"Medicine, Faculty of"@en . "Obstetrics and Gynaecology, Department of"@en . "DSpace"@en . "UBCV"@en . "Fuchisawa, Akiko"@en . "2009-11-04T21:00:49Z"@en . "2003"@en . "Master of Science - MSc"@en . "University of British Columbia"@en . "INTRODUCTION: Preeclampsia is a pregnancy-specific condition, and it still remains one of the most common causes of maternal mortality in the developed world. Although the exact cause of preeclampsia has not been identified, it is most widely accepted that preeclampsia results from incomplete placentation. Interestingly, normotensive intrauterine growth restriction also shows the same defect of placentation. In preeclampsia, the maternal syndrome develops from a number of alternative pathways leading to uteroplacental mismatch and, consequently, the release of endothelium-activating factors. This research is focused on neutrophil activation and the hypothesis for this research was that maternal neutrophils and monocytes are inappropriately activated in preeclampsia but not in normotensive intrauterine growth restriction. METHODS: This was a prospective controlled cohort study in a tertiary center. Subjects consisted of cases: women with early-onset preeclampsia (<34+0wk; EOPET), late-onset preeclampsia (\u00E2\u0089\u00A734+0wk; LOPET), normotensive intrauterine growth restriction and controls: normal pregnancy controls (NPC) and non-pregnant controls (non-preg). Peripheral blood leukocytes were analyzed immediately after phlebotomy for CD11b, CD18, CD14 and band forms (marrow production). Neutrophils were cultured for 18-24h and apoptosis was assessed by Annexin V binding and hypodiploid PI. All analyses were by flow cytometry. Kruskal Wallis, Mann Whitney U and Dunn's post test were used for statistical analysis (significance at p<0.05). RESULTS: CD11b/CD18 monocyte expression was increased in pregnancy (all groups). CD18 expression was reduced on preeclampsia lymphocytes. CD14 expression was reduced on preeclampsia granulocytes. Gestational age-dependent decrease was noted for apoptosis in normal pregnancy. Band forms were increased in all pregnancy groups, with no difference seen between normal and abnormal pregnancy. CONCLUSIONS: Normal pregnancy is a state of Th1-to-Th2 switch. In preeclampsia, some findings noted by ourselves and others were not reproduced, probably due to steroid effects. Decreased CD18 on preeclampsia lymphocytes implies excess Th1 activity in preeclampsia. Preeclampsia granulocytes shed CD14, probably in response to prior activation. There was a gestational age effect on neutrophil apoptosis, as previously noted. However, neutrophil apoptosis was not inappropriately delayed in preeclampsia, probably due to steroid effects. Marrow production of neutrophils is similarly increased in all pregnancy groups."@en . "https://circle.library.ubc.ca/rest/handle/2429/14664?expand=metadata"@en . "13910119 bytes"@en . "application/pdf"@en . "Maternal leukocyte CD markers, apoptosis and band forms in preeclampsia by Akiko Fuchisawa Associated Degree in Medical Technology, Hirosaki University, Japan, 1996 B. H. Sc., The National Institution for Academic Degrees, Japan, 1997 A THESIS SUBMITED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Obstetrics and Gynaecology; Reproductive and Developmental Sciences) We accept this thesis as conforming To the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 2003 \u00C2\u00A9 Akiko Fuchisawa, 2003 Library Authorization In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shaii not be allowed without my written permission. -Name of Author (please print) Date (dd/mm/yyyy) Title of Thesis: \\ pet e d | ueu W cj 7 e. C D M ^ r ? A ^ p T p - S f - 5 o^^-X ^>CK^-A Tvrhn s In r\u00C2\u00A3 \u00E2\u0082\u00AC-c{ (h w^ ps LJ^ Degree: , S o . Year: D \"3 Department of , . .Oksftfr ? CS cK^r\ Gryw <\JL.O>l The University of British Columbia \u00E2\u0080\u0094 Vancouver, BC Canada ABSTRACT INTRODUCTION: Preeclampsia is a pregnancy-specific condition, and it still remains one of the most common causes of maternal mortality in the developed world. Although the exact cause of preeclampsia has not been identified, it is most widely accepted that preeclampsia results from incomplete placentation. Interestingly, normotensive intrauterine growth restriction also shows the same defect of placentation. In preeclampsia, the maternal syndrome develops from a number of alternative pathways leading to uteroplacental mismatch and, consequently, the release of endothelium-activating factors. This research is focused on neutrophil activation and the hypothesis for this research was that maternal neutrophils and monocytes are inappropriately activated in preeclampsia but not in normotensive intrauterine growth restriction. METHODS: This was a prospective controlled cohort study in a tertiary center. Subjects consisted of cases: women with early-onset preeclampsia (<34+0wk; EOPET), late-onset preeclampsia (^34+0wk; LOPET), normotensive intrauterine growth restriction and controls: normal pregnancy controls (NPC) and non-pregnant controls (non-preg). Peripheral blood leukocytes were analyzed immediately after phlebotomy for C D l l b , CD18, CD14 and band forms (marrow production). Neutrophils were cultured for 18-24h and apoptosis was assessed by Annexin V binding and hypodiploid PI. All analyses were by flow cytometry. Kruskal Wallis, Mann Whitney U and Dunn's post test were used for statistical analysis (significance at p<0.05). RESULTS: CDllb/CD18 monocyte expression was increased in pregnancy (all groups). CD18 ii expression was reduced on preeclampsia lymphocytes. CD14 expression was reduced on preeclampsia granulocytes. Gestational age-dependent decrease was noted for apoptosis in normal pregnancy. Band forms were increased in all pregnancy groups, with no difference seen between normal and abnormal pregnancy. CONCLUSIONS: Normal pregnancy is a state of Thl-to-Th2 switch. In preeclampsia, some findings noted by ourselves and others were not reproduced, probably due to steroid effects. Decreased CD 18 on preeclampsia lymphocytes implies excess Thl activity in preeclampsia. Preeclampsia granulocytes shed CD 14, probably in response to prior activation. There was a gestational age effect on neutrophil apoptosis, as previously noted. However, neutrophil apoptosis was not inappropriately delayed in preeclampsia, probably due to steroid effects. Marrow production of neutrophils is similarly increased in all pregnancy groups. iii TABLE OF CONTENTS Abstract : ii Table of Contents iv List of Tables .vii List of Figures viii List of Abbreviations x Acknowledgements xii CHAPTER 1 Background .1 1.1 Preeclampsia 2 1.2 The pathophysiology of preeclampsia 2 1.2.1 Normotensive Intrauterin Growth Restriction (nRJGR) 7 1.2.2 Endothelial cells in preeclampsia . .8 1.3 Preeclampsia and systemic inflammatory response syndrome (SIRS) 12 1.3.1 SIRS \u00E2\u0080\u00A2. 12 1.3.2 The similarity of preeclampsia and SIRS: Clinical 12 1.3.3 The similarity of preeclampsia and SIRS: Neutrophil activation 15 1.3.4 The inflammatory response: endothelium 16 1.3.5 The inflammatory response: leukocytes 18 1.3.6 Peripheral blood lymphocytes in normal pregnancy and preeclampsia.20 1.3.7 Peripheral blood monocytes in normal pregnancy and preeclampsia... 22 1.3.8 Peripheral blood granulocytes in normal and preeclampsia 23 CHAPTER 2 Research Plan 26 2.1 Hypothesis 27 2.2 Aims ..27 2.3 Subjects 27 2.4 Rationale 29 2.4.1 Rationale: neutrophil apoptosis 29 2.4.2 Rationale: surface antigen expression 30 i v CHAPTER 3 Standardization of methods 3 2 3.1 Rationale 3 3 3.1.1 Flow cytometry 3 3 3.1.2 CD markers and rapid flow cytometry 3 7 3.1.3 Apoptosis \"40 3.2 Methods 44 3.2.1 Comparison of three methods for rapid flow cytometry 44 3.2.1.1 Methods '-44 3.2.1.2 Results 46 3.2.1.3 Interpretation \u00E2\u0080\u00A2\u00E2\u0080\u00A2\u00E2\u0080\u00A247 3.2.2 Time course 48 3.2.3 Methods and Subjects 48 3.2.4 Results 48 3.2.5 Interpretation 50 3.3 Summery 50 CHAPTER 4 CDllb, 18 and 14 expressions on peripheral blood leukocytes and neutrophil apoptosis in preeclampsia, normotensive IUGR, normal pregnancy and non-pregnancy 52 4.1 Rationale 53 4.1.1 CD markers 53 4.1.2 Apoptosis \u00E2\u0080\u00A2 53 4.2 Subjects 5 3 4.3 Method 56 4.3.1 Method: Surface antigen expression 56 4.3.2 Method: Neutrophil apoptosis 56 4.3.2.1 Neutrophil isolation and culture 56 4.3.2.2 Surface binding of Annexin V and PI 57 4.3.2.3 Propidium iodide DNA profiles 58 4.3.3 Band count 58 4.3.4 Analyses 59 4.4 Results \u00E2\u0080\u00A2 61 4.4.1 Surface antigen expression 63 4.4.2 Neutrophil apoptosis 80 4.4.3 Band counts --91 4.5 Summary 97 v CHAPTER 5 Discussion .\". 105 Bibliography 110 Appendix 124 v i LIST OF TABLES Table 1.1 SIRS criteria 12 Table 1.2 Preeclampsia and SIRS share clinical and laboratory features 14 Table 3 The recovery of cells following three different preparation techniques 46 Table 4.1 Clinical characteristics 62 Table 4.2 Table 4.2 Summary of CD marker findings by leukocyte cell type (mean channel brightness; median [range]) 95 Table 4.3 Table 4.3 Summary of spontaneous neutrophil apoptosis findings (%; median [range]) 96 Table 4.4 Summary of neutrophil band form findings (median [range]) 97 v i i LIST OF FIGURES Figure 1.1 The pathogenesis of preeclampsia 5 Figure 1.2 Endothelial cell and granulocyte interaction 17 Figure 1.3 The mechamism of leukocyteAmediated injury following ischemia-reperfusion 19 Figure 3.1 Flow cytometer, principles of design and setup 35 Figure 3.2 The size and granularity characteristics of peripheral blood leukocytes 37 Figure 3.3.a Flow cytometric analysis: CDllb on granulocyte 39 Figure 3.3.b Flow cytometric analysis: CDllb on monocyte 39 Figure 3.4 Annexin V binding 41 Figure 3.5 Annexin V binding 42 Figure 3.6 Hypodiploid DNA (PI) 43 Figure 3.7 Stability of surface antigen expression in samples kept at room temperature for up to 5 hours 49 Figure 3.8 Stability of surface antigen expression in samples kept at room temperature for up to 5 hours 50 Figure 4.1 CDllb-antepartum: Granulocyte 64 Figure 4.2 CDllb-antepartum: Monocyte 65 Figure 4.3 CDllb-postpartum: Granulocyte 66 Figure 4.4 CDllb-postpartum: Monocyte 67 Figure 4.5 CD18-antepartum: Granulocyte 69 Figure 4.6 CD18-antepartum: Monocyte 70 Figure 4.7 CD18-antepartum: Lymphocyte 71 Figure 4.8 CD18-postpartum: Granulocyte 72 Figure 4.9 CD18-postpartum Monocyte 73 Figure 4.10 CD18-postpartum: Lymphocyte 74 Figure 4.11 CD14-antepartum: Granulocyte 76 Figure 4.12 CD 14-antepartum: Monocyte 77 Figure 4.13 CD14-postpartum: Granulocyte 78 Figure 4.14 CD 14-postpartum: Monocyte 79 Figure 4.15 Hypodiploid Pl-antepartum 82 Figure 4.16 Hypodiploid Pl-postpartum 83 Figure 4.17 Annexin V-antepartum 84 Figure 4.18 Annexin V-postpartum 85 Figure 4.19 PI (necrosis)-anteparrum 86 Figure 4.20 PI (necrosis)-postaprtum 87 Figure 4.21 Annexin V-PI (necrosis)-antepartum 88 Figure 4.22 Annexin V-PI (necrosis)-postpartum 89 viii Figure 4.23 Spontaneous neutrophil apoptosis delay across gestation 90 Figure 4.24.a #Band-antepartum 92 Figure 4.24.b %Band-antepartum 92 Figure 4.25.a #Band-postpartum 9 3 Figure 4.2 5 .b %Band-postpartum \u00E2\u0080\u00A2 93 ix LIST OF ABBREVIATIONS ACD: Acid citrate dextrose solution ANOVA: Analysis of variance ATN: Acute tubular necrosis ARDS: Acute respiratory distress syndrome AST: L-asparate: 2oxoglutarate arjiinotransferase BW: Birth weight CSF: Colony stimulating factor DIC: Disseminated intravascular coagulation DMEM: Dulbecco modified eagle medium EDF: End diastolic flow EDTA: Ethylene diamine teraacetic acid EOPET: Early-onset preeclampsia FITC: Fluorescein isothiocyanate FMLP: n-forTxiyl-methionyl-leucyl-phenylalanine GA: Gestational age ICAM-1: Intracellular adhesion molecule-1 IL-1: Interleukin-1 IL-IRa: Interleukin-1 receptor antagonist IL-6: Interleukin-6 IL-8: Interleukin-8 INF- y : Interferon- y IUGR: Intrauterine growth restriction LOPET: Late-onset preeclampsia MFI: Mean fluorescence index nlUGR: Normotensive mtrauterine growth restriction NO: Nitric oxide NPC: Normal pregnancy control MAP: Mean arterial pressure PBL: Peripheral blood leukocyte PBS: Phosphate buffered saline PHA: Phytohemagglutmin PI: Propidium iodide PIH: Pregnancy-induced hypertension PMN: Polymorphonuclear neutrophil PMSF: Phenyhnethylsulphonylfluoride PS: Phosphatidylserine PWM: Pokeweed mitogen ROS: Reactive oxygen species SEM: Standard error of the mean SIRS: Systemic inflammatory response syndrome Thl: T helper cell-1 Th2: T helper cell-2 TLC: Total leukocyte count T N F - a : Tumor necrosis factor- a VCAM-1: Vascular cell adhesion molecule-1 vWF: von Willebrand factor A C K N O W L E D G E M E N T I wish to thank the following individuals for their support in conducting this research. First of all, I would like to thank my thesis committee, Drs von Dadelszen, Leung, Magee and van Eeden, for their support, encouragement and advice. At BC Women's Hospital, I would like to thank Terry Viczko, Shelley Soanes and Vesna Popovska for identifying, recruiting and arranging the sample taking from the women in this study. My thanks are extended to the clinical phlebotomy service and laboratory for initial sample handling. I am grateful to Dr. Duna Goswami for sharing her data with me. These experiments could not have been done without the help and support of Beth Whalen, MacDonald Research Laboratory. I owe her a debt of gratitude. My funding was from an Establishment Grant from the BC Research Institute for Children's and Women's Health (awarded to Dr. von Dadelszen). xii Chapter 1: Background 1.1 Preeclampsia Preeclampsia is a pregnancy-specific condition, and remains one of the two most common causes of maternal mortality in the developed world (1-4). Preeclampsia occurs in 6-8% of all pregnancies (5), and has two syndromes: the maternal syndrome characterized by hypertension and proteinuria and the fetal syndrome characterized by intrauterine growth restriction (IUGR). According to the National High Blood Pressure Education Program Working Group (NHBPEP), preeclampsia is defined as a pregnancy-specific increase in blood pressure developing after 20 weeks of gestation (diastolic blood pressure ^ 90mmHg, or systolic blood pressured 140mmHg) associated with^0.3g proteinuria/d. The elevations in blood pressure must be present on two occasions six hours or more apart (5). 