INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

PULMONARY HYPOPLASIA is a major cause of neonatal respiratory diseases (15). Although fetal lung growth has been shown to depend on a proper lung fluid balance and normal fetal breathing movements (9), the precise mechanism(s) by which these physical factors influence lung cell proliferation remains unknown. To investigate the effect of phasic strain generated by fetal breathing movements on lung cell growth, we used a strain system for organotypic cultures of mixed fetal rat lung cells (17). DNA synthesis and cell division of mixed lung cells isolated from the canalicular stage of lung development were increased by an optimized intermittent strain regimen (11), which is very similar to the reported frequency, amplitude, and periodicity of normal human fetal breathing movements in vivo (7). Strain-induced cell proliferation as a function of lung development and lung cell type has not been previously investigated.

Fetal rat lung development can be divided into four periods: 1) embryonic period (days 11-13), development of major airways; 2) pseudoglandular period (days 14-18), development of airways to terminal bronchioles; 3) canalicular period (days 19-20), development of the acinus and vascularization; and 4) terminal sac period (days 21-22, term = 22 days), subdivision of saccules by secondary crests. The fifth and final stage, formation of true alveoli, occurs postnatally in the rat. Fetal lung cells change their phenotype (e.g., morphology, growth rate, cellular function) during each of these periods, and it is likely that the response of cells to mechanical stimulation also changes with gestation.

As assessed by autoradiography, both major cell types in organotypic cultures, epithelial cells and fibroblasts, respond to strain with increased DNA synthesis (11). This observation that both cell types can respond to strain when combined in a mixed-cell organotypic culture (11) does not mean that each cell type when cultured in isolation can respond to mechanical stimulation. Epithelial-mesenchymal interactions play a critical role in lung development (1, 18), and tissue interactions may also influence cellular responses to mechanical strain.

In the present study, epithelial cells and fibroblasts were isolated from the pseudoglandular, canalicular, and saccular stages of rat lung development, incubated in Gelfoam sponges alone or recombined in various ratios, and subjected to intermittent strain. We report that mechanical strain enhanced proliferation of epithelial cells and fibroblasts when cultured alone or when recombined. However, the proliferative response of mixed cells to strain was found to be both gestation dependent and modulated by mesenchymal-epithelial interactions.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Materials. Female (200-250 g) and male (250-300 g) Wistar rats were purchased from Charles River and bred in our animal facilities. Cell culture media, antibiotics, and trypsin were obtained from GIBCO Canada (Burlington, ON). Fetal bovine serum (FBS) was from Flow Laboratories (McLean, VA), and collagenase and DNase were from Worthington (Freehold, NJ). Cell culture flasks were from Falcon (Becton Dickinson, Lincoln Park, NJ), and multiwell plates were from Costar (Johns Scientific, Toronto, ON). Gelfoam sponges were from Upjohn (Toronto, ON), and [³H]thymidine was from Amersham Canada (Oakville, ON). All other chemicals were from Sigma (St. Louis, MO).

Cell culture. The isolation of epithelial cells and fibroblasts has been described previously (5, 8, 14). Briefly, fetuses were removed from the dam at day 18 to day 21 of gestation (term = 22 days), and the fetal lungs were dissected out into cold Hanks' balanced salt solution without calcium or magnesium [HBSS()] and cleared of major airways and vessels. The lungs were washed twice in HBSS(), minced, and suspended in HBSS(). The lung tissue was digested with trypsin solution [0.125% (wt/vol) trypsin and 0.4 mg/ml of DNase] for 20 min. After the lung tissue was filtered through a 100-µm mesh nylon bolting cloth, minimal essential medium (MEM) with 10% (vol/vol) FBS was added, and the mixture was centrifuged. The pellet was resuspended in MEM containing 0.1% (wt /vol) collagenase. After a 15-min incubation, the collagenase activity was neutralized by adding MEM + 10% (vol/vol) FBS. Two differential adhesion periods of 1 h in tissue culture flasks allowed attachment of fibroblasts. The nonadherent cells were removed, transferred to new culture flasks, and incubated overnight for attachment of epithelial cells. Nonadherent cells were removed from cell cultures after overnight incubation. Viability and purity of the cultures were comparable to previously published data (5). Fibroblasts stained positively (>95%) with anti-vimentin, but staining was negative for cytokeratin and smooth muscle cell actin, indicating that the fibroblast cultures were not contaminated with epithelial and smooth muscle cells. Epithelial cell cultures were cytokeratin positive (>95%) and vimentin negative. Although we previously reported fibroblast heterogeneity in fetal rat lung (5), herein we did not attempt to isolate the different fibroblast populations. After a 48-h culture, fibroblasts and epithelial cells were floated with 0.25% (wt /vol) trypsin + 0.4 mM EDTA, as previously described (14). Fibroblasts and epithelial cells were inoculated alone or as various recombinations (fibroblast-to-epithelial cell ratio in percentages: 100:0, 80:20, 50:50, 20:80, and 0:100) onto 2 × 2 × 0.2-cm Gelfoam sponges at a density of 1.6 × 10⁶ cells per sponge. After inoculation, cells were incubated at 37°C for 1 h before the addition of MEM + 10% (vol/vol) FBS.

