- Email: [email protected]
Reduced endothelial nitric oxide synthase activity and concentration in fetal umbilical veins from maternal cigarette smokers
American Journal of Obstetrics and Gynecology (2004) 191, 346e51
www.elsevier.com/locate/ajog
Reduced endothelial nitric oxide synthase activity and concentration in fetal umbilical veins from maternal cigarette smokers Malene R. Andersen, MSc,a,* Line R. Walker, MD,b Steen Stender, MD, PhDa Department of Clinical Biochemistry,a and the Department of Obstetrics and Gynecology,b Gentofte Hospital, University of Copenhagen, Gentofte, Denmark Received for publication June 20, 2003; revised December 2, 2003; accepted December 30, 2003
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS Birth weight Endothelium Nitric oxide Smoking Umbilical vein
Objective: This study aimed to investigate the effect of maternal cigarette smoking on endothelial nitric oxide synthase (eNOS) activity and concentration in the fetal umbilical vein, and subsequently, to relate the findings to the size of the newborn. Study design: Forty-four nonsmoking and 30 smoking women were included in the study. Umbilical vein endothelial cells were isolated immediately after delivery. The eNOS activity was determined in the samples by the conversion of 14C-L-arginine to 14C-L-citrulline, and the eNOS concentration was determined by a human eNOS immunoassay. Results: Newborns of smokers had a lower weight (P = .014) and a smaller head circumference (P = .002) than those newborns of nonsmokers. The eNOS activity in fetal umbilical veins exposed to maternal smoking was 40% lower (P = .006), and the eNOS concentration 32% lower (P = .053) in newborns of smokers than in nonsmokers. Conclusion: The findings suggest that maternal smoking reduces nitric oxide production in the fetal circulation. This may contribute to retarded fetal growth caused by the subsequent endothelial dysfunction with reduction of dilatory capacity of the vessels. Ó 2004 Elsevier Inc. All rights reserved.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Numerous investigations demonstrate reduced fetal growth1 and increased risk of perinatal mortality2 and morbidity3 in maternal cigarette smokers. In addition, smoking4 affects the maternal and fetal circulation.
Supported by grants from the Danish Heart Foundation (98-1-321A-22567 and 98-1-3-21B-22568) to Malene R. Andersen. * Reprint requests: Malene R. Andersen, Perinatal Epidemiological Research Unit, Department of Obstetrics and Gynecology, Aarhus University Hospital, Skejby Hospital, Brendstrupgaardsvej, DK-8200 Aarhus, Denmark. E-mail: [email protected] 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2003.12.040
In vivo, smoking5 and nicotine infusion6 impair the endothelium-dependent relaxation mediated by nitric oxide (NO) in human arteries and veins. NO is synthesized from the amino acid L-arginine by the enzyme NO synthase (NOS).7 In this process, L-arginine is converted to L-citrulline.7 Two isoforms of NOS are expressed in the endothelial cells (ECs): (1) the calcium-dependent constitutive endothelial NOS (eNOS), and (2) the calcium-independent inducible NOS (iNOS).7 NO contributes to the regulation of the vascular resistance in the fetal-placental and utero-placental circulation.8-10 Furthermore, studies suggest that NO
347
Andersen, Walker, and Stender deficiency is related to intrauterine (fetal) growth restriction (IUGR).11-14 For instance, women with IUGR pregnancies have lower umbilical artery eNOS expression,12 lower activity of eNOS in placental villous tissue,14 and higher plasma concentrations of asymmetric dimethylarginine (ADMA), an inhibitor of eNOS that competes with L-arginine,11 than women with normal pregnancies. With respect to smoking during pregnancy, the amounts of L-arginine and L-citrulline are lower in the umbilical vein and arterial walls of smokers than of nonsmokers.15,16 Thus, we hypothesize that fetal growth exposed to maternal smoking is inhibited by loss of dilatory capacity in the fetal circulation, owing to lower endothelial NO production. We have recently established a sufficiently sensitive method for quantification of the eNOS activity by measuring the conversion of 14C-L-arginine to 14C-L-citrulline in ECs from small rabbit aorta segments.17 The present study employed this method to investigate the association between maternal smoking and endothelial NO synthesis in fetal umbilical veins, and subsequently, to relate the findings to the size of the newborn.
Material and methods Study participants Pregnant women (21 to 42 years old) living in a small geographical area just north of Copenhagen and attending routine antenatal care at the Department of Obstetrics and Gynecology were invited to participate in the study. For inclusion, white women with singleton births had to fit 1 of 2 categories: (1) lifelong nonsmokers and (2) smokers who had smoked at least 5 pack-years and at least 5 cigarettes per day during pregnancy. One pack-year was defined as 20 cigarettes per day for 1 year, or the equivalent. In addition, the women included in the study were normotensive, normocholesterolemic, did not have preeclampsia, diabetes mellitus, a family history of premature vascular disease, nor were they on regular medication. The nature and purpose of the study was explained verbally and in writing to each woman. Information about smoking status was obtained by interview, and written consent was obtained from those who agreed to participate. The investigation conforms with the principles outlined in the Declaration of Helsinki, and was approved by the local ethics committee of Copenhagen (KA 99086). The women were recruited on 59 consultation days evenly distributed over a period of 7 months from September, 1999 to April, 2000. Of 440 pregnant women, 295 were asked to participate; 36% fulfilled the inclusion criteria and agreed to participate (66 nonsmokers and 39 smokers), 47% did not fulfill the inclusion criteria (many had stopped smoking in connection with the pregnancy), 13% did not want to participate, and 4% of elderly nonsmoking women were excluded at the interview to keep
the 2 groups at a comparable age distribution. One baby died during the study, 3 women developed preeclampsia, and 10 women were given general anesthesia because of complications during delivery. These women were excluded, leaving 59 nonsmokers and 32 smokers. However, both because of communication problems between the 2 departments and holidays, some samples were lost, leaving 44 nonsmokers and 30 smokers in the main study. Deliveries by smokers and nonsmokers were evenly distributed over the 7 months from October to May. After delivery, sociodemographic and clinical characteristics were obtained from the case notes. The Registrar General’s social class was used to register the parental occupation, using the father’s occupation in the event of a joint registration; otherwise, the mother’s occupation was recorded (I, professional; II, intermediate; III, skilled manual and nonmanual; IV, partly skilled manual; V, unskilled manual).
