Vitamin C Prevents the Effects of Prenatal Nicotine on Pulmonary Function in Newborn Monkeys

Despite vigorous antismoking campaigns, approximately 12% of women still smoke during pregnancy, corresponding to more than 450,000 smoke-exposed infants born in the United States in 2002 (1). Smoking during pregnancy causes premature delivery and intrauterine growth retardation and has been estimated to cause 5 to 10% of all fetal and neonatal deaths (2, 3). Fetal lung is particularly sensitive to the effects of maternal smoking. Ln Mfcro-exposed infants have decreased pulmonary function at birth (4-7), which persists throughout childhood (8-10) and beyond (11), increased risk of hospitalization for respiratory illnesses (4, 12, 13), increased risk of childhood asthma (14-16), and increased risk of dying of sudden infant death syndrome (17, 18). Despite these multiple effects, the mechanism by which smoking affects fetal development is unclear. Previous work from our laboratory (19-22) and other laboratories (23-27) has demonstrated, however, that many of the effects of smoking during pregnancy are mediated by nicotine crossing the placenta to interact with nicotinic receptors in developing lung.

Fetal lung expresses multiple nicotinic receptor subtypes in airway epithelial cells, airway fibroblasts, and pulmonary type II cells (19, 28). Nicotine freely crosses the placenta, and levels of nicotine in amniotic fluid of human smokers are equal to or higher than maternal plasma levels (29, 30). These concentrations of nicotine are sufficient to modulate signaling by nicotinic receptors present in fetal lung (31). Previously, we have demonstrated that prenatal nicotine exposure to timed-pregnant rhesus monkeys produces changes in offspring's pulmonary function that are similar to the changes seen in offspring of human smokers (20). This includes decreases in expiratory flows and volumes, and these changes are associated with altered levels of collagen and elastin in the offspring lung (21). The characterization that nicotine itself specifically mediates changes in lung development suggests there may be ways to block these effects.

Multiple studies have shown a protective effect of vitamin C on lung function. Increased vitamin C intake is associated with decreased chronic obstructive pulmonary disease in adult smokers (32) and increased expiratory flows in adults (33, 34). Conversely, decreased vitamin C intake has been associated with adult-onset wheezing (35, 36) and lower forced expiratory volumes (FEVs) in children (37). Specifically suggesting that vitamin C might also be protective against the harmful effects of smoking on lung development, Maritz and van Wyk (38, 39) have demonstrated that vitamin C protects against some of the effects of nicotine on lung development in a rat model. Given that we have previously reported that prenatal nicotine alters pulmonary function as measured by decreased forced expiratory flows (FEFs) and FEVs, it seemed possible that vitamin C supplementation might also prevent these effects of nicotine. Our data presented here now show that vitamin C supplementation to nicotine-treated, pregnant rhesus monkeys prevents some of the deleterious effects of nicotine on newborn pulmonary function.

METHODS

Experimental Design

Female Rhesus monkeys were mated for 5 days, and on Days 23 to 25 of mating and 2 weeks thereafter, ultrasound examinations were performed to identify and establish conception date. Pregnant females were randomly allocated to a control group (n = 6), nicotine-treated group (n = 7), or nicotine plus vitamin C group (n = 7). On Day 26 of pregnancy, control animals received subcutaneous (midscapular) Alzct miniosmotic 2ML4 pumps containing either bacteriostatic water (Abbott Laboratories, North Chicago, IL) or nicotine bitartrate (Sigma, St. Louis, MO) in bacteriostatic water to deliver 2.0 mg/kg/day nicotine bitartrate (equivalent to 0.7 mg free nicotine/kg bodyweight/day). Pumps were replaced every 3 weeks. Amniocentesis was performed on Gestation Day 140. Animals received Cefazolin (150 mg/twice a day) for 3 days after pump insertion and/or amniocentesis. The vitamin C-treated group received 250 mg supplemental vitamin C per day in the form of a children's chewable vitamin. All groups were similarly fed Purina Mills (Richmond, IN) Lab Diet High-Protein Monkey Diet, which contains 0.75 mg of vitamin C per gram of diet, and average daily vitamin C intake from the diet was approximately 50 mg. In addition, all groups received multivitamins twice weekly (Zoo Chews; IVC Industries, Portland, OR), which contained 60 mg additional 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.

