Fetal tracheal occlusion (TO) reverses lung hypoplasia by inducing rapid lung growth. Although increases in lung size accompanied by increased numbers of alveoli and capillaries have been reported, effects of TO on lung development have not been formally assessed. In the present study, the objective was to verify our prediction that the main effect of TO would be to accelerate fetal lung development. We have developed and characterized a new fetal mouse model of TO to best realize this goal. At embryonic day 16.5, pregnant CD1 mice were operated under general anesthesia. One fetus per dam was selected to undergo surgical TO with a surgical clip or a sham operation. The fetuses were delivered 24 or 36 h postsurgery. The maturation of lung parenchyma, evaluated by counting the generations of alveolar saccules from the terminal bronchiole to the pleura, was significantly accelerated in the TO group with a complexity of the gas exchange region comparable with postnatal days 1 and 3after 24 or 36 h of TO. Cellular proliferation and apoptosis peaks, assessed by immunohistochemistry directed against PCNA and the active form of caspase-3, were significantly increased 24 h after surgery in the TO group compared with the sham group. However, in situ hybridization showed no significant difference in the density of type II pneumocytes expressing surfactant protein C mRNA. Our results show that brief TO during late gestation in fetal mice induces accelerated lung development with minimal effects on surfactant protein C mRNA expression.
- proliferating cell nuclear antigen
- congenital diaphragmatic hernia
- mouse model
- fetal therapy
tracheal occlusion(TO) during fetal life induces rapid lung growth (2). TO has been shown to prevent or reverse the pulmonary hypoplasia associated with various experimental and pathological conditions such as chronic amniotic fluid loss (leading to severe oligohydramnios), bilateral nephrectomy, lung liquid drainage, and, most recently, congenital diaphragmatic hernia (CDH) (4, 9, 16, 17, 24,49).
Experimental TO in fetal lambs produces striking changes in lung size and architecture (2, 22, 24, 35, 49). In a study by Alcorn et al. (2), the lung weight-to-body weight ratio (LW/BW) nearly doubled after 3–4 wk of TO; in a study by Wilson et al. (49), the LW/BW and alveolar surface area were greater by fourfold over normal controls after 1 mo of TO; the general morphology and histology of these lungs appeared normal (22). Moreover, a study reported that pulmonary vasculature develops harmoniously with the alveoli (17); however, we, as well as others (14, 20, 43), have shown that TO of normal fetal lambs decreases the number and function of type II pneumocytes, the cells that synthesize and secrete surfactant proteins. Recent evidence from lamb (30) and rabbit models (12) suggests that TO in late gestation does not have a deleterious effect on type II pneumocytes.
The effects of TO appear rapidly: 6 days of TO were sufficient to reverse preexisting lung hypoplasia in fetal sheep (37). The acceleration of lung growth, as measured by the rate of DNA synthesis, was observed after only 2 days (eightfold increase in [3H]thymidine incorporation). This acceleration dropped as early as 4 days and disappeared completely by 10 (38, 39).
The effect of lung distension on lung growth appears to be unique to the developing lungs (40). TO affects lung growth during the canalicular and saccular stage, a critical period of lung development when the functional gas exchange unit appears (47). This period is characterized not only by changes in lung volume, but by dramatic changes in lung architecture as well, including septation of the pseudoalveoli, thinning of the interstitial tissue, and microvasculature development (27). Moreover, the organ changes from very dense cellular tissue to mostly liquid-filled future air spaces. Thus proliferation indexes and changes in dry lung weight taken at this time are particularly confounded by apoptosis and decreased cellular density, respectively (processes integral to this developmental phase) (15, 29,46). Surprisingly, the only effects of TO assessed so far have been on lung growth, with concomitant effects on lung development being largely ignored. However, the increase in alveolar number (22,43) and the decrease in mean terminal bronchiole density (8, 9, 19, 26) associated with TO in fetal sheep suggest that TO does indeed have an accelerating effect on development as well.
