FGF-10 is decreased in bronchopulmonary dysplasia and suppressed ...

Am J Physiol Lung Cell Mol Physiol 292: L550–L558, 2007. First published October 27, 2006; doi:10.1152/ajplung.00329.2006.

FGF-10 is decreased in bronchopulmonary dysplasia and suppressed by Toll-like receptor activation John T. Benjamin,1 Rebekah J. Smith,1 Brian A. Halloran,1 Timothy J. Day,4 David R. Kelly,4 and Lawrence S. Prince1,2,3 Departments of 1Pediatrics, 2Physiology and Biophysics, 3Cell Biology, and 4Pathology and Laboratory Medicine, Children’s Hospital of Alabama, University of Alabama at Birmingham, Birmingham, Alabama Submitted 25 August 2006; accepted in final form 24 October 2006

Benjamin, John T., Rebekah J. Smith, Brian A. Halloran, Timothy J. Day, David R. Kelly, Lawrence S. Prince. FGF-10 is decreased in bronchopulmonary dysplasia and suppressed by Toll-like receptor activation. Am J Physiol Lung Cell Mol Physiol 292: L550–L558, 2007. First published October 27, 2006; doi:10.1152/ajplung.00329.2006.— Many extremely preterm infants continue to suffer from bronchopulmonary dysplasia, which results from abnormal saccular-stage lung development. Here, we show that fibroblast growth factor-10 (FGF10) is required for saccular lung development and reduced in the lung tissue of infants with bronchopulmonary dysplasia. Although exposure to bacteria increases the risk of bronchopulmonary dysplasia, no molecular target has been identified connecting inflammatory stimuli and abnormal lung development. In an experimental mouse model of saccular lung development, activation of Toll-like receptor 2 (TLR2) or Toll-like receptor 4 (TLR4) inhibited FGF-10 expression, leading to abnormal saccular airway morphogenesis. In addition, Toll-mediated FGF-10 inhibition disrupted the normal positioning of myofibroblasts around saccular airways, similar to the mislocalization of myofibroblasts seen in patients with bronchopulmonary dysplasia. Reduced FGF-10 expression may therefore link the innate immune system and impaired lung development in bronchopulmonary dysplasia. innate immunity; lung development; branching morphogensis

preterm infants has improved over the past two decades, many former preterm children suffer severe sequelae of being born early (13, 29). Each year, up to 10,000 infants in the United States develop bronchopulmonary dysplasia (BPD), a chronic lung disease characterized by airway obstruction and defective gas exchange (National Heart, Lung, and Blood Institute Diseases and Conditions Index; http://www.nhlbi.nih.gov/health/dci/index.html). BPD predisposes children to subsequent respiratory infections, right heart failure, and death during the first year of life (19, 38). The common hallmark of BPD is abnormal lung development, with decreased saccular airway branching and fewer, larger alveoli (10). The cellular and molecular mechanisms leading to abnormal lung development in BPD are unknown, and multiple factors may contribute to BPD pathogenesis. However, exposure to bacterial infections and inflammation increases the risk of developing BPD and may therefore inhibit lung morphogenesis (20). Lung development in BPD appears to go awry during the saccular stage of development. Infants born between 23 and 28 wk gestation are at the highest risk of BPD; the lungs of infants at this gestation are transitioning from the canalicular to the

WHILE THE SURVIVAL OF EXTREMELY

Address for reprint requests and other correspondence: Lawrence S. Prince, Dept. of Pediatrics, Division of Neonatology, Univ. of Alabama at Birmingham, VH648C, 1670 Univ. Boulevard, Birmingham, AL 35294 (e-mail: [email protected]). L550

saccular stage of development. During normal saccular lung development, terminal saccules randomly branch into alveolar ducts (9). Myofibroblasts and endothelial cells position themselves around the alveolar ducts, where they will later drive alveolar septation (8, 44). Although many factors regulating branching morphogenesis of the large-conducting airways and bronchi have been identified, less is known about branching during this saccular stage of development. Exposure of the saccular-stage lung to bacteria, either in utero or following delivery, appears to interfere with normal morphogenesis. Chorioamnionitis (infection and inflammation of the placenta and/or amniotic membranes) increases the likelihood that a mother will deliver preterm (35). In addition, infants born early due to chorioamnionitis are at increased risk of BPD (42). This does not appear to result from acute lung injury, as preterm infants exposed to chorioamnionitis often have less severe respiratory distress and improved lung function following delivery (11). These clinical observations suggest exposure to bacteria and inflammation can disrupt normal saccular lung development. Better understanding of saccular lung development and how inflammation could disrupt this process may provide insight into BPD pathogenesis. Fibroblast growth factor-10 (FGF-10) regulates branching morphogenesis during the earliest stages of lung development (4). Expressed by mesenchymal cells, FGF-10 promotes elongation and branching of developing bronchi. Mice lacking either FGF-10 or its receptor FGFR2 do not develop lungs (31, 36). Overexpression of a soluble, dominant-negative receptor for FGFR2 during fetal lung development inhibited lung branching and caused emphysema (15). These data demonstrate the important role of FGF-10 in lung development. In addition to being expressed in the lung mesenchyme very early in development, FGF-10 continues to be expressed in the saccular stage of lung development. However, the role of FGF-10 during the saccular stage of morphogenesis is unclear. BPD likely originates in the saccular stage, as the preterm infants at the highest risk of BPD are born at this time. Because abnormal saccular airway branching may contribute to BPD pathogenesis, we examined the function of FGF-10 in saccular development and its possible dysregulation in BPD. MATERIALS AND METHODS

Reagents and mice. Phenol-extracted, gel-purified Escherichia coli LPS (O55:B5) and mouse monoclonal anti-␣-smooth muscle actin (␣-SMA) antibody were from Sigma (St. Louis, MO). We obtained 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.