1.2 The pathophysiology of preeclampsia The exact cause of preeclampsia has not been identified although numerous theories of potential causes exist, including genetic, dietary, vascular, autoimmune factors and placental ischemia. These theories, once considered to be competing and mutually exclusive, appear to be converging as our understanding of the condition improves. It has been reported that women whose sisters or mothers had preeclampsia are at high risk to develop preeclampsia in their own pregnancies (6). This genetic theory, however, does not explain why preeclampsia most frequently occurs in their first pregnancy and why 2 preeclampsia is more likely to occur after a change of partner (7). It is likely that the maternal syndrome of preeclampsia is a final common pathway with many alternative routes to its inception (8;9). Hypertension associated with preeclampsia develops during pregnancy, and resolves after delivery, suggests that the imperfectly implanted placenta is an initiating agent (22). The root cause of preeclampsia has been considered to be reduced placental perfusion commonly caused by abnormal placentation (10), which becomes clinically manifest in later pregnancy, perhaps once the fetal/placental nutritional demands outstrip placental supply. Most commonly preeclampsia results from incomplete placentation (14). In preeclampsia, there is either deficient placentation or excessive fetoplacental demand (multiple pregnancy, macrosomia). In the first scenario, fetal cytotrophoblast cells fail to completely invade the uterine spiral arteries with inadequate endovascular trophoblast remodeling (22;24;25). This results in relatively narrowed spiral arteries and decreased uteroplacental perfusion of the placenta causing placental ischemia (10), and reduced nutrient and oxygen supply to the fetus as gestation progresses. Interestingly, both preeclampsia and normotensive IUGR show the same defect of placentation (26). The resultant uteroplacental mismatch (9), where the demands of the fetus outstrip the capacity of the uteroplacental arterial supply, may also arise due to excessive fetal demands, as occur in multiple pregnancy (11) or fetal overgrowth (12), or by the loss of functioning placental mass in the setting of thrombophilia (13). The intervillous space of the mismatched placenta releases factor(s) into the maternal circulation ('intervillous soup' (9), ; 3 leading to endothelial dysfunction and microangiopathic hemolysis (14), inflammatory mediator release (15), and neutrophil activation (16-18). It is likely that maternal factors are also involved as reduced placental perfusion also accompanies IUGR (10) and preterm birth (19) without evident maternal changes. What results is the phenotypic organ dysfunction of the clinical syndrome. Although hypertension is the most common manifestation, the maternal syndrome of preeclampsia is considerably more than pregnancy-induced hypertension (8). In fact, in the era of adequate blood pressure control (20), preeclampsia-associated mortality is most commonly due to either hepatic necrosis or the acute respiratory distress syndrome, both of which are the consequences of systemic inflammation (21). 4 immunological factors hypertension glomerular endotheliosis/ proteinuria/ ATN liver damage/ hematoma/ rupture microangiopathic hemolysis/ thrombocytopenia/ DIC Figure 1.1 The pathogenesis of preeclampsia. In this model of preeclampsia, the maternal syndrome develops from a number of alternative pathways leading to uteroplacental mismatch, whereby the fetoplacental demands outstrip the maternal circulatory supply. In response to the mismatch, and probably due in part to recurrent ischemia-reperfusion injury within the intervillous (maternal blood) space of the placenta and accelerated placental apoptosis, a soup of endothelium-damaging substrates is released with resulting endothelial cell activation and consequent development of the maternal syndrome of preeclampsia. Some elements of the soup, namely activated peripheral blood leukocytes, can cause direct end-organ damage. There is cross-talk between elements of the soup (not illustrated). ARDS: acute respiratory distress syndrome; ATN: acute tubular necrosis; DIC: disseminated intravascular coagulation; PBLs: peripheral blood leukocytes; PGs: eicosanoids; ROS: reactive oxygen species (from von Dadelszen etal. Crit Care Med 2002;30:1883-1892). Hypertension-induced cerebral hemorrhage used to be the most common cause o f maternal death associated with preeclampsia (3). However, with the development o f effective antihypertensive therapy (20), women with preeclampsia now most commonly die from either hepatocellular necrosis or acute respiratory distress ( A R D S ) (2-4). Both hepatocellular necrosis and A R D S are characterized by neutrophil (PMN) tissue infiltration (23). 5 Incomplete spiral arterial modification is the most well developed concept for the etiology of preeclampsia, fetal IUGR and hypoxia, and appears to be the most common etiological pathway for early-onset disease. Relative fetal overgrowth is more commonly associated with late-onset disease. It appears that early-onset preeclampsia (onset <34 weeks' gestation) differs from late-onset preeclampsia (onset >34+0 weeks' gestation) in terms of both fetal and maternal risks (122). Early-onset preeclampsia represents considerable additional maternal risk, as maternal mortality is some 20-fold higher at <32 weeks' gestation than when preeclampsia occurs at term (27). This is supported by observations that the pathophysiology of early-onset preeclampsia differs from late-onset disease, in terms of increased neutrophil function (9) and increased Th2 cytokine levels (15;28). Also, there is compelling epidemiological evidence that early-onset disease (defined as onset <28 weeks) is associated with greater risk for recurrence in later pregnancies (29-31), and increased risk for later cardiovascular disease (30;31). That perinatal morbidity and mortality are gestational age-dependent is a given, as, among diploid fetuses, gestational age is the most important determinant of perinatal outcome (9;32). Early-onset preeclampsia is an important predictor of intrauterine growth restriction, which is much less common in pregnancies complicated by preeclampsia at term (33). In fact recent data suggests an increase in large babies among women with preeclampsia delivering after 37 weeks gestation (12). A greater than 50% chance of intact survival for a fetus delivered of a woman with . 6 preeclampsia arises only when the gestational age at delivery is >21Mi weeks' and the birthweight >600g (9). For these reasons, women with early-onset preeclampsia may provide the most homogeneous data for differentiating the changes of preeclampsia from those of normal pregnancy as investigators attempt to advance knowledge. It is also known that trophoblastic debris is released into the maternal circulation to renew syncytiotrophoblast in normal pregnancy. This may partially explain the upregulation of the innate (Th2) immune system in normal pregnancy (34); this upregulation increases as pregnancies advance. Therefore, the large placenta associated with multiple pregnancies, releasing more debris, may cause the maternal syndrome of preeclampsia (35). 1.2.1 Normotensive Intrauterine Growth Restriction (nlUGR) as a control group Women who deliver growth restricted fetuses, but have remained clinically normotensive and without significant proteinuria, are an important control group in preeclampsia research. Normotensive IUGR shares the placental pathology of preeclampsia (10), but without a maternal response. Therefore, changes found in preeclampsia, but not nlUGR will be specific to preeclampsia, and therefore, more important in understanding the pathophysiology of preeclampsia. Using nlUGR as a control group in preeclampsia research was pioneered by our group (17), and has helped to identify specific pathophysiologies of preeclampsia, namely 7 apoptosis (17) and chronic infection (36). 1.2.2 Endothelial cells in preeclampsia Rather than being a quiescent monolayer of cells lining the vascular tree, the endothelium has many important biological functions. It controls communication, transportation and leukocyte migration between the intravascular and extravascular compartments, and also regulates anticoagulation, antiplatelet, and fibrinolysis functions and blood flow through release of different soluble factors. Roberts et al (14) suggested the following as a unifying concept for the pathogenesis of preeclampsia: \"The poorly perfused placental tissue releases a factor(s) into the systemic circulation that injures endothelial cells. The changes initiated by endothelial cell injury set in motion a dysfunctional cascade of coagulation, vasoconstriction, and intravascular fluid redistribution that results in the clinical syndrome of preeclampsia.\" The evidence for this concept is summarized below. 1.2.2.1 Endothelial cell morphology Renal glomerular endotheliosis and acute placental atherosis are the most characteristic vascular changes in preeclampsia. Glomerular endotheliosis is the pathogromonic vascular alternation of preeclampsia . 8 (37;38), and it correlates with its disease severity (39). It does not occur in non-pregnancy (40). The lumina of the glomerular capillaries are removed by swollen endothelial cells while the glomerular epithelium seems to be essentially normal by light microscopy (41). It is considered that the changes in the glomerular capillary endothelium contribute to the proteinuria characteristic of the condition. Acute atherosis is characterized by focal endothelial disruption, fibrinoid necrosis of the arterial wall, the infiltration of perivascular spaces by mononuclear cells and endovascular accumulation of lipid-laden macrophages. The affected arteries, such as the spiral arteries of the decidua vera and parietalis, the basal arteries traversing the myometrium and decidua basalis, arid the myometrial segments of the uteroplacental arteries, may become partially or completely obliterated. Damaged endothelial cells are also found in uterine venulae (42) and endomyocardial vessels (43) in preeclampsia. 1.2.2.2 Evidence for endothelial cell dysfunction in preeclampsia A number of indicators of endothelial cell dysfunction have been found in preeclampsia, von Willebrand factor (vWF) (44), was probably the first described factor. vWF is released by activated endothelium, following stimulation by agents such as thrombin (45). Hepatocytes are the primary source of plasma fibronectin (46). Fibronectin levels have been found to be increased in pregnancies complicated by either preeclampsia or IUGR (47). 9 Friedman and colleagues speculate that the endothelial involvement in IUGR is confined to the uteroplacental circulation, whereas it is systemic in preeclampsia (47). It has been proposed that preeclampsia is associated with disorders of nitric oxide production (48), and this may contribute to the vasoconstriction characteristic of the syndrome. It was initially thought to reflect solely endothelial cell activation, NO is also derived from circulating phagocytes (49), and increased phagocyte production of NO could contribute pro-oxidant stress of preeclampsia (50), through conversion to peroxynitrite (51). In vivo investigation of NO-mediated forearm arterial vascular reactivity in both normal pregnancy and preeclampsia (52;53) mitigates against changes in vascular NO activity accounting for the increased vascular tone of preeclampsia Lowe et al found, in preeclampsia, there is a relative deficiency of available NO and an excess of peroxynitrite. The combination of a directly and/or indirectly reduced NO activity could initiate a number of the physiological and serological changes associated with preeclampsia (54). In summary, it is likely that preeclampsia is not due to a lack of NO, and is associated with increased NO production, perversely resulting in a increased production of peroxynitrite and oxidative stress. Endothelin-1 is a small polypeptide with potent vasoconstrictive properties, and may contribute to the hypertension in preeclampsia (55). Its origin may be either endothelial or placental, and elevated serum levels correlate with elevation of other factors, such as neutrophil 10 elastase, fibronectin, and vWF (56;57). In normal pregnancy, the plasma level of prostacyclin and thromboxane are balanced. (7). In pregnancy, prostacyclin may be produced by both endothelium and the placenta (58). The plasma level of prostacyclin is significantly lower in preeclampsia and it correlates with the disease severity (59;60). Conversely, thromboxane, which is the constitutional antagonist to prostacyclin, is increased in preeclampsia in which its effects dominate (59). Endothelial cells are both a source of, and a target for, cytokines. Stimulation by inflammatory and other stimuli changes their function. Interleukin-1 (IL-1) and tumor necrosis factor- a (TNF- a) induce a pattern of Th2 responses common to both inflammation and immunity, responses which are enhanced by interferon- y (IFN- y ). In addition, endothelial cells, themselves, produce some of the cytokines, including IL-1, IL-6, colony-stimulating factors (CSFs) (G-, M - , GM-), and chemotactic cytokines (IL-8 and monocyte chemotactic protein). Interleukin-6 (IL-6) is produced by activated monocytes/macrophages (61), endothelial cells, activated T- and B-lymphocytes, and other cells (61). Target cells for IL-6 include T-cells (61), hepatocytes (62), and other leukocytes (61). Vince et al (15) have reported increased levels of IL-6 in preeclampsia, correlating with its disease severity. IL-6 levels that increase as normal pregnancy progresses, and further increase intrapartum (28). Greer et al (63) also found elevated level of IL-6 in the circulation of women with preeclampsia (64). 11 1.3 Preeclampsia and systemic inflammatory response syndrome (SIRS) 1.3.1 SIRS The systemic inflammatory response syndrome (SIRS) is the development of a severe and inappropriate systemic response, which persists following the removal of the inciting stimulus such as infection or trauma (65). Sepsis is associated with dysregulation of the immune system and can lead to SIRS (Table 1.1). Table. 1.1 SIRS criteria Document Criteria Result Tmax= Fever^38\u00C2\u00B0C (100.4\u00C2\u00B0 F), or hypothermia^36\u00C2\u00B0C (96.8\u00C2\u00B0 F) HR= Heart rate ^ 90 beats/minute RR= Respiratory rate^20breaths per minute, or a PaC02^32mmHg, or PaC0 2= mechanical ventilation for an acute process WBC= White blood cell (WBC) counts 12,000/mm3, or ^4,000/ mm 3, or >10% Immature neutrophils SIRS: Patients must have three or more of the above four SIRS criteria. If the patient is on drugs that modify the heart rate, the patient must have two of the four SIRS criteria. Modified from Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus AM, Schein RM, Sibbald WJ. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 1992; 101:1664-1665. 1.3.2 The similarity of preeclampsia and SIRS: Clinical There is an obvious similarity between preeclampsia and SIRS in terms of the clinical features. For example, SIRS can be characterized by hepatic failure and ARDS (66;67), that are both neutrophil (PMN)-mediated (23) and, these most commonly cause maternal mortality in 12 preeclampsia (2-4). Both disorders persist following the removal of the initiating agent, that is the poor placenta (preeclampsia) or gram-negative sepsis (SIRS), but do not occur uniformly in the presence of that respective initiating agent. Disordered activation of the clotting cascade occurs in both syndromes, resulting in disseminated intravascular coagulation (DIC) and microvascular thrombosis and hemolysis in preeclampsia (68) and SIRS (23;66;67;69). 