Role of vagal innervation on pulmonary surfactant system duringfetal development
Luxmi Gahlot,1 Francis H. Y. Green,2 Anita Rigaux,1 Jennifer M. Schneider,1 and Shabih U. Hasan1
Departments of 1Pediatrics and 2Pathology and Laboratory Medicine, Institute of Child and Maternal Health, Facultyof Medicine, University of Calgary, Calgary, Alberta, Canada
Submitted 21 May 2008; accepted in final form 17 March 2009
Gahlot L, Green FH, Rigaux A, Schneider JM, Hasan SU.Role of vagal innervation on pulmonary surfactant system duringfetal development. J Appl Physiol 106: 1641–1649, 2009. Firstpublished March 19, 2009; doi:10.1152/japplphysiol.90868.2008.—Vagally mediated afferent feedback and compliant lungs (surfactantsystem) play vital roles in the establishment of adequate alveolarventilation and pulmonary gas exchange at birth. Although the sig-nificance of vagal innervation in the establishment of normal breath-ing patterns is well recognized, the precise role of lung innervation inthe maturation of the surfactant system remains unclear. The specificaim of the present study was to investigate whether vagal denervationcompromises the surfactant system during fetal development. Exper-iments were performed on 12 time-dated fetal sheep: 8 underwentcervical vagal denervation, and 4 were sham operated. Vagal dener-vation was performed at 110–113 days gestation. Fetal lambs wereinstrumented in utero to record arterial pH and blood-gas tensions.The animals were delivered by cesarean section under general anes-thesia between 130 and 133 days gestation (term �147 days). Lungsamples were collected for wet-to-dry ratios, light and electron mi-croscopy, and overall lung morphology. In addition, total proteins,total phospholipids, and surfactant proteins A and B were analyzed inboth lung tissue and bronchoalveolar lavage fluid. Vagal denervationhad no effect on alveolar architecture, including type II cells or themorphology of lamellar bodies within them. Furthermore, surfactantproteins A and B and total phospholipids were similar in lung tissueand bronchoalveolar lavage fluid between the two groups. A signifi-cant correlation was observed between circulating cortisol concentra-tions and surfactant proteins in the bronchoalveolar lavage fluid andlung tissue. We provide definitive evidence that vagal innervation atmidgestation is not required for maturation of the pulmonary surfac-tant system during fetal development.
ONE OF THE MOST VITAL AND unique adaptations that occursimmediately after birth is the establishment of continuousbreathing (19) and adequate pulmonary gas exchange, whichrequires vagally mediated volume feedback and compliantlungs (surfactant system). We and others have shown thatvagal neural input is necessary for the establishment of alveolarventilation and pulmonary gas exchange shortly after birth andduring early postnatal life (15, 30, 32). Pulmonary surfactant,a developmentally regulated complex lipoprotein mixture, issynthesized by type II pulmonary epithelial cells. The surfac-tant is tightly packed in the lamellar bodies (120–180 per typeII cell) from where it is secreted as tubular myelin to form afilm at the alveolar air-liquid interface (3, 55). Lung surfactantplays a critical role in the transition from the fluid-filled fetal
lungs to the aerated newborn lungs by maintaining alveolarexpansion and stabilizing the alveoli (12, 26) and later in themaintenance of lung function (25) and immunoregulation (63).The mechanisms through which glucocorticoids and a numberof other hormones and growth factors affect surfactant synthe-sis, secretion, and re-uptake have been extensively investigated(3, 7, 48, 57). Similarly, the importance of physical factors inthe stimulation of lung growth (36) has been addressed byseveral elegant studies. However, very few studies have fo-cused on the role of in vivo neural input on the development ofthe pulmonary surfactant system during fetal life.
An earlier study showed abnormal pulmonary ultrastructuralchanges and absence of lamellar bodies in vagally denervatedfetal lambs (1), suggesting that vagal innervation plays animportant role in the development of the surfactant systemduring the perinatal period. However, recent studies in fetaland neonatal lambs do not support such an inference (30, 62).At least three distinct possibilities exist that could explain suchdivergent results: 1) in the later studies, vagal denervation wasperformed in the intrathoracic (62) as opposed to the cervicalregion (1), thus sparing the upper airway motor control, which,along with the other physical factors, plays an important role inlung development (20, 27); 2) the denervation was performedlate in the third trimester of pregnancy (62) at a time when thesurfactant system was approaching maturity (11); and, finally,in the earlier studies, 3) the absence of lamellar bodies couldhave been due to suboptimal histological and staining tech-niques, because the predominantly lipid structure of the lamel-lar bodies is modifiable by conventional fixation techniques forelectron microscopy (13, 45, 64).
In view of the critical role that surfactant plays in extrauter-ine adaptation and pulmonary gas exchange, it is vital that therole of vagal innervation in the pulmonary surfactant system beinvestigated in a definitive and systematic manner. The specificaim of the present study was to investigate the relationshipbetween vagal innervation and the surfactant system duringfetal development. We tested the hypothesis that cervical vagaldenervation performed in the developing fetus does not de-crease the lamellar bodies in pulmonary epithelial cells, thetotal lung phospholipids, or surfactant proteins A (SP-A) andB (SP-B).
MATERIALS AND METHODS
Animal preparation. All surgical procedures were performed inaccordance with the Canadian Council on Animal Care, and the studyprotocol was approved by the institutional animal care committee.Twelve fetal sheep underwent either vagal denervation in the cervicalregion (n � 8) or sham surgery (n � 4) at 110–113 days of gestation.Surgery was performed under general anesthesia using ketaminehydrochloride (8–9 mg/kg, Rogarsetic, Rogar/STB, London, Ontario)
Address for reprint requests and other correspondence: S. U. Hasan, Dept. ofPediatrics, Health Sciences Center, 3330 Hospital Drive NW, Calgary, Alberta,Canada T2N 4N1 (e-mail: [email protected]).
