Objective
We hypothesized that fetal innate immune responses to lipopolysaccharide-induced chorioamnionitis would alter postnatal systemic immune and airway responsiveness.
Study Design
Ewes received intraamniotic injections with saline or lipopolysaccharide at 90, 100, and 110 days of gestation. Immune status and airway responsiveness were evaluated at term and at 7 weeks of age.
Results
At term, lymphocytes, monocytes, and neutrophils were significantly increased (respectively, 24-fold, 127-fold, and 31,000-fold) in lungs and blood monocytes became Toll-like receptor 2 responsive after lipopolysaccharide exposures. Furthermore, CD4 and CD4/CD25 lymphocytes were increased in thymus and lymph nodes. At 7 weeks, airway reactivity decreased and concentrations of CD8 cytotoxic T lymphocytes changed in the lungs and thymus relative to controls.
Conclusion
Early gestational lipopolysaccharide exposure increased leukocyte responsiveness at term. Decreased airway reactivity and changes in lymphocytes at 7 weeks postnatal demonstrate persistent effects of fetal exposure to LPS.
Preterm birth (<37 weeks) is frequently associated with chorioamnionitis. Recently, chorioamnionitis was correlated with a 4-fold increased risk for developing asthma in preterm infants after correction for confounding factors. This observation is in contrast to the hygiene hypothesis, in which infections during early childhood are associated with a decreased risk of subsequent asthma and allergic disease. These conflicting observations suggest that the time of exposure to bacterial products during childhood may have different effects when compared with antenatal exposures. For example, antenatal exposure to bacterial products may increase the risk of asthma in childhood, whereas postnatal exposure to bacterial products may protect from asthma. The mechanisms to explain the different outcomes are still unclear.
Chorioamnionitis-mediated inflammation exposes the fetus to proinflammatory mediators primarily by fetal breathing and swallowing of amniotic fluid. The fetal lung responds to mediators such as lipopolysaccharide (LPS), interleukin-1 (IL-1), and live ureaplasma with a local inflammatory response. Previous experiments revealed that exposure to chorioamnionitis changes the lymphocyte population in the posterior mediastinal lymph node (PMLN) that drains the lung and causes systemic immune modulation of blood monocytes. Furthermore, the gut responses to LPS with altered inflammatory cells in the gut wall and regional lymph nodes. These acute fetal responses were modulated further with repeated exposure to induce LPS-tolerance and cross-tolerance to other proinflammatory Toll-like receptor (TLR) agonists. However, the relationship between exposure to antenatal inflammation and postnatal disease is confounded by gestational age and neonatal lung disease in human studies.
We hypothesized that fetal innate immune responses to LPS-induced chorioamnionitis would alter postnatal systemic immune and airway responsiveness resembling asthma later in life. To test this hypothesis, we evaluated immune status at birth and at 7 weeks after LPS-mediated chorioamnionitis. We evaluated leukocytes in the bronchoalveolar lavage fluid (BALF). Transforming growth factor-beta 1 (TGF-β 1 ) and elastin were measured in lung tissue. CD4, CD8, CD4/CD25, and gamma-delta subpopulations of T lymphocytes were measured in the PMLN and thymus. Monocytes isolated from blood were tested in vitro for responses to LPS and PamCysK4 (a ligand for TLRs 1 and 2). The immunologic measurements were performed at term and at 7 weeks of age and airway reactivity to metacholine (MCh) was tested at 7 weeks after inhalation of house dust mite (HDM).
Materials and methods
Antenatal treatment
All animal procedures were approved by the Animal Ethics Committee of the University of Western Australia, Australia. Ewes received intraamniotic (IA) injections with saline (controls) or 10 mg Escherichia coli 055:B5 endotoxin (LPS; Sigma-Aldrich, St. Louis, MO) at 90, 100, and 110 days of gestation ( Figure 1 ). At 147 days of gestation, lambs from LPS (n = 5) and control (n = 3) groups were delivered surgically to assess immune function of the fetus at term (term is approximately 147 days). The remaining ewes delivered spontaneously and airway responsiveness and immune status of lambs were evaluated at 7 weeks (LPS n = 5, Control n = 6). Lambs were killed with intravascular pentobarbital at term and at 7 weeks.