EC preparation Placenta including the umbilical cord was delivered 2 to 29 minutes after the baby, and was immediately transported from the Department of Obstetrics and Gynecology to the Department of Clinical Biochemistry where EC preparation was carried out using, with a few modifications, a method of Andersen and Stender.17 The umbilical vein segment (range, 3 to 5 cm) adjoining the placenta was removed, rinsed with 0.9% saline (7(C), and opened longitudinally. The segment was pinned to laboratory film on a cork board, and the ECs were obtained by a single scrape of the luminal surface with a razor blade. The ECs were placed in eppendorf tubes containing 140 mL of 50 mmol/L Tris buffer (pH 7.4), immediately frozen in liquid nitrogen, and stored at 80(C for less than 9 months until assayed. The period from delivery to freezing the sample was 20 to 60 minutes. Deliveries, with subsequent cell harvesting, were evenly distributed over the 24 hours. Each sample was given a number, and was collectively analyzed at the end of the study without knowledge of the mother’s smoking status. Controls of ECs from pig aortas were prepared as previously described.17
Cell counting The number of ECs in the sample was determined by duplicate cell counting.17 The cell numbers obtained per umbilical vein segment ranged from approximately 6 ! 104 to 2 ! 10,6 and were similar for nonsmokers and smokers.
Citrulline assay The eNOS activity was measured by duplicate determination of the conversion of 14C-L-arginine to 14C-L-citrulline.17 In short, ECs were homogenized by 5 cycles
348 Table I
Andersen, Walker, and Stender Maternal smoking status
Cigarettes/d before pregnancy Cigarettes/d during pregnancy Cigarettes/d during 1st trimester Cigarettes/d during 2nd trimester Cigarettes/d during 3rd trimester Pack-years Period from last cigarette to delivery (h)
Nonsmokers (n = 44)
Smokers (n = 30)
0 0 0 0 0 0
20 G 7 15 G 7 16 G 7 15 G 7 16 G 8 16 G 9 6G4
Plus-minus values are mean G SD.
of freeze-thawing, and 25 mL of homogenate was incubated for 30 minutes at 37(C in a Tris-reaction buffer containing 14C-L-arginine, calcium, calmodulin, tetrahydrobiopterin, flavin adenine dinucleotide (FAD), and b-nicotinamide adenine dinucleotide phosphate, reduced form (NADPH) (final volume 150 mL). 14C-L-citrulline was isolated by column chromatography and quantified by liquid scintillation counting. The eNOS activity was calculated from the mean formation of 14C-L-citrulline/ min and the mean number of ECs in the sample, and was given in pmol L-citrulline/min/106 ECs. For controls (ECs from pig aortas), the coefficients of variation (CV) within an assay (n = 8) and between assays (n = 6) were, when determined within 1 week, 8%. For the umbilical vein ECs, the CV within an assay (n = 8) was 10%.
eNOS immunoassay For the quantitative determination of eNOS concentration in the samples a human eNOS Quantikine Immunoassay kit (R&D Systems, Abingdon, UK) was used. The assay procedure of the kit was followed, but only 15 mL of cell homogenate in Tris buffer (final volume of 100 mL) was added per well. All samples were assayed in duplicate. The mean eNOS concentration was calculated and was given in ng eNOS/106 ECs. For controls of recombinant human eNOS standards, the CV within an assay (n = 8) was 4% and between assays (n = 6) was, when determined within 1 week, 5%. For the umbilical vein ECs, the CV within an assay (n = 8) was 7%.
Statistical analysis All results are given as mean G SD. Data of smokers and nonsmokers were analyzed by the Mann-Whitney U test or c2 test. To examine the association between 2 variables after adjusting for the effect of other variables, multiple linear regression analysis was performed and the bivariate r and partial r (rpartial) estimated. For this analysis, the eNOS activity and concentration
were transformed by logarithm. Statistical significance was set at P!.05.
Results Maternal smoking status The average cigarette consumption for smokers was 16 pack-years (range, 5 to 49) (Table I). The consumption was lowered from 20 cigarettes/day (range, 10 to 35) before pregnancy to 15 cigarettes/day (range, 7 to 35) throughout the pregnancy (Table I). The last exposure to cigarette smoke was about 6 hours (range, 1 to 21) before delivery (Table I).
Sociodemographic and clinical characteristics Table II shows that maternal smokers and nonsmokers had similar age, body mass index (BMI) before pregnancy, parity, and blood pressure when measured a few days before delivery. However, the nonsmokers had a significantly higher occupational status than the smokers (Table II). All infants were delivered after 37 completed weeks of gestation, which was calculated from the beginning of the last menstrual period. The infants were healthy and without congenital deformities, except for 1 child of a smoking mother who had cleft lip and palate, syndactyly, and clubfoot. No differences were found between the 2 groups in gestational age and placental weight (Table II). In addition, the methods employed in the delivery of the infant, whether per vagina, cesarean section, extraction by vacuum, administration of an oxytocic drug, and/or anesthesia, were similar between smokers and nonsmokers (data not shown). Infants born to smokers had a significantly lower birth weight and smaller head circumferences than those born to nonsmokers, but the length and abdominal circumference were similar (Table II). The Apgar score at 1 and 5 minutes, and the infant male/female ratio, were not different between the 2 groups (Table II).>
eNOS activity and concentration The eNOS activity in fetal umbilical veins exposed to maternal smoking was 40% lower (27.1 G 11.8 vs 45.0 G 34.3 pmol L-citrulline/min/106 ECs), and the eNOS concentration 32% (1.7 G 1.0 vs 2.5 G 1.9 ng eNOS/106 ECs) lower than in nonsmokers (Figure). Multiple regression analysis showed that the number of cigarettes consumed during pregnancy was associated with the eNOS activity (r = 0.33, P = .004), and slightly with the eNOS concentration (r = 0.23, P = .051). The last exposure to cigarette smoke before delivery was, however, not associated with eNOS activity.