To ensure all offspring were tested at the same gestational age, fetuses were delivered on Gestation Day 160 (term, 165 days) by cesarean section. To eliminate any postnatal effects of the maternal nicotine administration, the newborn monkeys were then transferred to the primate center nursery where they were hand-fed standard milk formula. At the time of delivery, maternal blood, cord blood, and amniotic fluid were collected for analyses, as described later. Nicotine levels were measured by gas chromatography, as previously described (40). Surfactant apoprotein levels were measured in amniotic fluid by ELISA, as previously described (41, 42). Cytokines were measured by ELISA, as previously described (43).

Pulmonary Function Testing

Pulmonary function testing was performed 1 day after birth (~ 24-26 hours), as described previously (20). In brief, animals were anesthetized with ketamine and tracheostomized, and pulmonary function tests were performed using the Buxco BioSystem for Maneuvers hardware and software (Buxco Electronics, Inc., Wilmington, NC) for invasive pulmonary maneuvers. Using the Buxco pulmonary maneuvers protocol, tidal breathing and fast flow volume maneuvers were performed. From the tidal breathing maneuver, VT, peak tidal expiratory flow (PTEF), time to PTEF as a proportion of total expiratory time (tPTEF:tE), specific airway resistance, and dynamic compliance were determined. From the fast flow maneuver, inspiratory capacity, expiratory residual volume, FEV in 100, 200, and 400 milliseconds, peak expiratory flow (PEF), mean midexpiratory flow (FEF^sub 25-75%^, the average flow between 25 and 75% of FEV), FEF after 25, 50, and 75% of FVC was expired, and FEV at PEF (FEV^sub PEF^) were determined. After pulmonary function maneuvers, animals were killed with pentobarbital. The abdominal aorta was transected; the lungs flushed with normal saline were removed, blotted, and weighed. The left lobes with intact trachea were fixed with zinc formalin (Fisher Scientific, Pittsburgh, PA) at 20 cm constant transpulmonary water pressure for 72 hours. Left lung volume was measured by water displacement, and total lung volume was extrapolated using total lung weight. The right lobes were separated and frozen for analysis of RNA and protein.

Biochemical, Histologic, and Molecular Analyses

Lung tissue containing major airways and vessels was separated from the peripheral parenchymal tissue and designated as "central" and "peripheral" lung tissue. This distinction was made to determine if there was a difference in the central and peripheral compartments of the lung. Total lung elastin and collagen in the central and peripheral lung tissue content were determined by measuring desmosine and hydroxyproline, the modified amino acids unique to those proteins, respectively. Desmosine was measured by radioimmunoassay and hydroxyproline by amino acid analysis (44). The expression patterns of collagen and elastin in lung sections were examined by Mason's trichrome and Hart's elastin stains, as previously described (19, 44). The intensity of staining was qualitatively estimated on a scale of 1 to 4 in large airways, large vessels, and lung parenchyma by two readers (B.J.P., H.S.S.) blinded to treatments. Real-time polymerase chain reaction to quantify rhesus elastin, collagen I, collagen III, and surfactant apoprotein-B (SP-B) mRNAs were performed, as previously described (21), using 18S RNA as an internal standard in duplex assays. A real-time polymerase chain reaction assay for rhesus chromogranin A was established first by cloning the 3' end of rhesus chromogranin A and then using Applied Biosystems (Foster City, CA) Primer Express software to design the needed primers and probes. The sequence of rhesus chromogranin A has been deposited in GenBank with accession AY646928.