J Appl Physiol 106: 1641–1649, 2009.First published March 19, 2009; doi:10.1152/japplphysiol.90868.2008.
and 4% halothane in oxygen for induction and 1.5–2.5% halothane formaintenance.
The fetal head and neck were partially exteriorized through midlinematernal abdominal and uterine incisions. With the use of steriletechniques, a 4-cm ventral neck incision was made to expose the fetalcarotid arteries, jugular veins, and vagosympathetic trunk. Carotidartery and jugular vein catheters (2.0 mm outer diameter and 1.0 mminner diameter, Portex, Hythe, Kent, UK) were inserted and secured inplace. The arterial catheter was used to draw blood for analysis of pHand blood-gas tensions, whereas the venous catheter was used toadminister antibiotics. Vagal denervation was performed in the cer-vical area in eight fetuses (denervated). Approximately 4–5 cm of thevagus nerve was cleared of the surrounding tissue and sectioned. Thesectioned ends were treated with 4% phenol and further folded overand tied with 2.0 silk sutures to avoid nerve regeneration. The vagiwere identified but not sectioned in four fetuses (sham operated).After the completion of instrumentation, the fetuses were returned tothe uterine cavity, and all incisions were sewn in layers. The fetalvascular catheters were exteriorized through an incision in the leftmaternal flank and stored in a cloth pouch secured to the maternalabdominal wall. A polyvinyl catheter was placed in the maternaljugular vein for infusion of antibiotics and fluids (3 mm outerdiameter, Tygon). The ewes were housed in large (�1.8 � 2.4 m)custom-made individual holding pens with free access to food andwater. Daily care included administration of 125 mg (for fetuses) and375 mg (for ewes) of cefazolin sodium in saline (Ancef, SmithKlineBeecham Pharma, Oakville, Ontario), 20 mg (for fetuses) and 80 mg(for ewes) of gentamicin sulfate (Garamycin Injectable, ScheringCanada, Ponta-Claire, Quebec) twice daily for 5 days, and heparinizedflush for the patency of vascular catheters.
Arterial pH and blood-gas tensions. Blood was drawn daily fromthe fetal arterial line for the first 5 postoperative days and then onalternate days, until 130 days of gestation, for measurement of arterialpH and blood-gas tensions (arterial PCO2 and arterial PO2). The arterialpH and blood-gas tensions were measured to ensure fetal well-beinguntil the day of delivery. Furthermore, fetal blood was drawn on thelast day of the experiment to determine plasma cortisol concentra-tions, and analysis was performed using competitive chemilumines-cent immunoassays (Centaur, Bayer Diagnostics, Terrytown, NY).
Postmortem collection of samples. The ewes were monitored for20–23 days postsurgery for any signs of infection or onset of labor.The fetuses were delivered via cesarean section performed undergeneral anesthesia at 130–133 days gestation. Thereafter, the ewesand fetuses were euthanized using Euthanyl (pentobarbitone 240mg/ml), according to the Canadian Council on Animal Care guide-lines. The fetuses were weighed, and complete sectioning of the vagalnerves was verified in each case by identifying the cut and folded endsof the nerve. Location of the denervated vagi was identified by the silksutures used for folding and securing the denervated ends. Similarly,the integrity of the vagal nerves was confirmed in the sham-operatedanimals. Lungs were exposed through a midline thoracotomy. Theright main bronchus was tied off, and the right upper and lower lobeswere removed and sampled for measurements of wet-to-dry weightratio, electron microscopy, and SP-A and -B.
Bronchoalveolar lavage fluid. The left main stem bronchus wascannulated, and bronchoalveolar lavage was performed using chillednormal saline solution (0.9% NaCl). One hundred milliliters perkilogram body weight in four aliquots were slowly infused usinggravitational method to avoid fluid leaks, as previously described (30).Bronchoalveolar lavage fluid (BALF) was used for the assays of totalphospholipids and SP-A and -B. Thereafter, the left lung was perfused(inflation fixed) with 10% freshly prepared formalin at 25 cmH2O for24 h for routine light microscopy. BALF was stored at 4°C andcentrifuged at 150 g for 10 min to remove cellular debris. Thesupernatant was frozen at �70°C for total phospholipids and SP-Aand -B. Total phospholipids in the supernatant were assayed using thewell-established methods of Bartlett (5). Total protein was measured
as described by Lowry et al. (37). Measurement of the total protein inBALF was used to normalize the samples for variations in samplingmethods. Some BALF samples may have more protein, resulting inhigher levels of SP-A and -B that are unrelated to the experimentalcondition. Normalizing to total protein may help in accounting forsome of the variations in the sample values. Furthermore, increasedalveolar-capillary permeability may also occur during bronchoalveo-lar lavage that could, in turn, affect a number of other variables, e.g.,lung wet-to-dry ratios.
Light microscopy. Samples of lung parenchyma of �1.0 � 1.0 �0.3 cm were taken from each lobe. In addition, blocks of tissue weretaken from the lobar bronchi perpendicular to their long axes. Fivemicrometer sections of the airways and the parenchyma were stainedby hematoxylin and eosin and elastic trichrome.
Morphometric analysis. To investigate differences in the airways ofthe sham-operated and denervated animals, morphometric analysiswas performed. Airways were analyzed if seen with cartilage and intrue cross section. The boundary of the airway was determined by thesurrounding lung parenchyma. The inner border of the basementmembrane defined the internal perimeter of the airway. The lumenwas defined as points falling internal to the basement membrane.
