Hannah Morgan

Role of Tumor Necrosis Factor-α and Natural Killer Cells in Uterine Artery Function and Pregnancy Outcome in the Stroke-Prone Spontaneously Hypertensive Rat

Women with chronic hypertension are at increased risk of maternal and fetal morbidity and mortality. We have previously characterized the stroke-prone spontaneously hypertensive rat (SHRSP) as a model of deficient uterine artery function and adverse pregnancy outcome compared with the control Wistar–Kyoto. The activation of the immune system plays a role in hypertension and adverse pregnancy outcome. Therefore, we investigated the role of tumor necrosis factor-&agr; in the SHRSP phenotype in an intervention study using etanercept (0.8 mg/kg SC) at gestational days 0, 6, 12, and 18 in pregnant SHRSP compared with vehicle-treated controls (n=6). Etanercept treatment significantly lowered systolic blood pressure after gestational day 12 and increased litter size in SHRSP. At gestational day 18, etanercept improved the function of uterine arteries from pregnant SHRSP normalizing the contractile response and increasing endothelium-dependent relaxation, resulting in increased pregnancy-dependent diastolic blood flow in the uterine arteries. We identified that the source of excess tumor necrosis factor-&agr; in the SHRSP was a pregnancy-dependent increase in peripheral and placental CD3– CD161+ natural killer cells. Etanercept treatment also had effects on placental CD161+ cells by reducing the expression of CD161 receptor, which was associated with a decrease in cytotoxic granzyme B expression. Etanercept treatment improves maternal blood pressure, pregnancy outcome, and uterine artery function in SHRSP by antagonizing signaling from excess tumor necrosis factor-&agr; production and the reduction of granzyme B expression in CD161+ natural killer cells in SHRSP.

Abnormal uterine artery remodelling in the stroke prone spontaneously hypertensive rat

Introduction The stroke prone spontaneously hypertensive rat (SHRSP) is an established model of human cardiovascular risk. We sought to characterise the uteroplacental vascular response to pregnancy in this model and determine whether this is affected by the pre-existing maternal hypertension. Methods Doppler ultrasound and myography were utilised to assess uterine artery functional and structural changes pre-pregnancy and at gestational day 18 in SHRSP (untreated and nifedipine treated) and in the normotensive Wistar-Kyoto (WKY) rat. Maternal adaptations to pregnancy were also assessed along with histology and expression of genes involved in oxidative stress in the placenta. Results SHRSP uterine arteries had a pulsatile blood flow and were significantly smaller (70906 ± 3903 μm2 vs. 95656 ± 8524 μm2 cross-sectional area; p < 0.01), had a significant increase in contractile response (57.3 ± 10.5 kPa vs 27.7 ± 1.9 kPa; p < 0.01) and exhibited impaired endothelium-dependent vasorelaxation (58.0 ± 5.9% vs 13.9 ± 4.6%; p < 0.01) compared to WKY. Despite significant blood pressure lowering, nifedipine did not improve uterine artery remodelling, function or blood flow in SHRSP. Maternal plasma sFLT-1/PlGF ratio (5.3 ± 0.3 vs 4.6 ± 0.1; p < 0.01) and the urinary albumin/creatinine ratio (1.9 ± 0.2 vs 0.6 ± 0.1; p < 0.01) was increased in SHRSP vs WKY. The SHRSP placenta had a significant reduction in glycogen cell content and an increase in Hif1α, Sod1 and Vegf. Discussion We conclude that the SHRSP exhibits a number of promising characteristics as a model of spontaneous deficient uteroplacental remodelling that adversely affect pregnancy outcome, independent of pre-existing hypertension.

Melatonin Increases Fetal Weight in Wild-Type Mice but Not in Mouse Models of Fetal Growth Restriction

Fetal growth restriction (FGR) presents with an increased risk of stillbirth and childhood and adulthood morbidity. Melatonin, a neurohormone and antioxidant, has been suggested as having therapeutic benefit in FGR. We tested the hypothesis that melatonin would increase fetal growth in two mouse models of FGR which together represent a spectrum of the placental phenotypes in this complication: namely the endothelial nitric oxide synthase knockout mouse (eNOS-/-) which presents with abnormal uteroplacental blood flow, and the placental specific Igf2 knockout mouse (P0+/-) which demonstrates aberrant placental morphology akin to human FGR. Melatonin (5 μg/ml) was administered via drinking water from embryonic day (E)12.5 in C57Bl/6J wild-type (WT), eNOS-/-, and P0+/- mice. Melatonin supplementation significantly increased fetal weight in WT, but not eNOS-/- or P0+/- mice at E18.5. Melatonin did, however, significantly increase abdominal circumference in P0+/- mice. Melatonin had no effect on placental weight in any group. Uterine arteries from eNOS-/- mice demonstrated aberrant function compared with WT but melatonin treatment did not affect uterine artery vascular reactivity in either of these genotypes. Umbilical arteries from melatonin treated P0+/- mice demonstrated increased relaxation in response to the nitric oxide donor SNP compared with control. The increased fetal weight in WT mice and abdominal circumference in P0+/-, together with the lack of any effect in eNOS-/-, suggest that the presence of eNOS is required for the growth promoting effects of melatonin. This study supports further work on the possibility of melatonin as a treatment for FGR.