Statistical Analysis

Data are expressed as means

RESULTS

This study consisted of three groups: control, nicotine-treated, and nicotine plus vitamin C-treated. Nicotine and cotinine levels in maternal serum, amniotic fluid, and cord blood at the time of delivery are shown in Table 1. These values were all in the range of those observed in pregnant human smokers (29, 30). Furthermore, levels of nicotine were significantly higher in the amniotic fluid of the vitamin C plus nicotine group, although maternal plasma levels were similar in both groups. Nicotine and cotinine were undetectable in amniotic fluid from control animals. No spontaneous abortion or fetal resorption occurred in any groups. Maternal weight gain of the nicotine-treated group and the nicotine plus vitamin C group did not differ from control animals. As observed in our previous studies, infant weight was slightly decreased by nicotine treatment (Table 2).

Consistent with our previous reports and studies of offspring of human smokers, FEFs were significantly decreased in the nicotine-exposed group (Figure 1). This reduction was, however, prevented by maternal vitamin C supplementation (250 mg/day). A similar trend was seen for FEVs, although effects were not significant (Figure 2). Specific airway resistance, dynamic compliance, and tPTEF:tE were not significantly changed between the three groups (Table 2).

As shown in Figure 3, prenatal nicotine exposure significantly increased levels of SP-B at 160 days' gestation in amniotic fluid, but vitamin C supplementation blocked the nicotine-induced increases in SP-B. Parallel, although nonsignificant, changes were observed in lung SP-B mRNA levels. As shown in Figure 3, nicotine exposure increased levels of SP-B mRNA in newborn monkey lung tissue and this increase was also prevented by vitamin C supplementation. No significant changes were seen in surfactant apoprotein-A levels (data not shown).

One of the possible mechanisms by which nicotine affects lung development is through the release of hormones and cytokines. The signaling molecules that have been shown to affect lung development include the following: cortisol; leptin; and the cytokines interleukin 1a (IL-1a), IL-1ß, IL-2, IL-6, IL-8, transforming growth factor ß, and tumor necrosis factor a. As shown in Table 3, none of these levels were significantly changed or elevated by nicotine alone or in combination with vitamin C in amniotic fluid. In addition, levels of a-fetoprotein, cortisol, and leptin were also measured in cord blood and similarly did not differ between the three groups (Table 3). Thus, the nicotine-induced changes in pulmonary function, surfactant expression, and the amelioration by vitamin C do not appear to be clearly mediated by cortisol, leptin, or the cytokines measured. It should be noted, however, that there were small increases in IL-1ß and transforming growth factor ß with nicotine treatment that were prevented by vitamin C, which deserve further study.

Another potential mediator of the effects of prenatal nicotine exposure on lung development is the secretion of bioactive molecules from pulmonary neuroendocrine cells (PNECs) (45-47). Previously, we (19) and others (48) have reported that prenatal nicotine exposure increases the number of PNECs at birth. As an index of PNEC number, levels of chromogranin A RNA levels were measured (49). As shown in Table 4, nicotine increased chromogranin A levels, although the change was not statistically significant. Vitamin C had no additional effect on chromogranin A levels, thus providing no evidence to suggest that changes in number of PNECs mediate the ability of vitamin C supplementation to prevent the effects of nicotine on pulmonary function.

Our previous studies have shown that prenatal nicotine increased collagen I, collagen III, and elastin mRNA levels, and that changes in connective tissue could explain the nicotine-induced changes in pulmonary function (21 ). In this study, consistent with our previous results, prenatal nicotine exposure slightly increased collagen I mRNA (24%), collagen III (13%), and elastin mRNA (22%) levels (Table 4). Vitamin C supplementation did not prevent these mRNA increases. Protein levels of collagen and elastin were measured by assaying levels of hydroxyproline and desmosine in lung, respectively. There were no changes in hydroxyproline levels (data not shown) between groups. However, lung elastin content as measured by lung desmosine levels was decreased by nicotine, and the nicotine-induced decrease in lung desmosine was partially prevented by vitamin C supplementation (Figure 4). Consistent with the relatively modest changes in desmosine content seen in Figure 4, no clear changes in elastin or collagen distribution were evident with nicotine and vitamin C treatment as determined by qualitative analysis of Hart's elastin stain or trichrome-stained tissue sections. Representative images of elastin expression in the lungs from the three groups is shown in Figure 5.

