l

x 20 e a -/- l eNOS (4)

R -/-

0 e

eNOS (8) 0

R % -10 -8 -6 -4 -10 -8 -6 -4

% x C [ACh] x 10xM D [SNP] x 10 M

9 1

120 9

6 6 4 120

100 6 U

0 8

U 100

C 80

C 80 o

t 60

o 60

i 40 t

n a

o 40

i x

20 t

a l

WT+DMSO (9) a WT+DMSO (7) e

x 20

R l

0 WT+Pitavastatin (9) WT+Pitavastatin(7) e

% 0 R -10 -8 -6 -4 -10 -8 -6 -4 [ACh] x 10xM % E F [SNP] x 10xM

1 120

9 6

1 a 4 6 120

100 6

8 U

100

C 80

8 E

C 80 o

t 60

o 60

i 40

n a

o 40

i t

a 20 l

eNOS+DMSO (8) a eNOS+DMSO (4) e

x 20

R l

0 eNOS+Pitavastatin (8) eNOS+Pitavastatin (4) e

% 0 R -10 -8 -6 -4 -10 -8 -6 -4 [ACh] x 10xM % [SNP] x 10xM

Figure 5.10: Effect of 2 hr pitavastatin incubation on relaxation responses of uterine arteries from WT and eNOS -/- mice.

Reduced relaxation of uterine arteries to ACh in eNOS-/- versus WT dams (p<0.0001) (A) but not for SNP (B) in eNOS-/- (ACh; N=8, n=16, SNP; N=4, n=8) and WT mice (ACH; N=9, n=18, SNP; N=7, n=14). Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant effect on uterine artery relaxation to ACh (C) or SNP (D) in WT mice or in eNOS-/- mice (E) and (F). ACh and SNP dose response curves were compared using two-way ANOVA with Sidak’s post hoc test, where applicable (Sidak was chosen as it is more power than the Bonferroni test and assumes comparisons are independent of one another) . Data are expressed as mean±SEM; N= number of litters, n=uterine arteries. ****p<0.0001

207

5.4.7 Diameter of umbilical arteries from WT and eNOS-/- mice at embryonic day 18.5 prior to statin exposure The diameter of umbilical arteries from eNOS-/- and WT mice was not significantly different prior to statin incubation ((Figure 5.11 (eNOS-/- vs WT; p=0.754).

Figure 5.11: Umbilical artery diameter in eNOS -/- and WT mice at E18.5 prior to statin exposure.

Umbilical artery diameter from eNOS-/- mice was not statistically different to those from WT mice prior to statin incubation (N=12 for eNOS-/- mice and N=9 for WT mice). Data presented as average diameter of umbilical arteries per litter. The horizontal line denotes median and data are expressed as median±IQR. Mann-Whitney U test.

208

5.4.8 Effect of pitavastatin (1 µM) on basal tone of umbilical arteries in WT and eNOS-/- mice Following a 2 hr incubation with pitavastatin (1 µM), there was no significant difference in basal tone of umbilical arteries (0.9 of L5.1KPa) compared to DMSO controls in either WT ((Figure 5.12 A (DMSO vs pitavastatin; p=0.134) or eNOS-/- mice (figure 5.12 B (DMSO vs pitavastatin; p=0.843).

A B

WT eNOS-/-

) 3.5 MICE

) 3.5

a a

P 3.0

( ( 2.5

2.5

n 2.0 n

o 2.0

i s

1.5 s 1.5

n e

1.0 e 1.0

0.5 0.5

0.0 0.0

e g

g -0.5 -0.5

n n

a a -1.0

-1.0 h h

C -1.5 C -1.5 -2.0 -2.0 O M O M S  S  1 1 M - M D in D - t n ta ti s ta a s v a a v it a P it P

Figure 5.12: Effect of 2 hr pitavastatin incubation on basal tone of umbilical arteries in eNOS-/- and WT mice at E18.5

Following a 2 hr incubation with pitavastatin (1 µM,) basal tone of umbilical arteries was not significantly different between statin-exposed group versus DMSO controls in eNOS-/- or WT mice (N=12 for eNOS-/- mice and N=9 for WT mice). The horizontal line denotes median and data expressed as median±IQR, Mann-Whitney U test.

209

5.4.9 Effect of pitavastatin (1 µM) on contraction responses of the umbilical artery in WT and eNOS-/- mice There was a trend towards attenuated contraction to U46619 in umbilical arteries from eNOS-/- mice compared to WT mice; (Figure 5.13 A; p=0.061). However, when data was expressed as contraction to U46619 dose as a percentage of the maximum KPSS contraction, there was no significant difference in umbilical artery contraction between eNOS-/- and WT mice (Figure 5.13 B; p=0.545).

Following a 2 hr incubation with pitavastatin at 1µM, there was no significant effect of U46619 on umbilical artery contraction in WT mice; Figure 5.13 C (DMSO vs pitavastatin; p=0.886). When U46619 contraction curves were expressed as maximum KPSS contraction, there was again no significant difference between pitavastatin-exposed umbilical arteries and DMSO control; Figure 5.13 D (DMSO vs pitavastatin; p=0.717).

Following a 2 hr incubation with pitavastatin at 1µM, there was no significant effect on umbilical artery contraction to U46619 in eNOS-/- mice; Figure 5.13 E (DMSO vs pitavastatin; p=0.677) and when U46619 contraction curves were expressed as maximum KPSS contraction there was again no significant difference between pitavastatin-exposed uterine arteries and comparable control; Figure 5.13 F (DMSO vs pitavastatin; p=0.945).

210

A B 12 WT (9) 140 WT (9) -/- 10 eNOS (12) -/- ) 120 eNOS (12)

a n

P 8 k

o 100

(

c n

i 80

o P

6 r i

n 60

n % e 4

o T

C 40 2 20

0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM C D 140 WT+DMSO (7) 12 WT+DMSO (7) WT+Pitavastatin (7) WT+Pitavastatin (7) 120

) 10

n a

o 100

P S t

k 8

( i

n t

o s 6

i 60

s %

n 4 C e 40

T 2 20

0 0 -10 -8 -6 -4 -10 -8 -6 -4 x x [U46619] x 10 M E [U46619] x 10 M F 12 eNOS+DMSO (8)

10 eNOS+Pitavastatin (8) 140 eNOS+DMSO (8) )

a 120 eNOS+Pitavastatin (8)

8 n

( o

100

S t

c o

6 i i

P 80

t s

n 60 n

e 4

% o T

C 40 2 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM

Figure 5.13: Effect of 2 hr pitavastatin incubation on contraction responses of umbilical arteries from WT and eNOS - /- mice.

Umbilical artery contraction to U46619 expressed as tension (kPa) (A) or as % KPSS (B) in WT (N=9, n=18) and eNOS-/- mice (N=12, n=24). Following a 2 hr incubation with pitavastatin at 1 µM, there was no significant effect on umbilical contraction expressed as kPa or %KPSS; in WT -/- mice (C-D) or in eNOS mice (E-F). U46619 dose response curves were compared using two- way ANOVA. Data are expressed as mean±SEM; N= number of litters n=umbilical arteries.

211

5.4.10 Effect of pitavastatin (1 µM) on relaxation responses of umbilical arteries in WT and eNOS-/- mice There was no significant difference in relaxation of umbilical arteries to SNP between eNOS- /- and WT mice; (Figure 5.14 A; p=0.881). Following a 2 hr incubation with pitavastatin at 1 µM, umbilical arteries from WT mice demonstrated a reduced relaxation to SNP following pitavastatin exposure (Figure 5.14 B (DMSO vs pitavastatin; p=0.005). There was no significant difference in umbilical artery relaxation to SNP between pitavastatin-exposed umbilical arteries and comparable control in eNOS-/- mice (Figure 5.14 C (DMSO vs

pitavastatin; p=0.867). A

9 1

6 120

6 4

100

0 8

C 80

o i

t 40

x a

l 20 WT (9) e

R -/- 0 eNOS (9) % -10 -8 -6 -4 [SNP] x 10xM B

9 1

6 120

6 4

U 100

C 80

o t

60 n

i t

a 40

x ** a l WT+DMSO (9) e 20

R WT+Pitavastatin (9)

% 0 -10 -8 -6 -4 x

C [SNP] x 10 M

1 6

6 120

4 U

100

0 8

C 80

o t

o i

t 40

a x

a eNOS+DMSO (9)

l 20 e

R eNOS+Pitavastatin (9) 0 % -10 -8 -6 -4 x [SNP] x 10 M Figure 5.14: Effect of 2 hr pitavastatin incubation on relaxation responses of umbilical arteries from WT and eNOS - /- mice.

Umbilical artery relaxation to SNP (A) in eNOS-/- (N=9, n=18) and WT mice (N=9, n=18). Following a 2 hr incubation with pitavastatin at 1 µM, there was a significant effect on umbilical artery

relaxation to SNP in WT mice (B) but not in eNOS-/- mice (C). SNP dose response curves were compared using two-way ANOVA with Sidak’s post hoc test, where applicable (Sidak was chosen as it is more power than the Bonferroni test and assumes comparisons are independent of one another). Data are expressed as mean±SEM; N= number of litters. n=umbilical arteries. ** P<0.01 212

Table 5.1 Summary of measurements of vascular reactivity with/without 2hr pitavastatin incubation in uterine and umbilical arteries from eNOS-/- and WT mice.

KPSS (120 mM high potassium salt solution); U46619 (2x10-6 M thromboxane-A2 mimetic); ACh; acetylcholine, SNP; sodium nitroprusside. Changes in the maximal response (Rmax) and sensitivity

(log EC50) to U46619 (kPa and %KPSS) in uterine and umbilical arteries untreated and incubated for 2 hr with pitavastatin and control DMSO shown. Residual constriction following relaxation to SNP and sensitivity (log EC50) are also shown. Data are shown as median [min - max]; A Mann-Whitney U test was performed to assess the effect of treatment on each group. ↔ arrow denote no change, ↑ arrow denotes an increase in a parameter compared to another relevant group and ↓ arrow denotes a decrease in a parameter compared to water control group.

Maximal contraction ACh relaxation SNP relaxation

Rmax Rmax logEC Residual logEC Residual logEC

U46619 constriction 50 constriction 50 (% to 50 (kPa) (kPa) (kPa) KPSS)

WT vs

eNOS-/- ↔ eNOS-/- ↔ eNOS-/- ↓ vs ↔ ↔ ↔ ↑ vs WT uterine WT artery

WT vs

eNOS-/- ↔ ↔ ↔ ↔ ↔

Umbilical artery

WT+ WT+

Pitavastatin ↔ ↔ Pit ↑ vs ↔ ↔ ↔ ↔ vs DMSO WT DMSO Uterine artery

WT+

Pitavastatin ↔ ↔ ↔ WT+ Pit ↓ ↔ vs DMSO vs WT DMSO Umbilical artery

213

eNOS-/- +pitavastati n vs DMSO ↔ ↔ ↔ ↔ ↔ ↔ ↔

uterine artery

eNOS-/- +pitavastati n vs DMSO ↔ ↔ ↔ ↔ ↔

umbilical artery

5.4.11 Results: In vivo exposure to pitavastatin in WT and eNOS-/- mice

5.4.12 Effect of maternal pitavastatin treatment on fluid intake and maternal bodyweight in WT and eNOS-/- mice When comparing pitavastatin treated mice to vehicle group, there was no significant difference in fluid intake over the treatment period (E10.5-E18.5) in eNOS-/- mice and WT mice (figure 5.15 A; p=0.211). There was a trend towards WT dams being heavier than eNOS-/- dams at E18.5 (p=0.057) but no significant difference in maternal weight gain from E0.5 to E18.5 between pitavastatin treated mice and the vehicle group in either eNOS-/- or WT mice (Figure 5.15 B; p=0.260). eNOS-/- mice had a significantly reduced hysterectomised weight at E18.5 compared with WT mice (p=0.0035). There was no significant difference in hysterectomised weight between pitavastatin treated mice relative to the vehicle group in either eNOS-/- mice or WT mice (Figure 5.15 C; p= 0.181).

214

)

A d

/ 11 e

u 10 o

m 9 /

l m

( 8

k 7 a

n I

d i

u l

F 0 n C C ti in M a M t C t C ta + s + s T a S a v v W ta O a i N it P e P + + T S W O N e B

) 50

( WT+CMC (6)

h WT+Pitavastatin (7) g i 40

e eNOS+CMC (6)

w eNOS+Pitavastatin (7)

y d

o 30

a n

r 20

a M 0 .5 .5 .5 0 0 8 1 1 C

d ** e

z 30 i

m 28

) c

g 26 e

(

r t

e 24

g i

y 22

w l

a n

e t a 0 M C in C in M t M t ta ta C s C s + a + a T v S v W a O a it N it P e P + + T S W O N e

Figure 5.15: Comparison of fluid intake and maternal bodyweight from pitavastatin-treated and vehicle groups in eNOS -/- and WT mice.

There was no significant difference in fluid intake (A); maternal bodyweight from E0.5-E18.5 (B) and maternal hysterectomized weight (C) following pitavastatin treatment in eNOS-/- and WT mice. N= 27 pregnant dams total split into the groups; horizontal line denotes median for graphs (A) and (C) while for (B), data are expressed as mean±SEM. two-way ANOVA performed. ** P<0.01 215

5.4.13 Effect of maternal pitavastatin treatment on maternal organ weight in WT and eNOS-/- mice There was no significant difference in maternal organ weights (Table 5.2) between pitavastatin-treated and vehicle control groups in eNOS-/- mice. The exception was the brain which was lighter in pitavastatin-treated eNOS-/- mice versus those on vehicle (2-Way ANOVA; p=0.042 when data expressed as % hysterectomized weight but as raw values there was no significant difference (p=0.894)). In WT mice, there was no significant difference in maternal organ weight between pitavastatin-treated and vehicle control groups (p>0.05). There was a trend towards livers from eNOS-/- mice being lighter than those from WT mice (p=0.056) and hearts from eNOS-/- mice were significantly lighter than those from WT mice (p=0.032).

Table 5.2 Summary table of maternal organ weights

Maternal organ weight data shown as raw values (g) and as a % of the maternal hysterectomised weight; for each dam, measurements were averaged and data expressed as median±range; 2-Way ANOVA. (CMC= carboxymethylcellulose vehicle used as a control). *P<0.05 pitavastatin versus CMC in eNOS-/- mice.

216

Maternal Organ WT+CMC WT+Pitavastatin eNOS+CMC eNOS+Pitavastatin weights

Brain

(g) 0.46 (0.33- 0.45 (0.42-0.46) 0.49 (0.47- 0.46 (0.40-0.48) 0.46) 0.50)

% of 1.61 (1.26- 1.68 (1.51-1.85) 1.91 (1.86- 1.76 (1.54-1.96)* hysterectomised 1.85) 2.30) weight

Liver

(g) 2.42 (2.03- 2.21 (2.08-2.40) 1.91 (1.69- 1.86 (1.58-2.22) 2.52) 2.09)

% of 8.61 (8.19- 8.41 ( 8.18-8.61) 8.15 (6.86- 8.01 (5.89-8.40) hysterectomised 8.92) 8.99) weight

Heart

(g) 0.21 (0.19- 0.15 (0.12-0.23) 0.13 (0.12- 0.14 (0.09-0.19) 0.23) 0.17)

% of 0.79 (0.70- 0.53 (0.46-0.91) 0.51 (0.45- 0.53 (0.40-0.75) hysterectomised 0.83) 0.70) weight

Kidney

(g) 0.15 (0.14- 0.15 (0.14-0.16) 0.12 0.13 (0.11-0.13) 0.15) (0.11-0.14)

% of 0.56 (0.52- 0.55 (0.51-0.59) 0.51 (0.46- 0.48 (0.47-0.53) hysterectomised 0.58) 0.55) weight

Spleen

(g) 0.09 0.10 (0.07-0.13) 0.07 (0.07- 0.09 (0.07-0.13) 0.09) (0.08-0.10)

% of 0.33 (0.30- 0.38 (0.31-0.44) 0.31 (0.28- 0.36 (0.28-0.50) hysterectomised 0.38) 0.36) weight

217

5.4.14 Effect of maternal pitavastatin treatment on fetal and placental weights in WT and eNOS knockout mice at embryonic day 18.5 eNOS-/- mice had reduced litter size versus WT mice (p= 0.0002) but treatment had no effect (figure 5.16 A: p=0.526). Litter size was significantly different in this dataset but was not significantly different in the ex vivo dataset, a reason for this could be due to the different suppliers. All C57Bl/6J mice used for the ex vivo studies were from Envigo whilst all C57Bl/6J mice used in the in vivo study were from Charles River. Across all WT mice (N=13), there was 1 resorption in the pitavastatin-treated group and 0 in the vehicle group; in eNOS-/- mice (N=14) there were 5 resorptions in the pitavastatin treated group relative to 2 resorptions in the vehicle group (p=0.099). Fetal weight was significantly reduced in eNOS-/- vs. WT mice (P<0.0001). Pitavastatin treatment had no significant effect on fetal weight in WT or eNOS-/- mice compared with corresponding vehicle groups (figure 5.16 B: p=0.306). A frequency distribution curve of fetal weights from vehicle and pitavastatin- treated pregnancies is shown in figure 5.16 C. The 5th percentile of WT control fetal weights was 0.957g with 35% of eNOS-/- vehicle fetuses falling below this 5th centile compared with 23% of eNOS-/- fetuses from pitavastatin-treated pregnancies (p=0.210, chi-squared test). There was no difference in placental weight between eNOS-/- and WT mice (p=0.218). Pitavastatin did not affect placental weight in WT or eNOS-/- mice compared with vehicle (figure 5.16 D: P=0.218). Fetal: placental weight ratio was significantly lower in eNOS-/- mice compared with WT mice (p=0.016) but pitavastatin treatment did not affect fetal: placental weight ratio in eNOS-/- or WT mice (Figure 5.16 E: (p=0.899)).

218

A B

*** 1.4 11 10 ) ****

g 1.3 (

e 8

h 1.2

z i

7 g

6 e

r 1.1 w

e 5

l t

i 4

a 1.0 t

L 3 e 2 F 0.9 0 0

n C in C n C ti C in M t M ti M a M t ta a C t C ta C C t + s + s + s + s T a S a T a S a v v v v W ta O a W ta O a i N it i N it P e P P e P + + + + T S T S W O W O N N e e

C 5 th C e n tile

1 5 W T + C M C (n = 5 8 )

n o

i W T + P IT (n = 5 7 )

t a l 1 0

u e N O S + C M C (n = 4 3 ) p

o e N O S + P IT (n = 4 4 )

f 5

0 0 .0 0 .5 1 .0 1 .5 2 .0 F e ta l w e ig h t (g )

D E o 0.14 i 16

t *

)

( t

0.12 14

g i

e 12 e

0.10 w

a 10

a t

0.08 n n

e 8

a l

0.06 p 6

t e 0.0 f 0 C in C in n n M t M t C i C i ta ta M t M t C s C s ta ta + a + a C s C s T v S v + a + a W a O a T v S v it N it W a O a P e P it N it + + P e P T S + + W O T S N W O e N e Figure 5.16: Fetal and placental weights from WT and eNOS - /- mice following pitavastatin treatment.

There was a significant reduction in litter size (A), fetal weight (B) and fetal: placental weight ratio (E) in eNOS-/- mice compared to WT mice. No difference was seen in fetal weight frequency distribution curve (C) or placental weight (D) in eNOS-/- and WT mice at embryonic day 18.5. Pregnant dams were allowed access to water with (N=7 WT+Pitavastatin and N=7 eNOS+Pitavastatin) or without (N=6 WT+CMC and N=6 N=eNOS+CMC) pitavastatin (6µg/ml). For all figures (except C), average (mean) value per litter is plotted. Horizontal line denotes median. In C, vertical black dashed line denotes the 5th centile of WT fetal weights. Statistical analyses for genotype and treatment were performed using two-way ANOVA with Sidak’s post hoc test (Sidak was chosen as it is more power than the Bonferroni test and assumes comparisons are independent of one another). *p<0.05, ***p<0.001 ****p<0.0001 219

5.4.15 Effect of maternal pitavastatin treatment on fetal biometric measurements in WT and eNOS-/- mice There were no effects of treatment or genotype-specific effects on fetal biometric measurements in WT and eNOS-/- mice (Table 5.3).

Table 5.3 Summary table comparing fetal biometric measurements.

Fetal biometric measurements for each litter. Measurements were averaged per litter and data expressed as median±range (fetuses: n=16 WT+CMC; n=19 WT+pitavastatin; n=11 eNOS+CMC; n=16 eNOS+pitavastatin); two-Way ANOVA, (CMC= carboxymethylcellulose vehicle used as a control).

WT+CMC WT+Pitavastatin eNOS+CMC eNOS+Pitavastatin Fetal Biometric Measurements Crown/Rump 28.7 28.3 28.0 27.3 length (mm) (28.0-29.0) (27.3-29.5) (27.5-28.5) (26.0-28.5)

Abdominal 26.4 26.0 25.5 25.5 Circumference (25.3-26.7) (23.7-26.7) (24.5-26.0) (25.0-27.0) (mm)

Head 24.7 25.3 25.3 25.0 Circumference (mm) (23.5-25.5) (23.0-25.5) (24.0-26.5) (23.3-26.5)

220

5.4.16 Effect of maternal pitavastatin treatment on uterine artery diameter from WT and eNOS-/- mice There was no significant difference in uterine artery diameter between WT and eNOS-/- mice in vehicle groups (Figure 5.17 A; p=0.429)), nor in pitavastatin versus vehicle treated groups in WT (Figure 5.17 B; p=0.548)) or eNOS-/- mice (Figure 5.17 C; p=0.469)).

400

M 300



e t

e 200

a i

D 100

0 - T -/ W S O N e 400

M WT

 300

e t

e 200

m a

i 100 D

C n M ti C ta s a v a it P

400 M

 300 r

e -/-

t eNOS

e 200

m a i 100 D 0

C n M ti C ta s a v a it P

Figure 5.17: Comparison of uterine artery diameter between pitavastatin-treated

and those on vehicle in eNOS - /- and WT mice at E18.5.

Uterine artery diameter from pitavastatin-treated mice was not different to that of the vehicle -/- group in eNOS or WT mice. Data presented as average diameter of uterine artery segments per mouse for each group. Horizontal line denotes median (median±IQR). Statistical analyses for genotype and treatment were performed using two-Way ANOVA. 221

5.4.17 Effect of maternal pitavastatin treatment on contraction responses of uterine artery There was no significant difference in uterine artery contraction to U46619 between vehicle groups of WT and eNOS-/- mice; (Figure 5.18 A; p=0.953)). Similarly, when data were expressed as contraction to U46619 dose as a % of the maximum KPSS contraction, no significant difference was observed (Figure 5.18 B; p=0.207)). No differences were observed in U46619 sensitivity (shown in table 5.4).

There was no significant effect on uterine artery contraction to U46619 between pitavastatin-treated and vehicle groups in WT mice; (Figure 5.18 C; p=0.999)). This trend did not change when U46619 contraction curves were expressed as maximum KPSS contraction; (Figure 5.18 D; p=0.889)). No differences were observed in U46619 sensitivity (shown in table 5.4).

In eNOS-/- mice, no significant difference was observed in uterine artery contraction to U46619 between pitavastatin-treated and vehicle groups; (Figure 5.18 E; p=0.973)) and when U46619 contraction curves were expressed as maximum KPSS contraction there was again no significant difference between pitavastatin-exposed uterine arteries and vehicle; (Figure 5.18 F; p=0.966)). No differences were observed in U46619 sensitivity (shown in table 5.4).

222

A B

25 WT+CMC (6) 160 WT+CMC (6)

140 eNOS + CMC (5)

) eNOS+CMC (5)

a 20

n 120

P i

( S

c 100

n t

K 80

i n

s 10

% 60

n C e 40 T 5 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x x [U46619] x 10 M [U46619] x 10 M C D

25 WT + CMC (6) 160 WT + CMC (6) 140

) WT + Pitavastatin (7)

20 WT+ Pitavastatin (7) a

n 120

(

15 c 100

t o

K 80

10 n

n 60

e C

T 5 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10 M [U46619] x 10xM E F

25 eNOS+ CMC (5) 160 eNOS +CMC (5) 140 eNOS + Pitavastatin (5)

) eNOS+Pitavastatin (5)

a 20

n 120

k t

( 100

S c 15

r n

t 80

i 10

n s

% 60

n e C 40

T 5 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM

Figure 5.18: Contraction responses of uterine arteries to U46619 in pitavastatin- treated and vehicle groups in WT and eNOS - /- mice.

There was no genotype-specific effects on uterine artery contraction to U46619 expressed as tension (kPa) (A) or as %KPSS (B) from vehicle groups in WT and eNOS-/- mice. Pitavastatin treatment did not alter contraction of uterine artery of either genotype when contraction was expressed as kPa or %KPSS; p>0.05 (C-F). U46619 dose response curves were compared using two -way ANOVA. Data are expressed as mean±SEM; (N)= number of litters.

223

5.4.18 Effect of maternal pitavastatin treatment on relaxation responses of uterine artery Blunted uterine artery relaxation to ACh was observed in eNOS-/- compared to WT mice (Figure 5.19 A; p<0.001). However, there was no significant difference in uterine artery relaxation to SNP between eNOS-/- and WT mice; (Figure 5.19 B; p=0.691).

Pitavastatin treatment did not enhance uterine artery relaxation to ACh relative to the vehicle control in WT (Figure 5.19 C; p=0.961). Similarly, uterine artery relaxation to SNP was not significantly different between pitavastatin and vehicle groups in WT mice; (Figure 5.19 D; p=0.988).

A significant increase in relaxation of uterine arteries to ACh following pitavastatin treatment was observed in eNOS-/- mice; (Figure 5.19 E; p=0.004)). However, uterine artery relaxation to SNP showed no significant difference between pitavastatin-exposed uterine arteries and comparable controls in eNOS-/- mice; (Figure 5.19 F; p=0.911)).

224

A B

9 9

1 120

6 4

4 100 U

U 100

8 8

C 80

t 60

i t

40 a 40 a

**** x

a l

20 e 20 WT+CMC (6) e

WT+CMC (6) R

eNOS+CMC (5) eNOS+CMC (5) % 0 % 0 -10 -8 -6 -4 -10 -8 -6 -4 x [SNP] x 10xM

C [ACh] x 10 M D

9 1

9 120 6

1 120

6 4

4 100

100 U

8 8

C 80

o t

t 60

o i

i 40 t

t 40

a a

l 20

l e

WT+CMC (6) e 20 WT+CMC (6)

0 WT+Pitavastatin (6) % WT+Pitavastatin (6) % 0 -10 -8 -6 -4 -10 -8 -6 -4 x [ACh] x 10 M [SNP] x 10xM

E F 9

1 120

6 1

120 6

6 4

6 100

4 U

100

0 80

C C

80 E

o 60

t 60

n n ** o

i 40

o t

i 40

x x

a 20 l

a 20 l e eNOS+CMC (5)

e eNOS+CMC (5)

R 0

0 eNOS+Pitavastatin (5) eNOS+Pitavastatin (5)

% % -10 -8 -6 -4 -10 -8 -6 -4 [ACh] x 10xM [SNP] x 10xM

Figure 5.19: Assessment of relaxation of uterine arteries from pitavastatin - treated and vehicle groups in WT and eNOS - /- mice.

There was a significant difference in uterine artery relaxation to Ach (p<0.05) (A) but no significant difference in uterine artery relaxation to SNP (B) in vehicle control groups from eNOS- /- and WT mice. There was no significant effect on uterine artery relaxation to ACh (C) or SNP (D) in WT mice following pitavastatin treatment. Pitavastatin-treated uterine arteries showed significantly enhanced ACh-induced relaxation relative to vehicle control (p<0.05) (E) but no significant effect on SNP-induced relaxation (F) in eNOS-/- mice. ACh and SNP dose response curves were compared using two-way ANOVA with Sidak’s post hoc test, where applicable (Sidak was chosen as it is more power than the Bonferroni test and assumes comparisons are independent of one another). Data are expressed as mean±SEM; N= number of litters. **P<0.01 ****P<0.0001.

225

5.4.19 Effect of maternal pitavastatin treatment on mesenteric artery diameter from WT and eNOS-/- mice There was no significant difference in the diameter of mesenteric arteries between vehicle groups in WT and eNOS-/- mice (Figure 5.20 A; p=0.222)). When the diameters of pitavastatin-treated mesenteric arteries were compared to those in the vehicle in WT mice, there was a trend towards the pitavastatin-treated group having a larger intraluminal diameter than the vehicle group (Figure 5.20 B; p=0.126)). No significant difference was seen between pitavastatin and vehicle groups in eNOS-/- mice (Figure 5.20 C; p=0.548)).

A 300

250 m

 200

e t

e 150 m

a 100

i D 50

0 /- T - W S O N B e

300 WT

250 m



r 200

e t

e 150 m

a 100

i D 50 0

C n M ti C ta s a v a C it P 300 eNOS-/- 250

m 

r 200

e t

e 150 m

a 100 i

D 50

C n M ti C ta s a v a it P Figure 5.20: Comparison of mesenteric artery diameter between pitavastatin - treated and vehicle groups in eNOS - /- and WT mice at E18.5

There was no significant difference in mesenteric artery diameter between vehicle groups between -/- WT and eNOS mice (A). The pitavastatin-treated group showed a trend towards increased intraluminal diameter in WT mice (B), however mesenteric artery diameter from pitavastatin- treated mice was not different to that in the vehicle in eNOS-/- mice (C). Data presented as average diameter of mesenteric arteries for each group. Horizontal line denotes median (median±IQR). Statistical analyses for genotype and treatment were performed using two-way ANOVA. 226

5.4.20 Effect of maternal pitavastatin treatment on contraction responses of maternal mesenteric artery There was no significant difference in contraction of mesenteric arteries to U46619 between vehicle groups of WT and eNOS-/- mice (Figure 5.21 A; p=0.998)). Similarly, when data was expressed as contraction to U46619 dose as a percentage of the maximum KPSS contraction, no significant difference was observed (Figure 5.21 B; p=0.991)). No differences were observed in U46619 sensitivity (shown in table 5.4).

There was no significant effect on mesenteric artery contraction to U46619 between pitavastatin-treated and vehicle groups in WT mice (Figure 5.21 C; p=0.946)). This trend did not change when U46619 contraction curves were expressed as maximum KPSS contraction (Figure 5.21 D; p=0.782). No differences were observed in U46619 sensitivity (shown in table 5.4).

In eNOS-/- mice, no significant difference was observed in mesenteric artery contraction to U46619 between pitavastatin-treated and vehicle groups (Figure 5.21 E; p=0.853)) and when U46619 contraction curves were expressed as maximum KPSS contraction, there was again no significant difference between pitavastatin-exposed uterine arteries and vehicle (Figure 5.21 F; p=0.984)). No differences were observed in U46619 sensitivity (shown in table 5.4).

227

A B

25 WT+CMC (5) 180 WT+CMC (5) eNOS+CMC (5) 160

) 20

a 140 eNOS+CMC (5)

k i

(

S 120 t

c n

i 100

s s

10 80

% e

o 60 T 5 C 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM

C D

25 WT+CMC (5) 180 WT+Pitavastatin (6) WT+CMC (5)

) 20 160

a WT+Pitavastatin (6) P

n 140

(

i S 15

t 120

o P

r 100

s s

10 80

% e

o 60 T 5 C 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10 M [U46619] x 10xM E F 25 eNOS+ CMC (5) 180 eNOS+CMC (5) eNOS+ Pitavastatin (5) 160 eNOS+ Pitavastatin (5)

20 )

a 140

i k

S 120 t

( 15

i n

P 100

s 80

10 n

n o

e 60

C T 5 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10 M [U46619] x 10xM Figure 5.21: Assessment of contraction of mesenteric arteries from pitavastatin - treated and vehicle groups in WT and eNOS - /- mice.

There was no significant difference in mesenteric artery contraction to U46619 expressed as tension (kPa) (A) or as % KPSS (B) from vehicle groups in WT and eNOS-/- mice. Pitavastatin treatment did not alter contraction in either genotype when contraction was expressed as kPa or % KPSS; p>0.05 (C-F). U46619 dose response curves were compared using 2-way ANOVA. Data are expressed as mean±SEM; N=number of litters. 228

5.4.21 Effect of maternal pitavastatin treatment on relaxation responses of maternal mesenteric artery Mesenteric artery relaxation to ACh was not different between vehicle groups from eNOS- /- and WT mice (Figure 5.22 A; p=0.949) nor was there any significant difference in mesenteric artery relaxation to SNP between eNOS-/- and WT mice (Figure 5.22 B; p=0.838).

Pitavastatin treatment did not enhance mesenteric artery relaxation to ACh relative to vehicle group in WT mice; (Figure 5.22 C; p=0.972). Similarly, mesenteric artery relaxation to SNP was not significantly different between pitavastatin-treated and vehicle groups in WT mice; (Figure 5.22 D; (p=0.525)).

A trend towards augmented mesenteric artery relaxation to ACh was observed in the pitavastatin-treated group but not vehicle groups in eNOS-/- mice; (Figure 5.22 E; p=0.063)). However, mesenteric artery relaxation to SNP showed no significant difference between pitavastatin-treated and vehicle groups in eNOS-/- mice; (Figure 5.22 F; p=0.933)).

229

A B

9 1

120 1 6

6 120

4 4

U 100

8 C

80 C

E 80

60 t

o t i 40 t

a 40

a l WT+CMC (5) a

20 l

e 20 WT+CMC (5) e

eNOS+CMC (5) R eNOS+CMC (5) % 0 0 -10 -8 -6 -4 % -10 -8 -6 -4 x

C [ACh] x 10 M D [SNP] x 10xM

9 9 120 1

1 120

6 4

4 100

U 100

0 8

8 80 C

C 80

o o

t 60

i i t 40

t 40

a a

a l

20 a l

e WT+CMC (5) 20 WT+CMC (5)

0 % WT+ Pitavastatin (6) 0 WT+Pitavastatin (6) -10 -8 -6 -4 % -10 -8 -6 -4 x x

E [ACh] x 10 M F [SNP] x 10 M

9 1

9 120 6

1 120

6 4

4 100

100 U

0 8

80 C 80

t 60 60

o i 40 t

t 40

x a

a 20 l l 20 e eNOS+CMC (5)

e eNOS+CMC (5)

0 eNOS+Pitavastatin (5) eNOS+Pitavastatin (5) % % 0 -10 -8 -6 -4 -10 -8 -6 -4 x [ACh] x 10xM [SNP] x 10 M

Figure 5.22: Assessment of relaxation of mesenteric arteries from pitavastatin - treated and vehicle control groups in WT and eNOS -/- mice.

