Main

During fetal development, the future airways of the lungs are filled with a liquid that is produced by the pulmonary epithelium and which leaves the lungs via the trachea. The retention of an appropriate volume of this liquid within the future airways is an important determinant of the growth and structural maturation of the developing lungs. In fetal sheep, prolonged reductions in the volume of lung liquid, caused by its continuous drainage via the trachea, leads to decreases in lung tissue weight(1) and DNA content(2, 3). On the other hand, prolonged increases in lung liquid volume, caused by obstructing the fetal trachea, greatly stimulates lung growth(1, 2, 4). Because increased lung expansion also greatly stimulates growth of hypoplastic lungs in utero(3), this procedure has been proposed as a potential method for stimulating lung growth in human fetuses with pulmonary hypoplasia(3–6). However, little is known about the physiologic, cellular, and molecular mechanisms by which alterations in fetal lung expansion influence lung growth. Owing to the therapeutic potential of tracheal obstruction in correcting lung growth deficiencies, we wished to gain a greater understanding of the factors that regulate the lung growth response to alterations in fetal lung expansion.

Previous studies in fetal sheep have shown that the increase in lung size, as measured by increases in DNA and protein contents, attained after 7 d of tracheal obstruction is similar to that after 25 d(2, 4). This indicates that the acceleration in fetal lung growth induced by tracheal obstruction occurs within the first 7 d and is followed by a return to control growth rates after this time. It is not known why fetal lung growth ceases after the initial period of rapid growth, although we have suggested that it results from an inability of the fetal lung to expand further, despite a sustained elevation in intraluminal pressure(4). Consequently, one aim of this study was to determine the relationships between the increases in lung liquid volume, intraluminal pressure, and lung growth after tracheal obstruction. We hypothesized that lung liquid volumes would increase linearly after tracheal obstruction and would closely parallel the increase in lung growth, ceasing after 7 d of tracheal obstruction when the increase in lung growth had stopped. Thus, we have measured the increase in lung DNA and protein contents at 4 d of tracheal obstruction, combined these data with data we have obtained previously after 2 and 7 d of tracheal obstruction(4, 7), and related these changes to changes in lung liquid volume and tracheal pressure over this time. In addition, we have measured the increase in lung DNA and protein contents after 10 d of tracheal obstruction to test our earlier suggestion that the acceleration in lung growth is complete by d 7(4). The combination of these data permits a detailed analysis of the lung growth and lung liquid volume changes that occur over the first 10 d of tracheal obstruction.

Because the acceleration in fetal lung growth after tracheal obstruction is time-dependent, it is important to define the time period of maximal lung growth if the underlying mechanisms are to be identified. Indeed, the cellular mechanisms responsible for the increased growth may be active only during the period of accelerated growth. In a recent study(7) we showed that only a small increase in lung DNA content could be detected within 2 d of obstructing the trachea in fetal sheep, although DNA synthesis rates were markedly elevated at this time(7). Because DNA synthesis rates are a sensitive measure of tissue growth, a second aim of this study was to determine the time period of maximum DNA synthesis rates after tracheal obstruction. Identification of this time period will indicate when the mechanisms that stimulate the acceleration in lung growth are most active and can be studied.

METHODS

Surgery was performed on 24 pregnant ewes (Border-Leicester × Merino) at 113-120 d of gestation (term is 145 d). Anesthesia was induced with thiopentone sodium (1 g I.V.) and was maintained, after tracheal intubation, with 1.5% halothane in O2-N2O (50:50 vol/vol). In each fetus, two silicone rubber cannulas (Dow Corning, catalog no. 601-365) were implanted 2-4 cm into the trachea, one directed toward the lungs and the other directed toward, but not entering, the larynx. Securing the rubber cannulas with silk ties during surgery ensures that there is no leakage of lung liquid around the catheter(8–10). These catheters were joined externally to form a tracheal loop that allowed continuous flow of lung liquid. Polyvinyl catheters were also implanted into a fetal carotid artery and jugular vein for blood sampling. Antibiotics (900 mg of Depomycin(Intervet, Castle Hill, Australia), intramuscularly) were administered to the fetus before the uterine incision was closed. Ewes and fetuses were allowed to recover for 5 d after surgery. Fetal arterial Po2, Pco2, pH, and percentage O2 saturation of Hb were measured daily (ABL30, Radiometer, Denmark).

