|Home | About | Journals | Submit | Contact Us | Français|
Epidemiological studies have shown a clear association of adverse intrauterine environment and an increased risk of cardiovascular diseases and hypertension in adult life. The present study tested the hypothesis that prenatal cocaine exposure causes reprogramming of vascular reactivity, leading to an increased risk of hypertension in adult offspring. Pregnant rats received cocaine (30 mgkg-1day-1) or saline from days 15 to 21 of gestational age and experiments were conducted in 3-month-old offspring. Cocaine had no effect on the baseline blood pressure, but significantly increased norepinephrine-stimulated blood pressure and decreased the baroreflex sensitivity in male but not female offspring. The cocaine treatment significantly increased norepinephrine-induced contractions in pressurized resistance-sized mesenteric arteries but not in aortas, which was primarily due to a loss of eNOS-mediated inhibition and an enhanced Ca2+ sensitivity in mesenteric arteries. Additionally, the cocaine treatment significantly attenuated the endothelium-dependent relaxation in mesenteric arteries in male but not female offspring. eNOS protein levels in aortas but not mesenteric arteries were significantly increased in the cocaine-treated animals. However, cocaine significantly decreased phosphorylation levels of eNOS in both aortas and mesenteric arteries. The results suggest that prenatal cocaine exposure programs vascular contractility via changes in eNOS-regulated Ca2+ sensitivity of myofilaments in the sex- and tissue-dependent manners in resistance arteries leading to an increased risk of hypertension in male offspring.
Epidemiological studies have shown a clear association of adverse intrauterine environment and an increased risk of cardiovascular disease and hypertension in adult life.1-4 In utero exposure to cocaine is a significant public health problem. Offspring born to mothers with a history of cocaine abuse have a high incidence of congenital cardiovascular malformations.5 Recently, we have demonstrated in a rat model that fetal exposure to cocaine during gestation results in an increase in heart susceptibility to ischemia and reperfusion injury in adult male offspring.6 Additionally, other studies have demonstrated that prenatal cocaine exposure results in an attenuated vasodilation and enhanced responses to vasoconstrictors in cerebral arteries in the neonate.7,8
It is unknown whether or to what extent fetal cocaine exposure on the vascular reactivity of resistance vessels in adult offspring. Given the importance of resistance arteries in the regulation of blood pressure (BP), dysfunction of resistance arteries is likely to contribute to the development of hypertension observed frequently in adult offspring who experienced an adverse intrauterine environment before birth.9, 10 To investigate the fetal programming of cardiovascular function in response to fetal cocaine exposure, the present study was designed to test the hypothesis that maternal cocaine administration during pregnancy causes reprogramming of vascular reactivity leading to an increased risk of hypertension in adult offspring. The specific aims of the present study were to determine whether and to what extent prenatal cocaine exposure affects the baseline and norepinephrine-stimulated BP in vivo, KCl- and norepinephrine-induced contractions, the endothelium-dependent relaxations, and endothelial nitric oxide synthase (eNOS) activity and protein levels in resistance-sized mesenteric arteries in adult offspring. Additionally, we determined the role of Ca2+ signaling in the cocaine-mediated programming of vascular reactivity. To determine the differential effects of prenatal cocaine on large and resistance vessels in the offspring, the studies were also performed in aortas. To investigate the potential sex effects of prenatal cocaine exposure, the studies were performed in both male and female offspring.
An expanded Methods section is available in the online data supplement at http://hyper.ahajournals.org.
Pregnant rats were divided into two groups: 1) saline control; and 2) cocaine 15 mg/kg i.p. twice daily from day 15 to 21 of gestational age, as described previously.6 Offspring were studied at 3-month old. All procedures and protocols were approved by the Institutional Animal Care and Use Committee guidelines.
Baseline and norepinephrine-stimulated BP was measured in offspring as described previously.11
Norepinephrine-induced contractions were measured in aortic rings in the absence or presence of L-NNA (100 μM, 20 min) at 37°C, as described previously.12 For relaxation studies, the tissues were pre-contracted with the sub-maximal concentration (1 μM) of norepinephrine followed by acetylcholine added in a cumulative manner.
Resistance-sized mesenteric arteries (~200 μm in diameter) were loaded with fura 2-AM and pressurized to 45 mmHg in an organ chamber as previously described.11,13,14 Norepinephrine-induced contractions were measured in the absence or presence of L-NNA (100 μM, 20 min). For relaxation studies, the arteries were pre-contracted with 3 μM norepinephrine followed by acetylcholine.