DISCUSSION

Smoking during pregnancy continues to be a major problem. Given the addictive nature of nicotine and continued advertising by tobacco companies, which in 2001 exceeded $11 billion in the United States alone (50, 51), it is a problem that is not likely to go away in the near future. Thus, looking for ways to lessen the impact of smoking during pregnancy are important, even while keeping a major focus on decreasing and eliminating smoking during pregnancy. Previous research from our laboratory (19-22) and other laboratories (23, 24, 26, 27, 38, 39) has demonstrated that some of the effects of smoking during pregnancy are mediated by nicotine, which can cross the placenta to interact with nicotinic receptors in developing lung. This finding suggests that the actions of nicotine may provide a target to block some of the effects of smoking during pregnancy. Studies by Maritz and van Wyk (38, 39) have demonstrated that vitamin C supplementation blocks some of the effects of nicotine on lung development in rats, and multiple studies have shown that vitamin C supplementation is protective against the effects of smoking on development of chronic obstructive pulmonary disease (32, 33, 35, 36). Thus, we examined whether vitamin C supplementation could prevent effects of prenatal nicotine exposure on pulmonary function in newborn rhesus monkeys.

Timed-pregnant rhesus monkeys were treated with 2 mg/kg/ day nicotine bitartrate from Gestation Days 26 to 160 (term, 165 days). This dose is equivalent to 0.7 mg/kg free nicotine per day and is similar to the dose of nicotine received by heavy smokers (40, 52). This dose accurately models human smoking during pregnancy because average concentrations of nicotine and cotinine in monkey amniotic fluid and maternal blood obtained are similar to that seen in human smokers (29). Vitamin C was given as a single dose of 250 mg/day. Vitamin C consumption by unsupplemented monkeys was approximately 50 mg/day; thus, this dose represented an approximate fivefold supplementation. This relatively high dose of vitamin C was chosen to maximize the likelihood of observing protective effects given the low toxicity of vitamin C during pregnancy (53-55). By contrast, the standard dose of vitamin C in human prenatal vitamins is 90 to 120 mg. Thus, the effects of lower doses of vitamin C need to be explored in follow-up studies. The levels of nicotine and cotinine in amniotic fluid and cord blood were higher in the vitamin C-exposed group compared with the nicotine group for reasons that are unclear. This finding does, however, suggest that vitamin C did not act by speeding up the metabolism of nicotine.

The most consistently observed effects of maternal smoking during pregnancy on offspring lung function are alterations in expiratory flows and volumes (5, 6, 8, 9, 14, 56). In our previous study (20), we reported that prenatal nicotine exposure produced significant alterations in expiratory flows and volumes in newborn monkeys. The present study confirmed our previous finding and found consistent significant reductions in FEFs. Importantly, these reductions in flows were prevented when nicotine-exposed animals were simultaneously treated with vitamin C. There was a downward trend in FEVs that was not statistically significant, possibly reflecting the small sample size of the study. That FVC and FEVs decreased proportionately suggest that the defect induced by nicotine is restrictive in nature (57), although obstructive defects as well cannot be ruled out.

