Objective
The purpose of this study was to determine whether human amnion epithelial cells (hAECs) can modulate the pulmonary developmental consequences of intrauterine inflammation in fetal sheep that are exposed to intraamniotic lipopolysaccharide (LPS) injection.
Study Design
At 117 days’ gestation, fetal sheep (n = 16) received intraamniotic LPS (20 mg). hAECs were delivered at 0, 6, and 12 hours into the fetal jugular vein (n = 4), trachea (n = 4), or both (n = 4). Controls (n = 6) received equivalent administration of saline solution. Lungs were collected at 124 days.
Results
Intraamniotic LPS caused pulmonary inflammation and altered lung structure and function. hAECs attenuated changes in lung function and structure that had been induced by LPS: lung volume, 40 cm H 2 O ( P < .05, intravenous + intratracheal hAECs vs LPS), tissue-to-airspace ratio ( P < .05, intravenous + intratracheal hAECs vs LPS), and septal crest density ( P < .001, all hAEC groups vs LPS). Leukocyte infiltration of the lungs was not reduced by hAECs; however, inflammatory cytokines were reduced (tumor necrosis factor–α, P < .01, vs LPS; interleukin-1b, P < .01, vs LPS; interleukin-6, P < .01 vs LPS). Surfactant protein A and C messenger RNA was increased by LPS, although this was not statistically significant ( P > .05 vs control); there were significant increases in all hAEC-treated animals (surfactant protein–A, P < .05 vs LPS; surfactant protein–C, P < .01 vs LPS).
Conclusion
Human amnion epithelial cells attenuate the fetal pulmonary inflammatory response to experimental intrauterine inflammation and reduce, but (as administered in our study) do not prevent, consequent alterations in lung development.
Human amnion epithelial cells (hAECs) possess both regenerative and antiinflammatory properties and display low immunogenicity, which have enabled their application in human therapies for wound repair, burns, and corneal surgery without the development of immune rejection or need for immune suppression. One benefit of hAECs is that they are sourced from placentae, which are usually discarded after birth. This resolves many of the ethical and logistic limitations of other stem cells, including embryonic stem cells and bone marrow mesenchymal stem cells. With this in mind, we recently demonstrated a method for hAEC collection, isolation, and storage that is suitable for clinical use.
There are a number of characteristics that make hAECs an attractive cell therapy. Morphologically, hAECs maintain a normal karyotype in culture, have conserved long telomere lengths, and, unlike embryonic stem cells, do not form teratomas in vivo. hAECs can also transdifferentiate down mesodermal, ectodermal, and endodermal tissue lineages. Additionally, hAECs suppress lymphocyte proliferation and express negligible HLA antigens. With regard to the lung, we and others have noted that hAECs express thyroid transcription factor Nkx2.1, which is an early lung development marker. Furthermore, hAECs that are cultured in small airway growth media induce lung epithelial-specific gene expression.
Animal experiments have demonstrated the ability of hAECs to mitigate lung injury, likely through immune modulation and possibly integration into the alveolar epithelium. Specifically, in mouse models of bleomycin-induced pulmonary fibrosis, hAECs reduce acute expression of proinflammatory cytokines tumor necrosis factor-alpha (TNF-α), IL (interleukin)-6 and -1, and interferon-γ, and subsequent lung scarring. Given such findings in adult models, it is plausible that hAECs could modulate fetal lung injury that results from in utero infection.
Inflammation during pregnancy is linked causally to preterm birth and an increased risk of several diseases of prematurity. In a series of sheep studies, intrauterine inflammation initiated a fetal pulmonary inflammatory response that resulted in increased surfactant production and structural lung changes, which improved preterm lung function in the short term. These findings are consistent with human data that show a reduced risk of respiratory distress syndrome after chorioamnionitis. However, in the longer term, some aspects of this response are akin to the abnormal lung development that characterizes bronchopulmonary dysplasia. In support of these observations, some clinical studies have associated chorioamnionitis with an increased risk of bronchopulmonary dysplasia.
In light of their effects in modulating acute injury in adult lung, we hypothesized that hAECs would attenuate fetal lung inflammation after injection of lipopolysaccharide (LPS) into the amniotic cavity of sheep and would normalize consequent changes in lung structure. Thus, we aimed to investigate the capacity of hAECs to modulate the fetal lung response to intrauterine inflammation. Although there is evidence that intraamniotic LPS acts locally on the fetal lungs to elicit developmental changes, there is also a modest systemic inflammatory response. Therefore, we administered hAECs through fetal intravenous (IV) or intratracheal (IT) (or both) routes.
Materials and Methods
Animals and experimental groups
All experimental procedures were approved by the relevant institutional Animal Ethics Committee and conducted in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes (2006).
Hysterotomy was undertaken at 110 days of gestation (term is approximately 147 days of gestation) to provide access to the fetus. Ewes were anesthetized with thiopentone sodium intravenously and maintained with 1.5-2% isoflurane in oxygen through an endotracheal tube and positive pressure ventilation. Ampicillin (1 g) was administered intravenously to the ewe. Using an aseptic method, the fetal head and neck were delivered. Polyvinyl catheters (Dural Plastics, Silverwater, Australia) were inserted into the fetal carotid artery, jugular vein, trachea, and amniotic cavity. The fetus was returned to the uterus, and catheters were exteriorized through an incision in the ewe’s flank.