The area fractions of selected features in the airway wall profilewere determined by a modified point-counting technique (59) using anAxioplan light microscope (Carl Zeiss model 451888), drawing tube,and square lattice grid containing 240 points, at �10 magnification.The point grid was superimposed onto a segment of the airway wall,and the number of points falling on each area of interest per grid wascounted. The features that were quantified included interstitium, cartilage,mucous gland, nerve, smooth muscle, blood vessel, epithelium, andlumen. Using stereological principles (59), the area proportion occupiedby the structures counted was calculated, using the following formula:area (�m2) � Z2 n, where Z is the distance between two points on the grid(magnification factor), and n is the number of points that land on thestructure of interest. Luminal surface length (internal perimeter) wasdetermined by using the following formula: luminal surface length(�m) � bmi�Z��/4, where bmi is the number of times the grid intersectswith the basement membrane, and Z is the magnification factor (or thedistance between two points on the grid). To determine the thickness(smooth muscle, cartilage), assuming the feature was uniformly dis-tributed around the basement membrane, the following formula wasused: thickness (�m) � area (�m2)/luminal surface length (�m).
Electron microscopy. Lung samples from the right upper and lowerlobes were fixed using a nonaqueous fixation technique optimized forthe preservation of mucoproteins and surfactant lipids (34, 51, 53).The samples were fixed with 1% osmium tetroxide (wt/vol) dissolvedin fluorocarbon FC-77 fluid for 90 min and further postfixed in 2.5%glutaraldehyde solution containing 1% cetylpyridinium chloride in0.05 M sodium cacodylate (CaCo) buffer, pH � 7.3, for 90 min atroom temperature. The tissues were then dehydrated in acetone,infiltrated with acetone/Epon 812, sectioned, stained with uranylacetate/lead citrate, and examined with a Hitachi 7000 TEM electronmicroscope.
Lung wet-to-dry ratio. To determine the lung wet-to-dry-weightratios, three pieces of lung tissue from upper and lower lobes were tiedand dissected from sham-operated and denervated animals, weighed,and dried in preweighed aluminum containers for 3 days at 120°C.
SP-A and SP-B assays. SP-A and SP-B concentrations were deter-mined as previously described, courtesy Dr. Jeffrey Whitsett andWilliam Hull, Cincinnati, OH (2, 43). Briefly, wells were coated with100 �l of 0.1 M sodium bicarbonate and left overnight at 4°C. Theplates were washed with wash buffer solution, and 100 �l of 82bbuffer with 5% goat serum for SP-A samples or 5% human albuminfor SP-B samples were added. The wash buffer comprised 0.1 M Tris,pH 8.0, and 0.05% (vol/vol) Tween 20. The 82b buffer comprised0.15 M NaCl, 0.01 M Tris, pH 7.4, and 5 mg/ml bovine serumalbumin. For SP-A, 5% (vol/vol) of 25% human serum albumin wasadded, and, for SP-B, 2.5% (vol/vol) of 25% human serum albumin
and 2.5% (vol/vol) of goat serum was added. After 15 min, the bufferwas removed, and 100 �l of standard at concentrations of 5, 10, 25,50, 75, and 100 ng/ml were added and incubated for 1–2 h at 37°C.One hundred microliters of sample were added per well, diluted inphosphate-buffered saline containing 5% (vol/vol) Nonidet P-40.Serial dilutions of 1:10 and 1:4, resulting in the dilutions of 1:10, 1:40,1:160, and 1:640, were used. The wells were washed three times withwash buffer, and 100 �l of horseradish peroxidase conjugate (goatanti-rabbit IgG), diluted 1:1,000 in buffer with 5% human plasma forSP-A samples and 5% human albumin for SP-B samples, were addedand incubated at 37°C for 1 h. Before the addition of the horseradishperoxidase conjugate, antibodies specific to SP-A and SP-B wereadded in a 1:5,000 dilution. The wells were again washed with washbuffer, 100 �l of substrate solution were added, and color devel-opment was observed. Color development was stopped with 100 �lof 50% sulfuric acid. The absorption was read at an absorbance of492 nm.
Statistical analysis. The gestational age at surgery and at delivery,fetal body weight at delivery, lung wet-to-dry ratios, plasma cortisolconcentrations, total phospholipids in BALF and lung tissue, morpho-metric analysis, and SP in tissues and BALF were analyzed using anindependent sample t-test. The effects of vagal denervation and shamsurgery on arterial pH and blood-gas tensions between sham-operatedand denervated animals were analyzed using two-way ANOVA forrepeated measure. Spearman’s rank correlation was used to test thedirection and strength of the plasma cortisol concentrations and SP-Aand -B in BALF and lung tissue. Since no significant differences wereobserved between sham-operated and denervated animals in Spear-man’s rank correlation analysis, data were combined to delineate therelationship between cortisol concentrations and SP-A and -B. Allvalues are given as means � SE, and P � 0.05 was consideredsignificant.
RESULTS
Arterial pH and blood-gas tensions. The arterial pH, arterialPCO2, and arterial PO2 (Torr) are given in Fig. 1. The pH and theblood-gas tensions decreased in sham-operated animals anddenervated animals during the early postoperative period andgradually returned to normal range.
Gestational ages, body weight, lung wet-to-dry ratio, andplasma cortisol concentrations. The gestational age at surgeryand at cesarean section was similar between the sham-operatedand denervated groups. No difference was observed in bodyweight at delivery, lung wet-to-dry-ratio, and plasma cortisolconcentrations between the two groups (Table 1). The section-ing and integrity of the vagi in the denervated and sham-operated animals, respectively was confirmed in each case.
Light and electron microscopy. The lungs of the sham-operated and denervated animals appeared grossly normal.Examples of the light microscopic appearances of the lungparenchyma and airways of sham-operated and denervatedanimals are shown in Figs. 2 and 3, respectively. The histo-logical appearances of the lung parenchyma were characteristicof normal development for the gestational age of the fetallambs in both groups. In both groups, mature type II cells wereseen within the alveoli. The airways were also histologicallynormal, with no visible differences between the two studygroups (Fig. 3).