The goal of the present study was to test the prediction that the most important effect of TO on the fetal lung is in fact acceleration of lung development. The study was also used as an opportunity to develop and characterize a fetal model of the mouse, a species in which the pattern of lung development is well established (27, 47).
MATERIALS AND METHODS
Animals and surgical procedures.
The procedure described here was developed in our laboratory with classic trial and error to give optimal results while remaining representative of the period in which the intervention is performed in human fetuses. All surgical procedures were approved by the Committee for Animal Protection of Laval University and complied withFederal Guidelines for the Care and Use of Laboratory Animals (www.ccac.ca). Pregnant CD1 mice dated at 16.5 days of gestation were used (early canalicular stage) (47). We chose CD1 mice because they are well characterized, genetically stable, and readily available. An overview of the fetal surgery is illustrated in Fig.1. After intraperitoneal injection of 3 μg of buprenorphine (analgesic), the mice underwent general anesthesia with isoflurane. A heating pad was used to reduce heat loss during the surgery. After prepping the abdomen, we performed a median laparotomy and exposed one uterine horn. A 6.0 silk suture was passed through the uterine wall and the amniotic membranes and looped around the head of a fetus to limit the amniotic fluid lost during the surgery. The head and the neck of the fetus were pulled out by a small hysterotomy, and the mouth and nose were immediately covered with gauze soaked in saline to prevent spontaneous air breathing and to keep the head hyperextended. The trachea was exposed through careful dissection of the neck under a stereoscopic zoom microscope at ×10 magnification. A 6.0 silk suture was slipped under the trachea, and the trachea was gently lifted. The trachea was occluded with a small surgical ligating clip (Horizon; Weck Closure Systems, Research Triangle Park, NC). We carefully examined the neck to assess the correct position of the clip and to rule out damage to the trachea or the great vessels of the neck. The head and the neck of the fetuses were put back into the uterus, and the uterine wall was closed with 6.0 silk suture. It is essential at this step to avoid any excessive pressure on the head. One drop of ritodrine (5 mg/ml) was applied to the uterus to minimize the risk of preterm labor. The peritoneum and skin were closed separately with 4.0 silk. Ringer lactate solution (0.5 ml sc) was injected to rehydrate the mice. The dam received progesterone (40 μg of Depo-Provera) ondays −1, 0, and +1 to prevent premature delivery. The mice were kept under close postsurgical observation. We operated on only one fetus per dam. Sham-operated fetuses were used as controls and underwent the same surgery without the clip. We used untouched, 17.5-day-old fetuses to assess the effect of the sham surgery on lung growth to ensure that oligohydramnios from amniotic fluid loss was not the cause of a pulmonary hypoplasia (1-3, 34, 50). In addition, intact fetuses atdays 16.5, 17.5, and 18.5 of gestation and newborns at the day of birth (postnatal day 1) and the second and third days of life (postnatal days 2 and3) were used to build an atlas of normal perinatal lung development to quantify the effect of TO on lung development.
The mice were killed by cervical dislocation under anesthesia with isoflurane 24 or 36 h after the surgery (corresponding to 17.5 and 18 days of gestation), and the operated-on fetuses were delivered by cesarean section, put on ice, and weighed. At delivery, three criteria were used to assess the fetus: the presence of heart beat, pink skin color, and spontaneous movements. The fetuses had to be positive for all three criteria to be retained for future analysis. In our experience, significant damage to the neck and the head prevents survival of fetus. Under the stereomicroscope, a median sternotomy was done, the lungs were excised, and the trachea was dissected below the clip in the TO group. The excess lung fluid was drained spontaneously, and the lungs were weighed. Lung growth was assessed by the wet LW/BW.
It was assumed that the delay needed to weigh the lungs was sufficient to equilibrate the intraluminal pressure with the atmospheric pressure. Lungs were fixed in 10% formalin for 24 h and embedded in paraffin, and 5-μm sections were sliced. Because the mouse fetus is so small, it is impossible to fix the lungs by intratracheal instillation under constant pressure; neither is vacuum insufflation (48) to obtain uniform inflation helpful, because fetal lungs are not filled with air. On the other hand, the presence of liquid within the lung generally prevents the collapse of the tissue during fixation.