Airway reactivity
At 7 weeks lambs were sedated 2 days after HDM exposure (nebulized 1 mg) with diazepam 0.25 mg/kg (Sigma NSW, Australia) and Ketamine 5 mg/kg (Parnell Laboratories, NSW, Australia). They were intubated with a 6.0 mm ID tracheal tube (Portex Ltd, UK) and ventilated (Humming V, Metran, Japan) with a peak inspiratory pressure of 25 cm H 2 O, an inspiratory time 0.7 second and a rate of 40 breaths per minutes. A continuous infusion of intravenous Propofol (0.3-0.6 mg/kg/h, Repose, 0.1 mg/kg/min; Norbrook Laboratories Ltd., Victoria, Australia) and Remifentanil (0.3 mg/kg/h, Ultiva 0.05 μg/kg/min; GlaxoSmithKline, Victoria, Australia) was commenced, and following confirmation of deep anesthesia, neuromuscular blockade was achieved with vecuronium (0.1 mg/kg IV; Essex Pharma, Germany). Lambs were stabilized for 10 minutes before baseline measurements, and then received for 1 minute aerosols (1 mL) at 5 minute intervals of saline, followed by increasing concentrations (0.01%, 0.03%, 0.1%, 0.3%, 1%; w/v) of methacholine (Acetyl-β-methylcholine chloride; Sigma). Partitioned respiratory impedance ( Z rs ) was measured using the low-frequency forced oscillation technique (LFOT) using the FlexiVent (Module 5; Scireq, Montreal, Canada). After airway occlusion at commencement of expiration, respiratory system input impedance (Z rs ) measurements were obtained using a primewave (17 mutually prime frequencies from 0.5-19.75 Hz) over a 6-second apneic interval. Measurements were repeated every 30 seconds until peak response was observed (normally approximately 3 minutes).
Measurements of pressure and flow were transformed to the frequency domain and corrected for the impedance of the tracheal tube and measurement system. The partitioned airway (airway resistance − R aw ; airway inertance − I aw ) and tissue mechanical variables (tissue damping – G; tissue elastance − H) were determined by fitting the resultant impedance spectrum to the constant phase model. Hysteresivity (η) was calculated as G/H. Changes in airway and tissue variables were expressed relative to measurements obtained at baseline.
Bronchoalveolar lavage
After collection of the lungs, thymus, and the PMLN, the left lung was lavaged 3 times with 0.9% NaCl. The BALFs were pooled and centrifuged at 500 rpm for 5 minutes. Differential cell counts were performed on cytospin preparations after staining with Pappenheim-staining (May-Grünwald, Giemsa).
Lung tissue
Lung tissue from the right lower lobe (RLL) was snap frozen and stored at −80°C. For homogenization, a mix of lysis buffer (RIPA buffer, Sigma) and protease inhibitor (Sigma) was added to lung tissue. The lung tissue was homogenized (PRO Quick Connect Generators part no. 02-07095; PRO Scientific Inc., Oxford, CT) and centrifuged at 12 rcf (11.4 rpm) for 5 minutes at 4°C.
Enzyme-linked immunosorbent assay of TGF-β 1
Free, bound, and total TGF-β 1 (referred to R&D enzyme-linked immunosorbent assay [ELISA] kit; Minneapolis, MN as active, latent, and total TGFβ 1 ) were measured with R&D ELISA kits (DuoSet ELISA, human TGF/β1, no. DY240; R&D Systems). Free TGF-β 1 was measured for the original sample. Total TGF-β 1 was measured after acid activation of 150 μL of the sample with 30 μL 1 M HCl for 10 minutes at room temperature and 33 μL 1 M NaOH HEPES was used to stop the activation. Bound TGF-β 1 was calculated as the difference between total TGF-β and free TGF-β 1 .
Elastin staining
We measured elastin foci of secondary crests (concentrated vs nonconcentrated) and elastin fiber deposition in the vessel walls. Elastin fibers were stained black with a Hart’s resorcin fuchsin solution, using paraffin slides that were dehydrated and stained at 60°C and counterstained with a tartrazine solution. A magnification of 20× was used to save 3 pictures randomly. Counting of elastin foci was blinded, using ImageJ 1.41o software (Rasband; National Institutes of Health, Bethesda, MD) for quantification.