Electron microscopy showed mature lamellar bodies in thetype II cells in both the sham-operated and denervated animals(Fig. 4, A and B), respectively. The number and size of the
Fig. 1. Arterial pH (pHa; A), arterial PCO2 (PaCO2; B), and arterial PO2 (PaO2; C) at various gestations in sham-operated and denervated animals. No significantdifferences were observed between the two groups.
lamellar bodies in type II alveolar epithelial cells betweensham-operated and denervated animals were similar by subjec-tive assessment.
Morphometric analysis. The result of the morphometricanalysis of airway features is illustrated in Fig. 5. There wereno significant differences in the area fractions (expressed as athickness in �m) of cartilage, smooth muscle, mucous glands,nerves, blood vessel, interstitium, and lumen in the sham-operated and denervated animals.
Total phospholipids and total proteins. The results of thetotal phospholipids and total proteins measured in BALF areshown in Fig. 6, A and B. Total phospholipids were 9.2 � 1.58and 6.8 � 0.64 mg/kg in the denervated and sham-operatedgroups, respectively. Total proteins were 17.2 � 1.6 and
14.1 � 2.2 mg/ml in the denervated and sham-operated groups,respectively.
SP-A and SP-B. Surfactant-associated proteins measured inBALF and lung tissues are shown in Fig. 6, C and D, respec-tively. SP-A and SP-B measured in lung tissue and BALFwere similar between the two study groups. However, asignificant positive correlation was observed between circu-lating cortisol concentrations and SP-A and -B in BALF andlung tissue in both sham-operated and vagally denervatedanimals (Figs. 7 and 8, respectively).
DISCUSSION
The remarkable increase in the survival of very low birthweight infants (35) has largely been attributed to the wide-spread administration of maternal glucocorticoids during pre-term labor to augment lung maturation (47) and postnataltreatment of infants with exogenous surfactant, which plays acritical role in reducing surface tension at the liquid-air inter-face (26). Although the subject of vagal innervation and itseffects on the surfactant system have been debated for the pastthree decades with divergent results, the present study providesimportant and unequivocal evidence that vagal input at mid-gestation is not necessary for type II pneumocyte maturation.Electron microscopy of sham-operated and denervated animalsrevealed no differences in size or maturity of the lamellar
Fig. 2. Light microscopy of lung parenchyma in sham-operated (A and B) and denervated (C and D) animals. Both photomicrographs show a normal membranousbronchiole in the center of the field of view, surrounded by alveoli of approximately equal size. At higher magnifications (B and D), both groups showed maturetype II cells (arrows) in the alveoli, with no apparent differences between sham-operated and denervated animals. A and C: original magnification �100; B andD: �400.
Table 1. Body weight, lung wet-to-dry ratio,and plasma cortisol concentrations at delivery
Sham Denervated
Gestational age at surgery, days 110�0.0 110�0.3Gestational age at cesarean section, days 131�0.3 131�0.4Body weight, kg 3.63�0.1 3.51�0.2Lung wet-to-dry ratio 7.5�1.4 7.8�0.3Plasma cortisol concentrations, ng/ml 23�12 51�14
bodies in type II pneumocytes. In addition, we provide impor-tant new data on total phospholipids and SP-A and -B in BALFand lung tissues. Kitterman et al. (28) have reported a corre-lation between plasma cortisol concentrations and saturatedphosphatidylcholine (PC). Other studies have shown that ad-ministration of glucocorticoids leads to reversible increase insurfactant-associated proteins (54); however, to our knowl-edge, a correlation between circulating cortisol concentrationsand surfactant proteins has not been investigated in the devel-oping ovine fetus. Thus our study provides new data showinga strong correlation between circulating cortisol concentrationsand SP-A and -B. We also show that motor innervation of thelarynx does not appear essential for lung growth in utero, asshown by the normal lung histology, lung weights, and mor-phometric indexes of the airways in the denervated animals.
Vagal innervation, lamellar body formation, and matura-tion. Using the experimental protocol similar to ours, Alcornet al. (1) reported normal lung development and architecture intheir vagally denervated lambs. However, they also observedvacuole-like, membrane bound intra-cytoplasmic empty spaceswithin type II cells, which they interpreted as abnormal lamel-lar bodies. In contrast, we report mature lamellar bodies withinthe type II cells of our denervated lambs. Lamellar bodies arethe intracellular storage sites for pulmonary surfactant, as theyare deficient in a number of lysosomal enzymes necessary forthe de novo synthesis of PC and phosphatidylglycerol. Thus
the loss of lipid material from tissue sections may be an artifactof fixation, embedding, or sectioning. Phospholipids, the majorcomponent of lamellar bodies and plasma membranes, areparticularly vulnerable to these artifacts, making them difficultto preserve for electron microscopy (13, 45, 64). The classicprocedure of gluteraldehyde fixation followed by alcohol de-hydration and epoxy resin embedding can result in the loss of73–91% of lipids (38, 64). Furthermore, differences in theviscosity of the epoxy resin components could lead to organelleshrinkage during embedding. Spurr resin, which was used inthe study by Alcorn et al. (1), is particularly vulnerable to thisshrinkage effect (45). Yang et al. (64) describe atypical andempty lamellar bodies following conventional fixation ofmouse lungs. Their illustration of abnormal lamellar bodies,resulting from fixation artifacts, are almost identical to thoseshown by Alcorn et al. (1) in both vagally denervated andsham-operated lambs, findings that suggest the empty vacuole-like spaces were the result of a fixation or processing artifact.