349
Andersen, Walker, and Stender Table II
Sociodemographic and clinical characteristics of the participants Nonsmokers (n = 44)
Maternal characteristics Age (y) BMI before pregnancy (kg/m2) Occupational status Parity Systolic BP (mm Hg) Diastolic BP (mm Hg) Infant characteristics Gestational age (d) Placenta weight (g) Birth weight (g) Length (cm) Head circumference (cm) Abdominal circumference (cm) 1-min Apgar score 5-min Apgar score Infant male/female ratio
Smokers (n = 30)
P value
33 G 4 23 G 4 2.1 G 0.9 1.7 G 0.8 126 G 10 76 G 9
31 G 5 23 G 5 2.8 G 0.8 1.6 G 0.7 126 G 11 73 G 7
NS NS 0.001 NS NS NS
282 G 8 674 G 147 3649 G 386 52.6 G 1.7 35.1 G 1.3 33.3 G 2.1y 9.6 G 0.6 10.0 G 0.2 24/20
280 G 8 632 G 107 3398 G 350 51.9 G 1.9 34.1 G 1.4 32.8 G 2.1z 9.6 G 0.9 9.9 G 0.5 14/16
NS NS 0.014 NS 0.002 NS NS NS NS
Plus-minus values are mean G SD. NS, Not significant; BMI, body mass index; BP, blood pressure. Data were compared by Mann-Whitney U test and the infant male/female ratios by c2 test. y n = 41. z n = 28.
The eNOS activity and concentration were also associated (r = 0.61, P!.001). This association persisted after adjusting for maternal smoking (rpartial = 0.58, P!.001). The 40% reduction in eNOS activity associated with smoking was partially explained by the reduced eNOS concentration. The reduction became 25% after adjusting for the reduction in eNOS concentration. The eNOS activity was associated with the newborn weight (r = 0.28, P = .014), but not with infant head circumference. The association between newborn weight and eNOS activity persisted after adjusting for smoking, gestational age, and infant sex (rpartial = 0.25, P = .032). The difference in birth weight between the 2 groups was 251 g (Table II). This difference was reduced to 215 g after adjusting for gestational age and infant sex, and to 149 g after adjusting for the eNOS activity, as well. The above-mentioned associations were not affected by adjusting for the sociodemographic and clinical characteristics of the mothers.
Comment The present study confirms previous findings that infants of smokers have reduced fetal growth at birth, as measured by weight and head circumference compared with infants of nonsmokers.1-3 More importantly, the study demonstrates that the eNOS activity in the fetal umbilical vein endothelium exposed to maternal smoking is significantly lower than in nonsmokers, and that the reduction is partially explained by a lower eNOS concentration.
As expected, the maternal smokers had a lower occupational status than the nonsmokers. However, adjusting for sociodemographic and clinical characteristics of the mothers did not affect the differences in eNOS activity, eNOS concentration, and birth weight18 between smokers and nonsmokers in the present study. Epidemiologic studies suggest that smoking cessation during pregnancy improves the birth weight.18 It seems reasonable to assume, therefore, that the present findings are more directly related to maternal smoking than to sociodemographic and clinical characteristics.18 The pathogenesis of restricted fetal growth is poorly understood. IUGR of the fetus has, however, been connected to a lower endothelial NO production. For instance, the umbilical artery eNOS expression was reduced,12 the eNOS activity in placental villus tissue was lowered,14 and the resistance in uterine arteries was attenuated by L-arginine infusion13 when compared with normal pregnancies. Surprisingly, the resistance in placental-sided, uterine, and umbilical arteries was not affected by L-arginine infusion in the latter study.13 Furthermore, the NO concentration in umbilical veins determined by measuring the plasma nitrite/nitrate levels was increased in IUGR pregnancies compared with normal pregnancies.19 The measurement of nitrite/nitrate is, however, controversial, and there are many confounding factors including diet and renal clearance. Doppler ultrasonography studies suggest that cigarette smoking during pregnancy increases the resistance in the maternal and fetal circulation.4 The main findings of the present study are that the eNOS activity was 40%
350
Andersen, Walker, and Stender
Figure The (A) eNOS activity and the (B) eNOS concentration in freshly isolated umbilical vein ECs of maternal nonsmokers (n = 44) and smokers (n = 30). The eNOS activity was significantly lower in the fetal umbilical vein of smokers than in nonsmokers. The lower eNOS concentration in the fetal umbilical vein of smokers was only borderline significant when compared with that of nonsmokers. Horizontal lines represent the mean values for each group. P values were determined by Mann-Whitney U test.
lower, and the eNOS concentration 32% lower, in the fetal umbilical vein endothelium exposed to maternal smoking, and that the activity of eNOS was associated with the eNOS concentration and the newborn weight. Multiple regression analysis suggested that 40% of the reduction in eNOS activity associated with smoking was explained by the reduced eNOS concentration, and 26% of the reduction in birth weight by the reduced eNOS activity. Considered together, the findings suggest that smoking during pregnancy reduces the endothelial NO synthesis in part by a lower eNOS concentration in the fetal vessels, and that reduced NO level followed by increased vasoconstriction inhibits the fetal growth. Reduced fetal growth by maternal cigarette smoking may be explained by the following sequence of events: smoking reduces the levels of estrogen and L-arginine in the maternal and fetal circulation (by an effect of nicotine on liver and placental metabolism). A reduced level of estrogen and L-arginine followed by increased level of ADMA decreases the eNOS activity in part by a reduced eNOS concentration, leading to a lower NO production. A lowered NO production attenuates the vasodilatory capacity, whereby fetal growth is retarded. Alternatively or additionally, nicotine from the smoking mother enters the fetal circulation and acts directly on the endothelium. The estrogen hypothesis originates from the observation that the estrogen level in female smokers is decreased,20,21 and that estrogen stimulates the endothelial NO production in experimental animals17,22 and humans.23 The estrogen hypothesis is reinforced by the find-
ings that in the 37th week of pregnancy the estriol level is reduced by 20% to 30% in smokers,20,21 and that the estriol level in mothers of infants in the highest birth weight category (O4500 g) was twice as high as that of mothers of infants in the lowest category (!2500 g).20 Estrogen may interfere directly or indirectly with the eNOS inhibitor ADMA. For instance, the ADMA levels were significantly lower in the first and third trimester of pregnancy than in nonpregnant women.24 This is consistent with other findings that estrogen replacement therapy significantly reduced the plasma concentration of ADMA in healthy postmenopausal women.25 The findings suggest that a reduction in circulating ADMA might facilitate NO-mediated vasodilation during normal pregnancy. This hypothesis is supported by the observation that the plasma ADMA concentration was raised in women who developed IUGR compared with that in women with a normal pregnancy outcome.11 However, whether the ADMA level is affected by smoking, as is the NO precursor, L-arginine,15,16 warrants further investigation. The detrimental effect of smoking on the fetal circulation may disappear in a few days as soon as the concentration of nicotine, estrogen, L-arginine, and ADMA in the blood of the infant has been normalized. It is, however, also possible that the initial endothelial dysfunction in the vascular system of the fetus is only partly reversible, rendering the arteries more sensitive to cholesterol accumulation many decades later. Reduced endothelial NO production in rabbits predicts cholesterol accumulation during subsequent
Andersen, Walker, and Stender hypercholesterolemia.17 If exposure to maternal smoking during fetal development is a risk factor for ischemic heart disease in humans much later in life awaits further studies. In conclusion, the present study demonstrates that infants of smokers have reduced weight and head circumference at birth compared with infants of nonsmokers, and that the eNOS activity and concentration are reduced in the fetal umbilical vein endothelium exposed to maternal cigarette smoking. The findings suggest that reduced fetal growth in maternal smokers is at least partially caused by loss of vasodilatory capacity from a lower endothelial production of NO in the fetal circulation.