13 Table 1.2 Preeclampsia and SIRS share clinical and laboratory features (from von Dadelszen et al). Clinical finding Laboratory finding Disseminated intravascular coagulation Endothelial function and inflammation Microvascular thrombosis and hemolysis T plasma TNF- a Thrombocytopenia T plasma IL-6* Acute renal failure f PAI-l;4PAI-2 Hyperdynamic state f thromboxane:prostacyclin ratio End organ hypoperfusion t endothelin t Vascular permeability T ceruloplasmin Hepatic necrosis t a i-antitrypsin Acute respiratory distress syndrome Complement activation Cardiomyopathy \u00E2\u0080\u00A2I plasma transferrin Differential response to initiating agent Hypoalbuminemia (placenta/trauma/infection) Persistence following removal of initiating agent ^Protein C* Lymphocyte activation Neutrophil activation TiROS Neutrophilia* and delayed neutrophil apoptosis t IgG production T plasma neutrophil elastase* t intracellular ionized calcium t neutrophil CD l ib expression T basal intracellular ionized calcium t oxidative stress Monocyte activation T phagocytosis TiROS T intracellular ionized calcium t \u00C2\u00AB * CD14 t C D l l b t CD62L t INF-7 t IL-IRa t M-CSF t: increased compared with normal; ^decreased compared with normal; *: compared with normal pregnancy; IL-6: interleukin-6; PAI: plasminogen activator inhibitor; TNF- a : tumor necrosis factor- a; intracellular reactive oxygen species; INF-y : interferon- y ; IL-IRa: interleukin-1 receptor antagonist; M-CSF: macrophage-colony stimulating factor 14 1.3.3 The similarity of preeclampsia and SIRS: Neutrophil activation The understanding of the pathogenesis of organ dysfunction in patients with SIRS suggests a central role for activated PMNs. In SIRS, stimuli, such as endotoxin, bacteria, injured tissue and ischemia, result in disseminated activation of the innate immune system attributed to the action of host-derived mediators on host tissues. Preeclampsia and SIRS commonly show P M N activation present, including increased plasma concentration of PMN elastase (23;70), increased oxidative stress (50;71-73), increased surface expression of CD1 lb (74-76), increased concentration of basal intracellular ionized calcium (16;77;78), and, in vitro, increased chemoattractant-induced PMN superoxide production (79; 80). Neutrophilia is prominent in normal pregnancy, preeclampsia and SIRS. There is a gestational age-related delay in apoptosis that explains the neutrophilia of normal pregnancy (17). In normotensive IUGR, PMN apoptosis is as delayed as it is in normal pregnancy (17). In contrast with both normal pregnancy and normotensive IUGR, preeclampsia is associated with both a relative neutrophilia and greater delays in PMN apoptosis (17). The neutrophilia of preeclampsia is a relative increase in PMN concentration over the neutrophilia of normal pregnancy. However, the neutrophilia of SIRS is relative to the physiological non-pregnancy state. Delayed P M N apoptosis may explain the neutrophilia of both preeclampsia (17) and SIRS (76). 15 1.3.4 The inflammatory response: endothelium. One of the functions of the endothelium is the ability to express molecules which promote adherence of leukocytes and facilitate initiation of localized inflammatory responses. Integrins are a large family of heterodimers that are composed of varied a - chains (CD 11 a, b, c) and a common j3 - chain (CD 18) and include plasma membrane receptors for intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) (81). They mediate cell-cell and cell-extracellular matrix interactions, requiring both chains for normal receptor expression and function. These moieties function as adherence mediators of leukocytes to activated endothelium, and play an important role for rolling, spreading, diapedesis sequence by which circulating leukocytes move from the intravascular compartment into the tissues (Figure i2). ; Adhesion molecules can be shed from cell surfaces, and serum levels of soluble adhesion molecules may be useful monitors of the magnitude of the inflammatory disease activity (82), with inflammation (83), infection (84), and malignancy (85). Increased levels of soluble adhesion molecules including have been found in preeclampsia serum (86-90) but not consistently in nlUGR (88-90). 16 EC EC \ A ' , , , , , , , , , & ( ] selectin \u00E2\u0080\u00A2 complement regulatory protein ^ endothelial growth factor dom ain 0 lectin-like domain \u00E2\u0080\u0094 - integrin \u00E2\u0080\u00A2aaaaaai immunoglobulin 4aa* sialyl Lew is\" (C D 15s) Figure 1.2 Endothelial ceil and granulocyte interaction M o l e c u l a r in te rac t ions d u r i n g g r a n u l o c y t e a d h e s i o n t o endo the l i a l c e l l s at si tes o f i n f l a m m a t i o n . N o t e e x p l o d e d v i e w s o f g r a n u l o c y t e - e n d o t h e l i a l c e l l i n te r face w i t h i n respec t i ve rec tang les . BM: basemen t m e m b r a n e ; EC: endo the l i a l c e l l ; ICAM-1; i n t e r ce l l u l a r a d h e s i o n m o l e c u l e ( C D 5 4 ) ; LFA-1: C D 1 l a / C D 1 8 ; MAC-1: C D 1 l b / C D 1 8 ; PMN: p o l y m o r p h o n u c l e a r l e u k o c y t e ( g r a n u l o c y t e ) . M o d i f i e d from R o s a l e s & B r o w n (1993) . 17 1.3.5 The inflammatory response: leukocytes. The inflammatory response is not a disease specific, but rather a non-specific response to various insults including immune activation. Leukocytes play an essential role in dealing with inflammation. Most acute inflammatory responses are localized to an area of injury and characterized by microvascular leakage and the accumulation of neutrophils, but, in some cases, they can extend to involve the entire vascular compartment, as in SIRS (6). Only approximately 50% of the peripheral blood neutrophils circulate freely and the rest are marginated or rolling along the endothelium (Figure 1.3). The intravascular neutrophil population accounts for 5% of the body's neutrophil pool, but in some kinds of infection-the marrow production of neutrophils is increased and the marrow pool is decreased (91). Once neutrophils leave for the damaged site from the bloodstream, they do not return from the tissues. One of the critical parts of the process of inflammation is a dramatic increase in endothelial cell surface expression of molecules that lead the adhesion of leukocytes stimulated by thrombin or histamine within minutes, or by endotoxin, IL-1 or TNF- a within hours (92). Neutrophils are the characteristic feature of acute inflammation histologically. Granulocytes have been implicated in the pathophysiology of acute vascular damage. Inflammatory response is basically a vascular phenomenon and acute inflammation alters microvascular permeability and neutrophil for exudation. Leukocyte-derived products, such as is proteolytic enzymes and reactive oxygen intermediates contribute to tissue or endothelial cell damage (6). VASCULAR OCCLUSION O VASCULAR INJURY iiB endothelial cell \u00E2\u0080\u00A2\u00E2\u0080\u00A2i activated endothelial cell O erythrocyte fibrin granulocyte monocyte platelet protein flux Figure 1.3 The mechanisms of leukocyte-mediated injury following ischemia-reperfusion Following a period of ischemia, at the time of reperfusion, those endothelial cells and peripheral blood leukocytes still living are activated. Consequently, leukocyte-endothelial cell adhesion occurs, the leukocytes releasing mediators injurious to endothelial cells. During reperfusion, intravascular activation may be triggered, resulting in microvascular occlusion. Inhibition of leukocyte adhesion by mAbs raised against integrins prevents vascular injury and reduces tissue damage. Modified from von Dadelszen. P (D. Phil thesis, 2000) mAbs: monoclonal antibodies. 19 The adult respiratory distress syndrome (ARDS) can complicate acute systemic inflammation (93) and preeclampsia specifically (4;95). ARDS is associated with pulmonary capillary leak, in the absence of increased hydrostatic pressure (94). The increased pulmonary capillary permeability is in response to either injury or activated leukocytes within capillaries. Hence, there are several reasons to consider whether or not, during preeclampsia, there is leukocyte activation. 1.3.6 Peripheral blood lymphocytes in normal pregnancy and preeclampsia Peripheral blood lymphocytes in normal pregnancy No consistent pattern of peripheral blood lymphocyte activation has been reported in pregnancy. Sacks et al (75) have reported that lymphocytes in normal third trimester pregnancy have increased intracellular reactive oxygen species (iROS) in comparison with non-pregnancy, von Dadelszen et al (96) found that lymphocytes in normal pregnancy have increased basal intracellular free-ionized calcium compared with non-pregnancy (16). Although plasma IgG levels are decreased in pregnancy, following PWM (pokeweed mitogen) stimulation there is a significant enhancement of IgG production by pregnancy B-lymphocytes in vitro (97). Both B- and T-lymphocyte proliferative responses to either PWM or PHA (post-implantahaemagglutinin) stimulation are reduced in the first half of pregnancy (98). 20 Cell-mediated sensitization of the mother to her fetus is not a regular event during human pregnancy (99). Wegmann and colleagues (100) have suggested that immunological adjustment, which allows trophoblast to invade the uterine spiral arteries occurs by Thl down-regulation and Th2 up-regulation. Thl is mainly responsible for cellular immunity which provides protection against foreign and infected cells. Down-regulation of the Thl element protects the fetus from immune rejection as the fetus carries paternal genes, which are considered to be foreign to the mother (101). Th2 is responsible for humoral and innate immunity which protects against extracellular pathogens. As will be discussed, circulating monocytes and granulocytes are activated in normal pregnancy, possibly to counterbalance the loss of Thl activity during pregnancy. While in normal pregnancy there is a Thl - to Th2 shift, in preeclampsia this shift either does not occur, or is reversed with the onset of clinical disease (102-105). Peripheral blood lymphocytes in preeclampsia (Table 1 . 2 ) In preeclampsia, the total peripheral blood lymphocyte count is largely unchanged and No consistent pattern has been found, concerning lymphocyte subtypes. Sacks et al (75) have found that lymphocytes in preeclampsia showed elevated iROS over those levels found in normal pregnancy. Compared with normotensive pregnancy, in vitro IgG production was significantly increased in women with pregnancy-induced hypertension (PIH) and proteinuria (preeclampsia), 21 but not in those women with PIH without proteinuria (106). von Dadelszen et al found that lymphocytes in women with preeclampsia had increased basal intracellular free-ionized calcium compared with both matched normal pregnancy and non-pregnancy controls (16). 1.3.7 Peripheral blood monocytes in normal pregnancy and preeclampsia Peripheral blood monocytes in normal pregnancy During normal pregnancy, there is a significant monocytosis (107-109). Sacks et al (75) have reported that both basal iROS levels and CD 14 expression were increased in third trimester monocytes compared with non-pregnancy, von Dadelszen et al found that normal pregnancy had similar basal intracellular free-ionized calcium compared with non-pregnancy group (16). There is a progressive increase in monocyte expression of CD 11 a, CD54 and CD64 in normal pregnancy (110). Peripheral blood monocytes in preeclampsia (Table 1.2) Peripheral blood monocyte counts are not altered by preeclampsia (74). Phagocytic cell activity, as measured by a chemiluminescence response, is significantly increased (111). Sacks et al (75) and Gervasi et al (112) found increased iROS levels in unstimulated preeclampsia monocytes, which express more CD14 than in normal pregnancy (75). However, Gervasi et al did not find increase CD 14 but found differences in C D l l b (increased) and CD62L (decreased). Two . 22 of the factors specific to monocyte activity are tumor necrosis factor-alpha (TNF- a), which is pro-inflammatory, and interleukin-1 receptor antagonist (IL-IRa), which is anti-inflammatory. Plasma TNF-a (15;68) and TNF-a mRNA in peripheral blood leukocytes (113) are both increased in preeclampsia, as is circulating plasma IL-IRa (114). von Dadelszen et al found that monocytes in preeclampsia show increased basal intracellular free-ionized calcium compared with both matched normal pregnancy and non-pregnancy controls (16). Macrophage colony-stimulating factor (M-CSF), produced by lymphocytes and endothelial cells, activates monocytes and macrophages which have the appropriate receptor. Its circulating levels have been reported to be increased in preeclampsia, which would be expected to lead to monocyte activation (115). i 1.3.8 Peripheral blood granulocytes in normal pregnancy and preeclampsia Peripheral blood granulocytes in normal pregnancy The numbers of peripheral blood neutrophils increase during normal pregnancy (116-118). As found with lymphocytes and monocytes, third trimester granulocytes contain greater amounts of iROS and express more C D l l b than in non-pregnancy (75). Neutrophil elastase levels, which is an indirect index of neutrophil activation and degranulation, is significantly higher in norrnal 23 pregnancy than in comparison with non-pregnant women, suggesting that neutrophil activation and degranulation are stimulated in normal pregnancy (70). von Dadelszen et al found that granulocytes in normal pregnancy group had similar basal intracellular free-ionized calcium compared with non-pregnancy group (16). They have also reported that normal pregnancy group showed gestation-dependent delayed neutrophil apoptosis and this may explain the neutrophilia of pregnancy (17). Peripheral blood granulocytes in preeclampsia (Table 1.2) Although it is suggested that peripheral blood granulocytes are already activated in normal pregnancy, there seems to be a further superimposed activation in preeclampsia. There is also greater basal (75) and fMLP-induced (79) superoxide production. Basal neutrophil C D l l b expression, which is considered to be an indicator of activation, is significantly greater compared with normal pregnant controls. (74;75). Serum neutrophil elastase is also significantly higher (70; 119). von Dadelszen et al found that granulocytes in women with preeclampsia had increased basal intracellular free-ionized calcium compared with both normal and non-pregnancy groups (16) . They also reported that there is more delayed neutrophil apoptosis in preeclampsia compared with normal pregnancy and that this may help to explain the greater neutrophilia in preeclampsia (17) . 24 In summary, there is general evidence of P B L activation in preeclampsia, which for granulocytes and monocytes is superimposed on the alterations found in normal pregnancy. The activation of PBLs is consistent with current theories of the pathogenesis of preeclampsia, its clinical course, and its complications. 25 Chapter 2: Research pi: 2.1 Hypothesis The hypothesis for this research is that maternal neutrophils and monocytes are inappropriately activated in preeclampsia but not in normotensive intrauterine growth restriction (IUGR). 2.2 Aims We have pursued the following two specific aims to test this hypothesis. We have designed a prospective controlled cohort study of women with preeclampsia and three control groups to test our hypothesis by; (1) measuring neutrophil, monocyte, and lymphocyte activation by assessing surface antigen expression and (2) measuring neutrophil apoptosis. In parallel, we also and investigated levels of placental debris (undertaken by Dr D Goswami, obstetrics and gynecology resident research elective in Oxford, UK) from the same clinical blood samples tested in this thesis (Appendix). 2.3 Subjects These investigations were performed on women in five groups of women. The first two groups of women (cases) were identified antenatally with preeclampsia. These two groups, 27 undelivered women with early onset (<34 weeks) and late onset (>34 + 0 weeks) preeclampsia, met the diagnostic criteria determined by NHBPEP: a pregnancy-specific increase in blood pressure (dBP>90mmHg, or sBP>140mmHg) associated with 0.3g proteinuria/d (5). These women had no known predisposing intercurrent illnesses, such as essential hypertension. Given that clinical and manifestations and risks (both maternal and perinatal) of late-onset preeclampsia are usually milder (120-122), these groups controlled for disease severity as reflected in the gestational age at disease onset. The third group consisted of a control group of antenatal women with normotensive IUGR defined by ultrasound estimations of either fetal weight and/or abdominal circumference <5* centile for gestational age (123). Their data were retained for analysis following delivery only if their infant delivered at <10th centile for gestational age and gender (124). As stated, normotensive IUGR provides a control group of women without a systemic maternal response, thus controlling for the underlying placental pathology in preeclampsia. Fourth were matched normal pregnancy controls. These were healthy women, matched to preeclampsia cases for age (+5y), parity (0, 1, >2) and gestational age (+ 14d) at the time of phlebotomy. These women had no history of major medical problems before the pregnancy. Their data were retained only if the remaindar of their pregnancy was uncomplicated. This group controlled for the influence of gestation on the inflammatory response; this response is known to 28 be gestational age-dependent (17). The fifth group comprised healthy non-pregnancy women (controls). These women were in the reproductive age range (20-40y) and were not using hormonal contraception as this medication causes a pseudopregnant state. Smokers were excluded. This group controlled for general pregnancy effects on the maternal innate inflammatory system. 