There was no significant difference in relaxation of mesenteric arteries to ACh) (A) and SNP (B) in vehicle groups from eNOS-/- and WT mice. There was no significant effect on mesenteric relaxation to ACh (C) or SNP (D) in WT mice following pitavastatin treatment. Pitavastatin- treated mesenteric arteries showed a trend towards enhanced ACh-induced relaxation relative to vehicle control (E) but no significant effect on SNP-induced relaxation (F) in eNOS-/- mice. ACh and SNP dose response curves were compared using two-Way ANOVA with Sidak’s post hoc test, where applicable (Sidak was chosen as it is more power than the Bonferroni test and assumes comparisons are independent of one another). Data are expressed as mean±SEM; N =number of litters.

230

5.4.22 Effect of maternal pitavastatin treatment on diameter of umbilical arteries from WT and eNOS knockout mice at embryonic day 18.5 The umbilical artery intraluminal diameter was significantly smaller in eNOS-/- vehicle versus WT vehicle versus groups (Figure 5.23 A; p=0.0087)). There was no significant difference in umbilical artery intraluminal diameter between pitavastatin-treated and vehicle groups in WT mice (Figure 5.23 B; p=0.234)) or eNOS-/- mice (Figure 5.23 C; p=0.945)). A ** 800

700 m

 600 r

t 500 e

m 400 a i

300 D

0 /- T - W S O N B e WT 800

700 W m

 600

r e

t 500 e

m 400

a i

D 300

C n M ti C ta s a v C a it P

800 eNOS-/-

700 m

 600

r e

t 500

e m

a 400 i D 300

C n M ti C ta s a v a - it Figure 5.23: Umbilical artery diameter in pitavastatinP - treated and vehicle eNOS /- and WT mice at E18.5

There was a significant difference in umbilical artery diameter between vehicle groups from WT and eNOS-/- mice (A). Umbilical diameter from pitavastatin-treated mice was not different to vehicle in WT or eNOS-/- mice (B and C). Data presented as average diameter of umbilical arteries for each group. Horizontal line denotes median (median±IQR). Statistical analyses for genotype and treatment were performed using two-way ANOVA. **p<0.01 231

5.4.23 Effect of maternal pitavastatin treatment on contraction responses of umbilical artery There was no significant difference in umbilical artery contraction to U46619 between vehicle groups from WT and eNOS-/- mice (Figure 5.24 A; p=0.863). Similarly, when data were expressed as contraction to U46619 dose as a % of the maximum KPSS contraction, no significant difference was observed (Figure 5.24 B; p=0.879). No differences were observed in U46619 sensitivity (data shown in table 5.4).

There was no significant effect on umbilical artery contraction to U46619 between pitavastatin-treated and vehicle groups in WT mice (Figure 5.24 C; p=0.968), nor was there an effect when expressed as maximum KPSS contraction (Figure 5.24 D; p=0.909). No differences were observed in U46619 sensitivity (shown in table 5.4).

In eNOS-/- mice, no significant difference was observed in umbilical artery contraction to U46619 between pitavastatin-treated and vehicle groups; (Figure 5.24 E; p=0.994) nor was there an effect when U46619 contraction curves were expressed as maximum KPSS contraction (Figure 5.24 F; p=0.938). No differences were observed in U46619 sensitivity (shown in table 5.4).

232

A B

15 WT+CMC (6) WT+CMC (6) eNOS+CMC (6) 160

) 140 eNOS+CMC (6)

a P

10 n 120

(

S t

n 100

r i

t 80

n 5

n e

% 60

o T C 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM C D

15 WT+CMC (6) 160 WT+CMC (6)

WT+Pitavastatin (7) 140 WT+Pitavastatin (7)

) n

a 120

o P

10 i

k t

( 100

t o

K 80

n %

n 5 60

e C T 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10 M [U46619] x 10xM E F

15 eNOS+CMC (6) 160 eNOS+CMC (6)

eNOS+Pitavastatin (7) 140 eNOS+Pitavastatin (7)

) a

n 120

P 10

( t

100

o t

i 80

n n

5 % 60

e T C 40 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10 M [U46619] x 10xM

Figure 5.24: Assessment of contraction of umbilical arteries from pitavastatin - treated and vehicle control groups in WT and eNOS -/- mice.

Umbilical artery contraction to U46619 expressed as tension (kPa) (A) or as % KPSS (B) from vehicle groups in WT and eNOS-/- mice. Pitavastatin treatment did not alter contraction in either genotype when contraction was expressed as kPa or % KPSS; (p>0.05) (C-F). U46619 dose response curves were compared using two-way ANOVA. Data are expressed as mean±SEM; N=number of litters.

233

5.4.24 Effect of maternal pitavastatin treatment on relaxation responses of umbilical artery Umbilical artery relaxation to SNP was not significantly different between vehicle groups from eNOS-/- and WT mice (Figure 5.25 A; p=0.988)). Pitavastatin treatment did not affect umbilical artery relaxation to SNP in WT mice (Figure 5.25 B; p=0.711). Similarly umbilical artery relaxation to SNP showed no significant difference between pitavastatin-treated and vehicle control groups in eNOS-/- mice; (Figure 5.25 C; p=0.920).

A 9

6 120

4 U

100 0 8

C 80

o t

o i

t 40

x a l 20 eNOS+CMC (5)

e R

WT+CMC (5) 0 % -10 -8 -6 -4 B [SNP] x 10xM

1 120

6 6 4 100

0 8

C 80 E

o t

o i

t 40

a l

e 20 WT+CMC (5) R WT+Pitavastatin (7) % 0 -10 -8 -6 -4 C [SNP] x 10xM

1 120

6 6

4 100

0 8

C 80

o t

n o

t 40

a l

e 20 eNOS+CMC (5) R eNOS+Pitavastatin (7) % 0 -10 -8 -6 -4 [SNP] x 10xM

Figure 5.25: Assessment of relaxation of umbilical arteries from pitavastatin - treated and vehicle control groups in WT and eNOS-/- mice.

There was no significant difference in umbilical artery relaxation to SNP (A) in vehicle groups from eNOS-/- and WT mice. There was no significant effect on umbilical relaxation to SNP (B) in

WT or eNOS-/- mice (C) following pitavastatin treatment. SNP dose response curves were compared using two-way ANOVA. Data are expressed as mean±SEM; N=litters. 234

Table 5.4 Summary of vascular reactivity measurements comparing pitavastatin-treated and vehicle groups in uterine, mesenteric and umbilical arteries from eNOS-/- and WT mice.

KPSS (120 mM high potassium salt solution); U46619 (2x10-6 M thromboxane-A2 mimetic); ACh; acetylcholine; SNP; sodium nitroprusside; CMC; carboxymethylcellulose. Changes in the maximal response (Rmax) and sensitivity (log EC50) to U46619 (kPa and %KPSS. Residual constriction following relaxation to ACh and SNP and sensitivity (log EC50) are also shown. ↔ arrow denote no change, ↑ arrow denotes an increase in a parameter compared to control group.

Maximal contraction ACh relaxation SNP relaxation

Rmax Rmax logEC50 % logEC50 % logEC50 U46619 relaxatio relaxation (kPa) (% to n KPSS)

Uterine artery:

Pitavastatin vs CMC

WT ↔ ↔ ↔ ↔ ↔ ↔ ↔

eNOS ↔ ↔ ↔ PIT ↑ ↔ ↔ ↔

vs CMC

Mesenteric artery:

Pitavastatin vs CMC

WT ↔ ↔ ↔ ↔ ↔ ↔ ↔

eNOS ↔ ↔ ↔ ↔ ↔ ↔ ↔

Umbilical artery:

Pitavastatin vs CMC

WT ↔ ↔ ↔ ↔ ↔

eNOS ↔ ↔ ↔ ↔ ↔

235

5.5 Discussion

5.5.1 Acute Pitavastatin Incubation Studies in eNOS-/- and WT Mice (ex vivo) This study utilised the eNOS-/- mouse, which demonstrates chronic hypertension and fetal growth restriction (FGR), to determine the short-term and long-term effects of pitavastatin treatment on vascular function. Short-term pitavastatin treatment (2 hr) did not reduce contraction or increase endothelial-dependent or independent relaxation in uterine or umbilical arteries in WT or eNOS-/- mice, although pitavastatin treatment increased basal tone and U46619-sensitivity in uterine arteries from WT mice and blunted SNP-mediated relaxation in umbilical arteries from WT mice.

5.5.2 In vivo Pitavastatin Treatment Studies in eNOS-/- and WT mice Pitavastatin (6 µg/ml) treatment from E10.5 until E18.5 did not affect litter size, fetal weight, placental weight or fetal: placental weight ratio in eNOS-/- or WT mice. There was no effect of pitavastatin treatment on uterine artery U46619-mediated contraction in eNOS-/- or WT mice. Pitavastatin enhanced ACh-mediated relaxation, but not relaxation to SNP, in uterine arteries from eNOS-/- mice only. In mesenteric arteries, there was no effect of pitavastatin on U46619-mediated contraction in eNOS-/- or WT mice. However, treatment with pitavastatin showed a trend towards increasing ACh-mediated relaxation, but no effect of SNP, in eNOS-/- mice. Pitavastatin did not affect vascular reactivity of umbilical arteries in eNOS-/- or WT mice.

5.5.3 Confirmation of the Pregnancy Phenotype Associated with the eNOS- /- Mouse In the present study, we confirmed the previously reported pregnancy phenotype of eNOS- /- mice (Duncan et al., 2005; Hefler et al., 2001a; Kusinski et al., 2012; Pallares and Gonzalez- Bulnes, 2010; Poudel et al., 2013; Stanley et al., 2012a; Van Der Heijden et al., 2005) including reduced fetal weight and reduced fetal: placental weight ratio, indicative of reduced placental transport efficiency, at E18.5 versus WT controls. Whilst we did not measure blood pressure in the present study, multiple studies have demonstrated a reproducible hypertensive phenotype in eNOS knockout mice both prior to, and during, pregnancy (Cabrales et al., 2005; Hefler et al., 2001b; Huang et al., 1995; Kusinski et al., 2012; Ortiz and Garvin, 2003; Shesely et al., 1996; Stanley et al., 2012a). There was no

236

difference in litter size or placental weight in eNOS-/- mice confirming observations by other groups (Stanley et al., 2012; Poudel et al., 2013) although some groups have reported a reduction in litter size in eNOS-/- mice compared to WT mice (Sazonova et al., 2012; Van Vliet and Chafe, 2007). In the present study, a reduction in maternal weight was observed at E18.5, which fits some (Kulandavelu et al., 2012; Van der Heijden et al., 2005) but not all (Poudel et al., 2013 and Finn-Sell et al., 2018) observations. Previous studies have shown that eNOS-/- fetuses show asymmetric growth with reductions in their abdominal circumference but no difference in head circumference, indicative of head sparing relative to the WT (Hefler et al., 2001a; Pallares and Gonzalez-Bulnes, 2010). In our study, there was no reduction in abdominal or head circumference or crown-rump length, but this observation could have been limited by a small sample size. Studies were powered on the basis of vascular function rather than effects on fetal weight/measurements. Our data is consistent with other studies that have shown head sparing in eNOS-/- fetuses (Hefler et al., 2001a; Pallares and Gonzalez-Bulnes, 2010; Stanley et al., 2012a).

This study also confirmed previous findings regarding altered vascular reactivity in uteroplacental and fetoplacental circulations following eNOS gene ablation. Increased vasoconstriction to U46619 (expressed as %KPSS only) and reduced relaxation to the endothelium-dependent vasodilator ACh was seen in uterine arteries in the current study, as described previously (Kusinski et al., 2012). There appeared to be a difference in SNP- mediated relaxation between eNOS-/- and WT mice however the sample size for this particular experiment was too small for this difference to be detected statistically. This outcome is in contrast to others who have shown no difference between eNOS-/- and WT mice (Poudel et al., 2013) or an increased uterine artery response to SNP in eNOS-/- mice (Finn-Sell et al., 2018). Both Poudel et al. and Finn-Sell et al. used phenylephrine to preconstrict their uterine arteries while the agonist U46619 was used in the present mouse study. There appears to be conflicting evidence regarding the NO-dependent intracellular signal transduction pathway in pregnant eNOS-/- mouse compared to WT mice. Kusinski et al. showed that umbilical arteries demonstrated no significant difference in contraction relative to WT mice and vessels responded normally to donated NO following SNP treatment. This outcome was in support of the findings in the present study, with the caveat that different agonists were used; phenylephrine in the study by Kusinski et al., (Kusinski et al., 2012) compared to U46619 in the current study. The agonist U46619 was

237

used as opposed to phenylephrine to enable consistency between mouse and human studies, allowing for comparison in vascular function.

5.5.4 Rationale Underpinning Myography Studies of Mouse Uterine and Umbilical Artery Function Uterine arteries are resistance vessels and thus regulate peripheral vascular resistance and thereby arterial blood pressure. They also provide approximately 80% of total uteroplacental blood flow (Osol and Mandala, 2009). In pregnancy, changes in uterine artery vascular reactivity result in enhanced vasodilation and reduced vasoconstriction following a considerable remodelling of the uterine vascular bed (Mandala and Osol, 2012; Rosenfeld, 2001). In rat uterine arteries, enhanced NO-mediated relaxation occurs as an adaptation to pregnancy (Ni et al., 1997). In humans, adverse outcomes in pregnancy are associated with increased resistance within the uterine arteries as measured by Doppler ultrasound velocimetry (Bower et al., 1998). Thus, performing myography studies on the uterine artery offers important insight into the vascular performance of the uteroplacental circulation. To assess vascular function in the fetoplacental vasculature, the umbilical artery was chosen; this vessel carries blood from the fetus to the placenta and thus is important for gaseous and nutrient exchange (Piers and Lee, 2019). In humans, Doppler ultrasound of the umbilical artery provides information about downstream placental vascular resistance; an abnormal umbilical artery Doppler may be indicative of placental dysfunction as observed in severe cases of FGR and may indicate fetal compromise (Niromanesh et al., 2017). Furthermore, assessing umbilical artery reactivity in the current study is important as it may offer insight into effects of pitavastatin in the fetoplacental circulation, given its potential to cross the placenta.

5.5.5 Vascular Reactivity of Uterine and Umbilical Arteries Following 2 hr Pitavastatin Exposure in WT and eNOS-/- Mice In the present study, it was hypothesized that pitavastatin would improve uterine artery vascular function in eNOS-/- mice whilst not having detrimental effects on umbilical arteries. There was an increased uterine artery contractile response to U46619 in eNOS-/- mice compared to WT mice, when expressed as %KPSS, suggesting that either vascular smooth muscle mass or activity was increased in eNOS-/- mice relative to WT mice. Following a 2 hr incubation with pitavastatin, there was no significant effect on U46619-mediated

238

contraction in uterine or umbilical arteries. Similarly, short-term exposure to pitavastatin did not enhance endothelium-dependent ACh relaxation in uterine arteries or endothelium-independent SNP relaxation in uterine or umbilical arteries.

With the hypothesis in mind, it was predicted that pitavastatin would improve vascular reactivity following a 2 hr incubation. In WT mice, it was expected that the eNOS enzyme would be the main mechanism for modulating vascular tone while in eNOS-/- mice this would be through an eNOS-independent mechanism or no effect of pitavastatin would be observed. NO production, through the actions of eNOS, has been suggested to play a key role in modulating vascular tone. Phosphorylation of eNOS, and especially at Ser1177, is believed to enhance its activity (Harris et al., 2004; Sessa, 2004; Wang et al., 2008; Wu et al., 2008). Rossini et al. demonstrated that statins, and in particular simvastatin, was able to promote eNOS activity through the activation of AMPK, which phosphorylates eNOS at Ser1177 (Rossini et al., 2011). Furthermore, statins are believed to also inhibit the Rho/ROCK pathway to activate P13k/Akt pathway thus phosphorylating eNOS (Wolfrum et al., 2004; Van der Heijden., 2008). Furthermore, statins have been shown to decrease caveolin-1 expression, thus promoting eNOS activity and NO production (Pelat et al., 2003). In addition, eNOS-independent effects of statins have also been demonstrated; Mraiche et al. showed through rat in vitro studies that pre-treatment with the Kv channel inhibitor, 4- aminopyridine had the ability to lessen the acute dilatory effects of simvastatin in aortic vessels that had been preconstricted to endothelin-1 and noradrenaline (Mraiche et al., 2005). Pitavastatin specifically has shown protective effects on endothelial function. In vitro experiments in human umbilical vein endothelial cell HUVECS have shown that pitavastatin and antimalarial drugs are activators of the vasoprotective molecule ERK5, which leads to inhibition of endothelial inflammation and dysfunction (Le et al., 2014); this was achieved by increasing eNOS expression. Furthermore, in patients with cardiovascular risks such as hypercholesterolemia and type 2 diabetes, pitavastatin was able to increase plasma VEGF, increase the phosphorylation of eNOS in endothelial progenitor cells (EPCs) and circulating levels of EPCs (Lin et al., 2014), an effect which would be possible in WT mice but not in eNOS-/- mice. This pitavastatin-mediated improvement in endothelial function was supported by Liu et al., who demonstrated that pitavastatin-filled nanoparticles were able to stimulate EPC proliferation by activating the P13K signalling pathway and promoting recovery of injured carotid arteries in the rat (Liu et al., 2017). The role of EPCs is very important, as they are critical in vascular homeostasis and endothelial repair, the increase 239

and mobilization of EPCs would be beneficial in the eNOS-/- mice where endothelial dysfunction is evident.

The current literature has shown effects on vascular function following acute treatment of vessels with statins. Rossini et al. conducted an acute treatment study similar to that of the present mouse study. They demonstrated that, following a 2 hr exposure with simvastatin to mesenteric arteries from non-pregnant male rats, there was an attenuation in U46619-mediated contraction (Rossini et al., 2011). Whilst the study by Rossini et al. used a similar methodology to that of the present study, it is important to note that the physiological state, sex and the species were all different. There are sex-specific differences in vascular function and the vascular response of vessels during pregnancy is altered compared to a non-pregnant state, where vessels may show reduced or increased sensitivity to certain agonists. Moreover, whilst mice and rats are rodents, their vascular responses may show subtle differences.

Mraiche et al. showed through an ex vivo study that, following the administration of cumulative concentrations of simvastatin (0.1–10 μM), relaxation of rat aortic vessels was promoted in a concentration-dependent manner following pre-constriction with endothelin-1, noradrenaline and potassium chloride (KCL) (Mraiche et al., 2005). Finally, Mukai et al. showed enhanced endothelium-dependent ACh relaxation of rat aortic vessels following a 2 hr incubation with 1 µM cerivastatin. In this study, another lipophilic statin, fluvastatin (1–300 µM), was able to relax vascular smooth muscle in the aorta and mesenteric arteries in a dose-dependent manner; vessels were pre-contracted with Prostaglandin F2α (Mukai et al., 2003). The study by Mukai et al. and Mraiche et al. used vascular beds from the systemic circulation while the present mouse study used vessels from the uteroplacental and fetoplacental circulation, the vascular responses may differ in response to vasoactive agents because of their functional roles in the body. Absi et al. have shown in healthy arteries and arteries from rats with pulmonary arterial hypertension (non- pregnant), that a 1 hr incubation with 5 µM simvastatin was able to attenuate phenylephrine and U46619 mediated constriction. However, in rats with pulmonary arterial hypertension, simvastatin promoted relaxation of pulmonary arteries to ACh but not SNP (preconstricted to U46619), suggesting an involvement of the endothelium. They proposed that simvastatin facilitated effects on vascular tone by inhibiting ROCK and reducing Ca2+ sensitisation, inhibiting Ca2+ entry through L-type Ca2+ channels and

240

disrupting GPCR signalling by preventing prenylation of Gγ subunits (Absi et al., 2019). The study by Absi et al. and other published literature have used cumulative dose response curves or acute incubations with higher statin concentrations, while the current study used a single dose of 1 µM, which could have contributed towards differences in results. Many of the studies that have shown positive acute effects ex-vivo have assessed lipophilic drugs; pitavastatin, used in the current study, is moderately lipophilic. However, this is not to say that pitavastatin does not have an effect on vascular tone but it may require more time for pleotropic effects to be seen. Lipophilic statins have the ability to easily diffuse across the plasma membrane of cells in both the liver and extrahepatic tissue, subsequently leading to more potent pleiotropic effects in peripheral tissues (Bełtowski and Jamroz-Wiśniewska, 2012).This observation could insinuate that the metabolic state of statins provide differential effects on vascular tone.

This is the first study to assess acute effects of statin treatment on umbilical artery function in mice (ex vivo) and hence direct comparisons are difficult. No pitavastatin- mediated effects were observed in eNOS mice-/-. An unexpected finding following pitavastatin exposure was increased U46619 sensitivity in uterine arteries as well as blunted SNP-mediated relaxation of umbilical arteries in WT mice. This could be suggestive of a harmful effect of pitavastatin, and could be attributed to the suprapharmacological dose used to treat these vessels. Alternatively, this effect could be due to the possible contribution statins may have towards oxidative stress via increased NO which can combine with superoxide to produce peroxynitrite. Grygiel-Gorniak et al. postulated that excess NO production can promote peroxynitrite formation, which leads to vascular damage and endothelial apoptosis (Grygiel-Gorniak et al 2014). The fact that this observation was only observed in WT mice suggests that the presence of eNOS is an important mechanism in these pitavastatin-mediated effects. Furthermore, Finn-Sell et al. showed that umbilical arteries from eNOS-/- mice treated with pomegranate juice (rich in polyphenols such as ellagitannin and punicalagin) in utero showed increased U46619 sensitivity (Finn-Sell et al., 2018). The polyphenol epigallocatechin gallate has been shown to increase prostacyclin production in endothelial cells in culture (Mizugaki et al., 2000). Prostacyclin production, although known to be a vasodilator, has been associated with constriction in umbilical vessels (Abramovich et al., 1984). Mevastatin, lovastatin and simvastatin have also been associated with arachidonic acid release and prostacyclin production in rat liver cells (Levine, 2003). However, the result observed in WT umbilical arteries should be interpreted 241

with caution because, in vivo, the concentration of pitavastatin in the umbilical vessels is likely to be considerably lower, due partly to the action of placental efflux transporters.

Limitations of this study are that only one dose of pitavastatin was utilised, a dose response curve involving higher concentrations may have provided additional information. Additionally, whilst the use of DMSO was an appropriate control, a further control (without DMSO exposure) would have provided further information on the effects of DMSO alone.

5.5.6 Effect of Maternal Pitavastatin Treatment in vivo on Maternal and fetal Measurements in eNOS-/- and WT Mice Pitavastatin, administered via drinking water, was well tolerated as evidenced by normal maternal fluid intake in WT and eNOS-/- mice in all groups. Pitavastatin was administered orally to try to mimic the route of choice in humans. Other groups have administered pitavastatin intramuscularly (Kubo et al., 2009), by adding it to standard chow diet (Kobayashi et al., 2009; Shinozaki et al., 2007), via oral gavage with 0.5% carboxymethylcellulose (Almeida and Ozaki, 2014; Kikuchi et al., 2011; Kitahara et al., 2010; Koladiya et al., 2008) or via drinking water (Yagi et al., 2010). This study used 0.05% carboxymethylcellulose (CMC) as the vehicle to dissolve pitavastatin prior to addition to drinking water. This was in contrast to the use of the vehicle DMSO, which was used in the short-term study discussed in section 5.5.5. Initially in this study, when a negligible amount of DMSO was dissolved in drinking water for the mice, it was not well tolerated, most probably due to the strong odour and/or taste, and so CMC was chosen instead for these studies. CMC has not been used as a vehicle control for vascular function studies in humans/mice and so could have been a confounding factor however there is evidence that DMSO has effects on vascular function. In an attempt to address this, the in vivo study involved ‘water only’ mice as a negative control and no significant difference in vascular function was seen compared to CMC mice (data in chapter 8: appendix). However, there was no ex vivo myography data gathered using CMC vehicle. Maternal food intake was not measured for either pitavastatin treated or vehicle control mice as maternal weight gain between E10.5 and E18.5 was not different between pitavastatin and vehicle treated groups in WT or eNOS-/- but may provide useful additional information in future studies.

At E18.5, eNOS-/- dams showed a trend towards being lighter than WT dams, most likely due to a trend towards reduced maternal liver weight (p=0.056) and reduced litter size and fetal weight, in eNOS-/- mice. To correct for this, hysterectomised weights were 242

recorded which excluded the uterine horn and the conceptuses within. When hysterectomised weights were compared for each genotype, a significantly reduced maternal weight remained (p=0.0035). This was most likely explained by the trend towards reduced liver mass seen in eNOS-/- mice compared to WT mice, as all other maternal organ weights were similar. Maternal brain weight in pitavastatin treated eNOS-/- mice was decreased relative to vehicle controls, when expressed as a % of hysterectomised weight only. There was no difference seen in WT mice. The mechanisms underpinning this are currently unknown but whilst this is a potential concern, data must be interpreted cautiously as raw values of brain weight showed no difference suggesting that the effects of altered maternal hysterectomised weight may be a contributor to this difference.

eNOS-/- mice, independent of treatment group, had significantly reduced litter sizes compared to WT mice. This was not observed in the animals that were included in our acute study; reasons for this could be differences in sample size since we powered studies to assess effects of vascular reactivity rather than differences in litter sizes or fetal weight. The WT controls in our in vivo study had relatively large litter sizes, however the litter sizes of eNOS-/- mice were consistent with observations from our group (Finn-Sell et al., 2018; Kusinski et al., 2012; Renshall et al., 2018). In eNOS-/- mice, the vehicle group showed reduced fetal weight, fetal: placental weight and litter size and the pitavastatin-treated group did not normalise this to WT levels or increase levels significantly; placental weight was unaltered compared to WT. In WT mice, pitavastatin treatment did not increase or decrease fetal weight, placental weight, fetal: placental weight or litter size compared to vehicle group.

Whilst pitavastatin has not previously been used in pregnancy, other statins, and particularly pravastatin, have been investigated in terms of safety and efficacy in a PE-like model. Following pravastatin administration (5 mg·kg−1·d−1; E8–E18 via intragastric administration) to mice administered with L-NAME (NOS-inhibitor) or lipopolysaccharide (to mimic PE), fetal and placental weights and fetal:placental weight were unaffected (Huai et al., 2018). However, Wyrwoll et al. developed a glucocorticoid excess (Hsd11b2-/-) mouse model that exhibited FGR of a similar degree to the eNOS-/- mouse. When pravastatin (20 μg/kg via IP) was administered from E6.5-E17.5, an increase in placental vascular endothelial growth factor A was observed along with increased placental vascularisation and normalisation of fetal weights in Hsd11b2-/- fetuses (Wyrwoll et al., 2016). Similarly, in

243

a sFlt-1 mouse model of PE, Kumasawa et al. demonstrated that pravastatin (1 mg/ml via intraperitoneal injection (IP), E7.5 - E18.5) reduced blood pressure and proteinuria, increased fetal growth, reduced maternal sFlt-1 levels, increased maternal PlGF, and enhanced endothelial cell proliferation (Kumasawa et al., 2011). In the current study, when pitavastatin (6 µg/ml) was added to the drinking water of the dam for eight days, no differences in litter size, fetal weight, placental weight or fetal:placental weight were seen compared to the vehicle control group in eNOS-/- or WT mice was observed. This suggests that, in mice at least, chronic pitavastatin treatment at this dose from mid-pregnancy does not have negative effects on fetal or placental growth. Neither did pitavastatin appear to affect fetal morphology, which appeared grossly normal. In mice, spiral artery remodelling occurs between E7.5-E10.5 (Woods et al., 2018), with the mature haemochorial placenta fully established by E14.5 and embryogenesis complete by E16.0 (Cross et al., 1994). Thus, pitavastatin administration in this study commenced following the process of spiral artery remodelling in the mouse but whilst placental growth and development was continuing.

These results are important in light of the current FDA statement suggesting that statin use in pregnancy could be harmful due to their classification as “category X,” thus are contraindicated in women who are or may become pregnant. This is due to previous studies in animals and humans demonstrating fetal abnormalities and the risk of the use of the drug in pregnant women outweighing any possible benefit, hence there is limited data about its effects on the human placenta and fetus available (Zarek and Koren, 2014). However, the present mouse study has proven that the statin pitavastatin administered via drinking water at a dose of 6µg/ml for 8 days in vivo does not appear to affect gross fetal development. In the present mouse study, pitavastatin was administered during embryogenesis, prior to the completion of organogenesis, which ends around E14.5. Mouse gestation ranges between 19 and 21 days while in humans it is 37-42 weeks; the time point of pitavastatin administration was therefore equivalent to the beginning of the second trimester to the third trimester of pregnancy in humans. However, direct comparisons between translating trimesters between humans and rodents should be treated with caution, particularly as for some organs such as the kidney, organogenesis continues postnatally in rodents.

Previous studies that have observed a difference in fetal or placental weight have administered pravastatin for ~10-11 days versus 8 days in the current study. 8 days for

244

pitavastatin administration was chosen as a balance between minimising exposure during critical windows of embryonic development but providing maximal opportunity for differences in fetal weight to be observed. It is worth reiterating however that vascular function was the primary outcome of these studies and power calculations were based upon these studies rather than the ability to detect differences in fetal weight. In order to detect a biologically relevant change for example a % change in fetal weight, a sample size of N=14 litters would be required, based upon a 5% significance level and 80% statistical power, with the unit of analysis being number of litters (Renshall et al., 2018). In addition, the mode of administration could be a factor for differences in fetal weight, as intraperitoneal and intragastric treatment are examples of parenteral administration while oral treatment is an example of enteral administration. Parenteral administration involves fewer biologic barriers as a result of the drug not being absorbed via the GI tract and thus drug bioavailability is promoted, which could lead to more direct and potent effects compared to that of the oral administration method used in this study for pitavastatin (Bardal et al., 2011). Plasma samples of pitavastatin-treated and vehicle groups in eNOS-/- and WT mice were stored in this study and so further investigations to determine circulating levels of pitavastatin would be beneficial in the future.

Pitavastatin’s effects on vascular reactivity following treatment in utero has not been investigated previously. However, pravastatin has been studied in PE and in a number of PE-like models and so these studies will be utilised for comparison. Singh et al. showed in their C1q knockout PE mice that norepinephrine constricted aortic vessels demonstrated normal ACh-induced relaxation compared with WT mice post-pravastatin treatment (Singh et al., 2011). This supports the notion that statins can stimulate relaxation of vessels, either by NO dependent or independent mechanisms. In another study, the ACh-mediated vasorelaxant responses of carotid arteries following pravastatin treatment (5 mg/kg/d; E9- E18) were significantly higher in WT mice (Costantine et al., 2010). Pitavastatin has shown protection against vascular injury in eNOS-/- mice modelling atherogenesis and limb ischemia by activating ERK1/2 and ERK5 in the aorta, as well as activating AMPK in the aorta and skeletal muscle (Mitsuhashi et al., 2018). Improved vasorelaxation following pitavastatin treatment has been shown in humans; Arao et al. showed, in a study with patients with coronary artery disease, 2 mg pitavastatin given for 6 months improved postprandial endothelium-dependent vasodilation in forearm resistance vessels (Arao et al., 2009). 245

In human pregnancy, there are examples of improved maternal outcomes following pravastatin treatment in pregnancy. In patients with antiphospholipid syndrome, thrombosis and PE, following treatment with pravastatin (20 mg), enoxaparin (40 mg) and aspirin (100 mg), BP and proteinuria improved and normalization of previous pathological uterine artery blood flow velocity (Doppler) of the uterine arteries was observed (Lefkou et al., 2016). Additionally, a pilot study in 4 women with pre-term PE (24+5 to 29+4 weeks), demonstrated that clinical signs of PE were improved following administration of 40 mg pravastatin (from the time of PE presentation - 23+0 to 30+0 weeks - till delivery); urine protein/creatinine ratio (UPCR) was stabilized or reduced in three out of the four women and no further change in antihypertensive medication was required (Brownfoot et al., 2015). In contrast to the studies previously mentioned, the StAmP trial involved administration of 40 mg pravastatin to women with early-onset PE from 24-31 weeks until delivery; however, pravastatin failed to slow down PE progression and to prolong gestation (data unpublished).

In the present mouse study, enhanced endothelium-dependent ACh relaxation was observed in both the uterine and mesenteric arteries from pitavastatin-treated eNOS-/- mice compared to the vehicle control group but there was no effect in WT mice. There was no significant effect on endothelial-independent relaxatory responses of uterine or mesenteric arteries to SNP. The dilatory response to SNP seen in the maternal uterine and mesenteric arteries corroborates with what other groups have seen in eNOS-/- mice. SNP contributes to cGMP production in the smooth muscle promoting maximal relaxation (Faraci et al., 1998; Huang et al., 1995). However, whilst uterine artery relaxation to ACh was seen in pitavastatin-treated eNOS-/- mice, this effect did not align with changes in fetal outcome in terms of increased fetal/placental weight. Clinically this change may not be biologically relevant enough to prevent incidences of FGR and preterm birth in humans. Interestingly, this increase in endothelium-dependent relaxation in uterine arteries was not seen in the acute study following a 2 hr treatment with pitavastatin. This implies that the length of treatment may have played a role in vascular effects observed.