Morphometric studies were done in midsagittal sections of the lungs stained with hematoxylin and eosin. It was not possible to use the radial alveolar count or to have a reliable estimate of the alveolar surface because the degree of lung distension at fixation affects these measurements (22). We adapted a previously reported method to assess the development of lung parenchyma (48, 50) and counted the generation of alveolar saccules from the terminal bronchiole to the pleura. An example of the method is illustrated in Fig. 3. The generation of alveolar saccules was not assessed in three dimensions (on multiple section of a same lung); therefore the value is a relative and not absolute number. The air space-tissue fraction was evaluated by the point-counting method (32) and was used to correct the density of cells expressing surfactant protein (SP)-C and the lung wet weight for the higher size of the saccules and alveoli in the TO group.
We processed tissue sections using the avidin-biotin bridge method with peroxide as a substrate. Immunohistochemical localization of proliferating cell nuclear antigen (PCNA) and the active form of caspase-3 protein was used to assess cellular division and apoptosis, respectively, according to a modified, previously reported method (11). The active form of caspase-3 was selected to assess apoptosis for its central role in apoptotic pathways (membrane and mitochondrial); moreover, unlike terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling (TUNEL) assay, caspase-3 labels apoptotic events leading to cell death, but not necrosis (21, 36, 42). Blocking serum used for PCNA was 1× Tris-buffered saline (TBS) containing 1.5% (vol/vol) normal goat serum and 1% (wt/vol) BSA; blocking serum used for active caspase-3 was 1× TBS containing 1.5% (vol/vol) normal goat serum and 0.5% (wt/vol) BSA. Incubation time in blocking serum was 45 and 30 min for PCNA and active caspase-3, respectively. Primary antibodies used were anti-PCNA (mouse monoclonal clone PC-10 antibody, no. 1 486 772; Roche, Laval, QC, Canada) and antiactive caspase-3 (rabbit polyclonal antibody, no. 557038; BD PharMingen, Mississauga, ON, Canada), diluted at 2.5 and 15 μg/ml, respectively, in blocking serum. Secondary antibodies were anti-mouse IgG-B (sc-2039 at 1.5 μl/ml; Santa Cruz) for PCNA and anti-rabbit IgG (5 μl/ml; Vector Laboratories) for active caspase-3. Antibody binding was revealed with Vectastain ABC Elite kit (Vector Laboratories), and the immunohistochemical reaction was visualized by incubation with 3,3′-diaminobenzidine (D-5637; Sigma). Two control sections for each antibody were treated identically, except that in the first the primary antibody was replaced by blocking serum, and in the other the secondary antibody was replaced by blocking serum. Counterstaining was performed with hematoxylin (Gill's formulation no. 2; Fisher Scientific).
Semiquantitative assessment of positively stained cells in each tissue compartment was done for the two primary antibodies. The proportion of positively stained cells was assessed on a scale of 0 to ++++. The analysis was done in duplicate by a person blinded to the experimental conditions. Moreover, the slides were randomly mixed for the analysis. For the TO fetuses, 10 and 5 mice were used from the 24- and 36-h time points, respectively; for the sham group, 7 and 4 mice were used from 24 and 36 h, respectively.
In situ hybridization.
In situ hybridization was used to assess specific cell populations. SP-C mRNA is generally accepted as a reliable marker of differentiated type II pneumocytes (23, 25). We performed in situ hybridization as previously described using [3H]riboprobe (5, 26, 43). Slides were coated in NBT-2 photographic emulsion (Kodak, Rochester, NY) stored at 4°C for 7 days until developed and counterstained with hematoxylin and eosin, before photomicrography. We evaluated nonspecific binding and background using sense probes on three sections of the TO group. For TO and sham, one section per animal was analyzed. We assessed the density of cells expressing SP-C by counting the number of positive cells per field on three or four randomly selected fields of each section depending on lung size. Cells were considered positive for SP-C if five or more silver grains were observed over the cell. To correct for confounding effects of the changes in alveolar saccule size (induced either by the TO or tissue fixation), we corrected the data for the tissue-air space fraction. The density of cells expressing SP-C was expressed as the number of positive cells per cm2 of tissue (excluding air space).