We previously reported (21, 22) that prenatal nicotine exposure increased levels of collagen and elastin mRNA in lung, and increased collagen protein but decreased elastin protein. Thus, we have hypothesized that nicotine may interact with nicotinic receptors on fibroblasts (the chief source of connective tissue synthesis) to modify connective tissue expression in lung and lead to altered pulmonary function (19-21). In this study, we again observed that prenatal nicotine exposure led to small increases in collagen and elastin mRNA in lung. To estimate collagen and elastin content, levels of hydroxyproline and desmosine, amino acids unique to collagen and elastin, were measured, respectively, in lung tissue. These assays provide the most accurate quantitative measurement of total collagen and elastin but lack a high degree of anatomic specificity. Using these assays in this study, increased collagen as reflected by levels of hydroxyproline was not observed. Our previous study used immunohistochemical methods to determine and measure the spatial distribution of connective tissue proteins in 132-day-gestation fetuses as opposed to this study in which hydroxyproline was measured in full-term newborns. Because collagen synthesis is significantly increased in the third trimester of gestation (58), the differences between our two studies may reflect gestational age of animals or methods used. However, consistent with previous reports (21, 22), this study showed that prenatal nicotine exposure significantly decreased elastin levels as determined by desmosine content. In addition, as shown in Figure 4, the nicotine-induced decrease in elastin was partially blocked by vitamin C supplementation, although the effect of vitamin C was not significant. Thus, it is possible that vitamin C may alleviate the effects of prenatal nicotine exposure on pulmonary function by partially limiting the reduction in elastin content. This finding is consistent with the results of Maritz and Woolward (23), who have also suggested that nicotine may affect lung development through alterations in elastin levels. Further studies to clarify the role of elastin are necessary, however.

Another potential mechanism by which prenatal nicotine exposure could decrease expiratory flows is by affecting airway structure. This could be altered airway geometry or thicker airway walls consistent with the reports of Elliot and coworkers (59), Collins and coworkers (60), and Maritz and Windvogel (24). Recent observations by Plopper and colleagues (61) of alterations in airway length and caliber in response to developmental insults also provide a potential mechanism. This is an intriguing possibility, and a complete morphometric analysis of the lungs from the animals described here is part of future studies. Our findings of significant decreases in expiratory flows but not significant changes in FEV are consistent with the finding of Gilliland and coworkers (10), who, in examining changes in pulmonary function of schoolchildren exposed to smoke in utero, similarly observed the largest changes as did we in mean mid-expiratory flow (MMEF, same as FEF^sub 25-75%^) and FEF^sub 75%^, and suggested these changes reflected defects in small airways. Furthermore, Gilliland and colleagues (37) have also observed decreases in MMEF in children with low vitamin C intake.

Another consistently observed change in human smoking during pregnancy is increased levels of surfactant for given term and decreased levels of respiratory distress syndrome for premature deliveries (19, 42, 62, 63). Consistent with this finding, our laboratory (19) has shown that prenatal nicotine administration to monkeys increases levels of surfactant apoprotein and number of type II cells, and Wuenschell and colleagues (64) and Maritz and Thomas (65) have also observed that prenatal nicotine increases surfactant and type II cells in mice. This study confirmed that nicotine increases SP-B protein levels and has shown that vitamin C supplementation blocks the nicotine-induced increase in SP-B expression. The mechanism by which smoking and nicotine affect surfactant expression is unknown. Potential mechanisms include the following: direct action of nicotine on type II cells, which express nicotinic receptors (19); a nonspecific stress response mediated by glucocorticoids; or a cytokine- or other signaling molecule-mediated effect.

To begin to explore the mechanism by which prenatal nicotine and vitamin C might affect lung function, we examined several possible mediators that are involved in lung development. First, it is well established that cortisol affects lung maturation and increases surfactant levels (66), but, as shown in Table 3, no significant changes in cortisol in amniotic fluid and cord blood were observed between the three groups. Next, because macrophages and lymphocytes express nicotinic receptors (19, 67), levels of cytokines that have been clearly linked to lung development were measured in amniotic fluid. As shown in Table 3, there were no significant changes between the three groups for the factors measured. Furthermore, although the changes did not reach statistical significance, levels of transforming growth factor ß and IL-1ß were increased in amniotic fluid of the nicotine-treated group and vitamin C prevented those increases. Given that both transforming growth factor ß and IL-1ß have been linked to increased connective tissue expression in lung (68, 69), these changes warrant further study.

