|Home | About | Journals | Submit | Contact Us | Français|
Intermittent hypoxia can evoke persistent increases in ventilation (ν̇ E) in neonates (i.e. long-term facilitation, LTF) (Julien et al. Am J Physiol Regul Integr Comp Physiol 294: R1356–R1366, 2008). Since prenatal nicotine (PN) exposure alters neonatal respiratory control (Fregosi & Pilarski. Respir. Physiol. Neurobiol. 164: 80–86, 2008), we hypothesized that PN would influence LTF of ventilation (ν̇ E) in neonatal rats. An osmotic minipump delivered nicotine (6 mg/kg/day) or saline to pregnant dams. ν̇ E was assessed in unanesthetized pups via whole body plethysmography at post-natal (P) days 9–11 or 15–17 during baseline (BL, 21% O2), hypoxia (10 × 5 min, 5% O2) and 30 min post-hypoxia. PN pups had reduced BL ν̇ E (p<0.05) but greater increases in ν̇ E during hypoxia (p<0.05). Post-hypoxia ν̇ E (i.e. LTF) showed an age × treatment interaction (p<0.01) with similar values at P9-11 but enhanced LTF in saline (30±8 %BL) vs. PN pups (6±5 %BL; p=0.01) at P15-17. We conclude that the post-natal developmental time course of hypoxia-induced LTF is influenced by PN.
Considerable evidence indicates that prenatal nicotine (PN) exposure alters the normal developmental progression of respiratory control mechanisms (reviewed in Hafström et al. 2005; Fregosi and Pilarski 2008; Mahliere et al. 2008). For example, both the eupneic pattern of breathing (Huang et al. 2004) and the ability to increase minute ventilation (ν̇ E) during hypoxia (Hafström et al. 2005; Mahliere et al. 2008) are altered in neonatal rats following PN. Recently, it has been shown that intermittent hypoxia exposures (i.e. intermittent hypoxia) can induce persistent increases in respiratory motor output in neonates (i.e. long term facilitation or LTF; Mckay et al. 2004; Tadjalli et al. 2008; Julien et al. 2008). While its physiological significance is not precisely known, recent studies indicate that LTF in humans is associated with reductions in upper airway airflow resistance (Pierchala et al. 2008). Accordingly, prolonged increases in upper airway muscle activity following intermittent hypoxia may be one important functional aspect of LTF (Harris et al. 2006). Studies in anesthetized adult animals indicate that activation of serotonin (5-hydroxytryptamine or 5-HT) receptors is both necessary and sufficient to induce LTF of respiratory motor output (Baker-Herman et al. 2004). Because PN alters the post-natal development of serotonergic neurons (Slotkin et al. 2007), we examined the impact of PN on LTF in neonatal rats. We hypothesized that PN exposure would attenuate LTF of ν̇ E following intermittent hypoxia in unanesthetized neonatal rats.
Pregnant female Sprague-Dawley rats were obtained from Charles River Laboratories (Wilmington, MA, USA). Alzet osmotic minipumps (2ML4, Alzet, Cupertino, CA, USA) containing nicotine or saline (see below) were implanted in pregnant dams on gestational day 5. All pups were then studied on P9-11 or 15-17 at baseline conditions (21% O2) and many were also studied during and following intermittent hypoxia (details provided below). At baseline conditions the sample sizes were as follows: P9-11, N=27 saline, N=26 nicotine; P15-17, N=28 saline, N=26 nicotine. During and following hypoxia, the total number of pups were: P9-11, N=16 saline, N=13 nicotine; P15-17, N=14 saline, N=16 nicotine.. Experiments were approved by the Institutional Animal Care and Use Committee at the University of Florida.
These procedures were based on prior reports (Huang et al. 2004; Luo et al. 2004). Pumps were filled with saline or nicotine hydrogen tartrate (6 mg/Kg/day, Sigma Chemical) and released fluid at ~2.5 μl/hr. Dams were briefly anesthetized (2–3% isoflurane) to a surgical plane and an incision was made over the scapulae to enable subcutaneous insertion of the pump. The incision was closed with wound clips. An analgesic (buprenorphine, 0.03 mg/kg, s.q.) and an anti-inflammatory drug (carprofen, 5.0 mg/kg, s.q.) were given every 12 hrs for 2 days.