Acknowledgments We thank Professor Michael Vaeth (Department of Biostatistics, Aarhus University) for statistical supervising, the midwives (Department of Obstetric and Gynecology) for their assistance in connection with the project, and the veterinarians at Danish Crown for providing pig aortas.
References 1. Olsen J. Cigarette smoking in pregnancy and fetal growth. Does the type of tobacco play a role? Int J Epidemiol 1992;21:279-84. 2. Wilcox AJ. Birth weight and perinatal mortality: the effect of maternal smoking. Am J Epidemiol 1993;137:1098-104. 3. Rantakallio P. The effect of maternal smoking on birth weight and the subsequent health of the child. Early Hum Dev 1978;2:371-82. 4. Morrow RJ, Ritchie JWK, Bull SB. Maternal cigarette smoking: the effects on umbilical and uterine blood flow velocity. Am J Obstet Gynecol 1988;159:1069-71. 5. Ueda S, Matsuoka H, Miyazaki H, Usui M, Okuda S, Imaizumi T. Tetrahydrobiopterin restores endothelial function in long-term smokers. J Am Coll Cardiol 2000;35:71-5. 6. Chalon S, Moreno H, Benowitz NL, Hoffman BB, Blaschke TF. Nicotine impairs endothelium-dependent dilatation in human veins in vivo. Clin Pharmacol Ther 2000;67:391-7. 7. Moncada S, Higgs A. The L-arginine-nitric oxide pathway [review]. N Engl J Med 1993;329:2002-12. 8. Myatt L, Brewer AS, Langdon G, Brockman DE. Attenuation of the vasoconstrictor effect of thromboxane and endothelin by nitric oxide in the human fetal-placental circulation. Am J Obstet Gynecol 1992;166:224-30. 9. Myatt L, Brewer A, Brockman DE. The action of nitric oxide in the perfused human fetal-placental circulation. Am J Obstet Gynecol 1991;164:687-92.
351 10. Poston L, McCarthy AL, Ritter JM. Control of vascular resistance in the maternal and feto-placental arterial beds [review]. Pharmacol Ther 1995;65:215-39. 11. Savvidou MD, Hingorani AD, Tsikas D, Fro¨lich JC, Vallance P, Nicolaides KH. Endothelial dysfunction and raised plasma concentrations of asymmetric dimethylarginine in pregnant women who subsequently develop pre-eclampsia. Lancet 2003;361: 1511-7. 12. Rutherford RAD, McCarthy A, Sullivan MHF, Elder MG, Polak JM, Wharton J. Nitric oxide synthase in human placenta and umbilical cord from normal, intrauterine growth-retarded and preeclamptic pregnancies. Br J Pharmacol 1995;116:3099-109. 13. Neri I, Mazza V, Galassi MC, Volpe A, Facchinetti F. Effects of L-arginine on utero-placental circulation in growth-retarded fetuses. Acta Obstet Gynecol Scand 1996;75:208-12. 14. Morris NH, Sooranna SR, Learmont JG, Poston L, Ramsey B, Pearson JD, et al. Nitric oxide synthase activities in placental tissue from normotensive, pre-eclamptic and growth retarded pregnancies. Br J Obstet Gynaecol 1995;102:711-4. 15. Obwegeser R, Oguogho A, Ulm M, Berghammer P, Sinzinger H. Maternal cigarette smoking increases F2-isoprostanes and reduces prostacyclin and nitric oxide in umbilical vessels. Prostaglandins Other Lipid Mediat 1999;57:269-79. 16. Ulm MR, Plo¨ckinger B, Pirich C, Gryglewski RJ, Sinzinger HF. Umbilical arteries of babies born to cigarette smokers generate less prostacyclin and contain less arginine and citrulline compared with those of babies born to control subjects. Am J Obstet Gynecol 1995;172:1485-7. 17. Andersen MR, Stender S. Endothelial nitric oxide synthase activity in aorta of normocholesterolemic rabbits: regional variation and the effect of estrogen. Cardiovasc Res 2000;47:192-9. 18. Butler NR, Goldstein H, Ross EM. Cigarette smoking in pregnancy: its influence on birth weight and perinatal mortality. BMJ 1972;2:127-30. 19. Lyall F, Greer IA, Young A, Myatt L. Nitric oxide concentrations are increased in the feto-placental circulation in intrauterine growth restriction. Placenta 1996;17:165-8. 20. Kaijser M, Granath F, Jacobsen G, Cnattingius S, Ekbom A. Maternal pregnancy estriol levels in relation to anamnestic and fetal anthropometric data. Epidemiology 2000;11:315-9. 21. Mochizuki M, Maruo T, Masuko K, Ohtsu T. Effects of smoking on fetoplacental-maternal system during pregnancy. Am J Obstet Gynecol 1984;149:413-20. 22. Cockell AP, Poston L. 17b-estradiol stimulates flow-induced vasodilatation in isolated small mesenteric arteries from prepubertal female rats. Am J Obstet Gynecol 1997;177: 1432-8. 23. Knock GA, Poston L. Bradykinin-mediated relaxation of isolated maternal resistance arteries in normal pregnancy and preeclampsia. Am J Obstet Gynecol 1996;175:1668-74. 24. Maeda T, Yoshimura T, Okamura H. Asymmetric dimethylarginine, an endogenous inhibitor of nitric oxide synthase, in maternal and fetal circulation. J Soc Gynecol Investig 2003;10:2-4. 25. Teerlink T, Neele SJM, De Jong S, Netelenbos JC, Stehouwer CDA. Oestrogen replacement therapy lowers plasma levels of asymmetrical dimethylarginine in healthy postmenopausal women. Clinical Science 2003;105:67-71.