We analyzed data by ANOVA using a factorial test except for the data comparing the untouched group with the sham group, which were analyzed by an unpaired t-test. Statistical significance was accepted at a P value < 0.05. Values were expressed as the means ± SE. Statistical differences between sham and TO groups for immunohistochemistry directed against PCNA and the active form of caspase-3 were tested with nonparametric Mann-WhitneyU-test. Statistical significance was accepted at tiedP value < 0.05.
Shams vs. untouched fetuses.
Sham-operated fetuses were compared with untouched twins to show sham surgery does not cause significant lung hypoplasia. No significant differences were observed in the LW/BW ratio (2.58 ± 0.07%,n = 10 vs. 2.74 ± 0.10%, n = 10), in the air space-tissue fraction (19 ± 2%,n = 9 vs. 18 ± 1%, n = 13), or in the generation of alveolar saccules (2.5 ± 0.2,n = 9 vs. 2.3 ± 0.2, n = 13) between the sham and untouched groups. On the basis of the absence of deleterious effects of the sham surgery on lung development, we deemed the sham group a suitable control for the remaining experiments.
Survival and success rate of TO.
A total of 73 pregnant mice underwent surgery. Of the 47 mice planned for the 24-h time-point experiment, five mice aborted before scheduled delivery. Of the 42 remaining, 21 had surviving operated fetus (11 TO and 10 sham, for a success rate of 44.7%). Of the 26 mice planned for the 36-h time point, five mice died from hypothermia or anesthetic complications during the surgery due to technical complications that were eventually rectified. Of the 21 remaining, nine had surviving operated fetus (five TO and four sham, for a success rate of 42.8%).
Body and lung wet weights.
The results are summarized in Table 1. The wet body weights of fetuses in the TO and sham groups were comparable for the two time points studied, whereas the wet lung weights and the LW/BW ratios were significantly greater in the TO group. There was a 37% increase in the relative lung weight between the sham group and the TO group after 24 h and a 102% increase between the two groups after 36 h.
Morphology of the lungs.
Microscopy clearly showed the dramatic effect of TO on lung morphology. The TO lungs appeared more developed compared with the sham group. We observed a significant increase in the air space-tissue fraction (Fig.2) and in the complexity of the gas exchange region confirmed by the highest number of generations of alveolar saccules (Table 2 and Fig.3). To compare the effect of TO with normal lung development, we measured the generation of alveolar saccules on days 16.5, 17.5, and 18.5of gestation, and on postnatal days 1, 2, and3 (Fig. 4). The linear progression of the generations of alveolar saccules between day 16.5 of gestation and postnatal day 2 confirmed the validity of the method to assess lung development. The comparison of the values of generation of alveolar saccules between intact fetuses and fetuses with 24 or 36 h of TO corroborated the positive effect of TO on lung development. As shown in Fig. 4, after 24 h of TO, the generation of alveolar saccules was comparable with the developmental stage of postnatal day 1 or 2. After 36 h, the developmental stage was superior topostnatal day 3. A short 24 h of TO resulted in lung development corresponding to >72 h of normal development or >200% acceleration in the kinetic of lung development. Furthermore, the interalveolar tissue thickness and cellularity appeared to decrease in the TO group (Fig. 2). However, representative measurement of the interalveolar thickness was not feasible because this variable is liable to change upon lung distension during chemical fixation.
Confounding effect of changes in the air space-tissue fraction upon increased LW/BW.