Experimental protocol. Animals were divided into four treatment groups: 1) in the 4-d obstruct group, the fetal trachea was obstructed for 4 d by occluding the exteriorized tracheal loop (124-128 d,n = 5); 2) in the 4-d control group, lung liquid was allowed to flow through the tracheal loop unimpeded for 4 d (124-128 d,n = 5); 3) in the 10-d obstruct group, the fetal trachea was obstructed (as for 4-d obstruct) for 10 d (118-128 d, n= 9); and 4) in the 10-d control group, lung liquid was allowed to flow through the tracheal loop unimpeded for 10 d (118-128 d,n = 5). The 10-d obstruct group was divided into two subgroups, in which we measured either DNA synthesis rates (n = 4) or lung liquid volumes, secretion rates, and tracheal pressures (n = 5). Pulmonary DNA and protein contents were measured in all fetuses.

At the end of the experimental period, fetuses were injected with[3H]thymidine (1 mCi/kg estimated fetal body weight, i.v., Amersham, UK) to measure relative DNA synthesis rates from the incorporation of[3H]thymidine into DNA. [3H]Thymidine was not administered to those fetuses in which lung liquid volume and secretion rate measurements were made because of the possibility that large changes in lung expansion, which are an integral part of making these measurements, may influence the rate of DNA synthesis and, therefore, the rate of [3H]thymidine uptake. Fetuses that received [3H]thymidine were killed 8 h later. The lungs of all fetuses were drained of liquid, and the tracheal loop was blocked before the ewe and fetus were killed by an overdose of sodium pentobarbitone administered to the ewe (130 mg/kg, i.v.). The fetal lungs were weighed, and portions were rapidly frozen in liquid nitrogen and stored at -70°C for subsequent analysis. All experimental procedures on animals were approved by the Monash University Committee for Ethics in Animal Experimentation.

Analytical methods. Commencing 1 d before tracheal obstruction(117 d), fetal tracheal pressures were recorded (Statham, P23ID) throughout the experimental period in the 10-d obstruct and 10-d control fetuses; amniotic sac pressure was electronically subtracted from the tracheal pressure before the differential pressure was displayed on a polygraph. Tracheal pressure was measured, during periods of fetal apnea, at 1-min intervals for 12 h during each day of the experimental period. Lung liquid volume and secretion rates were measured using an established method(11) of impermeant indicator dilution (blue dextran 2000, Pharmacia Chemical, Sweden). These measurements were made just before tracheal obstruction (d 0; 118 d) and then on d 1 (119 d), 2 (120 d), 4 (122 d), 7 (125 d), and 10 (128 d) after obstructing the trachea. In both the 10-d control and 10-d obstruct groups, measurements of lung liquid volume and secretion rates were made by connecting the distal tracheal catheter to a 100-mL glass syringe that was open to the atmosphere via a bacterial filter. Fetal lung liquid was then drained into the syringe before 2-5 mL (50 mg/mL) of indicator were added. The lung liquid was then alternately drained and returned to the lungs for 45 min to thoroughly mix the indicator. Samples of lung liquid (1-2 mL) were collected every 15 min over a 1.5-h period for determination of indicator concentration. The rate at which the indicator was diluted was used to measure the secretion rate(11). At the end of the measurement period, the liquid was returned to the lung lumen, and the syringe was disconnected from the tracheal catheter. The concentration of indicator in each lung liquid sample was measured using a multichannel absorbance meter(Titertek Multiscan, Flow Laboratories, UK), set at a wavelength of 620 nm.

Protein contents of the fetal lungs were determined using a standard protein assay (Bio-Rad, Australia). BSA standards were made up in distilled water and diluted to achieve concentrations of 100, 50, 25, 12.5, and 6.25μg/mL. The hydroxyproline(7, 12) and DNA contents and DNA synthesis rates(4) of fetal lung tissues were determined using established techniques. Estimates of fetal body weights at the time of the experiments were made from postmortem weights(13).

The data on pulmonary DNA and protein contents and DNA synthesis rates obtained from this study were combined with those of two previous studies conducted in our laboratory using either a 2-d(7) or a 7-d(4) tracheal obstruction period in fetal sheep at similar gestational ages (n = 5 for each group of control and tracheal obstructed fetuses). Tissues from these earlier experiments, which had been stored at -70°C, were reanalyzed for protein and DNA contents in the same assays as tissues from the present study to reduce interassay differences. The previous data were used to demonstrate the time course for the lung growth response to tracheal occlusion.

