Main

BPD is a chronic lung injury which is a major cause of mortality and morbidity in newborn infants requiring prolonged respiratory support. Since the original description of the condition in 1967(1), it has become generally accepted that pulmonary immaturity, O2 toxicity, and barotrauma all contribute to its pathogenesis, although their relative contributions remain uncertain. We, and others, have used an adult rat model of pulmonary O2 toxicity to determine changes in the expression of growth factors which may mediate the observed pneumocyte and fibroblast hyperplasias(2–5). The advantage of this model is that the adult lung is essentially a nonproliferative organ and that any changes in growth factor expression resulting from lung injury are observed against a background of low or absent expression. The disadvantage is that it does not replicate the situation in the neonate, in whom there is normally active postnatal lung growth. BPD is, in the most severe cases, characterized by a long-term global reduction in alveolar number and surface area(6). However, during the reparative phase after the initial acute necrotic injury, both fibroblast and pneumocyte hyperplasia are evident(7). Whether the factors which control these reparative attempts, which may lead to fibrosis, are the same factors as control normal postpartum lung cell growth is unknown.

In 1979, a National Institutes of Health-sponsored workshop on BPD recommended that emphasis be placed on the development of animal models that would allow a better understanding of factors which contribute to BPD(8). Progress in this direction has recently been comprehensively reviewed by deLemos and Coalson(9). The preterm baboon ventilated with increased O2 concentrations appears to be the best large animal model of human BPD currently available(9), but there is currently no good small animal model suitable for screening various antioxidant combinations for their effect on O2-mediated changes in lung structure. Randell et al.(10) found that neonatal rats exposed to >95% O2 for 1 wk, then allowed to recover in air, had arrested alveolarization with considerable catch-up during the recovery period. Histologic changes were of a modest reduction in alveolar number and an increase in the scatter of alveolar size measurements. A variety of exposure protocols, as reviewed by deLemos and Coalson(9), have resulted in bullous emphysematous lesions in rodent lungs. We wished to create an injury which more closely mimics the dysplastic lesions observed in BPD. We reasoned that either a more chronic exposure to a lesser concentration of O2, or recovery at an increased O2 concentration, might allow the development of a lesion characterized by poorly regulated cell proliferation. The two protocols used were an exposure to >95% O2 for 7 d, followed by 7 d in 60% O2 or an exposure to 60% O2 for the full 14 d. Neonatal rats exposed to >95% O2 for 7 d had marked inhibition of lung cell DNA synthesis, and their recovery in 60% O2 was not associated with recovery of DNA synthesis or gross parenchymal thickening. In contrast, a continuous 14-d exposure to 60% O2 was associated with patchy areas of parenchymal thickening. Although there was an overall reduction in lung cell DNA synthesis, areas of parenchymal thickening had active DNA synthesis. These histologic changes were accompanied by a reduced lung volume and elastance. Because of these findings, our further studies focused on the 14-d 60% O2 model. In this report, we describe how an exposure to 60% O2 can alter the pattern of expression of IGF and IGF-IR. The IGF were studied because the temporal changes in their postnatal expression(11) suggest a critical role in normal postnatal lung growth.