At 117 days of gestation, each fetus was assigned randomly to 1 of 6 groups: saline solution (control; fetuses were exposed to intraamniotic injection of saline solution; n = 6); LPS alone (fetuses were exposed to intraamniotic injection of LPS from Escherichia coli [055:B5 Sigma, St. Louis, MO]; n = 4); LPS + hAEC groups (fetuses received intraamniotic injection of LPS as described earlier in addition to hAECs that were administered by 1 of 3 routes to form 3 subgroups: IV [90 million hAECs were injected to the fetal jugular vein; n=4]; IT [180 million hAECs were delivered into the fetal trachea; n = 4]; IV + IT [90 million hAECs were delivered to the fetal jugular vein and fetal trachea at a total dose of 180 million cells; n = 4]). hAECs were administered at 3 time points: initially, with saline solution or LPS exposure at 0 hours, and again at 6 and 12 hours thereafter. Vascular and tracheal catheters of all fetuses were flushed with saline solutions at times that corresponded to hAEC administration.
Cell preparation and injection
Cells were isolated from term placentae that were donated by healthy volunteers who underwent elective caesarean section delivery; the procedures have been described in detail previously. Briefly, the amnion was peeled manually from the chorion and incubated in 0.05% trypsin for epithelial cell digestion. Once isolated, viability was assessed by Trypan blue exclusion (a minimum of 80% viability was required for use). Before injection, first passage hAECs were labeled passively with carboxyfluorescein succinimidyl ester (CFSE) and resuspended in sterile saline solution at a concentration of 6 million cells/mL. This isolation and cell suspension method results in only 1% of cells expressing the common leukocyte antigen (CD45), 94% EpCam-positive and <1% CD105- and CD90-positive cells in the resulting isolates, which indicates that these cells are 94% epithelial and <1% mesenchyme, respectively.
Fetal well-being was monitored periodically by an assessment of arterial partial pressure of oxygen, partial pressure of carbon dioxide, percent oxygen saturation, pH, hematocrit level, and lactate level with a blood gas analyzer (ABL 700; Radiometer, Copenhagen, Denmark).
Animals were killed at 124 days of gestation with a maternal overdose of IV pentobarbitone (Lethabarb Virbac Pty Ltd, Peakhurst, Australia). The fetus was removed, and total body and organ weights were documented. To assess lung function, we constructed deflation pressure-volume curves after air inflation of the lungs to 40 cm H 2 O pressure. Fetal lungs were separated; the left lung was ligated at the main stem bronchus, excised, and processed for analysis: RNA, protein- or fluorescence-activated cell sorting (FACS), which will be described later. The right lung was instilled with 4% paraformaldehyde at 20 cm H 2 O pressure, excised, and immersed in 4 % paraformaldehyde for 24 hours before histologic processing.
FACS analysis
CFSE-labeled hAECs were detected by FACS. Fresh tissue samples from the lung, brain, heart, liver, thymus, right kidney, and spleen (approximately 100 mg) were minced and passed through a 70-μm filter to obtain single cell suspensions and analyzed with a flow cytometry system (BD FACS Canto II Flow; BD Biosciences, New South Wales, Australia). Gates were optimized to exclude dead cells and autofluorescence. CFSE-positive cell numbers are expressed as a percentage of total cells per sample.
Point counting for tissue airspace ratio and septal crest density
Histologic sections (5 μm) were cut from 2 blocks per lobe of each right lung to provide 6 sections per fetus. Sections were stained with hematoxylin and eosin and analyzed to distinguish between tissue, septal crests, and airspaces. Researchers were blind to experimental group. This process was repeated for 5 random fields of view per section at ×400 magnification.
CD45 quantification
Histologic sections were prepared as described earlier. Immunohistochemical staining was performed with a mouse anti-sheep CD45 monoclonal antibody (MCA 2220, 1:50; AbD Serotec, Kidlington, UK). Sections were blocked with serum-free protein block (DAKO, New South Wales, Australia) before overnight primary incubation. Secondary staining was achieved with the LSAB-2 universal biotinylated link antibody kit (DAKO) with visualization by 3,3′-diaminobenzadine. CD45-positive cells were counted in a total of 5 fields of view per section with ImageJ software (National Institutes of Health, Bethesda, MD).
Protein isolation and multiplex analysis
Total protein was extracted from homogenized lung tissue that had been snap frozen. Protein concentration was determined with a Pierce BCA protein assay kit (Thermo Fisher Scientific, Waltham, MA). Protein was transferred to a 96-well plate and analyzed with a Luminex protein array system (Bio-Plex; Bio-Rad Laboratories Inc, Hercules, CA) for the presence of cytokines TNF-α, IL-1β, IL-6, interferon-γ, and IL-10. The assay is designed to detect murine cytokines and cross-reacts with sheep but has no cross-reactivity with human cytokines. Cytokine levels were normalized against total protein content.
RNA isolation and polymerase chain reaction analysis
RNA was isolated from snap-frozen lung tissue with a QIAGEN RNeasy Lipid Tissue Midi Kit (QIAGEN, Victoria, Australia) and converted to complementary DNA with the Superscript III Reverse Transcription System (Invitrogen, Victoria, Australia) according to the manufacturer’s instructions. Real-time polymerase chain reaction was performed with the Applied Biosystems Power SYBR Green PCR Master Mix and the Applied Biosystems 7900HT Fast Real-Time PCR System (Applied Biosystems, Victoria, Australia). Primers were directed against surfactant proteins A and C (SP-A, SP-C). Primer sequences and annealing temperatures are listed in Table 1 . Gene expression was normalized to 18S.
