INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: nicotine; smoking; pulmonary function; pregnancy; lung

Despite the compelling evidence that smoking increases the risk of spontaneous abortion, preterm delivery, low birth weight, and prenatal morbidity and mortality (1, 2), a significant number of women continue to smoke during pregnancy (3, 4). In addition, multiple epidemiological studies that have followed children from infancy through childhood have shown significantly increased incidences of bronchitis and hospital admissions for lower respiratory illness in offspring of mothers who smoked during pregnancy (5). These studies further suggest that it is prenatal, at least as much as postnatal smoke exposure, that correlates with the increased lower respiratory illnesses (5, 6). Along with increased respiratory illness, studies have demonstrated altered pulmonary mechanics in infants and children exposed to smoke in utero. Infants of mothers who smoked during pregnancy demonstrate decreased lung compliance (9- 11), decreased expiratory flow rates (3, 5, 12), decreases in the ratio of time to peak tidal expiratory flow to total expiratory time (T_PTEF:TE) (10, 13, 14), and increased airway resistance (11, 15) compared with infants of nonsmoking mothers. Strikingly the strong correlation between increased lower respiratory illness and compromised lung function in offspring of mothers who smoked during gestation persists well into adolescence (12, 16) and even to adulthood (17).

Although it is clear that prenatal smoke exposure adversely affects lung development, the mechanisms underlying those changes remain to be determined. Recently, our laboratory has demonstrated abundant expression of nicotinic acetylcholine receptors (nAChR) in nonneuronal cells in fetal monkey lung (18). Nicotinic acetylcholine receptors are pentameric ligand-gated ion channels. To date a total of 14 different subunits (1-4, 1-10) have been described and subunit compositions determine the receptor sensitivity to nicotine (19, 20). For example, nAChR composed of three 4 subunits and two 2 subunits (42-nAChR) are activated by low levels of nicotine, receptors composed of five 7 subunits (7-nAChR) are inactivated by very low levels of nicotine, and receptors composed of 3, 5, and 2 receptors are relatively insensitive to nicotine (19, 20). The abundant expression of nAChR in developing lung and the sensitivity of these receptors to nicotine make it likely that the interaction of nicotine with nicotinic receptors in developing lung underlies many of the effects of maternal smoking on lung development.

Consistent with this hypothesis, we have shown that in monkeys, prenatal nicotine exposure from Day 26 to 134 of gestation (term is 165 d) alters lung development and is associated with enlargement of airspaces and reduction in alveolar surface area (18). In addition, prenatal nicotine exposure markedly upregulates 7-nAChR expression in fetal monkey lung particularly in the walls of conducting airways and vessels (18). This suggests that interplay between nicotine and 7-nAChR in airway walls leads to altered airway morphology and pulmonary mechanics. Consistent with this hypothesis, Elliot and coworkers (21) reported that airway wall thickness was increased in infants who died of sudden infant death syndrome (SIDS) and whose mothers smoked during gestation. These findings suggest that in women who smoke during pregnancy, nicotine crosses the placenta to interact with nicotinic receptors in developing lung to cause changes in lung structure and subsequently influence lung function in offspring.

Although human studies clearly depict the consequences of prenatal smoke exposure on postnatal lung mechanics, the role of nicotine in cigarette smoke in causing these changes has yet to be determined. In this study, we present evidence that prenatal nicotine exposure produces alterations in lung function parallel to those observed in infants of mothers who smoked during pregnancy. Understanding the mechanism by which prenatal smoke exposure alters lung function may lead to therapeutic interventions to block those effects as well as help to further discourage smoking during pregnancy.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental Design