Statistical analysis. Data in the text are presented as mean± SEM. Differences in lung liquid secretion rates and volumes and tracheal pressures were analyzed by three-way ANOVA with treatment, gestational age, and animals as factors. When an interaction was found between treatment and gestational age, the data were analyzed by a two-way ANOVA with either gestational age and animals or treatment and animals as factors. Significant differences between values were then identified with a Student Newman-Keuls test. Lung DNA content and synthesis rates, protein and hydroxyproline contents, and protein-to-hydroxyproline ratios were analyzed using unpaired t tests. Relationships between the percent increases in lung DNA and protein content and the percent increases in lung liquid volume and tracheal pressure were calculated by linear regression analysis. The level of significance was p < 0.05 for all statistical analyses.

RESULTS

All fetuses were considered healthy according to their blood gas and acid base status before and throughout the experiments (pH, 7.36 ± 0.00; arterial Pco2, 45.9 ± 0.4 mm Hg; arterial Po2, 21.6± 0.5 mm Hg; oxygen saturation 67.3 ± 0.9%; total Hb, 9.1± 0.1 g/dL; hematocrit, 28.2 ± 0.3%). Body weights were not different between control and tracheal obstructed fetuses at 128 d of gestation (4-d control, 3.15 ± 0.30 kg; 4-d obstruct, 3.33 ± 0.11 kg; 10-d control, 2.70 ± 0.16 kg; and 10-d obstruct, 3.13 ± 0.16 kg).

Lung liquid volumes and secretion rates. The mean volume of lung liquid, adjusted for fetal body weight, did not change throughout the 10-d study period in control fetuses; it ranged between 29.7 and 33.8 mL/kg (Fig. 1a). Before tracheal obstruction (d 118), lung liquid volumes of control fetuses and fetuses with tracheal obstruction were similar (Fig. 1a). After 1 d of tracheal obstruction, lung liquid volumes increased to a mean of 55.0 ± 5.9 mL/kg, but there was no further increase at 2 d. The volumes then increased to 75.1 ± 5.7 mL/kg after 4 d of tracheal obstruction (Fig. 1a) and continued to increase until 7 d of tracheal obstruction (97.3 ± 15.2 mL/kg). After 10 d of tracheal obstruction, lung liquid volumes were not significantly different from those measured after 7 d of obstruction (Fig. 1a).

Figure 1
figure 1

The volume (a) and production rate(b) of lung liquid (mean ± SEM), adjusted for fetal body weight, at 118, 119, 120, 122, 125, and 128 d of gestation in control fetuses(open circles, control) and fetuses subjected to tracheal obstruction (solid circles, 10-d obstruct). The black bar represents the period of tracheal obstruction. Values that do not share a common letter are significantly different (p < 0.05) from each other.

Full size image

Adjusted for fetal body weight, lung liquid secretion rates in control fetuses remained constant throughout the 10-d study period (Fig. 1b). Before tracheal obstruction (118 d), lung liquid secretion rates were similar in both groups. After 1 d of tracheal obstruction, lung liquid secretion rates decreased to undetectable levels and remained so throughout the period of obstruction (Fig. 1b).

Tracheal pressure. In control fetuses, tracheal pressures(relative to amniotic sac pressure) ranged between 1.1 and 2.8 mm Hg throughout the experimental period (Fig. 2). On the day before tracheal obstruction (117 d of gestation) the mean tracheal pressure in the 10-d obstruct group was not different to that in the 10-d control group. However, within 1 d of tracheal obstruction, the mean tracheal pressure increased to 4.3 ± 0.4 mm Hg and remained above control values for the duration of the obstruction period (Fig. 2). There was no further significant increase in tracheal pressure after 1 d of tracheal obstruction.

Figure 2
figure 2

Tracheal pressure, from which amniotic sac pressure has been subtracted (mean ± SEM), in control fetuses (open circles) and fetuses which were exposed to 10 d of tracheal obstruction(solid circles). The black bar represents the period of tracheal obstruction. Values with different letters are significantly different (p < 0.05) from each other.

Full size image

Lung weights and total protein contents. Compared with control values, wet lung weights (adjusted for fetal weight) were increased by 53% after 4 d of tracheal obstruction and by 111% after 10 d of tracheal obstruction (Table 1). Similarly, total lung protein content, adjusted for fetal body weight (Table 1), was increased to 47% above control values after 4 d of tracheal obstruction and to 56% above control values after 10 d of tracheal obstruction(Table 1).