Previous studies in sheep have shown that the wet lung weight is reliable for assessing lung growth following TO (26). The caveat here is that the volume destined to become air space is filled with liquid in the fetal lung; thus changes in the air space-tissue fraction can contribute to changes in the wet lung weight. Given the labor-intensive procedure and the small quantity of lung tissue in a fetal mouse, measuring dry-weight lung was impractical. To assess the effect of the increase in liquid content (future air space) of the TO group, we corrected the wet weight of the lung tissue for the change in the air space-tissue fraction (%tissue multiplied by the lung wet weight, assuming a density of 1 g/cm3). After 24 h of TO, the tissue weight corrected for the air space-tissue fraction was no more different from the tissue weight of the shams (22.0 ± 1.8 mg, n = 9 vs. 18.7 ± 1.0 mg, n = 9; P = 0.21). However, after 36 h, corrected lung weight of the TO group was significantly higher than the lung weight of the sham group (30.5 ± 3.9 mg, n = 5 vs. 21.4 ± 2.1 mg, n = 4; P = 0.02).
Cellular proliferation and apoptosis.
Figure 5 shows an example and Table3 summarizes the results of the immunohistochemistry assay against PCNA and the active form of caspase-3. In addition to the results shown in Table 3 for the airway epithelium, the alveolar epithelium, and the interstitium, we also assessed the underlying cell layer of the airway epithelium, the endothelium, and muscle layer of the blood vessels. Because TO did not modify the staining of cells in any of these structures, results are not shown. Furthermore, negative slides where the first or the second antibody was omitted showed low background without positive cells for the immunohistochemistry assay against PCNA and no background for the active form of caspase-3 (data not shown).
We performed PCNA-specific immunohistochemistry to evaluate the cellular proliferation in sham-operated and TO groups. In both groups and both time points, the percentage of positively stained cells was higher in the lung periphery than in the central lung (Table 3). In the sham lungs, the proportion of positively stained cells was higher at 36 h than at 24 h, showing that an increase in proliferation rate is part of normal lung development. In the TO lungs, the proportion of positively stained cells was higher at 24 h than at 36 h, showing that, as in sheep, the increased proliferation induced by TO is an early and limited phenomenon (38, 39). When TO was compared with the sham group, a significant difference was observed at 24 h with more positive cells in the TO group in the airway and alveolar epithelia of the lung periphery (tied Pvalues of 0.02 and 0.03). Furthermore, a clear difference was observed at 36 h with fewer positive cells in the TO group in the interstitium, airway epithelium, and alveolar epithelium both in lung periphery and in the central lung (tied P values ≤ 0.05 in all structures). Therefore, the rate of proliferation increases earlier in TO lungs than in sham lungs; however, this phenomenon seems to be time limited.
We performed immunohistochemistry specific for active caspase-3 to assess the extent of the apoptosis in the sham-operated and TO groups. As for PCNA, positively stained cells were higher in the lung periphery than in the central lung both in sham and TO groups (Table3). In the sham lungs, the proportion of positively stained cells was higher at 36 h than at 24 h, showing, as reported by others, that increased apoptosis is part of normal lung development (15, 29, 46). In the TO group, the proportion of stained cells was comparable at 24 and 36 h. When TO was compared with the sham group, clear differences were observed at 24 h: TO strongly increased positively stained cells compared with shams both in lung periphery and in the central lung (tied P value ≤ 0.01 in all structures). However, at 36 h, the difference between sham and TO groups disappeared. This shows that the rate of apoptosis increases earlier in the occluded group than in the sham group. Thus TO precipitates the apoptotic process of normal lung development.
In situ hybridization.
Type II pneumocyte density and function were assessed by in situ hybridization of SP-C mRNA. For the two time points, SP-C mRNA expression appears to slightly decrease in the TO group (Fig.6). However, when the density of cells expressing SP-C was corrected for the tissue-air space fraction, the two groups were not different: 201 ± 26 compared with 199 ± 18 (mean/cm2) for the sham and ligated after 24 h, and 200 ± 30 compared with 255 ± 28 for the sham and ligated after 36 h. Moreover, we did not observe significant difference in the density of SP-C-positive cells between 24 and 36 h in either the sham or TO groups. In addition, the relative amount of SP-C mRNA per positive cell (number of grain per positive cell) did not vary with the treatment and the time (data not shown). Hybridization with sense RNA probe showed low background without positive cell (data not shown).
