Objective
We previously demonstrated that prenatal nicotine exposure decreases neonatal pulmonary function in nonhuman primates, and maternal vitamin C supplementation attenuates these deleterious effects. However, the effect of nicotine on placental perfusion and development is not fully understood. This study utilizes noninvasive imaging techniques and histological analysis in a nonhuman primate model to test the hypothesis that prenatal nicotine exposure adversely effects placental hemodynamics and development but is ameliorated by vitamin C.
Study Design
Time-mated macaques (n = 27) were divided into 4 treatment groups: control (n = 5), nicotine only (n = 4), vitamin C only (n = 9), and nicotine plus vitamin C (n = 9). Nicotine animals received 2 mg/kg per day of nicotine bitartrate (approximately 0.7 mg/kg per day free nicotine levels in pregnant human smokers) from days 26 to 160 (term, 168 days). Vitamin C groups received ascorbic acid at 50, 100, or 250 mg/kg per day with or without nicotine. All underwent placental dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) at 135–140 days and Doppler ultrasound at 155 days to measure uterine artery and umbilical vein velocimetry and diameter to calculate uterine artery volume blood flow and placental volume blood flow. Animals were delivered by cesarean delivery at 160 days. A novel DCE-MRI protocol was utilized to calculate placental perfusion from maternal spiral arteries. Placental tissue was processed for histopathology.
Results
Placental volume blood flow was significantly reduced in nicotine-only animals compared with controls and nicotine plus vitamin C groups ( P = .03). Maternal placental blood flow was not different between experimental groups by DCE-MRI, ranging from 0.75 to 1.94 mL/mL per minute ( P = .93). Placental histology showed increased numbers of villous cytotrophoblast cell islands ( P < .05) and increased syncytiotrophoblast sprouting ( P < .001) in nicotine-only animals, which was mitigated by vitamin C.
Conclusion
Prenatal nicotine exposure significantly decreased fetal blood supply via reduced placental volume blood flow, which corresponded with placental histological findings previously associated with cigarette smoking. Vitamin C supplementation mitigated the harmful effects of prenatal nicotine exposure on placental hemodynamics and development, suggesting that its use may limit some of the adverse effects associated with smoking during pregnancy.
Approximately 12% of women smoke during pregnancy, corresponding to more than 450,000 smoke-exposed infants born in the United States per year. Smoking during pregnancy poses a serious health risk to the fetus; it has been estimated to cause 5-10% of all fetal and neonatal deaths.
Nicotine is the major component of cigarettes and a key mediator of the negative effects of smoking. It readily crosses the placenta and accumulates in the amniotic fluid. Given the addictive nature of nicotine and lack of effective treatments for smoking cessation in pregnancy, it is an issue that is not likely to be resolved in the near future. Thus, looking for methods to lessen the impact of smoking during pregnancy is important, even while maintaining a focus on smoking reduction and cessation in pregnancy.
It has been previously demonstrated in both our pregnant nonhuman primate (NHP) model and in a rat model that prenatal nicotine exposure produces changes in an offspring’s pulmonary function (eg, decreased expiratory flow) similar to those in the offspring of human smokers. In addition, supplementation with vitamin C is found to ameliorate the harmful effects of smoking on fetal lung development.
Because some of these adverse changes correspond with changes induced by oxidants, it can be expected that natural antioxidants, such as vitamin C, will be protective to both mother and baby. Despite the protective benefit of vitamin C on nicotine exposure on fetal pulmonary function, its effect on placental development and perfusion has not been studied.
The mechanism and specific vasoactive effects of nicotine on the placenta and fetal development have not been identified. Nicotine acts via nicotinic receptors, which are widely distributed not only in the fetal lung but also in the placenta. Nicotine freely crosses the placenta and binds to the α-subunit of nicotinic acetylcholine receptors, expressed in all placental cell types along with endogenous acetylcholine, an essential placental signaling molecule. Nicotinic acetylcholine receptors are also located in the placental vessels in which nicotine is thought to be involved in the cholinergic regulation of placental blood flow and fluid volume.
Understanding the impact of nicotine exposure on placental perfusion and development is critical to reduce and prevent its deleterious effects. Doppler ultrasound (D-US) is an established antenatal surveillance method routinely used in the clinical management of pregnancy. It has the ability to semiquantitatively evaluate uterine blood flow for assessment of fetal well-being but not comprehensive placental perfusion.