Nicotine was measured in a subset of rats (n=2) to confirm that the delivery system was working. Dams were sacrificed at one or two weeks following implantation of an osmotic minipump, and plasma was collected and analyzed by HPLC using methods previously described by Gosheh et al (2000).
A whole-body plethysmography system (Buxco Inc., Wilmington, NC, USA) was used to study the pattern of breathing. The system was calibrated by injecting known gas volumes into a ~400 ml Plexiglas recording chamber using a 1 ml syringe. The chamber pressure, temperature and humidity, rat rectal temperature, and atmospheric pressure were used to calculate respiratory volumes including ν̇ E (mL/min) and tidal volume (VT, mL/breath) (Drorbaugh and Fenn, 1955). Calibration experiments indicated that the accuracy of volume measurements was ±10% over the range of 0.05 to 1.0 ml. In this study, however, our primary objective was to quantify relative changes in respiratory volumes (i.e. within an experimental session) in each animal. Overall breathing frequency (fR, min−1) and the duration of inspiration (TI) and expiration (TE) were calculated from the airflow traces. Gas mixtures flowed through the chamber at 1 L/min. Baseline recordings lasted ~ 30 min while the chamber was flushed with 21% O2 (balance N2). Rats were then exposed to 10 hypoxic episodes (5 min, 5% O2, balance N2) separated by baseline conditions. It took ~ 30 sec for complete equilibration of the chambers after switching gases. The relatively severe hypoxic challenge (5% O2) was chosen because both the relative magnitude and time course of LTF in neonates has been reported to correlate with the degree of hypoxia (Tadjalli et al. 2007). Ten hypoxic episodes were used because the probability of evoking ν̇ E LTF increases with greater number of hypoxic episodes (McGuire and Ling 2005). Chamber temperature was maintained at 34°C using an incubator.
Plethysmography data were analyzed over a stable 10-min period prior to hypoxia (i.e. baseline), over the last min of the first hypoxic challenge, and at 25–30 min post-intermittent hypoxia. A two-way analyses of variance (ANOVA) was used to compare outcome measures between groups: factor 1 = age (P9-11 or P15-17), factor 2 = treatment (nicotine or saline). To enable comparisons across groups, the hypoxia and LTF data were expressed relative to baseline values (Simakajornboon et al., 2004; Baker-Herman et al. 2004). Male and female pups were grouped together as in prior reports (Huang et al. 2004; Simakajornboon et al., 2004; Luo et al. 2004, 2007). Data are presented as the mean ± SEM; statistical significance was assumed when P<0.05.
Body weight was not significantly different between groups (Table 1). However, there was a tendency for PN pups to be slightly heavier (2 way ANOVA, treatment effect P=0.094), and accordingly we expressed respiratory volume data (ν̇ E, VT) relative to body weight. It was previously reported that PN pups are heavier than saline controls at day P21 (Mahlière et al. 2008). To confirm that the osmotic minipumps were functional, plasma nicotine and its metabolite cotinine were measured in a small sample (n=2) of dams. At one week following pump implantation plasma nicotine was 28 ng/ml and cotinine was 182 ng/ml; at two weeks nicotine was 24 ng/mL and cotinine was 170 ng/ml.
Representative airflow traces are provided in Fig. 1. Assessment of baseline fR revealed a significant age × treatment interaction (p=0.042; Fig. 2A). Similar age × treatment interactions were also observed for baseline TI (p=0.026) and TE (p=0.003) (Table 1). Accordingly, the normal developmental changes in the pattern of breathing are influenced by PN (Fig. 2A). Significant treatment effects were present for both baseline VT (p=0.044) and ν̇ E (p=0.045). Specifically, PN pups had reduced VT and ν̇ E relative to age matched saline controls (Fig. 2A).