Flow cytometry
Single cell suspensions of thymus and PMLN were made with a strainer, assessed for viability, and counted using a Neubauer chamber (Hawksley, England). The 10 6 cells were incubated with primary monoclonal antibodies (mAbs: CD4, CD8, CD25, and TcR-1 receptor [gamma-delta T lymphocytes]; VMRD USA) for 30 minutes at 4°C. Cells were washed and incubated with fluorescence-conjugated secondary mAbs for 30 minutes at 4°C in the dark (FITC and R-PE labeled antibodies from SEROTEC, Great Britain). Cells were analysed on a FACScalibur instrument (Becton Dickinson, Franklin Lakes, NJ) using CellQuest software (Becton Dickinson). CD4/CD25 percentages are expressed relative to the population of cells expressing CD4. Other lymphocyte percentages are expressed relative to the total population of thymic lymphocytes.
Hydrogen peroxide assay for blood monocytes
Monocytes and macrophages were isolated with Percoll gradients from cord blood (at term) and peripheral blood (at 7 weeks) to perform functional tests. In brief, the isolated cells were cultured in RPMI 1640 media supplemented by 10% heat-inactivated fetal calf serum for 6 hours with LPS (a ligand for TLR 4; 100 ng/mL; E coli O55:B5; Sigma-Aldrich) and PamCysK4 (a synthetic ligand for TLRs 1 and 2; 5 μg/mL; EMC Microcollection, Tuebingen, Germany). PamCysK4 signals through the TLR2 pathway and is involved in maturation of the immune system. Control cells were exposed to saline (media). After washing the cells with phosphate buffered saline (PBS), the production of hydrogen peroxide by 1 × 10 6 monocytes was measured with an assay based on the oxidation of ferrous iron (Fe 2+ ) to ferric iron (Fe 3+ ) by hydrogen peroxide under acidic conditions (Bioxytech H 2 O 2 -560 assay; OXIS International, Portland, OR).
Statistical analysis
Microsoft Excel and GraphPad Prism 5 were used for statistics. Data are shown as mean ± SEM. Statistical differences between LPS and control groups were evaluated with Student t test and the Mann-Whitney test. Values were considered significant if P ≤ .05 vs the control (saline) group.
Results
Body and organ weights at birth and at 7 weeks of age
The body weights of LPS-exposed lambs were similar to the control group at term and at 7 weeks ( Table 1 ) indicating similar growth and development. The organ weights (PMLN, thymus, spleen, total lung) were not different for the groups at birth or at 7 weeks of age.
| Antenatal treatment | Body weight, kg | PMLN, g/kg | Thymus, g/kg | Spleen, g/kg | Total lung, g/kg | Animals, n |
|---|---|---|---|---|---|---|
| At term | ||||||
| Control | 5.0 ± 1.0 | 0.13 ± 0.01 | 2.07 ± 1.07 | 1.37 ± 0.18 | 35.3 ± 3.3 | 3 |
| LPS | 5.5 ± 1.0 | 0.21 ± 0.02 | 3.59 ± 0.43 | 1.74 ± 0.20 | 31.8 ± 2.4 | 5 |
| 7 weeks | ||||||
| Control | 16.2 ± 1.3 | 0.23 ± 0.02 | 2.52 ± 0.19 | 4.20 ± 0.30 | 12.9 ± 0.4 | 6 |
| LPS | 15.5 ± 1.5 | 0.26 ± 0.02 | 2.68 ± 0.48 | 4.62 ± 0.50 | 13.7 ± 0.4 | 5 |
Airway reactivity
Measurements with methacholine challenge at 7 weeks, including heart rate, peak inspiratory pressure, respiratory rate, and blood gas values (PaCO 2 , PaO 2 ) were not different between fetal LPS exposed and control lambs ( Table 2 ). Representative impedance spectra are shown in Figure 2 . There were no differences in baseline measurements between the groups. Airway resistance ( R aw ; P = .008; Figure 2 , A) and tissue damping ( G ; P = .037; Figure 2 , B) did not increase at 1% w/v MCh in LPS-exposed lambs, whereas they increased in the saline control animals, indicating decreased airway reactivity with fetal LPS exposure. LPS-exposed lambs and control lambs did not respond to lower MCh dosing. Tissue elastance ( H ; Figure 2 , C) and hysteresivity (η; Figure 2 , D) were not different at any MCh concentration between LPS-exposed and control lambs.