Cycling female Rhesus monkeys were mated for 3 d and on Days 23- 25 of mating, ultrasound examinations were performed to identify pregnancy. Pregnant females were randomly allocated to a control (n = 7) or nicotine treatment (n = 7) group. On Day 26 of pregnancy, control monkeys received subcutaneous (mid-scapular region) Alzet mini osmotic 2ML4 pumps containing bacteriostatic water (Abbott Laboratories, North Chicago, IL) and the treatment monkeys received pumps containing nicotine bitartrate (Sigma, St. Louis, MO) in bacteriostatic water to deliver 1.5 mg nicotine/kg body weight/d. Pumps were replaced every 3 wk. On Day 118 and 139 of gestation, maternal blood samples were drawn, and amniocentesis was performed in both control and nicotine-treated animals to obtain amniotic fluid samples for determination of nicotine level. Animals received Cefazolin (150 mg/twice a day) for 3 d after pump insertion and/or amniocentesis. Nicotine and cotinine levels in amniotic fluid were measured by gas chromatography-mass spectrometry as previously described by Benowitz and coworkers (22).

To ensure all offspring were tested at the same gestational age, fetuses were delivered on gestation Day 160 (term = 165 d) 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.

Pulmonary Function Maneuvers

Pulmonary function testing was performed 1 d after birth (approximately 24-26 h). Animals were anesthetized with ketamine (10 mg/kg body weight), tracheostomized, and a polypropylene cannula (2 mm i.d.) was placed and secured. A water-filled polypropylene catheter with multiple holes at the distal end was inserted into the esophagus for pulmonary pressure measurements. Pulmonary function tests were performed using a Buxco apparatus and BioSystem for Maneuvers software (Buxco Electronics Inc., Sharon, CT) for invasive pulmonary maneuvers (23). In this system, a whole body plethysmograph with a built in reference chamber and a heat sink was equipped with a mouth pressure port manifold attached to a mouth pressure transducer, slow inspiration and expiration valves, and a functional residual capacity valve. The manifold was connected to a pressurized air source for fast expiration maneuvers. The transpulmonary pressure was measured using a liquid-coupled Cobe pressure transducer connected to an esophageal port and changes in flow were recorded using a flow transducer. High and low flow, and mouth and pulmonary pressure signals were calibrated, and calibration levels applied were kept within 5 and +5 V range and an effective range difference of less than 5% for calibrations was accepted.

Animals were placed in the whole body plethysmograph. The tracheal cannula was connected to a mouth pressure port and the esophageal catheter was connected to a pulmonary pressure port. Using the standard Buxco maneuvers protocol, tidal breathing, quasistatic pressure volume, and fast flow volume maneuvers were performed, and computer-generated data were used for analysis. From the tidal breathing maneuver, tidal volume, peak tidal expiratory flow (PTEF), pulmonary resistance, and dynamic compliance were determined. Using the fast flow maneuver (lungs inflated at +25 cm and then deflated at 25 cm H₂O pressure), inspiratory capacity (IC), expiratory residual volume (ERV), forced expiratory volume in 100 ms, 200 ms, and 400 ms (FEV_0.1, FEV_0.2, and FEV_0.4), peak expiratory flow (PEF), mid-expiratory flow (FEF_25%-75%), forced expiratory flow (FEF) after 25%, 50%, 75%, and 90% of forced vital capacity has expired, FEV_PEF (forced expiratory volume at peak expiratory flow), and Te-FEF (time for forced expiratory flow phase) were determined. Values of quasistatic chord lung compliance between 0 and 10 cm H₂O (Cst-chord) and static compliance at 50% forced vital capacity (Cst-FVC_50%) obtained using quasistatic pressure volume maneuver were used for analyses.

After pulmonary function maneuvers, animals were euthanized with pentobarbital. The abdominal aorta was transected, the lungs flushed with normal saline to minimize the contribution of residual blood volume, and lungs dissected out, blotted, and weighed. For morphometric analysis, the left lung with intact trachea was fixed with 4% zinc formalin at 20 cm constant transpulmonary water pressure for 72 h. Left lung volume was measured by water displacement and total lung volume was extrapolated using total lung weight.