www.elsevier.com/locate/ajog
Reduced endothelial nitric oxide synthase activity and concentration in fetal umbilical veins from maternal cigarette smokers Malene R. Andersen, MSc,a,* Line R. Walker, MD,b Steen Stender, MD, PhDa Department of Clinical Biochemistry,a and the Department of Obstetrics and Gynecology,b Gentofte Hospital, University of Copenhagen, Gentofte, Denmark Received for publication June 20, 2003; revised December 2, 2003; accepted December 30, 2003
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS Birth weight Endothelium Nitric oxide Smoking Umbilical vein
Objective: This study aimed to investigate the effect of maternal cigarette smoking on endothelial nitric oxide synthase (eNOS) activity and concentration in the fetal umbilical vein, and subsequently, to relate the findings to the size of the newborn. Study design: Forty-four nonsmoking and 30 smoking women were included in the study. Umbilical vein endothelial cells were isolated immediately after delivery. The eNOS activity was determined in the samples by the conversion of 14C-L-arginine to 14C-L-citrulline, and the eNOS concentration was determined by a human eNOS immunoassay. Results: Newborns of smokers had a lower weight (P = .014) and a smaller head circumference (P = .002) than those newborns of nonsmokers. The eNOS activity in fetal umbilical veins exposed to maternal smoking was 40% lower (P = .006), and the eNOS concentration 32% lower (P = .053) in newborns of smokers than in nonsmokers. Conclusion: The findings suggest that maternal smoking reduces nitric oxide production in the fetal circulation. This may contribute to retarded fetal growth caused by the subsequent endothelial dysfunction with reduction of dilatory capacity of the vessels. Ó 2004 Elsevier Inc. All rights reserved.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Numerous investigations demonstrate reduced fetal growth1 and increased risk of perinatal mortality2 and morbidity3 in maternal cigarette smokers. In addition, smoking4 affects the maternal and fetal circulation.
Supported by grants from the Danish Heart Foundation (98-1-321A-22567 and 98-1-3-21B-22568) to Malene R. Andersen. * Reprint requests: Malene R. Andersen, Perinatal Epidemiological Research Unit, Department of Obstetrics and Gynecology, Aarhus University Hospital, Skejby Hospital, Brendstrupgaardsvej, DK-8200 Aarhus, Denmark. E-mail: [email protected] 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2003.12.040
In vivo, smoking5 and nicotine infusion6 impair the endothelium-dependent relaxation mediated by nitric oxide (NO) in human arteries and veins. NO is synthesized from the amino acid L-arginine by the enzyme NO synthase (NOS).7 In this process, L-arginine is converted to L-citrulline.7 Two isoforms of NOS are expressed in the endothelial cells (ECs): (1) the calcium-dependent constitutive endothelial NOS (eNOS), and (2) the calcium-independent inducible NOS (iNOS).7 NO contributes to the regulation of the vascular resistance in the fetal-placental and utero-placental circulation.8-10 Furthermore, studies suggest that NO
347
Andersen, Walker, and Stender deficiency is related to intrauterine (fetal) growth restriction (IUGR).11-14 For instance, women with IUGR pregnancies have lower umbilical artery eNOS expression,12 lower activity of eNOS in placental villous tissue,14 and higher plasma concentrations of asymmetric dimethylarginine (ADMA), an inhibitor of eNOS that competes with L-arginine,11 than women with normal pregnancies. With respect to smoking during pregnancy, the amounts of L-arginine and L-citrulline are lower in the umbilical vein and arterial walls of smokers than of nonsmokers.15,16 Thus, we hypothesize that fetal growth exposed to maternal smoking is inhibited by loss of dilatory capacity in the fetal circulation, owing to lower endothelial NO production. We have recently established a sufficiently sensitive method for quantification of the eNOS activity by measuring the conversion of 14C-L-arginine to 14C-L-citrulline in ECs from small rabbit aorta segments.17 The present study employed this method to investigate the association between maternal smoking and endothelial NO synthesis in fetal umbilical veins, and subsequently, to relate the findings to the size of the newborn.
Material and methods Study participants Pregnant women (21 to 42 years old) living in a small geographical area just north of Copenhagen and attending routine antenatal care at the Department of Obstetrics and Gynecology were invited to participate in the study. For inclusion, white women with singleton births had to fit 1 of 2 categories: (1) lifelong nonsmokers and (2) smokers who had smoked at least 5 pack-years and at least 5 cigarettes per day during pregnancy. One pack-year was defined as 20 cigarettes per day for 1 year, or the equivalent. In addition, the women included in the study were normotensive, normocholesterolemic, did not have preeclampsia, diabetes mellitus, a family history of premature vascular disease, nor were they on regular medication. The nature and purpose of the study was explained verbally and in writing to each woman. Information about smoking status was obtained by interview, and written consent was obtained from those who agreed to participate. The investigation conforms with the principles outlined in the Declaration of Helsinki, and was approved by the local ethics committee of Copenhagen (KA 99086). The women were recruited on 59 consultation days evenly distributed over a period of 7 months from September, 1999 to April, 2000. Of 440 pregnant women, 295 were asked to participate; 36% fulfilled the inclusion criteria and agreed to participate (66 nonsmokers and 39 smokers), 47% did not fulfill the inclusion criteria (many had stopped smoking in connection with the pregnancy), 13% did not want to participate, and 4% of elderly nonsmoking women were excluded at the interview to keep
the 2 groups at a comparable age distribution. One baby died during the study, 3 women developed preeclampsia, and 10 women were given general anesthesia because of complications during delivery. These women were excluded, leaving 59 nonsmokers and 32 smokers. However, both because of communication problems between the 2 departments and holidays, some samples were lost, leaving 44 nonsmokers and 30 smokers in the main study. Deliveries by smokers and nonsmokers were evenly distributed over the 7 months from October to May. After delivery, sociodemographic and clinical characteristics were obtained from the case notes. The Registrar General’s social class was used to register the parental occupation, using the father’s occupation in the event of a joint registration; otherwise, the mother’s occupation was recorded (I, professional; II, intermediate; III, skilled manual and nonmanual; IV, partly skilled manual; V, unskilled manual).