We have recently developed a complementary noninvasive method of advanced imaging, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), to quantify blood flow to the entire placental unit simultaneously. Prior small animal studies have used DCE-MRI for the evaluation of placental blood flow by acquiring T 1 -weighted placental images prior to, during, and following the administration of a gadolinium-based contrast reagent.
In rodents, previous studies have quantified placental perfusion and maternal-fetal interface permeability, including a recent study that demonstrated different perfusion regions in the mouse placenta. Anatomically the primate placenta is distinct from those of small animals and is hemochorial like humans. Nonhuman primates also have a longer gestation and developmental ontogeny resembling that of humans. Given the similarity in vascular organization and development of the placenta, the NHP provides a powerful translational model for human pregnancy studies.
Our study focuses on the effects of chronic maternal nicotine exposure with and without vitamin C supplementation in a relevant NHP model on placental perfusion and development. We hypothesized that nicotine exposure during pregnancy would decrease placental perfusion and increase placental pathology reportedly associated with cigarette smoking and that these adverse effects would be ameliorated by maternal Vitamin C supplementation.
Materials and Methods
Experimental design
Time-mated female Rhesus macaques (n = 27) were randomly allocated to control (n = 5), nicotine only (n = 4), vitamin C only (n = 9), or nicotine plus vitamin C groups (n = 9). On gestational day 26 of pregnancy, control animals received subcutaneous (midscapular) Alzet miniosmotic 2ML4 pumps containing bacteriostatic water (Abbott Laboratories, North Chicago, IL), and nicotine animals received subcutaneous osmotic minipump infusions of nicotine bitartrate (Sigma, St. Louis, MO) in bacteriostatic water to deliver 2.0 mg/kg per day nicotine bitartrate. Pumps were replaced every 3 weeks and animals received cefazolin (150 mg, twice daily) for 3 days after pump insertion.
The vitamin C–treated group received supplemental ascorbic acid at 50, 100, or 250 mg/kg per day in the form of a children’s chewable vitamin. Standard dosing of vitamin C in human prenatal vitamins is 90–120 mg. Given the low toxicity of vitamin C in pregnancy, a range of doses were chosen to assess dose-response effects. All groups were fed the Purina Mills (Richmond, IN) high-protein monkey diet, containing 0.75 mg of vitamin C per gram of diet, with an average daily vitamin C intake from the diet of 50 mg. In addition, all groups received multivitamins twice weekly (Zoo Chews; IVC Industries, Portland, OR) containing 60 mg of vitamin C.
All animal procedures were approved by the Oregon Health and Science University Institutional Animal Care and Use Committee and conformed to all applicable regulations.
All animals underwent D-US at 155 days of gestation and placental DCE-MRI at 135–140 days’ gestation. Cesarean delivery was performed at gestational day 160 (term is 168 days) to ensure delivery of offspring prior to the onset of spontaneous labor. At the time of delivery, maternal blood, cord blood, and amniotic fluid were collected for metabolite analyses. Nicotine and cotinine (metabolite of nicotine) levels were measured by gas chromatography, as previously described. Placental weights and measures were recorded.
Imaging
Doppler ultrasound
D-US data were collected by a sonographer blinded to the treatment group using image-directed pulsed and color Doppler equipment (GE Voluson 730 Expert; Kretztechnik, Zipf, Austria) with a 5–9 MHz sector probe. Prior to D-US, animals were sedated with 10 mg/kg ketamine. The lowest high-pass filter level was used (100 Hz), and an angle of 15° or less between the Doppler beam was deemed acceptable. Blood flow velocity waveforms were obtained from the proximal portion of the uterine artery (Uta) as previously described.
Doppler waveforms measurements for the Uta, umbilical artery, and ductus venosus were performed using machine-specific software and the following measurements were obtained: pulsatility index (PI), velocity time integral, and fetal heart rate. The diameter of the Uta was measured using power angiography as previously described. The cross-sectional area (CSA) of the vessel was calculated as CSA = π(diameter/2) 2 . The calculated uterine artery blood flow volume (cQ Uta ) was calculated using the following formula: cQ Uta = velocity time integral × CSA × heart rate. For the placental volume blood flow (cQ UV ), the Doppler waveforms were obtained from the straight portion of the intraabdominal umbilical vein (UV) as previously described. The mean velocity was calculated as 0.5 of the maximum velocity. cQ UV was calculated as: mean velocity × CSA × 60.