Statistical Analysis

Means, standard errors of means, and 95% confidence intervals were calculated using Number Cruncher Statistical System ver.2000 (Iowa, CA). Differences between the means of the control and treatment groups were determined by t test.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Amniotic fluid nicotine and cotinine levels were measured on Days 119, 140, and 160 of gestation. Nicotine levels were 8.8 ng/ml ± 2.3 on Day 119, 17.7 ng/ml ± 8.7 on Day 140, and 13.8 ng/ml ± 4.0 on Day 160 of gestation; cotinine levels were 67.7 ng/ml = 13.9, 56.9 ng/ml ± 8.5 and 74.4 ng/ml ± 8.7, respectively. These values are similar to the range of those observed in amniotic fluid of pregnant human smokers (24, 25). Nicotine and cotinine were undetectable in the amniotic fluids from control animals. No spontaneous abortion or fetal resorption occurred during treatment in nicotine or control mothers. Maternal food intake (by observation) and maternal weight gain of the nicotine-treated group did not differ from control animals. Nicotine treatment during pregnancy did not significantly affect fetal somatic growth as body weight and body length of nicotine-treated neonates were only slightly lower than control animals (Table 1).

Lung Growth

In utero nicotine exposure diminished fetal lung growth. Absolute and specific (per 100 g body weight) lung weight was significantly lower (16% and 14%, respectively) in the nicotine-exposed group compared with the control group (Table 1). Similarly, fixed lung volume and lung volume normalized to body weight were also significantly decreased (14%, p = 0.001 and 11% p = 0.006, respectively) in the nicotine group (Table 1). These findings are in concordance with those observed in our previous studies and suggest that prenatal nicotine exposure decreases lung growth.

Pulmonary Function Tests

Nicotine exposure during gestation altered pulmonary functional capacity and mechanics of newborn monkeys. Neonates exposed to nicotine during gestation had significantly lower FEV_0.2 (23%) than saline control animals but decreases in FEV_0.2/FVC ratio (11%) and FEV_0.4 (17%) were not significant (Table 2). Forced vital capacity (FVC), expiratory reserve volume (ERV), and tidal volume (VT) were reduced (12%, 28%, and 8%, respectively) following prenatal nicotine treatment, but these changes did not reach a significant level (Table 2).

Peak tidal expiratory flow (PTEF) (Table 2) and mid-expiratory flow (FEF_25%-75%) (Table 3) diminished significantly (17% and 29%, respectively, p < 0.05) in neonates exposed to nicotine during gestation. The values of peak forced expiratory flow (PFEF), forced expiratory flow after 25% of FVC expired (FEF_25%), FEF_50%, FEF_75%, and FEF_90% were also lower in newborns exposed to nicotine in utero (21%, 24%, 27%, 24%, and 10%, respectively), but these changes were not significant (Table 3). These alterations indicated that prenatal nicotine exposure may have induced morphometic changes in conducting airways. Interestingly, forced expiratory volume at peak expiratory flow (FEV_PEF) and FEV_PEF/FVC% were significantly diminished following nicotine treatment suggesting that increased recoil at higher volume may have achieved peak flow with smaller expiratory volume (Table 3).

Pulmonary resistance and specific (corrected for lung volume) pulmonary resistance was significantly increased (132% and 100%, respectively) in in utero nicotine-treated neonatal monkeys compared with those of saline control animals (Figure 1). Conversely, dynamic lung compliance, static chord compliance, and static compliance at VC_50% decreased (14%, 15%, and 25%, respectively) in neonates that were exposed to nicotine during pregnancy but not significantly (Figure 2). These findings show that nicotine exposure during gestation leads to altered pulmonary function at birth.

View larger version (24K): [in this window] [in a new window]   Figure 1.   Changes in pulmonary resistance after prenatal nicotine exposure. (A) Pulmonary resistance (95% CI; control 0.118-0.272 versus nicotine 0.17-0.73) and (B) specific pulmonary resistance (calculated as pulmonary resistance × lung volume) (95% CI; control 0.38-0.84 versus nicotine 0.40-2.06) significantly increased in neonatal monkeys that were exposed to nicotine during gestation. Data shown as mean ± SEM; *p < 0.05.