EC preparation Placenta including the umbilical cord was delivered 2 to 29 minutes after the baby, and was immediately transported from the Department of Obstetrics and Gynecology to the Department of Clinical Biochemistry where EC preparation was carried out using, with a few modifications, a method of Andersen and Stender.17 The umbilical vein segment (range, 3 to 5 cm) adjoining the placenta was removed, rinsed with 0.9% saline (7(C), and opened longitudinally. The segment was pinned to laboratory film on a cork board, and the ECs were obtained by a single scrape of the luminal surface with a razor blade. The ECs were placed in eppendorf tubes containing 140 mL of 50 mmol/L Tris buffer (pH 7.4), immediately frozen in liquid nitrogen, and stored at 80(C for less than 9 months until assayed. The period from delivery to freezing the sample was 20 to 60 minutes. Deliveries, with subsequent cell harvesting, were evenly distributed over the 24 hours. Each sample was given a number, and was collectively analyzed at the end of the study without knowledge of the mother’s smoking status. Controls of ECs from pig aortas were prepared as previously described.17
Cell counting The number of ECs in the sample was determined by duplicate cell counting.17 The cell numbers obtained per umbilical vein segment ranged from approximately 6 ! 104 to 2 ! 10,6 and were similar for nonsmokers and smokers.
Citrulline assay The eNOS activity was measured by duplicate determination of the conversion of 14C-L-arginine to 14C-L-citrulline.17 In short, ECs were homogenized by 5 cycles
348 Table I
Andersen, Walker, and Stender Maternal smoking status
Cigarettes/d before pregnancy Cigarettes/d during pregnancy Cigarettes/d during 1st trimester Cigarettes/d during 2nd trimester Cigarettes/d during 3rd trimester Pack-years Period from last cigarette to delivery (h)
Nonsmokers (n = 44)
Smokers (n = 30)
0 0 0 0 0 0
20 G 7 15 G 7 16 G 7 15 G 7 16 G 8 16 G 9 6G4
Plus-minus values are mean G SD.
of freeze-thawing, and 25 mL of homogenate was incubated for 30 minutes at 37(C in a Tris-reaction buffer containing 14C-L-arginine, calcium, calmodulin, tetrahydrobiopterin, flavin adenine dinucleotide (FAD), and b-nicotinamide adenine dinucleotide phosphate, reduced form (NADPH) (final volume 150 mL). 14C-L-citrulline was isolated by column chromatography and quantified by liquid scintillation counting. The eNOS activity was calculated from the mean formation of 14C-L-citrulline/ min and the mean number of ECs in the sample, and was given in pmol L-citrulline/min/106 ECs. For controls (ECs from pig aortas), the coefficients of variation (CV) within an assay (n = 8) and between assays (n = 6) were, when determined within 1 week, 8%. For the umbilical vein ECs, the CV within an assay (n = 8) was 10%.
eNOS immunoassay For the quantitative determination of eNOS concentration in the samples a human eNOS Quantikine Immunoassay kit (R&D Systems, Abingdon, UK) was used. The assay procedure of the kit was followed, but only 15 mL of cell homogenate in Tris buffer (final volume of 100 mL) was added per well. All samples were assayed in duplicate. The mean eNOS concentration was calculated and was given in ng eNOS/106 ECs. For controls of recombinant human eNOS standards, the CV within an assay (n = 8) was 4% and between assays (n = 6) was, when determined within 1 week, 5%. For the umbilical vein ECs, the CV within an assay (n = 8) was 7%.
Statistical analysis All results are given as mean G SD. Data of smokers and nonsmokers were analyzed by the Mann-Whitney U test or c2 test. To examine the association between 2 variables after adjusting for the effect of other variables, multiple linear regression analysis was performed and the bivariate r and partial r (rpartial) estimated. For this analysis, the eNOS activity and concentration
were transformed by logarithm. Statistical significance was set at P!.05.
Results Maternal smoking status The average cigarette consumption for smokers was 16 pack-years (range, 5 to 49) (Table I). The consumption was lowered from 20 cigarettes/day (range, 10 to 35) before pregnancy to 15 cigarettes/day (range, 7 to 35) throughout the pregnancy (Table I). The last exposure to cigarette smoke was about 6 hours (range, 1 to 21) before delivery (Table I).
Sociodemographic and clinical characteristics Table II shows that maternal smokers and nonsmokers had similar age, body mass index (BMI) before pregnancy, parity, and blood pressure when measured a few days before delivery. However, the nonsmokers had a significantly higher occupational status than the smokers (Table II). All infants were delivered after 37 completed weeks of gestation, which was calculated from the beginning of the last menstrual period. The infants were healthy and without congenital deformities, except for 1 child of a smoking mother who had cleft lip and palate, syndactyly, and clubfoot. No differences were found between the 2 groups in gestational age and placental weight (Table II). In addition, the methods employed in the delivery of the infant, whether per vagina, cesarean section, extraction by vacuum, administration of an oxytocic drug, and/or anesthesia, were similar between smokers and nonsmokers (data not shown). Infants born to smokers had a significantly lower birth weight and smaller head circumferences than those born to nonsmokers, but the length and abdominal circumference were similar (Table II). The Apgar score at 1 and 5 minutes, and the infant male/female ratio, were not different between the 2 groups (Table II).>
eNOS activity and concentration The eNOS activity in fetal umbilical veins exposed to maternal smoking was 40% lower (27.1 G 11.8 vs 45.0 G 34.3 pmol L-citrulline/min/106 ECs), and the eNOS concentration 32% (1.7 G 1.0 vs 2.5 G 1.9 ng eNOS/106 ECs) lower than in nonsmokers (Figure). Multiple regression analysis showed that the number of cigarettes consumed during pregnancy was associated with the eNOS activity (r = 0.33, P = .004), and slightly with the eNOS concentration (r = 0.23, P = .051). The last exposure to cigarette smoke before delivery was, however, not associated with eNOS activity.