RNA was extracted from tissue samples with TRIzol, and eNOS mRNA abundance was determined by real-time RT-PCR using gene-specific primers.
Data are presented as means ± SEM. Experimental number (n) represents offspring from different dams. The differences were evaluated for statistical significance (P < 0.05) by two-way ANOVA or by t-test, where appropriate.
As shown in Table S1, prenatal cocaine exposure had no significant effects on baseline arterial SBP, DBP, MAP, HR, or body weight in either male or female offspring at 3-month old.
Norepinephrine produced time-dependent increases in arterial BP in both control and cocaine-treated offspring. In 3-month old male offspring, prenatal cocaine caused significant increases in norepinephrine-stimulated SBP, DBP and MAP (Figure 1). HR was not affected (Figure S1). The increased BP in response to norepinephrine resulted in a decrease in HR via baroreflex. Prenatal cocaine treatment resulted in a significant decrease in the baroreflex sensitivity (control: 0.83±0.11 vs. cocaine: 0.45±0.04 msec/mmHg, P < 0.05). In contrast to the findings in males, norepinephrine-induced changes of SBP, DBP and MAP in females were not significantly different between the control and cocaine-treated animals (Figure S2). Additionally, the baroreflex sensitivity was the same between the two groups (0.31 ± 0.07 vs. 0.31 ± 0.12 msec/mmHg).
Prenatal cocaine exposure had no significant effects on KCl-induced contractions of aortas (control, 1.9±0.1 g/mm2 vs. cocaine, 2.0±0.2 g/mm2; P > 0.05) and pressurized mesenteric arteries (control, 64.7±12.5% vs. cocaine, 68.6±11.9%, diameter changes at pressure of 45 mmHg; P > 0.05) in 3-month old male offspring. In aortas, norepinephrine-induced concentration-dependent contractions were not significantly altered by the cocaine treatment regardless the presence or absence of eNOS inhibitor L-NNA (Table 1). In resistance-sized mesenteric arteries, prenatal cocaine treatment resulted in significant increases in the pD2 value and maximum response of norepinephrine-induced contractions in the absence of L-NNA (Figure 2A, 2B and Table 1). In the control animals, inhibition of eNOS with L-NNA significantly potentiated norepinephrine-induced contractions and increased the norepinephrine-mediated maximal response (Figure 2A and Table 1). In contrast, in the cocaine-treated animals, norepinephrine-induced contractions were not significantly affected by L-NNA (Figure 2B and Table 1). In the presence of L-NNA, there was no significant difference in norepinephrine-induced contractions of mesenteric arteries between the control and cocaine-treated animals (Table 1).
[Ca2+]i were measured simultaneously in the same tissues of pressurized mesenteric arteries in which norepinephrine-mediated contractions were determined. Consistent with concentration-dependent vasoconstrictions and decreases in the arterial diameter of pressurized mesenteric arteries (Figure 2A and 2B), norepinephrine produced concentration-dependent increases in [Ca2+]i (Figure 2C and 2D). Prenatal cocaine treatment showed no significant effects on the pD2 values and maximum responses of norepinephrine-induced Ca2+ mobilization in the absence (pD2: 5.93±0.18 vs. 6.13±0.10, P > 0.05; Emax: 1.12±0.03 vs. 1.13±0.02 Rf340/f380, P > 0.05) or presence of L-NNA (pD2: 6.10±0.18 vs. 6.08±0.10, P > 0.05; Emax: 1.05±0.02 vs. 1.08±0.02 Rf340/f380, P > 0.05). Additionally, L-NNA did not significantly affect norepinephrine-induced Ca2+ mobilization.
The simultaneous measurement of [Ca2+]i and contractions in the same tissue allowed us to determine the norepinephrine-mediated [Ca2+]i–contraction relation in resistance mesenteric arteries. Figure 2E and 2F showed a positive correlation between increased [Ca2+]i and contractions induced by norepinephrine in mesenteric arteries of both control and cocaine-treated animals. Prenatal cocaine treatment resulted in a significant increase in the slope (Δdiameter/Δ[Ca2+]i) of norepinephrine-induced responses in the absence of L-NNA (734±22 vs. 496±25, P < 0.05). Whereas L-NNA significantly increased the slope of norepinephrine-induced [Ca2+]i and contractions in control animals (677±45 vs. 496±25, P < 0.05), it had no significant effect on the relation in cocaine-treated animals (783±31 vs. 734±22, P > 0.05). In the presence of L-NNA, there was no significant difference in norepinephrine-induced Ca2+ sensitivity in mesenteric arteries between the control and cocaine-treated animals (677±45 vs. 783±31, P > 0.05).