Another potential mediator of effects of nicotine and vitamin C are the PNECs. Nicotine has been shown to stimulate secretion from PNECs, and smoking increases the number of PNECs (45-47). PNECs secrete a variety of bioactive compounds, including gastrin-releasing peptide, serotonin, and calcitonin gene-related peptide (70, 71), which could all potentially affect lung development. Chromogranin A is present in the secretory granules of PNECs and is an excellent marker of PNECs (49), and, as expected, nicotine modestly increased levels of chromogranin A (Table 4). Vitamin C supplementation, however, had no additional effect on levels of chromogranin A mRNA (Table 4). This finding suggests that, whereas the stimulation of PNECs may mediate some of nicotine's effects, vitamin C does not appear to interact with the nicotine-induced increase in PNEC number, although an effect of vitamin C on a specific factor secreted by PNECs remains possible.

Thus, although our data show that vitamin C supplementation blocks some of the effects of prenatal nicotine exposure on offspring pulmonary function, the mechanism of action remains unclear. Our results suggest that potential mechanisms could involve elastin or cytokines, but clear conclusions as to how vitamin C acts to block the effects of nicotine cannot yet be drawn. Furthermore, vitamin C has recently also been shown to affect coronary blood flow and arterial stiffness in smokers (72, 73), suggesting there could be a potentially common mechanism by which vitamin C affects both blood vessel and airway function in smokers. Clearly, more studies are needed to characterize the mechanism underlying vitamin C's actions.

Although the data presented here suggest that vitamin C supplementation could be clinically useful in attenuating the effects of smoking on lung development, this study has a number of limitations. First, vitamin C did not prevent all the effects of prenatal nicotine exposure. Vitamin C supplementation had no effect on the nicotine-induced effects on body weight and chromogranin A (Tables 2 and 4). In addition, as reported by Slotkin and colleagues (74), vitamin C supplementation did not prevent the multiple effects of prenatal nicotine exposure on brain development, and even potentiated the effects of nicotine in some brain regions. Slotkin and colleagues have previously demonstrated multiple adverse effects of prenatal nicotine exposure on mouse brain development (75, 76), and have shown that maternal tobacco smoke exposure similarly affects brain development in rhesus monkeys (77). That these adverse effects of nicotine on brain development also occur in humans is shown by the multiple reports of detrimental effects of smoking during pregnancy on offspring behavior (78-80). Furthermore, although vitamin C may prevent some of the effects of nicotine on lung function, it may not affect other toxic mechanisms by which maternal smoking alters lung development, such as carbon monoxide or polycyclic hydrocarbons. It is also important to note that this study lacked a vitamin C-only control group; thus, whether vitamin C alone has effects on pulmonary function of offspring cannot be determined, so the specificity of the interaction of vitamin C with nicotine on pulmonary function can only be inferred. Finally, the safety of excess vitamin C supplementation during pregnancy must also be considered. This study was conducted with only a single high dose of vitamin C and clearly the efficacy of lower doses needs to be studied. Although high doses of vitamin C have been used in pregnancy to treat pre-eclampsia and other conditions and appear safe (53-55), the specific safety and efficacy of the appropriate dosage must be determined. However, if further animal studies establish its safety and efficacy, vitamin C supplementation may have the potential to be beneficial for highly nicotine-addicted pregnant women and as an adjunct to nicotine replacement therapy during pregnancy. Thus, our findings suggest that vitamin C supplementation has the potential to limit the deleterious effects of maternal smoking during pregnancy on offspring's lung function, although further studies are needed to confirm this effect.

Conflict of Interest Statement: B.J.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; H.S.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; J.A.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; S.L.L. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; Y.J. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; W.M.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; J.A.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; B.C.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; E.R.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

Acknowledgment: The authors thank Neal Benowitz for performing the nicotine and cotinine assays; Kevin Grove for assistance with the leptin assays and brain dissection; Michael Gravett for assistance with cytokine assays; John Fanton, Darla Jacobs, and the Division of Animal Resources for assistance with rhesus monkeys; and Gary Sexton, director of the Biostatistics Core of the Oregon Alzheimer's Disease Center, for assistance with statistical analysis.