METHODS

Materials. Radioisotopes and nylon membranes were purchased from Amersham Canada (Oakville, Ontario), and restriction enzymes and dextran sulfate were from Pharmacia (Baie D'Urfe, Quebec). BSA type V, Ficoll 400, polyvinylpyrrolidone, guanidinium thiocyanate, cesium chloride, and salmon sperm DNA were from Sigma Chemical Co. (St. Louis, MO). Organic solvents were of HPLC grade. The histone H3 cDNA (insert size, 2.1 kb) was a gift from Dr. G. S. Stein (Massachusetts Medical School, Worcester, MA). Mouse IGF-I cDNA(insert size, 0.7 kb) and IGF-II cDNA (insert size, 1.4 kb) were a gift from Dr. Graeme Bell (University of Chicago, Chicago, IL). The IGF-IR cDNA (insert size, 0.7 kb) was from the American Type Culture Collection (Rockville, MD). Rabbit anti-human factor VIII IgG was obtained from Dakopatts (Glostrup, Denmark). Guinea pig anti-rat SP-A IgG was a generous gift from Dr. J. Whitsett of the University of Cincinnati (Cincinnati, OH). Antisera raised in rabbits against recombinant human IGF-I and IGF-II were kindly provided by Dr. Victor Han (University of Western Ontario, London, Ontario) and were originally gifts from Drs. P. Gluckman and B. Brier (University of Auckland, New Zealand). A mouse monoclonal anti-rat macrophage (clone ED2) IgG was from Serotec Canada (Toronto, Ontario). The specificity of the IGF-I and IGF-II antibodies has also demonstrated by a loss of immunoreactivity after the primary antisera had been immunoabsorbed with either recombinant human IGF-I or IGF-II(11). Antiserum raised in rabbits against a synthetic peptide corresponding to residues 642-659 of the deduced cDNA sequence of the α-subunit of the human IGF-IR, which is specific for the human IGF-IR α-subunit and cross-reacts with the rat IGF-IR but not the IGF-II or insulin receptors, was purchased from UBI (Upstate Biotechnology Inc., Lake Placid, NY). The specificity of the IGF-IR antibody has also been demonstrated by a loss of immunoreactivity after the primary antiserum had been immunoabsorbed with the synthetic peptide(12). Kits for avidin-biotin-peroxidase and alkaline phosphatase complex immunohistochemical staining were from Vector Laboratories (Burlingame, CA), and α-aminopropyltriethoxysilane from Pierce Chemical Co. IL). A trichrome stain (Masson) kit for collagen was from Sigma Chemical Co.

Institutional review. All procedures involving animals were conducted according to criteria established by the Canadian Council for Animal Care, and prior approval was obtained from the Animal Care Review Committee at the Samuel Lunenfeld Research Institute, Mt. Sinai Hospital.

Exposure system. Pathogen-free timed pregnant Sprague-Dawley rats (250-275 g) were obtained from Charles River (St. Constant, Quebec). All experiments were conducted as paired exposures with one chamber receiving O2 and the other receiving air. On the anticipated day of delivery each animal was placed in a 60 × 48 × 25 cm plastic chamber with 12 h: 12 h light-dark cycles. Food and water were available ad libitum. Chambers were connected to an air or O2 source before placing animals in the chambers and delivery occurred in the selected gas. Oxygen concentrations were calibrated daily using an on-line SensorMedics O2 analyser (Summit Technologies, Oakville, Ontario). Gas flow was adjusted to maintain minimal chamber humidity with a CO2 concentration below 0.5% using an on-line SensorMedics CO2 analyzer (Summit Technologies, Oakville, Ontario). Dams were exchanged daily between O2 and air chambers to prevent maternal O2 toxicity. Litter sizes were maintained the same by removing an equal number of pups from air-exposure chambers as died in paired O2 chambers.