1040-0605/07 $8.00 Copyright © 2007 the American Physiological Society

http://www.ajplung.org

TOLL SIGNALING LEADS TO BRONCHOPULMONARY DYSPLASIA THROUGH SUPPRESSION OF FGF-10

recombinant FGF-10 from R&D Systems (Minneapolis, MN) and the TLR2 agonist Pam3Cys-Ser-(Lys)4 from Calbiochem (San Diego, CA). Anti-FGF-10 antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA) and R&D. SYTO13 and Alexa-conjugated secondary antibodies were purchased from Invitrogen (Carlsbad, CA). BALB/cJ and C.C3H.Tlr4Lpsd mice were obtained from Jackson Laboratories (Bar Harbor, ME) and maintained in pathogenfree facilities. For timed matings, the day of vaginal plug determination was defined as E0. All animal procedures were reviewed, approved, and performed in accordance with the policy of the Institutional Animal Care and Use Committee at the University of Alabama at Birmingham. Mouse fetal lung explant and mesenchyme isolation and culture. E16 female mice were euthanized with pentobarbital sodium (150 mg/kg ip). Using sterile technique, we isolated the fetal mouse lungs and dissected the lungs so that they were free of surrounding structures. The lung tissue was minced into 0.5–1 mm3 cubes using dissecting scissors and cultured on an air-liquid interface using permeable supports (Costar Transwell, Corning Incorporated, Acton, MA) and serum-free DMEM. Explants were cultured at 37°C in 95% air-5% CO2 for up to 72 h (12, 32, 33). LPS was included in culture media at a concentration of 250 ng/ml. Real-time PCR. We isolated total RNA from fetal lung explants and cultured mesenchyme using Trizol reagent and standard protocols. First-strand cDNA was synthesized using oligo-dT primers and Moloney murine leukemia virus (MMLV) reverse transciptase (Superscript II, Invitrogen). We designed real-time PCR primers using Beacon Design software (Bio-Rad, Hercules, CA). Each primer pair was validated by performing electrophoresis and melting temperature analysis of the PCR product. Standard concentration curves were done for each primer pair used. Two-step real-time PCR was performed with a Bio-Rad MyiQ thermocycler and SYBR Green detection system (Bio-Rad). We normalized gene expression to GAPDH in each sample. The 2⫺⌬⌬CT method was used to compare gene expression levels between samples. Gene silencing using small interfering RNA. We transfected lung explants with FGF-10 small interfering RNA (siRNA) oligonucleotides (Smartpool, Dharmacon, Chicago, IL) using Oligofectamine (Invitrogen). Sixty picomoles of siRNA oligonucleotides were complexed with 4 ␮l of oligofectamine in a total volume of 600 ␮l DMEM in a sterile 1.8-ml microcentrifuge tube. Freshly isolated E16 lung explants were placed in the siRNA-containing media and incubated for 4 h at 37°C. Explants were then removed and directly seeded onto Transwell filters for culture. Cy3-labeled siRNA oligonucleotides against cyclophilin B (CyB, Dharmacon) were used as a control in all of the transfection experiments. Luciferase measurement. We used a luciferase reporter cell line to assay for TGF-␤ activity. Briefly, conditioned media from control and LPS-treated explants were added to transfected mink lung cells (TMLC) expressing a plasminogen activator ihibitor (PAI)-luciferase reporter gene. The TMLC cells were cultured for up to 72 h. The cells were solubilized at different time points and luminescence was measured with a single-tube luminometer (Turner Biosciences, Sunnyvale, CA) and using Steady-Glo luciferase assay reagent (Promega, Madison, WI). Image analysis. We have previously described our procedure for quantifying saccular explant branching (32). The number of peripheral airway branches was normalized to explant area using Histometrix software (Andor Bioimaging, South Windsor, CT). Control explants from different litters in separate experiments formed 35– 42 branches/ mm2 after 72 h. Because of this inherent variability, separate control explants were included in each experiment. For each experiment, explants from at least three separate litters of mice were used. To measure airway length, we labeled nuclei of formaldehyde-fixed explants with SYTO13 and acquired optical sections along the explant periphery using a Leica TCS SP2 laser scanning confocal microscope (63⫻ oil objective). We then measured airway length (from branch AJP-Lung Cell Mol Physiol • VOL

