Main

Despite significant advances in neonatal resuscitation and intensive care, newborn infants with CDH still have a high mortality rate (1). In CDH, this high rate has been attributed to pulmonary hypoplasia and PPH (1). CDH lung is developmentally arrested in the canalicular or saccular stage of development (2). Structural and functional parenchymal immaturities have been demonstrated as the characteristics of hypoplastic CDH lung in many clinical reports and animal experiments (3, 4). They include the lack of adequate surfactant, poor compliance, thickened alveolar septa, and the association with hyaline membrane formation despite term gestation (3, 4).

TNF-α is a polypeptide that is a potent inhibitor of the biosynthesis of surfactant phospholipids. TNF-α is a principal mediator of cardiopulmonary shock and ARDS associated with invasive bacterial infection and reperfusion injury (5). TNF-α was increased in BAL and serum of patients with ARDS or sepsis syndrome (6, 7). Surfactant component composition and function were also disrupted in animal models and patients with ARDS. High doses of TNF-α supplementation on embryonic mouse lung promida diminished branching morphogenesis (8). Thus, elevated TNF-α is associated with disease complicated by surfactant dysfunction. A recent study from our laboratory for the first time reported that mRNA expression of TNF-α is markedly increased in human newborn and stillborn CDH hypoplastic lung, suggesting the important role of elevated pulmonary TNF-α in the development of CDH hypoplastic lung (9).

In vitro studies have demonstrated that the expression of TNF-α is suppressed by glucocorticoids at both protein and mRNA levels (10, 11). In experimental CDH models, antenatal glucocorticoid therapy improves structural and functional abnormalities in hypoplastic CDH lung, promoting parenchymal maturation (12, 13). The exact mechanism by which antenatal glucocorticoids treatment promotes parenchymal maturation in CDH hypoplastic lung is not fully understood. The aim of this study was to investigate the effect of antenatal glucocorticoid administration on TNF-α protein and mRNA expression in nitrofen-induced CDH hypoplastic lung in rats.

MATERIAL AND METHODS

Creation of CDH and protein extraction.

Adult Sprague-Dawley rats were bred after overnight controlled matings. Observation of positive smears was considered a proof of pregnancy; the day of observation was determined d 0. Water and food were supplied ad libitum. At d 9.5 of pregnancy (term = 22 d), a single dose consisting of 100 mg nitrofen (WAKO Chemical, Osaka, Japan) dissolved in olive oil was given via a stomach tube, under short anesthesia. Antenatal Dex therapy was administered 72 and 48 h before delivery' the lowest dose of dexamethasone (0.25 mg/kg intraperitoneally) was used on d 18.5 and 19.5 of gestation. Pregnancy was continued, and cesarean section was performed on d 21 of gestation. The animals were killed by intracardiac pentobarbital injection. To determine which litters had CDH, we opened the chest and abdomen of fetuses under the dissecting microscope. The fetuses were divided into three groups: group I, control (n= 12); group II, nitrofen-induced left CDH (n= 12); group III, nitrofen-induced left CDH with antenatal Dex treatment (n= 12). In control animals, the same dose of olive oil was given without nitrofen. The research project had been approved by the Department of Health, Ministry of Health, Ireland.

Tissues.

The fetuses were killed by cesarean section at term, and lungs were dissected out from the thoracic cavity under a dissecting microscope and fixed by intratracheal instillation of 4% paraformaldehyde at a water pressure of 20 cm. The dissected lung tissue was fixed in 4% paraformaldehyde solution for 24 h and cryopreserved; 8-μm cryosections were mounted on poly-L-lysine-coated glass slides for in situ hybridization. Protein content was extracted from each lung with commercially available TRIZOL reagent (Life Technologies, Paisley, United Kingdom), according to the recommended protocol. Briefly, each sample was homogenized by TRIZOL reagent and incubated in chloroform. After removal of supernatant, protein fraction was collected in ethanol. Precipitated protein in isopropyl alcohol was washed in a solution containing 0.3 M guanidine hydrochloride in 95% ethanol. Finally, the protein pellet was redissolved in 1% SDS.

In situ hybridization.