Several approaches for reducing loss of lipid material fromcell membranes and lamellar bodies have been used, with
Fig. 4. Electron microscopy of peripheral lung tissue in sham-operated (A) anddenervated (B) animals. Both samples were fixed with nonaqueous techniqueusing osmium/perfluorocarbon mixture. The type 2 cells from both groupsshow normal, mature lamellar bodies of similar size in the apical cytoplasm.Original magnification: �6,000.
Fig. 3. Light microscopy of pulmonary airways in sham-operated (A) anddenervated (B) animals. The airways from both groups were normal andsimilar in appearance. Original magnification: �25.
varying success. Such approaches include addition of 1%osmium tetroxide to the primary fixation (13), changing thecomposition of the embedding medium and substituting ace-tone for ethyl alcohol for dehydration (9), rapid freezing (60),freeze drying without chemical fixation or solvent dehydration(42), and freeze drying combined with osmium vapor fixation(64). We used nonaqueous fixation technique originally devel-oped by Sims et al. (53) and further modified by Lee et al. (34)and Schurch et al. (51). Using this technique, we show thatlamellar bodies are structurally normal in vagally denervatedlambs.
Vagal innervation, surfactant secretion, and dysfunction.The role of vago-sympathetic innervation on surfactant syn-thesis, secretion, and dysfunction is another area that has longbeen debated (1, 41, 46). Cholinergic and �-adrenergic mech-anisms, including direct stimulation of the vagal nerve andacetylcholine infusion, have been shown to stimulate the se-cretion of lung surfactant, both in vivo and in isolated perfusedlung preparations (10, 16, 41, 46, 48). Although isolated typeII cells contain receptors for both cholinergic and �-adrenergicagonists, it is only the �-adrenergic agonists that stimulatesecretion of surfactant from the alveolar type II cells (10). Arapid increase in surfactant flux (12- to 13-fold) and pool sizeshortly before birth has been documented in fetal sheep andother species (6, 44). Since our experimental protocol wasdesigned to deliver fetuses around 133 days gestation, it ispossible that the effects of vagosympathetic denervation on a
number of physiological and biochemical variables might havebeen different had the fetuses been delivered closer to termgestation. Sympathetic (- and �-adrenergic) and cholinergicsystems can have specific effects on surfactant synthesis andsecretion. Schellenberg et al. (50) investigated the effects ofchemical sympathectomy and �-adrenergic blockade on alve-olar and tissue phospholipids and saturated PC. They showedthat chemical sympathectomy suppressed the increase in alve-olar phospholipids in response to thyrotropin-releasing hor-mone and cortisol administration (50). However, the lungtissue phospholipids and/or saturated PC increased in a mannersimilar to or higher than the values observed in nonsympath-etectomized animals, suggesting that stimulation of pulmonarysynthesis is not dependent on sympathetic innervation. Inhibi-tion of surfactant secretion may be the reason for loweralveolar phospholipid values in sympathetectomized animals(50), indicating separate mechanisms for surfactant synthesisand secretion. It is possible that, if vagally denervated lambshad been delivered at term, they might have had intact lamellarbodies and lung tissue surfactant constituents, but lower alve-olar phospholipids and saturated PC concentrations.
Direct mechanical stretching, due to lung inflation, of thetype II cells is believed to be the primary physiological stim-ulus for the secretion of surfactant (49, 61). Stretching of thetype II cells increases intracellular Ca2 concentration andincreases surfactant release (61). A number of studies haveshown that air inflation of the lungs is a sufficient stimulus forsurfactant release (23, 39, 46), and that the initial expansion ofthe newborn lungs acts as the primary stimulus for surfactantrelease (33). In the present study, animals were delivered bycesarean section under general anesthesia, and the lungs werenot inflated with air or subjected to spontaneous breathing, thuseliminating these as potential confounding factors for thisstudy. Surfactant dysfunction after vagal denervation (8, 18,29) could result from pulmonary edema secondary to upperairway obstruction (8, 18, 29) and increased intrathoracicpressure, leading to damaged alveolar-capillary membrane.Such change in permeability could allow serum proteins,hemoglobin, fatty acids, and cellular degradation products intothe alveolar space, where they can impair the surface-activefunction of surfactant (24, 58). Vagal denervation performed inthe intrathoracic region, which spares the recurrent laryngealnerves, is not associated with a decrease in total phospholipidsor to histological changes of respiratory distress (30, 32).However, absence of the afferent feedback, critically requiredfor lung expansion at birth, could lead to an increase in theminimum surface tension of the lungs due to pulmonaryatelectasis (62). Both overexpansion (40) and/or atelectasis(56), which can be associated with vagal denervation, can alsolead to surfactant dysfunction. In summary, data from othersand our laboratory suggest that vagal denervation does not playa direct role in surfactant dysfunction (8, 24, 30, 32, 58, 62).
Motor innervation of the upper airway and morphometricanalysis. No significant differences were observed in the lungmorphometric analysis between the two groups. Motor inner-vation of the upper airway, which is dependent on intact vagiand fetal breathing movements, plays important roles in themaintenance of functional residual capacity and lung develop-ment. Vagal denervation leads to paralysis of all laryngealmuscles, causing the vocal cords to lie in a cadaveric position(53a). During fetal apnea, the laryngeal adduction (thyro-aryte-
Fig. 5. Morphometric analysis of conducting airways in sham-operated anddenervated animals for interstitium, cartilage, and mucous gland (A) and nerve,smooth muscle (Ms), and blood vessels (Vs; B). Values are expressed as athickness in millimeters. No significant differences were observed between thetwo groups.