After overnight incubation, sponges were washed twice with MEM to remove nonadhered cells. Attached cells were harvested from sponges by collagenase and trypsin digestion (11) and counted electronically to determine inoculation efficiency. In separate experiments, sponges were changed to MEM + 1% (vol/vol) FBS supplemented with 1 µCi/ml of [³H]thymidine and subjected to strain or static culture for 24 h.

Strain of fetal rat lung cells and assessment of cell proliferation. The mechanical strain device used in these studies has been described in detail elsewhere (11, 17). It consists of a programmable burst timer, a control unit, a direct-current power supply, and a set of solenoids. A culture dish with a Gelfoam sponge was placed in front of each solenoid. One end of each sponge was glued to the bottom of the dish, and the other end was attached to a movable metal bar, which was wrapped and sealed in plastic tubing. A magnetic force, generated through the solenoids, acted on the metal bar to apply strain to the organotypic cultures. Sponges were subjected for 48 h to a 5% elongation from their original length at 60 cycles/min for 15 min/h. We previously reported (11) that such a strain regimen optimally stimulated mixed fetal lung cell growth without cell injury, and the increase in proliferation, as assessed by thymidine incorporation, was not due to enhanced mixing of nutrients.

Cell proliferation was assessed by [³H]thymidine incorporation into DNA, as described previously (11). Results were determined as disintegrations per minute per sponge and are expressed as percentage of static control groups. In contrast to adult lung cells, especially adult type II pneumocytes, fetal lung cells undergo continuous proliferation both in vivo and in vitro. We have recently demonstrated that [³H]thymidine incorporation into DNA correlates directly with the change in () cell number in three-dimensional (3-D) organotypic cultures (12) of fetal lung cells. The correlation between disintegrations per minute per sponge and cell number per sponge was highly significant (r = 0.777, P < 0.001 for the unstrained control group; r = 0.940, P < 0.001 for the strained group) (12), implying that under the 3-D organotypic culture conditions thymidine incorporation into DNA mainly reflects DNA synthesis by proliferating fetal lung cells. Therefore, DNA synthesis, as measured by [³H]thymidine incorporation into DNA, was used as an index of cell proliferation.

Effect of conditioned media on DNA synthesis. Conditioned media were collected after a 24-h strain period (S-CM) and from static control cultures (C-CM). The conditioned media were centrifuged at 420 g to remove cell debris and then split in aliquots for storage at 20°C until mitogenic activities were measured. The mitogenic activity of conditioned media was assessed by measuring [³H]thymidine incorporation into DNA of monolayer cell cultures. One milliliter of a cell suspension (10⁴ cells/ml) in MEM + 10% (vol/vol) FBS, containing either day 19 or day 21 fetal rat lung fibroblasts or epithelial cells, was seeded in wells of 24-multiwell culture plates and incubated overnight. The wells were rinsed twice with fresh MEM. Cells were incubated for another 24 h in MEM + 1% (vol/vol) FBS for suboptimal cell growth. Cells were then cultured in the presence of either C-CM or S-CM supplemented with 1 µCi/ml of [³H]thymidine. Based on our previous findings (14), conditioned media were diluted 10 times with MEM. Because the conditioned media originally contained 1% (vol/vol) FBS, MEM supplemented with 0.1% (vol/vol) FBS was used as a control medium. After a 24-h incubation, media were aspirated, and cells were rinsed twice with ice-cold PBS. The amount of radioactive thymidine incorporated into DNA was then measured (14). Results were determined as disintegrations per minute per well and are expressed as percentage of C-CM values. It is worthwhile mentioning that we have also demonstrated that [³H]thymidine incorporation into DNA paralleled the increase in cell number (cells) in two-dimensional (2-D) monolayer cultures (13) of fetal lung cells. The correlation coefficient between disintegrations per minute per well and cells per well was highly significant ( = 0.918, P < 0.001). Thus [³H]thymidine incorporation into DNA is a reliable index of cell proliferation in 2-D cultures.