View larger version (19K): [in this window] [in a new window]   Figure 2.   Prenatal nicotine effects on (A) dynamic lung compliance (95% CI; control 1.37- 5.24 versus nicotine 1.15-4.52), (B) specific dynamic lung compliance (calculated as dynamic compliance/lung volume) (95% CI; control 0.05-0.16 versus nicotine 0.04-0.17), and (C) quasistatic chord lung compliance (Cst-chord) between 0-10 cm H2O and static compliance (Cst-FVC50%) at 50% forced vital capacity (95% CI; control 0.16-1.01 versus nicotine 0.14- 0.73). After in utero nicotine exposure dynamic lung compliance, static chord compliance and static lung compliance at FVC50% decreased but changes were not significant. Data shown as mean ± SEM.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Multiple clinical investigations have reported that prenatal in addition to postnatal smoke exposure is associated with increased frequency of lower respiratory illnesses and hospitalizations (5). This increase in respiratory illnesses correlates with impaired pulmonary function in infants and children whose mothers smoked during pregnancy (3, 5, 9). The exact alterations in lung structure that lead to this impaired function and the mechanism underlying smoking-induced alterations in lung development are unknown. Mechanistic hypotheses have tended to focus more on blood flow, oxygen, and toxins than on the specific effects of nicotine itself. However, the recent descriptions of extensive expression of nicotinic receptors in nonneuronal cells throughout peripheral organs (26) provide a possible basis to explain the effects of smoking on organ development at the molecular level of ligand-receptor interaction. Indeed, although the effects of nicotine on brain function have been extensively studied (19, 20), until recently, little attention has been focused on implications of peripheral expression of nicotinic receptors in nonneuronal cells. In our previous study (18), we reported the presence of nicotinic acetylcholine receptors in multiple cell populations in fetal monkey lung, and the inhibition of fetal lung development with associated morphometric changes in lung following prenatal nicotine exposure. These findings suggested that prenatal nicotine-induced structural changes in lung might be reflected by changes in neonatal lung function similar to those observed in infants of mothers who smoke during pregnancy. Now, we report that prenatal nicotine exposure in monkeys produces changes in lung mechanics highly similar to those reported in offspring of mothers who smoked during pregnancy. The pattern of monkey fetal lung development and growth is similar to that of humans and the duration of pregnancy in monkeys is long enough to simulate the effects of prenatal nicotine exposure in humans, so these findings are likely relevant to humans.

Timed-pregnant monkeys were treated subcutaneously with 1.5 mg/kg/day nicotine. This dose was chosen because it is similar to the dose of nicotine taken by heavy human smokers and prior studies indicated this would achieve amniotic fluid nicotine levels similar to those seen in pregnant human smokers (18, 24, 25). Average levels of nicotine and cotinine in amniotic fluid measured at 119, 140, and 160 d gestation were 13.2 ± 3 and 66.9 ± 6 ng/ml, respectively. These levels are very similar to those measured in human amniotic fluid levels, which range between 3.3 and 28 ng/ml (24) for nicotine and 93 and 109 ng/ml for cotinine (25). Thus the nicotine dose used for these experiments is appropriate to model effects of maternal smoking during pregnancy. Administering nicotine by minipump has been well characterized by Stlotkin and coworkers (30, 31) and avoids the extreme highs and lows caused by bolus injections. There was no incidence of fetal resorption or spontaneous abortion, and the observed food intake and maternal body weight gain in the nicotine-treated group did not differ from those of the control group. These findings indicated that although nicotine easily crossed the fetoplacental barrier it did not have major adverse effects on maternal health or pregnancy.

In the present study, in utero nicotine treatment did not have significant effects on somatic growth during fetal development as fetal body weight, fetal body weight corrected for maternal weight, and crown-rump length of nicotine-treated newborn monkeys were only slightly and not significantly decreased (3%, 12%, and 4%, respectively) compared with those of control animals. These findings are, however, consistent with those we have reported in our previous study (18) and are consistent with the magnitude of changes seen in offspring of human smokers (32, 33). Typically in human studies there is a 5% decrease in average birth weight and a 1.5% decrease in crown-heel length (32, 33) and the large N of epidemiological studies allows significance to be achieved. Thus although the decreases in weight and length observed here are consistent with those observed in studies of human smoker offspring, samples size prevents conclusions as to the relative role of nicotine versus other factors in cigarette smoke in causing fetal growth retardation.