Acetylcholine produced concentration-dependent relaxations in both aortas and pressurized resistance mesenteric arteries. In male offspring, prenatal cocaine exposure resulted in a significant decrease in the maximal relaxation induced by acetylcholine in mesenteric arteries (Emax: control, 80.0±2.8% vs. cocaine, 48.7±4.5%; P < 0.05) (Figure 3A), but not in aortas (Emax: control, 61.9±2.8% vs. cocaine, 67.0±3.1%; P > 0.05) (Figure S3). In females, cocaine had no significant effect on acetylcholine-induced relaxations in either aortas or mesenteric arteries (Figure S4).
Acetylcholine-mediated relaxation and reduction in vascular wall [Ca2+]i in mesenteric arteries were measured simultaneously in the same tissues. Consistent with acetylcholine-induced relaxations, it produced concentration-dependent reductions in [Ca2+]i in mesenteric arteries pre-contracted with norepinephrine, which were not significantly different in the arteries between control and cocaine-treated animals (Figure 3B). Analysis of the relations between the decrease in vascular wall [Ca2+]i and relaxation depicted from the results of simultaneous measurements of [Ca2+]i and diameter changes in pressurized resistance mesenteric arteries indicated that there was a significant difference in acetylcholine-induced -Δ[Ca2+]i-relaxation relation in mesenteric arteries between control and cocaine-treated animals, with a significant decreased relaxation at the same level of [Ca2+]i reduction in the arteries from cocaine-treated animals (Figure 3C).
eNOS mRNA and protein abundance and phospho-eNOSser1179 levels in aortas and mesenteric arteries in male offspring were determined by real-tine RT-PCR and immunoblotting. As shown in Figure 4A, prenatal cocaine treatment had no significant effect on eNOS mRNA abundance in either aortas or mesenteric arteries. However, eNOS protein levels were significantly increased in the aorta, but not in the mesenteric arteries, from cocaine-treated animals as compared with those from the control (Figure 4B). In contrast, the ratio of phospho-eNOSser1179 to total eNOS protein was significantly decreased in both aortas and mesenteric arteries from cocaine-treated animals (Figure 4C).
The present study has demonstrated in a rat model that prenatal cocaine exposure causes a hypertensive reactivity of resistance-sized mesenteric arteries in adult male offspring. The major findings of the present study are that 1) prenatal cocaine exposure caused a sex-dependent increase in arterial BP response to norepinephrine and a decrease in the baroreflex sensitivity in male offspring, 2) α1-adrenoreceptor agonist norepinephrine-induced vascular contractions are significantly enhanced in resistance mesenteric arteries but not in aortas in response to prenatal cocaine exposure, 3) this functional alteration of vascular reactivity is independent of changes in Ca2+ mobilization, but mainly depends on enhanced Ca2+ sensitivity in mesenteric arteries, 4) the basal eNOS activity inhibits norepinephrine-mediated contractions of resistance mesenteric arteries but not aortas, which is abolished by the cocaine treatment, 5) endothelium-dependent relaxation is significantly attenuated in mesenteric arteries but not in aortas in male offspring in a sex-dependent manner, which is mediated by increased Ca2+ sensitivity in mesenteric arteries, 6) fetal cocaine exposure significantly increased eNOS protein levels in aortas, but decreased the ratio of phosphorylated eNOS in both aortas and mesenteric arteries of adult offspring.
The finding that fetal cocaine exposure increased susceptibility of elevated BP in adult offspring in the present study is consistent with recent studies in humans and animal models showing a link between adverse intrauterine environments and fetal programming, resulting in an increased risk of hypertension and ischemic heart disease in adulthood.1,16-18 Previous studies suggested that intrauterine cocaine exposure increased a risk of persistently elevated blood pressure during early and later childhood.19-22 The present finding that prenatal cocaine affected male offspring predominantly is in agreement with previous studies showing that female offspring were less sensitive in manifestation of hypertension caused by adverse prenatal insults.3 Additionally, our recent study has demonstrated that fetal cocaine results in increased heart susceptibility to ischemia and reperfusion injury in adult male but not female offspring.6 These studies suggest sex-dependent fetal programming of cardiovascular dysfunction induced by prenatal cocaine exposure. It is likely that multiple mechanisms are involved in fetal programming of cardiovascular response. In the present study, we found that prenatal cocaine treatment caused a significant decrease in the baroreflex sensitivity in male offspring. Consistent with this finding, previous studies have demonstrated that cocaine suppresses baroreflex control of blood pressure.23,24 Studies in other models of fetal programming also showed a decrease in baroreflex function in offspring,11,25,26 suggesting a common mechanism of impaired baroreflex in fetal programming of cardiovascular response in offspring.