Statistical analysis. All values are shown as means ± SE. Statistical analysis was by Student's t-test or, for comparison of more than two groups, by one-way ANOVA, followed by Duncan's multiple range comparison test, with significance defined as P < 0.05 (19).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Cell attachment to Gelfoam sponges. To determine the inoculation efficiency of cells on sponges, epithelial cells and fibroblasts isolated from fetal lungs at day 18-21 of gestation were seeded on sponges separately or as various recombinations. After 24 h of incubation, nonadherent cells were removed and attached cells were collected and counted to determine the inoculation efficiency, which is expressed as percentage of number of inoculated cells (1.6 × 10⁶ cells/sponge). The average inoculation efficiency was 73.8 ± 1.95%. The efficiency did not change significantly as a function of development, cell type, or epithelial-mesenchymal recombination (Table 1).

Effect of mechanical strain on cell proliferation. To investigate the responses of fetal lung cells to mechanical stimulation, fibroblasts, epithelial cells, and various recombinations were subjected to static culture or intermittent strain. Cell proliferation was assessed by measuring DNA synthesis. To compare the responses of fibroblasts, epithelial cells, and their recombinants with strain, strain-induced DNA synthesis was plotted as a function of epithelial cell-to-fibroblast ratio of the recombinations. Table 2 shows the DNA synthesis of fetal lung cells during static culture. Independent of gestation, DNA synthesis decreased with an increasing amount of epithelial cells. Figure 1 shows that, except for day 18 fetal lung fibroblasts, cells from different gestations, individually or recombined, responded to strain with a significant increase in DNA synthesis compared with static control groups (P < 0.05). These data suggest that increased DNA synthesis is a common response to strain of both fetal lung epithelial cells and fibroblasts throughout late gestation. Regression analysis and ANOVA revealed that there was no significant additive effect of fibroblasts on strain-induced DNA synthesis of epithelial cells or vice versa at any day of gestation (Fig. 1). Because the regression lines did not change significantly at any day of gestation, we combined all the data of the 1:1 recombinations of each day of gestation to determine at which day of gestation cells were most responsive to mechanical strain. Strain-induced stimulation of DNA synthesis of the recombinations was maximal at the early canalicular stage of lung development at 19 days of gestation, after which there was a decline in strain-induced DNA synthesis (Fig. 2). The developmental pattern of strain-induced DNA synthesis of fibroblasts was similar to that of the recombinants. Strain-induced DNA synthesis of epithelial cells increased slightly but not significantly with advancing gestation of cells.

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Mechanical strain stimulates DNA synthesis of fetal rat lung cells. Epithelial cells and fibroblasts were isolated from pseudoglandular stage (day 18), canalicular stage (days 19 and 20), and saccular stage (day 21) of lung development, cultured on sponges alone or recombined in various ratios, and subjected to an intermittent strain or static culture for 24 h. [3H]thymidine incorporation into DNA was measured, and strain-induced DNA synthesis is expressed as percentage of control groups. Values are means ± SE. Data from 3-4 separate experiments carried out in duplicate were combined and plotted as a function of recombination ratios. Independent of gestation, except day 18 fibroblasts (# P > 0.05), and recombination ratio, DNA synthesis of strained cells was significantly greater compared with that of static control groups (P < 0.05). Fib, fibroblasts; Epi, epithelial cells.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Gestation-dependent proliferative response of mixed cells to mechanical strain originates in the mesenchyme. Epithelial cells (A) and fibroblasts (B) were isolated from pseudoglandular stage (day 18), canalicular stage (days 19 and 20), and saccular stage (day 21) of lung development, cultured on sponges alone or recombined in 1:1 ratios (C), and subjected to an intermittent strain or static culture for 24 h. [3H]thymidine incorporation into DNA was measured, and strain-induced DNA synthesis was expressed as percentage of static control groups. Values are means ± SE. Data from 3-4 separate experiments carried out in duplicate were combined and plotted as a function of gestation. Epithelial cells showed a slight but not significant increased proliferative response to mechanical strain with advancing gestation. Strain-induced DNA synthesis increased and peaked in fibroblasts at 19 days of gestation and declined thereafter. The gestational proliferative response of 1:1 recombinants to strain was identical to that of fibroblasts alone. * P < 0.05 compared with day 18 cells. # P < 0.05 compared with day 18 cells.