noid muscles) reduces the efflux of tracheal fluid, whereas, duringfetal breathing movements, active phasic laryngeal dilation medi-ated via posterior crico-arytenoid muscles leads to high net tra-cheal efflux (21). Thus it is likely that vocal cord paralysis mayallow increased efflux during fetal apnea (�50% of the total timeat 110–125 days gestation), while absence of phasic dilation inconjunction with reduced lung recoil during fetal breathing
movements (50% of the total time) may retard the tracheal fluidefflux, resulting in little net change in overall lung expansion.Furthermore, fetal sheep can defend its lung expansion byincreasing lung liquid secretion (22). Finally, it is likely thatvagal denervation performed at a relatively late gestation interms of structural pulmonary development and for a short
Fig. 6. A: total phospholipids (mg/kg body wt) measured in the BALF at 130–133 days gestation in sham-operated and denervated animals. B: total protein(mg/ml) measured in BALF at 130–133 days gestation in sham-operated and denervated animals. C: surfactant protein (SP)-A and -B measured in BALF (�g/ml).D: SP-A and -B in the lung tissue (�g/mg). No significant differences were observed between the sham-operated and denervated animals.
Fig. 7. SP-A and SP-B in BALF (�g/ml) as a correlation to plasma cortisolconcentrations (ng/ml). Significant correlation was observed between SP-A,SP-B, and cortisol concentrations (P � 0.003 and 0.012, respectively).
Fig. 8. SP-A and SP-B in lung tissue (�g/mg protein) as a correlation toplasma cortisol concentrations (ng/ml). Significant correlation was observedbetween SP-A, SP-B, and cortisol concentrations (P � 0.015 and 0.017,respectively).
duration (2–3 wk) would have little effect on morphometricanalysis.
Since glucocorticoids play a critical role in maturation of thesurfactant system (14, 44, 54) and may have been affected by fetaland maternal surgical procedures and sectioning of the vagosym-pathetic trunk, we measured plasma cortisol concentrations atdelivery of the animals. We found no statistically significantdifference in plasma cortisol concentrations between sham-oper-ated and denervated animals. Furthermore, these values are com-parable with the concentrations obtained from control animalsin our laboratory (35.01 � 10.21; mean � SE; unpublishedobservations). Also, these plasma cortisol values are similar tothose reported by Schellenberg et al. (50) in control animals.The increase in surfactant proteins is reversible, despite re-peated large doses of glucocorticoids (54). In our present study,we followed the experimental protocol, as described by Alcornet al. (1), to clarify the role of vagal innervation in pulmonarysurfactant during fetal development. However, there are cleardifferences in the presence of lamellar bodies between the twostudies, which suggest that the differences in the presence oflamellar bodies between the two studies cannot be explainedby the surgical stress alone and are the result of fixationartifact.
In conclusion, we provide definitive evidence that the mat-uration of the surfactant system during fetal development insheep is not dependent on intact vagal innervation. Further-more, the present study provides a number of gestation-specificimportant morphological and biochemical variables related tofetal lung development. Our findings do not contradict thestudies that have established the important role of vagal inner-vation in the maintenance of normal breathing patterns andpulmonary gas exchange during the early neonatal period (17,30, 31, 52). Instead, our findings strongly support the hypoth-esis that the effects of vagal denervation in the early neonatalperiod are secondary to the alterations in pulmonary mechan-ical forces induced by the denervation. Previous studies show-ing absence of lamellar bodies in vagally denervated fetallambs were probably an artifact resulting from suboptimal lungtissue fixation techniques. In fact, a trend toward an increasedplasma cortisol concentration, higher total phospholipids, andsurfactant-associated proteins in the denervated animals furtherlend support to our hypothesis that vagal innervation playslittle role in the maturation of the surfactant system during fetaldevelopment. Finally, it is important to note that our compre-hensive study addresses and resolves the long-debated relation-ship between vagal innervation and lamellar body maturationduring fetal development. Thus the present study represents asignificant and much needed contribution to the important areaof lung growth and maturation literature during fetal develop-ment.
ACKNOWLEDGMENTS
We express our appreciation to Dr. Jeffrey Whitsett and William Hull(Children’s Hospital Medical Center, Cincinnati, OH) for generous help withsurfactant protein assays. We also thank Dr. Tak Fung and Linda Brigan forstatistical and editorial assistance, respectively.
GRANTS
This study was supported by the Canadian Institutes of Health Research.
REFERENCES
1. Alcorn D, Adamson TM, Maloney JE, Robinson PM. Morphologicaleffects of chronic bilateral phrenectomy or vagotomy in the fetal lamblung. J Anat 130: 683–695, 1980.
3. Andreeva AV, Kutuzov MA, Voyno-Yasenetskaya TA. Regulation ofsurfactant secretion in alveolar type II cells. Am J Physiol Lung Cell MolPhysiol 293: L259–L271, 2007.
5. Bartlett GR. Phosphorus assay in column chromatography. J Biol Chem234: 466–468, 1959.
6. Batenburg JJ. Surfactant phospholipids: synthesis and storage. Am JPhysiol Lung Cell Mol Physiol 262: L367–L385, 1992.
7. Batenburg JJ. Biosynthesis, secretion, and recycling of surfactant com-ponents. In: Surfactant Therapy of Lung Disease, edited by Robertson B,Taeusch HW. New York: Dekker, 1995, p. 47–73.
8. Berry D, Ikegami M, Jobe A. Respiratory distress and surfactant inhi-bition following vagotomy in rabbits. J Appl Physiol 61: 1741–1748,1986.
9. Berryman MA, Rodewald RD. An enhanced method for post-embeddingimmunocytochemical staining which preserves cell membranes. J Histo-chem Cytochem 38: 159–170, 1990.
10. Brown LA, Longmore WJ. Adrenergic and cholinergic regulation oflung surfactant secretion in the isolated perfused rat lung and in thealveolar type II cell in culture. J Biol Chem 256: 66–72, 1981.