Our data are in alignment with vascular studies investigating effects of pitavastatin but also pravastatin on endothelial dysfunction. Studies have shown that in eNOS-/- mice, compensatory prostaglandin and EDHF-mediated relaxation can occur (Chataigneau et al., 1999; Waldron et al., 1999). Furthermore it has been shown in carotid, coronary, and

246

skeletal muscle arteries in non-pregnant eNOS-/- mice that compensatory mechanisms mediated by neuronal NOS, EDHF, and/or prostaglandins are observed in these mice (Godecke and Schrader, 2000). In the current study, uterine artery relaxation to acetylcholine was observed despite the absence of the eNOS enzyme, indicating that acetylcholine may be working through an alternative pathway. Acetylcholine exerts its effects via NO and COX pathways; in the eNOS-/- mice pitavastatin could be working through the COX arm of acetylcholine, in particular COX-2, to stimulate the release of prostacyclin. Interestingly, the enzyme COX-2 is involved in producing the prostaglandin prostacyclin but is only expressed in damaged or inflamed endothelium (Needleman and Isakson, 1997). Alternatively, pitavastatin could also be interacting with the NO-independent and non- prostaglandin EDHF to promote vasodilation in these vessels. Bradykinin and acetylcholine have been shown to stimulate the release of EDHF from endothelial cells (Cohen and Vanhoutte, 1995). Scotland et al. showed in mesenteric arteries from male and non- pregnant female eNOS−/− and COX-1−/− mice (8-12 weeks), that following a preconstriction to U46619 (EC50 –EC80), a single application of ACh produced ~60% relaxation. Moreover these relaxatory responses were unchanged even after NOS/COX blockade, indicating the role of EDHF, as the presence of raised [K+], ChTx/apamin, or Ba2+/ouabain profoundly suppressed these relaxatory responses (Scotland et al., 2005). As COX-2 is mainly expressed in damaged endothelium, it may explain why no pitavastatin-specific vasodilatory effect was observed in WT mice.

Finally, one cannot eliminate other mechanisms that pitavastatin could work through, such as NO-dependent mechanisms independent of eNOS. Lack of one NOS isoform could be compensated for by the increased activity of another isoform (Quinlan et al., 1998). It has been shown in eNOS-/- mice under normoxic conditions that elevated pulmonary blood pressure is seen, but this effect was partially compensated for by an increase in the expression of iNOS and nNOS (Steudel et al., 1997). In cerebral pial arterioles from the eNOS-/-, ACh has been shown to stimulate relaxation through NO-dependent and independent mechanisms, the NO-dependent mechanism being attributed to nNOS (Meng et al., 1996).Moreover, statins are able to promote antioxidant effects and protect endothelial health by inhibiting Rho/ROCK signalling (Ma and Han, 2005). This data from the current in vivo study suggests that length of time of a treatment could influence vascular response, as pitavastatin did not enhance endothelium dependent or independent relaxation after 2 hr incubation in the acute in-vitro study. 247

In a study by Singh et al. using the complement component 1, q subcomponent (C1q) knockout mouse model, pravastatin (5 mg/d) was given from E6-E15 in pregnant mice. Pravastatin treatment was able to normalise angiotensin II contractile response of aortic systemic vessels from C1q knockout to WT levels (Singh et al., 2011). Similarly, Constantine et al. demonstrated an attenuation in phenylephrine-induced contraction in carotid arteries from mice exposed to sFlt-1 treated with pravastatin compared to untreated controls although no difference was observed in responses to thromboxane A2 (U46619) (Costantine et al., 2010). This was in a study where pravastatin (5 mg/kg/d) was dissolved in drinking water and administered from E9-E18 to sFlt-1 mice (a PE-like model created by injecting adenovirus carrying sFlt-1 into pregnant CD1 mice at E8).

In the current study, there was no significant effect of pitavastatin treatment on the U46619-mediated contractile responses of maternal uterine or mesenteric arteries. The data from the current study partially agree with that of Costantine et al. who saw no effect of pravastatin on contraction when U46619 was used. While pitavastatin may not have attenuated contraction, it did not increase maximal contraction or sensitivity to U46619, which was observed in uterine arteries from WT mice in the acute pitavastatin study ex vivo previously reported here.

No difference was seen in umbilical artery contraction to U46619 or relaxation to SNP between pitavastatin-treated and vehicle control groups from eNOS-/- and WT mice. This complements the results seen in the short-term pitavastatin study, indicating that acute and chronic exposure to pitavastatin does not affect U46619 contraction of umbilical arteries. This outcome is desirable and demonstrates that when pitavastatin is administered to a pregnant mouse at a clinically relevant dose, at a time-point approximately at the beginning of the 2nd trimester – up to the third trimester, it does not negatively affect fetal vascular function as assessed via the umbilical artery. This outcome highlights the importance of in vivo studies as blunted umbilical artery relaxation was observed following 2 hr pitavastatin incubation in vitro study however, this observation did not continue after 8 day exposure to pitavastatin in an in vivo model.

From studies in non-pregnant humans, it has been that shown that pitavastatin administered orally can reach peak plasma concentration within 1 hour (Luo et al., 2015). Furthermore, pitavastatin is minimally metabolized by P450 enzymes making its bioavailability higher than other statins (Saito, 2011). This is important to note as the 248

metabolic rate of a mouse is predicted to be seven times greater than in humans (Demetrius, 2005). In this study, pitavastatin was administered via drinking water at a concentration of 6 µg/ml, which is approximately equivalent to a human oral dose of 4 mg (maximum dose for humans). In terms of the level of exposure of pitavastatin to the fetus, this is uncertain as the levels of pitavastatin achieved within the fetal circulation in this current study are unknown. However, studies in the dually-perfused human placenta imply that statin transfer can occur, although to a limited extent (Zarek et al., 2013). HPLC studies were not performed in the present study so the exact therapeutic concentration of pitavastatin and its metabolite in mouse plasma is uncertain, which is a limitation of the current study.

5.5.7 Summary The data from this current study suggests that apart from a small reduction in brain weight (as % of hysterectomized weight) pitavastatin was well tolerated and did not affect litter size or fetal weight. Akin to women, mice also demonstrate vascular adaptations in the uteroplacental circulation during normal pregnancy. In eNOS-/- mice, remodelling is inadequate resulting in increased vascular resistance, endothelial dysfunction and hypertension. Ex vivo, short-term exposure (2 hr) to pitavastatin at a suprapharmacological dose had no beneficial effect on vascular function in eNOS-/- mice and a potentially harmful effect in uterine and umbilical arteries from WT mice. In mice treated with pitavastatin chronically (8 days) in vivo, an enhanced eNOS-independent relaxation was seen in uterine and mesenteric arteries suggesting the potential influence of pitavastatin on vasodilators such as prostacyclin, EDHF and NO from other NOS isoforms. This all occurred without pitavastatin having any obvious detrimental effects on the fetus as evidenced by normal fetal/placental weight and litter size, at the same time there was no added benefit of treatment.

Previous studies have suggested that the eNOS-/- mouse might be usefully applied to characterizing a subpopulation of women with vascular disease, evident in some women with PE as well as those with chronic hypertension and superimposed PE. Using this model has also provided a framework to explore how pitavastatin could be used as a potential therapeutic in the treatment of endothelial dysfunction as well as to garner information about a target therapeutic window for pitavastatin. To our knowledge, this is the first description of pitavastatin’s effects on vascular function in a model of chronic hypertension

249

and FGR. In summary, these data suggest, in mice at least, that pitavastatin is safe for both the dam and fetus when given from mid-pregnancy and pitavastatin demonstrates modest improvements in uterine artery vascular reactivity following chronic exposure.

250

Chapter 6: General Discussion

251

6.0 Overview Pre-eclampsia (PE) is associated with significant maternal and fetal morbidity and mortality, affecting 3-5% of pregnancies, and is responsible for 70,000 maternal deaths and 500,000 fetal deaths worldwide per year (Phipps et al., 2019b). Currently, preterm delivery of the fetus and the placenta is the only way to resolve the clinical signs of PE (Kuklina et al., 2009; Wallis et al., 2008). Hypertensive disorders of pregnancy have also been shown to be associated with an increased risk for cardio- and cerebrovascular disease later in these women later in life (Bell, 2010). Despite medical advances in obstetric medicine, there is a scarcity of novel drugs to treat or prevent PE, with research focusing on the repurposing of existing drugs instead. Correcting the initial placental dysfunction or improving maternal clinical symptoms are key targets to prevent maternal complications and prolong gestation.

A limitation of existing studies investigating the effect of statins on vascular function has been that effects of statins have been demonstrated in in vivo animal models of PE or during pregnancy in humans in small clinical trials, with little focus on vascular function. Few or no studies have been performed ex vivo to assess the acute effects of statin treatment. As stated, very few preclinical and clinical studies have assessed the effects of such treatment on vascular function during pregnancy, and in particular, to investigate potential beneficial or detrimental effects on maternal or fetoplacental vessels following statin exposure. Finally, many studies that have tested statin use in pregnancy complications such as PE have predominantly used pravastatin, whilst the effects of other statins in pregnancy are unknown. This thesis investigated the potential for statins to be repurposed for the treatment of vascular dysfunction in PE as well as other hypertensive disorders of pregnancy such as chronic hypertension (CHT) and superimposed PE. This work represents the first study of the acute (ex vivo) exposure of statin treatment on human vascular function in omental (OAs) and chorionic plate arteries (CPAs) and the acute and chronic (in vivo) effects of statin treatment in mice uterine, mesenteric and umbilical arteries.

6.1 Key Findings A detailed summary and comparison of the results obtained from this study will be outlined in the sections below, with maternal and fetal vessels discussed separately. Novel findings of this study include the observation that statins do not appear to have detrimental effects on the vascular function of human CPAs. Secondly, whilst a pathological vascular 252

phenotype was not demonstrated in OAs from CHT or superimposed PE pregnancies, OAs from PE pregnancies constricted more when expressed as %KPSS in this study, but acute treatment with pitavastatin did not alter vascular function. Furthermore, in mice, acute treatment with pitavastatin had no effect on vascular function in maternal vessels from eNOS-/- mice but suggested a detrimental effect in maternal uterine and fetal umbilical arteries from WT mice. However, following chronic treatment of eNOS-/- and WT mice in vivo, there were no harmful effects of pitavastatin on the fetus observed, evidenced by normal fetal/placental weight and litter size and normal vascular reactivity in maternal and fetal vessels from eNOS-/- and WT mice. Moreover, there was an enhancement of endothelium-dependent relaxation in uterine arteries from pitavastatin treated eNOS-/- mice compared to the vehicle group. In addition to these novel findings, the in vivo study in mice helped to confirm features of the eNOS-/- phenotype.

6.1.1 Effect of Statin Treatment on Maternal Vessels In non-pregnant populations, statins have been shown to exert beneficial effects by lowering cholesterol and managing CVD risk (Mills et al., 2008). Pleiotropic effects of statins include anti-inflammatory properties, decreasing inflammatory cells in atherosclerotic plaques and increasing plaque stability by reducing macrophages and MMPs. Furthermore, statins can enhance endothelial function by activating the ROCK pathway through the inhibition of the isoprenylation of Rac and Rho. Statins have been shown to stimulate heme oxygenase-1 and enhance activity of NO synthase both of which offer vasoprotection (Ramma and Ahmed, 2014). Additionally, statins decrease phenylephrine-induced constriction in a PE sFlt-1 mouse model (Costantine et al., 2010). Other animal models of PE have shown encouraging roles of statins such as restoring the angiogenic imbalance (Kumasawa et al., 2011; Saad et al., 2014) in human umbilical vein endothelial cells, primary trophoblast cells and placenta (Brownfoot et al., 2016b). These properties of statins make them a candidate therapy to improve vascular function.

The hypothesis for maternal vessels in humans was:-

“Short-term (2h) exposure to statins will improve vascular reactivity of omental arteries from pre-eclamptic, chronic hypertensive and superimposed pre-eclamptic pregnancies.”

In acute studies in mice, the hypothesis was:-

253

“In eNOS knockout mice, short-term (2h) exposure to pitavastatin will result in improved uterine artery function.”

In studies in women, omental arteries (OAs) were employed as this vascular bed assists in the regulation of blood pressure during pregnancy by contributing to total peripheral vascular resistance (Hammond et al., 2011). Additionally, from a practical perspective, these vessels are also more accessible than, say, myometrial vessels during caesarean section. OAs can be used as a human vascular tissue model for both normal and pathological pregnancies (Dong et al., 2015). In mice, the uterine artery was representative of the uteroplacental circulation.

Previous studies examining the effects of statins on vascular function ex vivo have differed in species, disease of interest, concentration and choice of statin used, physiological state i.e. non-pregnant or pregnant, treatment period and the vascular bed investigated. Due to the large number of variables mentioned, this has led to heterogeneous results regarding pleiotropic effects of statins ex vivo. The acute ex vivo experiments described herein have attempted to eliminate some of these ambiguities by using a single methodology and keeping the choice of statin, treatment duration and dose constant in vascular beds from humans and mice. As a result, this study has demonstrated that acute statin exposure (2 hr) at the same concentration (1 µM) can have disparate effects on vascular function, depending on the vascular bed and whether vessels are healthy or have existing signs of endothelial dysfunction.

In the literature, endothelial dysfunction has been observed in vessels from women with PE (Ashworth et al., 1997; Mishra et al., 2011; Pascoal et al., 1998; Wareing et al., 2006b). Similarly, in the eNOS-/- mouse model, endothelial dysfunction has also been seen relative to WT mice (Finn-Sell et al., 2018; Kulandavelu et al., 2006; Kusinski et al., 2012). Interestingly, endothelial dysfunction was not seen in OAs from PE, CHT or superimposed PE pregnancies in the current study. The reasons for this could be a limitation of sample size as the studies assessing endothelial dysfunction, particularly in the CHT and superimposed PE groups, were likely underpowered due to a relative paucity of samples available. Furthermore, although DMSO was the appropriate control for the experiments focused upon the effects of pitavastatin, it was evident that DMSO appeared to mask the endothelial dysfunction. In this study, samples from hypertensive pregnancies, which focused on pitavastatin treatment, did not have a negative control (DMSO free) and 254

therefore it was not possible to comprehensively assess vascular function in the context of the pregnancy complications included. Pascoal et al. demonstrated reduced endothelial- dependent relaxation to the muscarinic agonist acetylcholine in OAs from PE women. However, relaxation to bradykinin was similar between PE and normotensive women (Pascoal et al., 1998). In contrast to previous studies demonstrating blunted relaxation of omental, myometrial or subcutaneous fat arteries in PE (Ashworth et al., 1997; Knock and Poston, 1996; Kublickiene et al., 1998; Mccarthy et al., 1993; Wareing et al., 2006b), the current study suggests that endothelial function did not differ between hypertensive and normotensive pregnancies. This highlights the importance of future investigations into endothelial dysfunction in hypertensive pregnancies, in particular taking into account the agonists used to relax vessels, vascular bed of choice and the length of time a women has had vascular disease. Following a 2 hr incubation with 1 µM pravastatin and pitavastatin, OAs from normal pregnancies showed no difference in U46619-mediated contraction or bradykinin-mediated relaxation. In the mouse, increased uterine artery constriction to U46619 was observed (expressed as %KPSS) and impaired relaxation to acetylcholine compared to WT mice, similar to previous reports in the literature. When pitavastatin was exposed to uterine arteries from WT mice, raised basal tone and increased sensitivity to U46619 was observed. This effect was not seen in eNOS-/- mice, suggesting the presence of eNOS could be involved in the mechanism behind these pitavastatin-mediated effects. eNOS is the enzyme in the endothelium responsible for regulating vascular tone (Forstermann and Sessa, 2012). It is uncertain as to the reasons why pitavastatin affected vasoconstriction in WT mice, however, statins are known to alter NO bioavailability and it

- - has been shown that NO can react with the radical O2 to form ONOO and promote eNOS uncoupling, hence disturbing vascular homeostasis (Landmesser et al., 2004). It is possible that, in a healthy endothelium, this could result in raised vascular tone. Statins are also known to possess antioxidant properties, however ROS have a physiological role in cell function such as activation of redox-sensitive transcription factors and protein kinases (Burton and Jauniaux, 2011; Mannaerts et al., 2018; Wu et al., 2016). A dysregulation of ROS in a healthy state could have a detrimental effect on vascular tone. Furthermore, there are prostaglandin-like compounds called isoprostanes formed through the reaction of free radicals and arachidonic acid (Duhig et al., 2016; Turpin et al., 2015). Statins have been shown to enhance production of arachidonic acid (Levine, 2003) which could have

255

interacted with physiological ROS, thus causing harm in the endothelium of WT mice. It is important to note that the dose of pitavastatin used was suprapharmacological, as is common place in these myography studies, and hence observations should be interpreted with caution. Given that these effects of pitavastatin in WT mice were not observed in human omental vessels, it may hint at differences in the mechanisms of statin according to the vascular bed, and also the species, studied. In OAs from hypertensive pregnancies, a 2 hr incubation with pitavastatin did not alter vascular function, but this should be tempered by the fact that endothelial function was not significantly different to OAs between pathological and normal pregnancies in the current study. Currently, no study has assessed acute effects of statins on maternal vascular function ex vivo so the study cannot be directly compared to others in the literature. Similarly, in the eNOS-/- mouse, a model of CHT and FGR presenting with endothelial dysfunction, pitavastatin did not attenuate this vascular dysfunction as assessed by U46619-mediated constriction or endothelium-dependent relaxation. There was a trend towards impaired endothelium-independent relaxation in eNOS-/- mice compared to WT mice, which was not improved by pitavastatin. The use of vascular beds from both humans and mice both confirmed that pitavastatin had no impact on vascular function following acute exposure. The data from this part of my study suggested that time of statin exposure may have been a limiting factor as discussed in the next paragraph.

Studies involving pitavastatin have suggested vasoprotective effects and underlying mechanisms, but statins have been given for different durations in these studies. In vitro studies have shown that 0.01 µM and 0.1 µM pitavastatin in HUVECs can increase NO production through eNOS following 6 hr treatment (Wang et al., 2005) and that pitavastatin can protect endothelial cell viability and enhance migration/proliferation at a concentration of 0.01 µM after 24 hrs (Katsumoto et al., 2005). Furthermore, in HUVECs, pitavastatin (0.1– 3.0 µM) given for 24 h potently increased the release of urocortin from HUVECs in a dose-dependent manner. Urocortin is a potent vasodilator that may lower blood pressure and increase K+ATP channel gene expression (Honjo et al., 2006). In-vivo studies in non-pregnant mice have shown that pitavastatin, given for 2 weeks, promotes antioxidant effects by suppressing NADPH expression (Yagi et al., 2008). Kobayashi et al. demonstrated in rats with heart failure that pitavastatin treatment for 7 weeks stimulates eNOS expression and phosphorylation through the PI3K-Akt signalling pathway and suppresses the ROCK pathway (Kobayashi et al., 2009). Additionally, Almeida et al. showed 256

that 15-day treatment with pitavastatin in hypercholesterolemic rabbits reduced lipid peroxidation resulting in reduced oxidative stress and reversal of endothelial dysfunction (Almeida and Ozaki, 2014). Finally, Mitsuhashi et al. showed in non-pregnant eNOS-/- mice, that a 2-week treatment with pitavastatin is capable of activating extracellular signal- regulated kinases ERK1/2 and ERK5 (Mitsuhashi et al., 2018); ERK5 has been reported to regulate endothelial integrity and protect against vascular dysfunction and disease (Hayashi et al., 2004). In addition, pitavastatin activated AMP kinase (AMPK), which maintains endothelial homeostasis, in response to various stimuli, including pro- inflammatory cytokines and oxidative damage (Schulz et al., 2008). Pitavastatin may not have been potent enough to demonstrate its pleiotropic effects on vascular function in the 2 hr timeframe, indicating a longer therapeutic window could alter this observation, as discussed in the next paragraph. Alternatively, a higher dose of pitavastatin for this short time period may be more effective.

When conducting the in vivo study in mice, focused upon pitavastatin, the hypothesis was

“In eNOS knockout mice, pitavastatin treatment from embryonic days 10.5 to 18.5 will improve maternal vascular (mesenteric artery and uterine artery) function without having detrimental effects on umbilical artery function or fetal growth.’

Taking into consideration that short-term pitavastatin exposure did not have any beneficial effect on maternal vessels from normal or pathological pregnancies in human or mouse, an in vivo study involving chronic exposure to pitavastatin (8 day treatment from E10.5 until E18.5) was implemented. A key finding from this part of the study was that pitavastatin was able to promote acetylcholine-mediated endothelium-dependent relaxation in uterine arteries from eNOS-/- mice. This effect was not seen in uterine arteries following a 2 hr acute exposure to pitavastatin, suggesting that length of treatment may have been a contributing factor. Furthermore, unlike the ex vivo study, pitavastatin treatment did not cause an increase in U46619 sensitivity in uterine arteries from WT mice. This could be due to a difference in dose as the dose used in vivo was more akin to that given in human studies. Moreover, the effects of pitavastatin in a closed system ex vivo are likely to be different to the effects it may have following its metabolism in vivo. In normal and hypertensive human pregnancies, pitavastatin had no effect on vascular function of OAs following acute treatment. However, in vivo, the mesenteric arteries from eNOS-/- mice showed a trend

257

towards increased acetylcholine mediated relaxation, but this should be considered in the context that mesenteric artery function is similar between WT and eNOS-/- mice. This is similar to OAs from human hypertensive pregnancies which did not exhibit a phenotype different from the hypertensive groups.

The uterine artery relaxation data in eNOS-/- mice suggested that pitavastatin could be working through eNOS-independent mechanisms to cause these vasodilatory effects, however further studies using inhibitors of the NO, COX and EDHF pathway would be required to confirm these observations. Pitavastatin showed modest effects on vascular function in vivo and no effect ex vivo following acute treatment, which suggests that prolonged exposure of pitavastatin is required to incur a beneficial effect. Furthermore, ex vivo studies in both maternal vessels from humans and mice were complicated by the use of DMSO as the diluent used in pitavastatin studies (Cayman, 2018) and its known effects on vascular function (Kaneda et al., 2016). DMSO was initially used in vivo but the taste/smell was not agreeable to the mice and hence the vehicle carboxymethylcellulose (CMC) was used instead. This diluent did not appear to have a significant effect on vascular function in the in vivo studies; however, there are no other ex vivo studies with which to compare its effects. It is also important to note that pitavastatin treatment did not have any detrimental effect on maternal water intake, maternal weight or maternal organ weight with the exception of maternal brain weight in pitavastatin-treated eNOS-/- mice; this appeared to be lower than that in the vehicle group (expressed as % hysterectomized weight). As this was the first study to use pitavastatin in vivo in pregnant mice, other groups should be encouraged to perform similar experiments to see if the effect of pitavastatin on maternal brain is indeed reproducible.

The study conducted in eNOS-/- and WT mice highlighted the complexity of transitioning from an in vitro model to an in vivo model. This study revealed that oral pitavastatin treatment was feasible and well tolerated (typically, statins are administered orally to patients). However, information regarding metabolite bioavailability and pharmacokinetics (absorption, protein bindings, metabolism, half-life and clearance) remained unknown. These factors have implications on the therapeutic dose to be used in vivo. The circulating concentrations of pitavastatin achieved, and thus required to promote uterine artery relaxation, was not determined in the current study but stored samples will be useful for answering this question in future. The dose of pitavastatin used in the in vivo

258

mouse study was 6 µg/ml/day based on a human equivalent dose of 4 mg/day (calculated using dose translation; (Reagan-Shaw et al., 2008). The dose achieved in mice may have been influenced by the intake of water given that mice will drink varying quantities each day. In women, pitavastatin can be administered as an oral tablet with the specific dose needed. The equivalent in mice would be to perform a daily gavage procedure but this is a stressful procedure. Giving a daily gavage for 8 consecutive days during mid to late gestation is far from ideal and so the decision was made to give pitavastatin via drinking water and to change the bottle daily. It is important to note that humans are more heterogeneous than inbred mice in their genetic make-up, with variable pharmacokinetic characteristics but also there is the issue of polypharmacy. Women with hypertensive pregnancies will likely be taking antihypertensive medication and aspirin and it is uncertain how statin interaction with these medications could affect vascular function. Mice, on the other hand, did not possess these confounding effects, making it slightly easier to study the isolated effects of pitavastatin on vascular function and fetal and maternal health.

6.1.2. Effect of Statin Treatment on Fetoplacental Vessels In humans, the hypothesis was:-

“Short-term (2h) exposure to statins will have no detrimental effects on fetoplacental chorionic plate artery function.”

While in mice, the hypothesis was:-

“In eNOS knockout mice, short-term (2h) exposure to pitavastatin will have no detrimental effects on umbilical artery function.”

In humans, CPAs were used as a proxy for the fetoplacental circulation whilst in mice, the umbilical artery was selected. In humans, there has been conflicting evidence regarding vascular reactivity of CPAs from PE pregnancies, with studies showing an increase (Benoit et al., 2007; Bertrand et al., 1993), decrease (Wareing and Baker, 2004) and no change in constriction (Ong et al., 2002) and blunted relaxation (Ong et al., 2002) compared to normotensive pregnancies. In eNOS-/- mice, Kusinski et al. reported no difference in either U46619-mediated constriction or SNP-mediated relaxation in umbilical arteries versus WT (Kusinski et al., 2012).

The key purpose of using these vascular beds was to investigate safety and assess whether statins had any detrimental effects on vascular function in fetoplacental vessels. 259

In the acute study (2 hr statin exposure) in human normotensive pregnancies, 1 µM simvastatin and 1/10 µM pravastatin did not have an effect on either constriction or relaxation, while 1 µM pitavastatin enhanced SNP-mediated relaxation and 10 µM pitavastatin attenuated constriction. In WT mice, 1 µM pitavastatin had no effect on constriction compared to the DMSO control but did blunt SNP-mediated relaxation, an effect not seen in human CPAs. The reasons for this are unclear but could be species- specific differences or related to the different vascular beds used. In addition, pitavastatin could be having an inhibitory effect by innervating adrenergic receptors in smooth muscle of the umbilical artery, antagonizing the vasorelaxatory effect.

When 1 µM pravastatin, simvastatin and pitavastatin were used in CPAs from PE pregnancies, there was no difference in U46619-mediated constriction or SNP-mediated relaxation, while 10 µM pitavastatin attenuated constriction. In mice, 1 µM pitavastatin had no effect on constriction to U46619 or relaxation to SNP in umbilical arteries of eNOS- /- mice. Considering that it was only the high dose of pitavastatin (10 µM) that attenuated U46619-mediated constriction of CPAs from PE pregnancies, this suggests that statins may not have any beneficial effect on CPAs clinically. However, given that fetoplacental vessels are not the primary therapeutic target in cases of maternal hypertension, this is not necessarily a key finding. As the maternal vascular bed would be the therapeutic target for statins to prevent or treat PE, the likely dose reaching the fetoplacental vasculature following metabolism of pitavastatin would be much lower than the doses used in this study. It is reassuring that no harmful effect on fetoplacental vessels was seen following acute statin treatment. However, if fetoplacental vascular reactivity was enhanced by statins, whilst the maternal vasculature was unaffected, this may have been useful for hypertensive pregnancies complicated by fetal growth restriction (FGR) and characterised by abnormal fetoplacental blood flow.

For the in vivo study, the hypothesis for fetoplacental vessels was:-

Following maternal treatment with pitavastatin in-vivo, there were no apparent detrimental effects on the fetus as evidenced by litter size, fetal weight and fetal biometry all being comparable between pitavastatin-treated and vehicle groups in both WT and

260

eNOS-/- mice. Similarly, umbilical artery function showed no difference to vehicle group post-treatment in both genotypes. Pitavastatin failed to increase fetal weight in eNOS-/- mice in the current study, while other studies have reported that other statins such as pravastatin given in utero increased placental and fetal weight (Ahmed et al., 2010; Wyrwoll et al., 2016). Unlike the ex vivo data, pitavastatin treatment in vivo did not cause blunted SNP-mediated relaxation and showed no difference in vascular function relative to the vehicle group in both WT and eNOS-/- mice. It was encouraging that pitavastatin, at a physiological dose and over a longer treatment period, did not cause harm to the fetus. Clinically, pitavastatin may not be the optimum choice of drug for normalising fetal weight, improving umbilical artery function and preventing preterm birth, but if given as a maternal treatment to relieve maternal complications and to prolong gestation; our data suggest it does not appear to affect fetal development. Long-term studies on offspring would be required to assess cardiovascular and metabolic health, behaviour and cognitive function of individuals exposed to statins in utero.

6.2 Application of Statins in Pregnancy and Safety Considerations When considering a drug for the treatment or prevention of PE, it is important to consider potential detrimental effects on the fetoplacental circulation and the fetus itself. Clinical treatments for PE should undergo substantial toxicological studies to provide safety data on effects on fetoplacental hemodynamics, fetal development and physiology as well as any detrimental effects on maternal health and physiology.

The data gathered in the study from the human CPAs and mouse umbilical arteries do not suggest that statins have negative effects on vascular function, which is in support of the hypothesis. The FDA categorise drugs according to teratogenic risk based on evidence from animal and human studies offering guidance to clinicians. The category with the lowest risk is category A, followed by B, C and D; however, drugs in category X are contraindicated for use during pregnancy (Sachdeva et al., 2009). Statins have been placed into category X. Through the use of ex vivo and in vivo studies in humans and mice, this thesis has shown that acute exposure of CPAs to pitavastatin in humans does not alter vascular function in normotensive or PE pregnancies. Although blunted relaxation was observed in umbilical arteries of WT mice, this observation was not mimicked in the in vivo studies, while umbilical artery function in eNOS-/- mice post-pitavastatin treatment was normal. In the in vivo studies in mice, fetal wellbeing post-pitavastatin treatment was 261

investigated in the form of fetal/placental weight, litter size and fetal biometric measurements such as crown-rump length, head and abdominal circumstance (indicators of fetal growth). These were all normal in both genotypes. Alongside this, clinical evidence from other groups investigating the use of statins for treatment or prevention of PE have not shown adverse effects of statin use in pregnancy (Ahmed, 2011; Brownfoot et al., 2015; Costantine et al., 2016; Lefkou et al., 2016; Wang et al., 2008). The data from this thesis, coupled with the existing literature, warrants consideration of a revision of the FDA’s decision to place statins in category X, as the teratogenic risk appears low. However, further studies investigating long-term consequences for offspring of statin-treated pregnancies are still required. It is important to note that CPAs are not the only representative vessels for the fetal circulation and hence cannot solely be used to determine that statins or other drugs do or do not have a vascular effect. A useful ex vivo method to assess drug transfer and vascular tone in human placenta is the dual-perfusion method, which has already been performed with pravastatin (Balan et al., 2017; Nanovskaya et al., 2013; Zarek et al., 2013). It would be beneficial to use this method to further assess the safety profile of pitavastatin and other statins. Whilst fetal vascular function was not detrimentally affected by pitavastatin treatment, nor was fetal growth affected, it is uncertain how statin treatment may affect other fetal organs such as the fetal brain, and this aspect was not explored in the current study. Although an effect of pitavastatin was seen on maternal brain weight in eNOS-/- treated mice, neurodevelopment is a major concern in relation to statins, given their role in cholesterol metabolism. These questions would need to be addressed using suitable animal models of PE. Evidence from epidemiological and animal studies have demonstrated that exposure of the fetus in utero to harmful exogenous stimuli such as toxic agents, environmental hazards or maternal stress could influence the developmental trajectory of the fetal brain (Horton et al., 2014).

The fetal brain has a high cholesterol component, and 70% of the cholesterol in the fetal brain is used for myelin sheath production, whilst also being used for glial and neuronal membranes (Dietschy, 2009; Goluszko and Nowicki, 2005). With statins known to inhibit HMG CoA reductase and to lower cholesterol in the liver, it is important that cholesterol levels are unaffected in extrahepatic cells such as those in the fetal brain. In the safety and efficacy trial conducted by Constantine et al. a decrease in maternal total plasma cholesterol (TC) and LDL was observed in pravastatin-treated women compared to a 262

placebo group. However, cord blood concentrations of TC and LDL were similar between the pravastatin- and placebo-exposed fetuses after 18 weeks of treatment (~14-32 weeks gestation). This outcome is encouraging; however, extrahepatic effects of other statins such as pitavastatin would also need to be investigated. Hydrophilic statins include rosuvastatin and pravastatin whilst cerivastatin, pitavastatin, simvastatin, fluvastatin, and atorvastatin are all hydrophobic statins. How hydrophilic or hydrophobic a statin is will determine its bioavailability, bioactivity and its ability to diffuse across the lipid bilayer of cells and even the blood brain barrier. More hydrophobic statins typically diffuse by passive diffusion while hydrophilic statins require passage via facilitated diffusion or active transport (Fong, 2014). The method of uptake of a statin into cells could potentially influence its safety and vascular effects in pregnancy.

When considering a drug for clinical treatment or prevention of pregnancy complications, alongside an attempt to attenuate maternal clinical signs, a reasonable objective is to also to prolong gestation of pregnancy in women with hypertension associated with FGR. The EPICure Study (1995) is a series of studies investigating survival and long-term health of babies and young people who were born prematurely from 22-26 weeks’ gestation. They showed that survival in fetuses born prior to 24 weeks was rare in the UK. The EPICure 2 study showed that survival for births at 24-25 weeks has risen significantly while babies born at 26 weeks continued to have higher survival as reported in the 1995 study. These changes reflect an improvement in standardised hospital care. Nevertheless, the EPICure 2 study showed improved fetal survival but no difference in acute neonatal morbidity and suggested that babies born extremely preterm (22-25) remain at high risk of lifelong health problems (Costeloe, 2006; Costeloe et al., 2012). Preterm birth, as well as the obvious health costs, also carries high economic costs, with studies showing that preterm infants are more likely to require additional hospital resources and access to community health services compared to fetuses born at term (Brooten et al., 1986; Mccormick et al., 1991). Studies like these will inform the development of future economic evaluations of interventions aimed at preventing/managing preterm birth (Petrou et al., 2006).