Although the effect of prenatal nicotine exposure on overall somatic growth of fetal monkeys was minimal, lung development was significantly impaired. Absolute lung weight, lung weight normalized to body weight, and fixed lung volume were all significantly lower with prenatal nicotine exposure. These findings are in concert with those reported for rats exposed to cigarette smoke during pregnancy (34) and for prematurely delivered fetal monkeys exposed to nicotine during gestation (18). The vital capacity, expiratory reserve volume, and tidal volume in nicotine-exposed newborn monkeys were also considerably lower (12%, 28%, and 8%, respectively) compared with control animals, but differences did not reach significance. The precise mechanisms by which nicotine influences fetal lung growth remain to be determined, but likely involve nicotine's passage across the placenta to interact with nicotinic acetylcholine receptors on pulmonary cells to modify cell proliferation. Consistent with this hypothesis, in organ culture of fetal monkey lung, nicotine inhibits cell division of airway epithelial cells and blocks their mitogenic response to gastrin- releasing peptide (35). It is likely that the effect of nicotine exposure on organ development is tissue specific depending on the location and subtypes of nicotine receptors expressed.

In the present study, nicotine exposure during gestation significantly reduced peak tidal expiratory flow and mid-expiratory flow in the newborn. Peak forced expiratory flow (PFEF) and forced expiratory flow (FEF) at 25%, 50%, 75%, and 90% of expired FVC were also lower in the nicotine-exposed newborn monkeys compared with control animals (Tables 2 and 3). These findings are strikingly similar to human studies in which multiple investigators have observed a reduction in maximum flow at functional residual capacity in infants of mothers who smoked during gestation (11, 36), though Tager and coworkers observed significant reductions only in female offspring (5). In humans these decreases in expiratory air flows persist into adolescence (39). How long the nicotine-induced changes will last in the monkey model remains to be determined. Strikingly, Upton and coworkers (17) found that the decreases in expiratory flows persisted into adulthood suggesting that the anatomic changes produced by maternal smoking that lead to altered flows are permanent. These alterations in airway flow may result from a myriad of changes in morphometric dimensions of airways, geometric airway relationship with parenchyma, airway morphogenesis, parenchymal development, connective tissue equilibrium, and/or surfactant protein tension forces. Future studies of the animal model used here should help determine the exact nature of the responsible anatomic changes.

The forced expiratory volume to achieve peak expiratory flow (FEV_PEF) and FEV_PEF/FVC% declined significantly and the time to complete the expiratory phase increased in nicotine-treated neonates. These findings suggest that peak expiratory flow was achieved with smaller expiratory volume. This may have occurred because of increased elastic recoil at higher lung volume with reduction in airway caliber and increased pulmonary resistance, thus extending the time of expiratory phase. Similarly, human infants of mothers who smoke during gestation show a decreased ratio of time to peak tidal expiratory flow to total expiratory time (T_PTEF:TE) (10, 13, 40).

Pulmonary resistance was increased in nicotine-exposed newborn monkeys (Figure 1). Pulmonary resistance corrected for lung volume and indirectly for body size is considered an indicator of airway dimensions (41, 42). Because the total lung volume of the nicotine-treated group was lower, it is important that pulmonary resistance normalized to the lung volume also remained significantly higher in the nicotine-treated group. The increased pulmonary resistance likely reflects altered airway geometry following the prenatal nicotine treatment. Our findings are in agreement to those of Dezateux and coworkers who found an increase in airway resistance in infants (< 13 wks of age) whose mothers were smokers during pregnancy (15) though Hanrahan and coworkers (43) did not see such an increase. Because neonates are nose breathers and the nose may account for more than 50% of total resistance, nasal resistance can influence interpretation of lung function analysis (44). By tracheostomizing animals prior to testing, contributions of extrabronchial structures (i.e., nasal passages) to total pulmonary resistance measurements were eliminated. It has been shown in monkeys (45) and in humans (46) that early amniocentesis or puncture of membranes can affect fetal lungs. However, as amniocentesis was performed in both nicotine-treated and control groups, it is unlikely that amniocentesis was responsible for the differences in lung function between the two groups. In addition effects of amniocentesis performed during the third (relative) trimester on lung development are likely less than effects of amniocentesis during early gestation.