11. Cockshutt AM, Possmayer F. Metabolism of the surfactant lipids andproteins in the developing lung. In: Pulmonary Surfactant–From Molec-ular Biology to Clinical Practice, edited by Robertson B, Van GoldeLMG, Batenburg JJ. Amsterdam: Elsevier, 1992, p. 339–377.
12. Cole FS, Nogee LM, Hamvas A. Defects in surfactant synthesis: clinicalimplications. Pediatr Clin North Am 53: 911–27, 2006.
13. Collet AJ. Preservation of alveolar type II pneumocyte lamellar bodies forelectron microscopic studies. J Histochem Cytochem 27: 989–996, 1979.
14. Creuwels LAJM, van Golde LMG, Haagsman HP. The pulmonarysurfactant system: biochemical and clinical aspects. Lung 175: 1–39, 1997.
15. Delacourt C, Canet E, Praud JP, Bureau MA. Influence of vagalafferents on diphasic ventilatory response to hypoxia in newborn lambs.Respir Physiol 99: 29–39, 1995.
16. Dobbs LG, Mason RJ. Pulmonary alveolar type II cells isolated fromrats. Release of phosphatidylcholine in response to beta-adrenergic stim-ulation. J Clin Invest 63: 378–387, 1979.
17. Fedorko L, Kelly EN, England SJ. Importance of vagal afferents indetermining ventilation in newborn rats. J Appl Physiol 65: 1033–1039,1988.
18. Goldenberg VE, Buckingham S, Sommers SC. Pulmonary alveolarlesions in vagotomized rats. Lab Invest 16: 693–705, 1967.
19. Greer JJ, Funk GD, Ballanyi K. Preparing for the first breath: prenatalmaturation of respiratory neural control. J Physiol 570: 437–444, 2006.
20. Harding R. Fetal pulmonary development: the role of respiratory move-ments. Equine Vet J Suppl 24: 32–39, 1997.
21. Harding R, Hooper SB. Regulation of lung expansion and lung growthbefore birth. J Appl Physiol 81: 209–224, 1996.
22. Harding R, Hooper SB, Han VKM. Abolition of fetal breathing move-ments by spinal cord transection leads to reductions in fetal lung liquidvolume, lung growth, and IGF-II gene expression. Pediatr Res 34: 148–153, 1993.
23. Hildebran JN, Goerke J, Clements JA. Surfactant release in excised ratlung is stimulated by air inflation. J Appl Physiol 51: 905–910, 1981.
24. Holm BA, Notter RH. Effects of hemoglobin and red blood cell mem-brane lipids on the biophysical and physiological activity of pulmonarysurfactant. J Appl Physiol 63: 1434–1442, 1987.
25. Ikegami M, Whitsett JA, Martis PC, Weaver TE. Reversibility of lunginflammation caused by SP-B deficiency. Am J Physiol Lung Cell MolPhysiol 289: L962–L970, 2005.
27. Joe P, Wallen LD, Chapin CJ, Lee CH, Allen L, Han VK, Dobbs LG,Hawgood S, Kitterman JA. Effects of mechanical factors on growth andmaturation of the lung in fetal sheep. Am J Physiol Lung Cell Mol Physiol272: L95–L105, 1997.
28. Kitterman JA, Liggins GC, Campos GA, Clements JA, Forster CS,Lee CH, Creasy RK. Prepartum maturation of the lung in fetal sheep:relation to cortisol. J Appl Physiol 51: 384–390, 1981.
29. Kunc L, Kuncova M, Holusa R, Soldan F. Physical properties andbiochemistry of lung surfactant following vagotomy. Respiration 35:192–197, 1978.
30. Lalani S, Remmers JE, Green FH, Bukhari A, Ford GT, Hasan SU.Effects of vagal denervation on cardiorespiratory and behavioral responsesin the newborn lamb. J Appl Physiol 91: 2301–2313, 2001.
31. Lalani S, Remmers JE, Hasan SU. Breathing patterns, pulmonarymechanics and gas exchange: role of vagal innervation in neonatal lamb.Exp Physiol 86: 803–810, 2001.
32. Lalani S, Remmers JE, MacKinnon Y, Ford GT, Hasan SU. Hypox-emia and low Crs in vagally denervated lambs result from reduced lungvolume and not pulmonary edema. J Appl Physiol 93: 601–610, 2002.
33. Lawson EE, Birdwell RL, Huang PS, Taeusch HW Jr. Augmentationof pulmonary surfactant secretion by lung expansion at birth. Pediatr Res13: 611–614, 1979.
34. Lee MM, Schurch S, Roth SH, Jiang X, Cheng S, Bjarnason S, GreenFH. Effects of acid aerosol exposure on the surface properties of airwaymucus. Exp Lung Res 21: 835–851, 1995.
35. Lemons JA, Bauer CR, Oh W, Korones SB, Papile LA, Stoll BJ,Verter J, Temprosa M, Wright LL, Ehrenkranz RA, Fanaroff AA,Stark A, Carlo W, Tyson JE, Donovan EF, Shankaran S, StevensonDK. Very low birth weight outcomes of the National Institute of Childhealth and human development neonatal research network, January 1995through December 1996. NICHD Neonatal Research Network. Pediatrics107: E1, 2001.
36. Liu M, Skinner SJ, Xu J, Han RN, Tanswell AK, Post M. Stimulationof fetal rat lung cell proliferation in vitro by mechanical stretch. Am JPhysiol Lung Cell Mol Physiol 263: L376–L383, 1992.
37. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measure-ment with the Folin-Phenol reagents 6790. J Biol Chem 193: 265–275,1951.