The acute study ex vivo in vessels from humans and mice assessed pitavastatin for treatment of vascular dysfunction in PE. PE is typically diagnosed in the 3rd trimester (post- 20 weeks) and so if statins were to be used for prevention it would be in a specific high-risk

263

cohort of women i.e. with a history of PE (previously diagnosed), with CHT, which could progress to superimposed PE, or women with diabetes or antiphospholipid syndrome. It has been strongly argued that statin exposure should be avoided during the 1st trimester, which is a critical period for organogenesis and early placental development. Furthermore, certain drugs, if used during the first trimester of pregnancy, can cause miscarriage, teratogenicity and reduced fetal weight. Hence the decision to use a drug during this period requires the benefits to outweigh the potential risks of the drug therapy (Scully, 2014). Studies have shown deleterious effects of statins in the first trimester based on evidence from human first trimester placental explants and have demonstrated how statins can affect placental development (Forbes et al., 2008; Forbes et al., 2015; Kenis et al., 2005). For these reasons, the 2nd trimester of gestation onwards has been targeted for treatment of PE. However, an element of risk still remains as drugs given at any time point in pregnancy can still affect the growth or development of the fetus (Nice, 2019b). However, by the 2nd trimester, all major organs are formed in the fetus and so although teratogenic effects are not eliminated, they are minimised. With the use of any drug during pregnancy, it is important to assess the long-term impact on maternal and fetal health post-drug exposure. Long-term fetal outcome studies in animals will be required to assess the effect of statins on neurodevelopment and behaviour as well as long-term follow up studies of fetuses exposed to statins in utero during clinical trials for PE treatment.

6.3 Methodological considerations and limitations of the study This study was subject to certain limitations, which should be considered when interpreting the results. Acute studies performed in humans and mice differed in the vasodilator used, which may have limited comparisons, but the protocols were identical in other respects. Furthermore, human omental biopsies were only obtained if the women consented to donation and if the surgeons were able to collect the sample without surgical complications. These factors restricted sample size. The limited availability of human biopsies (especially pathological samples) was a key driver in also testing pitavastatin in mice. When the effect of pitavastatin on vascular function ex vivo and in vivo was examined, a specific blocker was not used for NO. However, previous studies have shown that statins mediate their vascular effects predominantly through eNOS (Gelosa et al., 2007; Kavalipati et al., 2015; Liao and Laufs, 2005). Hence, eNOS was the focus of interest in the present study as the eNOS-/- mouse was used to determine if statins’ vascular effects were limited

264

to eNOS alone or if they could work through eNOS-independent mechanisms. The preliminary data showing effects of pitavastatin on uterine artery endothelium-dependent relaxation independent of eNOS in vivo may help inform future vascular studies. A limitation of the present study was the choice of dose used in the ex vivo acute study. A suprapharmacological dose was chosen because, for ex vivo myography experiments, a higher dose than those typically achieved in vivo is generally needed to result in similar effects. This meant that observations gained from ex vivo studies might have been slightly exaggerated compared to what would be expected in vivo. Additionally, only one statin was investigated across different vascular beds in humans and mice, which was mainly due to practical reasons such as time, availability of tissue and shared usage of myographs within the laboratory. Whilst uteroplacental vessels were not used in humans, in mice the uterine artery was investigated, which allowed for the assessment of pitavastatin’s effects after acute and chronic treatment. Finally, the use of a negative control would have aided interpretation of results in the acute study in humans. DMSO was the correct experimental control for the studies with pitavastatin as this was the solvent used to dissolve pitavastatin. However, having a water control or PSS only vessel for hypertensive pregnancy studies would have helped to identify if differences in endothelial function existed between normal and hypertensive pregnancies. Moreover, many of the hypertensive women used in our study had the added complication of FGR. In addition, there was no comprehensive assessment of endothelial function (BK dose response curve) prior to statin exposure in the acute studies in humans.

6.4 Clinical perspective Taking into account the current clinical trial data investigating pravastatin for treatment/prevention of PE as well as the data from this thesis, I would conclude that there is still scope for statin use as a therapeutic candidate for PE. The outcome of the in vivo and in vitro data does not provide convincing evidence for pitavastatin as an optimal therapeutic candidate for PE, but animal data has been supportive in showing effects of pravastatin in preventing its onset. These beneficial effects include improving placental blood flow, restoring angiogenic balance, preventing oxidative stress, attenuating endothelial dysfunction and reversing FGR (Ahmed et al., 2010; Bauer et al., 2013; Costantine et al., 2010; Kumasawa et al., 2011; Singh et al., 2011). Furthermore, various other statins such as atorvastatin, simvastatin and rosuvastatin have not been fully

265

investigated for their treatment or prevention potential in PE and other hypertensive pregnancy disorders. By targeting the preclinical stages of PE, statins could prevent onset of clinical symptoms of PE hence promoting maternal health and subsequently leading to better uteroplacental blood flow to the placenta and fetus, consequently prolonging gestation, increasing maturation of the fetus in utero and increasing chances of fetal survival.

There is currently a randomized trial in progress, assessing the role of pravastatin (10 mg, 20 mg and 40 mg) for the prevention of PE in 48 high-risk women (Costantine et al., 2016), along with another trial combining 2 x pravastatin (20 mg) with 80 mg aspirin (280 women) (clinical trial identifier: NCT03648970). Finally, a much larger, well-designed randomized trial with 1550 women is assessing the use of 20 mg pravastatin for prevention of PE in a high-risk population (clinical trial identifier NCT03944512). These studies will be useful in informing us whether statins administered during a specific therapeutic window is successful in preventing onset of PE. Nevertheless, given the existing preclinical and clinical data, the potential short and long-term benefits of statins for pregnancies complicated by PE have warranted further research and their position in category X by FDA should be re-evaluated.

Whilst research into the effectiveness of statins as a treatment for PE continues, it is important to acknowledge that there may be more novel, effective therapies that can better manage and treat PE cases such as those with early-onset PE, or new onset PE that can arise without any previous history. Nevertheless, whilst novelty is encouraged, it is also time-consuming as pregnant women are rarely used in clinical trials, due to the risk of fetal exposure, so there is a paucity of data on how pregnant women may respond to drugs in phase I trials. Repurposing of a drug may be more economically and clinically viable; more studies are looking into combination therapies, for example, metformin and proton pump inhibitors such as esomeprazole or statins and aspirin as alluded to in the study by Lefkou et al. in women with antiphospholipid syndrome (Lefkou et al., 2016). The additive effects of two separate treatments may be more effective for prophylaxis and/or acute treatment.

6.5 Conclusions The principle conclusions from this study are:-

266

 This is the first study to assess the acute effects of statins during pregnancy using vessels from both humans and animals.

 Following a short-term exposure (2 hr) with 1 µM pitavastatin to OAs from women with PE, CHT or superimposed PE, there was no change in endothelial function. However, altered endothelial function in hypertensive disorders of pregnancy was not observed in this study.

 Following short-term exposure (2 hr) with statins, no detrimental effects on vascular function of CPAs were seen in normotensive or PE women.  A 2 hr treatment with pitavastatin was not associated with improvement in vascular function in uterine arteries at E18.5 in eNOS-/- or WT mice. Umbilical artery function was unaltered by pitavastatin in eNOS-/- mice but increased U46619 sensitivity was observed in uterine arteries and blunted SNP-mediated umbilical artery relaxation in WT mice.  Pitavastatin administration in vivo was unable to offer beneficial increases in fetal/placental weight or litter size but neither did it have detrimental effects on these parameters. Also, in vivo, pitavastatin increased endothelium-dependent relaxation of uterine arteries to acetylcholine in eNOS-/- mice, whilst showing a trend towards increasing relaxation in mesenteric arteries.  Overall, pitavastatin had modest effects on vascular function and did not have detrimental effects on fetoplacental vascular function, but does not appear to be the best statin for treatment or prevention of PE and associated hypertensive pregnancies. Further research should continue to explore effects of alternative statins on maternal and fetal vascular function in pregnancy.

6.6 Future work The results obtained from this study have brought to light specific avenues for future research to take place to assess the role of statins in hypertensive pregnancies.

 Assess endothelial function of OAs (BK dose response curve) prior to statin exposure to assess differences in relaxation between PE and other hypertensive pregnancies and normotensive pregnancies.

267

 Determine transplacental transfer and distribution of pitavastatin or other lipophilic statins by performing dual perfusion of placental lobules in order to assess the transfer of pitavastatin in the maternal-to-fetal and fetal-to-maternal directions.

 Begin pitavastatin treatment at an earlier time point in vivo in mice, to determine if effects on vascular function are more pronounced.

 Ascertain the concentration of pitavastatin in maternal and fetal plasma in mice using HPLC, thus allowing comparison with concentrations in humans taking this medication. This will also offer increased understanding of the therapeutic concentration required to stimulate effects as well as estimates of the clearance rate of pitavastatin and its major metabolite over the treatment period.  Assess the mechanism through which pitavastatin promoted endothelium-dependent relaxation of uterine and mesenteric arteries in eNOS-/- mice. This could involve the use of NG-nitro-L-arginine methyl ester (L-NAME) or Nomega-Nitro-L-arginine (L-NNA) to rule out NO-dependent pathways but also indomethacin (COX-1 inhibitor) and 4- aminopyridine (K+ channel inhibitor) for NO-independent pathways.

268

Chapter 7: References

269

Aalkjaer, C., Danielsen, H., Johannesen, P., Pedersen, E. B., Rasmussen, A. & Mulvany, M. J. (1985). 'Abnormal vascular function and morphology in pre-eclampsia: a study of isolated resistance vessels', Clin Sci (Lond), 69(4), pp. 477-82. Abramovich, D. R., Page, K. R. & Parkin, A. M. (1984). 'The effect of prostaglandin D2 on the blood vessels of the perfused isolated cotyledon of the human placenta', Br J Pharmacol, 81(1), pp. 19-21. Absi, M., Eid, B. G., Ashton, N., Hart, G. & Gurney, A. M. (2019). 'Simvastatin causes pulmonary artery relaxation by blocking smooth muscle ROCK and calcium channels: Evidence for an endothelium-independent mechanism', PLoS One, 14(8), p. e0220473. Acharya, G., Sonesson, S. E., Flo, K., Rasanen, J. & Odibo, A. (2016). 'Hemodynamic aspects of normal human feto-placental (umbilical) circulation', Acta Obstet Gynecol Scand, 95(6), pp. 672-82. Ahmed (2011). Statins to ameliorate early on-set pre-eclampsia: University of Birmingham. Available at: www.stamp.bham.ac.uk. Ahmed, A., Williams, D. J., Cheed, V., Middleton, L. J., Ahmad, S., Wang, K., Vince, A. T., Hewett, P., Spencer, K., Khan, K. S. & Daniels, J. P. (2020). 'Pravastatin for early-onset preeclampsia: a randomised, blinded, placebo-controlled trial', Bjog, 127(4), pp. 478-488. Ahmed, A. (2011). 'New insights into the etiology of preeclampsia: identification of key elusive factors for the vascular complications', Thromb Res, 127 Suppl 3, pp. S72-5. Ahmed, A. (2011). Statins to ameliorate early on-set pre-eclampsia: University of Birmingham Available at: www.stamp.bham.ac.uk. Ahmed, A., Singh, J., Khan, Y., Seshan, S. V. & Girardi, G. (2010). 'A new mouse model to explore therapies for preeclampsia', PLoS One, 5(10), p. e13663. Akhlaq, M., Nagi, A. H. & Yousaf, A. W. (2012). 'Placental morphology in pre-eclampsia and eclampsia and the likely role of NK cells', Indian J Pathol Microbiol, 55(1), pp. 17-21. Al-Enazy, S., Ali, S., Albekairi, N., El-Tawil, M. & Rytting, E. (2017). 'Placental control of drug delivery', Adv Drug Deliv Rev, 116, pp. 63-72. Al Qahtani, N. (2011). 'Doppler ultrasound in the assessment of suspected intra-uterine growth restriction', Ann Afr Med, 10(4), pp. 266-71. Albaiges, G., Missfelder-Lobos, H., Lees, C., Parra, M. & Nicolaides, K. H. (2000). 'One-stage screening for pregnancy complications by color Doppler assessment of the uterine arteries at 23 weeks' gestation', Obstet Gynecol, 96(4), pp. 559-64. Almeida, E. A. & Ozaki, M. R. (2014). 'Effect of pitavastatin on vascular reactivity in hypercholesterolemic rabbits', Arq Bras Cardiol, 103(1), pp. 4-12. Almuti, K., Rimawi, R., Spevack, D. & Ostfeld, R. J. (2006). 'Effects of statins beyond lipid lowering: potential for clinical benefits', Int J Cardiol, 109(1), pp. 7-15. Altman, D., Carroli, G., Duley, L., Farrell, B., Moodley, J., Neilson, J. & Smith, D. (2002). 'Do women with pre-eclampsia, and their babies, benefit from magnesium sulphate? The Magpie Trial: a randomised placebo-controlled trial', Lancet, 359(9321), pp. 1877-90. Alvarez De Sotomayor, M., Herrera, M. D., Marhuenda, E. & Andriantsitohaina, R. (2000). 'Characterization of endothelial factors involved in the vasodilatory effect of simvastatin in aorta and small mesenteric artery of the rat', British journal of pharmacology, 131(6), pp. 1179-1187. Alvarez de Sotomayor, M., Perez-Guerrero, C., Herrera, M. D. & Marhuenda, E. (2001). 'Effect of simvastatin on vascular smooth muscle responsiveness: involvement of Ca(2+) homeostasis', Eur J Pharmacol, 415(2-3), pp. 217-24. Andersen MD , A. K. O. A., Christina Søndergaard Duvald, Emmeli Fredsgaard Ravnkilde Mikkelsen, Mikkel Holm Vendelbo, Per Glud Ovesen and Michael (2018). Animal Models of Fetal Medicine and Obstetrics. Anderson, T. J., Meredith, I. T., Yeung, A. C., Frei, B., Selwyn, A. P. & Ganz, P. (1995). 'The effect of cholesterol-lowering and antioxidant therapy on endothelium-dependent coronary vasomotion', N Engl J Med, 332(8), pp. 488-93. Antoniades, C., Bakogiannis, C., Leeson, P., Guzik, T. J., Zhang, M. H., Tousoulis, D., Antonopoulos, A. S., Demosthenous, M., Marinou, K., Hale, A., Paschalis, A., Psarros, C., Triantafyllou, C., 270

Bendall, J., Casadei, B., Stefanadis, C. & Channon, K. M. (2011). 'Rapid, direct effects of statin treatment on arterial redox state and nitric oxide bioavailability in human atherosclerosis via tetrahydrobiopterin-mediated endothelial nitric oxide synthase coupling', Circulation, 124(3), pp. 335-45. Anumba, D. O., Robson, S. C., Boys, R. J. & Ford, G. A. (1999). 'Nitric oxide activity in the peripheral vasculature during normotensive and preeclamptic pregnancy', Am J Physiol, 277(2 Pt 2), pp. H848-54. Anwar, M. A., Docherty, C., Poston, L. & Nathanielsz, P. W. (1999). 'A comparative study of vascular responsiveness of myometrial and omental small resistance arteries in late- gestation sheep', Am J Obstet Gynecol, 181(3), pp. 663-8. Aouache, R., Biquard, L., Vaiman, D. & Miralles, F. (2018). 'Oxidative Stress in Preeclampsia and Placental Diseases', Int J Mol Sci, 19(5). Arao, K., Yasu, T., Umemoto, T., Jinbo, S., Ikeda, N., Ueda, S., Kawakami, M. & Momomura, S. (2009). 'Effects of pitavastatin on fasting and postprandial endothelial function and blood rheology in patients with stable coronary artery disease', Circ J, 73(8), pp. 1523-30. Arndt, H., Smith, C. W. & Granger, D. N. (1993). 'Leukocyte-endothelial cell adhesion in spontaneously hypertensive and normotensive rats', Hypertension, 21(5), pp. 667-73. Ashworth, J. R., Warren, A. Y., Baker, P. N. & Johnson, I. R. (1996). 'A comparison of endothelium- dependent relaxation in omental and myometrial resistance arteries in pregnant and nonpregnant women', Am J Obstet Gynecol, 175(5), pp. 1307-12. Ashworth, J. R., Warren, A. Y., Baker, P. N. & Johnson, I. R. (1997). 'Loss of endothelium- dependent relaxation in myometrial resistance arteries in pre-eclampsia', Br J Obstet Gynaecol, 104(10), pp. 1152-8. Atallah, A., Lecarpentier, E., Goffinet, F., Doret-Dion, M., Gaucherand, P. & Tsatsaris, V. (2017). 'Aspirin for Prevention of Preeclampsia', Drugs, 77(17), pp. 1819-1831. Bader, R. A., Bader, M. E., Rose, D. F. & Braunwald, E. (1955). 'Hemodynamics at rest and during exercise in normal pregnancy as studies by cardiac catheterization', J Clin Invest, 34(10), pp. 1524-36. Balan, A., Szaingurten-Solodkin, I., Swissa, S. S., Feinshtein, V., Huleihel, M., Holcberg, G., Dukler, D. & Beharier, O. (2017a). 'The effects of pravastatin on the normal human placenta: Lessons from ex-vivo models', PLoS One, 12(2), p. e0172174. Bardal, S. K., Waechter, J. E. & Martin, D. S. (2011). 'Chapter 2 - Pharmacokinetics', in Bardal, S. K., Waechter, J. E. & Martin, D. S. (eds.) Applied Pharmacology. Philadelphia: Content Repository Only pp. 17-34. Baschat, A. A. (2004). 'Fetal responses to placental insufficiency: an update', Bjog, 111(10), pp. 1031-41. Bateman, B. T., Hernandez-Diaz, S., Fischer, M. A., Seely, E. W., Ecker, J. L., Franklin, J. M., Desai, R. J., Allen-Coleman, C., Mogun, H., Avorn, J. & Huybrechts, K. F. (2015). 'Statins and congenital malformations: cohort study', Bmj, 350, p. h1035. Bauer, A. J., Banek, C. T., Needham, K., Gillham, H., Capoccia, S., Regal, J. F. & Gilbert, J. S. (2013). 'Pravastatin attenuates hypertension, oxidative stress, and angiogenic imbalance in rat model of placental ischemia-induced hypertension', Hypertension, 61(5), pp. 1103-10. Bayhan G, Koçyigit Y, Atamer A, et al. Potential atherogenic roles of lipids, lipoprotein(a) and lipid peroxidation in preeclampsia. Gynecol Endocrinol. 2005;21(1):1–6.

Beevers, G., Lip, G. Y. & O'Brien, E. (2001). 'ABC of hypertension: The pathophysiology of hypertension', Bmj, 322(7291), pp. 912-6.

Belfort, M. A., Saade, G. R., Suresh, M., Kramer, W. & Vedernikov, Y. P. (1996). 'Effects of selected vasoconstrictor agonists on isolated omental artery from premenopausal nonpregnant women and from normal and preeclamptic pregnant women', Am J Obstet Gynecol, 174(2), pp. 687-93. Bell, M. J. (2010). 'A historical overview of preeclampsia-eclampsia', J Obstet Gynecol Neonatal Nurs, 39(5), pp. 510-8.

271

Bellosta, S., Ferri, N., Bernini, F., Paoletti, R. & Corsini, A. (2000). 'Non-lipid-related effects of statins', Ann Med, 32(3), pp. 164-76. Bełtowski, J. & Jamroz-Wiśniewska, A. (2012). 'Modulation of h(2)s metabolism by statins: a new aspect of cardiovascular pharmacology', Antioxidants & redox signaling, 17(1), pp. 81-94. Benirschke K, K. P., Baergen R. (2006). Chapter 12. Anatomy and pathology of the umbilical cord. (5th ed.). Benoit, C., Zavecz, J. & Wang, Y. (2007). 'Vasoreactivity of chorionic plate arteries in response to vasoconstrictors produced by preeclamptic placentas', Placenta, 28(5-6), pp. 498-504. Benton, S. J., McCowan, L. M., Heazell, A. E., Grynspan, D., Hutcheon, J. A., Senger, C., Burke, O., Chan, Y., Harding, J. E., Yockell-Lelievre, J., Hu, Y., Chappell, L. C., Griffin, M. J., Shennan, A. H., Magee, L. A., Gruslin, A. & von Dadelszen, P. (2016). 'Placental growth factor as a marker of fetal growth restriction caused by placental dysfunction', Placenta, 42, pp. 1-8. Bergmann, A., Ahmad, S., Cudmore, M., Gruber, A. D., Wittschen, P., Lindenmaier, W., Christofori, G., Gross, V., Gonzalves, A., Grone, H. J., Ahmed, A. & Weich, H. A. (2010). 'Reduction of circulating soluble Flt-1 alleviates preeclampsia-like symptoms in a mouse model', J Cell Mol Med, 14(6b), pp. 1857-67. Bertrand, C., Duperron, L. & St-Louis, J. (1993). 'Umbilical and placental vessels: modifications of their mechanical properties in preeclampsia', Am J Obstet Gynecol, 168(5), pp. 1537-46. Bhide, A., Acharya, G., Bilardo, C. M., Brezinka, C., Cafici, D., Hernandez-Andrade, E., Kalache, K., Kingdom, J., Kiserud, T., Lee, W., Lees, C., Leung, K. Y., Malinger, G., Mari, G., Prefumo, F., Sepulveda, W. & Trudinger, B. (2013). 'ISUOG practice guidelines: use of Doppler ultrasonography in obstetrics', Ultrasound Obstet Gynecol, 41(2), pp. 233-39. Billington, C. K. & Penn, R. B. (2003). 'Signaling and regulation of G protein-coupled receptors in airway smooth muscle', Respir Res, 4, p. 2. Billington, W. D. & Bell, S. C. (1983). 'Fetal histocompatibility antigens and maternal immune responses', Ciba Found Symp, 96, pp. 69-88. Bjorkhem-Bergman, L., Lindh, J. D. & Bergman, P. (2011). 'What is a relevant statin concentration in cell experiments claiming pleiotropic effects?', Br J Clin Pharmacol, 72(1), pp. 164-5. Blumenthal, R. S. (2000). 'Statins: effective antiatherosclerotic therapy', Am Heart J, 139(4), pp. 577-83. Boo, Y. C., Sorescu, G., Boyd, N., Shiojima, I., Walsh, K., Du, J. & Jo, H. (2002). 'Shear stress stimulates phosphorylation of endothelial nitric-oxide synthase at Ser1179 by Akt- independent mechanisms: role of protein kinase A', J Biol Chem, 277(5), pp. 3388-96. Boura, A. L., Walters, W. A., Read, M. A. & Leitch, I. M. (1994). 'Autacoids and control of human placental blood flow', Clin Exp Pharmacol Physiol, 21(10), pp. 737-48. Bower, S., Kingdom, J. & Campbell, S. (1998). 'Objective and subjective assessment of abnormal uterine artery Doppler flow velocity waveforms', Ultrasound Obstet Gynecol, 12(4), pp. 260-4. Bramham, K., Parnell, B., Nelson-Piercy, C., Seed, P. T., Poston, L. & Chappell, L. C. (2014). 'Chronic hypertension and pregnancy outcomes: systematic review and meta-analysis', Bmj, 348, p. g2301. Braunthal, S. & Brateanu, A. (2019). 'Hypertension in pregnancy: Pathophysiology and treatment', SAGE Open Med, 7, p. 2050312119843700. Brennecke, S. P., Gude, N. M., Di Iulio, J. L. & King, R. G. (1997). 'Reduction of placental nitric oxide synthase activity in pre-eclampsia', Clin Sci (Lond), 93(1), pp. 51-5. Bridges, L. E., Williams, C. L., Pointer, M. A. & Awumey, E. M. (2011). 'Mesenteric artery contraction and relaxation studies using automated wire myography', J Vis Exp, (55). Brooten, D., Kumar, S., Brown, L. P., Butts, P., Finkler, S. A., Bakewell-Sachs, S., Gibbons, A. & Delivoria-Papadopoulos, M. (1986). 'A randomized clinical trial of early hospital discharge and home follow-up of very-low-birth-weight infants', N Engl J Med, 315(15), pp. 934-9. Brosens, I., Pijnenborg, R., Vercruysse, L. & Romero, R. (2011). 'The "Great Obstetrical Syndromes" are associated with disorders of deep placentation', Am J Obstet Gynecol, 204(3), pp. 193- 201.

272

Brosens, I. A., Robertson, W. B. & Dixon, H. G. (1972). 'The role of the spiral arteries in the pathogenesis of preeclampsia', Obstet Gynecol Annu, 1, pp. 177-91. Brown, M. A., Magee, L. A., Kenny, L. C., Karumanchi, S. A., McCarthy, F. P., Saito, S., Hall, D. R., Warren, C. E., Adoyi, G. & Ishaku, S. (2018). 'Hypertensive Disorders of Pregnancy: ISSHP Classification, Diagnosis, and Management Recommendations for International Practice', Hypertension, 72(1), pp. 24-43. Brownfoot, F. C., Hannan, N. J., Cannon, P., Nguyen, V., Hastie, R., Parry, L. J., Senadheera, S., Tuohey, L., Tong, S. & Kaitu'u-Lino, T. J. (2019). 'KK', EBioMedicine, 41, pp. 636-648. Brownfoot, F. C., Hastie, R., Hannan, N. J., Cannon, P., Tuohey, L., Parry, L. J., Senadheera, S., Illanes, S. E., Kaitu'u-Lino, T. J. & Tong, S. (2016a). 'Metformin as a prevention and treatment for preeclampsia: effects on soluble fms-like tyrosine kinase 1 and soluble endoglin secretion and endothelial dysfunction', Am J Obstet Gynecol, 214(3), pp. 356.e1- 356.e15. Brownfoot, F. C., Tong, S., Hannan, N. J., Binder, N. K., Walker, S. P., Cannon, P., Hastie, R., Onda, K. & Kaitu'u-Lino, T. J. (2015). 'Effects of Pravastatin on Human Placenta, Endothelium, and Women With Severe Preeclampsia', Hypertension, 66(3), pp. 687-97; discussion 445. Brownfoot, F. C., Tong, S., Hannan, N. J., Hastie, R., Cannon, P. & Kaitu'u-Lino, T. J. (2016b). 'Effects of simvastatin, rosuvastatin and pravastatin on soluble fms-like tyrosine kinase 1 (sFlt-1) and soluble endoglin (sENG) secretion from human umbilical vein endothelial cells, primary trophoblast cells and placenta', BMC Pregnancy Childbirth, 16, p. 117. Brugts, J. J., Yetgin, T., Hoeks, S. E., Gotto, A. M., Shepherd, J., Westendorp, R. G., de Craen, A. J., Knopp, R. H., Nakamura, H., Ridker, P., van Domburg, R. & Deckers, J. W. (2009). 'The benefits of statins in people without established cardiovascular disease but with cardiovascular risk factors: meta-analysis of randomised controlled trials', Bmj, 338, p. b2376. Burke, S. D. & Karumanchi, S. A. (2013). 'Spiral artery remodeling in preeclampsia revisited', Hypertension, 62(6), pp. 1013-4. Burnstock, G. (1988). 'Sympathetic purinergic transmission in small blood vessels', Trends Pharmacol Sci, 9(4), pp. 116-7. Burton, G. J., Charnock-Jones, D. S. & Jauniaux, E. (2009a). 'Regulation of vascular growth and function in the human placenta', Reproduction, 138(6), pp. 895-902. Burton, G. J. & Fowden, A. L. (2012). 'Review: The placenta and developmental programming: balancing fetal nutrient demands with maternal resource allocation', Placenta, 33 Suppl, pp. S23-7. Burton, G. J., Fowden, A. L. & Thornburg, K. L. (2016). 'Placental Origins of Chronic Disease', Physiol Rev, 96(4), pp. 1509-65. Burton, G. J. & Jauniaux, E. (2011). 'Oxidative stress', Best Pract Res Clin Obstet Gynaecol, 25(3), pp. 287-99. Burton, G. J., Jauniaux, E. & Charnock-Jones, D. S. (2010). 'The influence of the intrauterine environment on human placental development', Int J Dev Biol, 54(2-3), pp. 303-12. Burton, G. J., Redman, C. W., Roberts, J. M. & Moffett, A. (2019). 'Pre-eclampsia: pathophysiology and clinical implications', Bmj, 366, p. l2381. Burton, G. J., Woods, A. W., Jauniaux, E. & Kingdom, J. C. (2009b). 'Rheological and physiological consequences of conversion of the maternal spiral arteries for uteroplacental blood flow during human pregnancy', Placenta, 30(6), pp. 473-82. Burton, G. J., Yung, H. W., Cindrova-Davies, T. & Charnock-Jones, D. S. (2009c). 'Placental endoplasmic reticulum stress and oxidative stress in the pathophysiology of unexplained intrauterine growth restriction and early onset preeclampsia', Placenta, 30 Suppl A, pp. S43-8. Bustamante, J. J., Dai, G. & Soares, M. J. (2008). 'Pregnancy and lactation modulate maternal splenic growth and development of the erythroid lineage in the rat and mouse', Reprod Fertil Dev, 20(2), pp. 303-10. Cabrales, P., Tsai, A. G., Frangos, J. A. & Intaglietta, M. (2005). 'Role of endothelial nitric oxide in microvascular oxygen delivery and consumption', Free Radic Biol Med, 39(9), pp. 1229-37. 273

Cameron, I. T., van Papendorp, C. L., Palmer, R. M. J., Smith, S. K. & Moncada, S. (1993). 'Relationship Between Nitric Oxide Synthesis and Increase in Systolic Blood Pressure in Women with Hypertension in Pregnancy', Hypertension in Pregnancy, 12(1), pp. 85-92. Camici, G. G., Savarese, G., Akhmedov, A. & Luscher, T. F. (2015). 'Molecular mechanism of endothelial and vascular aging: implications for cardiovascular disease', Eur Heart J, 36(48), pp. 3392-403. Carter, A. M., Kingston, M. J., Han, K. K., Mazzuca, D. M., Nygard, K. & Han, V. K. (2005). 'Altered expression of IGFs and IGF-binding proteins during intrauterine growth restriction in guinea pigs', J Endocrinol, 184(1), pp. 179-89. Cartwright, J. E., Fraser, R., Leslie, K., Wallace, A. E. & James, J. L. (2010). 'Remodelling at the maternal-fetal interface: relevance to human pregnancy disorders', Reproduction, 140(6), pp. 803-13. Cayman (2018). 'Pitavastatin (Calcium Salt)'. Cerdeira, A. S., Agrawal, S., Staff, A. C., Redman, C. W. & Vatish, M. (2018). 'Angiogenic factors: potential to change clinical practice in pre-eclampsia?', Bjog, 125(11), pp. 1389-1395. Challis, J. R. G., Matthews, S. G., Gibb, W. & Lye, S. J. (2000). 'Endocrine and paracrine regulation of birth at term and preterm', Endocr Rev, 21(5), pp. 514-50. Chang, J. & Streitman, D. (2012). 'Physiologic adaptations to pregnancy', Neurol Clin, 30(3), pp. 781-9. Chapman, A. B., Abraham, W. T., Zamudio, S., Coffin, C., Merouani, A., Young, D., Johnson, A., Osorio, F., Goldberg, C., Moore, L. G., Dahms, T. & Schrier, R. W. (1998). 'Temporal relationships between hormonal and hemodynamic changes in early human pregnancy', Kidney Int, 54(6), pp. 2056-63. Chappell, L. C., Duckworth, S., Seed, P. T., Griffin, M., Myers, J., Mackillop, L., Simpson, N., Waugh, J., Anumba, D., Kenny, L. C., Redman, C. W. & Shennan, A. H. (2013). 'Diagnostic accuracy of placental growth factor in women with suspected preeclampsia: a prospective multicenter study', Circulation, 128(19), pp. 2121-31. Chataigneau, T., Feletou, M., Huang, P. L., Fishman, M. C., Duhault, J. & Vanhoutte, P. M. (1999). 'Acetylcholine-induced relaxation in blood vessels from endothelial nitric oxide synthase knockout mice', Br J Pharmacol, 126(1), pp. 219-26. Chau, K., Hennessy, A. & Makris, A. (2017). 'Placental growth factor and pre-eclampsia', J Hum Hypertens, 31(12), pp. 782-786. Chaudhuri, G., Buga, G. M., Gold, M. E., Wood, K. S. & Ignarro, L. J. (1991). 'Characterization and actions of human umbilical endothelium derived relaxing factor', Br J Pharmacol, 102(2), pp. 331-6. Chen, C. W., Jaffe, I. Z. & Karumanchi, S. A. (2014). 'Pre-eclampsia and cardiovascular disease', Cardiovasc Res, 101(4), pp. 579-86. Chen, L., Haught, W. H., Yang, B., Saldeen, T. G., Parathasarathy, S. & Mehta, J. L. (1997). 'Preservation of endogenous antioxidant activity and inhibition of lipid peroxidation as common mechanisms of antiatherosclerotic effects of vitamin E, lovastatin and amlodipine', J Am Coll Cardiol, 30(2), pp. 569-75. Cheung, K. L. & Lafayette, R. A. (2013). 'Renal physiology of pregnancy', Adv Chronic Kidney Dis, 20(3), pp. 209-14. Choi, J. W., Im, M. W. & Pai, S. H. (2002). 'Nitric oxide production increases during normal pregnancy and decreases in preeclampsia', Ann Clin Lab Sci, 32(3), pp. 257-63. Cindrova-Davies, T., van Patot, M. T., Gardner, L., Jauniaux, E., Burton, G. J. & Charnock-Jones, D. S. (2015). 'Energy status and HIF signalling in chorionic villi show no evidence of hypoxic stress during human early placental development', Mol Hum Reprod, 21(3), pp. 296-308. (1994). 'CLASP: a randomised trial of low-dose aspirin for the prevention and treatment of preeclampsia among 9364 pregnant women. CLASP (Collaborative Low-dose Aspirin Study in Pregnancy) Collaborative Group', Lancet, 343(8898), pp. 619-29. Coats, P. & Hillier, C. (1999). 'Determination of an optimal axial-length tension for the study of isolated resistance arteries on a pressure myograph', Exp Physiol, 84(6), pp. 1085-94.