Table 1 Wet weight (g/kg), DNA, protein, and hydroxyproline contents (mg/kg of fetal body weight), DNA synthesis rates(dpm/mg DNA), and hydroxyproline-to-protein ratios of lung tissue collected from fetuses exposed to: 1) unimpeded tracheal flow for 4 d (4-d control), 2) obstruction of the fetal trachea for 4 d (4-d obstruct); 3) unimpeded tracheal flow for 10 d (10-d control), and 4) obstruction of the fetal trachea for 10 d(10-d obstruct)
Full size table

Total lung DNA content and DNA synthesis rates. When adjusted for fetal body weights, total lung DNA contents were increased to 39% above control values after 4 d of tracheal obstruction (Table 1). After 10 d of tracheal obstruction, total lung DNA contents were increased to 79% above control values (Table 1). The rate of [3H]thymidine incorporation into DNA was increased by 66% above control values after 4 d of tracheal obstruction, but was not different from control levels after 10 d of tracheal obstruction (Table 1).

Lung hydroxyproline content. Hydroxyproline content of the lung, adjusted for fetal body weight, tended to increase but was not significantly different from control values after 4 d of tracheal obstruction(Table 1). However, the hydroxyproline content of the lung had increased by 70% above control values after 10 d of tracheal obstruction(Table 1). The hydroxyproline-to-protein ratio was not significantly different from control values after either 4 or 10 d of tracheal obstruction (Table 1).

Relationship between changes in lung DNA and protein content and lung liquid volumes. Using the data on DNA and protein contents obtained in this study, in combination with our previous data(4, 7), we have related the time course of the percentage increase in lung liquid volume with the increases in DNA and protein contents at each time point (2, 4, 7, and 10 d; Fig. 3a). The percentage increase in lung DNA content measured after obstructing the fetal trachea was found to be significantly correlated with the percentage increase in lung liquid volume (r = 0.99, n = 4; Fig. 4a). Correlations were not found between the percentage increase in protein content and the percentage increase in lung liquid volume (r = 0.36,n = 4), nor between the percentage increases in either DNA or protein contents and the percentage increase in tracheal pressure (Fig. 4b). The increase in protein content was similar at 2, 4, 7, and 10 d of tracheal obstruction. In contrast, the percentage increase in DNA content increased linearly between d 2 and 7 of tracheal obstruction, but no further increase had occurred by d 10(Fig. 3a). After 2 d of tracheal obstruction, DNA synthesis rates were increased by 777% above control values but fell to 66% above control values after 4 d of obstruction. Although DNA synthesis rates tended to be elevated after 10 d of tracheal obstruction, they were not significantly different from control values.

Figure 3
figure 3

Time course for percentage increases, above control values, in (a) pulmonary protein (squares) and DNA(circles) contents and (b) DNA synthesis rates after 2, 4, 7, and 10 d of tracheal obstruction. Each symbol represents five control and five experimental fetuses, except for the 10-d value, which was obtained from nine experimental and five control fetuses in (a) and four experimental and five control fetuses in (b). Values obtained from our previous studies(4, 7) are indicated by open symbols. In (a), values with different letters are significantly different (p < 0.05) from each other. In(b), asterisks indicate a significant increase above control values; the values at 4 and 7 d are significantly increased, although this fact is obscured by the large range of the y axis.

Full size image
Figure 4
figure 4

Relationship between (a) the percentage increase in lung DNA content and lung liquid volume (r = 0.99,p = 0.002, n = 4 time points) and (b) the percentage increase in tracheal pressure and DNA content between d 2 and 10 of fetal tracheal obstruction. In total, the mean percentage increase in DNA contents above control values were obtained from 44 fetuses, whereas the mean percentage increases in lung liquid volume and tracheal pressure were obtained from 10 of these fetuses.

Full size image

DISCUSSION

The results of our study indicate that increases in lung liquid volume are likely to be an important factor in mediating the lung growth response to tracheal occlusion and that the pattern of growth follows a specific time course. The time course we have identified indicates that future studies directed toward identifying cellular and molecular mechanisms involved should focus on the relatively narrow time period when the growth response is at a maximum.