38. Maneta-Peyret L, Compere P, Moreau P, Goffinet G, Cassagne C.Immunocytochemistry of lipids: chemical fixatives have dramatic effectson the preservation of tissue lipids. Histochem J 31: 541–547, 1999.
39. Martinez F, Lewis J, Copland I, Engelberts D, Kavanagh BP, Post M,Schurch S, Belik J. Mechanical ventilation effect on surfactant content,function, and lung compliance in the newborn rat. Pediatr Res 56: 19–25,2004.
40. Maruscak AA, Vockeroth DW, Girardi B, Sheikh TA, Possmayer F,Lewis JF, Veldhuizen RA. Alterations to surfactant precede physiologi-cal deterioration during high tidal volume ventilation. Am J Physiol LungCell Mol Physiol 294: L974–L983, 2008.
41. Massaro D, Clerch L, Massaro GD. Surfactant secretion: evidence thatcholinergic stimulation of secretion is indirect. Am J Physiol Cell Physiol243: C39–C45, 1982.
42. McAteer JA, Terracio L. Pulmonary type II cell lamellar body ultra-structure preserved by rapid freezing and freeze drying. Anat Rec 209:355–362, 1984.
43. McMahan MJ, Mimouni F, Miodovnik M, Hull WM, Whitsett JA.Surfactant associated protein (SAP-35) in amniotic fluid from diabetic andnondiabetic pregnancies. Obstet Gynecol 70: 94–98, 1987.
44. Mescher EJ, Platzker AC, Ballard PL, Kitterman JA, Clements JA,Tooley WH. Ontogeny of tracheal fluid, pulmonary surfactant, and plasmacorticoids in the fetal lamb. J Appl Physiol 39: 1017–1021, 1975.
45. Mollenhauer HH. Artifacts caused by dehydration and epoxy embeddingin transmission electron microscopy. Microsc Res Tech 26: 496–512,1993.
46. Oyarzun MJ, Clements JA. Ventilatory and cholinergic control ofpulmonary surfactant in the rabbit. J Appl Physiol 43: 39–45, 1977.
47. Roberts D, Dalziel S. Antenatal corticosteroids for accelerating fetal lungmaturation for women at risk of preterm birth. Cochrane Database SystRev 3: CD004454, 2006.
48. Rooney SA. Regulation of surfactant secretion. Comp Biochem Physiol AMol Integr Physiol 129: 233–243, 2001.
49. Rooney SA, Young SL, Mendelson CR. Molecular and cellular process-ing of lung surfactant. FASEB J 8: 957–967, 1994.
50. Schellenberg JC, Liggins GC, Kitterman JA, Lee CC. Sympathectomyand beta-adrenergic blockade during lung maturation stimulated by TRHand cortisol in fetal sheep. J Appl Physiol 75: 141–147, 1993.
51. Schurch S, Green FH, Bachofen H. Formation and structure of surfacefilms: captive bubble surfactometry. Biochim Biophys Acta 1408: 180–202, 1998.
52. Schwieler GH. Respiratory regulation during postnatal development incats and rabbits and some of its morphological substrate. Acta PhysiolScand 304: 7–123, 1968.
53. Sims DE, Westfall JA, Kiorpes AL, Horne MM. Preservation oftracheal mucus by nonaqueous fixative. Biotech Histochem 66: 173–180,1991.
53a.Standring S. Gray’s Anatomy: The Anatomical Basis of Clinical Practice.Edinburgh: Churchill Livingstone, 2008.
54. Tan RC, Ikegami M, Jobe AH, Yao LY, Possmayer F, Ballard PL.Developmental and glucocorticoid regulation of surfactant protein mRNAs inpreterm lambs. Am J Physiol Lung Cell Mol Physiol 277: L1142–L1148,1999.
55. Veldhuizen R, Possmayer F. Phospholipid metabolism in lung surfactant.Subcell Biochem 37: 359–388, 2004.
56. Verbrugge SJ, Lachmann B. Mechanisms of ventilation-induced lunginjury: physiological rationale to prevent it. Monaldi Arch Chest Dis 54:22–37, 1999.
57. Voyno-Yasenetskaya TA, Dobbs LG, Williams MC. Regulation ofATP-dependent surfactant secretion and activation of second-messengersystems in alveolar type II cells. Am J Physiol Heart Circ Physiol 261:H105–H109, 1991.
58. Wang Z, Notter RH. Additivity of protein and nonprotein inhibitors oflung surfactant activity. Am J Respir Crit Care Med 158: 28–35, 1998.
59. Weibel ER, Gomez DM. A principle for counting tissue structures onrandom sections. J Appl Physiol 17: 343–348, 1962.
60. Williams MC. Freeze-fracture studies of tubular myelin and lamellarbodies in fetal and adult rat lungs. J Ultrastruct Res 64: 352–361, 1978.
61. Wirtz HR, Dobbs LG. Calcium mobilization and exocytosis after onemechanical stretch of lung epithelial cells. Science 250: 1266–1269, 1990.
62. Wong KA, Bano A, Rigaux A, Wang B, Bharadwaj B, Schurch S,Green F, Remmers JE, Hasan SU. Pulmonary vagal innervation isrequired to establish adequate alveolar ventilation in the newborn lamb.J Appl Physiol 85: 849–859, 1998.
63. Wright JR. Host defense functions of pulmonary surfactant. Biol Neonate85: 326–332, 2004.
64. Yang C, Terada N, Ohno N, Fujii Y, Ohno S. Morphological analysis oflamellar structures in mouse type II pneumocytes by quick-freezing andfreeze-drying with osmium tetroxide vapor-fixation. Med Mol Morphol39: 88–96, 2006.
![Muscle Innervation Chart II[1]](https://static.documents.pub/doc/80x56/55241db64a7959da488b45f0/muscle-innervation-chart-ii1.jpg)