Autoradiography and immunohistochemistry. Animals were anesthetized with intraperitoneal ketamine (80 mg/kg) and xylazine (20 mg/kg). A tracheal catheter was inserted and sutured in place to facilitate lung inflation. The anterior part of the chest wall was reflected upward. While the heart was beating, a catheter was inserted through the right ventricle into the main pulmonary artery, and an incision was made in the left atrial appendage to allow drainage. The pulmonary circulation was then flushed with PBS containing 1 U/mL heparin, during intermittent lung inflations, until the lung became white. For autoradiography animals received 1 μCi/g intraperitoneal [3H]thymidine 2 h before lung fixation. The lungs were cleared of blood as described above, then fixed with freshly prepared 4%(wt/vol) paraformaldehyde and 0.2% (vol/vol) glutaraldehyde by perfusion under a constant airway pressure of 10 cm H2O. The lungs were embedded in paraffin and cut in 5-μm sections. Sections were mounted onα-aminopropyltriethoxysilane-coated slides. After completion of immunohistochemical studies, using an avidin-biotin-peroxidase complex method(13), slides were lightly counterstained with Carazzi hematoxylin, dehydrated, cleared in xylene, and mounted. Dilutions of the primary antisera were 1:1000 for IGF-I, 1:500 for IGF-II, 1:800 for IGF-IR, and 1:100 for macrophage identification. Antibody specificity was verified by both omitting the primary antiserum or by replacing it with nonimmune rabbit IgG. The specificity of the IGF-I and IGF-II antibodies has also been confirmed by a lack of staining after the primary antisera had been immunoabsorbed with either recombinant human IGF-I or IGF-II(11). Slides were coated with Kodak NBT-3 emulsion for autoradiography, and developed after 4 wk at 4°C for examination by light microscopy to identify dense areas of silver granule concentration over cell nuclei undergoing active DNA synthesis. Positive labeling was defined using the criteria of Chwalinski et al.(14). For quantitative autoradiography slides were studied at ×40 magnification. The number of positive cells was counted in 10 random fields for each of two animals from different litters at each time point. The selected animals were chosen arbitrarily from the average sized pups of each litter. For the d-14 time point, slides were also studied after immunohistochemical staining for anti-human factor VIII antigen to identify endothelial cells, and for anti-rat SP-A to identify primarily type II pneumocytes as previously described(3).

Isolation of lung RNA. Total lung RNA was isolated as previously described(3). Briefly, the thoracic contents were removed en bloc, and the lungs were dissected away from vessels and large airways to be flash frozen in liquid nitrogen after weighing. Total(nuclear and cytoplasmic) RNA was isolated by lysing the tissue in 4 M guanidinium thiocyanate followed by centrifugation on a 5.7 M cesium chloride cushion to pellet RNA(15). After extraction with phenol/chloroform (1:1, vol/vol), the RNA was ethanol-precipitated and collected by centrifugation. This RNA was lyophilized and dissolved in water. Integrity of RNA was confirmed after fractionation on 1.2% (wt/vol) agarose-formaldehyde gels by staining the ribosomal RNA bands with ethidium bromide.

RNA analyses. Slot blot analyses were performed because of the large number of samples obtained from several animals at multiple time points. Slot blot analyses of 0.6-μg aliquots of RNA, which had been determined in preliminary studies to give results in the linear range for densitometry, were applied in a final 200-μL solution of 6.15 M formaldehyde and ×10 SSC(1.5 M sodium chloride, 0.15 M sodium citrate, pH 7.0). Ribosomal RNA was used as a negative control. The RNA was denatured at 65°C for 15 min and immediately immobilized on nylon membrane under slow vacuum filtration. After the solution had filtered through, the wells were rinsed with 500 μL of×10 SSC, and RNA was fixed by baking at 80°C for 2 h.

The specificity of IGF-I, IGF-II, and IGF-1R probes was confirmed by performing Northern blot analyses on total RNA from lungs exposed to air or to 60% O2. The total RNA samples were denatured in 1.7 M formaldehyde for 30 min at 65°C and electrophoresed (20 μg/lane) in 1% (wt/vol) agarose gels containing 2.2 M formaldehyde. The RNAs were transferred to Hybond N+ nylon membrane (Amersham Corp., Arlington Heights, IL) using the capillary transfer technique. After transfer the blots were baked at 80°C for 1 h. Thereafter both slot and Northern blots were processed in a similar fashion.