L551

to leading tip) using LCS software (Leica, Wetzlar, Germany). All image analysis data were stored and analyzed in Excel spreadsheets (Microsoft, Redmond, WA). Immunostaining. Formalin-fixed, paraffin-embedded tissue sections were prepared from the autopsy specimens of five patients with BPD and five age-matched controls. Ages at time of death ranged between 2 and 18 mo. The clinical diagnosis of BPD was made by the attending physician and confirmed by the pathologist performing the autopsy. Sections were stained by standard immunohistochemical techniques, including antigen recovery in sodium citrate (10 mM, pH 6) and quenching of endogenous peroxidase activity using 3% H2O2 in methanol. Immunostaining was detected using avidin-biotin-horseradish peroxidase complexes (Vector Laboratories, Burlington, CA) and aminoethylcarbazol. Ten random images were obtained from each slide. The number of FGF-10-positive cells per field was measured using Histometrix software. Investigators were blinded during image capture and analysis. Data were matched to an identification key following analysis. Use of human tissue was approved by the Institutional Review Board at the University of Alabama at Birmingham. For immunofluorescent labeling of mouse fetal lung explants, we fixed and permeabilized explants with 4% formaldehyde followed by 0.5% Triton X-100. Nonspecific binding was blocked with PBS containing 20% donkey serum and 5% BSA. We then incubated the explants overnight with primary antibody at 4°C followed by 2 h incubation with Alexa-594 donkey anti-mouse secondary. Nuclei were labeled with SYTO13, and explants were mounted between a glass slide and coverslip. We then obtained confocal images of explants using a Leica TCS SP2 laser-scanning confocal microscope. Images were acquired using a ⫻10 dry objective and a ⫻63 oil objective at 512 ⫻ 512 pixel resolution. Single scans were obtained using the ⫻10 objective in Fig. 7, C–E. Line-averaged z-sections obtained with the ⫻63 objective are shown in Fig. 7, F–H. All images were stored as TIFF files and exported to Adobe Photoshop for sizing. Statistical analysis. Data are reported as means ⫾ SE. Statistical analyses were performed using SigmaStat 3.1 software. For two group comparisons, unpaired Student’s t-tests were performed. For comparison of multiple groups to the same control, data were analyzed using one-way ANOVA. Intergroup comparisons were done using the Holm-Sidak or Dunn’s Method. A P value ⬍0.05 was used for significance. RESULTS

We initially tested whether FGF-10 expression was decreased in the lungs of infants with BPD by immunostaining human lung tissue sections obtained at autopsy. Using a polyclonal antibody against FGF-10, we compared lung tissue from five infants who died with BPD to five age-matched controls. In control lungs, FGF-10 protein was detected within the lung interstitium and adjacent to alveolar epithelia (Fig. 1, A and B). However, lung tissue from infants with BPD contained fewer FGF-10-positive cells (Fig. 1, C–E). These data suggested that FGF-10 expression was decreased in patients with BPD. FGF-10 is required for early embryonal lung development (36). In BPD, however, the saccular stage of lung development, which occurs later in gestation, is disrupted (10). We therefore examined the role of FGF-10 in saccular lung development using an E16 fetal mouse lung explant model of saccular-stage development. At this point in development, the fetal mouse lung is transitioning from the canalicular stage to the saccular stage of morphogenesis. E16 explants were isolated and cultured for up to 72 h in the presence of antibodies against FGF-10 or Escherichia coli LPS. Figure 2, A–D shows lowmagnification images of representative explants following 72 h

292 • FEBRUARY 2007 •

www.ajplung.org

L552


Fig. 1. Fibroblast growth factor-10 (FGF-10) is reduced in bronchopulmonary dysplasia (BPD). A–D: immunostaining for FGF-10 expression in human lung tissue sections. Representative images shown from tissue sections from a patient with BPD (C and D) and an age-matched control (A and B). Scale bar: 100 ␮m in A and B. Scale bar: 50 ␮m in C and D. E: mean number of FGF-10-positive cells per field in 5 BPD patients and 5 age-matched controls. Ten random sections were imaged for each individual (*P ⬍ 0.01 by unpaired Student’s t-test).

of culture. Both anti-FGF-10 antibodies and LPS appeared to prevent the normal branching and elongation of saccular airways along the periphery of the explants. Higher-magnification images further demonstrated the effects on saccular airway morphogenesis (Fig. 2, E–H). In control explants, saccular airways branched and extended through the mesenchyme (Fig. 2E). However, LPS-exposed airways had fewer branches and appeared shorter (Fig. 2H). Anti-FGF-10 antibodies decreased airway branching when added to the media, similar to the effects seen with LPS (Fig. 2, F and G). This effect appeared concentration dependent and suggested that interfering with FGF-10 could inhibit saccular airway branching and elongation (Fig. 2I). To further test the requirement of FGF-10 in saccular airway branching, we used siRNA to inhibit FGF-10 expression. Transfecting E16 saccular explants with FGF-10-specific siRNA oligonucleotides reduced FGF-10 expression compared with controls (Fig. 2J), and inhibited saccular airway branching (Fig. 2K). siRNA oligonucleotides against cyclophilin B (CyB) had no effect on FGF-10 expression or saccular branching. Therefore decreased FGF-10 expression leads to abnormal saccular lung development. Because exposure to bacteria increases the risk of BPD and patients with BPD have decreased FGF-10 expression, we tested whether bacterial products could inhibit FGF-10 expression. In our previous studies, LPS inhibited saccular airway branching both in vivo and in fetal mouse lung explants (32). We hypothesized that these changes were due to decreased FGF-10. To initially test this hypothesis, we measured FGF-10 expression in saccular-stage explants using real-time PCR. LPS decreased FGF-10 gene expression in explants to ⬃50% of control levels (Fig. 3A). Consistent with signaling through Toll-like receptors, LPS had no effect on FGF-10 expression in explants from C.C3H.Tlr4Lpsd (TLR4LPSd) mice that have a loss of function mutation in the tlr4 gene (41). If reduced FGF-10 expression were responsible for the decrease in airway branching seen with LPS, then addition of exogenous FGF-10 AJP-Lung Cell Mol Physiol • VOL