Dynamic contrast-enhanced MRI
DCE-MRI data were acquired on a 3T Siemens Tim Trio scanner (Siemens, Erlangen, Germany) dedicated for NHP imaging, equipped with a high-resolution, 15-channel extremity RF coil, as described previously. Prior to MRI, animals were sedated with 10 mg/kg ketamine, intubated, and maintained on 1.5% isoflurane. Subsequent to anatomic imaging using standard T1- and T2-weighted sequences for the localization of the placenta, a time series of highly T1-weighted, 3-dimensional spoiled gradient echo measurements spanning the entire uterus with 2.5 mm isotropic spatial resolution was acquired (repetition time, 86 milliseconds; echo time, 1.5 milliseconds; flip angle, 20°) with 5.5 seconds of temporal resolution. Data were acquired for at least 8 minutes, with a standard dose of 0.1 mmol/kg of gadoteridol (Prohance; Bracco Diagnostics Inc., Princeton, NJ) injected as a bolus at the end of the 10th time frame.
The custom postprocessing algorithm, implemented in MATLAB (Mathworks, Natick, MA) and previously described was slightly modified, as described in the following text, to improve robustness and performance. After manually generating a region of interest covering each lobe of the placenta, curves of contrast uptake were fit via nonlinear regression to the gamma capillary transit time model to obtain parametric maps of model flow (F) and contrast bolus delay time. The F map was smoothed by 3-dimensional convolution with a Gaussian of SD 3 mm to minimize spatial noise.
Local maxima were identified in the smoothed F map and used to initialize a region-growing segmentation algorithm based on the fast marching method, as implemented in MATLAB File Exchange submission number 24531 by Dirk-Jan Kroon). Once individual perfusion domains were segmented, the average volume of the region growing shells within each domain was plotted against the corresponding median delay time and resulting curves fit via linear regression to obtain volumetric flow in milliliters per minute. The total volume of each perfusion domain, determined by multiplication of the number of voxels contained within the segmented domain mask by the known magnetic resonance imaging voxel volume, was used to compute normalized blood flow (milliliters per minute per milliliter), defined as the total blood flow for the domain divided by the domain volume.
Histological analyses
Representative samples from each fresh placenta were collected using 4-quadrant sampling and fixed in 10% zinc formalin and paraffin embedded, and histological sections were stained for hematoxylin and eosin. A placental pathologist (T.K.M.) evaluated the placental histological sections while blinded to treatment group and scored for the presence of specific pathological features previously reported to be associated with smoking including villous cytotrophoblast cell islands related to abnormal differentiation, villous capillary density (with and without chorangiosis), and syncytiotrophoblastic sprouting. Findings had to be present in multiple villi in multiple sections for each case to be considered positive.
Immunohistochemistry
Immunochemistry was performed as previously described. Primary antibodies were as follows: 4-hydroxyalkenals,4-hydroxynonenal (4-HNE) mouse monoclonal (diluted to 1:250; Abcam, Cambridge, MA), 8-hydroxydeoxyguanosine (8-OHdG) goat polyclonal (diluted to 1:250; Millipore, Billerica, MA), and Ki67 mouse monoclonal (diluted to 1:500; Dako, Carpinteria, CA). Slides were incubated with the primary antibody overnight at 4°C, followed by 1:10,000 horseradish peroxidase–conjugated secondary antibody (Jackson ImmunoResearch Laboratory, West Grove, PA) for 60 minutes at room temperature prior to standard color reaction and hematoxylin counterstain procedures. Immunostained slides were examined by 2 blinded reviewers. Ten random high-power fields were observed per slide and graded using a 0–3 scale in which 0 indicated the absence of positive staining and 3 indicated intense positive staining.
Statistical analysis
Data are expressed as mean ± SD. All animals (n = 27) were analyzed, unless otherwise noted. Three different vitamin C dosing groups were studied, but because no differences were observed between groups, they were combined for all analyses of in vivo imaging and in vitro studies. For the maternal and fetal imaging data, differences between all groups were tested by a 1-way ANOVA. Fisher exact analysis was used for placental pathological analysis, and an independent sample t test was used for an immunohistochemistry analysis with significance designated at P < .05.