All probes were labeled with deoxycytidine 5′-[α-32P]triphosphate by a random primed labeling system(Amersham), with specific activities of 0.5-2.9 × 109 counts·min·μg DNA-1. Prehybridization (5 h overnight) and hybridization were performed in 50% (vol/vol) formamide, 750 mM NaCl, 75 mM sodium citrate, ×5 Denhardt's solution [= 0.4% (wt/vol) each of BSA, Ficoll, and polyvinylpyrrolidone]; 10% (wt/vol) dextran sulfate; 100 μg/mL denatured salmon sperm DNA for 24-48 h at 42°C. Washes varied with the probe used. IGF-I blots were washed twice using ×2 SSC and 0.2% (wt/vol) SDS at 42°C for 10 min, followed by ×1 SSC and 0.2% (wt/vol) SDS at 42°C for 10 min. Blots hybridized with the IGF-II probe were washed in×5 SSC and 0.2% (wt/vol) SDS at 42°C for 20 min followed by ×2 SSC and 0.2% (wt/vol) SDS at 42°C for 10 min. IGF-IR blots were washed in×SSC with 0.2% (wt/vol) SDS at 42°C for 15 min, followed by ×2 SSC and 0.2% (wt/vol) SDS at 42°C for 10 min. The blots were exposed for 24-48 h to Kodak XAR-5 film using Dupont Cronex intensifying screens. The films were quantified by an Ultroscan XL laser densitometer (LKB, Bromma, Sweden). All steady state levels of the mRNAs studied have been normalized to the densitometric value of the air control. Glyceraldehyde-3′-phosphate dehydrogenase (EC 1.2.1.12) and β-actin probes are commonly used as controls, but in studies of O2-mediated alterations of gene expression they cannot be used as the expression of both β-actin(16) and glyceraldehyde-3′-phosphate dehydrogenase(data not shown) mRNAs are affected by exposure to elevated concentrations of O2. To correct for variations in loading of gels and transfer to membranes, all results were normalized using methylene blue staining(17). After hybridization and exposure to x-ray film, the slot blot filters were soaked in 5% (vol/vol) acetic acid for 15 min at room temperature. The filters were then transferred to a solution of 0.5 M sodium acetate with 0.04% methylene blue (vol/vol) for 5-10 min at room temperature. This was then followed by a rinse in water for 5-10 min. The filter was then photographed, and a reverse negative was scanned by laser densitometry.

Other assays and measurements. Pups were killed using excess ether inhalation and weighed. After decapitation, blood was allowed to drain from the transected neck, collected, and sonicated before enzyme assays. The heart and lungs were removed en bloc and kept at 4°C. The heart and major airways were dissected off, and the lungs from each litter were pooled, blotted dry, and weighed to give an average lung wet weight/pup. The pooled lung tissue from each litter was placed in a preweighed tube and lyophilized. After reweighing to obtain an average dry weight/pup, the tissue was reduced to a fine powder using a pestle and mortar. This material was resuspended in 50 mM potassium phosphate, 0.1 mM EDTA, pH 7.8, at 1 mL/g of wet weight and sonicated, and insoluble material was removed by centrifugation at 1000 × g for 10 min. The supernatant was made up to 2 mL/g of wet weight, then stored at -70°C before assay. Catalase (EC 1.11.1.6) activity was measured as described by Bergmeyer(18), glutathione peroxidase (EC 1.11.1.9) activity by an indirect coupled assay as described by Günzler and Flohé(19), and both Cu,Zn- and Mn-superoxide dismutase activity (EC 1.15.1.1) as described by Hayatdavoudi et al.(20). Contamination of homogenates with blood was estimated as described by Cross et al.(21). Protein content was assayed by the method of Bradford(22).

Lung volume-pressure loops. Animals were anesthetized with pentobarbital sodium (5-10 mg/kg) and paralyzed with pancuronium bromide (0.3 mg/kg). They were tracheotomized and placed in a multicompartment plethysmograph as described by Lachmann et al.(23). The animals were ventilated simultaneously from a common ventilator. Airway pressure was monitored with a Validyne (Northridge, CA) MP45 pressure transducer. Flow from each compartment was measured with a Fleisch 00 pneumotachograph and a Validyne DP45 differential pressure transducer. By adjusting the length of the low pass filter attached to the outflow line, the tidal volume of each animal was set to 10 mL/kg. The animals were ventilated for 10 min with 100% O2, then the endotracheal tube was clamped at end-expiration for 5 min to degas the lungs. The tracheas were then connected to narrow glass tubes attached to a water reservoir. By raising the reservoir a stepwise increase in airway pressure was created. The volume displaced was recorded from the movement of the menisci, with appropriate correction for compression volume. The animals were inflated and deflated in 5 cm H2O steps with a 1 min equilibration at each step.