Our results show that fetal TO induces a remarkable acceleration in lung development: 24 h of TO recovers at least 3 days of normal lung development, and 36 h, more than 5 days. In our model, the effect of TO is not associated with a decrease in type II pneumocytes expressing SP-C mRNA. Moreover, we can speculate that the majority of the lung growth observed after 24 h of TO, as measured by the increase in wet lung weight, is generated by the increase in liquid content within the future air space. This also concurs with the increased rate of apoptosis in the TO group at 24 h and is consistent with the fact that the mature lung is an organ with a very low cellular density. However, at 36 h, the significant increase of the corrected tissue weight in the TO group confirms that TO-induced lung development is accompanied by lung growth, as is the case in normal lung development.
In addition, this report is the first to show feasible, in vivo fetal TO in mice. We did not observe any adverse effects of the sham surgery on lung growth. Our surgical method performed a short time before the cesarean section prevented any significant lung hypoplasia that could have been caused by oligohydramnios due to amniotic fluid loss during the surgery. Therefore, under these conditions, the sham operation seems to be the most appropriate control.
The timing of the intervention was assessed experimentally. We operated on CD1 pregnant mice at 16.5 days of gestation (which corresponds to the early canalicular stage). Earlier intervention was unsuccessful because of the small size and fragility of the fetal trachea. Moreover, earlier intervention may not be appropriate; as in the fetal lamb, early intervention was associated with an abnormal pattern of lung growth (44). Later intervention was not possible either because the majority of the mice fetuses had spontaneous air-breathing movement; air breathing at 16.5 or 17.5 days was invariably associated with fetal death upon return of the fetus into the uterus. However, in human trial for fetuses with CDH, TO was done during the canalicular stage (16–26 wk); thus our model is clinically relevant.
Clearly, TO in fetal mice is limited by the small size of the fetus and spare amount of lung tissue. The surgery is labor intensive, and operations are limited to not more than one fetus per dam, as interventions lasting longer than 60 min are associated with a marked increased in abortion. Thus it is important to carefully plan the use of lung tissue samples to reduce waste. Therefore, using tissue sections has an advantage over using total lungs. However, the small size of the organ and the presence of ligation on the trachea preclude fixation by instillation under constant pressure or by insufflation under negative pressure (48); thus the usual tools to study alveolar development are not as available. However, our results show that the generation of alveolar saccules, which is independent of lung distension at fixation, is a reliable way to study lung development from 16.5 days of gestation to postnatal day 3.
Numerous studies have shown increased cellular proliferation in the lungs after TO (10, 12, 13, 39), but very few have compared the time course of the response with the normal lung development. In the fetal lamb, maximum cellular proliferation (measured by the DNA synthesis rate) was observed after only 2 days of TO as an eightfold increase in [3H]thymidine incorporation. This acceleration was already markedly decreased byday 4 and had disappeared altogether by day 10(38, 39). Our results suggest that, in the mouse, increased cellular proliferation is part of normal saccular development; TO accelerates the apparition of the peak of proliferation.
The major constituent of a mature lung is air space. During the last phase of gestation, most of the interstitium disappears to leave the space for mature alveoli with very thin stroma between air spaces. Very few studies have looked at the effect of TO on apoptosis, which plays a crucial role in normal lung development (15, 29). Our results with the sham-operated fetuses confirm that apoptosis occurs in a large number of interstitial cells in the periphery of the lung during the early saccular stage. In the present study, we have clearly shown that TO in fetal mice promotes apoptosis within 24 h post-TO, suggesting that TO speeds up lung development.
