We sought to investigate the mechanisms of action by which pravastatin improves vascular reactivity in a mouse model of preeclampsia induced by overexpression of soluble Fms-like tyrosine kinase-1 (sFlt)-1.
Pregnant CD-1 mice were randomly allocated to tail vein injection with adenovirus carrying sFlt-1 or murine immunoglobulin G2 Fc (control), and thereafter to receive pravastatin (5 mg/kg/d) or water. Mice were sacrificed at gestational day 18. Protein expression of endothelial nitric oxide synthase (eNOS), vascular endothelial growth factor receptor-1, and hemeoxygenase-1 were assayed by Western blot in aorta, liver, and kidneys. Serum total cholesterol concentrations were measured.
Pravastatin up-regulated eNOS expression in the aorta of sFlt-1 mice by nearly 2-fold ( P = .005) to levels similar to control mice. Total cholesterol levels, vascular endothelial growth factor receptor-1, and hemeoxygenase-1 protein expression were similar across groups.
Pravastatin prevents vascular dysfunction in part by up-regulation of eNOS in the vasculature. Our data support a role for statins in preeclampsia prevention.
Preeclampsia complicates approximately 5-8% of pregnancies and is a leading cause of both fetal and maternal morbidity and mortality. Effective primary and secondary prevention strategies for preeclampsia remain elusive, possibly because the precise pathophysiologic mechanisms causing preeclampsia are still poorly understood. An imbalance between angiogenic factors such as vascular endothelial growth factor (VEGF) or placental growth factor (PlGF) and antiangiogenic factors, such as soluble Fms-like tyrosine kinase (sFlt)-1 and soluble endoglin, is now widely accepted to precede preeclampsia, yet causation has to be definitively established.
Preeclampsia shares biologic and pathologic plausibility and risk factors with adult cardiovascular disease, such as obesity, diabetes, and inflammation. Endothelial dysfunction and inflammation are essential for the initiation and progression of both disorders. Additionally, women with history of preeclampsia are predisposed to chronic hypertension, stroke, and cardiovascular morbidity later in life, as well as death from cardiovascular disease.
In contrast to the lack of effective preventive treatment options for preeclampsia, 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) reductase inhibitors, or statins have been shown in multiple studies to effectively reduce cardiovascular morbidity and mortality. These effects were noted even in individuals with only modestly elevated cholesterol levels, suggesting that statins, besides their lipid-lowering actions, are cardioprotective by other pleiotropic or cholesterol-independent mechanisms. Specifically, statins have been shown to up-regulate endothelial nitric oxide synthase (eNOS) in vivo and in vitro, and to induce in a concentration and time-dependent manner the expression of the antioxidant and antiinflammatory hemeoxygenase (HO)-1 in endothelial, vascular smooth muscle, and other cells. Activation of the latter pathway has been shown to inhibit the release of sFlt-1 and soluble endoglin from endothelial cells and placental explants, suggesting a possible role in preeclampsia.
Using an animal model of preeclampsia induced by overexpression of sFlt-1, we and others have shown that pregnant mice overexpressing sFlt-1 develop high blood pressure, altered vascular and endothelial responses, and other manifestations of preeclampsia. Additionally we have shown that treatment during pregnancy with pravastatin, a hydrophilic HMG-CoA reductase inhibitor in mice overexpressing sFlt-1, improves vascular function and reverses the angiogenic imbalance associated with preeclampsia by decreasing sFlt-1 concentrations. We hypothesize that in addition to sFlt-1 reduction, pravastatin treatment increases expression of eNOS, VEGF receptor (VEGFR)-1, and HO-1. Therefore, our aim in this study is to investigate the potential mechanisms of action by which pravastatin exerts its vascular protective effects, using this well-characterized murine model of a preeclampsia-like condition.
Materials and Methods
The Institutional Animal Care and Use Committee at the University of Texas Medical Branch, Galveston, TX, approved the study protocol and all procedures. Timed pregnant CD-1 mice were obtained from Charles River (Wilmington, MA). All animals were individually housed in temperature- and humidity-controlled facility, with automatically controlled 12-hour light and dark cycles. Mice were allowed to consume regular chow and water ad libitum. Certified personnel and veterinary staff provided regular maintenance and animal care according to Institutional Animal Care and Use Committee guidelines. The animals were sacrificed by carbon-dioxide inhalation per the American Veterinary Medical Association guidelines.