Data presentation. Unless otherwise stated all numerical values are shown as the χ ± SEM of four litters for each time point. Statistical significance (p < 0.05) was determined by an analysis of variance followed by assessment of differences using Dunnet's two-sided test(24), or Duncan's multiple range test(25). SEM bars are not evident for all data points shown in the figures because they fall within the plot point.

RESULTS

Cumulative mortality curves are shown in Fig. 1 for newborn rat pups exposed to three different O2-exposure protocols. In contrast to a 60% mortality in 95% O2 for 2 wk, animals exposed to 95% O2 for 1 wk then to 60% O2 for 1 wk had a reduced mortality of 44%. This was further reduced to 19% if animals were maintained in 60% O2 throughout the 2-wk exposure period. After 7 d in air the neonatal lung showed very active DNA synthesis, as assessed by [3H]thymidine autoradiography (Fig. 2A). DNA synthesis had declined considerably by d 14 in air (Fig. 2B). By d 7 in 95% O2, lung DNA synthesis was clearly reduced (Fig. 2C). After a further 7 d of 95% O2, DNA synthesis was essentially undetectable (data not shown), whereas recovery in 60% O2 allowed only a low level of DNA synthesis by d 14 (Fig. 2D). Patchy areas of bullous emphysema were observed with recovery in 60% O2 (data not shown), but no obvious areas of marked interstitial thickening. We therefore focused our attention on pups exposed to air or to 60% O2 for 14 d. Pups continuously exposed to 60% O2 had small increases in antioxidant enzyme activities during the 2nd wk of exposure, but the only statistically significant increase was in glutathione peroxidase activity after exposure to 60% O2 for 14 d (Table 1). As shown in Figure 3, exposure to 60% O2 caused a reduction in lung weight and lung/body weight ratio by d 4 of exposure, and an increase in the same parameters after a 14-d exposure. Although statistically significant (p < 0.05), the observed differences were small. There were no significant effects (p > 0.05) of exposure to 60% O2 on body weight (Fig. 3) or on lung wet/dry weight ratio (data not shown). Lung mechanics were, however, significantly impaired by a 14-d exposure to 60% O2, as assessed by volume pressure loops (Fig. 4). All points on the deflation limb from 30 to 5 cm H2O, and from 15 to 30 cm H2O on the inflation limb, showed a significant (p < 0.05) reduction in lung volume in the 60% O2-exposed pups. Gross morphologic changes, with patchy areas of both alveolar enlargement and of interstitial thickening, were evident after a 14-d exposure to 60% O2 (Fig. 5). Trichrome staining of lung sections did not suggest that this interstitial thickening was attributable to increased collagen deposition (data not shown). Counts of[3H]thymidine-labeled cells and analysis of histone H3 mRNA content, as indices of global lung cell division (Fig. 6), both suggested that the major impact of 60% O2 on DNA synthesis was in the 2nd wk of exposure. Inhibition of [3H]thymidine labeling by 60% O2 was statistically significant (p < 0.05) on d 4 and 14 of exposure, whereas suppression of histone H3 gene expression was statistically significant (p < 0.05) on d 10 of exposure. Inhibition of DNA synthesis was not equivalent for all cell types, as assessed by double immunolabelling for cell-specific markers (SP-A and factor VIII) and[3H]thymidine autoradiography. On d 14 of exposure to 60% O2 labeling of factor VIII-positive (endothelial) cells was reduced by 82% from 4.25 ± 0.97 in air-exposed lungs to 0.75 ± 0.60 cells/field, whereas labeling of parenchymal cells not immunoreactive for either factor VIII or SP-A (chiefly interstitial cells) was reduced by only 58% from 51.1± 1.3 to 21.4 ± 1.1 cells/field. In contrast, labeling of parenchymal cells staining for SP-A (chiefly type II pneumocytes) had increased 35% from 2.6 ± 1.2 to 3.5 ± 1.0 cells/field. Alveolar macrophages were infrequently seen in either air- or 60% O2-exposed neonatal lungs, rarely had evidence of active DNA synthesis and were not particularly localized to areas of active DNA synthesis by other cell types(data not shown).