In the present study, KCl-induced contractions of both aortas and resistance mesenteric arteries were not significantly different between the control and cocaine-treated animals, suggesting that electromechanic-mediated contractile signal pathway in arteries is not altered by prenatal cocaine exposure. This is in agreement with the previous finding that glucocorticoid exposure during early gestation had no significant effect on KCl-induced contractions in newborn lamb coronary arteries.27 In contrast, our recent study showed that KCl-induced vascular contractility was enhanced in adult rats that exposed to nicotine before birth.12 These findings suggest a stimuli-specificity in fetal programming of electromechanic coupling in vascular smooth muscle in adult offspring.
Unlike the finding with KCl-induced contractions, prenatal cocaine exposure significantly increased norepinephrine-induced contractions of resistance mesenteric arteries in adult offspring. In contrast, norepinephrine-mediated contractions of aortas were not affected by the cocaine treatment, suggesting that fetal cocaine causes a tissue-selective programming of resistance vascular reactivity to the sympathetic neurotransmitter in adult offspring. Consistent with the present finding of altered agonist-mediated vascular contractions, Yakubu et al.7 have reported that prenatal cocaine exposure significantly enhances 5-HT-, ET-1- and clonidine-induced constrictions of cerebral arterioles in neonate piglets.
It is likely that multiple mechanisms are involved in fetal programming of vascular hypertension. Previous studies demonstrated that in utero cocaine exposure altered α-adrenoreceptor mRNA levels and binding sites in the brain28-30 as well as myocardial β-adrenoceptor signaling pathway in neonatal rats.31 In the present study, inhibition of eNOS by L-NNA significantly increased norepinephrine-induced contractions of mesenteric arteries in control animals, suggesting a significant component of basal eNOS activity in suppressing α-adrenoreceptor-mediated contractions. The cocaine treatment abolished the effect of L-NNA, and there was no significant difference in norepinephrine-mediated contractions of mesenteric arteries between control and cocaine-treated animals in the presence of L-NNA. These results suggest that cocaine-mediated enhancement of norepinephrine-induced contractions in resistance mesenteric arteries are primarily due to the loss of eNOS-mediated relaxation component, rather than increased norepinephrine-induced contractions per se in the resistance arteries. This is supported by the finding of a significant decrease in acetylcholine-induced relaxation of mesenteric arteries, but not aortas, in cocaine-treated offspring. Consistent with the sex-dependent BP response observed, the decreased endothelium-dependent relaxation of mesenteric arteries was found only in male offspring. The involvement of endothelium/NO in fetal programming of vascular function has been studied and the results are controversial.12,32-35 Our recent study demonstrated that enhanced α1-adrenoceptor-mediated contractions of the aorta in male adult offspring after prenatal nicotine exposure was primarily due to the loss of the eNOS-mediated relaxation component in α1-adrenoceptor-mediated contractions.12 In contrast, fetal nicotine enhanced arterial sensitivity to angiotensin II primarily due to increased angiotensin II-induced contractions per se rather than the loss of the eNOS-mediated relaxation component.14 Taken together, these findings suggest tissue-specific and stimuli-dependent fetal programming of endothelial function and vascular reactivity.