349
Andersen, Walker, and Stender Table II
Sociodemographic and clinical characteristics of the participants Nonsmokers (n = 44)
Maternal characteristics Age (y) BMI before pregnancy (kg/m2) Occupational status Parity Systolic BP (mm Hg) Diastolic BP (mm Hg) Infant characteristics Gestational age (d) Placenta weight (g) Birth weight (g) Length (cm) Head circumference (cm) Abdominal circumference (cm) 1-min Apgar score 5-min Apgar score Infant male/female ratio
Smokers (n = 30)
P value
33 G 4 23 G 4 2.1 G 0.9 1.7 G 0.8 126 G 10 76 G 9
31 G 5 23 G 5 2.8 G 0.8 1.6 G 0.7 126 G 11 73 G 7
NS NS 0.001 NS NS NS
282 G 8 674 G 147 3649 G 386 52.6 G 1.7 35.1 G 1.3 33.3 G 2.1y 9.6 G 0.6 10.0 G 0.2 24/20
280 G 8 632 G 107 3398 G 350 51.9 G 1.9 34.1 G 1.4 32.8 G 2.1z 9.6 G 0.9 9.9 G 0.5 14/16
NS NS 0.014 NS 0.002 NS NS NS NS
Plus-minus values are mean G SD. NS, Not significant; BMI, body mass index; BP, blood pressure. Data were compared by Mann-Whitney U test and the infant male/female ratios by c2 test. y n = 41. z n = 28.
The eNOS activity and concentration were also associated (r = 0.61, P!.001). This association persisted after adjusting for maternal smoking (rpartial = 0.58, P!.001). The 40% reduction in eNOS activity associated with smoking was partially explained by the reduced eNOS concentration. The reduction became 25% after adjusting for the reduction in eNOS concentration. The eNOS activity was associated with the newborn weight (r = 0.28, P = .014), but not with infant head circumference. The association between newborn weight and eNOS activity persisted after adjusting for smoking, gestational age, and infant sex (rpartial = 0.25, P = .032). The difference in birth weight between the 2 groups was 251 g (Table II). This difference was reduced to 215 g after adjusting for gestational age and infant sex, and to 149 g after adjusting for the eNOS activity, as well. The above-mentioned associations were not affected by adjusting for the sociodemographic and clinical characteristics of the mothers.
Comment The present study confirms previous findings that infants of smokers have reduced fetal growth at birth, as measured by weight and head circumference compared with infants of nonsmokers.1-3 More importantly, the study demonstrates that the eNOS activity in the fetal umbilical vein endothelium exposed to maternal smoking is significantly lower than in nonsmokers, and that the reduction is partially explained by a lower eNOS concentration.
As expected, the maternal smokers had a lower occupational status than the nonsmokers. However, adjusting for sociodemographic and clinical characteristics of the mothers did not affect the differences in eNOS activity, eNOS concentration, and birth weight18 between smokers and nonsmokers in the present study. Epidemiologic studies suggest that smoking cessation during pregnancy improves the birth weight.18 It seems reasonable to assume, therefore, that the present findings are more directly related to maternal smoking than to sociodemographic and clinical characteristics.18 The pathogenesis of restricted fetal growth is poorly understood. IUGR of the fetus has, however, been connected to a lower endothelial NO production. For instance, the umbilical artery eNOS expression was reduced,12 the eNOS activity in placental villus tissue was lowered,14 and the resistance in uterine arteries was attenuated by L-arginine infusion13 when compared with normal pregnancies. Surprisingly, the resistance in placental-sided, uterine, and umbilical arteries was not affected by L-arginine infusion in the latter study.13 Furthermore, the NO concentration in umbilical veins determined by measuring the plasma nitrite/nitrate levels was increased in IUGR pregnancies compared with normal pregnancies.19 The measurement of nitrite/nitrate is, however, controversial, and there are many confounding factors including diet and renal clearance. Doppler ultrasonography studies suggest that cigarette smoking during pregnancy increases the resistance in the maternal and fetal circulation.4 The main findings of the present study are that the eNOS activity was 40%
350
Andersen, Walker, and Stender
Figure The (A) eNOS activity and the (B) eNOS concentration in freshly isolated umbilical vein ECs of maternal nonsmokers (n = 44) and smokers (n = 30). The eNOS activity was significantly lower in the fetal umbilical vein of smokers than in nonsmokers. The lower eNOS concentration in the fetal umbilical vein of smokers was only borderline significant when compared with that of nonsmokers. Horizontal lines represent the mean values for each group. P values were determined by Mann-Whitney U test.
lower, and the eNOS concentration 32% lower, in the fetal umbilical vein endothelium exposed to maternal smoking, and that the activity of eNOS was associated with the eNOS concentration and the newborn weight. Multiple regression analysis suggested that 40% of the reduction in eNOS activity associated with smoking was explained by the reduced eNOS concentration, and 26% of the reduction in birth weight by the reduced eNOS activity. Considered together, the findings suggest that smoking during pregnancy reduces the endothelial NO synthesis in part by a lower eNOS concentration in the fetal vessels, and that reduced NO level followed by increased vasoconstriction inhibits the fetal growth. Reduced fetal growth by maternal cigarette smoking may be explained by the following sequence of events: smoking reduces the levels of estrogen and L-arginine in the maternal and fetal circulation (by an effect of nicotine on liver and placental metabolism). A reduced level of estrogen and L-arginine followed by increased level of ADMA decreases the eNOS activity in part by a reduced eNOS concentration, leading to a lower NO production. A lowered NO production attenuates the vasodilatory capacity, whereby fetal growth is retarded. Alternatively or additionally, nicotine from the smoking mother enters the fetal circulation and acts directly on the endothelium. The estrogen hypothesis originates from the observation that the estrogen level in female smokers is decreased,20,21 and that estrogen stimulates the endothelial NO production in experimental animals17,22 and humans.23 The estrogen hypothesis is reinforced by the find-
ings that in the 37th week of pregnancy the estriol level is reduced by 20% to 30% in smokers,20,21 and that the estriol level in mothers of infants in the highest birth weight category (O4500 g) was twice as high as that of mothers of infants in the lowest category (!2500 g).20 Estrogen may interfere directly or indirectly with the eNOS inhibitor ADMA. For instance, the ADMA levels were significantly lower in the first and third trimester of pregnancy than in nonpregnant women.24 This is consistent with other findings that estrogen replacement therapy significantly reduced the plasma concentration of ADMA in healthy postmenopausal women.25 The findings suggest that a reduction in circulating ADMA might facilitate NO-mediated vasodilation during normal pregnancy. This hypothesis is supported by the observation that the plasma ADMA concentration was raised in women who developed IUGR compared with that in women with a normal pregnancy outcome.11 However, whether the ADMA level is affected by smoking, as is the NO precursor, L-arginine,15,16 warrants further investigation. The detrimental effect of smoking on the fetal circulation may disappear in a few days as soon as the concentration of nicotine, estrogen, L-arginine, and ADMA in the blood of the infant has been normalized. It is, however, also possible that the initial endothelial dysfunction in the vascular system of the fetus is only partly reversible, rendering the arteries more sensitive to cholesterol accumulation many decades later. Reduced endothelial NO production in rabbits predicts cholesterol accumulation during subsequent
Andersen, Walker, and Stender hypercholesterolemia.17 If exposure to maternal smoking during fetal development is a risk factor for ischemic heart disease in humans much later in life awaits further studies. In conclusion, the present study demonstrates that infants of smokers have reduced weight and head circumference at birth compared with infants of nonsmokers, and that the eNOS activity and concentration are reduced in the fetal umbilical vein endothelium exposed to maternal cigarette smoking. The findings suggest that reduced fetal growth in maternal smokers is at least partially caused by loss of vasodilatory capacity from a lower endothelial production of NO in the fetal circulation.