Figure 1
figure 1

Cumulative mortality of newborn rats exposed to 95% O2 for 14 d (▪), 95% O2 for 7 d then 60% O2 for 7 d(•) or 60% O2 for 14 d (○). Each point represents the combined survival of the total pool of pups from 16 litters at d 0 and 4, 12 litters at d 7, 8 litters at d 10, and 4 litters at d 14. Each litter had 10-14 pups at the onset of exposure.

Full size image
Figure 2
figure 2

[3H]Thymidine autoradiography of lung tissue using dark (A-D) and bright (E-H) field microscopy to show granules over cell nuclei in white or black respectively. A/E, Extremely active DNA synthesis at d 7 in air. Original magnification × 250. B/F, DNA synthesis has declined by d 14 in air. Original magnification × 250. C/G, Inhibition of DNA synthesis after 7 d in 95% O2. Original magnification × 250. D/H, After 7 d in 95% O2 then 7 d in 60% O2 there is minimal DNA synthesis. Original magnification × 250.

Full size image
Table 1 The influence of 60% O2 on lung antioxidant enzyme activities
Full size table
Figure 3
figure 3

The effect of exposure to air (○) or 60% O2(•) on lung weight, body weight and the ratio of lung weight(LW) to body weight (BW) × 103. Data points represent the χ ± SEM for four litters. Where error bars are not evident they fall within the plot point. *p < 0.05.

Full size image
Figure 4
figure 4

Lung volume pressure loops for rats exposed to air(○) or 60% O2 (•) for 14 d. Data points represent the mean± SEM for eight or nine pups. *p < 0.05.

Full size image
Figure 5
figure 5

Macroscopic appearance of the lungs of rat pups exposed to air (A) or 60% O2 (B) for 14 d. Lungs exposed to 60% O2 had a reduced volume with patchy areas of parenchymal thickening(D) compared with normal air-exposed lung tissue (C).

Full size image
Figure 6
figure 6

The effect of air (○) or 60% O2 (•) for 14 d on lung cell DNA synthesis as assessed by [3H]thymidine([3H]Tdr) autoradiography and histone H3 gene expression. Autoradiography data points represent the χ ± SEM for 20 microscope fields (10 fields/pup from two pups from different litters) at each time point. For histone H3 mRNA data points represent the χ ± SEM for four litters. Where error bars are not evident they fall within the plot point. *p < 0.05.

Full size image

Results of slot blot analyses for IGF-related mRNAs are shown in Fig. 7. The patterns of expression are similar for all three mRNAs with an increase after birth in air-exposed animals to a maximum at 6 d, followed by a lower level of expression at 10 and 14 d. This profile is similar to that seen for the mRNA marker of cell proliferation, histone H3. The 60% O2-exposed animals had a similar, but dampened, pattern. Specificity of the slot blot analyses was confirmed using Northern blot analyses for d 10 of exposure. Transcript sizes were 7.8, 4.4, 2.1, and 0.6 kb for IGF-I, 3.5 kb for IGF-II, and 11.9 kb for IGF-IR (Fig. 8).

Figure 7
figure 7

The effect of air (○) or 60% O2 (•) for 14 d on whole lung mRNA for IGF-I, IGF-II, and IGF-IR as assessed by slot blot analysis. All data points represent the χ ± SEM for four litters. Where error bars are not evident they fall within the plot point.

Full size image
Figure 8
figure 8

Northern analysis of total lung RNAs after a 10-day exposure to air or 60% O2. Transcript sizes are shown in kilobases.