To perform in situ hybridization, we used rat TNF-α-specific oligonucleotide probe (Biogenostic, Gottingen, Germany), a method we have described previously (14). The probe was labeled by DIG, with a DIG oligonucleotide tailing kit (Boehringer Mannheim, Mannheim, Germany). The slides were placed in freshly mixed 0.25% acetic anhydride in 0.1 M triethanolamine HCl, pH 8.0, for 10 min at room temperature, and were treated with graded ethanol and chloroform. After tissue preparation, the sections were prehybridized with hybridization buffer. Then the slides were incubated with hybridization buffer in which rat TNF-α probe was diluted to 100 ng/mL. Sections were processed for immunologic detection using alkaline phosphatase-conjugated anti-DIG serum. Nitroblue tetrazolium salt and 5-bromo-4-chloro-3-indolyl phosphate toluidinium salt were used as substrate. At least 10 random fields of alveolar region were examined. The number of TNF-α mRNA-positive cells were graded independently by two of us (H.S. and T.O.), without knowledge of the animal group, as (−), no TNF-α mRNA-positive cells; (±). occasional; (+), moderate; and (++), many. If the grades were not identical, sections were regraded and a consensus reached.

ELISA.

Rat TNF-α expression was measured in protein content from solubilized lung tissue extracts by ELISA analysis. ELISA procedure was done according to the protocol of the rat TNF-α ELISA kit (Genzyme Diagnostic, Cambridge, MA). This kit is a solid-phase ELISA that uses the multiple antibody sandwich principle. Briefly, a capturing anti-TNF-α antibody was coated onto microwell strips, and samples were incubated in each well. Then HRP-conjugated anti-TNF-α was added to each well and the plate was incubated. Each sample was incubated in the substrate regent, which was added in each well. After incubation, the substrate reaction was stopped by the addition of 1 M sulfuric acid.

Absorbance of colored reaction product was read with an ELISA microplate reader at 450 nm. Concentrations of pulmonary TNF-α in samples were determined by relating absorbance value to values from a standard curve generated with recombinant rat TNF-α. Samples were assayed in duplicate by two operators to assess interassay precision. Soluble protein concentrations of each tissue extract were measured by BCA protein assay kit (Pierce, Rockford, IL). Values of TNF-α for soluble protein were expressed as μg TNF/g soluble protein.

For statistical evaluation, we used the nonparametric Mann-Whitney U test. Differences were considered significant when p< 0.05.

RESULTS

The results are summarized in Table 1.

Table 1 TNF-α mRNA and protein expression in rat lung Abbreviation: (−) no TNF-α mRNA-positive cells, (±) occasional, (+) moderate, (++) many. *p< 0.01; vs group I, †p< 0.05; vs group II.
Full size table

In situ hybridization.

In control lung (group I), TNF-α mRNA expression was weak or absent in type II pneumocytes in mature lung parenchyma (Fig. 1a). In nitrofen-induced CDH lung (group II), abundant and strong TNF-α mRNA expression was demonstrated in type II pneumocytes and bronchiolar epithelium in immature lung parenchyma (Fig. 1b). In contrast, in nitrofen-induced CDH lung with antenatal Dex treatment (group III), TNF-α mRNA expression was markedly reduced in both type II pneumocytes and bronchiolar epithelium with an increase in air saccule size and thinning of septal walls (Fig. 1c). (C) Nitrofen-induced CDH lung with antenatal Dex treatment (group II). Weak and occasional TNF-α mRNA expression in type II pneumocytes with an increase in air saccule size and thinning of septal walls (×200). \.

Figure 1
figure 1figure 1figure 1

TNF-α mRNA expression. (A) Control lung (group I). Weak and occasional TNF-α mRNA expression in type II pneumocytes in mature lung parenchyma (×200). (B) Nitrofen-induced CDH lung (group II). Abundant and strong TNF-α mRNA expression in type II pneumocytes in immature lung parenchyma (×200).

Full size image

ELISA.

TNF-α levels (TNF-α/soluble protein) were significantly elevated in group II (mean ± SD, 0.42 ± 0.11) compared with control lung extract (0.23 ± 0.08); (p< 0.01) (Fig. 2). In group III, TNF-α protein (0.28 ± 0.17) were significantly decreased compared with nitrofen-induced CDH lung extract (p< 0.05) (Fig. 2).

Figure 2
figure 2

ELISA. The amount of TNF-α was elevated significantly in group II (mean ± SD, 0.42 ± 0.11) compared with group I (0.23 ± 0.08) (p< 0.01). In group III, TNF-α protein (0.28 ± 0.17) was significantly decreased compared with group II (p< 0.05).