might prevent the effects of LPS. Adding recombinant FGF-10 increased saccular airway branching in both control and LPStreated explants, preventing the effects of LPS (Fig. 3, B and C). These data also demonstrated that LPS did not render the saccular airways resistant to FGF-10. LPS may therefore decrease saccular airway branching through inhibiting FGF-10 expression. We next tested whether the effects of LPS on FGF-10 expression were specific and could lead to altered expression of genes downstream of FGF-10. LPS had no effect on the expression of FGF-7 or FGF-9 (Fig. 3D). LPS also inhibited expression of bone morphogenic protein 4 (Bmp4), bone morphogenic protein receptor 1A (Bmpr1A), and Perp (Fig. 3E), each induced by FGF-10 (26). However, LPS did not have the same effect on Erm and Pea3, which are regulated by both FGF-10 and FGF-7 (24). These effects of LPS required TLR4, as LPS did not alter expression of Bmp4 or Bmpr1a in explants from C.C3H.Tlr4Lpsd mice. Therefore, LPS activation of TLR4 inhibited expression of FGF-10 and subsequent downstream targets of FGF-10. E. coli LPS is a potent, well-described activator of the innate immune system (5). In addition to Gram-negative bacteria, other microbes, including streptococci, mycoplasma, and ureaplasma can cause prenatal inflammation and increase the risk of BPD in preterm infants (21). If reduced FGF-10 expression is a central mechanism in BPD pathogenesis, then we predicted that other bacterial activators of Toll-like receptors would also inhibit FGF-10 expression. To specifically target another TLR, we treated saccular explants with Pam3Cys-Ser-(Lys)4, a TLR2-specific peptide agonist (1). Pam3Cys-Ser-(Lys)4 increased TNF-␣ expression, confirming that activation of TLR2 in our model could induce inflammation (Fig. 4A). Like LPS, Pam3Cys-Ser-(Lys)4 inhibited FGF-10 expression and decreased saccular airway branching (Fig. 4, B and C). These data were consistent with the polymicrobial nature of BPD and suggested that activation of multiple TLRs can inhibit saccular lung development through decreased FGF-10 expression.

292 • FEBRUARY 2007 •

www.ajplung.org


L553

Fig. 2. FGF-10 is required for saccular-stage lung branching in mouse fetal lung explants. A–H: brightfield microscopy images of saccular airways. E16 explants were cultured for 72 h in the presence of nonimmune rabbit IgG, Escherichia coli LPS (250 ng/ml), or increasing concentrations of anti-FGF-10 polyclonal antibody. A–D: low-magnification images of explants following 72 h of culture. Scale bar: 500 ␮m. E–H: higher-magnification images of saccular airways at the explant periphery showing decreased branching and elongation in anti-FGF-10-treated (F and G) and LPS-treated explants (H). Airway lumina are indicated by *; the leading edges of newly formed saccular airways are identified with arrows. Scale bar: 100 ␮m. I: number of peripheral airway branches in explants from each experimental group was counted and normalized to explant area. Control explants that were untreated were included along with explants cultured in the presence of nonimmune IgG (*P ⬍ 0.01 by ANOVA; n ⫽ 8 –11). J: explants transfected with FGF-10 siRNA oligonucleotides had decreased FGF-10 expression as measured by real-time PCR (*P ⬍ 0.05 by ANOVA; n ⫽ 6). siRNA oligonucleotides against cyclophilin B (CyB) had no effect on FGF-10 expression. K: saccular airway branching was measured in control explants (ctrl) or explants transfected with siRNA oligonucleotides against cyclophilin B (CyB) or FGF-10 (*P ⬍ 0.01 by ANOVA; n ⫽ 23).

LPS and Toll-like receptor signaling could inhibit FGF-10 expression directly or through an intermediate pathway. Transforming growth factor ␤ (TGF-␤) and sonic hedgehog (shh) both suppress FGF-10 expression in early embryonal lung development (3, 6, 22, 37, 40). However, in our saccular-stage lung explant model, LPS did not increase TGF-␤1, TGF-␤2, TGF-␤3, or shh gene expression (Fig. 5, A and C). In addition, LPS did not increase the amount of AJP-Lung Cell Mol Physiol • VOL

bioactive TGF-␤1 released into the media when assayed by a cell reporter assay (Fig. 5B). We next isolated primary fetal lung mesenchymal cells from saccular-stage E16 mouse lungs. These cells constitutively express FGF-10, and adding LPS to the cell culture media inhibited FGF-10 expression (Fig. 5D). LPS can therefore inhibit FGF-10 gene expression in the fetal mouse lung mesenchyme in the absence of airway epithelia.