274

Cogolludo, A., Moreno, L., Bosca, L., Tamargo, J. & Perez-Vizcaino, F. (2003). 'Thromboxane A2- induced inhibition of voltage-gated K+ channels and pulmonary vasoconstriction: role of protein kinase Czeta', Circ Res, 93(7), pp. 656-63. Cohen, R. A. & Vanhoutte, P. M. (1995). 'Endothelium-dependent hyperpolarization. Beyond nitric oxide and cyclic GMP', Circulation, 92(11), pp. 3337-49. Conrad, K. P., Joffe, G. M., Kruszyna, H., Kruszyna, R., Rochelle, L. G., Smith, R. P., Chavez, J. E. & Mosher, M. D. (1993). 'Identification of increased nitric oxide biosynthesis during pregnancy in rats', Faseb j, 7(6), pp. 566-71. Cooke, C. L. & Davidge, S. T. (2003). 'Endothelial-dependent vasodilation is reduced in mesenteric arteries from superoxide dismutase knockout mice', Cardiovasc Res, 60(3), pp. 635-42. Coppage, K. H., Sun, X., Baker, R. S. & Clark, K. E. (2005). 'Expression of phosphodiesterase 5 in maternal and fetal sheep', Am J Obstet Gynecol, 193(3 Pt 2), pp. 1005-10. Costantine, M. M. & Cleary, K. (2013). 'Pravastatin for the prevention of preeclampsia in high-risk pregnant women', Obstet Gynecol, 121(2 Pt 1), pp. 349-53. Costantine, M. M., Cleary, K., Hebert, M. F., Ahmed, M. S., Brown, L. M., Ren, Z., Easterling, T. R., Haas, D. M., Haneline, L. S., Caritis, S. N., Venkataramanan, R., West, H., D'Alton, M. & Hankins, G. (2016). 'Safety and pharmacokinetics of pravastatin used for the prevention of preeclampsia in high-risk pregnant women: a pilot randomized controlled trial', Am J Obstet Gynecol, 214(6), pp. 720.e1-720.e17. Costantine, M. M., Tamayo, E., Lu, F., Bytautiene, E., Longo, M., Hankins, G. D. & Saade, G. R. (2010). 'Using pravastatin to improve the vascular reactivity in a mouse model of soluble fms-like tyrosine kinase-1-induced preeclampsia', Obstet Gynecol, 116(1), pp. 114-20. Costeloe, K. (2006). 'EPICure: facts and figures: why preterm labour should be treated', Bjog, 113 Suppl 3, pp. 10-2. Costeloe, K. L., Hennessy, E. M., Haider, S., Stacey, F., Marlow, N. & Draper, E. S. (2012). 'Short term outcomes after extreme preterm birth in England: comparison of two birth cohorts in 1995 and 2006 (the EPICure studies)', Bmj, 345, p. e7976. Cross, J. C., Hemberger, M., Lu, Y., Nozaki, T., Whiteley, K., Masutani, M. & Adamson, S. L. (2002). 'Trophoblast functions, angiogenesis and remodeling of the maternal vasculature in the placenta', Mol Cell Endocrinol, 187(1-2), pp. 207-12. Cross, J. C., Werb, Z. & Fisher, S. J. (1994). 'Implantation and the placenta: key pieces of the development puzzle', Science, 266(5190), pp. 1508-18. Cruz, M. A., Dominguez, A., Gallardo, V., Carrasco, G., Miguel, P. & Gonzalez, C. (2000). 'Vascular reactivity to angiotensin II and eicosanoid production in the human placenta from term and preterm pregnancy', Gynecol Obstet Invest, 50(4), pp. 247-53. Csige, I., Ujvarosy, D. & Szabo, Z. (2018). 'The Impact of Obesity on the Cardiovascular System', 2018, p. 3407306. Cudmore, M., Ahmad, S., Al-Ani, B., Fujisawa, T., Coxall, H., Chudasama, K., Devey, L. R., Wigmore, S. J., Abbas, A., Hewett, P. W. & Ahmed, A. (2007). 'Negative regulation of soluble Flt-1 and soluble endoglin release by heme oxygenase-1', Circulation, 115(13), pp. 1789-97. Curtis, N. E., Gude, N. M., King, R. G., Marriott, P. J., Rook, T. J. & Brennecke, S. P. (1995). 'Nitric Oxide Metabolites in Normal Human Pregnancy and Preeclampsia', Hypertension in Pregnancy, 14(3), pp. 339-349. Daud, A. N., Bergman, J. E., Oktora, M. P., Kerstjens-Frederikse, W. S., Groen, H., Bos, J. H., Hak, E. & Wilffert, B. (2017). 'Maternal use of drug substrates of placental transporters and the effect of transporter-mediated drug interactions on the risk of congenital anomalies', PLoS One, 12(3), p. e0173530. Davidge, S. T., Stranko, C. P. & Roberts, J. M. (1996). 'Urine but not plasma nitric oxide metabolites are decreased in women with preeclampsia', Am J Obstet Gynecol, 174(3), pp. 1008-13. Davignon, J. (2012). 'Pleiotropic effects of pitavastatin', British journal of clinical pharmacology, 73(4), pp. 518-535. Davison, J. M. & Dunlop, W. (1980). 'Renal hemodynamics and tubular function normal human pregnancy', Kidney Int, 18(2), pp. 152-61. 275

Davisson, R. L., Hoffmann, D. S., Butz, G. M., Aldape, G., Schlager, G., Merrill, D. C., Sethi, S., Weiss, R. M. & Bates, J. N. (2002). 'Discovery of a spontaneous genetic mouse model of preeclampsia', Hypertension, 39(2 Pt 2), pp. 337-42. DeFronzo, R. A. (1981). 'Insulin and renal sodium handling: clinical implications', Int J Obes, 5 suppl 1, pp. 93-104. Dekan, S., Linduska, N., Kasprian, G. & Prayer, D. (2012). 'MRI of the placenta - a short review', Wien Med Wochenschr, 162(9-10), pp. 225-8. Dekker G, Sibai BM: Etiology and pathogenesis of preeclampsia: Current concepts. Am J Obstet Gynecol 1998;179:1359–1375. Demetrius, L. (2005). 'Of mice and men. When it comes to studying ageing and the means to slow it down, mice are not just small humans', EMBO Rep, 6 Spec No, pp. S39-44. Diaz, P., Powell, T. L. & Jansson, T. (2014). 'The role of placental nutrient sensing in maternal-fetal resource allocation', Biol Reprod, 91(4), p. 82. Dietschy, J. M. (2009). 'Central nervous system: cholesterol turnover, brain development and neurodegeneration', Biol Chem, 390(4), pp. 287-93. Dilworth, M. R., Kusinski, L. C., Baker, B. C., Renshall, L. J., Greenwood, S. L., Sibley, C. P. & Wareing, M. (2011). 'Defining fetal growth restriction in mice: A standardized and clinically relevant approach', Placenta, 32(11), pp. 914-6. Dong, Y., Betancourt, A., Chauhan, M., Balakrishnan, M., Lugo, F., Anderson, M. L., Espinoza, J., Fox, K., Belfort, M. & Yallampalli, C. (2015). 'Pregnancy Increases Relaxation in Human Omental Arteries to the CGRP Family of Peptides', Biol Reprod, 93(6), p. 134. Doridot, L., Passet, B., Mehats, C., Rigourd, V., Barbaux, S., Ducat, A., Mondon, F., Vilotte, M., Castille, J., Breuiller-Fouche, M., Daniel, N., le Provost, F., Bauchet, A. L., Baudrie, V., Hertig, A., Buffat, C., Simeoni, U., Germain, G., Vilotte, J. L. & Vaiman, D. (2013). 'Preeclampsia-like symptoms induced in mice by fetoplacental expression of STOX1 are reversed by aspirin treatment', Hypertension, 61(3), pp. 662-8. Dorne, J. L. & Renwick, A. G. (2005). 'The refinement of uncertainty/safety factors in risk assessment by the incorporation of data on toxicokinetic variability in humans', Toxicol Sci, 86(1), pp. 20-6. Doupis, J., Rahangdale, S., Gnardellis, C., Pena, S. E., Malhotra, A. & Veves, A. (2011). 'Effects of diabetes and obesity on vascular reactivity, inflammatory cytokines, and growth factors', Obesity (Silver Spring), 19(4), pp. 729-35. Duhig, K., Chappell, L. C. & Shennan, A. H. (2016). 'Oxidative stress in pregnancy and reproduction', Obstetric medicine, 9(3), pp. 113-116. Duhig, K. E., Myers, J., Seed, P. T., Sparkes, J., Lowe, J., Hunter, R. M., Shennan, A. H. & Chappell, L. C. (2019a). 'Placental growth factor testing to assess women with suspected preeclampsia: a multicentre, pragmatic, stepped-wedge cluster-randomised controlled trial', Lancet, 393(10183), pp. 1807-1818. Duley, L. (2003). 'Pre-eclampsia and the hypertensive disorders of pregnancy', Br Med Bull, 67, pp. 161-76. Duley, L., Gulmezoglu, A. M., Henderson-Smart, D. J. & Chou, D. (2010a). 'Magnesium sulphate and other anticonvulsants for women with pre-eclampsia', Cochrane Database Syst Rev, (11), p. Cd000025. Duley, L., Henderson-Smart, D. J., Walker, G. J. & Chou, D. (2010b). 'Magnesium sulphate versus diazepam for eclampsia', Cochrane Database Syst Rev, (12), p. Cd000127. Duncan, K. R., Issa, B., Moore, R., Baker, P. N., Johnson, I. R. & Gowland, P. A. (2005). 'A comparison of fetal organ measurements by echo-planar magnetic resonance imaging and ultrasound', Bjog, 112(1), pp. 43-9. Duplain, H., Burcelin, R., Sartori, C., Cook, S., Egli, M., Lepori, M., Vollenweider, P., Pedrazzini, T., Nicod, P., Thorens, B. & Scherrer, U. (2001). 'Insulin resistance, hyperlipidemia, and hypertension in mice lacking endothelial nitric oxide synthase', Circulation, 104(3), pp. 342-5. Dymara-Konopka, W., Laskowska, M. & Blazewicz, A. (2018). 'Angiogenic Imbalance as a Contributor of Preeclampsia', Curr Pharm Biotechnol, 19(10), pp. 797-815. 276

Easterling, T. R., Benedetti, T. J., Carlson, K. C., Brateng, D. A., Wilson, J. & Schmucker, B. S. (1991). 'The effect of maternal hemodynamics on fetal growth in hypertensive pregnancies', Am J Obstet Gynecol, 165(4 Pt 1), pp. 902-6. Eddy, A. C., Bidwell, G. L., 3rd & George, E. M. (2018). 'Pro-angiogenic therapeutics for preeclampsia', Biol Sex Differ, 9(1), p. 36. Edison, R. J. & Muenke, M. (2004). 'Mechanistic and epidemiologic considerations in the evaluation of adverse birth outcomes following gestational exposure to statins', Am J Med Genet A, 131(3), pp. 287-98. Edwards, G. & Weston, A. H. (2004). 'Potassium and potassium clouds in endothelium-dependent hyperpolarizations', Pharmacol Res, 49(6), pp. 535-41. El-Sayed, A. A. F. (2017). 'Preeclampsia: A review of the pathogenesis and possible management strategies based on its pathophysiological derangements', Taiwan J Obstet Gynecol, 56(5), pp. 593-598. Elsby, R., Smith, V., Fox, L., Stresser, D., Butters, C., Sharma, P. & Surry, D. D. (2011). 'Validation of membrane vesicle-based breast cancer resistance protein and multidrug resistance protein 2 assays to assess drug transport and the potential for drug-drug interaction to support regulatory submissions', Xenobiotica, 41(9), pp. 764-83. Elsheikh, A., Creatsas, G., Mastorakos, G., Milingos, S., Loutradis, D. & Michalas, S. (2001). 'The renin-aldosterone system during normal and hypertensive pregnancy', Arch Gynecol Obstet, 264(4), pp. 182-5. Emet T, Ustüner I, Güven SG, et al. Plasma lipids and lipoproteins during pregnancy and related pregnancy outcomes. Arch Gynecol Obstet 2013;288(01):49-55.

Endo, A. (2010). 'A historical perspective on the discovery of statins', Proc Jpn Acad Ser B Phys Biol Sci, 86(5), pp. 484-93.

Eneroth-Grimfors, E., Lindblad, L. E., Westgren, M., Ihrman-Sandahl, C. & Bevegard, S. (1993). 'Noninvasive test of microvascular endothelial function in normal and hypertensive pregnancies', Br J Obstet Gynaecol, 100(5), pp. 469-71. Enquobahrie, D. A., Williams, M. A., Butler, C. L., Frederick, I. O., Miller, R. S. & Luthy, D. A. (2004). 'Maternal plasma lipid concentrations in early pregnancy and risk of preeclampsia', Am J Hypertens, 17(7), pp. 574-81. Euser, A. G. & Cipolla, M. J. (2009). 'Magnesium sulfate for the treatment of eclampsia: a brief review', Stroke, 40(4), pp. 1169-75. Facemire, C. S., Nixon, A. B., Griffiths, R., Hurwitz, H. & Coffman, T. M. (2009). 'Vascular endothelial growth factor receptor 2 controls blood pressure by regulating nitric oxide synthase expression', Hypertension, 54(3), pp. 652-8. Falloon, B. J., Stephens, N., Tulip, J. R. & Heagerty, A. M. (1995). 'Comparison of small artery sensitivity and morphology in pressurized and wire-mounted preparations', Am J Physiol, 268(2 Pt 2), pp. H670-8. Faraci, F. M., Sigmund, C. D., Shesely, E. G., Maeda, N. & Heistad, D. D. (1998). 'Responses of carotid artery in mice deficient in expression of the gene for endothelial NO synthase', Am J Physiol, 274(2), pp. H564-70. FDA (1980). ' Drug bulletin', Federal Regulation, 44, pp. 37434 – 67. Feghali, M., Venkataramanan, R. & Caritis, S. (2015). 'Pharmacokinetics of drugs in pregnancy', Semin Perinatol, 39(7), pp. 512-9. Fehsel, K., Jalowy, A., Qi, S., Burkart, V., Hartmann, B. & Kolb, H. (1993). 'Islet cell DNA is a target of inflammatory attack by nitric oxide', Diabetes, 42(3), pp. 496-500. Finn-Sell, S. L., Cottrell, E. C., Greenwood, S. L., Dilworth, M. R., Cowley, E. J., Sibley, C. P. & Wareing, M. (2018). 'Pomegranate Juice Supplementation Alters Utero-Placental Vascular Function and Fetal Growth in the eNOS(-/-) Mouse Model of Fetal Growth Restriction', Front Physiol, 9, p. 1145. Fisher, S. A. (2010). 'Vascular smooth muscle phenotypic diversity and function', Physiol Genomics, 42a(3), pp. 169-87. 277

Fitzgerald, B., Shannon, P., Kingdom, J. & Keating, S. (2011). 'Rounded intraplacental haematomas due to decidual vasculopathy have a distinctive morphology', J Clin Pathol, 64(8), pp. 729- 32. Flammer, A. J., Anderson, T., Celermajer, D. S., Creager, M. A., Deanfield, J., Ganz, P., Hamburg, N. M., Luscher, T. F., Shechter, M., Taddei, S., Vita, J. A. & Lerman, A. (2012). 'The assessment of endothelial function: from research into clinical practice', Circulation, 126(6), pp. 753- 67. Fleming, I. & Busse, R. (2003). 'Molecular mechanisms involved in the regulation of the endothelial nitric oxide synthase', Am J Physiol Regul Integr Comp Physiol, 284(1), pp. R1- 12. Foëx, P. & Sear, J. W. (2004). 'Hypertension: pathophysiology and treatment', BJA Education, 4(3), pp. 71-75. Fong, C. W. (2014). 'Statins in therapy: understanding their hydrophilicity, lipophilicity, binding to 3-hydroxy-3-methylglutaryl-CoA reductase, ability to cross the blood brain barrier and metabolic stability based on electrostatic molecular orbital studies', Eur J Med Chem, 85, pp. 661-74. Foo, F. L., Mahendru, A. A., Masini, G., Fraser, A., Cacciatore, S., MacIntyre, D. A., McEniery, C. M., Wilkinson, I. B., Bennett, P. R. & Lees, C. C. (2018). 'Association Between Prepregnancy Cardiovascular Function and Subsequent Preeclampsia or Fetal Growth Restriction', Hypertension, 72(2), pp. 442-450. Forbes, K., Hurst, L. M., Aplin, J. D., Westwood, M. & Gibson, J. M. (2008). 'Statins are detrimental to human placental development and function; use of statins during early pregnancy is inadvisable', J Cell Mol Med, 12(6a), pp. 2295-6. Forbes, K., Shah, V. K., Siddals, K., Gibson, J. M., Aplin, J. D. & Westwood, M. (2015). 'Statins inhibit insulin-like growth factor action in first trimester placenta by altering insulin-like growth factor 1 receptor glycosylation', Mol Hum Reprod, 21(1), pp. 105-14. Forbes, K. & Westwood, M. (2008). 'The IGF axis and placental function. a mini review', Horm Res, 69(3), pp. 129-37. Forstermann, U. & Sessa, W. C. (2012). 'Nitric oxide synthases: regulation and function', Eur Heart J, 33(7), pp. 829-37, 837a-837d. Fox, K. A., Longo, M., Tamayo, E., Kechichian, T., Bytautiene, E., Hankins, G. D., Saade, G. R. & Costantine, M. M. (2011). 'Effects of pravastatin on mediators of vascular function in a mouse model of soluble Fms-like tyrosine kinase-1-induced preeclampsia', Am J Obstet Gynecol, 205(4), pp. 366.e1-5. Frampton, G. K., Jones, J., Rose, M. & Payne, L. (2016). 'Placental growth factor (alone or in combination with soluble fms-like tyrosine kinase 1) as an aid to the assessment of women with suspected pre-eclampsia: systematic review and economic analysis', Health Technol Assess, 20(87), pp. 1-160. Frank, S., Stallmeyer, B., Kampfer, H., Schaffner, C. & Pfeilschifter, J. (1999). 'Differential regulation of vascular endothelial growth factor and its receptor fms-like-tyrosine kinase is mediated by nitric oxide in rat renal mesangial cells', Biochem J, 338 ( Pt 2), pp. 367-74. Fu, J., Zhao, L., Wang, L. & Zhu, X. (2015). 'Expression of markers of endoplasmic reticulum stress- induced apoptosis in the placenta of women with early and late onset severe preeclampsia', Taiwan J Obstet Gynecol, 54(1), pp. 19-23. Fujino, H., Yamada, I., Shimada, S., Yoneda, M. & Kojima, J. (2003). 'Metabolic fate of pitavastatin, a new inhibitor of HMG-CoA reductase: human UDP-glucuronosyltransferase enzymes involved in lactonization', Xenobiotica, 33(1), pp. 27-41. Furukawa, S., Kuroda, Y. & Sugiyama, A. (2014). 'A comparison of the histological structure of the placenta in experimental animals', J Toxicol Pathol, 27(1), pp. 11-8. Fushima, T., Sekimoto, A., Minato, T., Ito, T., Oe, Y., Kisu, K., Sato, E., Funamoto, K., Hayase, T., Kimura, Y., Ito, S., Sato, H. & Takahashi, N. (2016). 'Reduced Uterine Perfusion Pressure (RUPP) Model of Preeclampsia in Mice', PLoS One, 11(5), p. e0155426. Gallos ID, Sivakumar K, Kilby MD, et al. Pre-eclampsia is associated with, and preceded by, hypertriglyceridaemia: a meta-analysis. BJOG. 2013;120(11):1321–1332. 278

Garcia-Recio, S. & Gascon, P. (2015). 'Biological and Pharmacological Aspects of the NK1- Receptor', Biomed Res Int, 2015, p. 495704. Gardosi, J., Francis, A., Turner, S. & Williams, M. (2018). 'Customized growth charts: rationale, validation and clinical benefits', Am J Obstet Gynecol, 218(2s), pp. S609-s618. Garmendia, J. V., Gutierrez, Y., Blanca, I., Bianco, N. E. & De Sanctis, J. B. (1997). 'Nitric oxide in different types of hypertension during pregnancy', Clin Sci (Lond), 93(5), pp. 413-21. Gati, S., Papadakis, M., Papamichael, N. D., Zaidi, A., Sheikh, N., Reed, M., Sharma, R., Thilaganathan, B. & Sharma, S. (2014). 'Reversible de novo left ventricular trabeculations in pregnant women: implications for the diagnosis of left ventricular noncompaction in low-risk populations', Circulation, 130(6), pp. 475-83. Gelinas, D. S., Bernatchez, P. N., Rollin, S., Bazan, N. G. & Sirois, M. G. (2002). 'Immediate and delayed VEGF-mediated NO synthesis in endothelial cells: role of PI3K, PKC and PLC pathways', Br J Pharmacol, 137(7), pp. 1021-30. Gelosa, P., Cimino, M., Pignieri, A., Tremoli, E., Guerrini, U. & Sironi, L. (2007). 'The role of HMG- CoA reductase inhibition in endothelial dysfunction and inflammation', Vascular health and risk management, 3(5), pp. 567-577. Ghio A, Bertolotto A, Resi V, Volpe L, Di Cianni G. Triglyceride metabolism in pregnancy. Adv Clin Chem 2011;55:133-153. Giles, W., O'Callaghan, S., Read, M., Gude, N., King, R. & Brennecke, S. (1997). 'Placental nitric oxide synthase activity and abnormal umbilical artery flow velocity waveforms', Obstet Gynecol, 89(1), pp. 49-52. Gille J.H., Moore D.G., Sedgwick C. J. (1977). Placental infarction: a sign of pre-eclampsia in a patas monkey (Erythrocebus patas). Lab. Anim. Sci. 27, 1190121. Gittenberger-de Groot, A. C., Bartelings, M. M., Poelmann, R. E., Haak, M. C. & Jongbloed, M. R. (2013). 'Embryology of the heart and its impact on understanding fetal and neonatal heart disease', Semin Fetal Neonatal Med, 18(5), pp. 237-44. Giussani, D. A. (2016). 'The fetal brain sparing response to hypoxia: physiological mechanisms', J Physiol, 594(5), pp. 1215-30. Godecke, A. & Schrader, J. (2000). 'Adaptive mechanisms of the cardiovascular system in transgenic mice--lessons from eNOS and myoglobin knockout mice', Basic Res Cardiol, 95(6), pp. 492-8. Godfrey, L. M., Erramouspe, J. & Cleveland, K. W. (2012). 'Teratogenic risk of statins in pregnancy', Ann Pharmacother, 46(10), pp. 1419-24. Goluszko, P. & Nowicki, B. (2005). 'Membrane cholesterol: a crucial molecule affecting interactions of microbial pathogens with mammalian cells', Infect Immun, 73(12), pp. 7791-6. Gomez, O., Figueras, F., Fernandez, S., Bennasar, M., Martinez, J. M., Puerto, B. & Gratacos, E. (2008). 'Reference ranges for uterine artery mean pulsatility index at 11-41 weeks of gestation', Ultrasound Obstet Gynecol, 32(2), pp. 128-32. Gordijn, S. J., Beune, I. M., Thilaganathan, B., Papageorghiou, A., Baschat, A. A., Baker, P. N., Silver, R. M., Wynia, K. & Ganzevoort, W. (2016). 'Consensus definition of fetal growth restriction: a Delphi procedure', Ultrasound Obstet Gynecol, 48(3), pp. 333-9. Granger, J. P., LaMarca, B. B., Cockrell, K., Sedeek, M., Balzi, C., Chandler, D. & Bennett, W. (2006). 'Reduced uterine perfusion pressure (RUPP) model for studying cardiovascular-renal dysfunction in response to placental ischemia', Methods Mol Med, 122, pp. 383-92. Griffiths, S. K. & Campbell, J. P. (2014). 'Placental structure, function and drug transfer', BJA Education, 15(2), pp. 84-89. Grigsby, P. L. (2016). 'Animal Models to Study Placental Development and Function throughout Normal and Dysfunctional Human Pregnancy', Semin Reprod Med, 34(1), pp. 11-6. Grill, S., Rusterholz, C., Zanetti-Dallenbach, R., Tercanli, S., Holzgreve, W., Hahn, S. & Lapaire, O. (2009). 'Potential markers of preeclampsia--a review', Reprod Biol Endocrinol, 7, p. 70. Gude, N. M., Roberts, C. T., Kalionis, B. & King, R. G. (2004). 'Growth and function of the normal human placenta', Thromb Res, 114(5-6), pp. 397-407. Guedes-Martins, L. (2017). 'Superimposed Preeclampsia', Adv Exp Med Biol, 956, pp. 409-417. 279

guideline, N. (2019). Hypertension in pregnancy: diagnosis and management. Available at: https://www.nice.org.uk/guidance/ng133/chapter/Recommendations#management-of- chronic-hypertension-in-pregnancy (Accessed: 12.08.19). Gunn, A. J. & Bennet, L. (2009). 'Fetal hypoxia insults and patterns of brain injury: insights from animal models', Clin Perinatol, 36(3), pp. 579-93. Guy, G. P., Ling, H. Z., Garcia, P., Poon, L. C. & Nicolaides, K. H. (2017). 'Maternal cardiac function at 35-37 weeks' gestation: prediction of pre-eclampsia and gestational hypertension', Ultrasound Obstet Gynecol, 49(1), pp. 61-66. Haba, M., Kinoshita, H., Matsuda, N., Azma, T., Hama-Tomioka, K., Hatakeyama, N., Yamazaki, M. & Hatano, Y. (2009). 'Beneficial effect of propofol on arterial adenosine triphosphate- sensitive K+ channel function impaired by thromboxane', Anesthesiology, 111(2), pp. 279- 86. Hammond, S., Mathewson, A. M., Baker, P. N., Mayhew, T. M. & Dunn, W. R. (2011). 'Gap junctions and hydrogen peroxide are involved in endothelium-derived hyperpolarising responses to bradykinin in omental arteries and veins isolated from pregnant women', Eur J Pharmacol, 668(1-2), pp. 225-32. Han, H. C. (2009a). 'Blood vessel buckling within soft surrounding tissue generates tortuosity', J Biomech, 42(16), pp. 2797-801. Han, H. C. (2009b). 'The theoretical foundation for artery buckling under internal pressure', J Biomech Eng, 131(12), p. 124501. Han, H. C., Chesnutt, J. K., Garcia, J. R., Liu, Q. & Wen, Q. (2013). 'Artery buckling: new phenotypes, models, and applications', Ann Biomed Eng, 41(7), pp. 1399-410. Harold, J. G., Siegel, R. J., FitzGerald, G. A., Satoh, P. & Fishbein, M. C. (1988). 'Differential prostacyclin production by human umbilical vasculature', Arch Pathol Lab Med, 112(1), pp. 43-6. Harris, M. B., Blackstone, M. A., Sood, S. G., Li, C., Goolsby, J. M., Venema, V. J., Kemp, B. E. & Venema, R. C. (2004). 'Acute activation and phosphorylation of endothelial nitric oxide synthase by HMG-CoA reductase inhibitors', Am J Physiol Heart Circ Physiol, 287(2), pp. H560-6. Harvey Motulsky, A. C. (2004). 'Fitting Models to Biological Data Using Linear and Nonlinear Regression: A practical guide to curve fitting', pp. 261-262. Haschek, W. M., Rousseaux, C. G. & Wallig, M. A. (2013). 'Toxicologic Pathology: An Introduction', in Haschek, W. M., Rousseaux, C. G. & Wallig, M. A. (eds.) Haschek and Rousseaux's Handbook of Toxicologic Pathology (Third Edition). Boston: Academic Presspp. 1-9. Hassan AS, S. M. (2018). 'Doppler ultrasound hemodynamic assessment in preeclampsia. ', Med J Babylon, 15, pp. 191-4. Hayashi, M., Kim, S. W., Imanaka-Yoshida, K., Yoshida, T., Abel, E. D., Eliceiri, B., Yang, Y., Ulevitch, R. J. & Lee, J. D. (2004). 'Targeted deletion of BMK1/ERK5 in adult mice perturbs vascular integrity and leads to endothelial failure', J Clin Invest, 113(8), pp. 1138-48. Hayman, R., Warren, A., Brockelsby, J., Johnson, I. & Baker, P. (2000). 'Plasma from women with pre-eclampsia induces an in vitro alteration in the endothelium-dependent behaviour of myometrial resistance arteries', Bjog, 107(1), pp. 108-15. Hayward, C. E., Higgins, L., Cowley, E. J., Greenwood, S. L., Mills, T. A., Sibley, C. P. & Wareing, M. (2013). 'Chorionic plate arterial function is altered in maternal obesity', Placenta, 34(3), pp. 281-7. He, M., Li, F., Yang, M., Fan, Y., Beejadhursing, R., Xie, Y., Zhou, Y. & Deng, D. (2018). 'Impairment of BKca channels in human placental chorionic plate arteries is potentially relevant to the development of preeclampsia', Hypertens Res, 41(2), pp. 126-134. Heagerty, A. M., Heerkens, E. H. & Izzard, A. S. (2010). 'Small artery structure and function in hypertension', J Cell Mol Med, 14(5), pp. 1037-43. Hedberg, A., Mento, P. F., Liu, E. C., Hollander, A. M. & Wilkes, B. M. (1989). 'Evidence for functional thromboxane A2-prostaglandin H2 receptors in human placenta', Am J Physiol, 256(2 Pt 1), pp. E256-63.

280

Hefler, L. A., Reyes, C. A., O'Brien, W. E. & Gregg, A. R. (2001a). 'Perinatal development of endothelial nitric oxide synthase-deficient mice', Biol Reprod, 64(2), pp. 666-73. Hefler, L. A., Tempfer, C. B., Moreno, R. M., O'Brien, W. E. & Gregg, A. R. (2001b). 'Endothelial- derived nitric oxide and angiotensinogen: blood pressure and metabolism during mouse pregnancy', Am J Physiol Regul Integr Comp Physiol, 280(1), pp. R174-82. Henderson, K. K. & Byron, K. L. (2007). 'Vasopressin-induced vasoconstriction: two concentration- dependent signaling pathways', J Appl Physiol (1985), 102(4), pp. 1402-9. Hennessy A., Gillin A. G., Painter D.M., Kirwan P. J., Thompson J.F., Horvath J. S. (1997). Evidene for preeclampsia in a baboon pregnancy with twins. Hypertens. Pregnancy 16, 223-228. Herraiz, S., Pellicer, B., Serra, V., Cauli, O., Cortijo, J., Felipo, V. & Pellicer, A. (2012). 'Sildenafil citrate improves perinatal outcome in fetuses from pre-eclamptic rats', Bjog, 119(11), pp. 1394-402. Hibbard, J. U., Korcarz, C. E., Nendaz, G. G., Lindheimer, M. D., Lang, R. M. & Shroff, S. G. (2005). 'The arterial system in pre-eclampsia and chronic hypertension with superimposed preeclampsia', Bjog, 112(7), pp. 897-903. Hodzic, J., Izetbegovic, S., Muracevic, B., Iriskic, R. & Stimjanin Jovic, H. (2017). 'Nitric oxide biosynthesis during normal pregnancy and pregnancy complicated by preeclampsia', Med Glas (Zenica), 14(2), pp. 211-217. Holden, D. P., Fickling, S. A., Whitley, G. S. & Nussey, S. S. (1998). 'Plasma concentrations of asymmetric dimethylarginine, a natural inhibitor of nitric oxide synthase, in normal pregnancy and preeclampsia', Am J Obstet Gynecol, 178(3), pp. 551-6. Holtzman, C. W., Wiggins, B. S. & Spinler, S. A. (2006). 'Role of P-glycoprotein in statin drug interactions', Pharmacotherapy, 26(11), pp. 1601-7. Honjo, T., Inoue, N., Shiraki, R., Kobayashi, S., Otsui, K., Takahashi, M., Hirata, K., Kawashima, S., Yokozaki, H. & Yokoyama, M. (2006). 'Endothelial urocortin has potent antioxidative properties and is upregulated by inflammatory cytokines and pitavastatin', J Vasc Res, 43(2), pp. 131-8. Horton, M. K., Margolis, A. E., Tang, C. & Wright, R. (2014). 'Neuroimaging is a novel tool to understand the impact of environmental chemicals on neurodevelopment', Curr Opin Pediatr, 26(2), pp. 230-6. Hottinger, D. G., Beebe, D. S., Kozhimannil, T., Prielipp, R. C. & Belani, K. G. (2014). 'Sodium nitroprusside in 2014: A clinical concepts review', J Anaesthesiol Clin Pharmacol, 30(4), pp. 462-71. Huai, J., Yang, Z., Yi, Y. H. & Wang, G. J. (2018). 'Different Effects of Pravastatin on Preeclampsia- like Symptoms in Different Mouse Models', Chin Med J (Engl), 131(4), pp. 461-470. Huang, P. L., Huang, Z., Mashimo, H., Bloch, K. D., Moskowitz, M. A., Bevan, J. A. & Fishman, M. C. (1995). 'Hypertension in mice lacking the gene for endothelial nitric oxide synthase', Nature, 377(6546), pp. 239-42. Hunninghake, D. B. (1992). ' HMG-CoA reductase inhibitors', Current Opinion in Lipidology, 3, pp. 22-28. Hunter, R. A., Storm, W. L., Coneski, P. N. & Schoenfisch, M. H. (2013). 'Inaccuracies of nitric oxide measurement methods in biological media', Anal Chem, 85(3), pp. 1957-63. Hussein, W. & Lafayette, R. A. (2014). 'Renal function in normal and disordered pregnancy', Curr Opin Nephrol Hypertens, 23(1), pp. 46-53. Iftikhar U, Iqbal A, Shakoor S. Relationship between leptin and lipids during pre-eclampsia. J Pak Med Assoc. 2010; 60(6):432–435.

Ignarro, L. J., Harbison, R. G., Wood, K. S. & Kadowitz, P. J. (1986). 'Activation of purified soluble guanylate cyclase by endothelium-derived relaxing factor from intrapulmonary artery and vein: stimulation by acetylcholine, bradykinin and arachidonic acid', J Pharmacol Exp Ther, 237(3), pp. 893-900.