2.4 Rationale 2.4.1 Rationale: surface antigen expression Sacks et al have reported that increased granulocyte and monocyte CD 1 lb expression in preeclampsia and sepsis, and increased granulocyte and monocyte CD 14 expression increases in normal pregnancy but not in either preeclampsia or sepsis (75). As discussed in Chapter 1, V C A M , the ligand for CDllb/CD18 is increased in normal pregnancy and further increased in preeclampsia. The leukocyte adhesion molecules (CDllb or CDllb/CD18) are 2-integrins involved in neutrophil signaling, and leukocyte-endothelial adhesion. The lipopolysaccharide receptor (CD 14) is a marker of the inflammatory response both in sepsis (127) and pregnancy (75); CD 14 expression is controlled by the cytokines of the Thl and Th2 responses. An increase in CD14 expression has been reported in normal pregnancy. There is evidence to support the presence of 29 functional interactions between these CD molecules and the caspase cascade that controls apoptosis (125;129-135). Therefore, to extend the work done by Sacks et al (75), we designed a study to measure the cell surface antigens of different subtypes of leukocytes (neutrophils, monocytes and lymphocytes) in preeclampsia and, pregnancies complicated by normotensive IUGR and to rejate the findings with markers of disease activity (i.e. early onset vs late onset disease). These experiments were novel as they related CD marker expression to both preeclampsia and nlUGR, as well as CD expression and apoptosis in the same women. \u00E2\u0080\u00A2\u00E2\u0080\u00A2 2.4.2 Rationale: neutrophil apoptosis As stated, apoptosis is a form of cell death and also known as programmed cell death. It is characterized by cell shrinkage, nuclear condensation and DNA fragmentation. Its physiological functions are mainly to eliminate unwanted cells such as damaged cells, to maintain organ size, and to renew tissues. This process is primarily, and tightly, controlled by the caspase family of enzymes, some of which are pro-apoptotic and others of which are anti-apoptotic (125). As previously described, retarded PMN apoptosis occurs in both preeclampsia and SIRS (21), and may contribute to the perpetuation of these syndromes after removal of the inciting stimulus. Therefore, in cases destined to have protracted and more severe disease, PMN apoptosis : 30 may be further retarded than in women with milder clinical courses. Apoptotic PMNs are cleared from the circulation by the reticuloendothelial system once they bind Annexin V to exteriorized phosphatidyl serine (126). Therefore, PMNs are cultured for 18-24h before assessment of apoptosis. In the early phase of apoptosis, the cells lose the asymmetry of their membrane although the cell membrane is still maintained (125). The binding of Annexin V, a calcium and phosphohpid-binding protein, to exteriorized phosphatidyl serine from the internal surface of the cell membrane is the first phenotypic markers of apoptosis (128). As apoptosis advances, PMNs form apoptotic bodies, a process that includes DNA scission (127). The amount of DNA scission can be determined by propidium iodide profile analysis using flow cytometry (17). Therefore, it was decided to investigate apoptosis by testing both annexin V binding and propidium iodide profiles as had been done previously (17), and extend those investigations by relating them with other markers of disease activity (i.e. early onset vs late onset preeclampsia). 31 Chapter 3: Standardization of methods 3.1 Rationale The aim of this section of the thesis was to develop and modify existing methods of flow cytometric peripheral blood leukocyte analysis, so that a standardized protocol could be applied to the investigation of the clinical samples described in Chapter Four. These experiments were designed to investigate peripheral blood leukocyte expression of CD l ib (monocytes and granulocytes), its heterodimer, CD 18 (lymphocytes, monocytes, and granulocytes), and CD 14 (monocytes and granulocytes). CD expression had previously been found to be altered both in normal pregnancy and more so in preeclampsia (75). Also, spontaneous neutrophil apoptosis had been shown to be altered (reduced) in normal pregnancy, with further, perhaps inappropriate reduced apoptosis in preeclampsia, but not in normotensive intrauterine growth restriction (IUGR) (17). However, no previous examination of CD expression had included normotensive IUGR as an experimental group. Similarly, no previous investigation in preeclampsia had correlated CD expression with the processes of neutrophil apoptosis ex vivo. Therefore, two experimental designs were standardized, those for CD l ib , 18, and 14 expression, using rapid flow cytometry (136), and those for neutrophil apoptosis. 3.1.1 Flow cytometry Flow cytometry is a means of measuring certain physical and chemical characteristics of cells or particles as they travel in suspension one by one past a sensing point (137). A flow : 33 cytometer is like a specialized fluorescence microscope. The modern flow cytometer consists of a light source, collection optics, electronics and a computer to translate signals to data. In most modern cytometers, the light source of choice is a laser that emits consistent light,as a specified wavelength. Scattered and emitted fluorescent light is collected by two lenses (one set is front of the light source and one set at right angles) and by a series of optic, beam splitters and filters (Figure 3.1). 34 L A S E R laser steering mirror flow cell blocker bar Figure 3.1 Flow cytometer, principles of design and setup Intense monochromatic light produced by laser, and directed through flow cell. The forward detector receives forward-sccaterd light (proportional to optical size). Those collecting light at right angles (up to four detectors) measure different colors, typically sccatterd primary light (proportional to optical granularity), and green, orange and red fluorescences. Photodetectors convert photon pulses into electronic signals. Electonic and computational processing result in graphic display and statistical analyses of measurements. Modified from Carter & Meyer (1994). 35 Fluophores such as propidium iodide (PI) and fluorescein isothiocyanate (FITC) were used for the flow cytometric analyses. PI is not fluorescent, however, it becomes fluorescent once it binds to double-stranded nucleic acid (137). Dead cells loose their ability to exclude PI although viable cells are impermeable. It is excited at 488nm. (emition 622nm) FITC absorbs light at 460-510 nm and fluoresces at 510-560 nm. FITC can be readily conjugated to surface antigen antibodies and is one of the most common fluorescein compounds for the detection of the various antibodies. LDS-751 was used for rapid flow cytometry .The fluophore, LDS-751, binds to only nuclear DNA; it is excited at 488nm and emits at 630nm (in the red spectrum) (138). The advantage of flow cytometry is that leukocyte populations can be resolved by optical and phenotypic criteria without physical separation. It also allows us to measure physical characteristics as cell size and shape, as well as a wide variety of cellular components and function. The preparation of samples for flow cytometric analysis usually requires a number of centrifugation steps, so that the cells of interest are vulnerable to damage and stimulation during the sample preparation phase (139). Therefore, it is possible that some experimental findings may differ from in vivo, rather reflecting differential vulnerability of cells from differing clinical states to excitation during sample preparation. Recently, rapid flow cytometry, which does not require centrifugation during sample preparation, has been developed, thereby offering 36 a means by which these potential inadvertent cellular activations may be avoided (139). 3.1.2 C D markers and rapid flow cytometry Rapid flow cytometry is a flow cytometric technique, which does not require the centrifugation for the sample preparation. Flow cytometry (EPICS-XL, Coulter, Hileah, FL) with LDS-751 identifies lymphocytes, monocytes, and granulocytes by their constituent DNA (LDS751 positive cells, erythrocytes will remain LDS751 negative) and their respective size and granularity (Figure 3.2). 4> .a Granularity Figure 3.2 The size and granularity characteristics of peripheral blood leukocytes PBLs prepared from E D T A blood by hypotonic lysis and leukocyte identification with LDS-751 (DNA stain). (See text for details) G: Granulocyte, M . Monocyte, L: Lymphocyte 37 In these experiments, monoclonal antibodies conjugated to the fluorescein derivative, fluorescein isothiocyanate (FITC) were used. The monoclonal antibodies were mouse anti-human IgG, raised against the human CD markers, C D l l b , CD18, and CD14. To adjust for non-specific binding of antibodies, negative controls comprised a mixture of FITC-conjugated IgGl, IgG2a epitopes. 38 leee Fluorescence Intensity (574nm) Figure 3.3.b Flow cytometric analysis: CDllb on monocyte 10 ii L FITC-conjugated monoclonal antibody to CDllb and LDS-751 (11.4ju g/ml: dark for 15 min at room temperature. 300 p L distilled water was added and incubated for 30 sec. 300 p L lx PBS was then added. The samples were analyzed at 574 nm (FITC detection) by flow cytometry within 20 min (Red). FTIC-conjugated IgGl was used as negative control (Blue). 39 3.1.3 Apoptosis Flow cytometry is very useful for tracking apoptotic responses. In assessing apoptosis, we can examine three phenotypic stages. First, we can examine the earliest stage of phosphatidylserine being exteriorized through its binding to FITC-conjugated Annexin V (its natural ligand). It is known that the integrity of the cell membrane is maintained in the early phases of apoptosis but the cells lose the asymmetry of their membrane phospholipids. Phosphatidylserine (PS), a negatively charged phospholipid located in the internal surface of the plasma membrane, becomes exposed at the surface. Annexin V, a calcium and phospholipid-binding protein, binds preferentially to PS, with high affinity. Apoptotic cells are stained by annexin V before the dying cell changes its morphology and hydrolyzes its DNA (126-128). Next we can examine nuclear DNA breakdown through the presence of hypodiploid DNA in permeabilized cells that would otherwise be impermeable to the DNA dye, propidium iodide (PI). Then we can examine the loss of cellular membrane integrity as cells die, through their inability to exclude PI passage through non-permeabilized membranes. 40 annexin V cytosol Figure 3.4 Annexin V binding The integrity of the cell membrane is maintained in the early phrase of apoptosis but the cells lose the asymmetry of their membrane phospholipids. Phosphatidylserine (PS) becomes exposed at the surface and binds to Annexin V. 41 1 \u00E2\u0080\u00A2 .\"-if Z \u00E2\u0080\u00A2 \u00E2\u0080\u00A2\u00E2\u0080\u00A2 \u00E2\u0080\u00A2\u00E2\u0080\u00A2 ft * ..\" \u00E2\u0080\u00A2 Annexin V Figure 3.5 Annexin V binding Peripheral venous blood neutrophils (PMNs) were isolated and cultured for 18-24 h. Cultured non-permeabilized PMNs were suspended in binding buffer and incubated with Annexin V-FITC and PI solution in the dark for 10 min. Analysis were by a flow cytometer. Region 1: PI+ve/Annexin V-ve Region 2: PI+ve/Annexin V+ve Region 3: PI-ve/Annexin V-ve Region 4: PI-ve/Annexin V+ve As apoptosis advances, neutrophils form apoptotic bodies, a process that includes DNA fragmentation. Apoptotic neutrophil nuclei can be distinguished from normal neutrophil nuclei by the presence of a hypodiploid amount of DNA by flow cytometry following permeabilization and staining with the vital dye, propidium iodide (PI). 42 e > 188 188B Fluorescence Intensity (622nm) Figure 3.6 Hypodiploid DNA (PI) Peripheral venous blood neutrophils (PMNs) were isolated and cultured for 18-24h. Cultured PMNs were suspended in 500 fi L fluorochrome solution and stored in the dark for 10 min and then analyzed by a flow cytometer. 1: hypodiploid D N A 2: diploid D N A Dr. von Dadelszen had previously standardized the methods for the assessment of apoptosis, and no modification of those experiments was undertaken in this thesis. However, the methods were practiced at least ten times to ensure expertise prior to beginning the analysis of clinical samples. Therefore, the remainder of this chapter will focus on the CD marker methodology, in particular the use of rapid flow cytometry and any influence of time on experimental results. This was in response to the fact that clinical samples were to be drawn at the Children's and Women's Health Centre of BC and transported to St Paul's Hospital for later analysis and culture. 43 3.2 Methods. 3.2.1 Comparison of three methods for rapid flow cytometry Rapid flow cytometry is a flow cytometric technique that does not usually require centrifugation during sample preparation. LDS-751, which is a vital dye is used. LDS-751 binds only to nuclei, its positivity, therefore, will identify leukocytes and the leukocytes will be differentiated on histograms as lymphocytes, monocytes or neutrophils by size and granularity in contrast to the anuclear population of erythrocytes. 3.2.1.1 Methods In response to concerns about leukocyte preparation and inadvertent cell activation, I decided to compare and contrast three established methods of peripheral blood leukocyte preparation to optimize and standardize my experimental design. The three methods were: Hypotonic lysis method: Method-A (136), Unfixed method: Method-B (139) and Whole blood lysis method: Method-C (commercial method) were compared. Each protocol is described below. Method-A: Hypotonic lysis method (Leon W.M.M. Terstappen et al) 10 LtL FITC-conjugated monoclonal antibody to C D l l b , CD18 or CD14 and 20 U L LDS-751 (11.4 g/ml: final concentration 2/ig/mL (Molecular Probes, Oregon, USA)) were 44 added to 40 p. L whole blood drawn into ethylene diamine teraacetic acid (EDTA) tubes (Vacutainer, USA), and then incubated in the dark for 15min at room temperature. 300 fl L distilled water (Baxter corporation, Toronto, Canada) was added and incubated for 30 sec. 300//L lx phosphate buffered saline (PBS) was then added. PBS was made up and prepared to be ph7.4. The samples were analyzed by flow cytometry within 20 min. PBS was prepared mixing solution A (250mL 2xsaline (Sodium, Sigma-Aldrich, Canada), 0.9g K H 2 P 0 4 (Sigma-Aldrich), ph4.4) and solution B (250mL 2 x saline, 0.944g Na 2 HP0 4 (Sigma-Aldrich), ph8.8) and adjusted to ph7.4. Method-B: Unfixed method (Marion G. Macey et al.) 500 p. L LDS751(10// g/mL: final concentration 5 jl g/mL) and 17.3mg phenylmethylsulphonylfluoride (PMSF, Sigma-Aldrich, Canada) were added to 500 /J L whole blood and incubated for 1 min. at room temperature. A 25 fl L aliquot was sampled and 5 ll L FITC-conjugated monoclonal antibody to C D l l b , CD 18 and CD 14 was added, and incubated for 10 min. on ice. lmL cold Hank's buffered saline solution (HBSS, Gibco, NY, USA) with 5mM sodium azide (BDH, Toronto, Canada), 0.5mM PMSF (Sigma-Aldrich) and 1% bovine serum albumin (BSA, Sigma-Aldrich) was added and then analyzed by flow cytometry. 