Full size image

Immunohistochemistry for IGF-I and IGF-IR is shown in Fig. 9. Because our findings for IGF-I and IGF-II immunohistochemistry of the air-exposed pups are not significantly different from those recently reported by others(12), using the same antibodies, and because patchy thickening was only evident in 60% O2-exposed animals after 14 d, we have focused on that time point. By d 14 of life IGF-II immunoreactivity was minimal except for solitary cells around airways, assumed to be macrophages, and was not significantly altered by 60% O2 exposure (data not shown). IGF-I immunoreactivity was evident in the lungs of 14-d air-exposed pups. Although not particularly intense, it was widely distributed over a variety of cell types (Fig. 9A). In the 14-d 60% O2-exposed pups, IGF-I was detected in all areas of the lung with staining being particularly intense over the airway epithelium (Fig. 9B). At birth IGF-IR is localized almost exclusively to the vascular endothelium and to the smooth muscle layer surrounding the airway epithelium (Fig. 9C). Some atelectasis is evident in this section because the newborn lung was not inflated for fixation. Over the next 14 d this staining pattern became progressively less intense as staining also became evident over cells in the lung parenchyma (Fig. 9D). In the lungs of pups exposed to 60% O2 for 14 d, the intensity of IGF-IR immunoreactivity correlated well with the intensity of DNA synthesis, as assessed by [3H]thymidine autoradiography. Those areas with reduced proliferative activity had minimal staining for IGF-IR (Fig. 9E). Areas of tissue thickening and increased cell proliferation had a marked increase in the number of cells with IGF-IR immunoreactivity, particularly in the alveolar epithelium (Fig. 9F).

Figure 9
figure 9

Immunohistochemistry for IGF-I and IGF-IR in lung tissue of neonatal rats exposed to air or 60% O2. A, IGF-I at d 14 in air is detected in some airway (a) epithelial cells and multiple parenchymal cells (arrows). Original magnification ×320. B, IGF-I at d 14 in 60% O2 is detected over all lung tissue in an area of tissue thickening with maximum intensity of staining over airway (a) epithelial cells. Original magnification ×320. C, IGF-IR at birth was detected in vessel (v) endothelial cells and in airway(a) smooth muscle of uninflated lung. Original magnification×250. D, IGF-IR at d 14 in air had a similar distribution with occasional cells in the parenchyma also being stained (arrows). Original magnification ×320. E, IGF-IR at d 14 in 60% O2 in an area of the lung with low DNA synthesis as assessed by [3H]thymidine autoradiography. Only occasional cells had granules over their nuclei(arrows) and the distribution of IGF-IR was similar to that seen in animals exposed to air. Original magnification ×250. F, IGF-IR at d 14 in 60% O2 in an area of tissue thickening and active proliferation as assessed by the number of cells with granules over their nuclei. Staining intensity is increased over vessel (v) endothelial cells and airway (a) smooth muscle cells and multiple parenchymal cells now show immunoreactivity. Original magnification, × 320.