292 • FEBRUARY 2007 •

www.ajplung.org

L554


During early lung branching morphogenesis, FGF-10 directly stimulates airway elongation (4, 28, 43). To test whether LPS could inhibit saccular airway elongation through decreasing FGF-10 expression in an explant model, we measured the

lengths of the terminal airways by laser-scanning confocal microscopy (Fig. 6). LPS-treated saccular airways were shorter than controls, as would be expected if FGF-10 were required for saccular elongation (Fig. 6, B, D, E). The addition of

Fig. 3. LPS inhibits expression of FGF-10 through TLR4 activation. A: FGF-10 expression in wild-type BALBc/J and C.C3H.Tlr4Lpsd (TLR4Lpsd) explants following 72 h of culture with LPS. FGF-10 expression in each sample was measured by real-time PCR and compared with its own control (*P ⬍ 0.01 by unpaired Student’s t-test; n ⫽ 18). B: brightfield micrographs of control and LPStreated E16 fetal mouse lung explants with recombinant FGF-10 added to the media. Scale bar: 50 ␮m. C: saccular airway branching in control and LPS-treated explants with 50 ng/ml and 1,000 ng/ml recombinant FGF10. The addition of FGF-10 prevented the defective branching seen with LPS. Saccular branches were measured at 72 h of culture. (*P ⬍ 0.05 compared with control by ANOVA; #P ⬍ 0.01 compared with LPS by ANOVA; n ⫽ 7–11). D: FGF-7 and FGF-9 gene expression was measured in control and LPS-treated explants by real-time PCR. (n ⫽ 9 for FGF-7; n ⫽ 10 for FGF-9). E: gene expression of Bmp4, Bmpr1a, Perp, Erm, and Pea3 was measured in control and LPS-treated explants by real-time PCR. (*P ⬍ 0.01 by unpaired t-test; n ⫽ 27). Gene expression of Bmp4 and Bmpr1a was also measured in control and LPS-treated explants from C.C3HTlr4Lpsd mice (TLR4Lpsd, gray bars, n ⫽ 3).

AJP-Lung Cell Mol Physiol • VOL

292 • FEBRUARY 2007 •

www.ajplung.org


L555

Fig. 4. TLR2 activation decreases FGF-10 expression and inhibits saccular airway branching. A and B: gene expression of TNF-␣ (A) and FGF-10 (B) in control explants and explants treated with LPS (250 ng/ml) and the TLR2 agonist Pam3Cys-Ser-(Lys)4 (Pam3Cys; 0.1 ␮g/ml and 1 ␮g/ml). Expression was measured by real-time PCR. Both LPS and Pam3Cys-Ser-(Lys)4 increased TNF-␣ and decreased FGF-10 expression (*P ⬍ 0.05 by ANOVA; n ⫽ 6). C: quantification of saccular airway branching in explants treated with Pam3Cys-Ser-(Lys)4 for 72 h. (*P ⬍ 0.05 by ANOVA; n ⫽ 8).

FGF-10 increased airway length and prevented the effects of LPS. The effects of LPS on saccular airway branching and elongation appear consistent with decreased mesenchymal FGF-10 expression. Decreased saccular airway branching could reduce the lung surface area available for gas exchange. Although infants with BPD often have decreased lung surface area, they also lack normal alveolarization (16, 30). Alveolar septa form when myofibroblasts and endothelia migrate inward toward the alveolar lumen. Alveolar myofibroblasts expressed ␣-SMA (8,

44) and localized to the tips of alveolar septa and around alveolar ducts in human lung sections from control infants (Fig. 7A). However, in lung tissue from infants with BPD, thick bands of ␣-SMA-positive myofibroblasts were seen around terminal airways that lacked alveolar septa (Fig. 7B). We next tested whether LPS could regulate myofibroblast localization in saccular-stage lung explants. In control explants, myofibroblasts were found throughout the mesenchyme, lining the terminal saccular airways (Fig. 7, C and F). Myofibroblasts tended to be spaced evenly

Fig. 5. LPS decreases FGF-10 in fetal lung mesenchyme independent of TGF-␤ and sonic hedgehog (shh). A: LPS did not increase TGF-␤1, TGF-␤2, and TGF-␤3 expression in fetal mouse lung explants. Gene expression was measured by real-time PCR (n ⫽ 9). B: conditioned media from control and LPS-treated explants was added to transfected mink lung cells (TMLC) expressing a plasminogen activator inhibitor (PAI)-luciferase reporter gene. In these cells, luciferase expression is strongly induced by TGF-␤1. Both control and LPS-treated samples released some TGF-␤1 into the media by 72 h of culture, but there was no significant difference between control and LPS-treated samples. (*P ⬍ 0.05 by ANOVA; n ⫽ 6). C: real-time PCR was used to measure shh gene expression in control and LPStreated explants. LPS did not significantly change shh expression in fetal mouse lung explants (n ⫽ 12). D: FGF-10 gene expression in primary cultures of fetal mouse lung mesenchyme as measured by real-time PCR (*P ⬍ 0.05 by ANOVA; n ⫽ 9).


292 • FEBRUARY 2007 •

www.ajplung.org

L556


Fig. 6. LPS inhibits FGF-mediated saccular airway elongation. A–C: confocal images of saccular airways following 72 h of culture. Explants were labeled with SYTO13 nuclear stain and imaged using a laser-scanning confocal microscope. Representative sections are shown. Scale bar: 50 ␮m. D and E: measurement of saccular airway length in explants cultured for 48 h (D) or 72 h (E). (*P ⬍ 0.01 compared by ANOVA to control; #P ⬍ 0.01 compared by ANOVA to LPS; for each condition and time point, 144 –225 airways were measured).

around saccular airways. However, in LPS-treated explants, myofibroblasts formed thick bands of cells within the explant mesenchyme (Fig. 7, D and G). Very few myofibroblasts were found immediately adjacent to the most peripheral airways in LPS-treated explants. The addition of exogenous FGF-10 appeared to prevent this effect, as terminal airways exposed to LPS and FGF-10 were enveloped by individual ␣-SMA-positive myofibroblasts (Fig. 7, E and H). Inhibition of FGF-10 by LPS not only leads to shorter, more simplified saccular airways but also disrupts the positioning of myofibroblasts around these airways. DISCUSSION