281

Illingworth, D. R., Erkelens, D. W., Keller, U., Thompson, G. R. & Tikkanen, M. J. (1994). 'Defined daily doses in relation to hypolipidaemic efficacy of lovastatin, pravastatin, and simvastatin', Lancet, 343(8912), pp. 1554-5. Iqbal, M., Audette, M. C., Petropoulos, S., Gibb, W. & Matthews, S. G. (2012). 'Placental drug transporters and their role in fetal protection', Placenta, 33(3), pp. 137-42. Itoh, T., Ikebe, M., Kargacin, G. J., Hartshorne, D. J., Kemp, B. E. & Fay, F. S. (1989). 'Effects of modulators of myosin light-chain kinase activity in single smooth muscle cells', Nature, 338(6211), pp. 164-7. Ivankovich, A. D., Miletich, D. J. & Tinker, J. H. (1978). 'Sodium nitroprusside: metabolism and general considerations', Int Anesthesiol Clin, 16(2), pp. 1-29. Izumi, Y., Clifford, D. B. & Zorumski, C. F. (1992). 'Inhibition of long-term potentiation by NMDA- mediated nitric oxide release', Science, 257(5074), pp. 1273-6. Jackson, W. F. (2017). 'Potassium Channels in Regulation of Vascular Smooth Muscle Contraction and Growth', Adv Pharmacol, 78, pp. 89-144. Jadeja, R. N., Rachakonda, V., Bagi, Z. & Khurana, S. (2015). 'Assessing Myogenic Response and Vasoactivity In Resistance Mesenteric Arteries Using Pressure Myography', J Vis Exp, (101), p. e50997. Jansson, T. & Persson, E. (1990). 'Placental transfer of glucose and amino acids in intrauterine growth retardation: studies with substrate analogs in the awake guinea pig', Pediatr Res, 28(3), pp. 203-8. Jauniaux, E., Jurkovic, D., Henriet, Y., Rodesch, F. & Hustin, J. (1991). 'Development of the secondary human yolk sac: correlation of sonographic and anatomical features', Hum Reprod, 6(8), pp. 1160-6. Jo, T., Takauchi, Y., Nakajima, Y., Fukami, K., Kosaka, H. & Terada, N. (1998). 'Maternal or umblical venous levels of nitrite/nitrate during pregnancy and at delivery', In Vivo, 12(5), pp. 523-6. jtowicz, A., Zembala-Szczerba, M., gorzata, Babczyk, D., Ko, odziejczyk-Pietruszka, M., Lewaczy, ska, O. & Huras, H. (2019). 'Early- and Late-Onset Preeclampsia: A Comprehensive Cohort Study of Laboratory and Clinical Findings according to the New ISHHP Criteria %J International Journal of Hypertension', 2019, p. 9. Kanaide, H., Ichiki, T., Nishimura, J. & Hirano, K. (2003). 'Cellular mechanism of vasoconstriction induced by angiotensin II: it remains to be determined', Circ Res, 93(11), pp. 1015-7. Kanasaki, K., Palmsten, K., Sugimoto, H., Ahmad, S., Hamano, Y., Xie, L., Parry, S., Augustin, H. G., Gattone, V. H., Folkman, J., Strauss, J. F. & Kalluri, R. (2008). 'Deficiency in catechol-O- methyltransferase and 2-methoxyoestradiol is associated with pre-eclampsia', Nature, 453(7198), pp. 1117-21. Kane, A. D., Herrera, E. A., Hansell, J. A. & Giussani, D. A. (2012). 'Statin treatment depresses the fetal defence to acute hypoxia via increasing nitric oxide bioavailability', J Physiol, 590(2), pp. 323-34. Kaneda, T., Sasaki, N., Urakawa, N. & Shimizu, K. (2016). 'Endothelium-Dependent and - Independent Vasodilator Effects of Dimethyl Sulfoxide in Rat Aorta', Pharmacology, 97(3- 4), pp. 171-6. Karabucak, B., Walsch, H., Jou, Y.-T., Simchon, S. & Kim, S. (2005). 'The Role of Endothelial Nitric Oxide in the Substance P Induced Vasodilation in Bovine Dental Pulp', Journal of Endodontics, 31(10), pp. 733-736. Karumanchi, S. A. & Levine, R. J. (2010). 'How does smoking reduce the risk of preeclampsia?', Hypertension, 55(5), pp. 1100-1. Karumanchi, S. A., August, P. & Podymow, T. (2010). 'CHAPTER 42 - Renal Complications in Normal Pregnancy', in Floege, J., Johnson, R. J. & Feehally, J. (eds.) Comprehensive Clinical Nephrology (Fourth Edition). Philadelphia: Mosbypp. 504-515. Katsumoto, M., Shingu, T., Kuwashima, R., Nakata, A., Nomura, S. & Chayama, K. (2005). 'Biphasic effect of HMG-CoA reductase inhibitor, pitavastatin, on vascular endothelial cells and angiogenesis', Circ J, 69(12), pp. 1547-55. Kavalipati, N., Shah, J., Ramakrishan, A. & Vasnawala, H. (2015). 'Pleiotropic effects of statins', Indian J Endocrinol Metab, 19(5), pp. 554-62. 282

Kenis, I., Tartakover-Matalon, S., Cherepnin, N., Drucker, L., Fishman, A., Pomeranz, M. & Lishner, M. (2005). 'Simvastatin has deleterious effects on human first trimester placental explants', Hum Reprod, 20(10), pp. 2866-72. Khammy, M. M., Angus, J. A. & Wright, C. E. (2016). 'Vascular reactivity of rabbit isolated renal and femoral resistance arteries in renal wrap hypertension', Eur J Pharmacol, 773, pp. 32- 41. Khong, S. L., Kane, S. C., Brennecke, S. P. & da Silva Costa, F. (2015). 'First-trimester uterine artery Doppler analysis in the prediction of later pregnancy complications', Dis Markers, 2015, p. 679730. Kikuchi, R., Takeshita, K., Uchida, Y., Kondo, M., Cheng, X. W., Nakayama, T., Yamamoto, K., Matsushita, T., Liao, J. K. & Murohara, T. (2011). 'Pitavastatin-induced angiogenesis and arteriogenesis is mediated by Notch1 in a murine hindlimb ischemia model without induction of VEGF', Lab Invest, 91(5), pp. 691-703. Kilkenny, C., Browne, W. J., Cuthill, I. C., Emerson, M. & Altman, D. G. (2012). 'Improving bioscience research reporting: the ARRIVE guidelines for reporting animal research', Osteoarthritis Cartilage, 20(4), pp. 256-60. Kingdom, J. C. & Kaufmann, P. (1997). 'Oxygen and placental villous development: origins of fetal hypoxia', Placenta, 18(8), pp. 613-21; discussion 623-6. Kinlay, S., Schwartz, G. G., Olsson, A. G., Rifai, N., Leslie, S. J., Sasiela, W. J., Szarek, M., Libby, P. & Ganz, P. (2003). 'High-dose atorvastatin enhances the decline in inflammatory markers in patients with acute coronary syndromes in the MIRACL study', Circulation, 108(13), pp. 1560-6. Kinney, M. V., Lawn, J. E., Howson, C. P. & Belizan, J. (2012). '15 Million preterm births annually: what has changed this year?', Reprod Health, 9, p. 28. Kinoshita, H., Watanabe, K., Azma, T., Feng, G. G., Akahori, T., Hayashi, H., Sato, M., Fujiwara, Y. & Wakatsuki, A. (2017). 'Human serum albumin and oxidative stress in preeclamptic women and the mechanism of albumin for stress reduction', Heliyon, 3(8), p. e00369. Kitahara, M., Kanaki, T., Ishii, I. & Saito, Y. (2010). 'Atherosclerosis induced by chronic inhibition of the synthesis of nitric oxide in moderately hypercholesterolaemic rabbits is suppressed by pitavastatin', Br J Pharmacol, 159(7), pp. 1418-28. Kleiner-Assaf, A., Jaffa, A. J. & Elad, D. (1999). 'Hemodynamic model for analysis of Doppler ultrasound indexes of umbilical blood flow', Am J Physiol, 276(6), pp. H2204-14. Knock, G. A. & Poston, L. (1996). 'Bradykinin-mediated relaxation of isolated maternal resistance arteries in normal pregnancy and preeclampsia', Am J Obstet Gynecol, 175(6), pp. 1668- 74. Knot, H. J. & Nelson, M. T. (1998). 'Regulation of arterial diameter and wall [Ca2+] in cerebral arteries of rat by membrane potential and intravascular pressure', J Physiol, 508 ( Pt 1), pp. 199-209. Kobayashi, N., Takeshima, H., Fukushima, H., Koguchi, W., Mamada, Y., Hirata, H., Machida, Y., Shinoda, M., Suzuki, N., Yokotsuka, F., Tabei, K. & Matsuoka, H. (2009). 'Cardioprotective effects of pitavastatin on cardiac performance and remodeling in failing rat hearts', Am J Hypertens, 22(2), pp. 176-82. Koladiya, R. U., Jaggi, A. S., Singh, N. & Sharma, B. K. (2008). 'Ameliorative role of Atorvastatin and Pitavastatin in L-Methionine induced vascular dementia in rats', BMC Pharmacol, 8, p. 14. Kovo, M., Schreiber, L., Ben-Haroush, A., Gold, E., Golan, A. & Bar, J. (2012). 'The placental component in early-onset and late-onset preeclampsia in relation to fetal growth restriction', Prenat Diagn, 32(7), pp. 632-7. Krugner-Higby, L., Luck, M., Hartley, D., Crispen, H. M., Lubach, G. R. & Coe, C. L. (2009). 'High-risk pregnancy in rhesus monkeys (Macaca mulatta): a case of ectopic, abdominal pregnancy with birth of a live, term infant, and a case of gestational diabetes complicated by preeclampsia', J Med Primatol, 38(4), pp. 252-6. Kubes, P., Suzuki, M. & Granger, D. N. (1991). 'Nitric oxide: an endogenous modulator of leukocyte adhesion', Proc Natl Acad Sci U S A, 88(11), pp. 4651-5.

283

Kublickiene, K.-R., Grunewald, C., Lindblom, B. & Nisell, H. (1998). 'Myogenic and Endothelial Properties of Myometrial Resistance Arteries from Women with Preeclampsia', Hypertension in Pregnancy, 17(3), pp. 271-281. Kublickiene, K. R., Kublickas, M., Lindblom, B., Lunell, N. O. & Nisell, H. (1997). 'A comparison of myogenic and endothelial properties of myometrial and omental resistance vessels in late pregnancy', Am J Obstet Gynecol, 176(3), pp. 560-6. Kubo, M., Egashira, K., Inoue, T., Koga, J., Oda, S., Chen, L., Nakano, K., Matoba, T., Kawashima, Y., Hara, K., Tsujimoto, H., Sueishi, K., Tominaga, R. & Sunagawa, K. (2009). 'Therapeutic neovascularization by nanotechnology-mediated cell-selective delivery of pitavastatin into the vascular endothelium', Arterioscler Thromb Vasc Biol, 29(6), pp. 796-801. Kuklina, E. V., Ayala, C. & Callaghan, W. M. (2009). 'Hypertensive disorders and severe obstetric morbidity in the United States', Obstet Gynecol, 113(6), pp. 1299-306. Kulandavelu, S., Qu, D. & Adamson, S. L. (2006). 'Cardiovascular function in mice during normal pregnancy and in the absence of endothelial NO synthase', Hypertension, 47(6), pp. 1175- 82. Kulandavelu, S., Whiteley, K. J., Qu, D., Mu, J., Bainbridge, S. A. & Adamson, S. L. (2012). 'Endothelial nitric oxide synthase deficiency reduces uterine blood flow, spiral artery elongation, and placental oxygenation in pregnant mice', Hypertension, 60(1), pp. 231-8. Kumasawa, K., Ikawa, M., Kidoya, H., Hasuwa, H., Saito-Fujita, T., Morioka, Y., Takakura, N., Kimura, T. & Okabe, M. (2011). 'Pravastatin induces placental growth factor (PGF) and ameliorates preeclampsia in a mouse model', Proc Natl Acad Sci U S A, 108(4), pp. 1451-5. Kureishi, Y., Luo, Z., Shiojima, I., Bialik, A., Fulton, D., Lefer, D. J., Sessa, W. C. & Walsh, K. (2000). 'The HMG-CoA reductase inhibitor simvastatin activates the protein kinase Akt and promotes angiogenesis in normocholesterolemic animals', Nat Med, 6(9), pp. 1004-10. Kusinski, L. C., Baker, P. N., Sibley, C. P. & Wareing, M. (2009). 'In vitro assessment of mouse uterine and fetoplacental vascular function', Reprod Sci, 16(8), pp. 740-8. Kusinski, L. C., Stanley, J. L., Dilworth, M. R., Hirt, C. J., Andersson, I. J., Renshall, L. J., Baker, B. C., Baker, P. N., Sibley, C. P., Wareing, M. & Glazier, J. D. (2012). 'eNOS knockout mouse as a model of fetal growth restriction with an impaired uterine artery function and placental transport phenotype', Am J Physiol Regul Integr Comp Physiol, 303(1), pp. R86-93. Kuzkaya, N., Weissmann, N., Harrison, D. G. & Dikalov, S. (2003). 'Interactions of peroxynitrite, tetrahydrobiopterin, ascorbic acid, and thiols: implications for uncoupling endothelial nitric-oxide synthase', J Biol Chem, 278(25), pp. 22546-54. Kwek, K., Read, M. A., Khong, T. Y., Bisits, A. T. & Walters, W. A. (2001). 'Vasoactive effects of 8- epi-prostaglandin F(2alpha)in isolated human placental conduit and resistance blood vessels in vitro', Placenta, 22(6), pp. 526-33. Lafeber, H. N., Rolph, T. P. & Jones, C. T. (1984). 'Studies on the growth of the fetal guinea pig. The effects of ligation of the uterine artery on organ growth and development', J Dev Physiol, 6(6), pp. 441-59. Lager, S. & Powell, T. L. (2012). 'Regulation of nutrient transport across the placenta', J Pregnancy, 2012, p. 179827. LaMarca, B. B., Bennett, W. A., Alexander, B. T., Cockrell, K. & Granger, J. P. (2005). 'Hypertension produced by reductions in uterine perfusion in the pregnant rat: role of tumor necrosis factor-alpha', Hypertension, 46(4), pp. 1022-5. Landmesser, U., Hornig, B. & Drexler, H. (2004). 'Endothelial function: a critical determinant in atherosclerosis?', Circulation, 109(21 Suppl 1), pp. Ii27-33. LaRosa, J. C., Hunninghake, D., Bush, D., Criqui, M. H., Getz, G. S., Gotto, A. M., Jr., Grundy, S. M., Rakita, L., Robertson, R. M., Weisfeldt, M. L. & et al. (1990). 'The cholesterol facts. A summary of the evidence relating dietary fats, serum cholesterol, and coronary heart disease. A joint statement by the American Heart Association and the National Heart, Lung, and Blood Institute. The Task Force on Cholesterol Issues, American Heart Association', Circulation, 81(5), pp. 1721-33.

284

Lawton, P. F., Lee, M. D., Saunter, C. D., Girkin, J. M., McCarron, J. G. & Wilson, C. (2019). 'VasoTracker, a Low-Cost and Open Source Pressure Myograph System for Vascular Physiology', Front Physiol, 10, p. 99. Le, N. T., Takei, Y., Izawa-Ishizawa, Y., Heo, K. S., Lee, H., Smrcka, A. V., Miller, B. L., Ko, K. A., Ture, S., Morrell, C., Fujiwara, K., Akaike, M. & Abe, J. (2014). 'Identification of activators of ERK5 transcriptional activity by high-throughput screening and the role of endothelial ERK5 in vasoprotective effects induced by statins and antimalarial agents', J Immunol, 193(7), pp. 3803-15. Lean, S. C., Heazell, A. E. P., Dilworth, M. R., Mills, T. A. & Jones, R. L. (2017). 'Placental Dysfunction Underlies Increased Risk of Fetal Growth Restriction and Stillbirth in Advanced Maternal Age Women', Sci Rep, 7(1), p. 9677. Learmont, J. G. & Poston, L. (1996). 'Nitric oxide is involved in flow-induced dilation of isolated human small fetoplacental arteries', Am J Obstet Gynecol, 174(2), pp. 583-8. Lecarpentier, E. & Tsatsaris, V. (2016). 'Angiogenic balance (sFlt-1/PlGF) and preeclampsia', Ann Endocrinol (Paris), 77(2), pp. 97-100. Lefkou, E., Mamopoulos, A., Dagklis, T., Vosnakis, C., Rousso, D. & Girardi, G. (2016). 'Pravastatin improves pregnancy outcomes in obstetric antiphospholipid syndrome refractory to antithrombotic therapy', J Clin Invest, 126(8), pp. 2933-40. Leggio, M., Lombardi, M., Caldarone, E., Severi, P., D'Emidio, S., Armeni, M., Bravi, V., Bendini, M. G. & Mazza, A. (2017). 'The relationship between obesity and hypertension: an updated comprehensive overview on vicious twins', Hypertens Res, 40(12), pp. 947-963. Levine, L. (2003). 'Statins stimulate arachidonic acid release and prostaglandin I2 production in rat liver cells', Lipids Health Dis, 2, p. 1. Levine, R. J., Maynard, S. E., Qian, C., Lim, K. H., England, L. J., Yu, K. F., Schisterman, E. F., Thadhani, R., Sachs, B. P., Epstein, F. H., Sibai, B. M., Sukhatme, V. P. & Karumanchi, S. A. (2004). 'Circulating angiogenic factors and the risk of preeclampsia', N Engl J Med, 350(7), pp. 672-83. Levy, J. H. (2005). 'Management of systemic and pulmonary hypertension', Tex Heart Inst J, 32(4), pp. 467-71. Li, Y., Li, X. H. & Yuan, H. (2012). 'Angiotensin II type-2 receptor-specific effects on the cardiovascular system', Cardiovasc Diagn Ther, 2(1), pp. 56-62. Li, C., Raikwar, N. S., Santillan, M. K., Santillan, D. A. & Thomas, C. P. (2015). 'Aspirin inhibits expression of sFLT1 from human cytotrophoblasts induced by hypoxia, via cyclooxygenase 1', Placenta, 36(4), pp. 446-53. Li, J., LaMarca, B. & Reckelhoff, J. F. (2012). 'A model of preeclampsia in rats: the reduced uterine perfusion pressure (RUPP) model', Am J Physiol Heart Circ Physiol, 303(1), pp. H1-8. Lian, I. A., Loset, M., Mundal, S. B., Fenstad, M. H., Johnson, M. P., Eide, I. P., Bjorge, L., Freed, K. A., Moses, E. K. & Austgulen, R. (2011). 'Increased endoplasmic reticulum stress in decidual tissue from pregnancies complicated by fetal growth restriction with and without pre-eclampsia', Placenta, 32(11), pp. 823-9. Liao, J. K. (2005). 'Effects of statins on 3-hydroxy-3-methylglutaryl coenzyme a reductase inhibition beyond low-density lipoprotein cholesterol', Am J Cardiol, 96(5a), pp. 24f-33f. Liao, J. K. & Laufs, U. (2005). 'Pleiotropic effects of statins', Annu Rev Pharmacol Toxicol, 45, pp. 89-118. Lin, L. Y., Huang, C. C., Chen, J. S., Wu, T. C., Leu, H. B., Huang, P. H., Chang, T. T., Lin, S. J. & Chen, J. W. (2014). 'Effects of pitavastatin versus atorvastatin on the peripheral endothelial progenitor cells and vascular endothelial growth factor in high-risk patients: a pilot prospective, double-blind, randomized study', Cardiovasc Diabetol, 13, p. 111. Liu, A., Wen, S. W., Bottomley, J., Walker, M. C. & Smith, G. (2009). 'Utilization of health care services of pregnant women complicated by preeclampsia in Ontario', Hypertens Pregnancy, 28(1), pp. 76-84. Liu, H., Bao, P., Li, L., Wang, Y., Xu, C., Deng, M., Zhang, J. & Zhao, X. (2017). 'Pitavastatin nanoparticle-engineered endothelial progenitor cells repair injured vessels', Sci Rep, 7(1), p. 18067. 285

Liu, H. S., Chu, T. Y., Yu, M. H., Chang, Y. K., Ko, C. S. & Chao, C. F. (1998). 'Thromboxane and prostacyclin in maternal and fetal circulation in pre-eclampsia', Int J Gynaecol Obstet, 63(1), pp. 1-6. Liu, L., Johnson, H. L., Cousens, S., Perin, J., Scott, S., Lawn, J. E., Rudan, I., Campbell, H., Cibulskis, R., Li, M., Mathers, C. & Black, R. E. (2012). 'Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000', Lancet, 379(9832), pp. 2151-61. Llinas, M. T., Alexander, B. T., Seedek, M., Abram, S. R., Crell, A. & Granger, J. P. (2002). 'Enhanced thromboxane synthesis during chronic reductions in uterine perfusion pressure in pregnant rats', Am J Hypertens, 15(9), pp. 793-7. Lopez-Jaramillo, P., Arenas, W. D., Garcia, R. G., Rincon, M. Y. & Lopez, M. (2008). 'The role of the L-arginine-nitric oxide pathway in preeclampsia', Ther Adv Cardiovasc Dis, 2(4), pp. 261- 75. Lopez-Mendez, M. A., Martinez-Gaytan, V., Cortes-Flores, R., Ramos-Gonzalez, R. M., Ochoa- Torres, M. A., Garza-Veloz, I., Martinez-Acuña, M. I., Badillo-Almaraz, J. I. & Martinez- Fierro, M. L. (2013). 'Doppler ultrasound evaluation in preeclampsia', BMC research notes, 6, pp. 477-477. Lorquet, S., Pequeux, C., Munaut, C. & Foidart, J. M. (2010). 'Aetiology and physiopathology of preeclampsia and related forms', Acta Clin Belg, 65(4), pp. 237-41. Low, J. A., Galbraith, R. S., Muir, D. W., Killen, H. L., Pater, E. A. & Karchmar, E. J. (1985). 'The relationship between perinatal hypoxia and newborn encephalopathy', Am J Obstet Gynecol, 152(3), pp. 256-60. Lu, X. & Kassab, G. S. (2011). 'Assessment of endothelial function of large, medium, and small vessels: a unified myograph', Am J Physiol Heart Circ Physiol, 300(1), pp. H94-h100. Luo, Z., Zhang, Y., Gu, J., Feng, P. & Wang, Y. (2015). 'Pharmacokinetic Properties of Single- and Multiple-Dose Pitavastatin Calcium Tablets in Healthy Chinese Volunteers', Curr Ther Res Clin Exp, 77, pp. 52-7. Lyall F, B. M. P. R., Vercruysse L, Hanssens M, Van Assche F A . (2007). 'In: Lyall F, Belfort M, eds. Pre-eclampsia; ' Etiology and Clinical Practice, Cambridge:Cambridge University Press, pp. 3–19. Lyall, F., Greer, I. A., Young, A. & Myatt, L. (1996). 'Nitric oxide concentrations are increased in the feto-placental circulation in intrauterine growth restriction', Placenta, 17(2-3), pp. 165-8. Lyall, F., Young, A. & Greer, I. A. (1995). 'Nitric oxide concentrations are increased in the fetoplacental circulation in preeclampsia', Am J Obstet Gynecol, 173(3 Pt 1), pp. 714-8. Ma, F. X. & Han, Z. C. (2005). 'Statins, nitric oxide and neovascularization', Cardiovasc Drug Rev, 23(4), pp. 281-92. MacIntyre, D. E., Westwick, J. & Williams, T. J. (1978). 'Comparison of the effects of prostaglandin analogues on rabbit platelets, rabbit isolated vascular tissues and rabbit skin microvasculature [proceedings]', Br J Pharmacol, 62(3), pp. 418p-420p. Magee, L. A. (2001). 'Drugs in pregnancy. Antihypertensives', Best Pract Res Clin Obstet Gynaecol, 15(6), pp. 827-45. Magee, L. A., Singer, J. & von Dadelszen, P. (2015). 'Less-tight versus tight control of hypertension in pregnancy', N Engl J Med, 372(24), pp. 2367-8. Mandala, M. & Osol, G. (2012). 'Physiological remodelling of the maternal uterine circulation during pregnancy', Basic Clin Pharmacol Toxicol, 110(1), pp. 12-8. Mannaerts, D., Faes, E., Cos, P., Briede, J. J., Gyselaers, W., Cornette, J., Gorbanev, Y., Bogaerts, A., Spaanderman, M., Van Craenenbroeck, E. & Jacquemyn, Y. (2018). 'Oxidative stress in healthy pregnancy and preeclampsia is linked to chronic inflammation, iron status and vascular function', PLoS One, 13(9), p. e0202919. Maria Dahl Andersen, A. K. O. A. C. S. D. E. F. R. M. M. H. V. P. G. O. & Michael, P. (2018). Animal Models of Fetal Medicine and Obstetrics, Experimental Animal Models of Human Diseases - An Effective Therapeutic Strategy: IntechOpen.

286

Mason CW, Buhimschi IA, Buhimschi CS, Dong Y, Weiner CP, Swaan PW. ATP-binding cassette transporter expression in human placenta as a function of pregnancy condition. Drug Metab Dispos. 39 (2011) 1000-1007. Matsushima, S., Maeda, K., Kondo, C., Hirano, M., Sasaki, M., Suzuki, H. & Sugiyama, Y. (2005). 'Identification of the hepatic efflux transporters of organic anions using double- transfected Madin-Darby canine kidney II cells expressing human organic anion- transporting polypeptide 1B1 (OATP1B1)/multidrug resistance-associated protein 2, OATP1B1/multidrug resistance 1, and OATP1B1/breast cancer resistance protein', J Pharmacol Exp Ther, 314(3), pp. 1059-67. Matsuzawa, Y., Funahashi, T., Kihara, S. & Shimomura, I. (2004). 'Adiponectin and metabolic syndrome', Arterioscler Thromb Vasc Biol, 24(1), pp. 29-33. Matthews, T. J. & Hamilton, B. E. (2009). 'Delayed childbearing: more women are having their first child later in life', NCHS Data Brief, (21), pp. 1-8. Maynard, S. E., Min, J. Y., Merchan, J., Lim, K. H., Li, J., Mondal, S., Libermann, T. A., Morgan, J. P., Sellke, F. W., Stillman, I. E., Epstein, F. H., Sukhatme, V. P. & Karumanchi, S. A. (2003). 'Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia', J Clin Invest, 111(5), pp. 649- 58. McAdam, B. F., Catella-Lawson, F., Mardini, I. A., Kapoor, S., Lawson, J. A. & FitzGerald, G. A. (1999). 'Systemic biosynthesis of prostacyclin by cyclooxygenase (COX)-2: the human pharmacology of a selective inhibitor of COX-2', Proc Natl Acad Sci U S A, 96(1), pp. 272-7. McCarthy, A. L., Taylor, P., Graves, J., Raju, S. K. & Poston, L. (1994a). 'Endothelium-dependent relaxation of human resistance arteries in pregnancy', Am J Obstet Gynecol, 171(5), pp. 1309-15. McCarthy, A. L., Woolfson, R. G., Evans, B. J., Davies, D. R., Raju, S. K. & Poston, L. (1994b). 'Functional characteristics of small placental arteries', Am J Obstet Gynecol, 170(3), pp. 945-51. McCarthy, A. L., Woolfson, R. G., Raju, S. K. & Poston, L. (1993). 'Abnormal endothelial cell function of resistance arteries from women with preeclampsia', Am J Obstet Gynecol, 168(4), pp. 1323-30. McCarthy, F. P., Kingdom, J. C., Kenny, L. C. & Walsh, S. K. (2011). 'Animal models of preeclampsia; uses and limitations', Placenta, 32(6), pp. 413-9. McCormick, M. C., Bernbaum, J. C., Eisenberg, J. M., Kustra, S. L. & Finnegan, E. (1991). 'Costs incurred by parents of very low birth weight infants after the initial neonatal hospitalization', Pediatrics, 88(3), pp. 533-41. Melchiorre, K., Sutherland, G. R., Liberati, M. & Thilaganathan, B. (2012). 'Maternal cardiovascular impairment in pregnancies complicated by severe fetal growth restriction', Hypertension, 60(2), pp. 437-43. Meng, W., Ma, J., Ayata, C., Hara, H., Huang, P. L., Fishman, M. C. & Moskowitz, M. A. (1996). 'ACh dilates pial arterioles in endothelial and neuronal NOS knockout mice by NO-dependent mechanisms', Am J Physiol, 271(3 Pt 2), pp. H1145-50. Mills, E. J., Rachlis, B., Wu, P., Devereaux, P. J., Arora, P. & Perri, D. (2008). 'Primary prevention of cardiovascular mortality and events with statin treatments: a network meta-analysis involving more than 65,000 patients', J Am Coll Cardiol, 52(22), pp. 1769-81. Mills, T. A., Wareing, M., Bugg, G. J., Greenwood, S. L. & Baker, P. N. (2005). 'Chorionic plate artery function and Doppler indices in normal pregnancy and intrauterine growth restriction', Eur J Clin Invest, 35(12), pp. 758-64. Minsker, D. H., MacDonald, J. S., Robertson, R. T. & Bokelman, D. L. (1983). 'Mevalonate supplementation in pregnant rats suppresses the teratogenicity of mevinolinic acid, an inhibitor of 3-hydroxy-3-methylglutaryl-coenzyme a reductase', Teratology, 28(3), pp. 449-56. Mishra, N., Nugent, W. H., Mahavadi, S. & Walsh, S. W. (2011). 'Mechanisms of enhanced vascular reactivity in preeclampsia', Hypertension, 58(5), pp. 867-73.

287

Mistry, H. D., McCallum, L. A., Kurlak, L. O., Greenwood, I. A., Broughton Pipkin, F. & Tribe, R. M. (2011). 'Novel expression and regulation of voltage-dependent potassium channels in placentas from women with preeclampsia', Hypertension, 58(3), pp. 497-504. Mitsuhashi, T., Uemoto, R., Ishikawa, K., Yoshida, S., Ikeda, Y., Yagi, S., Matsumoto, T., Akaike, M. & Aihara, K. I. (2018). 'Endothelial Nitric Oxide Synthase-Independent Pleiotropic Effects of Pitavastatin Against Atherogenesis and Limb Ischemia in Mice', J Atheroscler Thromb, 25(1), pp. 65-80. Mizugaki, M., Ishizawa, F., Yamazaki, T. & Hishinuma, T. (2000). 'Epigallocatechin gallate increase the prostacyclin production of bovine aortic endothelial cells', Prostaglandins Other Lipid Mediat, 62(2), pp. 157-64. Mo, L. Y., Bascom, P. A., Ritchie, K. & McCowan, L. M. (1988). 'A transmission line modelling approach to the interpretation of uterine Doppler waveforms', Ultrasound Med Biol, 14(5), pp. 365-76. Molnar, M. & Hertelendy, F. (1992). 'N omega-nitro-L-arginine, an inhibitor of nitric oxide synthesis, increases blood pressure in rats and reverses the pregnancy-induced refractoriness to vasopressor agents', Am J Obstet Gynecol, 166(5), pp. 1560-7. Molnar, M., Suto, T., Toth, T. & Hertelendy, F. (1994). 'Prolonged blockade of nitric oxide synthesis in gravid rats produces sustained hypertension, proteinuria, thrombocytopenia, and intrauterine growth retardation', Am J Obstet Gynecol, 170(5 Pt 1), pp. 1458-66. Moncada, S., Palmer, R. M. & Higgs, E. A. (1989). 'Biosynthesis of nitric oxide from L-arginine. A pathway for the regulation of cell function and communication', Biochem Pharmacol, 38(11), pp. 1709-15. Mone, F., Mcauliffe, F. & Ong, S. (2015). 'The clinical application of Doppler ultrasound in obstetrics', The Obstetrician & Gynaecologist, 17, pp. 13–19. Mori, T., Watanabe, K., Iwasaki, A., Kimura, C., Matsushita, H., Shinohara, K. & Wakatsuki, A. (2014). 'Differences in vascular reactivity between pregnant women with chronic hypertension and preeclampsia', Hypertens Res, 37(2), pp. 145-50. Morton, J. S., Levasseur, J., Ganguly, E., Quon, A., Kirschenman, R., Dyck, J. R. B., Fraser, G. M. & Davidge, S. T. (2019). 'Characterisation of the Selective Reduced Uteroplacental Perfusion (sRUPP) Model of Preeclampsia', Sci Rep, 9(1), p. 9565. Mraiche, F., Cena, J., Das, D. & Vollrath, B. (2005). 'Effects of statins on vascular function of endothelin-1', Br J Pharmacol, 144(5), pp. 715-26. Mukai, Y., Shimokawa, H., Matoba, T., Hiroki, J., Kunihiro, I., Fujiki, T. & Takeshita, A. (2003). 'Acute vasodilator effects of HMG-CoA reductase inhibitors: involvement of PI3- kinase/Akt pathway and Kv channels', J Cardiovasc Pharmacol, 42(1), pp. 118-24. Mulvany, M. J. & Aalkjaer, C. (1990). 'Structure and function of small arteries', Physiol Rev, 70(4), pp. 921-61. Mulvany, M. J. & Halpern, W. (1977). 'Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats', Circ Res, 41(1), pp. 19-26. Mulvany, M. J. & Nyborg, N. (1980). 'An increased calcium sensitivity of mesenteric resistance vessels in young and adult spontaneously hypertensive rats', Br J Pharmacol, 71(2), pp. 585-96. Myatt, L. (1992). 'Control of vascular resistance in the human placenta', Placenta, 13(4), pp. 329- 41. Myatt, L., Eis, A. L., Brockman, D. E., Greer, I. A. & Lyall, F. (1997). 'Endothelial nitric oxide synthase in placental villous tissue from normal, pre-eclamptic and intrauterine growth restricted pregnancies', Hum Reprod, 12(1), pp. 167-72. Myatt, L. & Webster, R. P. (2009). 'Vascular biology of preeclampsia', Journal of Thrombosis and Haemostasis, 7(3), pp. 375-384. Nagatomo, T., Rashid, M., Abul Muntasir, H. & Komiyama, T. (2004). 'Functions of 5-HT2A receptor and its antagonists in the cardiovascular system', Pharmacol Ther, 104(1), pp. 59- 81.