4 5 Method-C: Whole blood lysis method (commercial method) 5 U L FITC-conjugated monoclonal antibody to C D l l b , CD 18 or CD 14 was added to 100 U L whole blood (EDTA) and incubated for 15 min. at room temperature. Cells were washed with lx PBS (1200rpm, 5min, no brake), lysing reagent (Whole Blood Lysing Reagent Kit, Beckman coulter, USA) was added, the cell pellet resuspended and incubated for 1 min. Then they were washed with lx PBS again (1200rpm, 5min). Cells were fixed by 1% formaldehyde and kept in the dark before analysis by flow cytometry. 3.2.1.2 Results Method-A and Method-C permitted sufficient recovery of cells, and their optical distribution was clear (Table. 1). However, Method-B took excessive time to run each sample and it did not permit recovery of an adequate number of cells. The use of PMSF was not practical in our hands, and it is also a highly toxic compound (140). Table 3 The recovery of cells following three different preparation techniques Method-A (n=5) Mean (SD) Method-B (n=l) Method-C (n=3) Mean (SD) # of cell/sec 248.35 (13.55) 6.63 1064.89(112.58) Neutrophil (%) 50.7(5.85) 22.6 42.17(6.45) Monocyte (%) 6.45(1.23) 9.57 5.95 (0.58) 4 6 3.2.1.3 Interpretation The hypotonic lysis method (136) was chosen considering these factors below. First, the PMSF used in Method-B was toxic and difficult to deal with. PMSF, therefore, is not an ideal anticoagulant. Because it uses unlysed blood, a large number of erythrocytes remain. Therefore, the sample analysis takes a long time. Method-C gave us enough neutrophils and clear cell distributions. However, it has been reported that the fixing procedure can affect the antigenicity of certain epitopes (141). This is despite the fact that whole blood is fixed rapidly, thus preventing subsequent ex vivo changes in the expression of C D l l b and 18 (142). Method-A was a simple procedure and was not time-consuming. It allows us to analyze samples immediately after phlebotomy. Having blood exposed only to LDS-751, a vital dye, FITC-conjugated antibodies and physiological buffer, this technique also avoids potential artifacts caused by the use of fixatives, erythrocyte lysing solution or leukocyte preparation procedure, such as Ficoll-Hypaque or dextrin sedimentation (143). There are, unfortunately, disadvantages. Analysis should be completed within 20 minutes after the sample preparation, although, for this thesis, this was not a practical problem. Otherwise, the cells will start dying. 47 3.2.2 Time course It was investigated how the time for which blood is left at room temperature affects subsequent antibody binding. 3.2.3 Methods EDTA-anticoagulated peripheral blood from five subjects was left at room temperature for 1,2 and 3 hours prior to sample preparation. Surface antigen expression of CD l ib, CD 18, and CD 14 was measured. One sample was left at room temperature up to 5 hours. The hypotonic lysis method was used (Section 3.2.1). 3.2.4 Results Examples of the relationship between time and Mean Fluorescence Index (MFI, standard error of the mean (SEM)) is shown in Figure 3.7 (a-c). Figure 3.8 shows the MFI change (CD 14) in each subject over the time course, and that one sample initially unstable explained the initial variability at 0.25h in Figure 3.7.a. All of the epitopes were stable up to 3-5h in all cell types except for Figure 3.7.a at 0.25h. 48 CD14 (Monocytes) 10 8 2 0 I 1 1 1 0.25 1 2 3 hours Figure 3.7.a CDllb (Granulocytes) 2 0 I 1 1 1 0.25 1 2 3 hours Figure 3.7. b CD18 (Granulocytes) 10 0.25 1 2 hours Figure 3.7. c Figure 3.7 Stability of surface antigen expression in samples kept at room temperature for up to 5 hours (n=5). 10 / /L FITC-conjugated monoclonal antibody to C D l l b , CD18 or CD14 and 20UL LDS-751 (11.4#g/ml: final concentration 2 U g/mL were added to 4 0 / / L whole blood, and then incubated in the dark for 15min at room temperature. 300 fx L distilled water was added and incubated for 30 sec. 300// L lx phosphate buffered saline was then added. The samples were analyzed at575nm (FITC detection) by flow cytometry within 20 min. 49 CD14 (Monocytes) 25 0 I 1 1 L 1 0.25 1 2 3 5 hours Figure 3.8 Subjects (n=5) were left at room temperature in different time course (1,2 and 3 hours) and then prepared. 10 jl L FITC-conjugated monoclonal antibody to CD 1 lb, CD 18 or CD 14 and 20 UL LDS-751 (11.4//g/ml: final concentration 2//g/mL were added to 40flL whole blood, and then incubated in the dark for 15min at room temperature. 300// L distilled water was added and incubated for 30 sec. 300uL lx phosphate buffered saline was then added. The samples were analyzed at575nm (FITC detection) by flow cytometry within 20 min. 3.2.5 Interpretation CD14, C D l l b and CD18 expressions were quite stable over the time course, up to 5 hours after phlebotomy. Figure 3.8 shows the extremely high MFI level in subject-3, which probably resulted from technical error. It is considered to be the cause of the high SE in CD 14 at 15 min. 3.3 Summary In response to the results of these standardization experiments, rapid flow cytometry using LDS-751 was used to measure surface antigen expression. As a result of the comparison of the three methods described above, the hypotonic lysis method was used for 50 the sample preparation. Apoptosis was detected by two different approaches: Annexin V binding and hypodiploid D N A profiling using PI. I did not determine whether or not within sample variability was a possible source of error in our experimental system. Previous work by both the van Eeden and von Dadelszen groups had not found this to be a problem. The decision was made to proceed to the investigation of clinical samples from non-pregnant women and women with either normal pregnancy, preeclampsia pregnancy or normotensive IUGR pregnancy. 51 Chapter 4: CD lib, 18 and 14 expression on peripheral blood leukocytes and neutrophil apoptosis in preeclampsia, normotensive IUGR, normal pregnancy and non-pregnancy 52 4.1 Rationale 4.1.1 CD markers Sacks et al have reported that CD l ib expression in preeclampsia and sepsis on both granulocytes and monocytes increases, and that CD 14 expression increases in normal pregnancy but not either preeclampsia or sepsis (75). In order to assess the surface antigen expression, rapid flow cytometry was used. 4.1.2 Apoptosis As discussed in Chapter 2, von Dadelszen et al found that retarded PMN apoptosis occurs in preeclampsia (17). It was decided to investigate apoptosis by testing both Annexin V binding and PI profiles. 4.2 Subjects Five groups of women were investigated. The first two groups of women were identified antenatally with preeclampsia. These two groups, undelivered women with early onset (<34 weeks) and late onset (>34 + 0 weeks) preeclampsia, met the diagnostic criteria determined by NHBPEP: a pregnancy-specific increase in blood pressure (dBP^90mmHg, or sBP^ 140mmHg) associated with ^0.3g proteinuria/d (5). These women had no known 53 predisposing intercurrent illnesses, such as essential hypertension. The symptoms and risks (both maternal and perinatal) of late-onset preeclampsia are usually milder (120-122). Therefore, these groups controlled for disease severity as reflected in the gestational age at disease onset. In my experiments, I did not take into consideration exposure to steroids given women to accelerate fetal lung maturation, as had been done by von Dadelszen et al. This was due to the pilot nature of the project and the fact that the local research group began with these sets of experiments. As will be described, this confounding influence on the inflammatory response may have influenced the results of my studies. The third group was undelivered women with normotensive IUGR defined by ultrasound estimations of either fetal weight and/or abdominal circumference <5th centile for gestational age (123). Their data were retained for analysis following delivery only if their infant delivered at < 10th centile for gestational age and gender (124). As stated, normotensive IUGR provides a control group of women without a systemic maternal response, thus controlling for the underlying placental pathology in preeclampsia. Fourth were matched pregnant controls. These were healthy women with normal pregnancies, matched for age (+5y), parity (0, 1, >2) and gestational age (+ 14d) at the time of phlebotomy. These women had no history of major medical problems before the pregnancy. This group controlled for the influence of gestation on the inflammatory response; this response is known to be gestational age-dependent (17). 54 The fifth group comprised healthy non-pregnant women. These women were in the reproductive age range (20-40y) and were not using hormonal contraception as this medication causes a pseudopregnant state. Smokers were excluded. This group controlled for general pregnancy effects on the maternal innate inflammatory system. For early-onset, late-onset preeclampsia and normotensive IUGR groups, phlebotomy was on either admission (for inpatients) or recruitment (for some inpatients admitted after hours or on weekends and for outpatients) and again within 72 hours after delivery. Not all women had postpartum samples for logistic reasons. These cases and normal controls were identified within either the maternal-fetal medicine service (cases) or family practice obstetric service (normal pregnancy controls) of the Children's and Women's Health Centre of British Columbia (CWHCBC) by Terry Viczko, RN, Shelley Soanes, RN, and Dr Vesna Popovska. Blood tests were performed following informed consent, and in accordance with the UBC Research Review Board and C&W Ethics Board approvals. All blood samples were drawn by the CWHCBC clinical phlebotomy service, and initial processing was performed by the clinical laboratory staff. 55 4.3 Methods 4.3.1 Method: Surface antigen expression Peripheral venous blood was sampled and added immediately to EDTA as anticoagulant. Rapid flow cytometry was used (see Section 3.1.2 for details). Briefly, the fluorescent nuclear vital dye, LDS-751 (final concentration 2mg/mL) and FITC-conjugated monoclonal antibodies raised against the epitopes of interest ( C D l l b , C D 14 or CD 18) were added to whole blood and incubated in the dark for 15 min. Distilled water was added, followed, 30sec later, by flooding with PBS. Samples were analyzed by flow cytometry within 20 min. Negative controls comprised either IgGl or IgG2a isotypes. Size and granularity characteristics and LDS-751 positivity were used to identify lymphocytes, monocytes and PMNs, and mean channel brightness at 575nm measured. 10,000 cells were analyzed twice per antigen per case and per control. 4.3.2 Method: Neutrophil apoptosis 4.3.2.1 Neutrophil isolation and culture Sterile procedures were used throughout neutrophil isolation, culture and analysis. Peripheral venous blood was sampled and immediately added to A C D (acid citrate dextrose solution) as an anticoagulant. Peripheral venous blood neutrophis (PMNs) were isolated by differential erythrocyte sedimentation with dextran (MW 188000, Sigma-Aldrich) and 56 differential centrifugation through a Ficoll-Hypaque density gradient (Sigma-Aldrich), the remaining erythrocytes being lysed by distilled water (Baxter Corporation, Toronto, Canada). The P M N s were cultured at l x l O 6 cells/ml in Dulbecco modified eagle medium ( D M E M , G I B C O , N Y , U S A ) , supplemented with 10% fetal calf serum (Sigma-Aldrich), 1% L-glutamine (Sigma-Aldrich), and 1% gentamycin solution (Gibco, U S A ) . Paired 1ml aliquots of P M N suspension alone (spontaneous apoptosis), in the presence of lOOng/mL mouse anti-Fas antibody (Immunotech, Beckman coulter, U S A ) (induced apoptosis) or with 1 ug/mL endotoxin (Sigma-Aldrich) (inhibited apoptosis) were cultured at 37% in 5% CO2 in air for 18-24h. Anti-Fas induces neutrophil apoptosis while endotoxin inhibits neutrophil apoptosis (17). 4.3.2.2 Surface binding of Annexin V and PI A commercial Annexin V - F I T C kit (Immunotech) was used. Cultured P M N s were suspended in binding buffer and incubated with annexin V - F I T C and PI solution in the dark for 10 min. Analysis was by f low cytometry, using an excitation wavelength o f 488nm and acquiring data from emissions at 520nm. 5000 cells were analysed per f low cytometry experiment. 57 4.3.2.3 Propidium iodide DNA profiles Cultured PMNs were suspended in 500 tl L fluorochrome solution (50 ll g/mL propidium iodide (Sigma-Aldrich), 3.4mM Na citrate (Fisher, NJ, USA), l.OmM ethylenediaminetetraacetie acid (EDTA, Fisher, NJ, USA), and 0.1% Triton X-100 (Sigma-Aldrich) and stored in the dark for lOmin before being analyzed with a flow cytometer. Permeabilization of the membrane was required as apoptotic cells have an intact cell membrane that excludes propidium iodide. Forward (optical size) and side scatter (optical granularity) were measured simultaneously. The propidium iodide fluorescence of individual nuclei with an acquisition of 605-635nm was registered on a logarithmic scale. At least 10,000 events were collected and analyzed. Apoptotic PMN nuclei were distinguished from normal PMN nuclei by their hypodiploid content of DNA. PMN debris was excluded by raising the forward threshold. 4.3.3 Method: Band count Blood smears were made immediately after the phlebotomy into EDTA, fixed by ethanol, and then stained by Wright-Giemsa. 300 neutrophils per slide were identified and cells were counted (x400) manually using a microscope (LABOPHOT-2, Nikon). Band cells were identified by the presence of an immature nuclear pattern. In response to advice from 58 the thesis committee, I began to analyze band forms after I had undertaken eight of non-pregnancy control experiments. Therefore, for the non-pregnancy group, eight samples were not subjected to band form analysis. Therefore, band cell analysis was performed on samples from four more non-pregnant women of reproductive age (20-40y), not using hormonal contraception. 4.3.4 Analyses A priori, nonparametric tests (ie such as Mann-Whitney U, Wilcoxon and Dunn's post test) and analysis of variance (ANOVA) were used for the analyses of this research as biological data were not likely to follow Gaussian distribution Therefore, medians were compared rather than means. Kruskal Wallis test This is a nonparametric unmatched analysis of variance (ANOVA). An ANOVA compares three or more groups and each group does not have to be the same size. If the P value is small, the possibility that the date sets do not differ (\"difference coincides\") will be rejected although it does not mean that every group differs from every other group but at least one group is different from one of the others (144). 59 Mann-Whitney U test This is a nonparametric test and is also called the rank sum test. It compares two unpaired groups. All the values are ranked from low to high. If two values are the same, they will be given the average of the two ranks for which they tie. The smallest number gets \"1\" and the largest gets \"N\" where N is the total number of values in the two groups, then these numbers are added in each group and compared. If the sums are different, P value will be small which means the possibility that the difference coincides is ignored (144). Wilcoxon test This is a nonparametric test that compares two matched groups. It compares the difference between each set of pairs and it also analyzes the difference between the lists. The absolute values of the differences are then ranked from low to high. The ranks of the differences where columm A was higher (positive ranks) are added and the ranks of the difference where columm B was higher (negative ranks) are added. If the sums are different, P value will be small which means the possibility that the difference coincides is ignored (144). 