Full size image

DISCUSSION

Exposure of adult rats to 100% O2 is rapidly lethal. Exposure to a sublethal 85% concentration of oxygen allows anti-oxidant enzyme induction, tolerance to subsequent 100% O2 exposure, and a reproducible lung injury involving both fibroblast and pneumocyte hyperplasia(26). In contrast, 60% O2 does not result in antioxidant enzyme induction in the adult lung, reduces tolerance to subsequent 100% O2 exposure, and leads to only modest histologic changes located primarily in small vessel endothelium(20). Although well defined in the adult rat, a predictable exposure protocol that will cause dysplastic fibroblast and pneumocyte hyperplasia in the neonatal rat lung has not been described. It has been reported that 40% O2 produces variable and unpredictable changes in lung histology(27). We observed that exposure to>95% O2 markedly reduces lung DNA synthesis by d 7 of exposure and has a high mortality rate by 14 d of exposure. A 7-d recovery in 60% O2 was not associated with any significant recovery of DNA synthesis and was associated with mild emphysematous changes, but no obvious localized areas of increased cell division and parenchymal thickening. We therefore elected to examine the effect of continuous exposure to 60% O2. Bucher and Roberts(27) and Yam et al.(28) have previously reported the profiles of postnatal changes in lung antioxidant enzymes of air-breathing rat pups. Allowing for our use of corrections for blood-derived antioxidant enzymes and the breakdown of superoxide dismutase activity into two components, which were not done in either of the earlier reports(27, 28), our observations of postnatal antioxidant enzyme activities are essentially no different from those previously reported. Bucher and Roberts(27) also looked at the effect of a 12-d exposure to either 40 or 80% O2. In contrast to the marked increases in antioxidant enzyme activities observed with 95% O2(27, 29), exposure to 40 or 80% O2 resulted in much smaller increases, which were similar in magnitude and timing to those reported here. Because our findings are entirely consistent with the small but statistically significant changes reported by others(27) with oxygen concentrations above and below 60%, we did not feel justified in sacrificing more animals so as to increase the statistical significance of our measurements by increasing the sample size. Brumley et al.(30) subjected newborn rats to 60% O2 for 21 d and observed a reduced total lung capacity, vital capacity, lung compliance, and diffusion capacity as well as a 50% increase in type II pneumocytes. Unfortunately, this exposure protocol was also associated with a reduction in body weight. Because altered nutrition can influence sensitivity to increased O2(31), we reduced the exposure period in an attempt to prevent significant differences in body weight between air- or O2-exposed pups while preserving a measurable degree of lung injury. Despite the reduction in exposure time the pups exposed to 60% O2 still had abnormal volume-pressure loops, as well as an increase in type II pneumocyte DNA synthesis at the end of a 14-d exposure. The abnormalities of the volume-pressure loops included a reduced lung volume and a decreased elastance entirely consistent with our findings on gross histologic examination. The lungs appeared shrunken and, although there were fewer obvious emphysematous areas than seen with 95% O2 followed by 60% O2, patchy areas of localized parenchymal thickening were evident by d 14 of exposure to 60% O2. Global measures of whole lung cell division indicated an overall reduction in DNA synthesis. Our preliminary observations, with a small number of animals, using [3H]thymidine autoradiography, suggested that inhibition of DNA synthesis was not homogeneous, either with respect to the degree of inhibition for different cell types or regions of the lung affected. A more detailed morphometric analysis, with larger numbers of animals, will be required for a definitive interpretation of cell type variations in susceptibility to O2-mediated inhibitions of DNA synthesis. The majority of the lung without patchy mesenchymal thickening had obviously reduced nuclear labeling, whereas thickened areas had multiple cells with nuclear labeling.

We were particularly interested in determining whether these areas of parenchymal thickening and cell division would reveal the presence of the same growth factors as found in the normal growing postnatal lung. We selected the IGF for examination as they appear to play a role in both prenatal(32) and postnatal(11) lung growth. Our findings for air-exposed animals by slot blot analysis and immunohistochemistry of IGF-I and IGF-II were essentially similar to those recently reported by Wallen and Han(11). Exposure to 60% O2 had markedly different effects, depending on whether the area of lung being viewed was a proliferative or nonproliferative region. In the nonproliferative regions there was no obvious effect on IGF-I distribution and a reduction of IGF-IR. In contrast, the patchy areas of parenchymal thickening and active cell division had a marked increase in immunoreactive IGF-I and IGF-IR. The source of the IGF-I could be from production by mesenchymal or epithelial lung cells(11), or it may reflect leakage into lung tissue from the circulation. We have previously observed increased IGF-I in the serum of animals exposed to increased O2 concentrations(33).

Our findings support the concept that dysplastic lung cell proliferation in O2-mediated neonatal lung injury of small animal models can be produced by an appropriate O2-exposure protocol. The range of O2 concentrations which will be sufficient to create a detectible and reproducible injury, yet will be insufficient to simply globally arrest DNA synthesis, may be very narrow. Our results suggest that in this model of O2 toxicity, as in human infants with bronchopulmonary dysplasia, there is a marked heterogeneity of response to injury in different lung regions as well as by different lung cell types. Despite this heterogeneity there are sufficient easily measurable global changes in mRNA expression, DNA synthesis, and lung mechanics to allow the model to be used as a screening tool for studies of various antioxidant interventions.