Obstructing the fetal trachea prevents lung liquid efflux, thereby causing secreted liquid to accumulate within the future airways. Lung liquid volumes were increased by 80% within 1 d of obstructing the trachea, but were not increased further between d 1 and 2, confirming our previous finding(7). However, after 2 d of tracheal obstruction, lung liquid volumes increased linearly, reaching maximal values by d 7. Similarly, the percentage increase in lung DNA content was linear between d 2 and 7 and had also reached a maximal value by d 7. Our finding of a highly significant correlation (r = 0.99) between the increase in fetal lung DNA content and the increase in lung liquid volume after tracheal obstruction suggests that these two parameters are closely related. In contrast, the changes in tracheal pressure between d 2 and 7 of tracheal obstruction were not significantly related to the increase in either lung liquid volume or total DNA and protein content of the lungs. These findings suggest that luminal volume changes are an important determinant of the acceleration in fetal lung growth induced by tracheal obstruction and that the dynamics of the relationship between lung liquid volume and tracheal pressure change during tracheal obstruction.

It is possible that the increase in lung liquid volume between d 2 and 7 of tracheal obstruction is a consequence (due to an increase in lung size), rather than a cause, of the acceleration in fetal lung growth. For instance, it is possible that the associated increase in intraluminal pressure is the essential factor. However, we consider this to be unlikely because the increase in lung DNA content was related to the increase lung liquid volume and not to the increase in intraluminal pressure, and both the increase in lung DNA content and lung liquid volume stopped at d 7 of tracheal obstruction, whereas the intraluminal pressure remained elevated at this time. Furthermore, when fetal lung growth is reduced as a result of oligohydramnios, the reduction in growth is accompanied by an increase in intraluminal pressure, whereas the volume of lung liquid is reduced(14, 15). This indicates that, unlike an increase in lung liquid volume, an increase in intraluminal pressure is not always associated with an increase in fetal lung growth. Thus, we suggest that the acceleration in fetal lung growth induced by tracheal occlusion is primarily stimulated by the increase in lung expansion and stops only because the lung is not able to expand further after 7 d, despite a sustained elevation in intraluminal pressure. However, further investigations are required to examine the relative contributions of these parameters in the lung growth response to tracheal occlusion as they are closely related. Indeed, it is possible that a reduction or abolition of the transpulmonary pressure gradient is an important factor in the cessation of lung growth at this time, although potential mechanisms are unknown.

If the increase in luminal volume is the principal determinant of the lung growth response to tracheal occlusion, it is possible that the growth rate is regulated by the rate at which the fetal lung expands with accumulated fluid. We suggest that, within the 1st d of obstructing the fetal trachea, the lungs expand to an initial limit. Indeed, the finding that lung liquid volumes do not increase significantly between d 1 and 2 after obstructing the trachea indicates that a limit to further expansion may have been reached after 1 d. Factors that could limit the increase in lung expansion at this time include 1) a lack of available space within the chest cavity or 2) a structural limit imposed by lung tissue. After d 2, intrathoracic space could be increased by the gradual flattening out and eventual eversion of the diaphragm(4). On the other hand, it is possible that intrathoracic space is not the limiting factor initially (at d 1 and 2) and that the lung expands to a limit imposed by the structural components of lung tissue (e.g. collagen and elastin). Thus, the subsequent expansion of the lung between 2 and 7 d, therefore, may result from restructuring of the extracellular matrix. Indeed, the progressive increase in lung liquid volume without a continued increase in tracheal pressure between 2 and 7 d of tracheal obstruction suggests that lung compliance must increase over this time, possibly due to restructuring of the extracellular matrix.

The inability of the fetal lungs to continue to expand after 7 d of tracheal occlusion most probably results from physical limitations imposed by the chest wall. If so, after 7 d of tracheal obstruction, further expansion of the fetal lungs must depend upon increasing the amount of intrathoracic space available to the lung. Such an increase could result from continued somatic, and hence rib cage growth, which may explain why fetal lung growth(i.e. DNA content and synthesis rates) appears to proceed at control levels after 7 d of tracheal occlusion. That is, further expansion and thus further growth of the lung is dependent upon the growth rate of the chest wall.