To evaluate endothelium-dependent mechanisms, eNOS protein abundance and phosphorylation levels were examined in both aortas and mesenteric arteries. In aortas, the cocaine treatment significantly enhanced eNOS protein, but not mRNA, abundance, suggesting that cocaine-mediated programming of eNOS expression in the aorta is at the translational level. The finding of attenuated eNOS phosphorylation levels suggests a counterbalance of eNOS protein expression and its activity, which may lead to the minimal effect on endothelium-dependent relaxation observed in the aorta of cocaine-treated animals. Unlike aortas, the finding of no difference in eNOS protein abundance but a significant decrease in eNOS phosphorylation levels in mesenteric arteries of cocaine-treated animals suggests that decreased eNOS activity contributes to the decreased endothelium-dependent relaxation and the increased vascular contractility of resistance arteries in adult offspring that exposed to cocaine before birth. This is in agreement with the previous studies showing that intrauterine malnutrition decreased eNOS activity and NO production without affecting eNOS gene expression in rat mesenteric arteries.36 Additionally, our recent study in a rat model of prenatal nicotine exposure demonstrated that alteration of eNOS activity without changes in eNOS protein abundance contributed to the vascular dysfunction.12 Although the present study provides evidence of a correlation between decreased phospho-eNOSser1179 levels and impaired endothelium-dependent relaxation, it should be noted that Ser1179 is not the only site that can be phosphorylated, and changes in Ser1179 phosphorylation and eNOS activity are not seen in all endothelial studies. Whether and to what extent fetal cocaine may cause programming of other phosphorylation events of eNOS remain an intriguing area for future investigation.
We determined further the downstream mechanisms of intracellular Ca2+ mobilization and Ca2+ sensitivity in cocaine-mediated fetal programming of resistance arterial contractility, and found that the enhanced norepinephrine-induced contractions of resistance arteries in cocaine-treated animals were not associated with increased [Ca2+]i. In the present study, the simultaneous measurement of [Ca2+]i with tension development in the same intact tissue allowed us to determine directly the precise relationship between [Ca2+]i and tension in arteries and thus to determine the Ca2+ sensitivity of myofilaments with unimpaired excitation-contraction coupling processes and retained regulatory targets for second messenger pathways. The finding of enhanced Δdiameter/Δ[Ca2+]i induced by norepinephrine in resistance mesenteric arteries of cocaine-treated animals indicates an increase in the Ca2+ sensitivity of myofilaments in the resistance arteries. Indeed, the increased Ca2+ sensitivity has been demonstrated to be critical in the fetal programming of vascular dysfunction and hypertension.11, 27 It is likely that inhibition of the eNOS activity plays a key role in fetal programming of the Ca2+ sensitivity in resistance arteries in cocaine-treated animals (see Figure S5 for a diagram). This is supported by the following evidence obtained in the present study. L-NNA-mediated increase in norepinephrine-induced contractions of resistance mesenteric arteries in control animals were not associated with increased [Ca2+]i but rather with an increase in Δdiameter/Δ[Ca2+]i, indicating that basal eNOS activity inhibits the Ca2+ sensitivity in resistance arteries. In the presence of L-NNA, there was no significant difference in norepinephrine-induced Ca2+ sensitivity in resistance mesenteric arteries between control and cocaine-treated animals. The role of eNOS in programming of the Ca2+ sensitivity in resistance arteries was further supported by the finding of a significant decrease in acetylcholine-induced relaxation at the same level of [Ca2+]i reduction in mesenteric arteries from cocaine-treated animals. It has been demonstrated in permeabilized vascular smooth muscle preparations that addition of cGMP-preactivated cGMP-dependent protein kinase reduces myosin phosphorylation and contractile tension proportionally despite a constant [Ca2+]i, indicating that cGMP inhibits the Ca2+ sensitivity of contractile proteins.37
The present study has demonstrated that prenatal cocaine exposure results in reprogramming of eNOS activity and enhanced Ca2+ sensitivity in resistance arteries leading to an increased resistance arterial contractile sensitivity and blood pressure response to norepinephrine in a sex-dependent manner. These findings indicate that fetal cocaine exposure not only causes increased perinatal morbidity and mortality as previously recognized,38 but also has long lasting effects and increases the susceptibility of elevated arterial blood pressure later in adult life. The present finding is consistent with recent studies in humans and animal models and re-enforces the notion that multiple in utero adverse factors during pregnancy cause fetal programming and increase the risk of hypertension in the adult. Given that cocaine abuse in pregnant women is a significant problem, the present finding suggests that intrauterine cocaine exposure may be a significant risk factor for cardiovascular disease in later life in male offspring. As is often the case with novel findings, the present study may raise more questions than it answers. For instance, are the effects mediated by a direct effect of cocaine on the fetus or indirectly through its effect on the mother? What are the epigenetic mechanisms involved in reprogramming of the eNOS activity in resistance arteries, which persist into adulthood? In addition, whether or to what extent do sex steroid hormones contribute to the sex differences in cocaine-mediated fetal programming of resistance vascular function? Undoubtedly, these questions warrant further investigations.
Sources of Funding: This work was supported in part by National Institutes of Health Grants HL82779 (LZ) and HL83966 (LZ), and S06GM073842 (SY).