Effect of heterologous recombinations on strain-induced DNA synthesis. To examine whether fibroblasts determine the stimulatory effect of strain on DNA synthesis, we recombined in a 1:1 ratio either day 19 epithelial cells and day 21 fibroblasts or day 21 epithelial cells and day 19 fibroblasts. Although DNA synthesis of all recombinations was enhanced by mechanical strain (P < 0.05 compared with static control groups), this effect was significantly reduced (>50%) in heterologous recombinations compared with homologous recombinations of day 19 fetal lung cells (P < 0.001; Fig. 3). In contrast, DNA synthesis of unstrained heterologous recombinants was increased compared with unstrained homologous recombinants (data not shown). To investigate whether this decreased mitogenic response of heterologous recombinants to strain is regulated by soluble factors, we collected conditioned media from day 19 and 21 fetal lung epithelial cells or fibroblasts after a 24-h static culture (C-CM) or intermittent strain (S-CM). Mitogenic activities of conditioned media were tested on day 19 and 21 fetal lung epithelial cells and fibroblasts. S-CM of day 19 fetal lung fibroblasts significantly increased DNA synthesis of day 19 and 21 fetal lung epithelial cells (P < 0.05), although the stimulatory effect decreased with advancing gestation of epithelial cells (Fig. 4A). S-CM of day 21 fetal lung fibroblasts had no mitogenic effect on day 19 or 21 fetal lung epithelial cells. S-CM from day 19 and 21 fetal lung epithelial cells slightly but not significantly inhibited [³H]thymidine incorporation into DNA of day 19 and 21 fibroblasts.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Mechanical strain-induced DNA synthesis is repressed in heterologous recombinations. Fibroblasts and epithelial cells isolated from day 19 and day 21 fetal rat lungs were cultured as gestational homologous (19F/19E) or heterologous (19F/21E or 21F/19E) recombinations and subjected to an intermittent strain or static culture for 24 h. [3H]thymidine incorporation into DNA was measured. Results are expressed as percentage of control groups. Values are means ± SE of 3 separate experiments in triplicate. The stimulatory effect of strain on DNA synthesis was attenuated when cells from different gestational ages were recombined. F, fibroblasts; E, epithelial cells; 19 or 21, day 19 or day 21. * P < 0.05 compared with 19F/19E recombinant.

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Mechanical strain-induced mitogenic activities alter with gestation. Mitogenic activity of conditioned media from strained (S-CM) fibroblasts (A) and epithelial cells (B) was assessed with epithelial cells (A) and fibroblasts (B) from day 19 and 21 fetal lung. Conditioned media were diluted 10 times with MEM and supplemented with 1 µCi/ml of [3H]thymidine for DNA synthesis measurements. S-CM results are expressed as percent change of static control medium (C-CM). Values are means ± SE of 3 separate experiments in quadruplicate. * P < 0.05 compared with C-CM.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The physical force experienced by fetal lung cells during fetal breathing movements is potentially an important mediator of normal fetal lung growth. We previously demonstrated that an intermittent strain regimen, simulating fetal breathing movements, enhanced mixed fetal lung cell proliferation in vitro (10-13). Uptake of [³H]thymidine was increased in both epithelial cells and fibroblasts when organotypic cultures of mixed cells were subjected to mechanical strain (11). Herein we show that mechanical strain also increased DNA synthesis of fetal lung epithelial cells and fibroblasts when both cell types were separately subjected to strain.