[OP.5D.01] INHIBITON OF TNF-ALPHA SIGNALLING USING ETANERCEPT IMPROVES DEFICIENT UTERINE ARTERY REMODELLING AND PREGNANCY OUTCOME IN THE STROKE PRONE SPONTANEOUSLY HYPERTENSIVE RAT

Objective: Impaired vascular remodelling has been extensively studied in pre-eclampsia but little is known about adaptation of the cardiovascular (CV) system to pregnancy in women with chronic hypertension. We have previously characterised the stroke prone spontaneously hypertensive rat (SHRSP) as a model of deficient uterine artery remodelling and identified an increase in pro-inflammatory TNF &agr; relative to the normotensive WKY strain during pregnancy. Design and method: SHRSP were treated with etanercept (0.8 mg/kg) or vehicle at gestational day (GD) 0, 6, 12 and 18. Animals were sacrificed at GD18. SHRSP, SHRSP treated with etanercept (ETN) and WKY (n=6) were used for vascular studies. An independent set of animals (n = 6) were used for flow cytometry analysis. Results: Etanercept significantly reduced systolic blood pressure in the SHRSP after GD 10 (&dgr;SBP GD 10–21 SHRSP 12.0 ± 4.17 vs. ETN 25.8 ± 4.27 mmHg; p < 0.05). Characterisation of uteroplacental blood flow using Doppler showed that etanercept significantly increased uterine artery diastolic blood flow and significantly reduced resistance index relative to SHRSP (SHRSP 0.79 ± 0.02 vs. ETN 0.61 ± 0.02 resistance index; p < 0.01). Analysis of GD18 uterine arteries using pressure and wire myography showed that etanercept did not change the size of the uterine arteries; but significantly reduced uterine artery contractile ability (SHRSP 57.3 ± 8.75 vs. ETN 35.2 ± 2.19 kPa; p < 0.01) and increased carbachol response (SHRSP 13.8 ± 3.8 % vs. ETN 40.1 ± 3.25 %; p < 0.05). Etanercept significantly increased litter size in the SHRSP (SHRSP 7.80 ± 0.44 vs. ETN 12.75 ± 0.94 fetuses), reduced resorption frequency (SHRSP 66.7% vs. ETN 25.0% dams with resorption) and decreased premature glycogen cell loss from the placenta. We sought to identify the source of excess TNF&agr; in the SHRSP. Inflammatory natural killer (NK) cells (CD3- CD161+) were significantly increased in the SHRSP relative to the WKY in the placenta (WKY 11.6 ± 2.39 vs. SHRSP 659.8 ± 201.2 cells/mg; p < 0.01). Etanercept significantly reduced the number of NK cells in the placenta of the SHRSP (SHRSP 659.8 ± 201.2 vs. ETN 148.0 ± 12.62 cells/mg; p < 0.01). Conclusions: Etanercept improves abnormal uterine artery function in the SHRSP with beneficial effects on pregnancy. Excess TNF&agr; production in this model may come from increased inflammatory NK cells.

[OP.5D.02] Transcriptomic changes in early pregnancy-associated uterine artery remodelling in stroke prone spontaneously hypertensive rat