In demonstrating that Toll-like receptor agonists can regulate FGF-10, this study links the innate immune system and a basic mechanism of fetal lung development. Toll signaling in Drosophila regulates both dorsal-ventral patterning in embryos and the innate response to pathogens in adult flies (23). Therefore, the same pathway can influence both development and the response to microbial products. We found that a similar overlap may exist between mammalian development and innate immunity. Activation of the Toll-like receptors TLR2 and TLR4 can disrupt normal fetal mouse lung development by inhibiting FGF-10 expression. Toll-like receptors are expressed in multiple tissues in addition to the fetal lung during development and pregnancy (2, 25, 33). The expression of TLR2 and TLR4 was increased in AJP-Lung Cell Mol Physiol • VOL

the chorioamniotic membranes of patients with chorioamnionitis and at risk of preterm delivery (21). This increase may result from both increased numbers of inflammatory cells and increased expression of Toll-like receptors on the surface of amniotic epithelia. Tissue or cellular injury may activate the innate immune pathway through release of endogenous Toll-like receptor agonists (34). However, low levels of bacterial products may also contribute to the injury response (39). While the role, if any, of Toll-like receptors and activation by endogenous ligands during in utero fetal development is not known, preterm infants are commonly exposed to bacteria as they try to complete normal fetal development. We have not yet determined whether direct activation of Toll-like receptors or the production of inflammatory cytokines or chemokines mediates the effects on lung development and FGF-10 expression we have measured. This issue is important in the extrapolation of data from our in vitro explant model, where the mesenchyme is exposed to LPS, to the in vivo situation, where bacterial products may have reduced access to the lung interstitium. However, inflammatory mediators produced downstream of localized Toll-like receptor activation may have greater access to a variety of tissues. The saccularstage explant model also lacks inflammatory cells that could be present in the lungs of patients with BPD. Our data show that FGF-10 signaling is critical for saccular airway development in addition to its previously described role during early lung budding. Saccular-stage airways branch randomly compared with the dichotomous branching of bronchi earlier in development. Although saccular airways are lined with differentiated distal type I and type II epithelia and not encircled with smooth muscle, their branching and elongation may well be regulated by similar mechanisms as the more proximal airways. The degree to which canalicular and saccular-stage development can progress in preterm infants determines their ability to survive with ventilatory assistance in the intensive care unit (14). Therefore, a more thorough understanding of the mechanisms regulating these stages of lung morphogenesis may have significant clinical implications. In the human lung specimens that we examined, infants that died with BPD had reduced FGF-10 protein expression. Because patients dying from BPD have complex clinical courses, human autopsy specimens may not be as experimentally rigorous as that from in vitro models. Additionally, our patient samples were from infants at a more advanced stage of lung development than our experimental explants. However, our FGF-10 immunostaining data are consistent with our hypothesis that reduced FGF-10 expression may contribute to BPD. We have not yet determined whether alterations in FGF-10 expression in preterm infants early in life can predict BPD or how FGF-10 levels might change during the course of hospitalization of preterm infants. Perhaps because FGF-10 is expressed in mesenchymal cells and avidly binds to extracellular matrix (17, 18), we have not been able to detect FGF-10 in blood or tracheal aspirate samples from normal or BPD patients. Nor have we been able to measure FGF-10 mRNA in formalin-fixed autopsy specimens. We hope to develop future approaches for measuring FGF-10 and its biological activity in vivo. While we recognize the possible multifactorial nature of BPD (7), our experimental data

292 • FEBRUARY 2007 •

www.ajplung.org


L557

Fig. 7. Abnormal localization of myofibroblasts in patients with BPD and LPS-treated saccular lung explants. A and B: myofibroblast localization in human lung tissue sections. A: representative ␣-SMA staining in three different control patients showing myofibroblasts at alveolar septal tips, evenly spaced around alveolar ducts. B: ␣-SMA staining in tissue sections from patients with BPD. Myofibroblasts form thick bands surrounding alveolar ducts. No alveolar septa are seen. Scale bar: 100 ␮m. C–E: low-magnification fluorescence images of explants cultured for 72 h under control conditions, with LPS, or with LPS and FGF-10 (200 ng/ml). Myofibroblasts are labeled red using an antibody against ␣-SMA. Nuclei are labeled green with SYTO13. Scale bar: 50 ␮m. F–H: confocal sections through saccular airways along the periphery of E16 explants. Airway lumina are indicated with *; arrows point out myofibroblasts lining saccular airways in control (F) and LPS ⫹ FGF-10 explants (H), but only along the explant periphery in LPS-treated explants (G). Scale bar: 50 ␮m.