288

Nanovskaya, T. N., Patrikeeva, S. L., Paul, J., Costantine, M. M., Hankins, G. D. & Ahmed, M. S. (2013). 'Transplacental transfer and distribution of pravastatin', Am J Obstet Gynecol, 209(4), pp. 373.e1-5. Napso, T., Yong, H. E. J., Lopez-Tello, J. & Sferruzzi-Perri, A. N. (2018). 'The Role of Placental Hormones in Mediating Maternal Adaptations to Support Pregnancy and Lactation', Front Physiol, 9, p. 1091. Nardozza, L. M. M., Zamarian, A. C. P. & Araujo Junior, E. (2017). 'New Definition of Fetal Growth Restriction: Consensus Regarding a Major Obstetric Complication', Rev Bras Ginecol Obstet, 39(7), pp. 315-316. Needleman, P. & Isakson, P. C. (1997). 'The discovery and function of COX-2', J Rheumatol Suppl, 49, pp. 6-8. Nelson SM, Matthews P, Poston L. Maternal metabolism and obesity: modifiable determinants of pregnancy outcome. Hum Reprod Update 2010;16(03):255-275. Nemenoff, R. A. (1998). 'Vasopressin signaling pathways in vascular smooth muscle', Front Biosci, 3, pp. d194-207. Ni, Y., Meyer, M. & Osol, G. (1997). 'Gestation increases nitric oxide-mediated vasodilation in rat uterine arteries', Am J Obstet Gynecol, 176(4), pp. 856-64. NICE (2011). Hypertension in pregnancy: diagnosis and management. . Available at: http://www.nice.org.uk (Accessed: 19 October). NICE (2019a). Hypertension in Pregnancy. Available at: http://pathways.nice.org.uk/pathways/hypertension-in-pregnancy (Accessed: 27.9.2019). NICE (2019b). Prescribing in pregnancy. Available at: https://bnf.nice.org.uk/guidance/prescribing-in-pregnancy.html (Accessed: 25.9.2019). Nicosia, S., Oliva, D., Noe, M. A., Corsini, A., Folco, G. C. & Fumagalli, R. (1987). 'PGI2 receptors in vasculature and platelets: 5Z-carbacyclin discriminates between them', Adv Prostaglandin Thromboxane Leukot Res, 17a, pp. 474-8. Niromanesh, S., Shirazi, M., Eftekhariyazdi, M. & Mortazavi, F. (2017). 'Comparison of umbilical artery Doppler and non-stress test in assessment of fetal well-being in gestational diabetes mellitus: A prospective cohort study', Electron Physician, 9(12), pp. 6087-6093. Nobunaga, T., Tokugawa, Y., Hashimoto, K., Kimura, T., Matsuzaki, N., Nitta, Y., Fujita, T., Kidoguchi, K. I., Azuma, C. & Saji, F. (1996). 'Plasma nitric oxide levels in pregnant patients with preeclampsia and essential hypertension', Gynecol Obstet Invest, 41(3), pp. 189-93. Noris, M., Todeschini, M., Cassis, P., Pasta, F., Cappellini, A., Bonazzola, S., Macconi, D., Maucci, R., Porrati, F., Benigni, A., Picciolo, C. & Remuzzi, G. (2004). 'L-arginine depletion in preeclampsia orients nitric oxide synthase toward oxidant species', Hypertension, 43(3), pp. 614-22. Norris, L. A., Higgins, J. R., Darling, M. R., Walshe, J. J. & Bonnar, J. (1999). 'Nitric oxide in the uteroplacental, fetoplacental, and peripheral circulations in preeclampsia', Obstet Gynecol, 93(6), pp. 958-63. O'Dell, T. J., Hawkins, R. D., Kandel, E. R. & Arancio, O. (1991). 'Tests of the roles of two diffusible substances in long-term potentiation: evidence for nitric oxide as a possible early retrograde messenger', Proc Natl Acad Sci U S A, 88(24), pp. 11285-9. O'Driscoll, G., Green, D. & Taylor, R. R. (1997). 'Simvastatin, an HMG-coenzyme A reductase inhibitor, improves endothelial function within 1 month', Circulation, 95(5), pp. 1126-31. O'Gorman, N., Nicolaides, K. H. & Poon, L. C. (2016). 'The use of ultrasound and other markers for early detection of preeclampsia', Womens Health (Lond), 12(2), pp. 199-207. O’Byrne, P. M. (2009). 'Chapter 23 - Prostanoids', in Barnes, P. J., Drazen, J. M., Rennard, S. I. & Thomson, N. C. (eds.) Asthma and COPD (Second Edition). Oxford: Academic Presspp. 275- 282. Ogura K Low-density lipoprotein particle diameter in normal pregnancy and preeclampsia .J Atheroscler Thromb 2002;9: 42–47. Olofsson, P., Laurini, R. N. & Marsal, K. (1993). 'A high uterine artery pulsatility index reflects a defective development of placental bed spiral arteries in pregnancies complicated by

289

hypertension and fetal growth retardation', Eur J Obstet Gynecol Reprod Biol, 49(3), pp. 161-8. Onda, K., Tong, S., Beard, S., Binder, N., Muto, M., Senadheera, S. N., Parry, L., Dilworth, M., Renshall, L., Brownfoot, F., Hastie, R., Tuohey, L., Palmer, K., Hirano, T., Ikawa, M., Kaitu'u- Lino, T. & Hannan, N. J. (2017). 'Proton Pump Inhibitors Decrease Soluble fms-Like Tyrosine Kinase-1 and Soluble Endoglin Secretion, Decrease Hypertension, and Rescue Endothelial Dysfunction', Hypertension, 69(3), pp. 457-468. Ong, S. S., Crocker, I. P., Warren, A. Y. & Baker, P. N. (2002). 'Functional characteristics of chorionic plate placental arteries from normal pregnant women and women with preeclampsia', Hypertens Pregnancy, 21(3), pp. 175-83. Ong, S. S., Moore, R. J., Warren, A. Y., Crocker, I. P., Fulford, J., Tyler, D. J., Gowland, P. A. & Baker, P. N. (2003). 'Myometrial and placental artery reactivity alone cannot explain reduced placental perfusion in pre-eclampsia and intrauterine growth restriction', Bjog, 110(10), pp. 909-15. Ortiz, P. A. & Garvin, J. L. (2003). 'Cardiovascular and renal control in NOS-deficient mouse models', Am J Physiol Regul Integr Comp Physiol, 284(3), pp. R628-38. Osol, G., Barron, C., Gokina, N. & Mandala, M. (2009). 'Inhibition of nitric oxide synthases abrogates pregnancy-induced uterine vascular expansive remodeling', J Vasc Res, 46(5), pp. 478-86. Osol, G. & Mandala, M. (2009). 'Maternal uterine vascular remodeling during pregnancy', Physiology (Bethesda), 24, pp. 58-71. Pacurari, M., Kafoury, R., Tchounwou, P. B. & Ndebele, K. (2014). 'The Renin-Angiotensin- aldosterone system in vascular inflammation and remodeling', Int J Inflam, 2014, p. 689360. Pallares, P. & Gonzalez-Bulnes, A. (2010). 'The effect of embryo and maternal genotypes on prolificacy, intrauterine growth retardation and postnatal development of Nos3-knockout mice', Reprod Biol, 10(3), pp. 241-8. Palmer A. E., London W. T., Sly D. L., Rice J. M. (1979). Spontaneous preeclamptic toxemia of pregnancy in the patas monkey (Erythroceus paras). Lab. Anim. Sci. 29, 102-106. Palmer, R. M., Ashton, D. S. & Moncada, S. (1988). 'Vascular endothelial cells synthesize nitric oxide from L-arginine', Nature, 333(6174), pp. 664-6. Panza, J. A., Casino, P. R., Kilcoyne, C. M. & Quyyumi, A. A. (1993). 'Role of endothelium-derived nitric oxide in the abnormal endothelium-dependent vascular relaxation of patients with essential hypertension', Circulation, 87(5), pp. 1468-74. Panza, J. A., Quyyumi, A. A., Brush, J. E., Jr. & Epstein, S. E. (1990). 'Abnormal endothelium- dependent vascular relaxation in patients with essential hypertension', N Engl J Med, 323(1), pp. 22-7. Papageorghiou, A. T., Yu, C. K., Bindra, R., Pandis, G. & Nicolaides, K. H. (2001). 'Multicenter screening for pre-eclampsia and fetal growth restriction by transvaginal uterine artery Doppler at 23 weeks of gestation', Ultrasound Obstet Gynecol, 18(5), pp. 441-9. Pascoal, I. F., Lindheimer, M. D., Nalbantian-Brandt, C. & Umans, J. G. (1998). 'Preeclampsia selectively impairs endothelium-dependent relaxation and leads to oscillatory activity in small omental arteries', J Clin Invest, 101(2), pp. 464-70. Pasupathy, D., McCowan, L. M. E., Poston, L., Kenny, L. C., Dekker, G. A., North, R. A. & consortium, S. (2012). 'Perinatal Outcomes in Large Infants Using Customised Birthweight Centiles and Conventional Measures of High Birthweight', 26(6), pp. 543-552. Pelat, M., Dessy, C., Massion, P., Desager, J. P., Feron, O. & Balligand, J. L. (2003). 'Rosuvastatin decreases caveolin-1 and improves nitric oxide-dependent heart rate and blood pressure variability in apolipoprotein E-/- mice in vivo', Circulation, 107(19), pp. 2480-6. Perinatal, I. (2019). Fetal Growth. Available at: https://www.perinatal.org.uk/FetalGrowth/fetalgrowth.aspx (Accessed: 20.09.2019). Petla, L. T., Chikkala, R., Ratnakar, K. S., Kodati, V. & Sritharan, V. (2013). 'Biomarkers for the management of pre-eclampsia in pregnant women', Indian J Med Res, 138, pp. 60-7.

290

Petrou, S., Henderson, J., Bracewell, M., Hockley, C., Wolke, D. & Marlow, N. (2006). 'Pushing the boundaries of viability: the economic impact of extreme preterm birth', Early Hum Dev, 82(2), pp. 77-84. Phillips, J. K., Janowiak, M., Badger, G. J. & Bernstein, I. M. (2010). 'Evidence for distinct preterm and term phenotypes of preeclampsia', J Matern Fetal Neonatal Med, 23(7), pp. 622-6. Phillips, T. J., Mootz, J. R. K. & Reed, C. (2016). 'Chapter Three - Identification of Treatment Targets in a Genetic Mouse Model of Voluntary Methamphetamine Drinking', in Bell, R. L. & Rahman, S. (eds.) International Review of Neurobiology: Academic Presspp. 39-85. Phipps, E., Prasanna, D., Brima, W. & Jim, B. (2016). 'Preeclampsia: Updates in Pathogenesis, Definitions, and Guidelines', Clin J Am Soc Nephrol, 11(6), pp. 1102-13. Phipps, E. A., Thadhani, R., Benzing, T. & Karumanchi, S. A. (2019). 'Pre-eclampsia: pathogenesis, novel diagnostics and therapies', Nat Rev Nephrol, 15(5), pp. 275-289. Pijnenborg, R., Anthony, J., Davey, D. A., Rees, A., Tiltman, A., Vercruysse, L. & van Assche, A. (1991). 'Placental bed spiral arteries in the hypertensive disorders of pregnancy', Br J Obstet Gynaecol, 98(7), pp. 648-55. Pijnenborg, R., Vercruysse, L. & Brosens, I. (2011). 'Deep placentation', Best Pract Res Clin Obstet Gynaecol, 25(3), pp. 273-85. Pijnenborg, R., Vercruysse, L. & Hanssens, M. (2006). 'The uterine spiral arteries in human pregnancy: facts and controversies', Placenta, 27(9-10), pp. 939-58. Podjarny, E., Bernheim, J., Katz, B., Green, J., Mekler, J. & Bursztyn, M. (1998). 'Chronic exogenous hyperinsulinemia in pregnancy: a rat model of pregnancy-induced hypertension', J Am Soc Nephrol, 9(1), pp. 9-13. Podjarny, E., Bernheim, J., Rathaus, M., Pomeranz, A., Tovbin, D., Shapira, J. & Bernheim, J. (1992). 'Adriamycin nephropathy: a model to study effects of pregnancy on renal disease in rats', Am J Physiol, 263(4 Pt 2), pp. F711-5. Podymow, T. & August, P. (2008). 'Update on the use of antihypertensive drugs in pregnancy', Hypertension, 51(4), pp. 960-9. Poston, L., Briley, A. L., Seed, P. T., Kelly, F. J. & Shennan, A. H. (2006). 'Vitamin C and vitamin E in pregnant women at risk for pre-eclampsia (VIP trial): randomised placebo-controlled trial', Lancet, 367(9517), pp. 1145-54. Poston, L., Caleyachetty, R., Cnattingius, S., Corvalan, C., Uauy, R., Herring, S. & Gillman, M. W. (2016). 'Preconceptional and maternal obesity: epidemiology and health consequences', Lancet Diabetes Endocrinol, 4(12), pp. 1025-1036. Poudel, R., Stanley, J. L., Rueda-Clausen, C. F., Andersson, I. J., Sibley, C. P., Davidge, S. T. & Baker, P. N. (2013). 'Effects of resveratrol in pregnancy using murine models with reduced blood supply to the uterus', PLoS One, 8(5), p. e64401. PrabhuDas, M., Bonney, E., Caron, K., Dey, S., Erlebacher, A., Fazleabas, A., Fisher, S., Golos, T., Matzuk, M., McCune, J. M., Mor, G., Schulz, L., Soares, M., Spencer, T., Strominger, J., Way, S. S. & Yoshinaga, K. (2015). 'Immune mechanisms at the maternal-fetal interface: perspectives and challenges', Nat Immunol, 16(4), pp. 328-34. Practice, A. C. o. O. (2002). 'Practice bulletin #33: diagnosis and management of preeclampsia and eclampsia', Obstetrics & Gynecology, 99(1), pp. 159-167. Prast, H. & Philippu, A. (2001). 'Nitric oxide as modulator of neuronal function', Prog Neurobiol, 64(1), pp. 51-68. Predoi, C. G., Grigoriu, C., Vladescu, R. & Mihart, A. E. (2015). 'Placental damages in preeclampsia - from ultrasound images to histopathological findings', J Med Life, 8 Spec Issue, pp. 62-5. Purdy R.E., M. D. L. (1989). Serotonin-induced Vasoconstriction and Contractile Synergism with Noradrenaline: Role of α-Adrenoceptors. . Satellite Symposia of the IUPHAR 10th International Congress of Pharmacology: Palgrave Macmillan, London. Quinlan, T. R., Laubach, V., Zhou, N. & Johns, R. A. (1998). 'Alterations in nitric oxide synthase isoform expression in NOS knockout mice exposed to normoxia or hypoxia', Chest, 114(1 Suppl), pp. 53s-55s. Raiteri, M., Arnaboldi, L., McGeady, P., Gelb, M. H., Verri, D., Tagliabue, C., Quarato, P., Ferraboschi, P., Santaniello, E., Paoletti, R., Fumagalli, R. & Corsini, A. (1997). 291

'Pharmacological control of the mevalonate pathway: effect on arterial smooth muscle cell proliferation', J Pharmacol Exp Ther, 281(3), pp. 1144-53. Ramesar, S. V., Mackraj, I., Gathiram, P. & Moodley, J. (2011). 'Sildenafil citrate decreases sFlt-1 and sEng in pregnant l-NAME treated Sprague-Dawley rats', Eur J Obstet Gynecol Reprod Biol, 157(2), pp. 136-40. Ramkumar, S., Raghunath, A. & Raghunath, S. (2016). 'Statin Therapy: Review of Safety and Potential Side Effects', Acta Cardiol Sin, 32(6), pp. 631-639. Ramma, W. & Ahmed, A. (2014). 'Therapeutic potential of statins and the induction of heme oxygenase-1 in preeclampsia', J Reprod Immunol, 101-102, pp. 153-60. Rana, S., Lemoine, E., Granger, J. & Karumanchi, S. A. (2019). 'Preeclampsia', Circ Res, 124(7), pp. 1094-1112. Rapoport, R. M., Draznin, M. B. & Murad, F. (1983). 'Endothelium-dependent relaxation in rat aorta may be mediated through cyclic GMP-dependent protein phosphorylation', Nature, 306(5939), pp. 174-6. Ray, K. K., Cannon, C. P., McCabe, C. H., Cairns, R., Tonkin, A. M., Sacks, F. M., Jackson, G. & Braunwald, E. (2005). 'Early and late benefits of high-dose atorvastatin in patients with acute coronary syndromes: results from the PROVE IT-TIMI 22 trial', J Am Coll Cardiol, 46(8), pp. 1405-10. Read, M. A., Boura, A. L. & Walters, W. A. (1993). 'Vascular actions of purines in the foetal circulation of the human placenta', Br J Pharmacol, 110(1), pp. 454-60. Reagan-Shaw, S., Nihal, M. & Ahmad, N. (2008). 'Dose translation from animal to human studies revisited', Faseb j, 22(3), pp. 659-61. Redman, C. W. & Sargent, I. L. (2005). 'Latest advances in understanding preeclampsia', Science, 308(5728), pp. 1592-4. Redman, C. W., Sargent, I. L. & Staff, A. C. (2014). 'IFPA Senior Award Lecture: making sense of pre-eclampsia - two placental causes of preeclampsia?', Placenta, 35 Suppl, pp. S20-5. Reilly, F. D. & Russell, P. T. (1977). 'Neurohistochemical evidence supporting an absence of adrenergic and cholinergic innervation in the human placenta and umbilical cord', Anat Rec, 188(3), pp. 277-86. Renna, N. F., de Las Heras, N. & Miatello, R. M. (2013). 'Pathophysiology of vascular remodeling in hypertension', Int J Hypertens, 2013, p. 808353. Renshall, L. J., Morgan, H. L., Moens, H., Cansfield, D., Finn-Sell, S. L., Tropea, T., Cottrell, E. C., Greenwood, S., Sibley, C. P., Wareing, M. & Dilworth, M. R. (2018). 'Melatonin Increases Fetal Weight in Wild-Type Mice but Not in Mouse Models of Fetal Growth Restriction', Front Physiol, 9, p. 1141. Roberge, S., Bujold, E. & Nicolaides, K. H. (2018). 'Aspirin for the prevention of preterm and term preeclampsia: systematic review and metaanalysis', Am J Obstet Gynecol, 218(3), pp. 287- 293.e1. Roberts, J. M. (2000). 'Preeclampsia: what we know and what we do not know', Semin Perinatol, 24(1), pp. 24-8. Roberts, J. M. & Cooper, D. W. (2001). 'Pathogenesis and genetics of pre-eclampsia', Lancet, 357(9249), pp. 53-6. Roberts JM, Pearson G, CutlerJ, Lindheimer M: Summary of the NHLBI Working Group on Research on Hypertension During Pregnancy. Hypertension 2003;41:437–445. Roberts, J. M. & Escudero, C. (2012). 'The placenta in preeclampsia', Pregnancy Hypertens, 2(2), pp. 72-83. Roccella, E. J. (2000). 'Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy', American Journal of Obstetrics and Gynecology, 183(1), pp. s1-s22. Rodriguez, I. & Gonzalez, M. (2014). 'Physiological mechanisms of vascular response induced by shear stress and effect of exercise in systemic and placental circulation', Front Pharmacol, 5, p. 209. Rolnik, D. L., Wright, D., Poon, L. C., O’Gorman, N., Syngelaki, A., de Paco Matallana, C., Akolekar, R., Cicero, S., Janga, D., Singh, M., Molina, F. S., Persico, N., Jani, J. C., Plasencia, W., 292

Papaioannou, G., Tenenbaum-Gavish, K., Meiri, H., Gizurarson, S., Maclagan, K. & Nicolaides, K. H. (2017). 'Aspirin versus Placebo in Pregnancies at High Risk for Preterm Preeclampsia', New England Journal of Medicine, 377(7), pp. 613-622. Rosenfeld, C. R. (2001). 'Mechanisms regulating angiotensin II responsiveness by the uteroplacental circulation', Am J Physiol Regul Integr Comp Physiol, 281(4), pp. R1025-40. Rossoni, L. V., Wareing, M., Wenceslau, C. F., Al-Abri, M., Cobb, C. & Austin, C. (2011). 'Acute simvastatin increases endothelial nitric oxide synthase phosphorylation via AMP-activated protein kinase and reduces contractility of isolated rat mesenteric resistance arteries', Clin Sci (Lond), 121(10), pp. 449-58. Rucker, D. & Dhamoon, A. S. (2019). 'Physiology, Thromboxane A2', StatPearls. Treasure Island (FL): StatPearls Publishing

StatPearls Publishing LLC. S Jurado, K. S., C Marceliano, VSouza and I Vieira (2019). 'Maternal and Fetal Complications Due to Decreased Nitric Oxide Synthesis during Gestation', in Abduljabbar, H. (ed.) Complications of Pregnancy: IntechOpen. Saad, A. F., Kechichian, T., Yin, H., Sbrana, E., Longo, M., Wen, M., Tamayo, E., Hankins, G. D., Saade, G. R. & Costantine, M. M. (2014). 'Effects of pravastatin on angiogenic and placental hypoxic imbalance in a mouse model of preeclampsia', Reprod Sci, 21(1), pp. 138-45. Sachdeva, P., Patel, B. G. & Patel, B. K. (2009). 'Drug use in pregnancy; a point to ponder!', Indian journal of pharmaceutical sciences, 71(1), pp. 1-7. Sahu S, Abraham R, Vedavalli R, et al. Study of lipid profile, lipid peroxidation and vitamin E in pregnancy induced hypertension. Indian J Physiol Pharmacol. 2009;53(4):365–369.

Saito, Y. (2011). 'Pitavastatin: an overview', Atheroscler Suppl, 12(3), pp. 271-6.

Sakabe, K., Fukuda, N., Fukuda, Y., Wakayama, K., Nada, T., Morishita, S., Shinohara, H. & Tamura, Y. (2008). 'Comparisons of short- and intermediate-term effects of pitavastatin versus atorvastatin on lipid profiles, fibrinolytic parameter, and endothelial function', Int J Cardiol, 125(1), pp. 136-8.

Samangaya, R. A., Mires, G., Shennan, A., Skillern, L., Howe, D., McLeod, A. & Baker, P. N. (2009). 'A randomised, double-blinded, placebo-controlled study of the phosphodiesterase type 5 inhibitor sildenafil for the treatment of preeclampsia', Hypertens Pregnancy, 28(4), pp. 369-82. Samanta, S., Jain, K., Bhardwaj, N., Jain, V., Singh, P. M., Samanta, S., Singla, V. & Saha, R. (2018). 'Labor analgesia in parturients of fetal growth restriction having raised umbilical Doppler vascular indices', J Anaesthesiol Clin Pharmacol, 34(1), pp. 11-17. Sand, A., Andersson, E. & Fried, G. (2006). 'Nitric oxide donors mediate vasodilation in human placental arteries partly through a direct effect on potassium channels', Placenta, 27(2-3), pp. 181-90. Sand, A. E., Andersson, E. & Fried, G. (2002). 'Effects of nitric oxide donors and inhibitors of nitric oxide signalling on endothelin- and serotonin-induced contractions in human placental arteries', Acta Physiol Scand, 174(3), pp. 217-23. Sandoo, A., van Zanten, J. J., Metsios, G. S., Carroll, D. & Kitas, G. D. (2010). 'The endothelium and its role in regulating vascular tone', Open Cardiovasc Med J, 4, pp. 302-12. Sankaralingam, S., Arenas, I. A., Lalu, M. M. & Davidge, S. T. (2006). 'Preeclampsia: current understanding of the molecular basis of vascular dysfunction', Expert Rev Mol Med, 8(3), pp. 1-20. Sattar N, Greer IA. Pregnancy complications and maternal cardiovascular risk: opportunities for intervention and screening? BMJ 2002;325(7356):157-160. Doi: 10.1136/bmj.325.7356.157 .

Saunders, M. (2009). 'Transplacental transport of nanomaterials', Wiley Interdiscip Rev Nanomed Nanobiotechnol, 1(6), pp. 671-84.

293

Say, L., Chou, D., Gemmill, A., Tuncalp, O., Moller, A. B., Daniels, J., Gulmezoglu, A. M., Temmerman, M. & Alkema, L. (2014). 'Global causes of maternal death: a WHO systematic analysis', Lancet Glob Health, 2(6), pp. e323-33. Sazonova, I. Y., Hoda, M. N., Zemskova, M. A. & Hess, D. C. (2012). 'Effect of Endothelial Nitric Oxide Synthase Genotype on Outcomes after Experimental Thromboembolic Stroke', 26(1_supplement), pp. 398.7-398.7. Schachter, M. (2005). 'Chemical, pharmacokinetic and pharmacodynamic properties of statins: an update', Fundam Clin Pharmacol, 19(1), pp. 117-25. Schlager G, L. J., Carrithers JA. (1989). 'Characteristics of the inbred hypertensive mouse strains', FASEB Journal, 3(A1315). Schulz, E., Dopheide, J., Schuhmacher, S., Thomas, S. R., Chen, K., Daiber, A., Wenzel, P., Munzel, T. & Keaney, J. F., Jr. (2008). 'Suppression of the JNK pathway by induction of a metabolic stress response prevents vascular injury and dysfunction', Circulation, 118(13), pp. 1347- 57. Schwartz, G. G., Olsson, A. G., Ezekowitz, M. D., Ganz, P., Oliver, M. F., Waters, D., Zeiher, A., Chaitman, B. R., Leslie, S. & Stern, T. (2001). 'Effects of atorvastatin on early recurrent ischemic events in acute coronary syndromes: the MIRACL study: a randomized controlled trial', Jama, 285(13), pp. 1711-8. Schwartz, R. B., Feske, S. K., Polak, J. F., DeGirolami, U., Iaia, A., Beckner, K. M., Bravo, S. M., Klufas, R. A., Chai, R. Y. & Repke, J. T. (2000). 'Preeclampsia-eclampsia: clinical and neuroradiographic correlates and insights into the pathogenesis of hypertensive encephalopathy', Radiology, 217(2), pp. 371-6. Scotland, R. S., Madhani, M., Chauhan, S., Moncada, S., Andresen, J., Nilsson, H., Hobbs, A. J. & Ahluwalia, A. (2005). 'Investigation of vascular responses in endothelial nitric oxide synthase/cyclooxygenase-1 double-knockout mice: key role for endothelium-derived hyperpolarizing factor in the regulation of blood pressure in vivo', Circulation, 111(6), pp. 796-803. Scully, C. (2014). '25 - Age and gender issues', in Scully, C. (ed.) Scully's Medical Problems in Dentistry (Seventh Edition). Oxford: Churchill Livingstonepp. 627-651. Seely, E. W. & Ecker, J. (2014). 'Chronic hypertension in pregnancy', Circulation, 129(11), pp. 1254- 61. Seligman, S. P., Buyon, J. P., Clancy, R. M., Young, B. K. & Abramson, S. B. (1994). 'The role of nitric oxide in the pathogenesis of preeclampsia', Am J Obstet Gynecol, 171(4), pp. 944-8. Sen, S., Rao, R. & Chaudhuri, G. (2013). 'Endothelial cell function in utero-placental circulation physiology and pathophysiology', Curr Vasc Pharmacol, 11(5), pp. 730-6. Serrano, M. A., Macias, R. I., Briz, O., Monte, M. J., Blazquez, A. G., Williamson, C., Kubitz, R. & Marin, J. J. (2007). 'Expression in human trophoblast and choriocarcinoma cell lines, BeWo, Jeg-3 and JAr of genes involved in the hepatobiliary-like excretory function of the placenta', Placenta, 28(2-3), pp. 107-17. Sessa, W. C. (2004). 'eNOS at a glance', Journal of Cell Science, 117(12), p. 2427. Shaamash, A. H., Elsnosy, E. D., Makhlouf, A. M., Zakhari, M. M., Ibrahim, O. A. & HM, E. L.-d. (2000). 'Maternal and fetal serum nitric oxide (NO) concentrations in normal pregnancy, pre-eclampsia and eclampsia', Int J Gynaecol Obstet, 68(3), pp. 207-14. Shaamash, A. H., Elsonosy, E. D., Zakhari, M. M., Radwan, S. H. & El-Dien, H. M. (2001). 'Placental nitric oxide synthase (NOS) activity and nitric oxide (NO) production in normal pregnancy, pre-eclampsia and eclampsia', Int J Gynaecol Obstet, 72(2), pp. 127-33. Shahid, M. & Buys, E. S. (2013). 'Assessing murine resistance artery function using pressure myography', J Vis Exp, (76). Sharma, D., Shastri, S. & Sharma, P. (2016). 'Intrauterine Growth Restriction: Antenatal and Postnatal Aspects', Clin Med Insights Pediatr, 10, pp. 67-83. Shaw, S. M., Fildes, J. E., Yonan, N. & Williams, S. G. (2009). 'Pleiotropic effects and cholesterol- lowering therapy', Cardiology, 112(1), pp. 4-12.

294

Shesely, E. G., Maeda, N., Kim, H. S., Desai, K. M., Krege, J. H., Laubach, V. E., Sherman, P. A., Sessa, W. C. & Smithies, O. (1996). 'Elevated blood pressures in mice lacking endothelial nitric oxide synthase', Proc Natl Acad Sci U S A, 93(23), pp. 13176-81. Shinozaki, K., Nishio, Y., Ayajiki, K., Yoshida, Y., Masada, M., Kashiwagi, A. & Okamura, T. (2007). 'Pitavastatin restores vascular dysfunction in insulin-resistant state by inhibiting NAD(P)H oxidase activity and uncoupled endothelial nitric oxide synthase-dependent superoxide production', J Cardiovasc Pharmacol, 49(3), pp. 122-30. Shizukuda, Y., Tang, S., Yokota, R. & Ware, J. A. (1999). 'Vascular endothelial growth factor- induced endothelial cell migration and proliferation depend on a nitric oxide-mediated decrease in protein kinase Cdelta activity', Circ Res, 85(3), pp. 247-56. Sibai, B. M. (2002). 'Chronic hypertension in pregnancy', Obstet Gynecol, 100(2), pp. 369-77. Sibai, B. M., Mirro, R., Chesney, C. M. & Leffler, C. (1989). 'Low-dose aspirin in pregnancy', Obstet Gynecol, 74(4), pp. 551-7. Sibley, C., Glazier, J. & D'Souza, S. (1997). 'Placental transporter activity and expression in relation to fetal growth', Exp Physiol, 82(2), pp. 389-402. Sibley, C. P. (2017). 'Treating the dysfunctional placenta', J Endocrinol, 234(2), pp. R81-r97. Sikkema, J. M., van Rijn, B. B., Franx, A., Bruinse, H. W., de Roos, R., Stroes, E. S. & van Faassen, E. E. (2001). 'Placental superoxide is increased in pre-eclampsia', Placenta, 22(4), pp. 304-8. Singh, J., Ahmed, A. & Girardi, G. (2011). 'Role of complement component C1q in the onset of preeclampsia in mice', Hypertension, 58(4), pp. 716-24. Sladek, S. M., Magness, R. R. & Conrad, K. P. (1997). 'Nitric oxide and pregnancy', Am J Physiol, 272(2 Pt 2), pp. R441-63. Smarason, A. K., Allman, K. G., Young, D. & Redman, C. W. (1997). 'Elevated levels of serum nitrate, a stable end product of nitric oxide, in women with pre-eclampsia', Br J Obstet Gynaecol, 104(5), pp. 538-43. Soma-Pillay, P., Nelson-Piercy, C., Tolppanen, H. & Mebazaa, A. (2016). 'Physiological changes in pregnancy', Cardiovasc J Afr, 27(2), pp. 89-94. Sorriento, D., Trimarco, B. & Iaccarino, G. (2011). 'Adrenergic mechanism in the control of endothelial function', Transl Med UniSa, 1, pp. 213-28. Sotiriadis, A., Hernandez-Andrade, E., da Silva Costa, F., Ghi, T., Glanc, P., Khalil, A., Martins, W. P., Odibo, A. O., Papageorghiou, A. T., Salomon, L. J. & Thilaganathan, B. (2019). 'ISUOG Practice Guidelines: role of ultrasound in screening for and follow-up of pre-eclampsia', Ultrasound Obstet Gynecol, 53(1), pp. 7-22. Spargo, B., Mc, C. C. & Winemiller, R. (1959). 'Glomerular capillary endotheliosis in toxemia of pregnancy', Arch Pathol, 68, pp. 593-9. Spiers, A. & Padmanabhan, N. (2005). 'A guide to wire myography', Methods Mol Med, 108, pp. 91-104. Spivack, M. (1943). 'On the presence or absence of nerves in the umbilical vessels of man and guinea pig', Anatomical Record, 85, pp. 85-109. Spracklen, C. N., Smith, C. J., Saftlas, A. F., Robinson, J. G. & Ryckman, K. K. (2014). 'Maternal hyperlipidemia and the risk of preeclampsia: a meta-analysis', Am J Epidemiol, 180(4), pp. 346-58. Spradley, F. T., Tan, A. Y., Joo, W. S., Daniels, G., Kussie, P., Karumanchi, S. A. & Granger, J. P. (2016). 'Placental Growth Factor Administration Abolishes Placental Ischemia-Induced Hypertension', Hypertension, 67(4), pp. 740-7. Sprague, B., Chesler, N. C. & Magness, R. R. (2010). 'Shear stress regulation of nitric oxide production in uterine and placental artery endothelial cells: experimental studies and hemodynamic models of shear stresses on endothelial cells', Int J Dev Biol, 54(2-3), pp. 331-9. Staff, A.C., Dechend, R., and Redman, C.W. Review: Preeclampsia, acute atherosis of the spiral arteries and future cardiovascular disease: two new hypotheses. Placenta. 2013; 34: S73– S78. Staff, A.C., Johnsen, G.M., Dechend, R., and Redman, C.W. Preeclampsia and uteroplacental acute atherosis: immune and inflammatory factors. J Reprod Immunol. 2014; 101-102: 120–126. 295

Stancu, C. & Sima, A. (2001). 'Statins: mechanism of action and effects', J Cell Mol Med, 5(4), pp. 378-87.

Stanley, J. L., Andersson, I. J., Hirt, C. J., Moore, L., Dilworth, M. R., Chade, A. R., Sibley, C. P., Davidge, S. T. & Baker, P. N. (2012a). 'Effect of the anti-oxidant tempol on fetal growth in a mouse model of fetal growth restriction', Biol Reprod, 87(1), pp. 25, 1-8.