The hypoxic ventilatory response (HVR) (i.e. ν̇ E expressed as % baseline) was greater in PN vs. saline pups (p=0.046) which is consistent with at least two prior reports (Fregosi et al. 2006; Mahliere et al. 2008). The increase in fR during hypoxia was indistinguishable between the two treatment groups (p=0.920, Fig. 2B) suggesting that the enhanced ν̇ E was driven by greater VT in PN pups. However, differences in hypoxic VT did not reach statistical significance as evaluated with 2 way ANOVA (p=0.103, Fig. 2B). TI was not substantially altered by hypoxia in P9-11 pups (saline: 105±3; nicotine: 107±3 % baseline). In contrast, at P15-17 hypoxia reduced TI to 82±3 % baseline in both groups (age effect: p<0.001). In P9-11 pups, hypoxic TE was 81±3 (PN) and 83±3 % baseline (saline). These values were reduced to 58±3 (PN) and 59±3% (saline) at P15-17 (age effect: p<0.001). No treatment effects were found for hypoxia TI (p=0.649) or TE (p=0.607).
ν̇ E was assessed at 30 min post-intermittent hypoxia to evaluate LTF. A significant age × treatment interaction (p=0.005) was observed for ν̇ E (Fig. 2C). Post-hoc tests indicate that LTF (i.e. ν̇ E, % baseline) was not different between saline and PN pups at P9-11 (p=0.204), but was greater in saline pups at P15-17 (p=0.011; Fig. 2C). An age × treatment interaction was also observed for VT LTF (p=0.020) indicating that ν̇ E LTF in P15-17 saline pups was driven by primarily by VT (Fig. 2C). Indeed, VT (% baseline) tended to be greater saline vs. PN pups at P15-17 (p=0.051). At 30 min post-hypoxia, fR was not different between PN and saline pups (p=0.277) and the age × treatment interaction did not reach significance (p=0.078). Consistent with this, TI (p=0.155) and TE (p=0.709) were not different between PN and saline pups at 30 min post-hypoxia.
Additional experiments were conducted in which pups remained in the plethysmograph chamber for a similar length of time but without exposure to intermittent hypoxia (i.e. a “time control”). In saline pups, ν̇ E was 103±5 (P9-P11, N=6) and 98±6 % baseline (P15-17, N=7) at the end of the time control trial. PN nicotine pups had similar values of 102±5 (P9-P11, N=6) and 92±7 % baseline (P15-P17, N=7). These time control data indicate that the ν̇ E LTF observed in P15-P17 saline pups (Fig. 2C) was not the result of the extended period in the plethysmograph recording chamber.
These data are consistent with earlier work showing that PN can influence the normal developmental progression of breathing pattern in neonatal animals. We observed that PN exposure attenuates ν̇ E during eupnea but enhances the ν̇ E response to a 5% O2 challenge. In addition, we report for the first time that intermittent hypoxia-induced LTF of ν̇ E is blunted after PN in P15-17 rat pups.
There is variability across prior studies regarding how PN alters the pattern of breathing during the post-natal period (Bamford et al. 1996; St. John and Leiter, 1999; Huang et al. 2004; Simakajornboon et al., 2004; Mahlière et al. 2008; Pendelbury et al. 2008). Over the age ranges used in our study, there are reports of unchanged (Simakajornboon et al., 2004), reduced (Huang et al. 2004) and finally increased (Mahlière et al. 2008) ν̇ E during normoxic or “eupneic” conditions following PN. The present data, however, are consistent with the findings of Huang et al. (2004) who observed reduced VT (p<0.05) and a tendency for reduced ν̇ E (p=0.067) in PN pups at P14 and P18 during eupnea. In contrast, Mahlière et al. (2008) report that baseline ν̇ E was increased in both P7 and P11 PN pups reflecting greater VT. A commentary regarding these apparently conflicting results is provided below (section 4.2).
The observation that the HVR was enhanced in PN pups could theoretically reflect an adaptation to an impaired metabolic response to hypoxia. For example, early neonatal rats typically achieve hyperventilation via a reduction in metabolic rate (Mortola 2001). This effect is most pronounced in very young pups (e.g. P2) and as pups age they begin to hyperventilate during hypoxia by increasing ventilation with progressively lower changes in metabolism (Mortola 2001). We did not measure metabolic rate in the current experiments, and this should be an important component of future studies. However, preliminary evidence indicates that PN does not impact overall metabolic rate during hypoxia (R.F. Fregosi, personal communication).