60 Dunn's post test. This compares the difference in the sum of ranks between two groups with the expected average difference considering the number of groups and their sizes within GraphPad Prism, the calculation will give us P values of >0.5, <0.05, <0.01 or O.001. It corrects for multiple comparisons (144). Sample sizes To enable non-parametric comparisons between groups all experimental groups required a mimimal sample size of 6 women. 4.4 Results The characteristics of the women from all groups and clinical characteristics are summarized in Table 4. 61 Table 4.1 Clinical characteristics (n (%), median [range]) Variable E O P E T (n=ll) L O P E T (n=7) nlUGR (n=ll) Normal pregnancy (n=22) Non-pregnancy (n=10) Age 33 [18-42] 35 [30-42] 30 [16-37] 31.5 [23-41] 31.5 [23-40] Primigravid 72.7 71.4 72.7 68.2 G A at 32 37 33 34 ^ ^ ^ ^ ^ ^ ^ ^ ^ sampling (wks) [28, 33] [34, 39] [30, 39] [27, 40] M A P 114 119 93 91 [100, 161] [100, 143] [75, 127] [83, 105] T L C (xl06/ml) 15.2 8.6 10.3 10.0 5.2 (antepartum) [11.2, 17.8] [7.3,15.0] [7.2, 14.0] [7.6, 16.9] [4.2, 9.5] Platelets 184 204 [136,356] [114, 251] AST 28 26 [19,193] [16,46] Uric acid 379 413 [260,505] [325,513] Steroid 10/11 0/7 2/11 0/11 0 10 Absent or 2(20) 0(0) 2(25) 0(0) reversed E D F G A at 33.1 38.1 36.4 40.6 delivery [29.0, 39.7] [35.9, 39.9] [34.3, 40.0] [39.1,41.0] BW<3% ile 8/11 2/7 11/11 0/22 T L C (xl06/ml) 14.2 15.8 12.0 13.3 Day 1-3 [9.52,18.0] [9.75, 20.7] [6.46,17.0] [8.72,17.8] (postpartum) Placental 9(100) 7(100) 6(100) 0(0) abnormality* (n=9) (n=7) (n=6) (n=l) EOPET: early-onset preeclampsia nlUGR: normotencive IUGR L O E T : late-onset preeclampsia NPC: normal pregnant comtrol non-pregnant: non-pregnant control * defined as acute atherosis, syncytial knots, infarction, pervillus thrombosis, villitis of unknown etiology, and advanced villus maturation 62 Differential counts (neutrophils, monocytes and lymphocytes) did not vary between pregnancy groups. The data for CD marker expression; neutrophil apoptosis follows in figures 4.1-4.25. 4.4.1 Surface antigen expression We did not find any difference in CD1 lb level on granulocytes before delivery (Fig. 4.1). However, we found monocyte C D l l b expression similarly increased in pregnancy states including preeclampsia, nlUGR and normal pregnancy compared with non-pregnancy; comparisons of both EOPET and NPC with non-pregnancy reached statistical significance (Figure 4.2). 63 tf) JC .2> JQ O 5.0H c c CO JC u c (0 2 2 . 5 H Q O 0.0-CD11b - antepartum I uu rx o UJ UJ 0. O I or o c T o i o CL JC ,g> \"la ja \u00E2\u0080\u00A25 ioH c c (0 o c CO E J 2 Q O CD11b - postpartum *KWp<0.05 h-UJ 0-o U J I 1 J L kMWp<0.05 H UJ 0-cc o c o Q L 2 I T CD d) \u00C2\u00A32. C o c Group Figure 4.4 CD11 b-postpartum: Monocyte 10U L FITC-conjugated monoclonal antibody to CD1 lb and 20 jU L LDS-751(11.4// g/ml: final concentration 2 / / g/mL were added to 40 / / L whole blood, and then incubated in the dark for 15min at room temperature. 300 U L distilled water was added and incubated for 30 sec. 300// L lx phosphate buffered saline was then added. The samples were analyzed at 575nm (FITC detection) by flow cytometry within 20 min. EOPET: early-onset preeclampsia L O E T . late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.05 **non-pregnancy vs each one of others, by Mann-Whitney U test, p<0.05 ***EOPET vs non-pregnancy, by Dunn's post test, p<0.05 Normal Pregnancy vs non-pregnancy, by Dunn's post test, p<0.05 non-pregnant: non-pregnant control 67 Antenatally, granulocyte CD 18 expression did not differ between groups (Figure 4.5). However, as with C D l l b , monocyte CD18 expression appeared to be higher in all pregnancy groups; statistical significance in comparison with normal pregnancy was reached for all but the EOPET group (Figure 4.6). In contrast, lymphocyte CD18 expression was significantly lower in the EOPET group, compared with other pregnancy groups and the non-pregnancy controls (Figure 4.7). Postnatally, similar results were seen. There were no differences seen groups with respect to CD1 lb expression (Figure 4.8). All pregnancy groups appeared to show increased CD 18 expression compared with non-pregnancy; only the comparison between nlUGR and non-pregnancy failed to reach statistical significance (Figure 4.9). Lymphocyte CD18 expression was significantly lower in the EOPET group, although significantly so only compared with normal pregnancy and non-preganncy (Figure 4.10). 68 CD18 - antepartum 10.0n tn tn o c x: 0) c c \u00C2\u00A3. u c (0 7.5H 5.0H $2 2.5-Q O o.o-I X 1 I X o I t-UJ 0-O w LU O o c o I o 0-z o ) O) CD i Group Figure 4.5 CD18-antepartum: Granulocyte 10 U L FITC-conjugated monoclonal antibody to C D 18 and 20 U L LDS-751 (11.4 a g/ml: final concentration 2 U g/mL were added to 40 fi L whole blood, and then incubated in the dark for 15min at room temperature. 300 U L distilled water was added and incubated for 30 sec. 300ll L lx phosphate buffered saline was then added. The samples were analyzed at 575nm (FITC detection) by flow cytometry within 20 min. EOPET: early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control 6 9 CD18 - antepartum 17.5-(8 15.0-5 c .2> 12.5-1 o c 10.0H c co o c CO 00 Q O 7 . 5 -5 . 0 -2 . 5 -0 .0 X X h-UI 0_ o UJ KWp<0.05 X 'PpfO.OI -L ** MWp<0.05 h UJ 0 . o or o c O Q- 0 Q . C Group Figure 4.6 CD18-antepartum: Monocyte 10 U L FITC-conjugated monoclonal antibody to C D 18 and 20 y. L LDS-751 (11.4 U g/ml: final concentration 2 jU g/mL were added to 40 ju L whole blood, and then incubated in the dark for 15min at room temperature. 300 11 L distilled water was added and incubated for 30 sec. 300// L lx phosphate buffered saline was then added. The samples were analyzed at 575nm (FITC detection) by flow cytometry within 20 min. EOPET: early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.05 * * L O P E T vs non-pregnancy, by Mann-Whitney U test, p<0.05 nlUGR vs non-pregnancy, by Mann-Whitney U test, p<0.05 Normal Pregnancy vs non-pregnancy, by Mann-Whitney U test, p<0.05 ***Normal Pregnancy vs non-pregnancy, by Dunn's post test, p<0.01 70 CD18 - antepartum 4-1 (0 Hi o c c c .c o c (0 00 T-Q O *KWp<0.05 I 1 1 MWp<0.05 LU Q. O LU H LU 0_ O c O 2 D) P C o c Group Figure 4.7 CD18-antepartum: Lymphocyte 10 fi L FITC-conjugated monoclonal antibody to CD18 and 20 fi L LDS-751 (11.4 U g/ml: final concentration 2 u g/mL were added to 40 H L whole blood, and then incubated in the dark for 15min at room temperature. 300 fi L distilled water was added and incubated for 30 sec. 300 fi L lx phosphate buffered saline was then added. The samples were analyzed at 575nm (FITC detection) by flow cytometry within 20 min. EOPET: early-onset preeclampsia L O E T : late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.05 **EOPET vs each one of others, by Mann-Whitney U test, p<0.05 71 CD18 - postpartum 8-i (0 (0 CD 7-c JZ .2> 6-turn JQ \"53 5-c c CO JZ 4 -o c 3-E 00 2-T\u00E2\u0080\u0094 Q 1-O 0-T CD t-tu CL o UJ UJ CL O or o 3 c T o z Q . I c O c Group Figure 4.8 CD18-postpartum: Granulocyte 10 u L FITC-conjugated monoclonal antibody to CD18 and 20 V L LDS-751 (11.4 u g/ml: final concentration 2 U g/mL were added to 40// L whole blood, and then incubated in the dark for 15min at room temperature. 300 fx L distilled water was added and incubated for 30 sec. 300// L lx phosphate buffered saline was then added. The samples were analyzed at 575nm (FITC detection) by flow cytometry within 20 min. EOPET: early-onset preeclampsia L O E T : late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control 72 CD18 - postpartum 20-tn % 15-o c c js io-o c ro 00 T-Q O 'KWp<0.05 X i *MWp<0.05 I Dpf0.05 LU Q. O LU L U 0. CD c O O) 9-c o c Group Figure 4.9 CD18-postpartum: Monocyte 10 fi L FITC-conjugated monoclonal antibody to CD 1 8 and 20 fi L LDS-751 ( 1 1 . 4 fi g/ml: final concentration 2 fi g/mL were added to 40 fi L whole blood, and then incubated in the dark for 15min at room temperature. 300 fi L distilled water was added and incubated for 30 sec. 300 fi L lx phosphate buffered saline was then added. The samples were analyzed at 575nm (FITC detection) by flow cytometry within 20 min. EOPET: early-onset preeclampsia L O E T : late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.05 **EOPET vs non-pregnancy, by Mann-Whitney U test, p<0.05 L O P E T vs non-pregnancy, by Mann-Whitney U test, p<0.05 Normal Pregnancy vs non-pregnancy, by Mann-Whitney U test, p<0.05 ***Normal Pregnancy vs non-pregnancy, by Dunn's post test, p<0.05 73 CD18 - postpartum (A (0 CD C \u00E2\u0080\u00A2*-> JSP 3-i _ .Q O) C c CO c 03 QJ E Q O *KWp<0.05 i 2A X x *MWp<0.05 k**Dp<0.05 i-LU 0-o U J U J D_ or CD D O z 0 ) c o c Group Figure 4.10 GDI8-postpartum: Lymphocyte 10 / /L FITC-conjugated monoclonal antibody to CD 18 and 20 U L LDS-751 (11.4 g/ml: final concentration 2 fi g/mL were added to 40 fi L whole blood, and then incubated in the dark for 15min at room temperature. 300 fi L distilled water was added and incubated for 30 sec. 300 fi L lx phosphate buffered saline was then added. The samples were analyzed at 575nm (FITC detection) by flow cytometry within 20 min. EOPET: early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.05 **EOPET vs Normal pregnancy, by Mann-Whitney U test, p<0.05 EOPET vs non-pregnancy, by Mass-Whitney U test, p<0.05 ***EOPET vs non-pregnancy, by Dunn's post test, p<0.05 74 We did not find any difference of CD 14 expression on monocytes between our groups (Figure 4.12). We found granulocyte CD14 levels in both preeclampsia groups were lower than in normotensive pregnancy and non-pregnant states, with nlUGR appearing to be similar to normal pregnancy and non-preganncy (Figure 4.11). No differences between groups were seen for monocytes (Figure 4.12). Postpartum, CD 14 expression was lower only in LOPET compared with normal pregnancy and non-preganncy (Figure 4.13); CD14 expression was also significantly lower in normal pregnancy compared with non-pregnancy. No differences between groups were seen for monocytes (figure 4.14). 75 CD14 - antepartum 2.CH (0 tn 9 c O) 1.5-C c jS 1.0-c 0.54 Q O 0.0- u CD I r-LU Q. O LU kKWp<0.05 CD i i-LU 0-*MWp<0.05 CD I or CD c CD ( o z CD co Q) Q. i Group Figure 4.11 CD14-antepartum: Granulocyte 10 fx L FITC-conjugated monoclonal antibody to C D 14 and 20 fi L LDS-751 (11.4 fi g/ml: final concentration 2 fl g/mL were added to 40 (1L whole blood, and then incubated in the dark for 15min at room temperature. 300 fi L distilled water was added and incubated for 30 sec. 300 fi L lx phosphate buffered saline was then added. The samples were analyzed at 575nm (FITC detection) by flow cytometry within 20 min. EOPET: early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.05 **EOPET vs non-pregnancy, by Mann-Whitney U test, p<0.05 L O P E T vs non-pregnancy, by Mann-Whitney U test, p<0.05 ***EOPET vs non-pregnancy, by Dunn's post test, p<0.05 L O P E T vs non-pregnancy, by Dunn's post test, p<0.05 76 CD14 - antepartum 5 i w to S c \u00C2\u00A3 4 cu c c (0 JC o c re CD E Q O 2H H T UJ o UJ LU CC o c o 0_ CO (1) c o c Group Figure 4.12 CD14-antepartum: Monocyte 10 fi L FITC-conjugated monoclonal antibody to CD14 and 20/J L LDS-751 (11.4// g/ml: final concentration 2 // g/mL were added to 40 ju L whole blood, and then incubated in the dark for 15min at room temperature. 300 fi L distilled water was added and incubated for 30 sec. 300 fi L lx phosphate buffered saline was then added. The samples were analyzed at 575nm (FITC detection) by flow cytometry within 20 min. EOPET: early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control 7 7 CD14 - postpartum 1.75-1 0> C c CO JZ o c (0 E Q o 1.504 I a c 1.00H 0.75H 0.50-0.254 0.00-'KWp<0.01 k-Dp c X CD c C 20-< 30-10H X c o CL co h-LU CL O LU kKWp=0.0042 tDp<0.05 *MWp<0.05 c o Q . w I I-LU CL T c o c o c o 8- Q . CO ? 0- o SP CD C L 2 1 c non Group Figure 4.16 Hypodiploid Pl-posl part urn Peripheral venous blood neutrophils (PMNs) were isolated and cultured for 18-24h. Cultured PMNs were suspended in 500 u L fluorochromoe solution and stored in the dark for 10 min and then analyzed by a flow cytometer. EOPET: early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant, non-pregnant control \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.01 **EOPET vs LOPET, by Mann-Whitney U test, p<0.05 non-pregnancy vs LOPET, by Mann-Whitney U test, p<0.05 * * * L O P E T vs non-pregnancy, by Dunn's post test, p<0.01 83 100-, Q) > + > C \"I C c < 754 504 254 Annexin V - antepartum *KWp<0.05 i c o C o CL tn i h- I\u00E2\u0080\u0094 LU LU QL QL O O LU T X *Dp<0.05 JL kMWp<0.05 o 3 c O 0_ \u00C2\u00A3 Q-C o c Group Figure 4.17 Annexin V-antepartum Peripheral venous blood neutrophils (PMNs) were isolated and cultured for 18-24h. Cultured PMNs were suspended in binding buffer and incubated with Annexin V-FITC and PI solution in the dark for 10 min. Analysis were by a flow cytometer EOPET. early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.05 * * L O P E T vs nlUGR, by Mann-Whitney U test, p<0.05 Normal Pregnancy vs nlUGR, by Mann-Whitney U test, p<0.05 Non-pregnancy vs nlUGR, by Mann-Whitney U test, p<0.05 ***nIUGR vs non-pregnancy, by Dunn's post test, p<0.05 84 Annexin V - postpartum 100 e 75 + c o c c < 50-25-X X X c I 8. to r\u00E2\u0080\u0094 r -L U L U CL CL o o U J _ J c o CL co or o c c o CL CO o CL c o Q . co CO C o c Group Figure 4.18 Annexin V-pos (par turn Peripheral venous blood neutrophils (PMNs) were isolated and cultured for 18-24h. Cultured PMNs were suspended in binding buffer and incubated with Annexin V-FITC and PI solution in the dark for 10 min. Analysis were by a flow cytometer EOPET: early-onset preeclampsia L O E T . late-onset preeclampsia nlUGR. normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control 85 70 60-\ 50H ^ 40 T3 CO s \"\u00C2\u00B0 30 20H 10H PI (dead) - antepartum tKWp=0.0002 c 8. LU 0-O LU kMWp<0.05 c a CO LU CL a i CC o 3 c a w o o. z T c o cx CO CD c o c Group Figure 4.19 PI (deaa>antepartuni Peripheral venous blood neutrophils (PMNs) were isolated and cultured for 18-24h. Cultured PMNs were suspended in binding buffer and incubated with Annexin V-FITC and PI solution in the dark for 10 min. Analysis were by a flow cytometer EOPET: early-onset preeclampsia L O E T : late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.001 ** non-pregnancy vs each one of others, by Mann-Whitney U test, p<0.05 ***EOPET vs non-pregnancy, by Dunn's post test, p<0.05 nlUGR vs non-pregnancy, by Dunn's post test, p<0.01 Normal Pregnancy vs non-pregnancy, by Dunn's post test, p<0.001 86 PI (dead) - postpartum \u00E2\u0080\u00A2o (0 CD \u00E2\u0080\u00A2o 70-i 60-50-40-30-20-10-0-cz o CL CO h-LU CL O LU 'KWp=0.0008 T *MWp<0.05 1 \u00C2\u00A5 \u00C2\u00BB** T c o Q. CO I h LU CL O c o o. CO t OH CD c c o Q. co O CL C o Q. co i CO p c o c Group Figure 4.20 PI (dead^postpartum Peripheral venous blood neutrophils (PMNs) were isolated and cultured for 18-24h. Cultured PMNs were suspended in binding buffer and incubated with Annexin V-FITC and PI solution in the dark for 10 min. Analysis were by a flow cytometer EOPET: early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.05 **non-pregnancy vs each one of others, by Mann-Whitney U test, p<0.05 ***EOPET vs non-pregnancy, by Dunn's post test, p<0.05 L O P E T vs non-pregnancy, by Dunn's post test, p<0.01 N l U G R vs non-pregnancy, by Dunn's post test, p<0.05 87 Ann V - PI (dead) - antepartum 100-. 