Stanley, J. L., Andersson, I. J., Poudel, R., Rueda-Clausen, C. F., Sibley, C. P., Davidge, S. T. & Baker, P. N. (2012b). 'Sildenafil citrate rescues fetal growth in the catechol-O-methyl transferase knockout mouse model', Hypertension, 59(5), pp. 1021-8. Stanley, J. L., Sulek, K., Andersson, I. J., Davidge, S. T., Kenny, L. C., Sibley, C. P., Mandal, R., Wishart, D. S., Broadhurst, D. I. & Baker, P. N. (2015). 'Sildenafil Therapy Normalizes the Aberrant Metabolomic Profile in the Comt(-/-) Mouse Model of Preeclampsia/Fetal Growth Restriction', Sci Rep, 5, p. 18241. Staud, F., Cerveny, L. & Ceckova, M. (2012). 'Pharmacotherapy in pregnancy; effect of ABC and SLC transporters on drug transport across the placenta and fetal drug exposure', J Drug Target, 20(9), pp. 736-63. Steudel, W., Ichinose, F., Huang, P. L., Hurford, W. E., Jones, R. C., Bevan, J. A., Fishman, M. C. & Zapol, W. M. (1997). 'Pulmonary vasoconstriction and hypertension in mice with targeted disruption of the endothelial nitric oxide synthase (NOS 3) gene', Circ Res, 81(1), pp. 34- 41. Stuart, M. J., Clark, D. A., Sunderji, S. G., Allen, J. B., Yambo, T., Elrad, H. & Slott, J. H. (1981). 'Decrease prostacyclin production: a characteristic of chronic placental insufficiency syndromes', Lancet, 1(8230), pp. 1126-8. Subramaniam, V. (2007). 'Seasonal variation in the incidence of preeclampsia and eclampsia in tropical climatic conditions', BMC Womens Health, 7, p. 18. Sunderland, N. S., Thomson, S. E., Heffernan, S. J., Lim, S., Thompson, J., Ogle, R., McKenzie, P., Kirwan, P. J., Makris, A. & Hennessy, A. (2011). 'Tumor necrosis factor alpha induces a model of preeclampsia in pregnant baboons (Papio hamadryas)', Cytokine, 56(2), pp. 192- 9. Suvas, S. (2017). 'Role of Substance P Neuropeptide in Inflammation, Wound Healing, and Tissue Homeostasis', The Journal of Immunology, 199(5), p. 1543. Suzuki, Y., Kajikuri, J., Suzumori, K. & Itoh, T. (2000). 'Mechanisms underlying the reduced endothelium-dependent relaxation in human omental resistance artery in pre-eclampsia', J Physiol, 527 Pt 1, pp. 163-74. Sybertz, E. J., Vander Vliet, G. & Baum, T. (1983). 'Analysis of the vasoconstrictor responses to potassium depolarization and norepinephrine and their antagonism by differing classes of vasodilators in the perfused rat hindquarters', J Pharmacol Exp Ther, 227(3), pp. 621-6. Taguchi, N., Rubin, E. T., Hosokawa, A., Choi, J., Ying, A. Y., Moretti, M. E., Koren, G. & Ito, S. (2008). 'Prenatal exposure to HMG-CoA reductase inhibitors: effects on fetal and neonatal outcomes', Reprod Toxicol, 26(2), pp. 175-7. Takemoto, M. & Liao, J. K. (2001). 'Pleiotropic effects of 3-hydroxy-3-methylglutaryl coenzyme a reductase inhibitors', Arterioscler Thromb Vasc Biol, 21(11), pp. 1712-9. Takimoto, E., Ishida, J., Sugiyama, F., Horiguchi, H., Murakami, K. & Fukamizu, A. (1996). 'Hypertension induced in pregnant mice by placental renin and maternal angiotensinogen', Science, 274(5289), pp. 995-8. Talabi, E. a. (2014). 'Vascular Effects of Histamine', Niger. J. Physiol. Sci, 29, pp. 007 –010. Tartakover-Matalon, S., Cherepnin, N., Kuchuk, M., Drucker, L., Kenis, I., Fishman, A., Pomeranz, M. & Lishner, M. (2007). 'Impaired migration of trophoblast cells caused by simvastatin is associated with decreased membrane IGF-I receptor, MMP2 activity and HSP27 expression', Hum Reprod, 22(4), pp. 1161-7. Thadhani, R., Hagmann, H., Schaarschmidt, W., Roth, B., Cingoez, T., Karumanchi, S. A., Wenger, J., Lucchesi, K. J., Tamez, H., Lindner, T., Fridman, A., Thome, U., Kribs, A., Danner, M., Hamacher, S., Mallmann, P., Stepan, H. & Benzing, T. (2016). 'Removal of Soluble Fms-Like

296

Tyrosine Kinase-1 by Dextran Sulfate Apheresis in Preeclampsia', J Am Soc Nephrol, 27(3), pp. 903-13. Thorp, A. A. & Schlaich, M. P. (2015). 'Relevance of Sympathetic Nervous System Activation in Obesity and Metabolic Syndrome', J Diabetes Res, 2015, p. 341583. Timor-Tritsch, I. E. & Monteagudo, A. (2009). 'Chapter 3 - Scanning techniques in obstetrics and gynaecology', in Wladimiroff, J. W. & Eik-Nes, S. H. (eds.) Ultrasound in Obstetrics and Gynaecology. Edinburgh: Elsevierpp. 33-56. Tkachenko, O., Shchekochikhin, D. & Schrier, R. W. (2014). 'Hormones and hemodynamics in pregnancy', Int J Endocrinol Metab, 12(2), p. e14098. Todd, P. A. & Goa, K. L. (1990). 'Simvastatin. A review of its pharmacological properties and therapeutic potential in hypercholesterolaemia', Drugs, 40(4), pp. 583-607. Tomimatsu, T., Mimura, K., Endo, M., Kumasawa, K. & Kimura, T. (2017). 'Pathophysiology of preeclampsia: an angiogenic imbalance and long-lasting systemic vascular dysfunction', Hypertens Res, 40(4), pp. 305-310. Tousoulis, D., Kampoli, A. M., Tentolouris, C., Papageorgiou, N. & Stefanadis, C. (2012). 'The role of nitric oxide on endothelial function', Curr Vasc Pharmacol, 10(1), pp. 4-18. Townsend, R., O'Brien, P. & Khalil, A. (2016). 'Current best practice in the management of hypertensive disorders in pregnancy', Integr Blood Press Control, 9, pp. 79-94. Tranquilli, A. L., Dekker, G., Magee, L., Roberts, J., Sibai, B. M., Steyn, W., Zeeman, G. G. & Brown, M. A. (2014). 'The classification, diagnosis and management of the hypertensive disorders of pregnancy: A revised statement from the ISSHP', Pregnancy Hypertens, 4(2), pp. 97- 104. Trapani, A., Jr., Goncalves, L. F., Trapani, T. F., Vieira, S., Pires, M. & Pires, M. M. (2016). 'Perinatal and Hemodynamic Evaluation of Sildenafil Citrate for Preeclampsia Treatment: A Randomized Controlled Trial', Obstet Gynecol, 128(2), pp. 253-9. Treasure, C. B., Klein, J. L., Weintraub, W. S., Talley, J. D., Stillabower, M. E., Kosinski, A. S., Zhang, J., Boccuzzi, S. J., Cedarholm, J. C. & Alexander, R. W. (1995). 'Beneficial effects of cholesterol-lowering therapy on the coronary endothelium in patients with coronary artery disease', N Engl J Med, 332(8), pp. 481-7. Trudinger, B. J., Giles, W. B., Cook, C. M., Bombardieri, J. & Collins, L. (1985). 'Fetal umbilical artery flow velocity waveforms and placental resistance: clinical significance', Br J Obstet Gynaecol, 92(1), pp. 23-30. Trudinger, B. J. & Ishikawa, K. (1990). 'Use of Doppler ultrasound in the high-risk pregnancy', Clin Diagn Ultrasound, 26, pp. 119-37. Turpin, C. A., Sakyi, S. A., Owiredu, W. K., Ephraim, R. K. & Anto, E. O. (2015). 'Association between adverse pregnancy outcome and imbalance in angiogenic regulators and oxidative stress biomarkers in gestational hypertension and preeclampsia', BMC Pregnancy Childbirth, 15, p. 189. Vallance, P., Collier, J. & Moncada, S. (1989). 'Effects of endothelium-derived nitric oxide on peripheral arteriolar tone in man', Lancet, 2(8670), pp. 997-1000. van der Heijden, O. W., Essers, Y. P., Fazzi, G., Peeters, L. L., De Mey, J. G. & van Eys, G. J. (2005). 'Uterine artery remodeling and reproductive performance are impaired in endothelial nitric oxide synthase-deficient mice', Biol Reprod, 72(5), pp. 1161-8. Van Vliet, B. N. & Chafe, L. L. (2007). 'Maternal endothelial nitric oxide synthase genotype influences offspring blood pressure and activity in mice', Hypertension, 49(3), pp. 556-62. VanBavel, E., Mooij, T., Giezeman, M. J. & Spaan, J. A. (1990). 'Cannulation and continuous cross- sectional area measurement of small blood vessels', J Pharmacol Methods, 24(3), pp. 219- 27. Vaughan, O. R. & Fowden, A. L. (2016). 'Placental metabolism: substrate requirements and the response to stress', Reprod Domest Anim, 51 Suppl 2, pp. 25-35. Vedernikov, Y., Saade, G. R. & Garfield, R. E. (1999). 'Vascular reactivity in preeclampsia', Semin Perinatol, 23(1), pp. 34-44.

297

Vedernikov, Y. P., Belfort, M. A., Saade, G. R. & Moise, K. J., Jr. (1995). 'Pre-eclampsia does not alter the response to endothelin-1 in human omental artery', J Cardiovasc Pharmacol, 26 Suppl 3, pp. S233-5. Veerareddy, S., Cooke, C. L., Baker, P. N. & Davidge, S. T. (2002). 'Vascular adaptations to pregnancy in mice: effects on myogenic tone', Am J Physiol Heart Circ Physiol, 283(6), pp. H2226-33. Venkatesha, S., Toporsian, M., Lam, C., Hanai, J., Mammoto, T., Kim, Y. M., Bdolah, Y., Lim, K. H., Yuan, H. T., Libermann, T. A., Stillman, I. E., Roberts, D., D'Amore, P. A., Epstein, F. H., Sellke, F. W., Romero, R., Sukhatme, V. P., Letarte, M. & Karumanchi, S. A. (2006). 'Soluble endoglin contributes to the pathogenesis of preeclampsia', Nat Med, 12(6), pp. 642-9. Veteriankey (2016). 'https://veteriankey.com/wire-myography-to-study-vascular-tone-and- vascular-structure-of-isolated-mouse-arteries/ (Accessed: 23.4.19)'. Vinnars, M. T., Nasiell, J., Ghazi, S., Westgren, M. & Papadogiannakis, N. (2011). 'The severity of clinical manifestations in preeclampsia correlates with the amount of placental infarction', Acta Obstet Gynecol Scand, 90(1), pp. 19-25. Wagner, A. H., Kohler, T., Ruckschloss, U., Just, I. & Hecker, M. (2000). 'Improvement of nitric oxide-dependent vasodilatation by HMG-CoA reductase inhibitors through attenuation of endothelial superoxide anion formation', Arterioscler Thromb Vasc Biol, 20(1), pp. 61-9. Waldron, G. J., Ding, H., Lovren, F., Kubes, P. & Triggle, C. R. (1999). 'Acetylcholine-induced relaxation of peripheral arteries isolated from mice lacking endothelial nitric oxide synthase', Br J Pharmacol, 128(3), pp. 653-8. Wallenburg, H. C., Dekker, G. A., Makovitz, J. W. & Rotmans, P. (1986). 'Low-dose aspirin prevents pregnancy-induced hypertension and pre-eclampsia in angiotensin-sensitive primigravidae', Lancet, 1(8471), pp. 1-3. Wallis, A. B., Saftlas, A. F., Hsia, J. & Atrash, H. K. (2008). 'Secular trends in the rates of preeclampsia, eclampsia, and gestational hypertension, United States, 1987-2004', Am J Hypertens, 21(5), pp. 521-6. Walsh, S. K., English, F. A., Johns, E. J. & Kenny, L. C. (2009). 'Plasma-mediated vascular dysfunction in the reduced uterine perfusion pressure model of preeclampsia: a microvascular characterization', Hypertension, 54(2), pp. 345-51. Walsh, S. W. (1985). 'Preeclampsia: an imbalance in placental prostacyclin and thromboxane production', Am J Obstet Gynecol, 152(3), pp. 335-40. Wang, J., Tokoro, T., Matsui, K., Higa, S. & Kitajima, I. (2005). 'Pitavastatin at low dose activates endothelial nitric oxide synthase through PI3K-AKT pathway in endothelial cells', Life Sci, 76(19), pp. 2257-68. Wang, J., Xu, Z., Kitajima, I. & Wang, Z. (2008). 'Effects of different statins on endothelial nitric oxide synthase and AKT phosphorylation in endothelial cells', Int J Cardiol, 127(1), pp. 33- 9. Wang, Y. & Zhao, S. (2010). 'Integrated Systems Physiology: from Molecules to Function to Disease', Vascular Biology of the Placenta. San Rafael (CA): Morgan & Claypool Life Sciences

Copyright (c) 2010 by Morgan & Claypool Life Sciences. Wang Y, Z. S. (2010). Vascular Biology of the Placenta. San Rafael (CA): Morgan & Claypool Life Sciences;. Wareing, M. & Baker, P. N. (2004). 'Vasoconstriction of small arteries isolated from the human placental chorionic plate in normal and compromised pregnancy', Hypertens Pregnancy, 23(3), pp. 237-46. Wareing, M., Crocker, I. P., Warren, A. Y., Taggart, M. J. & Baker, P. N. (2002). 'Characterization of Small Arteries Isolated From the Human Placental Chorionic Plate', Placenta, 23(5), pp. 400-409. Wareing, M., Greenwood, S. L., Fyfe, G. K. & Baker, P. N. (2006a). 'Reactivity of human placental chorionic plate vessels from pregnancies complicated by intrauterine growth restriction (IUGR)', Biol Reprod, 75(4), pp. 518-23.

298

Wareing, M., Myers, J. E., O'Hara, M., Kenny, L. C., Taggart, M. J., Skillern, L., Machin, I. & Baker, P. N. (2006b). 'Phosphodiesterase-5 inhibitors and omental and placental small artery function in normal pregnancy and pre-eclampsia', Eur J Obstet Gynecol Reprod Biol, 127(1), pp. 41-9. Wareing, M., Myers, J. E., O'Hara, M., Kenny, L. C., Warren, A. Y., Taggart, M. J., Skillern, L., Machin, I. & Baker, P. N. (2004). 'Effects of a phosphodiesterase-5 (PDE5) inhibitor on endothelium-dependent relaxation of myometrial small arteries', Am J Obstet Gynecol, 190(5), pp. 1283-90. Wassmann, S., Laufs, U., Baumer, A. T., Muller, K., Ahlbory, K., Linz, W., Itter, G., Rosen, R., Bohm, M. & Nickenig, G. (2001). 'HMG-CoA reductase inhibitors improve endothelial dysfunction in normocholesterolemic hypertension via reduced production of reactive oxygen species', Hypertension, 37(6), pp. 1450-7. Webb, R. C. (2003). 'SMOOTH MUSCLE CONTRACTION AND RELAXATION', Advances in Physiology Education, 27(4), pp. 201-206. Webster, L. M., Myers, J. E., Nelson-Piercy, C., Harding, K., Cruickshank, J. K., Watt-Coote, I., Khalil, A., Wiesender, C., Seed, P. T. & Chappell, L. C. (2017). 'Labetalol Versus Nifedipine as Antihypertensive Treatment for Chronic Hypertension in Pregnancy: A Randomized Controlled Trial', Hypertension, 70(5), pp. 915-922. Webster, L. M., Myers, J. E., Nelson-Piercy, C., Mills, C., Watt-Coote, I., Khalil, A., Seed, P. T., Cruickshank, J. K. & Chappell, L. C. (2019). 'Longitudinal changes in vascular function parameters in pregnant women with chronic hypertension and association with adverse outcome: a cohort study', Ultrasound Obstet Gynecol, 53(5), pp. 638-648. Wei, J., Liu, C. X., Gong, T. T., Wu, Q. J. & Wu, L. (2015). 'Cigarette smoking during pregnancy and preeclampsia risk: a systematic review and meta-analysis of prospective studies', Oncotarget, 6(41), pp. 43667-78. Wei, X., Zhang, Y., Yin, B., Wen, J., Cheng, J. & Fu, X. (2018). 'The expression and function of KCNQ potassium channels in human chorionic plate arteries from women with normal pregnancies and pre-eclampsia', PLoS One, 13(3), p. e0192122.

Weiler, J., Tong, S. & Palmer, K. R. (2011). 'Is fetal growth restriction associated with a more severe maternal phenotype in the setting of early onset pre-eclampsia? A retrospective study', PLoS One, 6(10), p. e26937. Weissgerber, T. L., Milic, N. M., Milin-Lazovic, J. S. & Garovic, V. D. (2016). 'Impaired Flow- Mediated Dilation Before, During, and After Preeclampsia: A Systematic Review and Meta-Analysis', Hypertension, 67(2), pp. 415-23. Whitley, G. S. & Cartwright, J. E. (2010). 'Cellular and molecular regulation of spiral artery remodelling: lessons from the cardiovascular field', Placenta, 31(6), pp. 465-74. Wight, E., Kung, C. F., Moreau, P., Takase, H., Bersinger, N. A. & Luscher, T. F. (2000). 'Aging, serum estradiol levels, and pregnancy differentially affect vascular reactivity of the rat uterine artery', J Soc Gynecol Investig, 7(2), pp. 106-13. Wild R,Weedin EA,Wilson D. Dyslipidemia in pregnancy. Cardiol Clin 2015;33(02):209-215. Wilhelm, D., Mansmann, U., Neudeck, H., Matejevic, D., Vetter, K. & Graf, R. (1999). 'Increase of segments of elastic-type blood vessel walls in fetal placental stem villi during preeclampsia at term', Anat Embryol (Berl), 200(6), pp. 597-605. Wink, D. A., Kasprzak, K. S., Maragos, C. M., Elespuru, R. K., Misra, M., Dunams, T. M., Cebula, T. A., Koch, W. H., Andrews, A. W., Allen, J. S. & et al. (1991). 'DNA deaminating ability and genotoxicity of nitric oxide and its progenitors', Science, 254(5034), pp. 1001-3. Winterfeld, U., Allignol, A., Panchaud, A., Rothuizen, L. E., Merlob, P., Cuppers- Maarschalkerweerd, B., Vial, T., Stephens, S., Clementi, M., De Santis, M., Pistelli, A., Berlin, M., Eleftheriou, G., Manakova, E. & Buclin, T. (2013). 'Pregnancy outcome following maternal exposure to statins: a multicentre prospective study', Bjog, 120(4), pp. 463-71.

299

Withagen, M. I., Wallenburg, H. C., Steegers, E. A., Hop, W. C. & Visser, W. (2005). 'Morbidity and development in childhood of infants born after temporising treatment of early onset preeclampsia', Bjog, 112(7), pp. 910-4. Wolff, K., Kublickiene, K. R., Kublickas, M., Lindblom, B., Lunell, N. O. & Nisell, H. (1996). 'Effects of endothelin-1 and the ETA receptor antagonist BQ-123 on resistance arteries from normal pregnant and preeclamptic women', Acta Obstet Gynecol Scand, 75(5), pp. 432-8. Wolfrum, S., Dendorfer, A., Rikitake, Y., Stalker, T. J., Gong, Y., Scalia, R., Dominiak, P. & Liao, J. K. (2004). 'Inhibition of Rho-kinase leads to rapid activation of phosphatidylinositol 3- kinase/protein kinase Akt and cardiovascular protection', Arterioscler Thromb Vasc Biol, 24(10), pp. 1842-7. Wong, A. Y., Kulandavelu, S., Whiteley, K. J., Qu, D., Langille, B. L. & Adamson, S. L. (2002). 'Maternal cardiovascular changes during pregnancy and postpartum in mice', Am J Physiol Heart Circ Physiol, 282(3), pp. H918-25. Woods, A. K., Hoffmann, D. S., Weydert, C. J., Butler, S. D., Zhou, Y., Sharma, R. V. & Davisson, R. L. (2011). 'Adenoviral delivery of VEGF121 early in pregnancy prevents spontaneous development of preeclampsia in BPH/5 mice', Hypertension, 57(1), pp. 94-102. Woods, L., Perez-Garcia, V. & Hemberger, M. (2018). 'Regulation of Placental Development and Its Impact on Fetal Growth-New Insights From Mouse Models', Front Endocrinol (Lausanne), 9, p. 570. Wu, F., Tian, F. J., Lin, Y. & Xu, W. M. (2016). 'Oxidative Stress: Placenta Function and Dysfunction', Am J Reprod Immunol, 76(4), pp. 258-71. Wu, H., Lu, D., Jiang, H., Xiong, Y., Qu, C., Li, B., Mahmood, A., Zhou, D. & Chopp, M. (2008). 'Increase in phosphorylation of Akt and its downstream signaling targets and suppression of apoptosis by simvastatin after traumatic brain injury', J Neurosurg, 109(4), pp. 691-8. Wyrwoll, C. S., Noble, J., Thomson, A., Tesic, D., Miller, M. R., Rog-Zielinska, E. A., Moran, C. M., Seckl, J. R., Chapman, K. E. & Holmes, M. C. (2016). 'Pravastatin ameliorates placental vascular defects, fetal growth, and cardiac function in a model of glucocorticoid excess', Proc Natl Acad Sci U S A, 113(22), pp. 6265-70. Yagi, S., Aihara, K., Ikeda, Y., Sumitomo, Y., Yoshida, S., Ise, T., Iwase, T., Ishikawa, K., Azuma, H., Akaike, M. & Matsumoto, T. (2008). 'Pitavastatin, an HMG-CoA reductase inhibitor, exerts eNOS-independent protective actions against angiotensin II induced cardiovascular remodeling and renal insufficiency', Circ Res, 102(1), pp. 68-76. Yagi, S., Akaike, M., Aihara, K., Ishikawa, K., Iwase, T., Ikeda, Y., Soeki, T., Yoshida, S., Sumitomo- Ueda, Y., Matsumoto, T. & Sata, M. (2010). 'Endothelial nitric oxide synthase-independent protective action of statin against angiotensin II-induced atrial remodeling via reduced oxidant injury', Hypertension, 55(4), pp. 918-23. Yamazaki, M., Kobayashi, K. & Sugiyama, Y. (1996). 'Primary active transport of pravastatin across the liver canalicular membrane in normal and mutant Eisai hyperbilirubinaemic rats', Biopharm Drug Dispos, 17(8), pp. 645-59. Yeboah, D., Sun, M., Kingdom, J., Baczyk, D., Lye, S. J., Matthews, S. G. & Gibb, W. (2006). 'Expression of breast cancer resistance protein (BCRP/ABCG2) in human placenta throughout gestation and at term before and after labor', Can J Physiol Pharmacol, 84(12), pp. 1251-8. Yoshida, O., Kondo, T., Kureishi-Bando, Y., Sugiura, T., Maeda, K., Okumura, K. & Murohara, T. (2010). 'Pitavastatin, an HMG-CoA reductase inhibitor, ameliorates endothelial function in chronic smokers', Circ J, 74(1), pp. 195-202. Young, J. (1914). 'The AEtiology of Eclampsia and Albuminuria and their Relation to Accidental Haemorrhage: (An Anatomical and Experimental Investigation.)', Proc R Soc Med, 7(Obstet Gynaecol Sect), pp. 307-48. Yzydorczyk, C., Comte, B., Huyard, F., Cloutier, A., Germain, N., Bertagnolli, M. & Nuyt, A. M. (2013). 'Developmental programming of eNOS uncoupling and enhanced vascular oxidative stress in adult rats after transient neonatal oxygen exposure', J Cardiovasc Pharmacol, 61(1), pp. 8-16.

300

Zafar, R. (2015). 'An Insight into Pathogenesis of Cardiovascular Diseases', Journal of Cardiovascular Disease and Diagnosis, 3(197). Zarek, J., DeGorter, M. K., Lubetsky, A., Kim, R. B., Laskin, C. A., Berger, H. & Koren, G. (2013). 'The transfer of pravastatin in the dually perfused human placenta', Placenta, 34(8), pp. 719- 21. Zarek, J., Delano, K. E., Nickel, C., Laskin, C. A. & Koren, G. (2013). 'Are statins teratogenic in humans? Addressing the safety of statins in light of potential benefits during pregnancy', Expert Review of Obstetrics & Gynecology, 8(6), pp. 513-524. Zarek, J. & Koren, G. (2014). 'The fetal safety of statins: a systematic review and meta-analysis', J Obstet Gynaecol Can, 36(6), pp. 506-9. Zeiher, A. M., Fisslthaler, B., Schray-Utz, B. & Busse, R. (1995). 'Nitric oxide modulates the expression of monocyte chemoattractant protein 1 in cultured human endothelial cells', Circ Res, 76(6), pp. 980-6. Zhang, Y. & Kaufman, S. (2000). 'Effect of nitric oxide synthase inhibition on cardiovascular and hormonal regulation during pregnancy in the rat', Can J Physiol Pharmacol, 78(5), pp. 423- 7. Zhong, Y., Tuuli, M. & Odibo, A. O. (2010). 'First-trimester assessment of placenta function and the prediction of preeclampsia and intrauterine growth restriction', Prenat Diagn, 30(4), pp. 293-308. Zhou, L. & Zhu, D. Y. (2009). 'Neuronal nitric oxide synthase: structure, subcellular localization, regulation, and clinical implications', Nitric Oxide, 20(4), pp. 223-30. Zhou, Q., Ruan, Z. R., Jiang, B., Yuan, H. & Zeng, S. (2013). 'Simvastatin pharmacokinetics in healthy Chinese subjects and its relations with CYP2C9, CYP3A5, ABCB1, ABCG2 and SLCO1B1 polymorphisms', Pharmazie, 68(2), pp. 124-8. Zhou, Y., Damsky, C. H. & Fisher, S. J. (1997). 'Preeclampsia is associated with failure of human cytotrophoblasts to mimic a vascular adhesion phenotype. One cause of defective endovascular invasion in this syndrome?', J Clin Invest, 99(9), pp. 2152-64. Zigic, Z., Markovic, S., Grbesa, D., Ramic, S. & Halilovic, A. (2010). 'Quantitative research of capillaries in terminal villi of mature placentae', Bosn J Basic Med Sci, 10(2), pp. 147-52.

301

Chapter 8: Appendix

302

8.1 Vascular reactivity of Uterine arteries

8.1.1 Contraction responses of uterine artery (including water control)

A B 25 WT+CMC (6) 25 eNOS+CMC (5)

WT+Pitavastatin (7) eNOS+Pitavastatin (5) )

) 20 20 a

a WT+Water (3) eNOS+Water (4) P

(

15 15

i s

10 s 10 n

e T 5 T 5

0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10xM [U46619] x 10 M

C D

160 WT+CMC (6) 160 eNOS+CMC (5)

140 WT+Pitavastatin (7) 140 eNOS+Pitavastatin (5) n

120 n 120 eNOS+Water (4) o

WT+Water (3) o

S t

t 100 100

r P

r t

t 80 80

n 60 60

o o

40 C 40 C 20 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10xM [U46619] x 10 M

Figure 8.1: Contraction responses of uterine arteries to U46619 in pitavastatin - treated, vehicle and water control groups in WT and eNOS- /- mice.

Contraction of uterine artery to U46619 expressed as tension (kPa) (A and B) or as % KPSS (C and D) in WT and eNOS-/- mice. There was no significant difference in uterine artery contraction between the three groups in WT mice eNOS-/- mice (p>0.05). U46619 dose response curves were compared using two-way ANOVA. Data are expressed as mean±SEM; (N) = number of litters.

303

8.1.2 Relaxation responses of uterine artery (including water control)

A B 9

1 6

120 6 120

100 100 0

8 C

80 C 80

60 60 n

o i

i eNOS+CMC (5) t

40 WT+CMC (6) t 40

x x

a eNOS+Pitavastatin (5)

WT+Pitavastatin (7) a

l 20

l 20 e

R WT+Water (3) eNOS+Water (4)

0 0

% -10 -8 -6 -4 % -10 -8 -6 -4 x [ACh] x 10 M [ACh] x 10xM

C D

9 1

9 6

1 120 6

120 6

4 U

100

U 0

100

8 0

8 C

C 80

80 E

t 60

60 n

i i t 40

t 40 WT+CMC (6) eNOS+CMC (5)

x x

a eNOS+Pitavastatin (5)

a WT+Pitavastatin (6)

l 20

l 20 e e

WT+Water (3) R eNOS+Water (4)

0 0 % % -10 -8 -6 -4 -10 -8 -6 -4 [SNP] x 10xM [SNP] x 10xM

Figure 8.2: Assessment of relaxation of uterine arteries from pitavastatin - treated, vehicle and water control groups in WT and eNOS - /- mice. There was no significant difference in uterine artery relaxation to ACh (A and B) or SNP (C and D) in WT or eNOS-/- mice between the three groups (p>0.05). ACh and SNP dose response curves were compared using two-way ANOVA with Sidak’s post hoc test, where applicable. Data are expressed as mean±SEM; N= number of litters

304

8.2 Vascular reactivity of mesenteric artery

8.2.1 Contraction responses of mesenteric arteries (including water control)

A B 30 WT+CMC (5) 30 eNOS+CMC (5) 25 WT+Pitavastatin (6) ) eNOS+Pitavastatin (5)

) 25

a P 20 WT+Water (3) P eNOS+ Water (3)

k 20

(

n 15 n

o 15

s n

10 n 10

e T 5 T 5

0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM C D 180 WT+CMC (5) 180 eNOS+CMC (5) 160

WT+Pitavastatin (6) 160 eNOS+Pitavastatin (5) n

n 140 140 o

WT+Water (3) o eNOS+Water (3)

S S

t 120

i i

P 100

r 100

80 80

% % o 60

o 60 C 40 C 40 20 20 0 0 -10 -8 -6 -4 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM

Figure 8.3: Contraction responses of mesenteric arteries to U46619 in pitavastatin-treated, vehicle and water control groups in WT and eNOS -/- mice.

Contraction of mesenteric artery to U46619 expressed as tension (kPa) (A and B) or as % KPSS (C and D) in WT and eNOS-/- mice. There was no significant difference in mesenteric artery contraction between the three groups in WT mice eNOS-/- mice (p>0.05). U46619 dose response curves were compared using two-way ANOVA. Data are expressed as mean±SEM; (N) = number of litters.

305

8.2.2 Relaxation responses of mesenteric artery (including water control)

9 A B

1 6

6 120 6

6 120

100 100

8 C

80 C 80

E E

t t

o i

WT+CMC (5) i

t 40 40 eNOS+CMC (5) t

a x

WT+Pitavastatin (6) x eNOS+Pitavastatin (5)

a 20 a

l 20

l e

WT+Water (3) e eNOS+Water (3)

0 R 0 -10 -8 -6 -4 % -10 -8 -6 -4 x % [ACh] x 10 M [ACh] x 10xM C

9 D

1 6

6 120 6

6 120

100 100 0

8 C

80 C 80

t t

i i

t 40 40

WT+CMC (5) t eNOS+CMC (5)

a x WT+Pitavastatin (6) x a 20 eNOS+Pitavastatin (5)

a 20 l

l e

WT+Water (3) e eNOS+Water (3)

R R 0 0 -10 -8 -6 -4

% -10 -8 -6 -4 % [SNP] x 10xM [SNP] x 10xM

Figure 8.4: Assessment of relaxation of mesenteric arteries from pitavastatin-treated, vehicle and water control groups in WT and eNOS -/- mice. There was no significant difference in mesenteric artery relaxation to ACh (A and B) or SNP (C and D) in WT or eNOS-/- mice between the three groups, although mesenteric arteries from the water control group in eNOS-/- mice showed a trend towards blunted SNP-mediated relaxation (p=0.089). ACh and SNP dose response curves were compared using two-way ANOVA with Sidak’s post hoc test, where applicable. Data are expressed as mean±SEM; N= number of litters.

306

8.3 Vascular reactivity of umbilical artery

8.3.1 Contraction responses of umbilical arteries (including water control)

A B 15 WT+CMC (6) 15 eNOS+CMC (6) WT+Pitavastatin (7)

) eNOS+Pitavastatin (7)

)

a a

P WT+Water (3) 10 P eNOS+Water (4)

k 10

(

i s 5 s

n 5

n e

T T

0 0 -10 -8 -6 -4 -10 -8 -6 -4 x [U46619] x 10 M [U46619] x 10xM

C D 180 WT+CMC (6) 180 eNOS+CMC (6) 160 WT+Pitavastatin (7) 160 eNOS+Pitavastatin (7)

n 140

WT+Water (3) n 140

o i

o eNOS+Water (4)

S 120

i S

t 120

i S

100 c

i t

P 100

80 K

n 80

n o

60 % o

C 60

40 C 40 20 20 0 -10 -8 -6 -4 0 -10 -8 -6 -4 [U46619] x 10xM [U46619] x 10xM

Figure 8.5: Contraction responses of umbilical arteries to U46619 in pitavastatin - treated, vehicle and water control groups in WT and eNOS - /- mice. Contraction of umbilical artery to U46619 expressed as tension (kPa) (A and B) or as % KPSS (C and D) in WT and eNOS-/- mice. There was no significant difference in umbilical artery -/- contraction between the three groups in WT mice eNOS mice (p>0.05). U46619 dose response curves were compared using two-way ANOVA. Data are expressed as mean±SEM; (N) = number of litters.

307

8.3.2. Relaxation responses of umbilical artery (including water control)

9 A B

6 6

6 120 120 4

100 100

8 C

C 80 80

o t

t 60

o i

i 40 40 t

t WT+CMC (5) eNOS+CMC (5) a

a x

x WT+Pitavastatin (7) 20 eNOS+Pitavastatin (7) a

a 20

l e

e WT+Water (3) eNOS+Water (4) R

0 R 0

-10 -8 -6 -4 -10 -8 -6 -4

% % x x [SNP] x 10 M [SNP] x 10 M

Figure 8.6: Assessment of relaxation of umbilical arteries from pitavastatin - treated, vehicle and water control groups in WT and eNOS - /- mice.

There was no significant difference in umbilical artery relaxation to SNP (A and B) in WT or eNOS- /- mice between the three groups, (p>0.05). SNP dose response curves were compared using two-way ANOVA with Sidak’s post hoc test, where applicable. Data are expressed as mean±SEM; N= number of litters.

308