The elevated HVR following PN occurred due to an increased VT (Fig. 2). Previously, Mahliere et al. (2008) also reported that PN pups (P11) had an increased HVR (10% O2 challenge, 10 min duration) associated with an increased VT. Similarly, Fregosi and colleagues (R.F. Fregosi et al., The FASEB Journal. 20:A1213, 2006) found an increased HVR at P12 and P18 in PN pups (10% O2, 3 min), and this response also reflected an increased VT. To the contrary, two studies have reported no changes in the HVR in similarly aged pups following PN (Bamford et al. 1996; Simakajornboon et al. (2004). Bamford et al. (1996) assessed ν̇ E in the 11th min of a 10% O2 challenge and observed similar HVR in PN and control pups at P19. It should be noted, however, that in Bamford’s study an increase in the breathing frequency response to hypoxia (P<0.001) was observed in slightly younger (P8) PN pups. Simakajornboon and colleagues found change in the HVR at P10, P 15, and P 20 following PN. However, similar to Bamford et al. (1996), some differences were noted in younger pups as PN was associated with a decreased HVR and attenuated hypoxic ventilatory roll-off at P5.
Differences across published reports likely do not reflect nicotine dosage as each of the studies reviewed here report a dose of 6 mg/kg/day. We speculate that differences in genetics (Fuller et al. 2001), temperature (Pendelbury et al. 2008) or possibly the nicotine delivery method could contribute to the conflicting results. For example, respiratory-related plasticity can differ between rat substrains (e.g. Sprague-Dawley rats originating from different breeding colonies) (Fuller et al. 2001). In regards to temperature, Pendlebury and colleagues (2008) found that the ambient temperature influenced breathing more profoundly in P7 rats exposed to prenatal cigarette smoke vs. control animals. Finally, studies producing apparently conflicting results have used surgically implanted pellets to release nicotine (Mahlière et al. 2008) vs. osmotic minipumps (Huang et al. 2004; current study). Thus the method of nicotine delivery may ultimately influence the development of plasticity. In any case, despite the conflicting results regarding the manner in which breathing pattern changes after PN across laboratories and experimental preparations, there is a growing consensus that PN alters the normal development of respiratory control mechanisms (Hafström et al. 2005; Fregosi & Pilarski 2008).
Respiratory LTF has been observed in anesthetized P3-P7 rat pups (Mckay et al. 2004), in situ heart-brainstems from P15-P25 rats (Tadjalli et al. 2008), and unanesthetized P10 pups (Julien et al. 2008). In the latter study robust ν̇ E LTF was induced by 10 episodes of hypoxia (10% O2) and was driven by increased VT with little change in fR. Since the ν̇ E LTF observed by Julien et al. was not statistically significant until ~50 min post-hypoxia, their finding does not contradict our observation that ν̇ E LTF was absent at 30 min post-hypoxia in P9-P11 rats. In this paper we extend these earlier observations by reporting that PN blunts LTF in P15-17 pups. In our preparation, LTF was not apparent in the P9-11 pups of either treatment group.
These studies were not intended to address the mechanism(s) responsible for the impact of PN on changes in breathing during or following hypoxia, but possibilities include changes in the ability of nicotinic acetylcholine receptors to modulate the pre-synaptic release of GABA and glutamate (Fregosi and Pilarski 2008) as well as changes in serotonergic function (Slotkin et al. 2007). A relationship between altered LTF and serotonergic function after PN is an attractive hypothesis because LTF is serotonin-dependent (at least in adults; Baker-Herman et al. 2004), and PN can lead to persistent alterations in the serotonergic nervous system (Slotkin et al. 2007). Another intriguing hypothesis is that PN may affect LTF expression via changes in the neuropeptide orexin. Recent work has implicated orexin in the expression of respiratory LTF. For example, LTF of both ventilation (Terada et al. 2008) and phrenic activity (Toyama et al. 2009) is blunted in orexin knockout mice. Nicotine can activate orexinergic neurons (Pasumarthi et al. 2006; Pasumarthi and Fadel 2008), and orexinergic neurons in turn send dense projections to brainstem serotonergic neurons (Baldo et al. 2003; Peyron et al. 1998). In addition, orexin release, can be regulated by serotonergic feedback (Muraki et al. 2004).
Support for this work was provided by a grant from the James & Esther King Biomedical Research Program. Support was also provided by NIH grant T32HD043730T32 (BJD). The authors thank Dr. Ralph F. Fregosi (University of Arizona) for help with the osmotic pumps and advice regarding the data. We also thank Ms. Heather Carr for help with data collection and analyses.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.