In the present study we found that lung DNA synthesis rates, as measured by incorporation of [3H]thymidine into DNA, were elevated at 4 d (66%) but were not different from control values after 10 d of tracheal obstruction. We have previously found that DNA synthesis rates were increased by 776% and 54% after 2 and 7 d of tracheal obstruction, respectively(4, 7). The maintenance of elevated DNA synthesis rates up to, and including, the 7th d of tracheal obstruction explains the continued increase in DNA content over this time. In addition, the time course of the changes in DNA synthesis rates indicates that the cellular mechanisms responsible for the acceleration in lung growth are most active during the first 2 d. We cannot readily explain the discrepancy between the time course of the increases in lung DNA synthesis rates and lung DNA content. However, it may relate to the very large difference in sensitivity between these two measurements of tissue growth, combined with the low levels of DNA synthesis rates relative to the total amount of DNA present in the lung at that time. Thus, despite the very large percentage increase in DNA synthesis rates observed at 2 d, the number of cells undergoing division over an 8-h period is relatively small compared with the total number of cells present within the lung. Consequently, it must take some time before a change in cell number could be detected in measurements of DNA content.

The cellular mechanisms whereby increased fetal lung expansion leads to increased rates of lung tissue growth are unknown. Increased release of tissue growth factors is one possibility, which is supported by our finding that 7 d of tracheal obstruction increased IGF-II mRNA levels in lung tissue(4). Other growth factors have also been found within the developing fetal lung, including vascular endothelial growth factor(16), platelet-derived growth factor(17), and epidermal growth factor(18), all of which may be involved in the growth response. Whatever the mechanisms, they would appear to be most active in stimulating DNA synthesis within the first 2 d of tracheal obstruction and are unlikely to be still active after 7 d. This finding may explain the recent observation that 30 d of tracheal obstruction failed to increase IGF-II mRNA levels in the fetal lung(19).

The maximum percentage increase in lung protein content occurred within 2 d of tracheal obstruction, whereas 7 d were required to increase DNA contents to maximal levels; thus, cellular hypertrophy apparently precedes hyperplasia. We have previously shown that lung collagen content was not increased after 2 d of tracheal obstruction(7). Thus, we suggested that structural remodeling of the lung, induced by tracheal obstruction, resulted from increased tissue growth without increased synthesis of the major structural components of the lung such as collagen(7). However, our finding that the hydroxyproline-protein ratio had returned to control levels after 7 d of tracheal obstruction indicates that lung hydroxyproline content increased over the same time period as the maximum increase in lung DNA content (between 2 and 7 d). Thus, increased synthesis of collagen occurs in parallel with the increase in DNA content and may be an important component of the growth response to tracheal occlusion; it is possible that restructuring of the extracellular matrix may lead to an increase in lung compliance, allowing a greater accumulation of lung liquid.

Lung liquid production rates were reduced to undetectable levels within 1 d of tracheal obstruction and remained reduced throughout the obstruction period. Although lung liquid production rates remained below detectable levels, liquid must have continued to accumulate within the lung at ≈0.2 mL/h/kg to explain the increase in lung liquid volume between d 2 and 7 of tracheal obstruction. The virtual cessation of lung liquid production supports our previous suggestion(3, 7) that changes in production can result from alterations in the hydrostatic pressure within the lung lumen. The current model for lung liquid secretion is based on the active transport of chloride ions from fetal plasma into the lung lumen, which establishes an osmotic pressure gradient promoting the movement of water into the lung(8). However, lung liquid production rates are likely to be determined by the balance of all forces acting across the pulmonary epithelium. Under normal conditions, the osmotic gradient across the pulmonary epithelium must exceed the opposing hydrostatic pressure gradient(by 1-2 mm Hg) leading to net lung liquid production. It is likely that, after tracheal obstruction, intraluminal pressure increases until the hydrostatic pressure required to expand the lung further (4-5 mm Hg) equals, and thereby counterbalances, the osmotic pressure driving liquid movement into the lung lumen. This concept may have important implications for the therapeutic use of tracheal obstruction in reversing lung growth deficits in human fetuses. For example, we have previously shown that growth of hypoplastic fetal lungs is dependent upon their reexpansion by liquid(3). If, however, the pressure required to inflate these small lungs is greater than the osmotic pressure driving lung liquid secretion, the lungs will not expand and lung growth will not be stimulated after tracheal occlusion.

We conclude that the fetal lung growth response to tracheal obstruction is activated within 2 d and is completed within 7 d, after which time the rate of lung growth returns to control levels. Our data indicate that the molecular and cellular mechanisms responsible for the acceleration in lung growth are primarily activated by distension of lung tissue, rather than an increase in intraluminal pressure per se, and are most active within the first 2 d. Furthermore, because the increase in fetal lung growth induced by tracheal occlusion is strongly related to the increase in lung liquid volume, increases in luminal volume may be a good indicator of the degree of lung growth that has occurred after obstruction of the fetal trachea.