Although not at the level of significance, dynamic lung compliance, chord compliance at 10 cm H₂O, and compliance at FVC_50% declined in the nicotine-exposed newborns. However, volume-corrected compliance in the nicotine-treated group did not differ from the control group. In human studies, reports documenting changes in compliance in infants of mothers who smoked during pregnancy are conflicting, with some investigators observing a significant decrease in compliance (9) and others not (13, 43). Thus the small decreases in compliance seen here are consistent with the clinical studies.

How then does prenatal nicotine lead to altered pulmonary function at birth? Our previous work in fetuses (18) and that of others in adults (26) shows the presence of nicotinic acetylcholine receptors in airway epithelial cells (7, 42, 352or4 subtypes), air space parenchymal cells (7, 352or4 subtypes), and fibroblast layers surrounding blood vessels and airways (7 subtype). Most abundant is the expression of 7-nAChR in the fibroblast layers (18). With prenatal nicotine exposure, levels of 7 expression increase in the fibroblast layers and airway wall thickness and collagen expression increases in parallel with the increase in 7 expression (18, 47). Collagen accounts for the bulk of extracellular matrix proteins in the lung. Although collagen type I is important for tensile strength and rigidity, collagen type III along with elastin is important for recoil properties in the lung. Increased wall thickness and accumulation of collagen in the airway wall could reduce the caliber of airways and make them less pliable. Thus alterations in the morphometric dimensions or airway compliance may produce greater changes in pulmonary resistance and expiratory flow rate.

In our previous study (18), we also observed that prenatal nicotine exposure increased the volume density of alveolar air while decreasing the volume density of bronchial air. Consequently the dysanaptic index (volume density of bronchial air divided by alveolar air) decreased considerably in the nicotine group (0.061 ± 0.016) compared with the control group (0.095 ± 0.025). These findings suggest that fetal lung development following prenatal nicotine treatment is anatomically dysanaptic (where two components of an organ grow disconcordantly) (48). This could result from inhibition of airway growth or stimulation of alveolar growth. Nicotine has been shown to inhibit airway epithelial cell mitogenesis (35). On the other hand, nicotine may also stimulate alveolar maturation as a number of studies have shown prenatal nicotine exposure stimulates type II proliferation and surfactant expression (18, 49, 50). Further studies are needed to determine the relative importance of these two mechanisms to the alterations in pulmonary mechanics caused by nicotine.

In summary, in utero nicotine exposure adversely affected fetal lung development as reflected by decreased lung weight and lung volume in the nicotine-exposed newborn monkey. In utero nicotine exposure also adversely affected pulmonary function at birth as reflected by decreased expiratory flow rates and increased pulmonary resistance. These findings demonstrate that prenatal nicotine exposure compromises both lung growth and pulmonary mechanics. These are the first studies that demonstrate that prenatal nicotine exposure alters pulmonary function at birth. The pulmonary function changes observed in monkeys exposed to nicotine in utero are strikingly similar to those reported in the literature studying human infants of mothers who smoked during pregnancy. This study suggests that nicotine, transported across the placenta, may be the key constituent of cigarette smoke to impair fetal lung development and lead to altered lung function and increased respiratory illness in offspring. These findings also have implications for the safety of nicotine replacement therapy during pregnancy, though the dose of nicotine delivered by gum or patch is typically less than that from cigarettes (51).

These findings serve to further emphasize the importance of active intervention to discourage women from smoking or using tobacco products during pregnancy. Key questions that remain are the anatomic changes produced by nicotine, the mechanism underlying these changes, and what, if any, is a safe dose and/or dosing regimen of nicotine during pregnancy.