suggest that reduced FGF-10 due to bacterial exposure and Toll-mediated signaling may play a significant role in BPD pathogenesis. Reduced FGF-10 expression leads to decreased saccular branching and abnormal myofibroblast localization around terminal airways, both histological features of BPD. FGF-10 is similarly required for smooth muscle localization during early lung development (27). Although the role of myofibroblasts during saccular airway branching is not completely understood, myofibroblasts migrate into secondary septa and alveolar clefts during the later alveolar stage of development (44). Defects in myofibroblast migration are associated with defective alveolarization (8). The thick bands of myofibroblasts seen both in BPD lung sections and in LPS-treated mouse lung explants could be a similar cellular finding in abnormal lung morphogenesis. If exposure to bacterial products alters the cell morphology and positioning of myofibroblasts around saccular airways, then these cells may not be in the correct location to AJP-Lung Cell Mol Physiol • VOL

promote alveolar septation and maturation of the alveolar-capillary bed. Therefore, a cellular defect during saccular development may lead to abnormal morphogenesis in later stages. While we have not identified all of the factors regulating FGF-10 expression during saccular development, bacterial products clearly inhibit FGF-10 expression in the lung mesenchyme. In developing therapeutic strategies for BPD, we should investigate how the fetal lung mesenchyme responds to inflammation and pathogens. Maintaining mesenchymal FGF-10 expression or directing delivery of FGF-10 to the mesenchyme may represent novel approaches for BPD prevention and treatment. GRANTS This work was supported by the American Lung Association, the Parker Francis Families Foundation, and the Child Health Research Center at the University of Alabama at Birmingham.

292 • FEBRUARY 2007 •

www.ajplung.org

L558


REFERENCES 1. Agrawal S, Agrawal A, Doughty B, Gerwitz A, Blenis J, Van Dyke T, Pulendran B. Cutting edge: different Toll-like receptor agonists instruct dendritic cells to induce distinct Th responses via differential modulation of extracellular signal-regulated kinase-mitogen-activated protein kinase and c-Fos. J Immunol 171: 4984 – 4989, 2003. 2. Beijar EC, Mallard C, Powell TL. Expression and subcellular localization of TLR-4 in term and first trimester human placenta. Placenta 27: 322–326, 2006. 3. Bellusci S, Furuta Y, Rush MG, Henderson R, Winnier G, Hogan BL. Involvement of Sonic hedgehog (Shh) in mouse embryonic lung growth and morphogenesis. Development 124: 53– 63, 1997. 4. Bellusci S, Grindley J, Emoto H, Itoh N, Hogan BL. Fibroblast growth factor 10 (FGF10) and branching morphogenesis in the embryonic mouse lung. Development 124: 4867– 4878, 1997. 5. Beutler B, Hoebe K, Du X, Ulevitch RJ. How we detect microbes and respond to them: the Toll-like receptors and their transducers. J Leukoc Biol 74: 479 – 485, 2003. 6. Bitgood MJ, McMahon AP. Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo. Dev Biol 172: 126 –138, 1995. 7. Bland RD. Neonatal chronic lung disease in the post-surfactant era. Biol Neonate 88: 181–191, 2005. 8. Bostrom H, Willetts K, Pekny M, Leveen P, Lindahl P, Hedstrand H, Pekna M, Hellstrom M, Gebre-Medhin S, Schalling M, Nilsson M, Kurland S, Tornell J, Heath JK, Betsholtz C. PDGF-A signaling is a critical event in lung alveolar myofibroblast development and alveogenesis. Cell 85: 863– 873, 1996. 9. Burri PH. Fetal and postnatal development of the lung. Annu Rev Physiol 46: 617– 628, 1984. 10. Coalson JJ. Pathology of new bronchopulmonary dysplasia. Semin Neonatol 8: 73– 81, 2003. 11. Dempsey E, Chen MF, Kokottis T, Vallerand D, Usher R. Outcome of neonates less than 30 weeks gestation with histologic chorioamnionitis. Am J Perinatol 22: 155–159, 2005. 12. Dieperink HI, Blackwell TS, Prince LS. Hyperoxia and apoptosis in developing mouse lung mesenchyme. Pediatr Res 59: 185–190, 2006. 13. Herber-Jonat S, Schulze A, Kribs A, Roth B, Lindner W, Pohlandt F. Survival and major neonatal complications in infants born between 22 0/7 and 24 6/7 weeks of gestation (1999 –2003). Am J Obstet Gynecol 195: 16 –22, 2006. 14. Hislop AA. Airway and blood vessel interaction during lung development. J Anat 201: 325–334, 2002. 15. Hokuto I, Perl AK, Whitsett JA. Prenatal, but not postnatal, inhibition of fibroblast growth factor receptor signaling causes emphysema. J Biol Chem 278: 415– 421, 2003. 16. Husain AN, Siddiqui NH, Stocker JT. Pathology of arrested acinar development in postsurfactant bronchopulmonary dysplasia. Hum Pathol 29: 710 –717, 1998. 17. Izvolsky KI, Shoykhet D, Yang Y, Yu Q, Nugent MA, Cardoso WV. Heparan sulfate-FGF10 interactions during lung morphogenesis. Dev Biol 258: 185–200, 2003. 18. Izvolsky KI, Zhong L, Wei L, Yu Q, Nugent MA, Cardoso WV. Heparan sulfates expressed in the distal lung are required for Fgf10 binding to the epithelium and for airway branching. Am J Physiol Lung Cell Mol Physiol 285: L838 –L846, 2003. 19. Jacob SV, Coates AL, Lands LC, MacNeish CF, Riley SP, Hornby L, Outerbridge EW, Davis GM, Williams RL. Long-term pulmonary sequelae of severe bronchopulmonary dysplasia. J Pediatr 133: 193–200, 1998. 20. Jobe AH. Antenatal factors and the development of bronchopulmonary dysplasia. Semin Neonatol 8: 9 –17, 2003. 21. Kim YM, Romero R, Chaiworapongsa T, Kim GJ, Kim MR, Kuivaniemi H, Tromp G, Espinoza J, Bujold E, Abrahams VM, Mor G. Toll-like receptor-2 and -4 in the chorioamniotic membranes in spontaneous labor at term and in preterm parturition that are associated with chorioamnionitis. Am J Obstet Gynecol 191: 1346 –1355, 2004. 22. Lebeche D, Malpel S, Cardoso WV. Fibroblast growth factor interactions in the developing lung. Mech Dev 86: 125–136, 1999.