Full size image

DISCUSSION

TNF-α, well-recognized as a proinflammatory cytokine, generated mainly by macrophages, is implicated in the pathogenesis of a variety of lung diseases, such as adult respiratory distress syndrome, neonatal RDS, and pulmonary fibrosis (15–18). It has been demonstrated that TNF-α is synthesized locally in lung and functions in an autocrine/paracrine mode (16, 18). This potent cytokine is reported to generate chemotactic factors, inflammatory mediators, and induce various changes in pulmonary tissue, vascular endothelium, and the surfactant system (19). It has recently been shown that TNF-α interferes with the synthesis of surfactant. A previous study from our laboratory demonstrated markedly increased TNF-α mRNA expression in human newborns and stillborns CDH hypoplastic lungs (9). It was suggested that the up-regulated pulmonary TNF-α gene expression may contribute to development of the surfactant deficiency, the decreased pulmonary compliance, and the persistent pulmonary hypertension observed in CDH hypoplastic lung. The present study confirmed our previous observations that hypoplastic lungs in nitrofen-induced CDH in rats demonstrated markedly increased TNF-α mRNA expression and TNF-α protein levels similar to those seen in human CDH hypoplastic lung.

CDH lung shows structural and functional parenchymal immaturity, as though it was developmentally arrested in the canalicular or saccular stage of development (2). Lung morphogenesis requires epithelial-mesenchymal interactions with precise regulatory controls (20). Specific hormones and growth factors have been implicated in this regulation (21–23). A recent study reported that TNF-α has a marked dose-dependent stimulatory effect on branching morphogenesis in the development of fetal lung (8). In contrast, with the higher concentration of TNF-α, TNF-α has a markedly inhibitory effect on branching morphogenesis in the developing lung (8). It has been shown that TNF-α is a potent inhibitor of the biosynthesis of surfactant phospholipids and proteins by isolated human type II pneumocytes (24). Over-production of pulmonary TNF-α in hypoplastic CDH lung may cause structural and functional parenchymal immaturity.

The present study demonstrated that antenatal dexamethasone treatment dramatically down-regulated TNF-α mRNA and protein levels in nitrofen-induced CDH hypoplastic lung in rat. Glucocorticoids are widely used in infants with RDS and result in improving pulmonary compliance, reduce alveolar proteinase activity, and frequently decrease the duration of ventilation and improve outcome (25). Over-expression of TNF-α has been implicated in the pathogenesis of ARDS, inducing the alterations in surfactant synthesis, composition, and function (7). Dexamethasone treatment has been shown to reduce the concentration of TNF-α in the bronchoalveolar secretions of patients with ARDS. Glucocorticoids influence numerous physiologic and pathologic processes. Glucocorticoids have been reported to have a wide range of effects on lung morphogenesis, promoting pulmonary maturation, type II pneumocyte histodifferentiation, and surfactant protein A expression (26, 27). Glucocorticoids enter the target cell primary by diffusion, interact with intracellular receptors, and then mediate their effects by regulation of specific genes or their products (28). There is a large body of evidence that glucocorticoids suppress TNF-α production (29). In vitro studies have demonstrated that dexamethasone will suppress the production of TNF-α in human and murine macrophages (30). It has been shown that glucocorticoids inhibit TNF-α at both the transcriptional and translational level by blocking gene transcription and mRNA mobilization (30). The therapeutic effects of glucocorticoid treatment in ARDS may be mediated by reduced production of pulmonary TNF-α (7). Histologic, morphologic, and quantitative biochemical similarities have been shown between the fetus/newborn with CDH and the surfactant-deficient newborn with RDS (3, 4). Recent studies demonstrated that antenatal glucocorticoid treatment in experimentally induced CDH hypoplastic lung results in an increase in air saccule size, thinning of septal walls, maturation of pulmonary parenchymal interstitium, acceleration of surfactant expression in type II cells, and increase in lung compliance, distensibility and functional residual capacity (12, 13, 31).

The precise mechanism by which antenatal glucocorticoids treatment promotes parenchymal maturation in CDH hypoplastic lung is not known. Steroidal growth of lung morphogenesis is mediated by regulating the effects of growth factors, cytokines, and their receptors (32). Our findings of barely detectable TNF-α mRNA expression and markedly reduced TNF-α protein levels in the dexamethasone-treated CDH lung suggest that the reduction of local TNF-α, which is up-regulated in CDH hypoplastic lung, may be one contributing mechanism by which antenatal glucocorticoid therapy improves lung parenchymal immaturity, including surfactant.