5 75. + > c I 50-< u I f 25-Q. ta c o C L to kKWp=0.0044 X X 1 *MWp<0.05 c o C L in c o Q. in c o C L t o C o C L to LU 0-O LU LU a. o 3 o Q . Z Group C O 0) Figure 4.21 Annevin V-PI (dead)-antepartum Peripheral venous blood neutrophils (PMNs) were isolated and cultured for 18-24h. Cultured PMNs were suspended in binding buffer and incubated with Annexin V-FITC and PI solution in the dark for 10 min. Analysis were by a flow cytometer EOPET: early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control \u00E2\u0080\u00A2Between groups, by Kruskal Wallis test, p<0.01 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 E O P E T vs non-pregnancy, by Mann-Whitney U test, p<0.05 L O P E T vs non-pregnancy, by Mann-Whitney U test, p<0.05 Normal Pregnancy, by Mann-Whitney U test, p<0.05 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 L O P E T vs non-pregnancy, by Dunn's post test, p<0.05 Normal Pregnancy vs non-pregnancy, by Dunn's post test, p<0.05 88 Ann V - PI (dead) -postpartum 100-. Group Figure 4.22 Annexin V-PI (dead)-postparttim Peripheral venous blood neutrophils (PMNs) were isolated and cultured for 18-24h. Cultured PMNs were suspended in binding buffer and incubated with Annexin V-FITC and PI solution in the dark for 10 min. Analysis were by a flow cytometer EOPET. early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control 89 I also compared the rate of spontaneous neutrophil apoptosis with gestational age, as had been done by von Dadelszen et al superimposing my data on these of von Dadelszen et al (Figure 4.23), I found that spontaneous neutrophil apoptosis decreased with increasing gestational age in normal pregnancy (as did von Dadelszen et al). nlUGR did not differ from normal pregnancy, as was expected. However, EOPET did not vary, which, as stated previously, was an unexpected finding. 70i ~ 6 0 fe < 50 Q 2 40 jg C =5 30 o a JS 20 104 o o I i o 0 . - \u00E2\u0080\u00A2 r 2 = 0.48 p < 0.001 \u00E2\u0080\u00A2 P E T \u00E2\u0080\u00A2 N P C \u00E2\u0080\u00A2 IUGR \u00E2\u0080\u00A2 non-preg V akiko-EOPET A akiko-LOPET o akiko-NPC o akiko-nlUGR O akiko-non-preg V \u00E2\u0080\u00A2 WO \u00C2\u00B0M A o 0 a o a t A 10 20 30 \u00E2\u0080\u0094i 40 gestational age (wk) Figure 4.23 Spontaneous neutrophil apoptosis delay across gestation Peripheral venous blood neutrophils (PMNs) were isolated and cultured for 18-24h. Cultured PMNs were suspended in 500 fi L fluorochromoe solution and stored in the dark for 10 min and then analyzed by a flow cytometer. Apoptotic population was that hypodiploid DNA. Data points, Individual values; horizontal bar, mean; solid line, linear regression; dashed line, 95% confidence intervals; solid bar in non-preng group: median value. New data superimposed on data from von Dadelszen et al. 1999 (17). Legend-1 (PET, NPC, IUGR, non-preg-1): from von Dadelszen et al. paper (17) Legend-2 (EOPET, L O P E T , n l U G R , N P C , non-preg-2): from this study 90 4.4.3 Band counts We found that the marrow production of neutrophils were increased in all pregnancy states, when compared with non-pregnancy for both antepartum (Figure 4.24.a, b) and postpartum samples (Figure 4.25.a, b). The marrow production of neutrophils in reflected in the concentration of circulating band cells (Figure 4.24.b). That the percentage of band cells was somewhat decreased in EOPET reflected the neutrophilia of preeclampsia, and certainly did not lend support to increased marrow production of neutrophils in preeclampsia above that of normal pregnancy (Figure 4.24.a). As for the other experiments in this thesis, I did not find any reversal in band counts within 72h of delivery (Figure 4.25). 91 #Band-antepartum CO O o o 1500-1 loooH 500 KWpjcO.01 ZL MWp<0.01 r -LU CL o LU I-LU D. 0C CD c O CL z CO a. Group Figure 4.24.a #Band-antepartum %Band-antepartum 10n KWprtO.001 5H MWp<0.({i UJ Q. o LU LU CL O or CD 3 O CL CD Group Figure 4.24.b %Baiui-antepartum Blood smears were made immediately after the phlebotomy and fixed by ethanol, and then stained by Wright-Giemsa. 300 neutrophils per slide were and cells were counted (x400) manually using a microscope. For the non-pregnancy group, four samples were from women not included in the other analyses. EOPET: early-onset preeclampsia LOET: late-onset preeclampsia nlUGR: normotencive IUGR NPC: normal pregnant comtrol non-pregnant: non-pregnant control 92 /=34wk), 3) women with nlUGR), 4) matched normal pregnant women (NPC), and 5) non-pregnant women. An ELISA using the anti-trophoblast antibody NDOG2 was used to measure STBM levels in peripheral venous plasma. Non-parametric analyses, with statistical significance set at p<0.05. Results. Preeclampsia was associated with increased STBM levels (EOPET (median): 44ng/ml, n=12; LOPET: 45ng/ml, n=8) compared with normal pregnancy (18ng/ml, n=14). 125 EOPET STBM levels were higher than LOPET when corrected for gestational age. In nlUGR (18ng/ml n=8) STBM levels were the same as normal pregnancy. Background levels in non-pregnant plasma were 0.49ng/ml, n=10. Conclusions. Increased STBM levels were found in preeclampsia but not nlUGR providing further evidence for their role in the pathogenesis of maternal syndrome. MeSH headings: preeclampsia, pregnancy; fetal growth retardation; placenta; apoptosis, syncytiotrophoblast. 126 Introduction. Preeclampsia remains one of the most common causes of maternal mortality in the developed world (1-6). There are two syndromes in preeclampsia: maternal, characterized by hypertension and proteinuria, and fetal, manifested by intrauterine growth restriction (IUGR). The maternal syndrome defines the disease. Given that incomplete placentation is shared by preeclampsia and normotensive IUGR (7), the latter is the fetal syndrome in isolation. The cogent model for the pathogenesis of the maternal syndrome of preeclampsia describes a process by which a placental factor is released into the maternal circulation, which damages the maternal endothelium, causing a syndrome of systemic endothelial dysfunction (8). It is now apparent that this endothelial dysfunction is part of a wider maternal systemic inflammatory response which occurs in normal pregnancy but is far more intense in preeclampsia(9-ll). The placental factor responsible is not known but candidates include peroxides, eicosanoids, cytokines and syncytiotrophoblast microparticles (STBM).. We have previously shown that STBM prepared from normal placentas cause endothelial cell dysfunction in vitro (12) and in isolated vessels (13) and that pre-eclampsia plasma inhibits endothelial cell proliferation (14). STBM are detectable in the plasma of pregnant women by both flow cytometry and enzyme-linked immunosorbent assay (ELISA) (15) and significantly higher levels were found in women with preeclampsia (15). A significant 127 correlation was found between the plasma concentration of STBM and endothelial inhibition, suggesting that STBM may contribute to the maternal endothelial dysfunction (15). There is also an excess of circulating cellular syncytial debris in preeclampsia (16). The release of syncytiotrophoblast debris into the maternal circulation is thought to be the result of syncytial apoptosis, which is part of a normal process of turnover and repair (17). Syncytiotrophoblast apoptosis is increased in pre-eclampsia (18) and this could explain the increased debris in the maternal circulation. It has been proposed that this increase in apoptosis may result from oxidative stress in the placenta caused by a failure of spiral artery adaptation leading to a poorly developed blood supply (17). The other consequence of this placental pathology is intrauterine growth restriction (IUGR) of the fetus. The poor placentation and fetal growth restriction seen in some cases of preeclampsia however is not unique to this disorder. Similar pathology is also seen in cases of normotensive IUGR. Interestingly, increased syncytial apoptosis has also been reported in these pregnancies (19) which would be expected to result in the increased shedding of syncytiotrophoblast debris. According to hypothesis, this should precipitate the maternal syndrome which it clearly does not. This could be due either to a lack of increased shedding in normotensive IUGR or a difference in the way that the mother's innate immune system and endothelial cells respond in this condition. The purpose of this study was therefore to 128 measure STBM levels in the maternal circulation in normal pregnancy and to compare them with those seen in pre-eclampsia, normotensive IUGR, and non-pregnancy. 129 Methods. This was a prospective case-control study using clinical plasma samples obtained from the maternity services at a tertiary referral centre (Children's and Women's Health centre of British Columbia). These samples were frozen at -80\u00C2\u00B0C and transported to Oxford, UK, for analysis. Preeclampsia was defined by the criteria of the National High Blood Pressure Education Program (20). Only singletons were investigated. IUGR was defined as either an ultrasound estimate of fetal weight or an ultrasound measurement of the fetal abdomen <5th centile for gestational age, confirmed at delivery and associated with neither aneuploidy, structural anomalies, nor congenital infection. The histopathology diagnoses of all women were reviewed, when available, to confirm the presence or absence of abnormal placental findings in cases and controls, respectively. Following informed consent, peripheral venous blood was drawn from the following: 1. 12 women with early onset pre-eclampsia (<34 weeks' gestation), 2. 8 women with late onset pre-eclampsia (>34 weeks' gestation), 3. 8 women with normotensive IUGR (estimated fetal weight <3rd centile for gestational age confirmed postnatally, excluding both aneuploidy and congenital infections), 4. 14 normal pregnant women matched for age, gestation and parity (one control per case in 130 groups 1-3), and 5. Ten non-pregnant women aged 20-40y, not using hormonal contraception. The sample collection was co-ordinated by a dedicated full-time research co-ordinator and was approved by both the University of British Columbia Clinical Research Ethics Board and the Children's and Women's Health Centre of British Columbia Ethics Board. Following informed consent, 5ml of antecubital vein blood was taken antenatally. The , plasma was prepared from this lithium heparin anticoagulated peripheral venous blood by high speed centrifugation, and stored at -80\u00C2\u00B0C for transport from Vancouver to Oxford. The tube containing the plasma was thawed to room temperature and 2ml of plasma was used per sample assay. The sample was topped up with endotoxin free phosphate buffered saline (PBS-E, Sigma, St Louis, MO), ensuring the sample was diluted at least 1:2. The plasma/PBS-E mixture was then transferred to an ultracentrifuge tube (14 x 89mm Ultra-Clear tube, Beckman Coulter, High Wycombe, Bucks, UK). To pellet any STBM, the samples were spun on a Beckman L8-80M Ultracentrifuge at 150,000 xg for 45 minutes at 4\u00C2\u00B0C. This was based on a protocol known to pellet ribosomes. The supernatant was discarded and the final pellet was resuspended in 350 n 10.1% bovine serum albumin (BSA, Research Diagnostics Inc, Flanders, NJ) in PBS-E. The samples were then transferred to 0.7ml screw-top tubes and kept at -80\u00C2\u00B0C until use. 131 Standards for the STBM enzyme-linked immunosorbent assay (ELISA) Syncytiotrophoblast microparticles (STBM) were prepared from normal placentas by a modification of the method of Smith et al (Samarason, 1993) and used as standards for the STBM ELISA. The protein content of the STBM suspension was 9.9 mg/ml. 50 p 1 of this was added to lml of diluting buffer (1% BSA, 0.05% Tween 20 (Sigma) in PBS-E) to make a stock solution of 495 p g/ml. Eight quadruple dilutions from 4000ng/ml down to lng/ml were then prepared, using the diluting buffer. Measurement offree STBM in plasma samples by ELISA The STBM ELISA was developed 'in house' by Dr S Kumar (DPhil Thesis, University of Oxford). A 96-well Maxi Sorp plate (Nunc plasticware, Life Technologies, Paisley, UK) was coated with NDOG2 antibody, at a concentration of 10 p g/ml (in PBS-E), using 100 ii \ per well. NDOG2 is an antitrophoblast antibody which recognizes placental alkaline phosphatase (15). The plate was then incubated overnight at room temperature under moist conditions in a covered box. The following day, the plate was washed five times, by hand, with wash buffer (0.05% Tween 20 in TBS). To prevent any non-specific binding, 300 p 1 of blocking buffer (5% BSA in PBS-E) was added per well and left for at least 3 hours at room temperature. Following this blocking period, the plate was then washed a further five times with wash buffer. The plasma samples and standards were then added in triplicate to the appropriate wells (100 132 \x 1 per well) and incubated overnight at room temperature in moist conditions in a covered box. The following day, the plate was washed 10 times with wash buffer. The final step was an ELISA Amplification System (Gibco BRL, Life Technologies, Paisely, Scotland, UK). This utilized endogenous alkaline phosphatase on the STBM microparticles as the enzyme for the colorimetric reaction. 50 n 1 of neat substrate per well was added and left for 1 hour at room temperature. Without any further washes, 50 ii 1 of neat amplifier per well was then added. The colour started to develop immediately. The plates were read at 2 time points, after 5 minutes, on a M R X Microplate reader (Dynex Technologies, Billinghurst, W Sussex, UK) at 490nm. The best standard curve was obtained only after 5 minutes. The standard curve was used to determine the STBM concentration in each 350 u 1 sample in ng/ml. As the samples had been concentrated by ultracentrifugation, this figure had to be divided by a concentration factor of x/0.35 (where x = volume of plasma in ml) in order to calculate the STBM level in the original plasma sample. Non-parametric (Mann-Whitney U and Wilcoxon, as appropriate) and Kruskal-Wallis ANOVA (multiple comparisons, Dunn's post test) analyses were used for continuous variables, and x f \u00C2\u00B0 r categorical variables. Statistical significance was set at p<0.05. Statistical calculations were made using Prism 3.0 (GraphPad Software Inc, San Diego, CA). 133 Results. The patient details are summarized in Table 1. # of the ## women with early-onset preeclampsia and # of the ## women with late-onset preeclampsia delivered infants below the 3 r d centile for sex and gestational age. Two cases of women identified antenatally as normotensive IUGR were not confirmed postnatally, and their data were removed from the analyses. Histopathology results were available for nine, seven, and six women with pregnancies complicated by early-onset preeclampsia, late-onset preeclampsia, and normotensive IUGR, respectively. All cases had confirmed placental abnormalities (eg acute atherosis, syncytial knots, infarction, perivillus thrombosis, villitis of unknown etiology, and advanced villus maturation). The plasma concentration of STBM for each group is shown in Figure 1. The background levels for the assay are shown in the non-pregnancy samples. As previously reported, preeclampsia, both of early- and late-onset, was associated with increased levels of STBM compared to the normal pregnant controls. However, STBM levels in normotensive IUGR were similar to those seen in normal pregnancy. The influence of gestational age on STBM concentration is shown in Figure 2. There was a linear relationship between STBM concentrations and gestational age for normal pregnancy, which was collinear with the results for normotensive IUGR. 23% of the variation in STBM 134 concentration could be explained by gestational age alone. Using the regression line of STBM concentration in normal pregnancy to determine 'expected values,' individual preeclampsia and normotensive IUGR results can be represented as 'observed/expected' for gestational age. These data are presented in Figure 3, and reveal that early-onset preeclampsia varies more from normal pregnancy than does late-onset disease. There was no relationship between STBM concentrations and parameters of clinical disease severity (mean arterial pressure, total leukocyte count, uric acid, platelet count, mean platelet volume, fibrinogen, aspartate transaminase, alanine transaminase, and plasma albumin) among the preeclampsia cases (data not shown). 