23. Lemaitre B, Nicolas E, Michaut L, Reichhart JM, Hoffmann JA. The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell 86: 973–983, 1996. 24. Liu Y, Jiang H, Crawford HC, Hogan BL. Role for ETS domain transcription factors Pea3/Erm in mouse lung development. Dev Biol 261: 10 –24, 2003. 25. Lotz M, Gutle D, Walther S, Menard S, Bogdan C, Hornef MW. Postnatal acquisition of endotoxin tolerance in intestinal epithelial cells. J Exp Med 203: 973–984, 2006. 26. Lu J, Izvolsky KI, Qian J, Cardoso WV. Identification of FGF10 targets in the embryonic lung epithelium during bud morphogenesis. J Biol Chem 280: 4834 – 4841, 2005. 27. Mailleux AA, Kelly R, Veltmaat JM, De Langhe SP, Zaffran S, Thiery JP, Bellusci S. Fgf10 expression identifies parabronchial smooth muscle cell progenitors and is required for their entry into the smooth muscle cell lineage. Development 132: 2157–2166, 2005. 28. Park WY, Miranda B, Lebeche D, Hashimoto G, Cardoso WV. FGF-10 is a chemotactic factor for distal epithelial buds during lung development. Dev Biol 201: 125–134, 1998. 29. Parker RA, Lindstrom DP, Cotton RB. Improved survival accounts for most, but not all, of the increase in bronchopulmonary dysplasia. Pediatrics 90: 663– 668, 1992. 30. Parker TA, Abman SH. The pulmonary circulation in bronchopulmonary dysplasia. Semin Neonatol 8: 51– 61, 2003. 31. Peters K, Werner S, Liao X, Wert S, Whitsett J, Williams L. Targeted expression of a dominant negative FGF receptor blocks branching morphogenesis and epithelial differentiation of the mouse lung. EMBO J 13: 3296 –3301, 1994. 32. Prince LS, Dieperink HI, Okoh VO, Fierro-Perez GA, Lallone RL. Toll-like receptor signaling inhibits structural development of the distal fetal mouse lung. Dev Dyn 233: 553–561, 2005. 33. Prince LS, Okoh VO, Moninger TO, Matalon S. Lipopolysaccharide increases alveolar type II cell number in fetal mouse lungs through Toll-like receptor 4 and NF-kappaB. Am J Physiol Lung Cell Mol Physiol 287: L999 –L1006, 2004. 34. Rifkin IR, Leadbetter EA, Busconi L, Viglianti G, Marshak-Rothstein A. Toll-like receptors, endogenous ligands, and systemic autoimmune disease. Immunol Rev 204: 27– 42, 2005. 35. Romero R, Mazor M, Wu YK, Sirtori M, Oyarzun E, Mitchell MD, Hobbins JC. Infection in the pathogenesis of preterm labor. Semin Perinatol 12: 262–279, 1988. 36. Sekine K, Ohuchi H, Fujiwara M, Yamasaki M, Yoshizawa T, Sato T, Yagishita N, Matsui D, Koga Y, Itoh N, Kato S. Fgf10 is essential for limb and lung formation. Nat Genet 21: 138 –141, 1999. 37. Serra R, Pelton RW, Moses HL. TGF beta 1 inhibits branching morphogenesis and N-myc expression in lung bud organ cultures. Development 120: 2153–2161, 1994. 38. Smith VC, Zupancic JA, McCormick MC, Croen LA, Greene J, Escobar GJ, Richardson DK. Rehospitalization in the first year of life among infants with bronchopulmonary dysplasia. J Pediatr 144: 799 – 803, 2004. 39. Tsan MF, Gao B. Endogenous ligands of Toll-like receptors. J Leukoc Biol 76: 514 –519, 2004. 40. Urase K, Mukasa T, Igarashi H, Ishii Y, Yasugi S, Momoi MY, Momoi T. Spatial expression of Sonic hedgehog in the lung epithelium during branching morphogenesis. Biochem Biophys Res Commun 225: 161–166, 1996. 41. Vogel SN, Wax JS, Perera PY, Padlan C, Potter M, Mock BA. Construction of a BALB/c congenic mouse, C. C3H-Lpsd, that expresses the Lpsd allele: analysis of chromosome 4 markers surrounding the Lps gene. Infect Immun 62: 4454 – 4459, 1994. 42. Watterberg KL, Demers LM, Scott SM, Murphy S. Chorioamnionitis and early lung inflammation in infants in whom bronchopulmonary dysplasia develops. Pediatrics 97: 210 –215, 1996. 43. Weaver M, Dunn NR, Hogan BL. Bmp4 and Fgf10 play opposing roles during lung bud morphogenesis. Development 127: 2695–2704, 2000. 44. Yamada M, Kurihara H, Kinoshita K, and Sakai T. Temporal expression of alpha-smooth muscle actin and drebrin in septal interstitial cells during alveolar maturation. J Histochem Cytochem 53: 735–744, 2005.

292 • FEBRUARY 2007 •

www.ajplung.org