Results
Nicotine and cotinine levels in maternal serum, amniotic fluid, and cord blood at the time of delivery are shown in Table 1 . Maternal plasma levels were similar in both groups. Nicotine and cotinine were undetectable in the amniotic fluid from control animals. Nicotine levels were significantly higher in the amniotic fluid of the vitamin C plus nicotine group, whereas cotinine showed a less notable increase, compared with the nicotine-only group.
| Group | Maternal plasma | Amniotic fluid | Cord plasma | |||
|---|---|---|---|---|---|---|
| Nicotine | Cotinine | Nicotine | Cotinine | Nicotine | Cotinine | |
| Nicotine | 29.8 ± 8.4 | 101.1 ± 18.1 | 13.1 ± 1.0 | 74.6 ± 11.9 | 10.7 ± 1.9 | 92.7 ± 12.9 |
| Nic plus vit C | 28.8 ± 6.0 | 126.7 ± 7.9 | 24.8 ± 3.8 a | 97.2 ± 2.6 | 15.3 ± 3.2 | 120 ± 11.8 |
a P < .05 by Tukey-Kramer test after multivariate ANOVA compared with nicotine alone.
Maternal, fetal birth, and placental weights are given in Table 2 . There were no significant differences in these parameters between any of the 4 groups. By D-US, we demonstrated a significant reduction in the nicotine group compared with each of the other 3 groups ( P = .024; Figure 1 ). There was no significant difference in cQuta as well as the ductus venosus, umbilical artery, and uterine artery pulsatility indices ( Table 3 ).
| Parameter | Control (n = 3) | Vit C (n = 8) | Nicotine (n = 3) | Nic plus Vit C (n = 8) |
|---|---|---|---|---|
| Maternal weight, kg | 8 ± 0.26 | 8.7 ± 1.1 | 8.5 ± 0.57 | 8.1 ± 0.63 |
| Fetal birthweight, g | 469 ± 67 | 524 ± 53 | 504 ± 38 | 482 ± 24 |
| Placental weight, g | 153 ± 36 | 147 ± 22 | 150 ± 19 | 122 ± 20 |
| Parameter | Control (n = 5) | Vit C (n = 9) | Nicotine (n = 4) | Nic plus Vit C (n = 9) |
|---|---|---|---|---|
| Ductus venosus PI | 0.28 ± 0.12 | 0.31 ± 0.1 | 0.27 ± 0.08 | 0.34 ± 0.13 |
| Umbilical artery PI | 0.96 ± 0.06 | 1.1 ± 0.14 | 1.05 ± 0.26 | 1.1 ± 0.16 |
| Uterine artery PI | 0.88 ± 0.31 | 0.84 ± 0.15 | 0.75 ± 0.11 | 0.82 ± 0.14 |
| cQuta, mL/min per kilogram | 1.35 ± 0.41 | 1.34 ± 0.59 | 1.35 ± 0.4 | 1.35 ± 0.29 |
Maternal perfusion of the placental intervillous space was evaluated using DCE-MRI. Secondary to incomplete imaging acquisition, data were available for analysis in 22 animals: control (n = 3), vitamin C only (n = 8), nicotine only (n = 3), and nicotine with vitamin C (n = 8). The average placental blood flow (milliliters per minute) and normalized placental blood flow (milliliters per minute per milliliter), which adjusts for the placental volume, were not significantly different between any of the groups, and all fell within the range of 59.7–109.2 mL/min and 0.75–1.94 mL/min per milliliter, respectively ( P = .927; Table 4 ).
| Parameter | Control (n = 3) | Vit C (n = 8) | Nicotine (n = 3) | Nic plus Vit C (n = 8) |
|---|---|---|---|---|
| Placental volume, mL | 61.7 ± 6.2 | 62.1 ± 5.7 | 74.5 ± 5.6 | 67.5 ± 27.4 |
| Placental blood flow, mL/min | 96.0 ± 26.8 | 59.7 ± 25.7 | 109.2 ± 24.5 | 86.1 ± 29.1 |
| Normalized placental blood flow, mL/min per milliliter | 1.6 ± 0.4 | 1.0 ± 0.4 | 1.50 ± 0.5 | 1.4 ± 0.4 |
A subset of 24 animals was available for a placental histological analysis: controls (n = 5), nicotine only (n = 4), vitamin C only (n = 7), and nicotine plus vitamin C (n = 8). Placentas exposed to nicotine showed conspicuous villous cytotrophoblast cell islands and syncytiotrophoblast sprouting, which was less common in controls and largely absent in animals treated with vitamin C ( Figure 2 ). Specifically, villous cytotrophoblast cell islands were seen in all nicotine-exposed animals compared with only 40% in the control group and 13% in the nicotine plus vitamin C group ( P < .05; Table 5 ).