The active form of caspase-3 was selected to assess apoptosis instead of the TUNEL assay because caspase-3 labels apoptotic events leading to cell death, but not necrosis (21, 36,42). However, we observed a much higher number of apoptotic cells in our study compared with those using the TUNEL method (29, 46), both in the TO and sham groups. This can be explained by the fact that the peak of induction of active caspase-3 preceded the peak of DNA fragmentation, as demonstrated in neuronal death induced by permanent middle cerebral artery occlusion (45). Indeed, this study showed a peak in the induction of caspase-3 at 8 h, whereas TUNEL labeling was maximal at 24 h. It is therefore inaccurate to compare cells undergoing apoptosis as assessed by either of these two methods concurrently. Furthermore, caspase-3 activation is sustained during the apoptotic process. In this way, positively stained cells for active caspase-3 include the majority of TUNEL-positive cells but not the opposite. This could very well explain the much higher number of apoptotic cells found in our study. Moreover, protocols for the immunostaining were adapted to maximize sensitivity and minimize background staining but could be detrimental for the specificity. For this reason, the fractions of positive staining for cellular apoptosis reported here could not be compared with other reports because they originated from different protocols. However, the detection of activated caspase-3 is a valuable and appropriate tool for the relative comparison of apoptotic cells identified in sham vs. TO groups as we consistently use this method across groups. However, we must acknowledge that without a study comparing the TUNEL assay with an immunohistochemistry against the active form of caspase-3, the level of apoptosis we report should be used only for comparison within the same study rather than as an absolute number.
Prolonged TO has been associated with an important decrease in type II pneumocyte density, which leads to severe surfactant deficiency. Interestingly, in the 1970s, several studies looked into the effect of in vivo decapitation in lung development in the fetal rat and focused on the lack of pituitary-adrenal-thyroid axes on the type II pneumocyte differentiation (6, 7, 18). However, the decapitation was associated with tracheal stenosis, and it is now clear that the vast majority of the observations are related to TO. We previously reported that 1–3 wk of TO in the fetal lamb produces a dramatic decrease in both the number and the function of type II pneumocytes (43). This was confirmed by other studies in the fetal sheep (31, 39), rabbit (13), rat (28), and fetal mouse explants (10). A recently published study suggests that increases in fetal lung expansion induce differentiation of type II pneumocytes into type I pneumocytes via an intermediate cell type (20). The effect on type II pneumocytes seems to be dependent on gestational age, as TO late in gestation in rabbits does not affect the type II pneumocyte density in contrast to intervention in midgestation (12). Other studies have shown that shorter periods of TO can preserve surfactant-producing type II pneumocytes (33, 41). We propose that it is the briefer period of TO and the later time in fetal life at which it was performed in our experiments that, in fact, spare the expression of SP-C by type II pneumocytes. We do concede, however, that interspecies variation could have an effect as well.
This study confirms that the principal effect of TO during the early canalicular stage is indeed acceleration of alveolar-saccular development. Despite this, determining lung growth by estimating changes in lung weight and LW/BW does not reveal the entire response to TO, as its effect on lung architecture cannot be accounted for. During this stage of lung development, an increase in the air space-tissue fraction by apoptosis of the interstitial cells plays a crucial role. Therefore, in addition to its therapeutic role in treating fetal lung hypoplasia, TO offers a unique opportunity to study the process of alveolarization. This stage of lung development is still poorly understood, even though aberrant alveolarization is at the base of several diseases such as chronic lung disease of the premature infant. The development of our in vivo fetal mouse model offers a unique tool to study the biomolecular mechanisms of late lung development.
This work was supported by the Canadian Institute of Health Research, the Fond de Recherche en Santé du Québec, The Jeanne and Jean-Louis Lévesque Chair in Perinatology, and the Foundation for Research into Children Diseases.
Presented at the 24th Annual Perinatal Investigators' Meeting, November 9–11, 2000, Kingston, Ontario, Canada, and at the Pediatric Academic Societies' 2001 Annual Meeting, April 28-May 1, Baltimore, Maryland.
Address for reprint requests and other correspondence: B. Piedboeuf, Unité de recherche de pédiatrie, Centre de recherche du CHUL, Centre Hospitalier Universitaire de Québec, 2705 Boulevard Laurier, Sainte-Foy, Québec G1V 4G2, Canada (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published December 13, 2002;10.1152/ajplung.00079.2002
- Copyright © 2003 the American Physiological Society