Recently, we have shown that the incorporation of [³H]thymidine into DNA by fetal lung cells correlates directly with changes in cell number, indicating that fetal lung cells in 3-D organotypic cultures (12) and 2-D monolayer cultures (13) progress through the cell cycle without interruption. Thus measurement of DNA synthesis is a reliable indicator of fetal lung cell proliferation in organotypic and monolayer culture. Because of the tight correlation between DNA synthesis and cell number, increases in DNA synthesis observed in the present study appear small, but it should be noted that a 25% increase of [³H]thymidine incorporation into DNA reflects a similar net increase in cells.

Although fetal lung cells responded mitogenically to strain throughout late gestation, there was a gestation-dependent effect. The proliferative response of mixed (1:1 recombinants) fetal cells to strain was low at the pseudoglandular stage on day 18, peaked during the early canalicular stage on day 19, and fell again during the late canalicular and saccular stages at days 20 and 21 of gestation. A similar ontogenic profile of strain-induced DNA synthesis was observed for fibroblasts but not epithelial cells, suggesting that the gestation-dependent component of the proliferative response of mixed fetal lung cells to strain originates in the mesenchyme. The developmental growth pattern of mixed fetal lung cells subjected to intermittent strain was identical to that of whole fetal lung (4), i.e., ontogenic pattern of DNA synthesis of the mixed fetal lung cells mimicked developmental expression pattern of growth-related genes (c-myc, c-fos, histone 3) in whole fetal lung. In contrast, developmental DNA synthesis profiles of intermittently strained epithelial cells and fibroblasts differed from those of static cultured cells (3, 6). Despite the usual caution regarding cell culture findings, these results suggest that fetal lung cells subjected to an intermittent strain regimen, simulating fetal breathing movements, more closely reflect the in vivo situation than static cell cultures.

Our initial findings suggested that mesenchymal-epithelial interactions did not influence the mitogenic response of fetal lung to strain in that all homologous recombinants, independent of tissue ratio and gestational age, responded to intermittent strain similarly to individual cells. However, the reduced growth-promoting response of heterologous recombinants to strain did suggest a role for tissue interactions. It is plausible that tissue interactions in homologous recombinants maintain the cells at a proper growth and differentiation state to sense mechanical signals. Interference with these interactions by changing the developmental phenotype of one of the cell types in the heterologous recombinants leads to a diminished biological response to mechanical strain.

Studies with static cell cultures have suggested that fetal lung cell growth may be controlled by mesenchymal-epithelial interactions (1, 5, 20). We previously demonstrated that S-CM of day 19 mixed fetal rat lung cells stimulated proliferation of day 19 epithelial cells but not that of day 19 fibroblasts (14). Herein we found that S-CM collected from day 19 fibroblasts enhanced day 19 epithelial cell proliferation. S-CM of day 21 fetal lung fibroblasts was not mitogenic for day 19 fetal lung epithelial cells. Thus the reduced mitogenic response of the heterologous (day 19/day 21) recombinants to strain compared with homologous (day 19/day 19) recombinants may be partially due to a reduced release of mitogenic activities by fibroblasts. Enhanced growth factor expression in response to mechanical strain may be a common reaction shared by various cell types. Bishop et al. (2) have reported that the mitogenic activity of conditioned media from an embryonic lung fibroblast cell line is increased after mechanical strain. Sadoshima and Izumo (16) found that conditioned media from strained neonatal cardiac myocytes can induce hypertrophy of nonstrained cells. We have recently observed that mechanical strain rapidly upregulates platelet-derived growth factor-B mRNA expression and protein production in mixed fetal lung cells (10).

In response to mechanical strain, cells may release not only growth-promoting factors but also inhibitory factors. Studies with static fetal lung cell cultures have shown that epithelial cells release inhibitory growth factors for fetal lung fibroblasts (6). Herein we found that S-CM of epithelial cells inhibited fibroblast growth independently of gestation. The S-CM of mixed fetal rat lung cells appeared to be more inhibitory than that of fetal lung epithelial cells alone (14), again indicating that mesenchymal-epithelial interaction may influence the production of soluble factors.

Taken together, these data are consistent with a gestation-dependent proliferative response of fetal lung cells to mechanical strain that originates in the mesenchyme.