Acknowledgments We thank Professor Michael Vaeth (Department of Biostatistics, Aarhus University) for statistical supervising, the midwives (Department of Obstetric and Gynecology) for their assistance in connection with the project, and the veterinarians at Danish Crown for providing pig aortas.
References 1. Olsen J. Cigarette smoking in pregnancy and fetal growth. Does the type of tobacco play a role? Int J Epidemiol 1992;21:279-84. 2. Wilcox AJ. Birth weight and perinatal mortality: the effect of maternal smoking. Am J Epidemiol 1993;137:1098-104. 3. Rantakallio P. The effect of maternal smoking on birth weight and the subsequent health of the child. Early Hum Dev 1978;2:371-82. 4. Morrow RJ, Ritchie JWK, Bull SB. Maternal cigarette smoking: the effects on umbilical and uterine blood flow velocity. Am J Obstet Gynecol 1988;159:1069-71. 5. Ueda S, Matsuoka H, Miyazaki H, Usui M, Okuda S, Imaizumi T. Tetrahydrobiopterin restores endothelial function in long-term smokers. J Am Coll Cardiol 2000;35:71-5. 6. Chalon S, Moreno H, Benowitz NL, Hoffman BB, Blaschke TF. Nicotine impairs endothelium-dependent dilatation in human veins in vivo. Clin Pharmacol Ther 2000;67:391-7. 7. Moncada S, Higgs A. The L-arginine-nitric oxide pathway [review]. N Engl J Med 1993;329:2002-12. 8. Myatt L, Brewer AS, Langdon G, Brockman DE. Attenuation of the vasoconstrictor effect of thromboxane and endothelin by nitric oxide in the human fetal-placental circulation. Am J Obstet Gynecol 1992;166:224-30. 9. Myatt L, Brewer A, Brockman DE. The action of nitric oxide in the perfused human fetal-placental circulation. Am J Obstet Gynecol 1991;164:687-92.
351 10. Poston L, McCarthy AL, Ritter JM. Control of vascular resistance in the maternal and feto-placental arterial beds [review]. Pharmacol Ther 1995;65:215-39. 11. Savvidou MD, Hingorani AD, Tsikas D, Fro¨lich JC, Vallance P, Nicolaides KH. Endothelial dysfunction and raised plasma concentrations of asymmetric dimethylarginine in pregnant women who subsequently develop pre-eclampsia. Lancet 2003;361: 1511-7. 12. Rutherford RAD, McCarthy A, Sullivan MHF, Elder MG, Polak JM, Wharton J. Nitric oxide synthase in human placenta and umbilical cord from normal, intrauterine growth-retarded and preeclamptic pregnancies. Br J Pharmacol 1995;116:3099-109. 13. Neri I, Mazza V, Galassi MC, Volpe A, Facchinetti F. Effects of L-arginine on utero-placental circulation in growth-retarded fetuses. Acta Obstet Gynecol Scand 1996;75:208-12. 14. Morris NH, Sooranna SR, Learmont JG, Poston L, Ramsey B, Pearson JD, et al. Nitric oxide synthase activities in placental tissue from normotensive, pre-eclamptic and growth retarded pregnancies. Br J Obstet Gynaecol 1995;102:711-4. 15. Obwegeser R, Oguogho A, Ulm M, Berghammer P, Sinzinger H. Maternal cigarette smoking increases F2-isoprostanes and reduces prostacyclin and nitric oxide in umbilical vessels. Prostaglandins Other Lipid Mediat 1999;57:269-79. 16. Ulm MR, Plo¨ckinger B, Pirich C, Gryglewski RJ, Sinzinger HF. Umbilical arteries of babies born to cigarette smokers generate less prostacyclin and contain less arginine and citrulline compared with those of babies born to control subjects. Am J Obstet Gynecol 1995;172:1485-7. 17. Andersen MR, Stender S. Endothelial nitric oxide synthase activity in aorta of normocholesterolemic rabbits: regional variation and the effect of estrogen. Cardiovasc Res 2000;47:192-9. 18. Butler NR, Goldstein H, Ross EM. Cigarette smoking in pregnancy: its influence on birth weight and perinatal mortality. BMJ 1972;2:127-30. 19. Lyall F, Greer IA, Young A, Myatt L. Nitric oxide concentrations are increased in the feto-placental circulation in intrauterine growth restriction. Placenta 1996;17:165-8. 20. Kaijser M, Granath F, Jacobsen G, Cnattingius S, Ekbom A. Maternal pregnancy estriol levels in relation to anamnestic and fetal anthropometric data. Epidemiology 2000;11:315-9. 21. Mochizuki M, Maruo T, Masuko K, Ohtsu T. Effects of smoking on fetoplacental-maternal system during pregnancy. Am J Obstet Gynecol 1984;149:413-20. 22. Cockell AP, Poston L. 17b-estradiol stimulates flow-induced vasodilatation in isolated small mesenteric arteries from prepubertal female rats. Am J Obstet Gynecol 1997;177: 1432-8. 23. Knock GA, Poston L. Bradykinin-mediated relaxation of isolated maternal resistance arteries in normal pregnancy and preeclampsia. Am J Obstet Gynecol 1996;175:1668-74. 24. Maeda T, Yoshimura T, Okamura H. Asymmetric dimethylarginine, an endogenous inhibitor of nitric oxide synthase, in maternal and fetal circulation. J Soc Gynecol Investig 2003;10:2-4. 25. Teerlink T, Neele SJM, De Jong S, Netelenbos JC, Stehouwer CDA. Oestrogen replacement therapy lowers plasma levels of asymmetrical dimethylarginine in healthy postmenopausal women. Clinical Science 2003;105:67-71.