135 Discussion. These data are the first to show that the increased concentration of STBM previously noted in preeclampsia (15) is specific to preeclampsia and not shared by the condition which shares the same defects of placentation, namely normotensive IUGR. This implies a central role for increased STBM shedding in either the pathogenesis and/or the maintenance of the maternal syndrome of preeclampsia (12;14-16;21). The characteristic endothelial cell (22) and innate immune cell (10; 11) activation of preeclampsia may well be secondary to this excess circulating trophoblast debris. The adverse effects of STBM on cultured endothelial cell function (12) and isolated small arterial function (13) have been well-described. Perturbation of in vitro endothelial cell function is plasma-specific (14), and the effect of preeclampsia plasma on endothelial cell function is directly related to STBM concentration (15). It appears that the disruption of normal endothelial cell function in vitro is mediated by adhesion molecules expressed on the surface of STBM (23) and not by intrinsic proteases (24). Also, the conditioned medium from endothelial cells co-cultured with STBM fragments activates peripheral blood leukocytes in vitro (25). The results from this study may reflect the differential rate of syncytiotrophoblast apoptosis noted between preeclampsia and normotensive IUGR placentae (19), although Ishihara and colleagues found differences between normotensive IUGR and normal pregnancy placentae 136 that we could not confirm by examination of circulating trophoblast debris. The data points for normotensive IUGR STBM concentrations were collinear with the gestational age vs STBM regression line noted for normal pregnancies. It is important to note that the cases of normotensive IUGR included pregnancies with abnormal uterine arterial (one case) and umbilical arterial (one case: absent end diastolic flow, one case: reversed end diastolic flow) Doppler velocimetry waveforms. We found no association between Doppler study findings and STBM concentrations. We accept that some of the normotensive IUGR cases may have been constitutionally small fetuses, as uterine arterial Doppler studies and placental pathology were not performed on all cases. These data support the mechanistic evidence of changes that are specific to the maternal syndrome of preeclampsia and not shared by the fetal syndrome of preeclampsia in isolation, namely normotensive IUGR. We have previously found that abnormally delayed neutrophil apoptosis (26) and levels of antibodies against atherogenic organisms (27) were specific to preeclampsia, and not found in normotensive IUGR. The Poston group has found biomarkers (leptin, placenta growth factor, the plasminogen activator inhibitor (PAI-l)/PAI-2 ratio, and uric acid) that selectively predict the development of later preeclampsia, whereas reduced ascorbate levels predicted the later development of both preeclampsia and normotensive IUGR (28). 137 We have confirmed the gestational age effect on STBM shedding into the maternal circulation in normal pregnancy, previously noted by us (15). Therefore, although the absolute concentrations of STBM were similar in both early- and late-onset preeclampsia, when corrected for the effect of gestational age (observed/expected data), the increase in circulating placental debris was greater in women with early-onset disease. This may explain, in part, the tendency for early-onset preeclampsia to be more homogeneous in its presentation, more severe, and more frequently associated with both poor placentation and abnormal fetal growth. It appears that early-onset preeclampsia differs considerably from late-onset preeclampsia in this and many other regards (29). What remains unknown is whether or not increased placental debris can be detected in the circulation prior to the onset of clinical disease. This might help to determine whether the excess shedding of STBM into the maternal circulation plays a central role in the development of the maternal syndrome, or whether that excess shedding occurs in response to events such as the development of acute atherosis and plays a central role in the maintenance of the condition. In addition, the pattern of the excessive STBM shedding needs to be determined, as that might reflect recurrent ischaemia-reperfusion injuries during the evolution of acute atherosis. We speculate that placental ischaemia-reperfusion events may underlie the episodic spikes in 138 maternal blood pressure and transient fluctuations in platelet counts and liver enzyme abnormalities that can be observed in women expectantly managed remote from term. Timing phlebotomy to coincide with these transient events in women, and during intervals between them, might help to determine the mechanisms that underlie the deteriorating clinical syndrome that compels clinicians to deliver women remote from term. These findings contribute to our understanding of the central role played by apoptotic trophoblast debris in the maternal syndrome of preeclampsia. They help to differentiate between those mechanisms that are specific to preeclampsia and not shared with normotensive IUGR. 139 Acknowledgements. This study was funded, in Canada, by the Canadian Foundation for Women's Health, the BC Research Institute for Children's and Women's Health, and the BC Women's Hospital and Health Centre Foundation. In the UK, Dionne Tannetta was supported by Action Research. We gratefully acknowledge the help and support of Terry Viczko, Shelley Soanes, and Vesna Popovska for recruiting both cases and controls for this study, and for our medical, nursing, and clinical laboratory colleagues for their support. Finally, we thank Thurl Wilkins for her scientific and technical expertise brought to this and many other projects over decades. 140 References 1. Department of Health. Report on confidential enquiries into maternal deaths in the United Kingdom 1982-1984. 1989. London, HMSO. 2. Department of Health. Report on confidential enquiries into maternal deaths in the United Kingdom 1985-1987. 1991. London, HMSO. 3. Department of Health. Report on confidential enquiries into maternal deaths in the United Kingdom 1988-1990. 1994. London, HMSO. 4. Department of Health. Report on confidential enquiries into maternal deaths in the United Kingdom 1991-1993. 1996. London, HMSO. 5. Department of Health. Why women die. Report on confidential enquiries into maternal deaths in the United Kingdom 1994-1996. 1999. London, HMSO. 6. National Institute for Clinical Excellence. Why women die. Report on confidential enquiries into maternal deaths in the United Kingom, 1997-1999. 2001. London, R C O G Press. 7. Khong TY, de Wolf F, Robertson WF3, and Brosens I. 1986. Inadequate maternal vascular response to placentation in pregnancies complicated by pre-eclampsia and by small-for-gestational age infants. Br J Obstet Gynaecol 93:1049-1059. 8. Roberts JM, Taylor RN, Musci TJ, Rodgers G M , Hubel CA, and McLaughlin M K . 1989. Preeclampsia: an endothelial cell disorder. Am J Obstet Gynecol 161:1200-1204. 9. Sacks,G.P, Studena,K, Sargent,K, and Redman,C.W.G. 1998. Normal pregnancy and preeclampsia both produce inflammatory changes in peripheral blood leukocytes akin to those of sepsis. Am J Obstet Gynecol 179:80-86. 10. Redman,C.W.G., Sacks,G.R, and Sargentl.L. 1999. Preeclampsia: an excessive maternal inflammatory response to pregnancy. Am J Obstet Gynecol 180:499-506. 11. von Dadelszen P, Magee L A , Marshall JR, and Rotstein OD. 2000. The maternal syndrome of preeclampsia: a forme fruste of the systemic inflammatory response syndrome. Sepsis 4:43-47. 12. Smarason,A.K., SargentJ.L., Starkey,P.M., and Redman,C.W. 1993. The effect of placental syncytiotrophoblast microvillous membranes from normal and pre-eclamptic 141 women on the growth of endothelial cells in vitro. Br J Obstet Gynaecol 100:943-949. 13. Cockell AP, Learmont JG, Smarason A K , Redman CWG, Sargent IL, and Poston L. 1997. Human placental syncytiotrophoblast microvillous membranes impair maternal vascular endothelial function. Br J Obstet Gynaecol 104:235-240. 14. Smarason,A.K., SargentJ.L., and Redman,C.W. 1996. Endothelial cell proliferation is suppressed by plasma but not serum from women with preeclampsia. Am J Obstet Gynecol 174:787-793. 15. Knight M , Redman CWG, Linton E A , and Sargent IL. 1998. Shedding of syncytiotrophoblast microvilli into the maternal circulation in pre-eclamptic pregnancies. Br J Obstet Gynaecol 105:632-640. 16. Johansen,M., Redman,C.WG., Wilkins,T., and SargentJ.L. 1999. Trophoblast deportation in human pregnancy\u00E2\u0080\u0094its relevance for pre- eclampsia. Placenta 20:531-539. 17. Huppertz,B., Kingdom,!, Caniggia,!, Desoye,G., Black,S., Korr,H., and Kaufmann,P. 2003. Hypoxia favours necrotic versus apoptotic shedding of placental syncytiotrophoblast into the maternal circulation. Placenta 24:181-190. 18. Allaire AD, Ballenger K A , Wells SR, McMahon MJ, and Lessey BA. 2000. Placental apoptosis in preeclampsia. Obstet Gynecol 96:271-276. 19. Ishihara N, Matsuo H, Murakoshi H, Laoag-Fernandez JB, Samato T, and Maruo T. 2002. Increased placental apoptosis in the syncytiotrophoblast in human term placentas complicated by either preeclampsia or intrauterine growth retardation. Am J Obstet Gynecol 186:158-166. 20. 2000. Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy. Am J Obstet Gynecol 183:S1-S22. 21. Redman CWG and Sargent IL. 2000. Placental debris, oxidative stress and pre-eclampsia. Placenta 21:597-602. 22. Roberts JM, Taylor RN, Musci TJ, Rodgers G M , Hubel CA, and McLaughlin M K . 1989. Preeclampsia: an endothelial cell disorder. Am J Obstet Gynecol 161:1200-1204. 23. Kertesz Z, Linton EA, and Redman CWG. 2000. Adhesion molecules of syncytiotrophoblast microvillous membranes inhibit proliferation of human umbilical 142 vein endothelial cells. Placenta 21:150-159. 24. de Jager CA, Linton E A , Spyropoulou I, Sargent IL, and Redman CWG. 2003. Matrix metalloprotease-9, placental syncytiotrophoblast and the endothelial dysfunction of pre-eclampsia. Placenta 24:84-91. 25. von Dadelszen,R, Hurst,G., and Redman,C.W. 1999. Supernatants from co-cultured endothelial cells and syncytiotrophoblast microvillous membranes activate peripheral blood leukocytes in vitro. Hum Reprod 14:919-924. 26. von Dadelszen P, Watson RWG, Noorwali F, Marshall JC, Parodo J, Farine D, Lye SJ, Ritchie JWK, and Rotstein OD. 1999. Maternal neutrophil apoptosis in normal pregnancy, preeclampsia, and normotensive intrauterine growth restriction. Am J Obstet Gynecol 181:408-414. 27. von Dadelszen P, Magee L A , Krajden M , Alasaly K, Popovska V, Devarakonda R M , Money D M , Patrick D, and Brunham RC. 2003. Titre of antibodies against cytomegalovirus and Chlamydia pneumoniae are increased in early onset pre-eclampsia, compared with late onset pre-eclampsia, normotensive intrauterine growth restriction, and normal pregnancy. Br J Obstet Gynaecol (in press). 28. Chappell L C , Seed PT, Briley A, Kelly FJ, Hunt BJ, Charnock-Jones DS, Mallet AI, and Poston L. 2002. A longitudinal study of biochemical variables in women at risk of preeclampsia. Am J Obstet Gynecol 187:127-136. 29. von Dadelszen P, Magee L A , Roberts JM, and for the VIPER group. 2003. Subclassification of preeclampsia. Hypertens Pregn 22:143-148. ,143 Table I. Clinical characteristics (n (%), median [range]) Variable E O P E T L O P E T nlUGR Normal Non-pregnancy (n-12) (n=8) (n=8) p r e g n a n c y . ( n = 1 Q ) (n=14) Age Primigravid G A at 32.0 38.3 32.4 34.9 sampling [28.3,33.4] [34.0,39.9] [30.0,39.1] [27.9,40.0] (wks) M A P 114 119 [100, 161] [100,143] Platelets 184 204 [136,356] [114,251] AST 28 26 [19,193] [16,46] Uric acid 379 413 [260,505] [325,513] Absent or 2(20) 0(0) 2(25) 0(0) reversed E D F G A at 33.1 38.1 36.4 40.6 delivery [29.0,39.7] [35.9,39.9] [34.3,40.0] [39.1,41.0] Birthweight B W < 3%ile Placental 9(100) 7(100) 6(100) 0(0) abnormality ( n = Q ) ( n = ? ) ( f f = 6 ) ( n = = 1 ) 144 ELISA (ng/ml) 175-150-125-S100-1 CM O O 754 50-25-0' MWu p=0.0197 0 \" MWu p=0.0631 o \u00E2\u0080\u00A2 MWu p=0.0111 o oo T \u00E2\u0080\u00A2 MWu p=0.498 O \u00C2\u00B0\ oo \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 PET PET PET I 1 = O HI 3 c O all PET \u00E2\u0080\u00A2 EOPET \u00E2\u0080\u00A2 LOPET \u00E2\u0080\u00A2 nlUGR \u00E2\u0080\u00A2 NPC \u00E2\u0080\u00A2 nonpreg vs all pregnancy groups p<0.0001 Parameter Value LO PET nlUGR Table Analyzed ELISA (ng/ml) Kruskal-Wallis test P value P<0.0001 Exact or approximate P value? Gaussian Approximation P value summary *** Do the medians vary signif. (P < 0.05) Yes Number of groups 5 Kruskal-Wallis statistic 29.04 Dunn's Multiple Comparison Test Difference in rank sum P value Summary EO PET vs LO PET 1.208 P>0.05 ns EO PET vs nlUGR 10.58 P>0.05 ns EO PET vs NPC 11.51 P>0.05 ns EO PET vs nonpreg 32.08 P < 0.001 MM LOPETvsnlUGR 9.375 P>0.05 ns LO PET vs NPC 10.30 P>0.05 ns LO PET vs nonpreg 30.88 P < 0.001 *** nlUGR vs NPC 0.9286 P>0.05 ns nlUGR vs nonpreg 21.50 P<0.05 * NPC vs nonpreg 20.57 P<0.05 * Figure 1. Peripheral venous blood syncytiotrophoblast microparticle (NDOG2) concentrations in women with early-onset preeclampsia (EO PET), late-onset preeclampsia (LO PET), normotensive intrauterine growth restriction (nlUGR), normal pregnancy (NPC), and nonpregnancy (nonpreg). Horizontal bars represent median values. MWu: Mann-Whitney U test. 145 GA v ND0G2 60-50H T nlUGR \u00E2\u0080\u00A2 NPC I\"* O 9 20H 10-1 1 1 1 I I 1 1 25.0 27.5 30.0 32.5 35.0 37.5 40.0 42.5 GA (wks) C1 Variables Slope 1.501 \u00C2\u00B10.6034 Y-intercept -28.89 \u00C2\u00B1 20.92 X-intercept 19.24 1/slope 0.6660 95% Confidence Intervals Slope 0.2463 to 2.757 Y-intercept -72.40 to 14.62 Goodness of Fit r2 0.2277 Sy.x 10.54 Is slope significantly non-zero? F 6.191 DFn, DFd 1.000,21.00 P value 0.0213 Deviation from zero? Significant Data Number of X values 23 Maximum number of Y replicates 1 Total number of values 23 Number of missing values 0 Figure 2. Concentration of circulating syncytiotrophoblast microparticles (NDOG2) is determined by gestational age (GA) for normal pregnancy controls (NPC) and normotensive intrauterine growth restriction (nlUGR) cases. Linear regression line and 95% confidence intervals presented. 146 observed/expected 13-112-\"5) 11-^ 10 (N \u00C2\u00A9 9 o 9 \u00C2\u00B0 8 2. B O 6 a X 0) T3 CD CD I \u00E2\u0080\u00A2 EOPET \u00E2\u0080\u00A2 LOPET \u00E2\u0080\u00A2 nlUGR \u00E2\u0080\u00A2 NPC L U O L U I -L U Q . o c o 0. z Parameter Value LO PET nlUGR Table Analyzed observed/expected Kruskal-Wallis test P value 0.0218 Exact or approximate P value? Gaussian Approximation P value summary * Do the medians vary signif. (P < 0.05) Yes Number of groups 4 Kruskal-Wallis statistic 9.653 Dunn's Multiple Comparison Test Difference in rank sum P value Summary EO PET vs LOPET 8232 P > 0.05 rts EO PET vs nlUGR 11.45 P > 0.05 ns EOPETvsNPC 15.03 P < 0.05 * LO PET vs nlUGR 3222 P > 0.05 ns LOPETvsNPC 6.794 P > 0.05 ns nlUGR vs NPC 3.571 P > 0.05 ns Figure 3. Early onset preeclampsia (EO PET) differs from normal pregnancy (NPC) and normotensive intrauterine growth restriction (nlUGR) more than does late-onset preeclampsia (LO PET). Expected values for [ N D O G 2 ] calculated from regression equation derived for NPC and nlUGR samples, and corrected for gestational age at sampling.. 147 "@en . "Thesis/Dissertation"@en . "2003-11"@en . "10.14288/1.0091280"@en . "eng"@en . "Reproductive and Developmental Sciences"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Maternal leukocyte CD markers, apoptosis and band forms in preeclampsia"@en . "Text"@en . "http://hdl.handle.net/2429/14664"@en .