The preparation of adenovirus carrying sFlt-1 and murine immunoglobulin G2 Fc (mFc) fragments, as well as the generation and validation of this preeclampsia-like model have been described in detail elsewhere. In brief, we used a replication-deficient recombinant adenovirus vector that leads, after a single injection, to hepatocyte transduction which allows sustained in vivo transgene expression of sFlt-1 and its secretion into the systemic circulation, by utilizing the cellular machinery. At day 8 of gestation, we randomly allocated mice to receive injection via the tail vein of either adenovirus carrying sFlt-1 (10 9 plaque-forming U in 100 μL) or adenovirus carrying the mFc fragment (10 9 plaque-forming U in 100 μL) as the virus control. The next day, gestational day 9, mice from both groups were assigned to receive either pravastatin (Sigma-Aldrich, St Louis, MO) dissolved in drinking water at a dose of 5 mg/kg/d or water alone (control group). The concentration of pravastatin used was based on the daily water consumption of pregnant mice, as determined in a preliminary study. Four groups were thus available for comparison (n = 4-6 mice per group): sFlt-1- and mFc-injected mice treated with pravastatin (sFlt-1-Pra and mFc-Pra, respectively) and sFlt-1- and mFc-injected mice drinking plain water (sFlt-1 and mFc groups, respectively). There was no difference in the drinking behavior or amount of water intake per day among the 4 groups.
On day 8 of gestation, blood (100 μL) was collected from all mice from the tail vein and centrifuged for 10 minutes at 3000 rpm. The serum was collected and stored at –80°C until time of testing. On day 18 of gestation, animals were sacrificed and maternal and pup weights and litter counts were recorded. Maternal blood was collected by cardiac puncture after opening the thoracic and abdominal cavities, and serum collected and stored as above. Liver, kidney, and aorta were harvested and stored at –80°C.
Serum total cholesterol levels were determined using colorimetric enzyme-linked immunosorbent assay kits (Wako Chemicals USA, Richmond, VA) according to manufacturer’s instructions and an automated spectrophotometer (Fusion 5.0; Ortho Clinical Diagnostics, Rochester, NY). All samples were batched and run in duplicate.
For Western blot analysis, tissues were homogenized and protein extracted using a Bullet Blender (Next Advance, Inc, Averill Park, NY) and radioimmunoprecipitation assay buffer. Proteins were quantified using BCA Protein Assay Kit (Pierce, Rockford, IL). Equal protein samples were loaded on sodium dodecyl sulfate polyacrylamide gels (Bio-Rad, Hercules, CA) and run according to the manufacturer’s instructions. Polyvinylidene fluoride membranes (Invitrogen Corp, Carlsbad, CA) containing protein were incubated overnight with the following antibodies at optimized dilutions: monoclonal mouse antibodies for eNOS type III (dilution 1:500; BD Biosciences, Franklin Lakes, NJ), rabbit monoclonal VEFGR-1 antibody (dilution 1:500; Abcam Inc, Cambridge, MA), and HO-1 mouse monoclonal antibody (dilution 1:250; Enzo Life Sciences, Plymouth Meeting, PA). Chemiluminescence was detected using the Amersham ECL Plus detection reagent (GE Healthcare, Little Chalfont, Buckinghamshire, UK). Detected bands were then analyzed densitometrically and normalized to monoclonal anti-β-actin at a dilution of 1:15,000 (Sigma-Aldrich) using ImageJ 1.44 software (U.S. National Institutes of Health, Bethesda, MD).
The data are reported as mean ± SEM. Statistical analyses were performed using either Kruskal-Wallis with Dunn post hoc tests or 1-way analysis of variance as appropriate, after assessing normality. For the comparisons between days 8-18 of gestation in each group, paired Student t test was used. Prism 5.0c for Mac OS X (v. 5.0c; GraphPad Software Inc, La Jolla, CA) was used. Two-tailed P values < .05 were considered significant.
Maternal weights at baseline (day 8) and at sacrifice (day 18), gestational weight gain, litter counts, and total litter weights were not significantly different among the 4 groups of mice (mFc, sFlt-1, sFlt-1-Pra, and mFc-Pra, respectively) ( Table ).
|Maternal weight gain gestational day 8, g||32.4 +/− 1.5||29.9 +/− 0.9||28.7 +/− 0.8||29.7 +/− 1.5||.43|
|Maternal weight gain gestational day 18, g||51.0 +/− 2.4||48.5 +/− 2.4||46.8 +/− 3.9||45.6 +/− 2.3||.17|
|Maternal weight gain, g||18.8 +/− 2.0||18.7 +/− 1.8||18.2 +/− 3.3||15.8 +/− 0.97||.25|
|Mean litter weight, g||15.5 +/− 1.4||17.5 +/− 1.6||17.8 +/− 1.4||15.7 +/− 2.0||.49|
|Mean pup count, #||12.58 +/− 0.3||12.10 +/− 1.1||12.52 +/− 0.4||11.83 +/− 0.6||.79|
Serum total cholesterol levels (mg/dL) were not significantly different between groups on day 8 or 18. In all groups, cholesterol levels decreased slightly over time, although not statistically significantly, with the exception of the mFc control group ( P = .02) ( Figure 1 ).