Objective: Deficient spiral artery remodelling is an early gestational defect linked to a number of pregnancy complications such as pre-eclampsia and gestational hypertension. We have previously demonstrated abnormal uterine artery remodelling at gestational day 18 in the stroke prone spontaneously hypertensive (SHRSP) rat. In this study, we examined the early uterine artery structural and functional changes in SHRSP and normotensive Wistar Kyoto (WKY) rats alongside changes in the transcriptome profile. Design and method: Female SHRSP and female WKY rats were time mated at 12 weeks of age ± 4 days. At gestational day 6 (GD6) uterine arteries were dissected and pressure and wire myography performed to assess vascular structure and functional properties. RNAseq (Illumina platform; ribosome depleted, 50 million reads) was performed in uterine arteries from SHRSP and WKY at GD6 and from non-pregnant rats. RNAseq data (FDR < 0.05, FPKM > 1.0) was interpreted using Ingenuity® Pathway Analysis. Results: No significant structural or functional differences were observed in uterine arteries from SHRSP and WKY rats at GD6; cross-sectional area (57.7 ± 1.2 × 103 &mgr;m2 vs 67.5 ± 4.2 × 103 &mgr;m2), maximal constriction to noradrenaline (41.7 ± 8.1 kPa vs 49.5 ± 9.4 kPa), vasorelaxation to carbachol (76.1 ± 3.9% vs 82.2 ± 3.2%) and SNP (60.5 ± 6.2% vs 58.9 ± 5.8%) (% of maximum constriction) for WKY and SHRSP respectively. RNAseq conducted at GD6, prior to vascular remodelling or functional changes, identified 551 differentially expressed genes between pregnant WKY and pregnant SHRSP; including genes involved in uterine artery vasculogenesis (p-value = 3.78 × 10−10), e.g. angiotensinogen (3.2 fold change, (fc)); angiogenesis (p-value = 3.10 × 10−11), e.g. thrombospondin 4 (4.9 fc); endothelial cell proliferation (p-value = 3.94 × 10−8), e.g. vascular endothelial growth factor D (2.1 fc); and the endothelial nitric oxide synthase pathway (p-value = 3.16 × 10−4), e.g. ryanodine receptor 2 (2.0 fc). Pregnancy resulted in SHRSP-specific increase in expression of NADPH oxidase regulatory subunits p47phox (2.1 fc), p67phox (2.0 fc), p40phox (2.9 fc) p22phox (1.8 fc) and gp91phox (1.8 fc). Conclusions: Uterine artery gene expression during early pregnancy is markedly different between SHRSP and WKY prior to any observed functional and structural changes. Transcriptome profiling implicates pathogenic changes in vasculogenesis and angiogenesis pathways, and endothelial dysfunction, as well as evidence of abnormal oxidative stress response in SHRSP uterine arteries.

175 Inhibiton of Tumour Necrosis Factor Alpha Signalling Improves Vascular Remodelling and Decreases the Pro-Inflammatory and Cytotoxic Phenotype of Peripheral Natural Killer Cells in a Model of Chronic Hypertension in Pregnancy

Objective Pregnancy induces extensive yet relatively rapid remodelling of the cardiovascular (CV) system, however, little is known about this adaptation in women with pre-existing CV disease. We have previously characterised the stroke prone spontaneously hypertensive rat (SHRSP) as a model of deficient uterine artery remodelling and identified an increase in pro-inflammatory TNFÎ ± relative to the normotensive WKY strain during pregnancy. Design and method SHRSP were treated with etanercept (0.8 mg/kg) or vehicle at gestational day (GD) 0, 6, 12 and 18. Animals were sacrificed at GD18. SHRSP, SHRSP treated with etanercept (ETN) and WKY (n = 6) were used for vascular studies. An independent set of animals (n = 6) were used for flow cytometry analysis. Results Etanercept significantly reduced systolic blood pressure in the SHRSP after GD 12 (ΔSBP GD 10–21 SHRSP 12.0 ± 4.17 vs. ETN 25.8 ± 4.27 mmHg; p < 0.05). Analysis of GD18 uterine arteries showed that etanercept significantly reduced uterine artery contractile ability (SHRSP 57.3 ± 8.75 vs. ETN 35.2 ± 2.19 kPa; p < 0.01) and increased carbachol response (SHRSP 13.8 ± 3.8% vs. ETN 40.1 ± 3.25%; p < 0.05). Characterisation of uteroplacental blood flow using Doppler showed that etanercept significantly reduced resistance index relative to SHRSP (SHRSP 0.79 ± 0.02 vs. ETN 0.61 ± 0.02 resistance index; p < 0.01). Etanercept significantly increased litter size in the SHRSP (SHRSP 7.80 ± 0.44 vs. ETN 12.75 ± 0.94 fetuses), reduced resorption frequency (SHRSP 66.7% vs. ETN 25.0% dams with resorption) and decreased premature glycogen cell loss from the placenta. Further, we sought to identify the source of excess TNFÎ ± in the SHRSP. Inflammatory natural killer (NK) cells (CD3- CD161+) were significantly increased in the SHRSP relative to the WKY in the placenta (WKY 11.6 ± 2.39 vs. SHRSP 659.8 ± 201.2 cells/mg; p < 0.01). Etanercept reduced the percentage of NK cells which produced TNFÎ ± in the maternal circulation and placenta in the SHRSP. Additionally, etanercept significantly reduced the number of CD161+ NK cells in the placenta of the SHRSP (SHRSP 659.8 ± 201.2 vs. ETN 148.0 ± 12.62 cells/mg; p < 0.01) by inducing a phenotypic switch to a granzyme Blow, CD161low population. Conclusions Etanercept improves uterine artery function and pregnancy outcome in the SHRSP. We propose that this is through the limitation of both damaging TNFÎ ± release and cytotoxicity from NK cells.